Calculus 2nd Edition - Single Variable Solutions

This page intentionally left blank
Students Solutions Manual

to accompany Jon Rogawskis
Single Variable
CALCULUS
SECOND EDITION
BRIAN BRADIE
Christopher Newport University
ROGER LIPSETT
W. H. FREEMAN AND COMPANY
NEW YORK
2012 by W. H. Freeman and Company
ISBN-13: 978-1-4292-4290-5
ISBN-10: 1-4292-4290-6
All rights reserved
Printed in the United States of America
First Printing
W. H. Freeman and Company, 41 Madison Avenue, New York, NY 10010
Houndmills, Basingstoke RG21 6XS, England
www.whfreeman.com
CONTENTS
Chapter 1 PRECALCULUS REVIEW 1
1.1 Real Numbers, Functions, and Graphs 1
1.2 Linear and Quadratic Functions 8
1.3 The Basic Classes of Functions 13
1.4 Trigonometric Functions 16
1.5 Technology: Calculators and Computers 23
Chapter Review Exercises 27
Chapter 2 LIMITS 31
2.1 Limits, Rates of Change, and Tangent Lines 31
2.2 Limits: A Numerical and Graphical Approach 37
2.3 Basic Limit Laws 46
2.4 Limits and Continuity 49
2.5 Evaluating Limits Algebraically 57
2.6 Trigonometric Limits 61
2.7 Limits at Innity 66
2.8 Intermediate Value Theorem 73
2.9 The Formal Denition of a Limit 76
Chapter 3 DIFFERENTIATION 91
3.1 Definition of the Derivative 91
3.2 The Derivative as a Function 101
3.3 Product and Quotient Rules 112
3.4 Rates of Change 119
3.5 Higher Derivatives 126
3.6 Trigonometric Functions 132
3.7 The Chain Rule 138
3.8 Implicit Differentiation 147
3.9 Related Rates 157
Chapter 4 APPLICATIONS OF THE DERIVATIVE 174
4.1 Linear Approximation and Applications 174
4.2 Extreme Values 181
4.3 The Mean Value Theorem and Monotonicity 191
4.4 The Shape of a Graph 198
4.5 Graph Sketching and Asymptotes 206
4.6 Applied Optimization 220
4.7 Newtons Method 236
4.8 Antiderivatives 242
Chapter 5 THE INTEGRAL 260
5.1 Approximating and Computing Area 260
5.2 The Denite Integral 274
5.3 The Fundamental Theorem of Calculus, Part I 284
5.4 The Fundamental Theorem of Calculus, Part II 290
5.5 Net Change as the Integral of a Rate 296
5.6 Substitution Method 300
Chapter 6 APPLICATIONS OF THE INTEGRAL 317
6.1 Area Between Two Curves 317
6.2 Setting Up Integrals: Volume, Density, Average Value 328
6.3 Volumes of Revolution 336
6.4 The Method of Cylindrical Shells 346
6.5 Work and Energy 355
Chapter 7 EXPONENTIAL FUNCTIONS 370
7.1 Derivative of f (x) = b
x
and the Number e 370
7.2 Inverse Functions 378
7.3 Logarithms and Their Derivatives 383
7.4 Exponential Growth and Decay 393
7.5 Compound Interest and Present Value 398
7.6 Models Involving y
= k ( y b) 401
7.7 LHpitals Rule 407
7.8 Inverse Trigonometric Functions 415
7.9 Hyperbolic Functions 424
Chapter 8 TECHNIQUES OF INTEGRATION 446
8.1 Integration by Parts 446
8.2 Trigonometric Integrals 457
8.3 Trigonometric Substitution 467
8.4 Integrals Involving Hyperbolic and Inverse Hyperbolic
Functions 481
8.5 The Method of Partial Fractions 485
8.6 Improper Integrals 503
8.7 Probability and Integration 520
8.8 Numerical Integration 525
Chapter 9 FURTHER APPLICATIONS OF THE
INTEGRAL AND TAYLOR
POLYNOMIALS 555
9.1 Arc Length and Surface Area 555
9.2 Fluid Pressure and Force 564
9.3 Center of Mass 569
9.4 Taylor Polynomials 577
iii
iv C A L C U L U S CONTENTS
Chapter 10 INTRODUCTION TO DIFFERENTIAL
EQUATIONS 601
10.1 Solving Differential Equations 601
10.2 Graphical and Numerical Methods 614
10.3 The Logistic Equation 621
10.4 First-Order Linear Equations 626
Chapter 11 INFINITE SERIES 646
11.1 Sequences 646
11.2 Summing an Innite Series 658
11.3 Convergence of Series with Positive Terms 669
11.4 Absolute and Conditional Convergence 683
11.5 The Ratio and Root Tests 690
11.6 Power Series 697
11.7 Taylor Series 710
Chapter 12 PARAMETRIC EQUATIONS, POLAR
COORDINATES, AND CONIC
SECTIONS 742
12.1 Parametric Equations 742
12.2 Arc Length and Speed 759
12.3 Polar Coordinates 766
12.4 Area and Arc Length in Polar Coordinates 780
12.5 Conic Sections 789
June 7, 2011 LTSV SSM Second Pass
1 PRECALCULUS REVIEW
1.1 Real Numbers, Functions, and Graphs
Preliminary Questions
1. Giveanexampleof numbersa andb suchthat a |b|.
solution Takea = 3andb = 1. Thena 1= |b|.
2. Whichnumberssatisfy|a| = a?Whichsatisfy|a| = a?What about |a| = a?
solution Thenumbersa 0satisfy|a| = a and| a| = a. Thenumbersa 0satisfy|a| = a.
3. Giveanexampleof numbersa andb suchthat |a +b| < |a| +|b|.
solution Takea = 3andb = 1. Then
|a +b| = | 3+1| = | 2| = 2, but |a| +|b| = | 3| +|1| = 3+1= 4.
Thus, |a +b| < |a| +|b|.
4. What arethecoordinatesof thepoint lyingat theintersectionof thelinesx = 9andy = 4?
solution Thepoint (9, 4) liesat theintersectionof thelinesx = 9andy = 4.
5. Inwhichquadrant dothefollowingpointslie?
(a) (1, 4) (b) (3, 2) (c) (4, 3) (d) (4, 1)
solution
(a) Becauseboththex- andy-coordinatesof thepoint (1, 4) arepositive, thepoint (1, 4) liesintherst quadrant.
(b) Becausethex-coordinateof thepoint (3, 2) is negativebut they-coordinateis positive, thepoint (3, 2) lies in
thesecondquadrant.
(c) Becausethex-coordinateof thepoint (4, 3) is positivebut they-coordinateis negative, thepoint (4, 3) lies in
thefourthquadrant.
(d) Becauseboththex- andy-coordinatesof thepoint(4, 1) arenegative, thepoint(4, 1) liesinthethirdquadrant.
6. What istheradiusof thecirclewithequation(x 9)
2
+(y 9)
2
= 9?
solution Thecirclewithequation(x 9)
2
+(y 9)
2
= 9hasradius3.
7. Theequationf (x) = 5hasasolutionif (chooseone):
(a) 5belongstothedomainof f .
(b) 5belongstotherangeof f .
solution Thecorrect responseis(b): theequationf (x) = 5hasasolutionif 5belongstotherangeof f .
8. What kindof symmetrydoesthegraphhaveif f (x) = f (x)?
solution If f (x) = f (x), thenthegraphof f issymmetricwithrespect totheorigin.
Exercises
1. Useacalculator tondarational number r suchthat |r
2
| < 10
4
.
solution r must satisfy
2
10
4
< r <
2
+ 10
4
, or 9.869504 < r < 9.869705. r = 9.8696 =
12337
1250
would
beonesuchnumber.
Whichof (a)(f) aretruefor a = 3andb = 2?
(a) a 0
(d) 3a < 3b (e) 4a < 4b (f)
1
a
<
1
b
In Exercises 38, express the interval in terms of an inequality involving absolute value.
3. [2, 2]
solution |x| 2
(4, 4)
5. (0, 4)
solution Themidpoint of theinterval isc = (0+ 4)/2 = 2, andtheradiusisr = (4 0)/2 = 2; therefore, (0, 4)
canbeexpressedas|x 2| < 2.
[4, 0]
7. [1, 5]
solution Themidpoint of theinterval is c = (1+ 5)/2 = 3, andtheradius is r = (5 1)/2 = 2; therefore, the
interval [1, 5] canbeexpressedas|x 3| 2.
(2, 8) 1
2 C HA P T E R 1 PRECALCULUS REVIEW
In Exercises 912, write the inequality in the form a < x < b.
9. |x| < 8
solution 8< x < 8
|x 12| < 8
11. |2x +1| < 5
solution 5< 2x +1< 5so6< 2x < 4and3< x < 2
|3x 4| < 2
In Exercises 1318, express the set of numbers x satisfying the given condition as an interval.
13. |x| < 4
solution (4, 4)
|x| 9
15. |x 4| < 2
solution Theexpression|x 4| < 2isequivalent to2< x 4< 2. Therefore, 2< x < 6, whichrepresentsthe
interval (2, 6).
|x +7| < 2
17. |4x 1| 8
solution Theexpression|4x 1| 8isequivalent to8 4x 1 8or 7 4x 9. Therefore,
7
4
x
9
4
,
whichrepresentstheinterval [
7
4
,
9
4
].
|3x +5| < 1
In Exercises 1922, describe the set as a union of nite or innite intervals.
19. {x : |x 4| > 2}
solution x 4> 2or x 4< 2 x > 6or x < 2 (, 2) (6, )
{x : |2x +4| > 3}
21. {x : |x
2
1| > 2}
solution x
2
1> 2or x
2
1< 2 x
2
> 3or x
2
< 1(thiswill never happen) x >
3or x <
3
(,
3) (
3, ).
{x : |x
2
+2x| > 2}
23. Match(a)(f) with(i)(vi).
(a) a > 3 (b) |a 5| <
1
3
(c)
a
1
3
< 5 (d) |a| > 5

(e) |a 4| < 3 (f) 1 a 5
(i) a liestotheright of 3.
(ii) a liesbetween1and7.
(iii) Thedistancefroma to5islessthan
1
3
.
(iv) Thedistancefroma to3isat most 2.
(v) a islessthan5unitsfrom
1
3
.
(vi) a lieseither totheleft of 5or totheright of 5.
solution
(a) Onthenumber line, numbersgreater than3appear totheright; hence, a > 3isequivalenttothenumberstotheright
of 3: (i).
(b) |a 5| measuresthedistancefroma to5; hence, |a 5| <
1
3
issatisedbythosenumberslessthan
1
3
of aunitfrom
5: (iii).
(c) |a
1
3
| measuresthedistancefroma to
1
3
; hence, |a
1
3
| < 5issatisedbythosenumberslessthan5unitsfrom
1
3
: (v).
(d) Theinequality|a| > 5isequivalenttoa > 5or a < 5; thatis, either a liestotherightof 5or totheleftof 5: (vi).
(e) Theinterval describedbytheinequality|a 4| < 3hasacenter at 4andaradiusof 3; that is, theinterval consists
of thosenumbersbetween1and7: (ii).
(f) Theinterval describedbytheinequality1< x < 5hasacenter at 3andaradiusof 2; that is, theinterval consistsof
thosenumberslessthan2unitsfrom3: (iv).
Describe
_
x :
x
x +1
< 0
_
asaninterval.
25. Describe{x : x
2
+2x < 3} asaninterval. Hint: Plot y = x
2
+2x 3.
solution Theinequalityx
2
+2x < 3isequivalent tox
2
+2x 3< 0. Inthegurebelow, weseethat thegraphof
y = x
2
+ 2x 3falls belowthex-axis for 3 < x < 1. Thus, theset {x : x
2
+ 2x < 3} corresponds totheinterval
3< x < 1.
S E C T I ON 1.1 Real Numbers, Functions, and Graphs 3
4 3 2
2
2
4
6
8
10
y
x
1 2
y = x
2
+ 2x 3
Describetheset of real numberssatisfying|x 3| = |x 2| +1asahalf-inniteinterval.
27. Showthat if a > b, thenb
1
> a
1
, providedthat a andb havethesamesign. What happensif a > 0andb < 0?
solution Case1a: If a andb arebothpositive, thena > b 1>
b
a

1
b
>
1
a
.
Case1b: If a andb arebothnegative, thena > b 1
1
a
(again, sinceb isnegative).
Case2: If a > 0andb < 0, then
1
a
> 0and
1
b
< 0so
1
b
<
1
a
. (SeeExercise2f for anexampleof this).
Whichx satisfyboth|x 3| < 2and|x 5| < 1?
29. Showthat if |a 5| <
1
2
and|b 8| <
1
2
, then|(a +b) 13| < 1. Hint: Usethetriangleinequality.
solution
|a +b 13| = |(a 5) +(b 8)|
|a 5| +|b 8| (bythetriangleinequality)
<
1
2
+
1
2
= 1.
Supposethat |x 4| 1.
(a) What isthemaximumpossiblevalueof |x +4|?
(b) Showthat |x
2
16| 9.
31. Supposethat |a 6| 2and|b| 3.
(a) What isthelargest possiblevalueof |a +b|?
(b) What isthesmallest possiblevalueof |a +b|?
solution |a 6| 2guaranteesthat4 a 8,while|b| 3guaranteesthat3 b 3.Therefore1 a +b 11.
It followsthat
(a) thelargest possiblevalueof |a +b| is11; and
(b) thesmallest possiblevalueof |a +b| is1.
Provethat |x| |y| |x y|. Hint: Applythetriangleinequalitytoy andx y.
33. Expressr
1
= 0.27asafraction. Hint: 100r
1
r
1
isaninteger. Thenexpressr
2
= 0.2666. . . asafraction.
solution Let r
1
= 0.27. Weobservethat 100r
1
= 27.27. Therefore, 100r
1
r
1
= 27.270.27= 27and
r
1
=
27
99
=
3
11
.
Now, let r
2
= 0.2666. Then10r
2
= 2.666and100r
2
= 26.666. Therefore, 100r
2
10r
2
= 26.6662.666= 24and
r
2
=
24
90
=
4
15
.
Represent 1/7and4/27asrepeatingdecimals.
35. Thetext states: If the decimal expansions of numbers a and b agree to k places, then |a b| 10
k
. Showthat the
converseisfalse: For all k therearenumbersa andb whosedecimal expansionsdo not agree at all but |a b| 10
k
.
solution Let a = 1andb = 0.9(seethediscussionbeforeExample1). Thedecimal expansions of a andb donot
agree, but |10.9| < 10
k
for all k.
Plot eachpair of pointsandcomputethedistancebetweenthem:
(a) (1, 4) and(3, 2) (b) (2, 1) and(2, 4)
(c) (0, 0) and(2, 3) (d) (3, 3) and(2, 3)
37. Findtheequationof thecirclewithcenter (2, 4):
(a) withradiusr = 3.
(b) that passesthrough(1, 1).
solution
(a) Theequationof theindicatedcircleis(x 2)
2
+(y 4)
2
= 3
2
= 9.
(b) First determinetheradiusasthedistancefromthecenter totheindicatedpoint onthecircle:
r =
_
(21)
2
+(4(1))
2
=
26.
Thus, theequationof thecircleis(x 2)
2
+(y 4)
2
= 26.
Findall pointswithinteger coordinateslocatedat adistance5fromtheorigin. Thenndall pointswithinteger
coordinateslocatedat adistance5from(2, 3).
39. Determinethedomainandrangeof thefunction
f : {r, s, t, u} {A, B, C, D, E}
denedbyf (r) = A, f (s) = B, f (t ) = B, f (u) = E.
solution Thedomainistheset D = {r, s, t, u}; therangeistheset R = {A, B, E}.
Giveanexampleof afunctionwhosedomainD hasthreeelementsandwhoserangeR hastwoelements. Doesa
functionexist whosedomainD hastwoelementsandwhoserangeR hasthreeelements?
In Exercises 4148, nd the domain and range of the function.
41. f (x) = x
solution D : all reals; R : all reals
g(t ) = t
4
43. f (x) = x
3
solution D : all reals; R : all reals
g(t ) =

2t
45. f (x) = |x|
solution D : all reals; R : {y : y 0}
h(s) =
1
s
47. f (x) =
1
x
2
solution D : {x : x = 0}; R : {y : y > 0}
g(t ) = cos
1
t
In Exercises 4952, determine where f (x) is increasing.
49. f (x) = |x +1|
solution A graphof thefunctiony = |x +1| isshownbelow. Fromthegraph, weseethat thefunctionisincreasing
ontheinterval (1, ).
x
3 2 1
1
2
1
y
f (x) = x
3
51. f (x) = x
4
solution A graphof thefunctiony = x
4
isshownbelow. Fromthegraph, weseethat thefunctionisincreasingon
theinterval (0, ).
x
2 1 1 2
12
4
8
y
f (x) =
1
x
4
+x
2
+1
In Exercises 5358, nd the zeros of f (x) and sketch its graph by plotting points. Use symmetry and increase/decrease
information where appropriate.
53. f (x) = x
2
4
solution Zeros: 2
Increasing: x > 0
Decreasing: x < 0
Symmetry: f (x) = f (x) (evenfunction). So, y-axissymmetry.
2
2
4
4
y
x
2 1 1 2
f (x) = 2x
2
4
55. f (x) = x
3
4x
solution Zeros: 0, 2; Symmetry: f (x) = f (x) (oddfunction). Sooriginsymmetry.
5
5
10
10
y
x
2 1 1 2
f (x) = x
3
57. f (x) = 2x
3
solution Thisisanx-axisreectionof x
3
translatedup2units. Thereisonezeroat x =
3
2.
10
10
20
20
y
x
2 1 1 2
f (x) =
1
(x 1)
2
+1
59. Whichof thecurvesinFigure26isthegraphof afunction?
(A)
x
y
(B)
x
y
(C)
x
y
(D)
x
y
FIGURE 26
solution (B) isthegraphof afunction. (A), (C), and(D) all fail thevertical linetest.
Determinewhether thefunctioniseven, odd, or neither.
(a) f (x) = x
5
(b) g(t ) = t
3
t
2
(c) F(t ) =
1
t
4
+t
2
61. Determinewhether thefunctioniseven, odd, or neither.
(a) f (t ) =
1
t
4
+t +1
1
t
4
t +1
(b) g(t ) = 2
t
2
t
(c) G() = sin +cos (d) H() = sin(
2
)
solution
(a) Thisfunctionisoddbecause
f (t ) =
1
(t )
4
+(t ) +1
1
(t )
4
(t ) +1
=
1
t
4
t +1
1
t
4
+t +1
= f (t ).
(b) g(t ) = 2
t
2
(t )
= 2
t
2
t
= g(t ), sothisfunctionisodd.
(c) G() = sin() + cos() = sin + cos which is equal to neither G() nor G(), so this function is
neither oddnor even.
(d) H() = sin(()
2
) = sin(
2
) = H(), sothisfunctioniseven.
Writef (x) = 2x
4
5x
3
+12x
2
3x +4asthesumof anevenandanoddfunction.
63. Determinetheinterval onwhichf (x) =
1
x 4
isincreasingor decreasing.
solution Agraphof thefunctionisshownbelow. Fromthisgraphwecanseethatf (x) isdecreasingon(, 4) and
alsodecreasingon(4, ).
2
2
0
4
6
2
4
6
2 4 6 8 10
x
y
Statewhether thefunctionisincreasing, decreasing, or neither.
(a) Surfaceareaof asphereasafunctionof itsradius
(b) Temperatureat apoint ontheequator asafunctionof time
(c) Priceof anairlineticket asafunctionof thepriceof oil
(d) Pressureof thegasinapistonasafunctionof volume
In Exercises 6570, let f (x) be the function shown in Figure 27.
1 2 3 4
0
1
2
3
4
x
y
FIGURE 27
65. Findthedomainandrangeof f (x)?
solution D : [0, 4]; R : [0, 4]
Sketchthegraphsof f (x +2) andf (x) +2.
67. Sketchthegraphsof f (2x), f
_
1
2
x
_
, and2f (x).
solution Thegraphof y = f (2x) isobtainedbycompressingthegraphof y = f (x) horizontallybyafactor of 2(see
thegraphbelowontheleft). Thegraphof y = f (
1
2
x) isobtainedbystretchingthegraphof y = f (x) horizontallybya
factor of 2(seethegraphbelowinthemiddle). Thegraphof y = 2f (x) isobtainedbystretchingthegraphof y = f (x)
verticallybyafactor of 2(seethegraphbelowontheright).
y
x
1
2
3
4
1 2 3 4
f (2x)
y
x
1
2
3
4
2 4 6 8
f (x/2)
y
x
2
4
6
8
1 2 3 4
2f (x)
Sketchthegraphsof f (x) andf (x).
69. Extendthegraphof f (x) to[4, 4] sothat it isanevenfunction.
solution Tocontinuethegraphof f (x) totheinterval [4, 4] asanevenfunction, reect thegraphof f (x) across
they-axis(seethegraphbelow).
2 4
x
2 4
y
1
2
3
4
Extendthegraphof f (x) to[4, 4] sothat it isanoddfunction.
71. Supposethat f (x) hasdomain[4, 8] andrange[2, 6]. Findthedomainandrangeof:
(a) f (x) +3 (b) f (x +3)
(c) f (3x) (d) 3f (x)
solution
(a) f (x) + 3 is obtained by shifting f (x) upward threeunits. Therefore, thedomain remains [4, 8], whiletherange
becomes[5, 9].
(b) f (x +3) isobtainedbyshiftingf (x) leftthreeunits. Therefore, thedomainbecomes[1, 5], whiletherangeremains
[2, 6].
(c) f (3x) is obtained by compressingf (x) horizontally by afactor of three. Therefore, thedomain becomes [
4
3
,
8
3
],
whiletherangeremains[2, 6].
(d) 3f (x) isobtainedbystretchingf (x) verticallybyafactor of three. Therefore, thedomainremains[4, 8], whilethe
rangebecomes[6, 18].
Let f (x) = x
2
. Sketchthegraphover [2, 2] of:
(a) f (x +1) (b) f (x) +1
(c) f (5x) (d) 5f (x)
73. Supposethat thegraphof f (x) = sinx is compressedhorizontally by afactor of 2andthenshifted5units to the
right.
(a) What istheequationfor thenewgraph?
(b) What istheequationif yourst shift by5andthencompressby2?
(c) Verifyyour answersbyplottingyour equations.
solution
(a) Let f (x) = sinx. After compressingthegraphof f horizontally by afactor of 2, weobtainthefunctiong(x) =
f (2x) = sin2x. Shiftingthegraph5unitstotheright thenyields
h(x) = g(x 5) = sin2(x 5) = sin(2x 10).
(b) Letf (x) = sinx. After shiftingthegraph5unitstotheright, weobtainthefunctiong(x) = f (x 5) = sin(x 5).
Compressingthegraphhorizontallybyafactor of 2thenyields
h(x) = g(2x) = sin(2x 5).
(c) The gure below at the top left shows the graphs of y = sinx (the dashed curve), the sine graph compressed
horizontallybyafactor of 2(thedash, doubledotcurve) andthenshiftedright5units(thesolidcurve). Comparethislast
graphwiththegraphof y = sin(2x 10) shownat thebottomleft.
Thegurebelowat thetopright showsthegraphsof y = sinx (thedashedcurve), thesinegraphshiftedtotheright
5units(thedash, doubledotcurve) andthencompressedhorizontallybyafactor of 2(thesolidcurve). Comparethislast
graphwiththegraphof y = sin(2x 5) shownat thebottomright.
1
1
y
x
6 4 2 6 4 2
1
1
y
x
6 4 2 6 4 2
1
1
y
x
6 4 2 6 4 2
1
1
y
x
6 4 2 6 4 2
Figure28showsthegraphof f (x) = |x| +1. Matchthefunctions(a)(e) withtheir graphs(i)(v).
(a) f (x 1) (b) f (x) (c) f (x) +2
(d) f (x 1) 2 (e) f (x +1)
75. Sketchthegraphof f (2x) andf
_
1
2
x
_
, wheref (x) = |x| +1(Figure28).
solution Thegraphof y = f (2x) is obtainedby compressingthegraphof y = f (x) horizontally by afactor of 2
(seethegraphbelowontheleft). Thegraphof y = f (
1
2
x) isobtainedbystretchingthegraphof y = f (x) horizontally
byafactor of 2(seethegraphbelowontheright).
x
1
2
4
6
2 3 1 2 3
y
f (2x)
x
1
2
4
6
2 3 1 2 3
y
f (x/2)
Findthefunctionf (x) whosegraphisobtainedbyshiftingtheparabolay = x
2
threeunitstotheright andfour
unitsdown, asinFigure29.
77. Denef (x) tobethelarger of x and2x. Sketchthegraphof f (x). Whatareitsdomainandrange?Expressf (x)
intermsof theabsolutevaluefunction.
solution
x
1
1
2
1 2 3
y
Thegraphof y = f (x) isshownabove. Clearly, thedomainof f isthesetof all real numberswhiletherangeis{y | y 1}.
Noticethegraphhas thestandardV-shapeassociatedwiththeabsolutevaluefunction, but thebaseof theV has been
translatedtothepoint (1, 1). Thus, f (x) = |x 1| +1.
For eachcurveinFigure30, statewhether it issymmetricwithrespect tothey-axis, theorigin, both, or neither.
79. Showthat thesumof twoevenfunctionsisevenandthesumof twooddfunctionsisodd.
solution Even: (f +g)(x) = f (x) +g(x)
even
= f (x) +g(x) = (f +g)(x)
Odd: (f +g)(x) = f (x) +g(x)
odd
= f (x) +g(x) = (f +g)(x)
Supposethat f (x) andg(x) arebothodd. Whichof thefollowingfunctionsareeven?Whichareodd?
(a) f (x)g(x) (b) f (x)
3
(c) f (x) g(x) (d)
f (x)
g(x)
81. Provethat theonlyfunctionwhosegraphissymmetricwithrespect toboththey-axisandtheoriginisthefunction
f (x) = 0.
solution Supposef issymmetricwithrespecttothey-axis. Thenf (x) = f (x). If f isalsosymmetricwithrespect
totheorigin, thenf (x) = f (x). Thusf (x) = f (x) or 2f (x) = 0. Finally, f (x) = 0.
Further Insights and Challenges
Provethetriangleinequalitybyaddingthetwoinequalities
|a| a |a|, |b| b |b|
83. Showthat afractionr = a/b inlowest termshasanite decimal expansionif andonlyif
b = 2
n
5
m
for somen, m 0.
Hint: Observethat r hasanitedecimal expansionwhen10
N
r isaninteger for someN 0(andhenceb divides10
N
).
solution Supposer hasanitedecimal expansion. Thenthereexistsaninteger N 0suchthat 10
N
r isaninteger,
call it k. Thus, r = k/10
N
. Becausetheonlyprimefactorsof 10are2and5, it followsthat whenr iswritteninlowest
terms, itsdenominator must beof theform2
n
5
m
for someintegersn, m 0.
Conversely, suppose r = a/b in lowest with b = 2
n
5
m
for some integers n, m 0. Then r =
a
b
=
a
2
n
5
m
or
2
n
5
m
r = a. If m n, then2
m
5
m
r = a2
mn
or r =
a2
mn
10
m
andthus r has anitedecimal expansion(less thanor
equal tom terms, tobeprecise). Ontheother hand, if n > m, then2
n
5
n
r = a5
nm
or r =
a5
nm
10
n
andonceagainr has
anitedecimal expansion.
Let p = p
1
. . . p
s
beaninteger withdigitsp
1
, . . . , p
s
. Showthat
p
10
s
1
= 0.p
1
. . . p
s
Usethistondthedecimal expansionof r =
2
11
. Notethat
r =
2
11
=
18
10
2
1
85. A functionf (x) issymmetricwithrespect tothevertical linex = a if f (a x) = f (a +x).
(a) Drawthegraphof afunctionthat issymmetricwithrespect tox = 2.
(b) Showthat if f (x) issymmetricwithrespect tox = a, theng(x) = f (x +a) iseven.
solution
(a) Therearemanypossibilities, oneof whichis
x
1
1
2
5 4 3 2 1
y
y = |x 2|
(b) Let g(x) = f (x +a). Then
g(x) = f (x +a) = f (a x)
= f (a +x) symmetrywithrespect tox = a
= g(x)
Thus, g(x) iseven.
Formulateaconditionfor f (x) tobesymmetricwithrespect tothepoint (a, 0) onthex-axis.
1.2 Linear and Quadratic Functions
1. What istheslopeof theliney = 4x 9?
solution Theslopeof theliney = 4x 9is4, givenbythecoefcient of x.
2. Arethelinesy = 2x +1andy = 2x 4perpendicular?
solution Theslopesof perpendicular linesarenegativereciprocalsof oneanother. Becausetheslopeof y = 2x +1
is2andtheslopeof y = 2x 4is2, thesetwolinesarenot perpendicular.
3. Whenisthelineax +by = c parallel tothey-axis?Tothex-axis?
solution Thelineax +by = c will beparallel tothey-axiswhenb = 0andparallel tothex-axiswhena = 0.
4. Supposey = 3x +2. What isy if x increasesby3?
solution Becausey = 3x +2isalinear functionwithslope3, increasingx by3will leadtoy = 3(3) = 9.
S E C T I ON 1.2 Linear and Quadratic Functions 9
5. What istheminimumof f (x) = (x +3)
2
4?
solution Because(x +3)
2
0, it followsthat(x +3)
2
4 4. Thus, theminimumvalueof (x +3)
2
4is4.
6. What istheresult of completingthesquarefor f (x) = x
2
+1?
solution Becausethereisnox terminx
2
+1, completingthesquareonthisexpressionleadsto(x 0)
2
+1.
Exercises
In Exercises 14, nd the slope, the y-intercept, and the x-intercept of the line with the given equation.
1. y = 3x +12
solution Becausetheequationof thelineis giveninslope-intercept form, theslopeis thecoefcient of x andthe
y-intercept istheconstant term: that is, m = 3andthey-intercept is12. Todeterminethex-intercept, substitutey = 0
andthensolvefor x: 0= 3x +12or x = 4.
y = 4x
3. 4x +9y = 3
solution Todeterminetheslopeandy-intercept, werst solvetheequationfor y toobtaintheslope-intercept form.
This yields y =
4
9
x +
1
3
. Fromhere, weseethat theslopeis m =
4
9
and they-intercept is
1
3
. To determinethe
x-intercept, substitutey = 0andsolvefor x: 4x = 3or x =
3
4
.
y 3=
1
2
(x 6)
In Exercises 58, nd the slope of the line.
5. y = 3x +2
solution m = 3
y = 3(x 9) +2
7. 3x +4y = 12
solution First solvetheequationfor y toobtaintheslope-intercept form. Thisyieldsy =
3
4
x +3. Theslopeof the
lineisthereforem =
3
4
.
3x +4y = 8
In Exercises 920, nd the equation of the line with the given description.
9. Slope3, y-intercept 8
solution Usingtheslope-intercept formfor theequationof aline, wehavey = 3x +8.
Slope2, y-intercept 3
11. Slope3, passesthrough(7, 9)
solution Usingthepoint-slopeformfor theequationof aline, wehavey 9= 3(x 7) or y = 3x 12.
Slope5, passesthrough(0, 0)
13. Horizontal, passesthrough(0, 2)
solution Ahorizontal linehasaslopeof 0. Usingthepoint-slopeformfor theequationof aline, wehavey (2) =
0(x 0) or y = 2.
Passesthrough(1, 4) and(2, 7)
15. Parallel toy = 3x 4, passesthrough(1, 1)
solution Becausetheequationy = 3x 4isinslope-intercept form, wecanreadilyidentifythat it hasaslopeof 3.
Parallel lineshavethesameslope, sotheslopeof therequestedlineisalso3. Usingthepoint-slopeformfor theequation
of aline, wehavey 1= 3(x 1) or y = 3x 2.
Passesthrough(1, 4) and(12, 3)
17. Perpendicular to3x +5y = 9, passesthrough(2, 3)
solution Westart bysolvingtheequation3x +5y = 9for y toobtaintheslope-intercept formfor theequationof a
line. Thisyields
y =
3
5
x +
9
5
,
fromwhichweidentifytheslopeas
3
5
. Perpendicular lineshaveslopesthat arenegativereciprocalsof oneanother, so
theslopeof thedesiredlineism
=
5
3
. Usingthepoint-slopeformfor theequationof aline, wehavey 3=
5
3
(x 2)
or y =
5
3
x
1
3
.
Vertical, passesthrough(4, 9)
19. Horizontal, passesthrough(8, 4)
solution Ahorizontal linehasslope0. Usingthepointslopeformfortheequationof aline, wehavey 4= 0(x 8)
or y = 4.
Slope3, x-intercept 6
21. Findtheequationof theperpendicularbisectorof thesegmentjoining(1, 2) and(5, 4) (Figure11). Hint: Themidpoint
Qof thesegment joining(a, b) and(c, d) is
_
a +c
2
,
b +d
2
_
.
Q
(1, 2)
(5, 4)
Perpendicular
bisector
x
y
FIGURE 11
solution Theslopeof thesegment joining(1, 2) and(5, 4) is
m =
42
51
=
1
2
andthemidpoint of thesegment (Figure11) is
midpoint =
_
1+5
2
,
2+4
2
_
= (3, 3)
Theperpendicular bisector hasslope1/m = 2andpassesthrough(3, 3), soitsequationis: y 3 = 2(x 3) or
y = 2x +9.
Intercept-Intercept Form Showthat if a, b = 0, thenthelinewithx-intercept x = a andy-intercept y = b
hasequation(Figure12)
x
a
+
y
b
= 1
23. Findanequationof thelinewithx-intercept x = 4andy-intercept y = 3.
solution FromExercise22,
x
4
+
y
3
= 1or 3x +4y = 12.
Findy suchthat (3, y) liesonthelineof slopem = 2through(1, 4).
25. Determinewhether thereexistsaconstant c suchthat thelinex +cy = 1:
(a) Hasslope4 (b) Passesthrough(3, 1)
(c) Ishorizontal (d) Isvertical
solution
(a) Rewritingtheequationof thelineinslope-intercept formgivesy =
x
c
+
1
c
. Tohaveslope4requires
1
c
= 4or
c =
1
4
.
(b) Substitutingx = 3andy = 1intotheequationof thelinegives3+c = 1or c = 2.
(c) From(a), weknowtheslopeof thelineis
1
c
. Thereisnovaluefor c that will makethisslopeequal to0.
(d) Withc = 0, theequationbecomesx = 1. Thisistheequationof avertical line.
Assumethat thenumber N of concert ticketsthat canbesoldat apriceof P dollarsper ticket isalinear function
N(P) for 10 P 40. DetermineN(P) (calledthedemandfunction) if N(10) = 500andN(40) = 0. What isthe
decreaseN inthenumber of ticketssoldif thepriceisincreasedbyP = 5dollars?
27. Materialsexpandwhenheated. Consider ametal rodof lengthL
0
at temperatureT
0
. If thetemperatureischanged
by anamount T , thentherodslengthchangesby L = L
0
T , where isthethermal expansioncoefcient. For
steel, = 1.2410
5
C
1
.
(a) A steel rodhaslengthL
0
= 40cmat T
0
= 40
C. Finditslengthat T = 90
C.
(b) Finditslengthat T = 50
C if itslengthat T
0
= 100
C is65cm.
(c) ExpresslengthL asafunctionof T if L
0
= 65cmat T
0
= 100
C.
solution
(a) WithT = 90
C andT
0
= 40
C, T = 50
C. Therefore,
L = L
0
T = (1.2410
5
)(40)(50) = 0.0248 and L = L
0
+L = 40.0248cm.
(b) WithT = 50
C andT
0
= 100
C, T = 50
C. Therefore,
L = L
0
T = (1.2410
5
)(65)(50) = 0.0403 and L = L
0
+L = 64.9597cm.
(c) L = L
0
+L = L
0
+L
0
T = L
0
(1+T ) = 65(1+(T 100))
Dothepoints(0.5, 1), (1, 1.2), (2, 2) lieonaline?
29. Findb suchthat (2, 1), (3, 2), and(b, 5) lieonaline.
solution Theslopeof thelinedeterminedbythepoints(2, 1) and(3, 2) is
2(1)
32
= 3.
Tolieonthesameline, theslopebetween(3, 2) and(b, 5) must alsobe3. Thus, werequire
52
b 3
=
3
b 3
= 3,
or b = 4.
S E C T I ON 1.2 Linear and Quadratic Functions 11
Findanexpressionfor thevelocityv asalinear functionof t that matchesthefollowingdata.
t (s) 0 2 4 6
v (m/s) 39.2 58.6 78 97.4
31. TheperiodT of apendulumismeasuredfor pendulumsof several different lengthsL. Basedonthefollowingdata,
doesT appear tobealinear functionof L?
L (cm) 20 30 40 50
T (s) 0.9 1.1 1.27 1.42
solution Examinetheslopebetweenconsecutivedatapoints. Therst pair of datapointsyieldsaslopeof
1.10.9
3020
= 0.02,
whilethesecondpair of datapointsyieldsaslopeof
1.271.1
4030
= 0.017,
andthelast pair of datapointsyieldsaslopeof
1.421.27
5040
= 0.015
Becausethethreeslopesarenot equal, T doesnot appear tobealinear functionof L.
Showthat f (x) islinear of slopem if andonlyif
f (x +h) f (x) = mh (for all x andh)
33. Findtherootsof thequadraticpolynomials:
(a) 4x
2
3x 1 (b) x
2
2x 1
solution
(a) x =
3
94(4)(1)
2(4)
=
3
25
8
= 1or
1
4
(b) x =
2
4(4)(1)(1)
2
=
2
8
2
= 1
2
In Exercises 3441, complete the square and nd the minimum or maximum value of the quadratic function.
y = x
2
+2x +5
35. y = x
2
6x +9
solution y = (x 3)
2
; therefore, theminimumvalueof thequadraticpolynomial is0, andthisoccursat x = 3.
y = 9x
2
+x
37. y = x
2
+6x +2
solution y = x
2
+ 6x + 9 9+ 2 = (x + 3)
2
7; therefore, theminimumvalueof thequadratic polynomial is
7, andthisoccursat x = 3.
y = 2x
2
4x 7
39. y = 4x
2
+3x +8
solution y = 4x
2
+3x +8= 4(x
2
3
4
x +
9
64
) +8+
9
16
= 4(x
3
8
)
2
+
137
16
; therefore, themaximumvalue
of thequadraticpolynomial is
137
16
, andthisoccursat x =
3
8
.
y = 3x
2
+12x 5
41. y = 4x 12x
2
solution y = 12(x
2
x
3
) = 12(x
2
x
3
+
1
36
) +
1
3
= 12(x
1
6
)
2
+
1
3
; therefore, themaximumvalueof the
quadraticpolynomial is
1
3
, andthisoccursat x =
1
6
.
Sketchthegraphof y = x
2
6x +8byplottingtherootsandtheminimumpoint.
43. Sketchthegraphof y = x
2
+4x +6byplottingtheminimumpoint, they-intercept, andoneother point.
solution y = x
2
+ 4x + 4 4+ 6= (x + 2)
2
+ 2sotheminimumoccursat (2, 2). If x = 0, theny = 6andif
x = 4, y = 6. Thisisthegraphof x
2
movedleft 2unitsandup2units.
4 3 2 1
2
4
6
8
10
y
x
If thealleles A andB of thecystic brosis geneoccur inapopulationwithfrequencies p and1 p (wherep
isafractionbetween0and1), thenthefrequencyof heterozygouscarriers(carrierswithbothalleles) is2p(1p).
Whichvalueof p givesthelargest frequencyof heterozygouscarriers?
45. For whichvaluesof c doesf (x) = x
2
+cx +1haveadoubleroot? Noreal roots?
solution A doubleroot occurswhenc
2
4(1)(1) = 0or c
2
= 4. Thus, c = 2.
Therearenoreal rootswhenc
2
4(1)(1) < 0or c
2
< 4. Thus, 2< c < 2.
Let f (x) be a quadratic function and c a constant. Which of the following statements is correct? Explain
graphically.
(a) Thereisauniquevalueof c suchthat y = f (x) c hasadoubleroot.
(b) Thereisauniquevalueof c suchthat y = f (x c) hasadoubleroot.
47. Provethat x +
1
x
2for all x > 0. Hint: Consider (x
1/2
x
1/2
)
2
.
solution Let x > 0. Then
_
x
1/2
x
1/2
_
2
= x 2+
1
x
.
Because(x
1/2
x
1/2
)
2
0, it followsthat
x 2+
1
x
0 or x +
1
x
2.
Let a, b > 0. Showthat thegeometric mean

ab isnot larger thanthearithmetic mean (a + b)/2. Hint: Usea
variationof thehint giveninExercise47.
49. If objects of weights x and w
1
are suspended fromthe balance in Figure 13(A), the cross-beamis horizontal if
bx = aw
1
. If thelengthsa andb areknown, wemayusethisequationtodetermineanunknownweightx byselectingw
1
suchthat thecross-beamishorizontal. If a andb arenot knownprecisely, wemight proceedasfollows. First balancex
byw
1
ontheleft asin(A). Thenswitchplacesandbalancex byw
2
ontheright asin(B). Theaverage x =
1
2
(w
1
+w
2
)
givesanestimatefor x. Showthat x isgreater thanor equal tothetrueweight x.
w
1
(A)
a
x
b
(B)
w
2
x
a b
FIGURE 13
solution First notebx = aw
1
andax = bw
2
. Thus,
x =
1
2
(w
1
+w
2
)
=
1
2
_
bx
a
+
ax
b
_
=
x
2
_
b
a
+
a
b
_
x
2
(2) byExercise47
= x
Findnumbersx andy withsum10andproduct 24. Hint: Findaquadraticpolynomial satisedbyx.
51. Findapair of numberswhosesumandproduct arebothequal to8.
solution Let x andy benumberswhosesumandproduct arebothequal to8. Thenx +y = 8andxy = 8. Fromthe
secondequation, y =
8
x
. Substitutingthisexpressionfor y intherst equationgivesx +
8
x
= 8or x
2
8x +8= 0. By
thequadraticformula,
x =
8
6432
2
= 42
2.
If x = 4+2
2, then
y =
8
4+2
2
=
8
4+2
42
2
42
2
= 42
2.
Ontheother hand, if x = 42
2, then
y =
8
42
2
=
8
42
4+2
2
4+2
2
= 4+2
2.
Thus, thetwonumbersare4+2
2and42
2.
Showthat theparabolay = x
2
consists of all points P suchthat d
1
= d
2
, whered
1
is thedistancefromP to
_
0,
1
4
_
andd
2
isthedistancefromP totheliney =
1
4
(Figure14).
S E C T I ON 1.3 The Basic Classes of Functions 13
53. Showthat if f (x) andg(x) arelinear, thensoisf (x) +g(x). Isthesametrueof f (x)g(x)?
solution If f (x) = mx +b andg(x) = nx +d, then
f (x) +g(x) = mx +b +nx +d = (m+n)x +(b +d),
whichislinear. f (x)g(x) isnot generallylinear. Take, for example, f (x) = g(x) = x. Thenf (x)g(x) = x
2
.
Showthat if f (x) andg(x) arelinear functionssuchthat f (0) = g(0) andf (1) = g(1), thenf (x) = g(x).
55. Showthaty/x for thefunctionf (x) = x
2
over theinterval [x
1
, x
2
] isnotaconstant, butdependsontheinterval.
Determinetheexact dependenceof y/x onx
1
andx
2
.
solution For x
2
,
y
x
=
x
2
2
x
2
1
x
2
x
1
= x
2
+x
1
.
UseEq. (2) toderivethequadraticformulafor therootsof ax
2
+bx +c = 0.
57. Let a, c = 0. Showthat therootsof
ax
2
+bx +c = 0 and cx
2
+bx +a = 0
arereciprocalsof eachother.
solution Let r
1
andr
2
betheroots of ax
2
+ bx + c andr
3
andr
4
betheroots of cx
2
+ bx + a. Without loss of
generality, let
r
1
=
b +
_
b
2
4ac
2a
1
r
1
=
2a
b +
_
b
2
4ac
b
_
b
2
4ac
b
_
b
2
4ac
=
2a(b
_
b
2
4ac)
b
2
b
2
+4ac
=
b
_
b
2
4ac
2c
= r
4
.
Similarly, youcanshow
1
r
2
= r
3
.
Show, bycompletingthesquare, that theparabola
y = ax
2
+bx +c
iscongruent toy = ax
2
byavertical andhorizontal translation.
59. ProveVites Formulas: Thequadratic polynomial with and asrootsisx
2
+ bx + c, whereb = and
c = .
solution If aquadraticpolynomial hasroots and, thenthepolynomial is
(x )(x ) = x
2
x x + = x
2
+( )x +.
Thus, b = andc = .
1.3 The Basic Classes of Functions
1. Giveanexampleof arational function.
solution Oneexampleis
3x
2
2
7x
3
+x 1
.
2. Is|x| apolynomial function?What about |x
2
+1|?
solution |x| isnot apolynomial; however, becausex
2
+1> 0for all x, it followsthat |x
2
+1| = x
2
+1, whichis
apolynomial.
3. Whatisunusual aboutthedomainof thecompositefunctionf g forthefunctionsf (x) = x
1/2
andg(x) = 1|x|?
solution Recall that (f g)(x) = f (g(x)). Now, for any real number x, g(x) = 1 |x| 1< 0. Becausewe
cannot takethesquareroot of anegativenumber, it follows that f (g(x)) is not denedfor any real number. Inother
words, thedomainof f (g(x)) istheemptyset.
4. Isf (x) =
_
1
2
_
x
increasingor decreasing?
solution Thefunctionf (x) = (
1
2
)
x
is anexponential functionwithbaseb =
1
2
< 1. Therefore, f is adecreasing
function.
5. Giveanexampleof atranscendental function.
solution Onepossibilityisf (x) = e
x
sinx.
Exercises
In Exercises 112, determine the domain of the function.
1. f (x) = x
1/4
solution x 0
g(t ) = t
2/3
3. f (x) = x
3
+3x 4
solution All reals
h(z) = z
3
+z
3
5. g(t ) =
1
t +2
solution t = 2
f (x) =
1
x
2
+4
7. G(u) =
1
u
2
4
solution u = 2
f (x) =
x
x
2
9
9. f (x) = x
4
+(x 1)
3
solution x = 0, 1
F(s) = sin
_
s
s +1
_
11. g(y) = 10
y+y
1
solution y > 0
f (x) =
x +x
1
(x 3)(x +4)
In Exercises 1324, identify each of the following functions as polynomial, rational, algebraic, or transcendental.
13. f (x) = 4x
3
+9x
2
8
solution Polynomial
f (x) = x
4
15. f (x) =

x
solution Algebraic
f (x) =
_
1x
2 17. f (x) =
x
2
x +sinx
solution Transcendental
f (x) = 2
x
19. f (x) =
2x
3
+3x
97x
2
solution Rational
f (x) =
3x 9x
1/2
97x
2
21. f (x) = sin(x
2
)
solution Transcendental
f (x) =
x
x +1
23. f (x) = x
2
+3x
1
solution Rational
f (x) = sin(3
x
)
25. Isf (x) = 2
x
2
atranscendental function?
solution Yes.
Showthat f (x) = x
2
+3x
1
andg(x) = 3x
3
9x +x
2
arerational functionsthat is, quotientsof polyno-
mials.
In Exercises 2734, calculate the composite functions f g and g f , and determine their domains.
27. f (x) =

x, g(x) = x +1
solution f (g(x)) =

x +1; D: x 1, g(f (x)) =

x +1; D: x 0
f (x) =
1
x
, g(x) = x
4
29. f (x) = 2
x
, g(x) = x
2
solution f (g(x)) = 2
x
2
; D: R, g(f (x)) = (2
x
)
2
= 2
2x
; D: R
f (x) = |x|, g() = sin
31. f () = cos, g(x) = x
3
+x
2
solution f (g(x)) = cos(x
3
+x
2
); D: R, g(f ()) = cos
3
+cos
2
; D: R
f (x) =
1
x
2
+1
, g(x) = x
2
33. f (t ) =
1
t
, g(t ) = t
2
solution f (g(t )) =
1
_
t
2
; D: Not validfor anyt , g(f (t )) =
_
1
t
_
2
=
1
t
; D: t > 0
f (t ) =

t , g(t ) = 1t
3
S E C T I ON 1.3 The Basic Classes of Functions 15
35. Thepopulation(inmillions) of acountryasafunctionof timet (years) isP(t ) = 30.2
0.1t
. Showthatthepopulation
doublesevery 10years. Showmoregenerally that for any positiveconstantsa andk, thefunctiong(t ) = a2
kt
doubles
after 1/k years.
solution Let P(t ) = 30 2
0.1t
. Then
P(t +10) = 30 2
0.1(t +10)
= 30 2
0.1t +1
= 2(30 2
0.1t
) = 2P(t ).
Hence, thepopulationdoublesinsizeevery10years. Inthemoregeneral case, let g(t ) = a2
kt
. Then
g
_
t +
1
k
_
= a2
k(t +1/k)
= a2
kt +1
= 2a2
kt
= 2g(t ).
Hence, thefunctiong doublesafter 1/k years.
Findall valuesof c suchthat f (x) =
x +1
x
2
+2cx +4
hasdomainR. Further Insights and Challenges
In Exercises 3743, we dene the rst differencef of a function f (x) by f (x) = f (x +1) f (x).
37. Showthat if f (x) = x
2
, thenf (x) = 2x +1. Calculatef for f (x) = x andf (x) = x
3
.
solution f (x) = x
2
: f (x) = f (x +1) f (x) = (x +1)
2
x
2
= 2x +1
f (x) = x: f (x) = x +1x = 1
f (x) = x
3
: f (x) = (x +1)
3
x
3
= 3x
2
+3x +1
Showthat (10
x
) = 9 10
x
and, moregenerally, that (b
x
) = (b 1)b
x
.
39. Showthat for anytwofunctionsf andg, (f +g) = f +g and(cf ) = c(f ), wherec isanyconstant.
solution (f +g) = (f (x +1) +g(x +1)) (f (x) g(x))
= (f (x +1) f (x)) +(g(x +1) g(x)) = f (x) +g(x)
(cf ) = cf (x +1) cf (x) = c(f (x +1) f (x)) = cf (x).
SupposewecanndafunctionP(x) suchthat P = (x + 1)
k
andP(0) = 0. Provethat P(1) = 1
k
, P(2) =
1
k
+2
k
, and, moregenerally, for everywholenumber n,
P(n) = 1
k
+2
k
+ +n
k
41. First showthat
P(x) =
x(x +1)
2
satisesP = (x +1). ThenapplyExercise40toconcludethat
1+2+3+ +n =
n(n +1)
2
solution Let P(x) = x(x +1)/2. Then
P(x) = P(x +1) P(x) =
(x +1)(x +2)
2

x(x +1)
2
=
(x +1)(x +2x)
2
= x +1.
Also, notethat P(0) = 0. Thus, byExercise40, withk = 1, it followsthat
P(n) =
n(n +1)
2
= 1+2+3+ +n.
Calculate(x
3
), (x
2
), and(x).Thenndapolynomial P(x) of degree3suchthatP = (x +1)
2
andP(0) = 0.
Concludethat P(n) = 1
2
+2
2
+ +n
2
.
43. ThisexercisecombinedwithExercise40showsthatforall wholenumbersk, thereexistsapolynomial P(x) satisfying
Eq. (1). ThesolutionrequirestheBinomial Theoremandproof byinduction(seeAppendixC).
(a) Showthat (x
k+1
) = (k +1) x
k
+ , wherethedotsindicatetermsinvolvingsmaller powersof x.
(b) Showbyinductionthat thereexistsapolynomial of degreek +1withleadingcoefcient 1/(k +1):
P(x) =
1
k +1
x
k+1
+
suchthat P = (x +1)
k
andP(0) = 0.
solution
(a) BytheBinomial Theorem:
(x
n+1
) = (x +1)
n+1
x
n+1
=
_
x
n+1
+
_
n +1
1
_
x
n
+
_
n +1
2
_
x
n1
+ +1
_
x
n+1
=
_
n +1
1
_
x
n
+
_
n +1
2
_
x
n1
+ +1
Thus,
(x
n+1
) = (n +1) x
n
+
wherethedotsindicatetermsinvolvingsmaller powersof x.
(b) For k = 0, notethat P(x) = x satisesP = (x +1)
0
= 1andP(0) = 0.
Nowsupposethepolynomial
P(x) =
1
k
x
k
+p
k1
x
k1
+ +p
1
x
whichclearlysatisesP(0) = 0alsosatisesP = (x +1)
k1
. Wetrytoprovetheexistenceof
Q(x) =
1
k +1
x
k+1
+q
k
x
k
+ +q
1
x
suchthat Q = (x +1)
k
. Observethat Q(0) = 0.
If Q = (x +1)
k
andP = (x +1)
k1
, then
Q = (x +1)
k
= (x +1)P = xP(x) +P
Bythelinearityof (Exercise39), wendQP = xP or (QP) = xP. Bydenition,
QP =
1
k +1
x
k+1
+
_
q
k

1
k
_
x
k
+ +(q
1
p
1
)x,
so, bythelinearityof ,
(QP) =
1
k +1
(x
k+1
) +
_
q
k

1
k
_
(x
k
) + +(q
1
p
1
) = x(x +1)
k1
Bypart (a),
(x
k+1
) = (k +1)x
k
+L
k1,k1
x
k1
+. . . +L
k1,1
x +1
(x
k
) = kx
k1
+L
k2,k2
x
k2
+. . . +L
k2,1
x +1
.
.
.
(x
2
) = 2x +1
wheretheL
i,j
arereal numbersfor eachi, j.
Toconstruct Q, wehavetogrouplikepowersof x onbothsidesof Eq. (43b). Thisyieldsthesystemof equations
1
k +1
_
(k +1)x
k
_
= x
k
1
k +1
L
k1,k1
x
k1
+
_
q
k

1
k
_
kx
k1
= (k 1)x
k1
.
.
.
1
k +1
+
_
q
k

1
k
_
+(q
k1
p
k1
) + +(q
1
p
1
) = 0.
Therst equationis identically true, andthesecondequationcanbesolvedimmediately for q
k
. Substitutingthevalue
of q
k
intothethirdequationof thesystem, wecanthensolvefor q
k1
. Wecontinuethisprocessuntil wesubstitutethe
valuesof q
k
, q
k1
, . . . q
2
intothelast equation, andthensolvefor q
1
.
1.4 Trigonometric Functions
1. Howisit possiblefor twodifferent rotationstodenethesameangle?
solution Workingfromthesameinitial radius, two rotations that differ by awholenumber of full revolutions will
havethesameendingradius; consequently, thetworotationswill denethesameangleeventhoughthemeasuresof the
rotationswill bedifferent.
2. Givetwodifferent positiverotationsthat denetheangle/4.
solution Theangle/4is denedby any rotationof theform
4
+ 2k wherek is aninteger. Thus, two different
positiverotationsthat denetheangle/4are
4
+2(1) =
9
4
and

4
+2(5) =
41
4
.
S E C T I ON 1.4 Trigonometric Functions 17
3. Giveanegativerotationthat denestheangle/3.
solution Theangle/3isdenedbyanyrotationof theform
3
+2k wherek isaninteger. Thus, anegativerotation
that denestheangle/3is
3
+2(1) =
5
3
.
4. Thedenitionof cos usingright trianglesapplieswhen(choosethecorrect answer):
(a) 0< <
2
(b) 0< < (c) 0< < 2
solution Thecorrect responseis(a): 0< <

2
.
5. What istheunit circledenitionof sin?
solution Let O denotethecenter of theunit circle, andlet P beapoint ontheunit circlesuchthat theradius OP
makesanangle withthepositivex-axis. Then, sin isthey-coordinateof thepoint P.
6. Howdoestheperiodicityof sin andcos followfromtheunit circledenition?
solution Let O denotethecenter of theunit circle, andlet P beapoint ontheunit circlesuchthat theradiusOP
makesanangle withthepositivex-axis. Then, cos andsin arethex- andy-coordinates, respectively, of thepoint
P. Theangle +2 isobtainedfromtheangle bymakingonefull revolutionaroundthecircle. Theangle +2 will
thereforehavetheradiusOP asitsterminal side. Thus
cos( +2) = cos and sin( +2) = sin.
Inother words, sin andcos areperiodicfunctions.
Exercises
1. Findtheanglebetween0and2 equivalent to13/4.
solution Because13/4> 2, werepeatedlysubtract 2 until wearriveat aradianmeasurethat isbetween0and
2. After onesubtraction, wehave13/42 = 5/4. Because0< 5/4< 2, 5/4istheanglemeasurebetween
0and2 that isequivalent to13/4.
Describe = /6byanangleof negativeradianmeasure.
3. Convert fromradianstodegrees:
(a) 1 (b)

3
(c)
5
12
(d)
3
4
solution
(a) 1
_
180
_
=
180
57.3
(b)

3
_
180
_
= 60
(c)
5
12
_
180
_
=
75
23.87
(d)
3
4
_
180
_
= 135
Convert fromdegreestoradians:
(a) 1
(b) 30
(c) 25
(d) 120
5. Findthelengthsof thearcssubtendedbytheangles and radiansinFigure20.

4
q = 0.9
f = 2
FIGURE 20 Circleof radius4.
solution s = r = 4(.9) = 3.6; s = r = 4(2) = 8
Calculatethevaluesof thesixstandardtrigonometricfunctionsfor theangle inFigure21.
7. Fill intheremainingvaluesof (cos, sin) for thepointsinFigure22.
p
2
0 (0, 0) p
5p
6
7p
6
11p
6
3p
4
5p
4
7p
4 4p
3
5p
3
3p
2
2p
3
( , )
p
6

2
3 1
2
( , )
2
3 1
2
p
3
( , )
p
4

2
2
2
2
FIGURE 22
solution

2
2
3
3
4
5
6

7
6
(cos, sin) (0, 1)
_
1
2
,
3
2
_ _
2
2
,
2
2
_ _
3
2
,
1
2
_
(1, 0)
_
3
2
,
1
2
_
5
4
4
3
3
2
5
3
7
4
11
6
(cos, sin)
_
2
2
,

2
2
_ _
1
2
,

3
2
_
(0, 1)
_
1
2
,

3
2
_ _
2
2
,

2
2
_ _
3
2
,
1
2
_
Findthevaluesof thesixstandardtrigonometricfunctionsat = 11/6.
In Exercises 914, use Figure 22 to nd all angles between 0and 2 satisfying the given condition.
9. cos =
1
2
solution =

3
,
5
3
tan = 1
11. tan = 1
solution =
3
4
,
7
4
csc = 2
13. sinx =
3
2
solution x =

3
,
2
3
sect = 2
15. Fill inthefollowingtableof values:
6

4

3

2
2
3
3
4
5
6
tan
sec
solution
6

4

3

2
2
3
3
4
5
6
tan
1
3
1

3 und
3 1
1
3
sec
2
3

2 2 und 2
2
2
3
Completethefollowingtableof signs:
sin cos tan cot sec csc
0< <
2
+ +
2
< <
< <
3
2
3
2
< < 2
17. Showthat if tan = c and0 < /2, thencos = 1/
_
1+c
2
. Hint: Drawaright trianglewhoseoppositeand
adjacent sideshavelengthsc and1.
solution Because0 < /2, wecanusethedenitionof thetrigonometric functionsintermsof right triangles.
tan is theratioof thelengthof thesideoppositetheangle tothelengthof theadjacent side. Withc =
c
1
, welabel
thelengthof theoppositesideasc andthelengthof theadjacent sideas1(seethediagrambelow). BythePythagorean
theorem, thelengthof thehypotenuseis
_
1+c
2
. Finally, weusethefactthatcos istheratioof thelengthof theadjacent
sidetothelengthof thehypotenusetoobtain
cos =
1
_
1+c
2
.
q
1+ c
2
c
1
Supposethat cos =
1
3
.
(a) Showthat if 0 < /2, thensin = 2
2/3andtan = 2
2.
(b) Findsin andtan if 3/2 < 2.
In Exercises 1924, assume that 0 < /2.
19. Findsin andtan if cos =
5
13
.
solution Consider thetrianglebelow. Thelengthsof thesideadjacent totheangle andthehypotenusehavebeen
labeledsothatcos =
5
13
. Thelengthof thesideoppositetheangle hasbeencalculatedusingthePythagoreantheorem:
_
13
2
5
2
= 12. Fromthetriangle, weseethat
sin =
12
13
and tan =
12
5
.
5
12 13
Findcos andtan if sin =
3
5
.
21. Findsin, sec, andcot if tan =
2
7
.
solution If tan =
2
7
, thencot =
7
2
. For theremainingtrigonometric functions, consider thetrianglebelow. The
lengthsof thesidesoppositeandadjacenttotheangle havebeenlabeledsothattan =
2
7
. Thelengthof thehypotenuse
hasbeencalculatedusingthePythagoreantheorem:
_
2
2
+7
2
=

53. Fromthetriangle, weseethat
sin =
2
53
=
2
53
53
and sec =
53
7
.
2
q
53
7
Findsin, cos, andsec if cot = 4.
23. Findcos2 if sin =
1
5
.
solution Usingthedoubleangleformulacos2 = cos
2
sin
2
andthefundamental identitysin
2
+cos
2
= 1,
wendthat cos2 = 12sin
2
. Thus, cos2 = 12(1/25) = 23/25.
Findsin2 andcos2 if tan =

2.
25. Findcos andtan if sin = 0.4and/2 < .
solution Wecandeterminethemagnitude of cos andtan usingthetriangleshownbelow. Thelengths of the
sideoppositetheangle andthehypotenusehavebeenlabeledsothat sin = 0.4=
2
5
. Thelengthof thesideadjacent
totheangle wascalculatedusingthePythagoreantheorem:
_
5
2
2
2
=

21. Fromthetriangle, weseethat
|cos| =
21
5
and |tan| =
2
21
=
2
21
21
.
Because/2 < , bothcos andtan arenegative; consequently,
cos =
21
5
and tan =
2
21
21
.
2
5
q
21
Findcos andsin if tan = 4and < 3/2.
27. Findcos if cot =
4
3
andsin < 0.
solution Wecandeterminethemagnitudeof cos usingthetriangleshownbelow. Thelengthsof thesidesopposite
andadjacent totheangle havebeenlabeledsothat cot =
4
3
. Thelengthof thehypotenusewascalculatedusingthe
Pythagoreantheorem:
_
3
2
+4
2
= 5. Fromthetriangle, weseethat
|cos| =
4
5
.
Becausecot =
4
3
> 0andsin < 0, theangle mustbeinthethirdquadrant; consequently, cos will benegativeand
cos =
4
5
.
4
3
5

Findtan if sec =

5andsin < 0.
29. Findthevaluesof sin, cos, andtan for theanglescorrespondingtotheeight pointsinFigure23(A) and(B).
(0.3965, 0.918)
(A) (B)
(0.3965, 0.918)
FIGURE 23
solution Letsstart withthefour pointsinFigure23(A).
Thepoint intherst quadrant hascoordinates(0.3965, 0.918). Therefore,
sin = 0.918, cos = 0.3965, and tan =
0.918
0.3965
= 2.3153.
Thecoordinatesof thepoint inthesecondquadrant are(0.918, 0.3965). Therefore,
sin = 0.3965, cos = 0.918, and tan =
0.3965
0.918
= 0.4319.
Becausethepoint inthethirdquadrant issymmetrictothepoint intherst quadrant withrespect totheorigin, its
coordinatesare(0.3965, 0.918). Therefore,
sin = 0.918, cos = 0.3965, and tan =
0.918
0.3965
= 2.3153.
Becausethepoint inthefourthquadrant issymmetrictothepoint inthesecondquadrant withrespect totheorigin,
itscoordinatesare(0.918, 0.3965). Therefore,
sin = 0.3965, cos = 0.918, and tan =
0.3965
0.918
= 0.4319.
Nowconsider thefour pointsinFigure23(B).
Thepoint intherst quadrant hascoordinates(0.3965, 0.918). Therefore,
sin = 0.918, cos = 0.3965, and tan =
0.918
0.3965
= 2.3153.
Thepoint inthesecondquadrant isareectionthroughthey-axisof thepoint intherst quadrant. Itscoordinates
aretherefore(0.3965, 0.918) and
sin = 0.918, cos = 0.3965, and tan =
0.918
0.3965
= 2.3153.
Becausethepoint inthethirdquadrant issymmetrictothepoint intherst quadrant withrespect totheorigin, its
coordinatesare(0.3965, 0.918). Therefore,
sin = 0.918, cos = 0.3965, and tan =
0.918
0.3965
= 2.3153.
Becausethepoint inthefourthquadrant issymmetrictothepoint inthesecondquadrant withrespect totheorigin,
itscoordinatesare(0.3965, 0.918). Therefore,
sin = 0.918, cos = 0.3965, and tan =
0.918
0.3965
= 2.3153.
Refer toFigure24(A). Expressthefunctionssin, tan, andcsc intermsof c.
31. Refer toFigure24(B). Computecos, sin, cot, andcsc.
c
1 1
0.3
(B) (A)

FIGURE 24
solution By thePythagoreantheorem, thelengthof thesideoppositetheangle inFigure24(B) is
_
10.3
2
=
0.91. Consequently,
cos =
0.3
1
= 0.3, sin =
0.91
1
=
0.91, cot =
0.3
0.91
and csc =
1
0.91
.
Expresscos
_
+

2
_
andsin
_
+

2
_
intermsof cos andsin. Hint: Findtherelationbetweenthecoordinates
(a, b) and(c, d) inFigure25.
33. Usetheadditionformulatocomputecos
_
3
+

4
_
exactly.
solution
cos
_
3
+

4
_
= cos
3
cos
4
sin
3
sin
4
=
1
2
2
2

3
2

2
2
=
6
4
.
Usetheadditionformulatocomputesin
_
3

4
_
exactly.
In Exercises 3538, sketch the graph over [0, 2].
35. 2sin4
solution
2
1
2
1
y
x
6 5 4 3 2 1
cos
_
2
_

2
__
37. cos
_
2

2
_
solution
1
0.5
1
0.5
y
x
6 5 4 3 2 1
sin
_
2
_

2
_
+
_
+2
39. Howmany pointslieontheintersectionof thehorizontal liney = c andthegraphof y = sinx for 0 x < 2?
Hint: Theanswer dependsonc.
solution Recall that for anyx, 1 sinx 1. Thus, if |c| > 1, thehorizontal liney = c andthegraphof y = sinx
never intersect. If c = +1, theny = c andy = sinx intersect at thepeak of thesinecurve; that is, they intersect at
x =

2
. Ontheother hand, if c = 1, theny = c andy = sinx intersect at thebottomof thesinecurve; that is, they
intersect at x =
3
2
. Finally, if |c| < 1, thegraphsof y = c andy = sinx intersect twice.
Howmanypointslieontheintersectionof thehorizontal liney = c andthegraphof y = tanx for 0 x < 2?
In Exercises 4144, solve for 0 < 2 (see Example 4).
41. sin2 +sin3 = 0
solution sin = sin when = + 2k or = + + 2k. Substituting = 2 and = 3, wehave
either2 = 3 +2k or2 = +3 +2k. Solvingeachof theseequationsfor yields =
2
5
k or = 2k.
Thesolutionsontheinterval 0 < 2 arethen
= 0,
2
5
,
4
5
, ,
6
5
,
8
5
.
sin = sin2
43. cos4 +cos2 = 0
solution cos = cos when + = +2k or = + +2k. Substituting = 4 and = 2, wehave
either 6 = +2k or 4 = 2 + +2k. Solvingeachof theseequationsfor yields =

6
+

3
k or =

2
+k.
Thesolutionsontheinterval 0 < 2 arethen
=

6
,
2
,
5
6
,
7
6
,
3
2
,
11
6
.
sin = cos2
In Exercises 4554, derive the identity using the identities listed in this section.
45. cos2 = 2cos
2
1
solution Startingfromthedoubleangleformulafor cosine, cos
2
=
1
2
(1+cos2), wesolvefor cos2. Thisgives
2cos
2
= 1+cos2 andthencos2 = 2cos
2
1.
cos
2

2
=
1+cos
2
47. sin
2
=
_
1cos
2
solution Substitutex = /2intothedoubleangleformulafor sine, sin
2
x =
1
2
(1cos2x) toobtainsin
2
_
2
_
=
1cos
2
. Takingthesquareroot of bothsidesyieldssin
_
2
_
=
_
1cos
2
.
sin( +) = sin
49. cos( +) = cos
solution Fromtheadditionformulafor thecosinefunction, wehave
cos( +) = cos cos sin sin = cos(1) = cos
tanx = cot
_
2
x
_ 51. tan( ) = tan
solution UsingExercises48and49,
tan( ) =
sin( )
cos( )
=
sin( +())
cos( +())
=
sin()
cos()
=
sin
cos
= tan.
Thesecondtolast equalityoccursbecausesinx isanoddfunctionandcosx isanevenfunction.
tan2x =
2tanx
1tan
2
x
53. tanx =
sin2x
1+cos2x
solution Usingtheadditionformulafor thesinefunction, wend
sin2x = sin(x +x) = sinx cosx +cosx sinx = 2sinx cosx.
ByExercise45, weknowthat cos2x = 2cos
2
x 1. Therefore,
sin2x
1+cos2x
=
2sinx cosx
1+2cos
2
x 1
=
2sinx cosx
2cos
2
x
=
sinx
cosx
= tanx.
sin
2
x cos
2
x =
1cos4x
8
55. UseExercises48and49toshowthat tan andcot areperiodicwithperiod.
solution ByExercises48and49,
tan( +) =
sin( +)
cos( +)
=
sin
cos
= tan,
and
cot( +) =
cos( +)
sin( +)
=
cos
sin
= cot.
Thus, bothtan andcot areperiodicwithperiod.
Usetheidentityof Exercise45toshowthat cos
8
isequal to
_
1
2
+
2
4
.
57. UsetheLawof CosinestondthedistancefromP toQinFigure26.
P
Q
8
10
7/9
FIGURE 26
solution BytheLawof Cosines, thedistancefromP toQis
_
10
2
+8
2
2(10)(8) cos
7
9
= 16.928.
S E C T I ON 1.5 Technology: Calculators and Computers 23
UseFigure27toderivetheLawof CosinesfromthePythagoreanTheorem.
59. Usetheadditionformulatoprove
cos3 = 4cos
3
3cos
solution
cos3 = cos(2 +) = cos2 cos sin2 sin = (2cos
2
1) cos (2sin cos) sin
= cos(2cos
2
12sin
2
) = cos(2cos
2
12(1cos
2
))
= cos(2cos
2
12+2cos
2
) = 4cos
3
3cos
Usetheadditionformulasfor sineandcosinetoprove
tan(a +b) =
tana +tanb
1tana tanb
cot(a b) =
cota cotb +1
cotb cota
61. Let betheanglebetweentheliney = mx +b andthex-axis[Figure28(A)]. Provethat m = tan.
y = mx + b
q
x
r
s
(A)
y
q
x
(B)
y
L
2
L
1
FIGURE 28
solution UsingthedistanceslabeledinFigure28(A), weseethat theslopeof thelineisgivenbytheratior/s. The
tangent of theangle isgivenbythesameratio. Therefore, m = tan.
LetL
1
andL
2
bethelinesof slopem
1
andm
2
[Figure28(B)]. Showthattheangle betweenL
1
andL
2
satises
cot =
m
2
m
1
+1
m
2
m
1
.
63. Perpendicular Lines UseExercise62toprovethat twolineswithnonzeroslopesm
1
andm
2
areperpendicular if
andonlyif m
2
= 1/m
1
.
solution If linesareperpendicular, thentheanglebetweenthemis = /2
cot(/2) =
1+m
1
m
2
m
1
m
2
0=
1+m
1
m
2
m
1
m
2
m
1
m
2
= 1 m
1
=
1
m
2
Applythedouble-angleformulatoprove:
(a) cos
8
=
1
2
_
2+
2
(b) cos

16
=
1
2
_
2+
_
2+
2
Guessthevaluesof cos

32
andof cos

2
n
for all n.
1.5 Technology: Calculators and Computers
1. Isthereadenitewayof choosingtheoptimal viewingrectangle, or isit best toexperiment until youndaviewing
rectangleappropriatetotheproblemat hand?
solution It isbest toexperiment withthewindowsizeuntil oneisfoundthat isappropriatefor theproblemat hand.
2. Describethecalculator screenproducedwhenthefunctiony = 3+x
2
isplottedwithviewingrectangle:
(a) [1, 1] [0, 2] (b) [0, 1] [0, 4]
solution
(a) Usingtheviewingrectangle[1, 1] by[0, 2], thescreenwill displaynothingastheminimumvalueof y = 3+x
2
isy = 3.
(b) Usingtheviewingrectangle[0, 1] by [0, 4], thescreenwill display theportionof theparabolabetweenthepoints
(0, 3) and(1, 4).
3. AccordingtotheevidenceinExample4, it appearsthat f (n) = (1+1/n)
n
never takesonavaluegreater than3for
n > 0. Doesthisevidenceprove that f (n) 3for n > 0?
solution No, thisevidencedoesnot constituteaproof that f (n) 3for n 0.
4. Howcanagraphingcalculator beusedtondtheminimumvalueof afunction?
solution Experiment with theviewing window to zoomin on thelowest point on thegraph of thefunction. The
y-coordinateof thelowest point onthegraphistheminimumvalueof thefunction.
Exercises
The exercises in this section should be done using a graphing calculator or computer algebra system.
1. Plot f (x) = 2x
4
+3x
3
14x
2
9x +18intheappropriateviewingrectanglesanddetermineitsroots.
solution Usingaviewingrectangleof [4, 3] by[20, 20], weobtaintheplot below.
10
20
20
10
y
x
4 2 3 1 1 2 3
Now, therootsof f (x) arethex-interceptsof thegraphof y = f (x). Fromtheplot, wecanidentifythex-interceptsas
3, 1.5, 1, and2. Therootsof f (x) arethereforex = 3, x = 1.5, x = 1, andx = 2.
Howmanysolutionsdoesx
3
4x +8= 0have?
3. Howmanypositive solutionsdoesx
3
12x +8= 0have?
solution Thegraphof y = x
3
12x + 8shownbelowhastwox-interceptstotheright of theorigin; thereforethe
equationx
3
12x +8= 0hastwopositivesolutions.
20
40
60
60
40
20
y
x
4 2 4 2
Doescosx +x = 0haveasolution?A positivesolution?
5. Findall thesolutionsof sinx =

x for x > 0.
solution Solutionstotheequationsinx =

x correspondtopointsof intersectionbetweenthegraphsof y = sinx
andy =

x. Thetwographsareshownbelow; theonlypointof intersectionisatx = 0. Therefore, therearenosolutions
of sinx =

x for x > 0.
x
1
2
5 4 3 2 1
y
1
Howmanysolutionsdoescosx = x
2
have?
7. Let f (x) = (x 100)
2
+ 1000. What will thedisplay showif yougraphf (x) intheviewingrectangle[10, 10]
by[10, 10]? Findanappropriateviewingrectangle.
solution Because(x 100)
2
0for all x, it followsthat f (x) = (x 100)
2
+1000 1000for all x. Thus, using
aviewingrectangleof [10, 10] by [10, 10] will display nothing. Theminimumvalueof thefunctionoccurs when
x = 100, soanappropriateviewingrectanglewouldbe[50, 150] by[1000, 2000].
Plot f (x) =
8x +1
8x 4
inanappropriateviewingrectangle. What arethevertical andhorizontal asymptotes?
9. Plotthegraphof f (x) = x/(4x) inaviewingrectanglethatclearlydisplaysthevertical andhorizontal asymptotes.
solution Fromthegraphof y =
x
4x
shownbelow, weseethat thevertical asymptoteisx = 4andthehorizontal
asymptoteisy = 1.
2
2
y
x
8 4
4 8 12 16
Illustratelocal linearityfor f (x) = x
2
byzoominginonthegraphat x = 0.5(seeExample6).
11. Plotf (x) = cos(x
2
) sinx for 0 x 2. Thenillustratelocal linearityatx = 3.8bychoosingappropriateviewing
rectangles.
solution The following three graphs display f (x) = cos(x
2
) sinx over the intervals [0, 2], [3.5, 4.1] and
[3.75, 3.85]. Thenal graphlookslikeastraight line.
S E C T I ON 1.5 Technology: Calculators and Computers 25
x
1
1
1 2 3 4 5 6
y
x
1
1
3.5 3.6 3.7 3.8 3.9 4
y
x
0.2
0.4
0.2
3.76 3.8 3.78 3.82 3.84
y
If P
0
dollarsaredepositedinabankaccountpaying5%interestcompoundedmonthly, thentheaccounthasvalue
P
0
_
1+
0.05
12
_
N
after N months. Find, tothenearestinteger N, thenumber of monthsafter whichtheaccountvalue
doubles.
In Exercises 1318, investigate the behavior of the function as n or x grows large by making a table of function values
and plotting a graph (see Example 4). Describe the behavior in words.
13. f (n) = n
1/n
solution Thetableandgraphsbelowsuggest that asn getslarge, n
1/n
approaches1.
n n
1/n
10 1.258925412
10
2
1.047128548
10
3
1.006931669
10
4
1.000921458
10
5
1.000115136
10
6
1.000013816
x
y
1
0 2 4 6 8 10
x
y
1
0 200 400 600 800 1000
f (n) =
4n +1
6n 5
15. f (n) =
_
1+
1
n
_
n
2
solution Thetableandgraphsbelowsuggest that asn getslarge, f (n) tendstoward.
n
_
1+
1
n
_
n
2
10 13780.61234
10
2
1.63582871110
43
10
3
1.19530660310
434
10
4
5.34178331210
4342
10
5
1.70233305410
43429
10
6
1.83973874910
434294
x
y
10,000
0 2 4 6 8 10
x
y
1 10
43
0 20 40 60 80 100
f (x) =
_
x +6
x 4
_
x
17. f (x) =
_
x tan
1
x
_
x
solution Thetableandgraphsbelowsuggest that asx getslarge, f (x) approaches1.
x
_
x tan
1
x
_
x
10 1.033975759
10
2
1.003338973
10
3
1.000333389
10
4
1.000033334
10
5
1.000003333
10
6
1.000000333
x
y
1
1.1
1.2
1.3
1.4
1.5
5 10 15 20
x
20 40 60 80 100
y
1
1.1
1.2
1.3
1.4
1.5
f (x) =
_
x tan
1
x
_
x
2
19. Thegraphof f () = Acos + B sin isasinusoidal wavefor any constantsA andB. Conrmthisfor (A, B) =
(1, 1), (1, 2), and(3, 4) byplottingf ().
solution Thegraphsof f () = cos +sin, f () = cos +2sin andf () = 3cos +4sin areshownbelow.
y
x
2 2
(A, B) = (1, 1)
4 6 8
1
1
y
x
2 2
(A, B) = (1, 2)
4 6 8
2
1
2
1
y
x
2 2
(A, B) = (3, 4)
4 6 8
4
2
4
2
Find themaximumvalueof f () for thegraphs produced in Exercise19. Can you guess theformulafor the
maximumvalueintermsof A andB?
21. Findtheintervalsonwhichf (x) = x(x +2)(x 3) ispositivebyplottingagraph.
solution Thefunctionf (x) = x(x + 2)(x 3) is positivewhenthegraphof y = x(x + 2)(x 3) lies abovethe
x-axis. Thegraphof y = x(x + 2)(x 3) isshownbelow. Clearly, thegraphliesabovethex-axisandthefunctionis
positivefor x (2, 0) (3, ).
20
40
20
y
x
4 2 2 4
Findtheset of solutionstotheinequality(x
2
4)(x
2
1) < 0byplottingagraph.
23. Let f
1
(x) = x and dene a sequence of functions by f
n+1
(x) =
1
2
(f
n
(x) + x/f
n
(x)). For example,
f
2
(x) =
1
2
(x +1). Useacomputer algebrasystemtocomputef
n
(x) for n = 3, 4, 5andplotf
n
(x) together with
x for
x 0. What doyounotice?
solution Withf
1
(x) = x andf
2
(x) =
1
2
(x +1), wecalculate
f
3
(x) =
1
2
_
1
2
(x +1) +
x
1
2
(x +1)
_
=
x
2
+6x +1
4(x +1)
f
4
(x) =
1
2
x
2
+6x +1
4(x +1)
+
x
x
2
+6x+1
4(x+1)
=
x
4
+28x
3
+70x
2
+28x +1
8(1+x)(1+6x +x
2
)
and
f
5
(x) =
1+120x +1820x
2
+8008x
3
+12870x
4
+8008x
5
+1820x
6
+120x
7
+x
8
16(1+x)(1+6x +x
2
)(1+28x +70x
2
+28x
3
+x
4
)
.
A plot of f
1
(x), f
2
(x), f
3
(x), f
4
(x), f
5
(x) and
x isshownbelow, withthegraphof
x shownasadashedcurve. It
seemsasif thef
n
areasymptoticto
x.
y
x
4
8
12
40 20 60 80 100
Set P
0
(x) = 1 and P
1
(x) = x. The Chebyshev polynomials (useful in approximation theory) are dened
inductivelybytheformulaP
n+1
(x) = 2xP
n
(x) P
n1
(x).
(a) Showthat P
2
(x) = 2x
2
1.
(b) ComputeP
n
(x) for 3 n 6usingacomputer algebrasystemor byhand andplot P
n
(x) over [1 1]
CHAPTER REVIEW EXERCISES
1. Express(4, 10) asaset {x : |x a| < c} for suitablea andc.
solution Thecenter of theinterval (4, 10) is
4+10
2
= 7andtheradiusis
104
2
= 3. Therefore, theinterval (4, 10) is
equivalent totheset {x : |x 7| < 3}.
Expressasaninterval:
(a) {x : |x 5| < 4} (b) {x : |5x +3| 2}
3. Express{x : 2 |x 1| 6} asaunionof twointervals.
solution Theset {x : 2 |x 1| 6} consistsof thosenumbersthat areat least 2but at most 6unitsfrom1. The
numbers larger than1that satisfy theseconditions are3 x 7, whilethenumbers smaller than1that satisfy these
conditionsare5 x 1. Therefore{x : 2 |x 1| 6} = [5, 1] [3, 7].
Giveanexampleof numbersx, y suchthat |x| +|y| = x y.
5. Describethepairsof numbersx, y suchthat |x +y| = x y.
solution First consider thecasewhenx +y 0. Then|x +y| = x +y andweobtaintheequationx +y = x y.
Thesolutionof thisequationisy = 0. Thus, thepairs(x, 0) withx 0satisfy|x +y| = x y. Next, consider thecase
whenx +y < 0. Then|x +y| = (x +y) = x y andweobtaintheequationx y = x y. Thesolutionof this
equationisx = 0. Thus, thepairs(0, y) withy < 0alsosatisfy|x +y| = x y.
Sketchthegraphof y = f (x +2) 1, wheref (x) = x
2
for 2 x 2.
In Exercises 710, let f (x) be the function shown in Figure 1.
1 2 3 4
1
2
0
3
x
y
FIGURE 1
7. Sketchthegraphsof y = f (x) +2andy = f (x +2).
solution Thegraphof y = f (x) + 2isobtainedby shiftingthegraphof y = f (x) up2units(seethegraphbelow
at theleft). Thegraphof y = f (x + 2) isobtainedby shiftingthegraphof y = f (x) totheleft 2units(seethegraph
belowat theright).
x
yy
x
1
2
3
4
5
1 2 3 4
f (x) + 2
1 2
x
yy
x
1
2
3
4
5
1 2 3 4
f (x + 2)
1 2
Sketchthegraphsof y =
1
2
f (x) andy = f
_
1
2
x
_
.
9. Continuethegraphof f (x) totheinterval [4, 4] asanevenfunction.
solution Tocontinuethegraphof f (x) totheinterval [4, 4] asanevenfunction, reect thegraphof f (x) across
they-axis(seethegraphbelow).
1 4 2 3
x
1 2 3 4
y
1
2
3
Continuethegraphof f (x) totheinterval [4, 4] asanoddfunction.
In Exercises 1114, nd the domain and range of the function.
11. f (x) =

x +1
solution Thedomainof thefunctionf (x) =

x +1is{x : x 1} andtherangeis{y : y 0}.
f (x) =
4
x
4
+1
13. f (x) =
2
3x
solution Thedomainof thefunctionf (x) =
2
3x
is{x : x = 3} andtherangeis{y : y = 0}.
f (x) =
_
x
2
x +5
15. Determinewhether thefunctionisincreasing, decreasing, or neither:
(a) f (x) = 3
x
(b) f (x) =
1
x
2
+1
(c) g(t ) = t
2
+t (d) g(t ) = t
3
+t
solution
(a) Thefunctionf (x) = 3
x
canberewrittenasf (x) = (
1
3
)
x
. Thisisanexponential functionwithabaselessthan1;
therefore, thisisadecreasingfunction.
(b) Fromthegraphof y = 1/(x
2
+1) shownbelow, weseethatthisfunctionisneither increasingnor decreasingfor all
x (thoughit isincreasingfor x < 0anddecreasingfor x > 0).
x
3 2 1 1 2 3
y
0.2
0.4
0.6
0.8
1
(c) Thegraphof y = t
2
+t isanupwardopeningparabola; therefore, thisfunctionisneither increasingnor decreasing
for all t . By completingthesquarewendy = (t +
1
2
)
2
1
4
. Thevertex of this parabolais thenat t =
1
2
, so the
functionisdecreasingfor t <
1
2
andincreasingfor t >
1
2
.
(d) Fromthegraphof y = t
3
+t shownbelow, weseethat thisisanincreasingfunction.
20
20
y
x
1 1 2 3 2 3
Determinewhether thefunctioniseven, odd, or neither:
(a) f (x) = x
4
3x
2
(b) g(x) = sin(x +1)
(c) f (x) = 2
x
2
In Exercises 1722, nd the equation of the line.
17. Linepassingthrough(1, 4) and(2, 6)
solution Theslopeof thelinepassingthrough(1, 4) and(2, 6) is
m =
64
2(1)
=
2
3
.
Theequationof thelinepassingthrough(1, 4) and(2, 6) isthereforey 4=
2
3
(x +1) or 2x 3y = 14.
Linepassingthrough(1, 4) and(1, 6)
19. Lineof slope6through(9, 1)
solution Usingthepoint-slopeformfor theequationof aline, theequationof thelineof slope6andpassingthrough
(9, 1) isy 1= 6(x 9) or 6x y = 53.
Lineof slope
3
2
through(4, 12)
21. Linethrough(2, 3) parallel toy = 4x
solution Theequation y = 4 x is in slope-intercept form; it follows that theslopeof this lineis 1. Any line
parallel toy = 4 x will havethesameslope, sowearelookingfor theequationof thelineof slope1andpassing
through(2, 3). Theequationof thislineisy 3= (x 2) or x +y = 5.
Horizontal linethrough(3, 5)
23. Doesthefollowingtableof marketdatasuggestalinear relationshipbetweenpriceandnumber of homessoldduring
aone-year period? Explain.
Price(thousandsof $) 180 195 220 240
No. of homessold 127 118 103 91
solution Examinetheslopebetweenconsecutivedatapoints. Therst pair of datapointsyieldsaslopeof
118127
195180
=
9
15
=
3
5
,
whilethesecondpair of datapointsyieldsaslopeof
103118
220195
=
15
25
=
3
5
andthelast pair of datapointsyieldsaslopeof
91103
240220
=
12
20
=
3
5
.
Becauseall threeslopes areequal, thedatadoes suggest alinear relationshipbetweenpriceandthenumber of homes
sold.
Doesthefollowingtableof revenuedatafor acomputer manufacturer suggest alinear relationbetweenrevenue
andtime? Explain.
Year 2001 2005 2007 2010
Revenue(billionsof $) 13 18 15 11
25. Findtherootsof f (x) = x
4
4x
2
andsketchitsgraph. Onwhichintervalsisf (x) decreasing?
solution Theroots of f (x) = x
4
4x
2
areobtainedby solvingtheequationx
4
4x
2
= x
2
(x 2)(x + 2) = 0,
whichyieldsx = 2, x = 0andx = 2. Thegraphof y = f (x) isshownbelow. Fromthisgraphweseethat f (x) is
decreasingfor x lessthanapproximately1.4andfor x between0andapproximately1.4.
10
20
y
x
1 1
2 3 2 3
Let h(z) = 2z
2
+12z +3. Completethesquareandndtheminimumvalueof h(z).
27. Let f (x) bethesquareof thedistancefromthepoint (2, 1) toapoint (x, 3x +2) ontheliney = 3x +2. Showthat
f (x) isaquadraticfunction, andnditsminimumvaluebycompletingthesquare.
solution Letf (x) denotethesquareof thedistancefromthepoint(2, 1) toapoint(x, 3x +2) ontheliney = 3x +2.
Then
f (x) = (x 2)
2
+(3x +21)
2
= x
2
4x +4+9x
2
+6x +1= 10x
2
+2x +5,
whichisaquadraticfunction. Completingthesquare, wend
f (x) = 10
_
x
2
+
1
5
x +
1
100
_
+5
1
10
= 10
_
x +
1
10
_
2
+
49
10
.
Because(x +
1
10
)
2
0for all x, it followsthat f (x)
49
10
for all x. Hence, theminimumvalueof f (x) is
49
10
.
Provethat x
2
+3x +3 0for all x.
In Exercises 2934, sketch the graph by hand.
29. y = t
4
solution
x
1 0.5 1 0.5
y
0.2
0.4
0.6
0.8
1
y = t
5 31. y = sin
2
solution
y
x
5 5 10
0.5
1
0.5
1
y = 10
x
33. y = x
1/3
solution
x
1 2 3 4 1 2 3 4
y
1
2
1
2
y =
1
x
2
35. Showthatthegraphof y = f
_
1
3
x b
_
isobtainedbyshiftingthegraphof y = f
_
1
3
x
_
totheright3b units. Usethis
observationtosketchthegraphof y =

1
3
x 4
.
solution Let g(x) = f (
1
3
x). Then
g(x 3b) = f
_
1
3
(x 3b)
_
= f
_
1
3
x b
_
.
Thus, thegraphof y = f (
1
3
x b) isobtainedbyshiftingthegraphof y = f (
1
3
x) totheright 3b units.
Thegraphof y = |
1
3
x 4| isthegraphof y = |
1
3
x| shiftedright 12units(seethegraphbelow).
y
x
1
2
3
4
0 5 10 15 20
Leth(x) = cosx andg(x) = x
1
. Computethecompositefunctionsh(g(x)) andg(h(x)), andndtheirdomains.
37. Findfunctionsf andg suchthat thefunction
f (g(t )) = (12t +9)
4
solution Onepossiblechoiceisf (t ) = t
4
andg(t ) = 12t +9. Then
f (g(t )) = f (12t +9) = (12t +9)
4
asdesired.
Sketchthepoints ontheunit circlecorrespondingto thefollowingthreeangles, andndthevalues of thesix
standardtrigonometricfunctionsat eachangle:
(a)
2
3
(b)
7
4
(c)
7
6
39. What istheperiodof thefunctiong() = sin2 +sin
2
?
solution Thefunctionsin2 hasaperiodof , andthefunctionsin(/2) hasaperiodof 4. Because4 isamultiple
of , theperiodof thefunctiong() = sin2 +sin/2is4.
Assumethat sin =
4
5
, where/2< < . Find:
(a) tan (b) sin2 (c) csc
2
41. Giveanexampleof valuesa, b suchthat
(a) cos(a +b) = cosa +cosb (b) cos
a
2
=
cosa
2
solution
(a) Takea = b = /2. Thencos(a +b) = cos = 1but
cosa +cosb = cos
2
+cos
2
= 0+0= 0.
(b) Takea = . Then
cos
_
a
2
_
= cos
_
2
_
= 0
but
cosa
2
=
cos
2
=
1
2
=
1
2
.
Let f (x) = cosx. Sketchthegraphof y = 2f
_
1
3
x

4
_
for 0 x 6.
43. Solvesin2x +cosx = 0for 0 x < 2.
solution Using thedoubleangleformulafor thesinefunction, werewritetheequation as 2sinx cosx + cosx =
cosx(2sinx + 1) = 0. Thus, either cosx = 0 or sinx = 1/2. Fromhereweseethat thesolutions arex = /2,
x = 7/6, x = 3/2andx = 11/6.
Howdoesh(n) = n
2
/2
n
behavefor largewhole-number valuesof n? Doesh(n) tendtoinnity?
45. Useagraphingcalculator todeterminewhether theequationcosx = 5x
2
8x
4
hasanysolutions.
solution Thegraphsof y = cosx andy = 5x
2
8x
4
areshownbelow. Becausethegraphsdonot intersect, there
arenosolutionstotheequationcosx = 5x
2
8x
4
.
x
1 1
y
y = cos x
y = 5x
2
8x
4
1
1
Usingagraphingcalculator, ndthenumber of real rootsandestimatethelargest root totwodecimal places:
(a) f (x) = 1.8x
4
x
5
x
(b) g(x) = 1.7x
4
x
5
x
2 LIMITS
2.1 Limits, Rates of Change, and Tangent Lines
1. Averagevelocityisequal totheslopeof asecant linethroughtwopointsonagraph. Whichgraph?
solution Averagevelocity is theslopeof asecant linethroughtwopoints onthegraphof positionas afunctionof
time.
2. Caninstantaneousvelocitybedenedasaratio? If not, howisinstantaneousvelocitycomputed?
solution Instantaneous velocity cannot bedened as aratio. It is dened as thelimit of averagevelocity as time
elapsedshrinkstozero.
3. What isthegraphical interpretationof instantaneousvelocityat amoment t = t
0
?
solution Instantaneousvelocityat timet = t
0
istheslopeof thelinetangent tothegraphof positionasafunctionof
timeat t = t
0
.
4. What isthegraphical interpretationof thefollowingstatement?Theaveragerateof changeapproachestheinstanta-
neousrateof changeastheinterval [x
0
, x
1
] shrinkstox
0
.
solution Theslopeof thesecant lineover theinterval [x
0
, x
1
] approachestheslopeof thetangent lineat x = x
0
.
5. Therateof changeof atmospheric temperaturewithrespect toaltitudeisequal totheslopeof thetangent linetoa
graph. Whichgraph?What arepossibleunitsfor thisrate?
solution Therateof changeof atmospherictemperaturewithrespect toaltitudeistheslopeof thelinetangent tothe
graphof atmospherictemperatureasafunctionof altitude. Possibleunitsfor thisrateof changeare
F/ft or
C/m.
Exercises
1. A ball droppedfromastateof rest at timet = 0travelsadistances(t ) = 4.9t
2
mint seconds.
(a) Howfar doestheball travel duringthetimeinterval [2, 2.5]?
(b) Computetheaveragevelocityover [2, 2.5].
(c) Computetheaveragevelocityforthetimeintervalsinthetableandestimatetheballsinstantaneousvelocityatt = 2.
Interval [2, 2.01] [2, 2.005] [2, 2.001] [2, 2.00001]
Average
velocity
solution
(a) Duringthetimeinterval [2, 2.5], theball travelss = s(2.5) s(2) = 4.9(2.5)
2
4.9(2)
2
= 11.025m.
(b) Theaveragevelocityover [2, 2.5] is
s
t
=
s(2.5) s(2)
2.52
=
11.025
0.5
= 22.05m/s.
(c)
timeinterval [2, 2.01] [2, 2.005] [2, 2.001] [2, 2.00001]
averagevelocity 19.649 19.6245 19.6049 19.600049
Theinstantaneousvelocityat t = 2is19.6m/s.
A wrenchreleasedfromastateof rest at timet = 0travelsadistances(t ) = 4.9t
2
mint seconds. Estimatethe
instantaneousvelocityat t = 3.
3. Let v = 20
T asinExample2. Estimatetheinstantaneousrateof changeof v withrespect toT whenT = 300K.

solution
T interval [300, 300.01] [300, 300.005]
averagerateof change 0.577345 0.577348
T interval [300, 300.001] [300, 300.00001]
averagerateof change 0.57735 0.57735
Theinstantaneousrateof changeisapproximately0.57735m/(s K).
31
32 C HA P T E R 2 LIMITS
Computey/x for theinterval [2, 5], wherey = 4x 9. What istheinstantaneousrateof changeof y with
respect tox at x = 2?
InExercises5and6, astoneistossedverticallyintotheair fromgroundlevel withaninitial velocityof 15m/s. Itsheight
at timet ish(t ) = 15t 4.9t
2
m.
5. Computethestonesaveragevelocityover thetimeinterval [0.5, 2.5] andindicatethecorrespondingsecant lineon
asketchof thegraphof h(t ).
solution Theaveragevelocityisequal to
h(2.5) h(0.5)
2
= 0.3.
Thesecant lineisplottedwithh(t ) below.
2
0.5 1 1.5 2 2.5 3
4
6
8
10
t
h
Compute the stones average velocity over the time intervals [1, 1.01], [1, 1.001], [1, 1.0001] and [0.99, 1],
[0.999, 1], [0.9999, 1], andthenestimatetheinstantaneousvelocityat t = 1.
7. Withaninitial deposit of $100, thebalanceinabankaccount after t yearsisf (t ) = 100(1.08)
t
dollars.
(a) What aretheunitsof therateof changeof f (t )?
(b) Findtheaveragerateof changeover [0, 0.5] and[0, 1].
(c) Estimatetheinstantaneousrateof changeat t = 0.5by computingtheaveragerateof changeover intervalstothe
left andright of t = 0.5.
solution
(a) Theunitsof therateof changeof f (t ) aredollars/year or $/yr.
(b) Theaveragerateof changeof f (t ) = 100(1.08)
t
over thetimeinterval [t
1
, t
2
] isgivenby
f
t
=
f (t
2
) f (t
1
)
t
2
t
1
.
timeinterval [0, .5] [0, 1]
averagerateof change 7.8461 8
(c)
timeinterval [0.5, 0.51] [0.5, 0.501] [0.5, 0.5001]
averagerateof change 8.0011 7.9983 7.9981
timeinterval [0.49, 0.5] [0.499, 0.5] [0.4999, 0.5]
averagerateof change 7.9949 7.9977 7.998
Therateof changeat t = 0.5isapproximately$8/yr.
Thepositionof aparticleat timet is s(t ) = t
3
+ t . Computetheaveragevelocity over thetimeinterval [1, 4]
andestimatetheinstantaneousvelocityat t = 1.
9. Figure8shows theestimatednumber N of Internet users inChile, basedondatafromtheUnitedNations
StatisticsDivision.
(a) Estimatetherateof changeof N at t = 2003.5.
(b) Doestherateof changeincreaseor decreaseast increases? Explaingraphically.
(c) Let R betheaveragerateof changeover [2001, 2005]. ComputeR.
(d) Istherateof changeat t = 2002greater thanor lessthantheaveragerateR? Explaingraphically.
2001 2002 2003 2004 2005
3.5
4.0
4.5
N (Internet users in Chilein millions)
t (years)
FIGURE 8
S E C T I ON 2.1 Limits, Rates of Change, and Tangent Lines 33
solution
(a) Thetangent lineshowninFigure8appearstopassthroughthepoints(2002, 3.75) and(2005, 4.6). Thus, therateof
changeof N at t = 2003.5isapproximately
4.63.75
20052002
= 0.283
millionInternet usersper year.
(b) Ast increases, wemovefromlefttorightalongthegraphinFigure8. Moreover, aswemovefromlefttorightalong
thegraph, theslopeof thetangent linedecreases. Thus, therateof changedecreasesast increases.
(c) Thegraphof N(t ) appear topass throughthepoints (2001, 3.1) and(2005, 4.5). Thus, theaveragerateof change
over [2001, 2005] isapproximately
R =
4.53.1
20052001
= 0.35
millionInternet usersper year.
(d) For thegurebelow, weseethat theslopeof thetangent lineat t = 2002islarger thantheslopeof thesecant line
through theendpoints of thegraph of N(t ). Thus, therateof changeat t = 2002 is greater than theaveragerateof
changeR.
3.0
2001 2002 2003 2004 2005
x
y
3.5
4.0
4.5
TheatmospherictemperatureT (in
C) ataltitudeh metersaboveacertainpointonearthisT = 150.0065h

for h 12,000m. What aretheaverageandinstantaneousratesof changeof T withrespect toh?Whyaretheythe
same? Sketchthegraphof T for h 12,000.
InExercises1118, estimatetheinstantaneousrateof changeat thepoint indicated.
11. P(x) = 3x
2
5; x = 2
solution
x interval [2, 2.01] [2, 2.001] [2, 2.0001] [1.99, 2] [1.999, 2] [1.9999, 2]
averagerateof change 12.03 12.003 12.0003 11.97 11.997 11.9997
Therateof changeat x = 2isapproximately12.
f (t ) = 12t 7; t = 4 13. y(x) =
1
x +2
; x = 2
solution
x interval [2, 2.01] [2, 2.001] [2, 2.0001] [1.99, 2] [1.999, 2] [1.9999, 2]
Therateof changeat x = 2isapproximately0.06.
y(t ) =

3t +1; t = 1
15. f (x) = 3
x
; x = 0
solution
x interval [0.01, 0] [0.001, 0] [0.0001, 0] [0, 0.01] [0, 0.001] [0, 0.0001]
Therateof changeisbetwenn1.0986and1.0987.
f (x) = 3
x
; x = 3
17. f (x) = sinx; x =

6
solution
x interval
_
6
0.01,

6
_ _
6
0.001,

6
_ _
6
0.0001,

6
_ _
6
,

6
+0.01
_ _
6
,

6
+0.001
_ _
6
,

6
+0.001
_
Therateof changeat x =

6
isapproximately0.866.
f (x) = tanx; x =

4
19. Theheight(incentimeters)attimet (inseconds)of asmall massoscillatingattheendof aspringish(t ) = 8cos(12t ).
(a) Calculatethemasssaveragevelocityover thetimeintervals[0, 0.1] and[3, 3.5].
(b) Estimateitsinstantaneousvelocityat t = 3.
solution
(a) Theaveragevelocityover thetimeinterval [t
1
, t
2
] isgivenby
h
t
=
h(t
2
) h(t
1
)
t
2
t
1
.
timeinterval [0, 0.1] [3, 3.5]
averagevelocity 144.721cm/s 0cm/s
(b)
timeinterval [3, 3.0001] [3, 3.00001] [3, 3.000001] [2.9999, 3] [2.99999, 3] [2.999999, 3]
averagevelocity 0.5685 0.05685 0.005685 0.5685 0.05685 0.005685
Theinstantaneousvelocityat t = 3secondsisapproximately0cm/s.
Thenumber P(t ) of E. coli cellsat timet (hours) inapetri dishisplottedinFigure9.
(a) Calculatetheaveragerateof changeof P(t ) over thetimeinterval [1, 3] anddrawthecorrespondingsecantline.
(b) Estimatetheslopem of thelineinFigure9. What doesm represent?
21. AssumethattheperiodT (inseconds) of apendulum(thetimerequiredfor acompleteback-and-forthcycle)
isT =
3
2
L, whereL isthependulumslength(inmeters).
(a) What aretheunitsfor therateof changeof T withrespect toL? Explainwhat thisratemeasures.
(b) Whichquantitiesarerepresentedbytheslopesof linesA andB inFigure10?
(c) Estimatetheinstantaneousrateof changeof T withrespect toL whenL = 3m.
Period (s)
Length (m)
1 3
A
B
2
FIGURE 10 TheperiodT isthetimerequiredfor apendulumtoswingbackandforth.
solution
(a) Theunitsfor therateof changeof T withrespect toL aresecondsper meter. Thisratemeasuresthesensitivity of
theperiodof thependulumtoachangeinthelengthof thependulum.
(b) Theslopeof thelineB representstheaveragerateof changeinT fromL = 1mtoL = 3m. Theslopeof theline
A representstheinstantaneousrateof changeof T at L = 3m.
(c)
timeinterval [3, 3.01] [3, 3.001] [3, 3.0001] [2.99, 3] [2.999, 3] [2.9999, 3]
averagevelocity 0.4327 0.4330 0.4330 0.4334 0.4330 0.4330
Theinstantaneousrateof changeat L = 1misapproximately0.4330s/m.
ThegraphsinFigure11represent thepositionsof movingparticlesasfunctionsof time.
(a) Dotheinstantaneousvelocitiesat timest
1
, t
2
, t
3
in(A) formanincreasingor adecreasingsequence?
(b) Istheparticlespeedingupor slowingdownin(A)?
(c) Istheparticlespeedingupor slowingdownin(B)?
23. Anadvertisingcampaignboostedsalesof CrunchyCrustfrozenpizzatoapeaklevel of S
0
dollarsper month.
A marketingstudyshowedthat after t months, monthlysalesdeclinedto
S(t ) = S
0
g(t ), whereg(t ) =
1
1+t
.
Do sales declinemoreslowly or morerapidly as timeincreases? Answer by referringto asketchof thegraphof g(t )
together withseveral tangent lines.
solution Wenoticefromthegurebelowthat, astimeincreases, theslopesof thetangent linestothegraphof g(t )
becomelessnegative. Thus, salesdeclinemoreslowlyastimeincreases.
2
0.2
0.4
0.6
0.8
1.0
y
x
4 6 8 10 12
S E C T I ON 2.1 Limits, Rates of Change, and Tangent Lines 35
Thefractionof acityspopulationinfectedbyauvirusisplottedasafunctionof time(inweeks) inFigure12.
(a) Whichquantitiesarerepresentedbytheslopesof linesA andB? Estimatetheseslopes.
(b) Istheuspreadingmorerapidlyat t = 1, 2, or 3?
(c) Istheuspreadingmorerapidlyat t = 4, 5, or 6?
25. ThegraphsinFigure13represent thepositionss of movingparticlesasfunctionsof timet . Matcheachgraphwith
adescription:
(a) Speedingup
(b) Speedingupandthenslowingdown
(c) Slowingdown
(d) Slowingdownandthenspeedingup
(B) (A) (D) (C)
t
s
t
s
t
s
t
s
FIGURE 13
solution When aparticleis speeding up over atimeinterval, its graph is bent upward over that interval. When a
particleisslowingdown, itsgraphisbent downwardover that interval. Accordingly,
Ingraph(A), theparticleis(c) slowingdown.
Ingraph(B), theparticleis(b) speedingupandthenslowingdown.
Ingraph(C), theparticleis(d) slowingdownandthenspeedingup.
Ingraph(D), theparticleis(a) speedingup.
Anepidemiologist ndsthat thepercentageN(t ) of susceptiblechildrenwhowereinfectedonday t duringthe
rst threeweeksof ameaslesoutbreakisgiven, toareasonableapproximation, bytheformula(Figure14)
N(t ) =
100t
2
t
3
+5t
2
100t +380
(a) Drawthesecant linewhoseslopeistheaveragerateof changeininfectedchildrenover theintervals[4, 6] and
[12, 14]. Thencomputetheseaveragerates(inunitsof percent per day).
(b) Istherateof declinegreater at t = 8or t = 16?
(c) Estimatetherateof changeof N(t ) onday12.
27. ThefungusFusariumexosporiuminfectsaeldof axplantsthroughtherootsandcausestheplantstowilt.Eventually,
theentireeldisinfected. Thepercentagef (t ) of infectedplantsasafunctionof timet (indays) sinceplantingisshown
inFigure15.
(a) What aretheunitsof therateof changeof f (t ) withrespect tot ?What doesthisratemeasure?
(b) Usethegraphto rank (fromsmallest to largest) theaverageinfectionrates over theintervals [0, 12], [20, 32], and
[40, 52].
(c) Usethefollowingtabletocomputetheaverageratesof infectionover theintervals[30, 40], [40, 50], [30, 50].
Days 0 10 20 30 40 50 60
Percent infected 0 18 56 82 91 96 98
(d) Drawthetangent lineat t = 40andestimateitsslope.
Percent infected
Days after planting
10 20 30 40 50 60
100
80
60
40
20
FIGURE 15
solution
(a) Theunitsof therateof changeof f (t ) withrespecttot arepercent/dayor %/d. Thisratemeasureshowquicklythe
populationof axplantsisbecominginfected.
(b) Fromsmallest tolargest, theaverageratesof infectionarethoseover theintervals[40, 52], [0, 12], [20, 32]. Thisis
becausetheslopesof thesecant linesover theseintervalsarearrangedfromsmallest tolargest.
(c) Theaverageratesof infectionover theintervals[30, 40], [40, 50], [30, 50] are0.9, 0.5, 0.7%/d, respectively.
(d) Thetangent linesketchedinthegraphbelowappearstopassthroughthepoints(20, 80) and(40, 91). Theestimate
of theinstantaneousrateof infectionat t = 40daysistherefore
9180
4020
=
11
20
= 0.55%/d.
10 20 30 40 50 60
100
80
60
40
20
Letv = 20
T asinExample2. Istherateof changeof v withrespecttoT greater atlowtemperaturesor high

temperatures? Explainintermsof thegraph.
29. If anobject inlinear motion(but withchangingvelocity) coverss metersint seconds, thenitsaverage
velocity is v
0
= s/t m/s. Showthat it wouldcover thesamedistanceif it traveledat constant velocity v
0
over the
sametimeinterval. Thisjustiesour callings/t theaveragevelocity.
solution At constant velocity, thedistancetraveledisequal tovelocity timestime, soanobject movingat constant
velocityv
0
for t secondstravelsv
0
t meters. Sincev
0
= s/t , wend
distancetraveled= v
0
t =
_
s
t
_
t = s
Sotheobject coversthesamedistances bytravelingat constant velocityv
0
.
Sketchthegraphof f (x) = x(1 x) over [0, 1]. Refer tothegraphand, without makingany computations,
nd:
(a) Theaveragerateof changeover [0, 1]
(b) The(instantaneous) rateof changeat x =
1
2
(c) Thevaluesof x at whichtherateof changeispositive
31. WhichgraphinFigure16has thefollowingproperty: For all x, theaveragerateof changeover [0, x] is
greater thantheinstantaneousrateof changeat x. Explain.
(B)
x
y
(A)
x
y
FIGURE 16
solution
(a) Theaveragerateof changeover [0, x] isgreater thantheinstantaneousrateof changeat x: (B).
(b) Theaveragerateof changeover [0, x] islessthantheinstantaneousrateof changeat x: (A)
Thegraphin(B) bends downward, so theslopeof thesecant linethrough(0, 0) and(x, f (x)) is larger thantheslope
of thetangent lineat (x, f (x)). Ontheother hand, thegraphin(A) bends upward, so theslopeof thetangent lineat
(x, f (x)) islarger thantheslopeof thesecant linethrough(0, 0) and(x, f (x)).
Theheight of aprojectileredintheair verticallywithinitial velocity25m/sis
h(t ) = 25t 4.9t
2
m.
(a) Computeh(1). Showthat h(t ) h(1) canbefactoredwith(t 1) asafactor.
(b) Usingpart (a), showthat theaveragevelocityover theinterval [1, t ] is20.14.9t .
(c) Usethis formulato ndtheaveragevelocity over several intervals [1, t ] witht closeto 1. Thenestimatethe
instantaneousvelocityat timet = 1.
33. Let Q(t ) = t
2
. Asinthepreviousexercise, ndaformulafor theaveragerateof changeof Qover theinterval [1, t ]
anduseit to estimatetheinstantaneous rateof changeat t = 1. Repeat for theinterval [2, t ] andestimatetherateof
changeat t = 2.
solution Theaveragerateof changeis
Q(t ) Q(1)
t 1
=
t
2
1
t 1
.
Applyingthedifferenceof squaresformulagivesthattheaveragerateof changeis((t +1)(t 1))/(t 1) = (t +1) for
t = 1. Ast getscloser to1, thisgetscloser to1+1= 2. Theinstantaneousrateof changeis2.
For t
0
= 2, theaveragerateof changeis
Q(t ) Q(2)
t 2
=
t
2
4
t 2
,
whichsimpliestot + 2for t = 2. Ast approaches2, theaveragerateof changeapproaches4. Theinstantaneousrate
of changeistherefore4.
Showthat theaveragerateof changeof f (x) = x
3
over [1, x] isequal to
x
2
+x +1.
Usethistoestimatetheinstantaneousrateof changeof f (x) at x = 1.
S E C T I ON 2.2 Limits: A Numerical and Graphical Approach 37
35. Findaformulafor theaveragerateof changeof f (x) = x
3
over [2, x] anduseit toestimatetheinstantaneousrate
of changeat x = 2.
solution Theaveragerateof changeis
f (x) f (2)
x 2
=
x
3
8
x 2
.
Applyingthedifferenceof cubesformulatothenumerator, wendthat theaveragerateof changeis
(x
2
+2x +4)(x 2)
x 2
= x
2
+2x +4
for x = 2. Thecloser x getsto2, thecloser theaveragerateof changegetsto2
2
+2(2) +4= 12.
LetT =
3
2
L asinExercise21. Thenumbersinthesecondcolumnof Table4areincreasing, andthoseinthe

lastcolumnaredecreasing. Explainwhyintermsof thegraphof T asafunctionof L. Also, explaingraphicallywhy
theinstantaneousrateof changeat L = 3liesbetween0.4329and0.4331.
2.2 Limits: A Numerical and Graphical Approach
1. What isthelimit of f (x) = 1asx ?
solution lim
x
1= 1.
2. What isthelimit of g(t ) = t ast ?
solution lim
t
t = .
3. Is lim
x10
20equal to10or 20?
solution lim
x10
20= 20.
4. Canf (x) approachalimit asx c if f (c) isundened? If so, giveanexample.
solution Yes. Thelimitof afunctionf asx c doesnotdependonwhathappensatx = c, onlyonthebehavior of
f asx c. Asanexample, consider thefunction
f (x) =
x
2
1
x 1
.
Thefunctionisclearlynot denedat x = 1but
lim
x1
f (x) = lim
x1
x
2
1
x 1
= lim
x1
(x +1) = 2.
5. What doesthefollowingtablesuggest about lim
x1
f (x) and lim
x1+
f (x)?
x 0.9 0.99 0.999 1.1 1.01 1.001
f (x) 7 25 4317 3.0126 3.0047 3.00011
solution Thevaluesinthetablesuggest that lim
x1
f (x) = andlim
x1+
f (x) = 3.
6. Canyoutell whether lim
x5
f (x) existsfromaplot of f (x) for x > 5? Explain.
solution No. Byexaminingvaluesof f (x) for x closetobut greater than5, wecandeterminewhether theone-sided
limit lim
x5+
f (x) exists. Todeterminewhether lim
x5
f (x) exists, wemust examinevalueof f (x) onbothsidesof
x = 5.
7. If youknowinadvancethat lim
x5
f (x) exists, canyoudetermineitsvaluefromaplot of f (x) for all x > 5?
solution Yes. If lim
x5
f (x) exists, thenbothone-sidedlimitsmust exist andbeequal.
Exercises
InExercises14, ll inthetablesandguessthevalueof thelimit.
1. lim
x1
f (x), wheref (x) =
x
3
1
x
2
1
.
x f (x) x f (x)
1.002 0.998
1.001 0.999
1.0005 0.9995
1.00001 0.99999
solution
x 0.998 0.999 0.9995 0.99999 1.00001 1.0005 1.001 1.002
f (x) 1.498501 1.499250 1.499625 1.499993 1.500008 1.500375 1.500750 1.501500
Thelimit asx 1is
3
2
.
lim
t 0
h(t ), whereh(t ) =
cost 1
t
2
. Notethat h(t ) iseven; that is, h(t ) = h(t ).
t 0.002 0.0001 0.00005 0.00001
h(t )
3. lim
y2
f (y), wheref (y) =
y
2
y 2
y
2
+y 6
.
y f (y) y f (y)
2.002 1.998
2.001 1.999
2.0001 1.9999
solution
y 1.998 1.999 1.9999 2.0001 2.001 2.02
f (y) 0.59984 0.59992 0.599992 0.600008 0.60008 0.601594
Thelimit asy 2is
3
5
.
lim
0
f (), wheref () =
sin
3
.
0.002 0.0001 0.00005 0.00001
f ()
5. Determine lim
x0.5
f (x) for f (x) asinFigure9.
0.5
1.5
x
y
1
f(x)
FIGURE 9
solution Thegraphsuggeststhat f (x) 1.5asx 0.5.
Determine lim
x0.5
g(x) for g(x) asinFigure10.
InExercises7and8, evaluatethelimit.
7. lim
x21
x
solution Asx 21, f (x) = x 21. Youcanseethis, for example, onthegraphof f (x) = x.
lim
x4.2
3
InExercises 916, verifyeachlimit usingthelimit denition. For example, inExercise9, showthat |3x 12| canbe
madeassmall asdesiredbytakingx closeto4.
9. lim
x4
3x = 12
solution |3x 12| = 3|x 4|. |3x 12| canbemadearbitrarilysmall bymakingx closeenoughto4, thusmaking
|x 4| small.
lim
x5
3= 3
11. lim
x3
(5x +2) = 17
solution |(5x + 2) 17| = |5x 15| = 5|x 3|. Therefore, if you make|x 3| small enough, you can make
|(5x +2) 17| assmall asdesired.
lim
x2
(7x 4) = 10
13. lim
x0
x
2
= 0
solution Asx 0, wehave|x
2
0| = |x +0||x 0|. Tosimplifythings, supposethat |x| < 1, sothat |x +0||x
0| = |x||x| < |x|. Bymaking|x| sufcientlysmall, sothat |x +0||x 0| = x
2
isevensmaller, youcanmake|x
2
0|
assmall asdesired.
lim
x0
(3x
2
9) = 9
15. lim
x0
(4x
2
+2x +5) = 5
solution Asx 0, wehave|4x
2
+2x +55| = |4x
2
+2x| = |x||4x +2|. If |x| < 1, |4x +2| canbenobigger
than6, so|x||4x +2| < 6|x|. Therefore, bymaking|x 0| = |x| sufcientlysmall, youcanmake|4x
2
+2x +55| =
|x||4x +2| assmall asdesired.
lim
x0
(x
3
+12) = 12
In Exercises 1736, estimate the limit numerically or state that the limit does not exist. If innite, state whether the
one-sidedlimitsareor .
17. lim
x1
x 1
x 1
solution
x 0.9995 0.99999 1.00001 1.0005
f (x) 0.500063 0.500001 0.49999 0.499938
Thelimit asx 1is
1
2
.
lim
x4
2x
2
32
x +4
19. lim
x2
x
2
+x 6
x
2
x 2
solution
x 1.999 1.99999 2.00001 2.001
f (x) 1.666889 1.666669 1.666664 1.666445
Thelimit asx 2is
5
3
.
lim
x3
x
3
2x
2
9
x
2
2x 3
21. lim
x0
sin2x
x
solution
x 0.01 0.005 0.005 0.01
f (x) 1.999867 1.999967 1.999967 1.999867
Thelimit asx 0is2.
lim
x0
sin5x
x
23. lim
0
cos 1
solution
x 0.05 0.001 0.001 0.05
f (x) 0.0249948 0.0005 0.0005 0.0249948
Thelimit asx 0is0.
lim
x0
sinx
x
2
25. lim
x4
1
(x 4)
3
solution
x 3.99 3.999 3.9999 4.0001 4.001 4.01
f (x) 10
6
10
9
10
12
10
12
10
9
10
6
Thelimit doesnot exist. Asx 4, f (x) ; similarly, asx 4+, f (x) .
lim
x1
3x
x 1
27. lim
x3+
x 4
x
2
9
solution
x 3.01 3.001 3.0001 3.00001
f (x) 16.473 166.473 1666.473 16666.473
Asx 3+, f (x) .
lim
h0
3
h
1
h
29. lim
h0
sinhcos
1
h
solution
h 0.01 0.001 0.0001 0.0001 0.001 0.01
f (h) 0.008623 0.000562 0.000095 0.000095 0.000562 0.008623
Thelimit asx 0is0.
lim
h0
cos
1
h
31. lim
x0
|x|
x
solution
x 0.05 0.001 0.00001 0.00001 0.001 0.05
f (x) 1.161586 1.006932 1.000115 0.999885 0.993116 0.860892
Thelimit asx 0is1.
lim
x0
2
x
3
x
x
33. lim
4
tan 2sin cos

4
solution

4
0.01

4
0.001

4
0.0001

4
+0.0001

4
+0.001

4
+0.01
f () 1.96026 1.99600 1.99960 2.00040 2.00400 2.04027
Thelimit asx

4
isapproximately2.
lim
r0
(1+r)
1/r 35. lim
0
1cos
2
solution
0.01 0.001 0.0001 0.0001 0.001 0.01
f () 0.499996 0.500000 0.500000 0.500000 0.500000 0.499996
Thelimit as 0appearstobe0.5.
lim
0
1cos
3
37. Thegreatestinteger functionisdenedby[x] = n, wheren istheuniqueinteger suchthat n x < n +1. Sketch
thegraphof y = [x]. Calculate, for c aninteger:
(a) lim
xc
[x] (b) lim
xc+
[x]
solution Hereisagraphof thegreatest integer function:
2
1
1 2 3 1
x
y
(a) Fromthegraph, weseethat, for c aninteger,
lim
xc
[x] = c 1.
(b) Fromthegraph, weseethat, for c aninteger,
lim
xc+
[x] = c.
Determinetheone-sidedlimitsat c = 1, 2, and4of thefunctiong(x) showninFigure11, andstatewhether the
limit existsat thesepoints.
InExercises3946, determinetheone-sidedlimitsnumericallyor graphically. Ifinnite, statewhether theone-sidedlimits
are or , anddescribethecorrespondingvertical asymptote. InExercise46, [x] is thegreatest integer function
denedinExercise37.
39. lim
x0
sinx
|x|
solution
x 0.2 0.02 0.02 0.2
f (x) 0.993347 0.999933 0.999933 0.993347
Theleft-handlimit is lim
x0
f (x) = 1, whereastheright-handlimit is lim
x0+
f (x) = 1.
lim
x0
|x|
1/x 41. lim
x0
x sin|x|
x
3
solution
x 0.1 0.01 0.01 0.1
f (x) 199.853 19999.8 0.166666 0.166583
Theleft-handlimitis lim
x0
f (x) = , whereastheright-handlimitis lim
x0+
f (x) =
1
6
. Thus, thelinex = 0isavertical
asymptotefromtheleft for thegraphof y =
xsin|x|
x
3
.
lim
x4
x +1
x 4
43. lim
x2
4x
2
+7
x
3
+8
solution Thegraphof y =
4x
2
+7
x
3
+8
for x near 2isshownbelow. Fromthisgraph, weseethat
lim
x2
4x
2
+7
x
3
+8
= while lim
x2+
4x
2
+7
x
3
+8
= .
Thus, thelinex = 2isavertical asymptotefor thegraphof y =
4x
2
+7
x
3
+8
.
3.0 2.5 2.0 1.5 1.0
x
lim
x3
x
2
x
2
9
45. lim
x1
x
5
+x 2
x
2
+x 2
solution Thegraphof y =
x
5
+x2
x
2
+x2
for x near 1isshownbelow. Fromthisgraph, weseethat
lim
x1
x
5
+x 2
x
2
+x 2
= 2.
2
0.5 1.0 1.5
x
y
4
6
lim
x2
cos
_
2
(x [x])
_ 47. Determinetheone-sidedlimits at c = 2, 4of thefunctionf (x) inFigure12. What arethevertical asymptotes of
f (x)?
5
4 2
15
5
10
x
y
FIGURE 12
solution
For c = 2, wehave lim
x2
f (x) = and lim
x2+
f (x) = .
For c = 4, wehave lim
x4
f (x) = and lim
x4+
f (x) = 10.
Thevertical asymptotesarethevertical linesx = 2andx = 4.
Determinetheinniteone- andtwo-sidedlimitsinFigure13.
InExercises4952, sketchthegraphof afunctionwiththegivenlimits.
49. lim
x1
f (x) = 2, lim
x3
f (x) = 0, lim
x3+
f (x) = 4
solution
2
4
6
1 2 3 4
y
x
lim
x1
f (x) = , lim
x3
f (x) = 0, lim
x3+
f (x) =
51. lim
x2+
f (x) = f (2) = 3, lim
x2
f (x) = 1, lim
x4
f (x) = 2= f (4)
solution
1
1
2
3
1 2 3 4 5
y
x
lim
x1+
f (x) = , lim
x1
f (x) = 3, lim
x4
f (x) =
53. Determinetheone-sidedlimitsof thefunctionf (x) inFigure14, at thepointsc = 1, 3, 5, 6.
1
2
3
4
1
2
3
4
5
y
x
1 2 3 4 5 6 7 8
FIGURE 14 Graphof f (x)
solution
lim
x1
f (x) = lim
x1+
f (x) = 3
lim
x3
f (x) =
lim
x3+
f (x) = 4
lim
x5
f (x) = 2
lim
x5+
f (x) = 3
lim
x6
f (x) = lim
x6+
f (x) =
Doeseither of thetwooscillatingfunctionsinFigure15appear toapproachalimit asx 0?
InExercises5560, plot thefunctionandusethegraphtoestimatethevalueof thelimit.
55. lim
0
sin5
sin2
solution
2.42
2.44
2.46
2.48
2.50
y
Fromthegraphof y =
sin5
sin2
shownabove, weseethat thelimit as 0is
5
2
.
lim
x0
12
x
1
4
x
1
57. lim
x0
2
x
cosx
x
solution
0.6935
0.6940
0.6930
0.6925
0.6920
y
y =
2
x
cosx
x
Thelimit asx 0isapproximately0.693. (Theexact answer isln2.)
lim
0
sin
2
4
cos 1
59. lim
0
cos7 cos5
2
solution
12.0
11.8
11.6
11.4
y
Fromthegraphof y =
cos7 cos5
2
shownabove, weseethat thelimit as 0is12.
lim
0
sin
2
2 sin4
4
61. Let n beapositiveinteger. For whichn arethetwoinniteone-sidedlimits lim
x0
1/x
n
equal?
solution First, supposethat n iseven. Thenx
n
0for all x, and
1
x
n
> 0for all x = 0. Hence,
lim
x0
1
x
n
= lim
x0+
1
x
n
= .
Next, supposethat n isodd. Then
1
x
n
> 0for all x > 0but
1
x
n
< 0for all x < 0. Thus,
lim
x0
1
x
n
= but lim
x0+
1
x
n
= .
Finally, thetwoinniteone-sidedlimitsareequal whenever n iseven.
Let L(n) = lim
x1
_
n
1x
n

1
1x
_
for n apositiveinteger. InvestigateL(n) numericallyfor several valuesof
n, andthenguessthevalueof of L(n) ingeneral.
63. Insomecases, numerical investigationscanbemisleading. Plot f (x) = cos
x
.
(a) Does lim
x0
f (x) exist?
(b) Show, byevaluatingf (x) at x =
1
2
,
1
4
,
1
6
, . . . , that youmight beabletotrick your friendsintobelievingthat
thelimit existsandisequal toL = 1.
(c) Whichsequenceof evaluationsmight trickthemintobelievingthat thelimit isL = 1.
solution Hereisthegraphof f (x).
0.05
0.5
0.5
y
x
0.05
(a) Fromthegraphof f (x), whichshowsthat thevalueof f (x) oscillatesmoreandmorerapidlyasx 0, it follows
that lim
x0
f (x) doesnot exist.
(b) Noticethat
f
_
1
2
_
= cos

1/2
= cos2 = 1;
f
_
1
4
_
= cos

1/4
= cos4 = 1;
f
_
1
6
_
= cos

1/6
= cos6 = 1;
and, ingeneral, f (
1
2n
) = 1for all integersn.
(c) At x = 1,
1
3
,
1
5
, . . ., thevalueof f (x) isalways1.
Light wavesof frequency passingthroughaslit of widtha produceaFraunhofer diffractionpatternof light
anddarkfringes(Figure16). Theintensityasafunctionof theangle is
I () = I
m
_
sin(Rsin)
Rsin
_
2
whereR = a/ andI
m
isaconstant. Showthattheintensityfunctionisnotdenedat = 0. Thenchooseanytwo
valuesfor R andchecknumericallythat I () approachesI
m
as 0.
65. Investigate lim
0
sinn
numericallyfor several valuesof n. Thenguessthevalueingeneral.

solution
For n = 3, wehave
0.1 0.01 0.001 0.001 0.01 0.1
sinn
2.955202 2.999550 2.999996 2.999996 2.999550 2.955202

Thelimit as 0is3.
For n = 5, wehave
0.1 0.01 0.001 0.001 0.01 0.1
sinn
4.794255 4.997917 4.999979 4.999979 4.997917 4.794255

Thelimit as 0is5.
Wesurmisethat, ingeneral, lim
0
sinn
= n.
Shownumerically that lim
x0
b
x
1
x
for b = 3, 5appearstoequal ln3, ln5, wherelnx isthenatural logarithm.
Thenmakeaconjecture(guess) for thevalueingeneral andtest your conjecturefor twoadditional valuesof b.
67. Investigate lim
x1
x
n
1
x
m
1
for (m, n) equal to (2, 1), (1, 2), (2, 3), and (3, 2). Then guess thevalueof thelimit in
general andcheckyour guessfor twoadditional pairs.
solution
x 0.99 0.9999 1.0001 1.01

x 1
x
2
1
0.502513 0.500025 0.499975 0.497512
Thelimit asx 1is
1
2
.
x 0.99 0.9999 1.0001 1.01
x
2
1
x 1
1.99 1.9999 2.0001 2.01
Thelimit asx 1is2.
x 0.99 0.9999 1.0001 1.01
x
2
1
x
3
1
0.670011 0.666700 0.666633 0.663344
Thelimit asx 1is
2
3
.
x 0.99 0.9999 1.0001 1.01
x
3
1
x
2
1
1.492513 1.499925 1.500075 1.507512
Thelimit asx 1is
3
2
.
For general m andn, wehave lim
x1
x
n
1
x
m
1
=
n
m
.
x 0.99 0.9999 1.0001 1.01

x 1
x
3
1
0.336689 0.333367 0.333300 0.330022
Thelimit asx 1is
1
3
.
x 0.99 0.9999 1.0001 1.01
x
3
1
x 1
2.9701 2.9997 3.0003 3.0301
Thelimit asx 1is3.
x 0.99 0.9999 1.0001 1.01
x
3
1
x
7
1
0.437200 0.428657 0.428486 0.420058
Thelimit asx 1is
3
7
0.428571.
Findbynumerical experimentationthepositiveintegersk suchthat lim
x0
sin(sin
2
x)
x
k
exists.
69. Plot thegraphof f (x) =
2
x
8
x 3
.
(a) ZoominonthegraphtoestimateL = lim
x3
f (x).
(b) Explainwhy
f (2.99999) L f (3.00001)
UsethistodetermineL tothreedecimal places.
solution
(a)
5.555
5.565
5.545
5.535
5.525
y
x = 3
y =
2
x
8
x 3
(b) It is clear that thegraphof f rises as wemoveto theright. Mathematically, wemay express this observationas:
whenever u < v, f (u) < f (v). Because
2.99999< 3= lim
x3
f (x) < 3.00001,
it followsthat
f (2.99999) < L = lim
x3
f (x) < f (3.00001).
Withf (2.99999) 5.54516andf (3.00001) 5.545195, theaboveinequality becomes 5.54516 < L < 5.545195;
hence, tothreedecimal places, L = 5.545.
Thefunctionf (x) =
2
1/x
2
1/x
2
1/x
+2
1/x
isdenedfor x = 0.
(a) Investigate lim
x0+
f (x) and lim
x0
f (x) numerically.
(b) Plot thegraphof f anddescribeitsbehavior near x = 0.
2.3 Basic Limit Laws
1. StatetheSumLawandQuotient Law.
solution Supposelim
xc
f (x) andlim
xc
g(x) bothexist. TheSumLawstatesthat
lim
xc
(f (x) +g(x)) = lim
xc
f (x) + lim
xc
g(x).
Providedlim
xc
g(x) = 0, theQuotient Lawstatesthat
lim
xc
f (x)
g(x)
=
lim
xc
f (x)
lim
xc
g(x)
.
2. Whichof thefollowingisaverbal versionof theProduct Law(assumingthelimitsexist)?
(a) Theproduct of twofunctionshasalimit.
(b) Thelimit of theproduct istheproduct of thelimits.
(c) Theproduct of alimit isaproduct of functions.
(d) A limit producesaproduct of functions.
solution Theverbal versionof theProduct Lawis(b): Thelimit of theproduct istheproduct of thelimits.
3. Whichstatement iscorrect?TheQuotient Lawdoesnot holdif:
(a) Thelimit of thedenominator iszero.
(b) Thelimit of thenumerator iszero.
solution Statements(a)iscorrect. TheQuotient Lawdoesnot holdif thelimit of thedenominator iszero.
Exercises
InExercises124, evaluatethelimit usingtheBasicLimit Lawsandthelimits lim
xc
x
p/q
= c
p/q
and lim
xc
k = k.
1. lim
x9
x
solution lim
x9
x = 9.
lim
x3
14
3. lim
x
1
2
x
4
solution lim
x
1
2
x
4
=
_
1
2
_
4
=
1
16
.
lim
z27
z
2/3
5. lim
t 2
t
1
solution lim
t 2
t
1
= 2
1
=
1
2
.
lim
x5
x
2
7. lim
x0.2
(3x +4)
solution UsingtheSumLawandtheConstant MultipleLaw:
lim
x0.2
(3x +4) = lim
x0.2
3x + lim
x0.2
4
= 3 lim
x0.2
x + lim
x0.2
4= 3(0.2) +4= 4.6.
S E C T I ON 2.3 Basic Limit Laws 47
lim
x
1
3
(3x
3
+2x
2
)
9. lim
x1
(3x
4
2x
3
+4x)
solution UsingtheSumLaw, theConstant MultipleLawandthePowersLaw:
lim
x1
(3x
4
2x
3
+4x) = lim
x1
3x
4
lim
x1
2x
3
+ lim
x1
4x
= 3 lim
x1
x
4
2 lim
x1
x
3
+4 lim
x1
x
= 3(1)
4
2(1)
3
+4(1) = 3+24= 1.
lim
x8
(3x
2/3
16x
1
)
11. lim
x2
(x +1)(3x
2
9)
solution UsingtheProduct Law, theSumLawandtheConstant MultipleLaw:
lim
x2
(x +1)
_
3x
2
9
_
=
_
lim
x2
x + lim
x2
1
__
lim
x2
3x
2
lim
x2
9
_
= (2+1)
_
3 lim
x2
x
2
9
_
= 3(3(2)
2
9) = 9.
lim
x
1
2
(4x +1)(6x 1) 13. lim
t 4
3t 14
t +1
solution UsingtheQuotient Law, theSumLawandtheConstant MultipleLaw:
lim
t 4
3t 14
t +1
=
lim
t 4
(3t 14)
lim
t 4
(t +1)
=
3lim
t 4
t lim
t 4
14
lim
t 4
t + lim
t 4
1
=
3 414
4+1
=
2
5
.
lim
z9
z
z 2
15. lim
y
1
4
(16y +1)(2y
1/2
+1)
solution UsingtheProduct Law, theSumLaw, theConstant MultipleLawandthePowersLaw:
lim
y
1
4
(16y +1)(2y
1/2
+1) =
_
lim
y
1
4
(16y +1)
__
lim
y
1
4
(2y
1/2
+1)
_
=
_
16 lim
y
1
4
y + lim
y
1
4
1
__
2 lim
y
1
4
y
1/2
+ lim
y
1
4
1
_
=
_
16
_
1
4
_
+1
__
2
_
1
2
_
+1
_
= 10.
lim
x2
x(x +1)(x +2) 17. lim
y4
1
6y +1
solution UsingtheQuotient Law, thePowersLaw, theSumLawandtheConstant MultipleLaw:
lim
y4
1
6y +1
=
1
lim
y4
6y +1
=
1
_
6 lim
y4
y +1
=
1
6(4) +1
=
1
5
.
lim
w7
w +2+1
w 31
19. lim
x1
x
x
3
+4x
solution UsingtheQuotient Law, theSumLaw, thePowersLawandtheConstant MultipleLaw:
lim
x1
x
x
3
+4x
=
lim
x1
x
lim
x1
x
3
+4 lim
x1
x
=
1
(1)
3
+4(1)
=
1
5
.
lim
t 1
t
2
+1
(t
3
+2)(t
4
+1)
21. lim
t 25
3
t
1
5
t
(t 20)
2
solution UsingtheQuotient Law, theSumLaw, theConstant MultipleLawandthePowersLaw:
lim
t 25
3
t
1
5
t
(t 20)
2
=
3
_
lim
t 25
t
1
5
lim
t 25
t
_
lim
t 25
t 20
_
2
=
3(5)
1
5
(25)
5
2
=
2
5
.
lim
y
1
3
(18y
2
4)
4
23. lim
t
3
2
(4t
2
+8t 5)
3/2
solution UsingthePowersLaw, theSumLawandtheConstant MultipleLaw:
lim
t
3
2
(4t
2
+8t 5)
3/2
=
_
4 lim
t
3
2
t
2
+8 lim
t
3
2
t 5
_
3/2
= (9+125)
3/2
= 64.
lim
t 7
(t +2)
1/2
(t +1)
2/3
25. UsetheQuotient Lawtoprovethat if lim
xc
f (x) existsandisnonzero, then
lim
xc
1
f (x)
=
1
lim
xc
f (x)
solution Since lim
xc
f (x) isnonzero, wecanapplytheQuotient Law:
lim
xc
_
1
f (x)
_
=
_
lim
xc
1
_
_
lim
xc
f (x)
_ =
1
lim
xc
f (x)
.
Assumingthat lim
x6
f (x) = 4, compute:
(a) lim
x6
f (x)
2
(b) lim
x6
1
f (x)
(c) lim
x6
x
_
f (x)
InExercises2730, evaluatethelimit assumingthat lim
x4
f (x) = 3and lim
x4
g(x) = 1.
27. lim
x4
f (x)g(x)
solution lim
x4
f (x)g(x) = lim
x4
f (x) lim
x4
g(x) = 3 1= 3.
lim
x4
(2f (x) +3g(x)) 29. lim
x4
g(x)
x
2
solution Since lim
x4
x
2
= 0, wemayapplytheQuotient Law, thenapplyingthePowersLaw:
lim
x4
g(x)
x
2
=
lim
x4
g(x)
lim
x4
x
2
=
1
_
lim
x4
x
_
2
=
1
16
.
lim
x4
f (x) +1
3g(x) 9
31. CantheQuotient Lawbeappliedtoevaluate lim
x0
sinx
x
? Explain.
solution ThelimitQuotientLawcannot beappliedtoevaluate lim
x0
sinx
x
since lim
x0
x = 0. Thisviolatesacondition
of theQuotient Law. Accordingly, therulecannot beemployed.
Showthat theProduct Lawcannot beusedtoevaluatethelimit lim
x/2
_
x

2
_
tanx.
33. Giveanexamplewhere lim
x0
(f (x) +g(x)) existsbut neither lim
x0
f (x) nor lim
x0
g(x) exists.
solution Let f (x) = 1/x and g(x) = 1/x. Then lim
x0
(f (x) +g(x)) = lim
x0
0= 0. However, lim
x0
f (x) =
lim
x0
1/x and lim
x0
g(x) = lim
x0
1/x donot exist.
S E C T I ON 2.4 Limits and Continuity 49
Show that if both lim
xc
f (x) g(x) and lim
xc
g(x) exist and lim
xc
g(x) = 0, then lim
xc
f (x) exists. Hint: Write
f (x) =
f (x) g(x)
g(x)
.
35. Supposethat lim
t 3
tg(t ) = 12. Showthat lim
t 3
g(t ) existsandequals4.
solution Wearegiventhat lim
t 3
tg(t ) = 12. Since lim
t 3
t = 3= 0, wemayapplytheQuotient Law:
lim
t 3
g(t ) = lim
t 3
tg(t )
t
=
lim
t 3
tg(t )
lim
t 3
t
=
12
3
= 4.
Provethat if lim
t 3
h(t )
t
= 5, then lim
t 3
h(t ) = 15.
37. Assumingthat lim
x0
f (x)
x
= 1, whichof thefollowingstatementsisnecessarilytrue?Why?
(a) f (0) = 0 (b) lim
x0
f (x) = 0
solution
(a) Given that lim
x0
f (x)
x
= 1, it is not necessarily true that f (0) = 0. A counterexample is provided by f (x) =
_
x, x = 0
5, x = 0
.
(b) Giventhat lim
x0
f (x)
x
= 1, it isnecessarilytruethat lim
x0
f (x) = 0. For notethat lim
x0
x = 0, whence
lim
x0
f (x) = lim
x0
x
f (x)
x
=
_
lim
x0
x
__
lim
x0
f (x)
x
_
= 0 1= 0.
Provethat if lim
xc
f (x) = L = 0and lim
xc
g(x) = 0, thenthelimit lim
xc
f (x)
g(x)
doesnot exist.
39. Supposethat lim
h0
g(h) = L.
(a) Explainwhy lim
h0
g(ah) = L for anyconstant a = 0.
(b) If weassumeinsteadthat lim
h1
g(h) = L, isit still necessarilytruethat lim
h1
g(ah) = L?
(c) Illustrate(a) and(b) withthefunctionf (x) = x
2
.
solution
(a) Ash 0, ah 0aswell; hence, if wemakethechangeof variablew = ah, then
lim
h0
g(ah) = lim
w0
g(w) = L.
(b) No. Ash 1, ah a, soweshouldnot expect lim
h1
g(ah) = lim
h1
g(h).
(c) Let g(x) = x
2
. Then
lim
h0
g(h) = 0 and lim
h0
g(ah) = lim
h0
(ah)
2
= 0.
Ontheother hand,
lim
h1
g(h) = 1 while lim
h1
g(ah) = lim
h1
(ah)
2
= a
2
,
whichisequal tothepreviouslimit if andonlyif a = 1.
Assumethat L(a) = lim
x0
a
x
1
x
existsfor all a > 0. Assumealsothat lim
x0
a
x
= 1.
(a) ProvethatL(ab) = L(a) +L(b) for a, b > 0. Hint: (ab)
x
1= a
x
(b
x
1) +(a
x
1). ThisshowsthatL(a)
behaves likealogarithm. Wewill seethat L(a) = lna inSection7.3.
(b) Verifynumericallythat L(12) = L(3) +L(4).
2.4 Limits and Continuity
1. Whichpropertyof f (x) = x
3
allowsustoconcludethat lim
x2
x
3
= 8?
solution Wecanconcludethat lim
x2
x
3
= 8becausethefunctionx
3
iscontinuousat x = 2.
2. What canbesaidabout f (3) if f iscontinuousand lim
x3
f (x) =
1
2
?
solution If f iscontinuousandlim
x3
f (x) =
1
2
, thenf (3) =
1
2
.
3. Supposethat f (x) < 0if x ispositiveandf (x) > 1if x isnegative. Canf becontinuousat x = 0?
solution Sincef (x) < 0whenxispositiveandf (x) > 1whenxisnegative, it followsthat
lim
x0+
f (x) 0 and lim
x0
f (x) 1.
Thus, lim
x0
f (x) doesnot exist, sof cannot becontinuousat x = 0.
4. Is it possible to determine f (7) if f (x) = 3 for all x < 7 and f is right-continuous at x = 7? What if f is
left-continuous?
solution No. To determinef (7), weneedto combineeither knowledgeof thevalues of f (x) for x < 7withleft-
continuityor knowledgeof thevaluesof f (x) for x > 7withright-continuity.
5. Arethefollowingtrueor false? If false, stateacorrect version.
(a) f (x) iscontinuousat x = a if theleft- andright-handlimitsof f (x) asx a exist andareequal.
(b) f (x) iscontinuousat x = a if theleft- andright-handlimitsof f (x) asx a exist andequal f (a).
(c) If theleft- andright-handlimitsof f (x) asx a exist, thenf hasaremovablediscontinuityat x = a.
(d) If f (x) andg(x) arecontinuousat x = a, thenf (x) +g(x) iscontinuousat x = a.
(e) If f (x) andg(x) arecontinuousat x = a, thenf (x)/g(x) iscontinuousat x = a.
solution
(a) False. Thecorrect statement isf (x) iscontinuousat x = a if theleft- andright-handlimitsof f (x) asx a exist
andequal f (a).
(b) True.
(c) False. Thecorrect statement isIf theleft- andright-handlimitsof f (x) asx a areequal but not equal tof (a),
thenf hasaremovablediscontinuityat x = a.
(d) True.
(e) False. ThecorrectstatementisIf f (x) andg(x) arecontinuousatx = a andg(a) = 0, thenf (x)/g(x) iscontinuous
at x = a.
Exercises
1. ReferringtoFigure14, statewhether f (x) isleft- or right-continuous(or neither) ateachpointof discontinuity. Does
f (x) haveanyremovablediscontinuities?
1 2 3 4 5 6
x
5
4
3
2
1
y
FIGURE 14 Graphof y = f (x)
solution
Thefunctionf isdiscontinuousat x = 1; it isright-continuousthere.
Thefunctionf isdiscontinuousat x = 3; it isneither left-continuousnor right-continuousthere.
Thefunctionf isdiscontinuousat x = 5; it isleft-continuousthere.
However, thesediscontinuitiesarenot removable.
Exercises24refer tothefunctiong(x) inFigure15.
1 2 3 4 5 6
x
5
4
3
2
1
y
FIGURE 15 Graphof y = g(x)
Statewhether g(x) isleft- or right-continuous(or neither) at eachof itspointsof discontinuity.
3. At whichpoint c doesg(x) havearemovablediscontinuity? Howshouldg(c) beredenedtomakeg continuousat
x = c?
solution Becauselim
x3
g(x) exists, thefunctiong has aremovablediscontinuity at x = 3. Assigningg(3) = 4
makesg continuousat x = 3.
Findthepointc
1
atwhichg(x) hasajumpdiscontinuitybutisleft-continuous. Howshouldg(c
1
) beredenedto
makeg right-continuousat x = c
1
?
5. InFigure16, determinetheone-sidedlimitsatthepointsof discontinuity. Whichdiscontinuityisremovableandhow
shouldf beredenedtomakeit continuousat thispoint?
4 2 2
6
x
y
FIGURE 16
solution Thefunctionf isdiscontinuousat x = 0, at which lim
x0
f (x) = and lim
x0+
f (x) = 2. Thefunctionf
is also discontinuous at x = 2, at which lim
x2
f (x) = 6and lim
x2+
f (x) = 6. Becausethetwo one-sidedlimits exist
andareequal at x = 2, thediscontinuityat x = 2isremovable. Assigningf (2) = 6makesf continuousat x = 2.
Supposethat f (x) = 2for x < 3andf (x) = 4for x > 3.
(a) What isf (3) if f isleft-continuousat x = 3?
(b) What isf (3) if f isright-continuousat x = 3?
InExercises716, usetheLawsof ContinuityandTheorems2and3toshowthat thefunctioniscontinuous.
7. f (x) = x +sinx
solution Sincex andsinx arecontinuous, soisx +sinx byContinuityLaw(i).
f (x) = x sinx
9. f (x) = 3x +4sinx
solution Sincex andsinx arecontinuous, soare3x and4sinx byContinuityLaw(ii).Thus3x +4sinx iscontinuous
byContinuityLaw(i).
f (x) = 3x
3
+8x
2
20x
11. f (x) =
1
x
2
+1
solution
Sincex iscontinuous, soisx
2
byContinuityLaw(iii).
Recall that constant functions, suchas1, arecontinuous. Thusx
2
+1iscontinuous.
Finally,
1
x
2
+1
iscontinuousbyContinuityLaw(iv) becausex
2
+1isnever 0.
f (x) =
x
2
cosx
3+cosx
13. f (x) = cos(x
2
)
solution The function f (x) is a composite of two continuous functions: cosx and x
2
, so f (x) is continuous by
Theorem5, whichstatesthat acompositeof continuousfunctionsiscontinuous.
f (x) = sin(4
x
)
15. f (x) = 2
x
cos3x
solution 2
x
andcos3x arecontinuous, so2
x
cos3x iscontinuousbyContinuityLaw(iii).
f (x) = tan
_
1
x
2
+1
_
InExercises 1734, determinethepoints of discontinuity. Statethetypeof discontinuity(removable, jump, innite, or
noneof these) andwhether thefunctionisleft- or right-continuous.
17. f (x) =
1
x
solution Thefunction1/x isdiscontinuousatx = 0, atwhichthereisaninnitediscontinuity. Thefunctionisneither
left- nor right-continuousat x = 0.
f (x) = |x| 19. f (x) =
x 2
|x 1|
solution Thefunction
x 2
|x 1|
isdiscontinuousat x = 1, at whichthereisaninnitediscontinuity. Thefunctionis
neither left- nor right-continuousat x = 1.
f (x) = [x] 21. f (x) =
_
1
2
x
_
_
1
2
x
_
isdiscontinuousat evenintegers, at whichtherearejumpdiscontinuities. Because
lim
x2n+
_
1
2
x
_
= n
but
lim
x2n
_
1
2
x
_
= n 1,
it followsthat thisfunctionisright-continuousat theevenintegersbut not left-continuous.
g(t ) =
1
t
2
1
23. f (x) =
x +1
4x 2
solution Thefunction f (x) =
x +1
4x 2
is discontinuous at x =
1
2
, at which thereis an innitediscontinuity. The
functionisneither left- nor right-continuousat x =
1
2
.
h(z) =
12z
z
2
z 6
25. f (x) = 3x
2/3
9x
3
solution Thefunctionf (x) = 3x
2/3
9x
3
isdenedandcontinuousfor all x.
g(t ) = 3t
2/3
9t
3
27. f (x) =
x 2
|x 2|
x = 2
1 x = 2
solution For x > 2, f (x) =
x 2
(x 2)
= 1. For x < 2, f (x) =
(x 2)
(2x)
= 1. Thefunctionhasajumpdiscontinuity
at x = 2. Because
lim
x2
f (x) = 1= f (2)
but
lim
x2+
f (x) = 1= f (2),
it followsthat thisfunctionisleft-continuousat x = 2but not right-continuous.
f (x) =
_
cos
1
x
x = 0
1 x = 0
29. g(t ) = tan2t
solution Thefunctiong(t ) = tan2t =
sin2t
cos2t
isdiscontinuouswhenever cos2t = 0; i.e., whenever
2t =
(2n +1)
2
or t =
(2n +1)
4
,
wheren is aninteger. At every suchvalueof t thereis aninnitediscontinuity. Thefunctionis neither left- nor right-
continuousat anyof thesepointsof discontinuity.
f (x) = csc(x
2
)
31. f (x) = tan(sinx)
solution Thefunctionf (x) = tan(sinx) iscontinuouseverywhere. Reason: sinx iscontinuouseverywhereandtanu
iscontinuouson
_
2
,

2
_
andinparticular on1 u = sinx 1. Continuityof tan(sinx) followsbythecontinuity
of compositefunctions.
f (x) = cos([x]) 33. f (x) =
1
2
x
2
x
solution Thefunctionf (x) =
1
2
x
2
x
isdiscontinuousat x = 0, at whichthereisaninnitediscontinuity. The
functionisneither left- nor right-continuousat x = 0.
f (x) = 2
_
x
2
_
4
_
x
4
_
InExercises3548, determinethedomainof thefunctionandprovethat it iscontinuousonitsdomainusingtheLawsof
Continuityandthefactsquotedinthissection.
35. f (x) = 2sinx +3cosx
solution Thedomainof 2sinx +3cosx isall real numbers. Bothsinx andcosx arecontinuousonthisdomain, so
2sinx +3cosx iscontinuousbyContinuityLaws(i) and(ii).
f (x) =
_
x
2
+9
37. f (x) =

x sinx
solution Thisfunctionisdenedaslongasx 0. Since
x andsinx arecontinuous, sois
x sinx byContinuity
Law(iii).
f (x) =
x
2
x +x
1/4
39. f (x) = x
2/3
2
x
solution Thedomainof x
2/3
2
x
isall real numbersasthedenominator of therational exponentisodd. Bothx
2/3
and
2
x
arecontinuousonthisdomain, sox
2/3
2
x
iscontinuousbyContinuityLaw(iii).
f (x) = x
1/3
+x
3/4
41. f (x) = x
4/3
solution This functionis denedfor all x = 0. Becausethefunctionx
4/3
is continuous andnot equal to zero for
x = 0, it followsthat
x
4/3
=
1
x
4/3
iscontinuousfor x = 0byContinuityLaw(iv).
f (x) = cos
3
x
43. f (x) = tan
2
x
solution Thedomainof tan
2
x isall x = (2n 1)/2wheren isapositiveinteger. Becausetanx iscontinuouson
thisdomain, it followsfromContinuityLaw(iii) that tan
2
x isalsocontinuousonthisdomain.
f (x) = cos(2
x
)
45. f (x) = (x
4
+1)
3/2
solution Thedomainof (x
4
+ 1)
3/2
isall real numbersasx
4
+ 1 > 0for all x. Becausex
3/2
andthepolynomial
x
4
+1arebothcontinuous, soisthecompositefunction(x
4
+1)
3/2
.
f (x) = 3
x
2
47. f (x) =
cos(x
2
)
x
2
1
solution Thedomain for this function is all x = 1. Becausethefunctions cosx and x
2
arecontinuous on this
domain, soisthecompositefunctioncos(x
2
). Finally, becausethepolynomial x
2
1iscontinuousandnotequal tozero
for x = 1, thefunction
cos(x
2
)
x
2
1
iscontinuousbyContinuityLaw(iv).
f (x) = 9
tanx
49. Showthat thefunction
f (x) =
x
2
+3 for x < 1
10x for 1 x 2
6x x
2
for x > 2
is continuous for x = 1, 2. Then computetheright- and left-hand limits at x = 1, 2, and determinewhether f (x) is
left-continuous, right-continuous, or continuousat thesepoints(Figure17).
6 2 1
9
y = 10 x
y = 6x x
2
y = x
2
+ 3
x
y
FIGURE 17
solution Letsstart withx = 1, 2.
Becausex iscontinuous, soisx
2
byContinuityLaw(iii). Theconstant function3isalsocontinuous, sox
2
+3is
continuousbyContinuityLaw(i). Therefore, f (x) iscontinuousfor x < 1.
Becausex andtheconstant function10arecontinuous, thefunction10 x is continuous by Continuity Law(i).
Therefore, f (x) iscontinuousfor 1< x < 2.
Becausex is continuous, x
2
is continuous by Continuity Law(iii) and6x is continuous by Continuity Law(ii).
Therefore, 6x x
2
iscontinuousbyContinuityLaw(i), sof (x) iscontinuousfor x > 2.
At x = 1, f (x) hasajumpdiscontinuitybecausetheone-sidedlimitsexist but arenot equal:
lim
x1
f (x) = lim
x1
(x
2
+3) = 4, lim
x1+
f (x) = lim
x1+
(10x) = 9.
Furthermore, theright-handlimit equalsthefunctionvaluef (1) = 9, sof (x) isright-continuousat x = 1. At x = 2,
lim
x2
f (x) = lim
x2
(10x) = 8, lim
x2+
f (x) = lim
x2+
(6x x
2
) = 8.
Theleft- andright-handlimitsexist andareequal tof (2), sof (x) iscontinuousat x = 2.
SawtoothFunction Draw the graph of f (x) = x [x]. At which points is f discontinuous? Is it left- or
right-continuousat thosepoints?
InExercises5154, sketchthegraphof f (x). At eachpoint of discontinuity, statewhether f isleft- or right-continuous.
51. f (x) =
_
x
2
for x 1
2x for x > 1
solution
1
1
1
x
y
1 2 3
Thefunctionf iscontinuouseverywhere.
f (x) =
x +1 for x < 1
1
x
for x 1
53. f (x) =
x
2
3x +2
|x 2|
x = 2
0 x = 2
solution
1
1
2 4 6
2
3
4
5
y
x
Thefunctionf isneither left- nor right-continuousat x = 2.
f (x) =
x
3
+1 for < x 0
x +1 for 0< x < 2
x
2
+10x 15 for x 2
55. Showthat thefunction
f (x) =
x
2
16
x 4
x = 4
10 x = 4
hasaremovablediscontinuityat x = 4.
solution Toshowthat f (x) hasaremovablediscontinuityat x = 4, wemust establishthat
lim
x4
f (x)
existsbut doesnot equal f (4). Now,
lim
x4
x
2
16
x 4
= lim
x4
(x +4) = 8= 10= f (4);
thus, f (x) hasaremovablediscontinuityat x = 4. Toremovethediscontinuity, wemust redenef (4) = 8.
Denef (x) = x sin
1
x
+2forx = 0. Plotf (x). Howshouldf (0) bedenedsothatf iscontinuousatx = 0?
InExercises5759, ndthevalueof theconstant (a, b, or c) that makesthefunctioncontinuous.
57. f (x) =
_
x
2
c for x < 5
4x +2c for x 5
solution Asx 5, wehavex
2
c 25 c = L. Asx 5+, wehave4x + 2c 20+ 2c = R. Matchthe
limits: L = R or 25c = 20+2c impliesc =
5
3
.
f (x) =
_
2x +9x
1
for x 3
4x +c for x > 3
59. f (x) =
x
1
for x < 1
ax +b for 1 x
1
2
x
1
for x >
1
2
solution Asx 1, x
1
1whileasx 1+, ax + b b a. For f tobecontinuousat x = 1, we
must thereforehaveb a = 1. Now, as x
1
2
, ax + b
1
2
a + b whileas x
1
2
+, x
1
2. For f to be
continuous at x =
1
2
, wemust thereforehave
1
2
a + b = 2. Solvingthesetwo equations for a andb yields a = 2and
b = 1.
Dene
g(x) =
x +3 for x < 1
cx for 1 x 2
x +2 for x > 2
Findavalueof c suchthat g(x) is
(a) left-continuous (b) right-continuous
Ineachcase, sketchthegraphof g(x).
61. Deneg(t ) = 2
1/(t 1)
for t = 0. Answer thefollowingquestions, usingaplot if necessary.
(a) Cang(1) bedenedsothat g(t ) iscontinuousat t = 1?
(b) Howshouldg(1) bedenedsothat g(t ) isleft-continuousat t = 1?
solution
(a) Fromthegraphof g(t ) shownbelow, weseethat g approaches 0as t 1fromtheleft andbecomes inniteas
t 1fromtheright. Therefore, g(1) cannot bedenedsothat g iscontinuousat t = 1.
1
0.5
0
1.0 1.5 2.0
2
3
4
5
y
t
(b) Tomakeg(t ) left-continuousat t = 1, weshoulddene
g(1) = lim
t 1
2
1/(t 1)
= 0
Each of the following statements is false. For each statement, sketch the graph of a function that provides a
counterexample.
(a) If lim
xa
f (x) exists, thenf (x) iscontinuousat x = a.
(b) If f (x) hasajumpdiscontinuityat x = a, thenf (a) isequal toeither lim
xa
f (x) or lim
xa+
f (x).
InExercises6366, drawthegraphof afunctionon[0, 5] withthegivenproperties.
63. f (x) isnot continuousat x = 1, but lim
x1+
f (x) and lim
x1
f (x) exist andareequal.
solution
5 4 3 2 1
1
2
3
4
y
x
f (x) isleft-continuousbut not continuousat x = 2andright-continuousbut not continuousat x = 3.
65. f (x) hasaremovablediscontinuityat x = 1, ajumpdiscontinuityat x = 2, and
lim
x3
f (x) = , lim
x3+
f (x) = 2
solution
5 4 3 2 1
1
2
3
4
y
x
f (x) is right- but not left-continuous at x = 1, left- but not right-continuous at x = 2, and neither left- nor
right-continuousat x = 3.
InExercises6780, evaluateusingsubstitution.
67. lim
x1
(2x
3
4)
solution lim
x1
(2x
3
4) = 2(1)
3
4= 6.
lim
x2
(5x 12x
2
)
69. lim
x3
x +2
x
2
+2x
solution lim
x3
x +2
x
2
+2x
=
3+2
3
2
+2 3
=
5
15
=
1
3
lim
x
sin
_
x
2
_ 71. lim
x
4
tan(3x)
solution lim
x
4
tan(3x) = tan(3

4
) = tan(
3
4
) = 1
lim
x
1
cosx
73. lim
x4
x
5/2
solution lim
x4
x
5/2
= 4
5/2
=
1
32
.
lim
x2
_
x
3
+4x
75. lim
x1
(18x
3
)
3/2
solution lim
x1
(18x
3
)
3/2
= (18(1)
3
)
3/2
= 27.
lim
x2
_
7x +2
4x
_
2/3
77. lim
x3
10
x
2
2x
solution lim
x3
10
x
2
2x
= 10
3
2
2(3)
= 1000.
lim
x
2
3
sinx
79. lim
x
3
sin
2
( sin
2
x)
solution lim
x
3
sin
2
( sin
2
x) = sin
2
_
lim
x
3
sin
2
x
_
= sin
2
_
sin
2
_
3
_
_
= sin
2
_
3
4
_
=
1
2
lim
x1
tan
_
2
x1
_ 81. Supposethat f (x) andg(x) arediscontinuousat x = c. Doesit followthat f (x) +g(x) isdiscontinuousat x = c?
If not, giveacounterexample. Doesthiscontradict Theorem1(i)?
solution Evenif f (x) andg(x) arediscontinuousatx = c, itisnot necessarilytruethatf (x) +g(x) isdiscontinuous
atx = c. Forexample, supposef (x) = x
1
andg(x) = x
1
. Bothf (x) andg(x) arediscontinuousatx = 0; however,
thefunctionf (x) +g(x) = 0, whichiscontinuouseverywhere, includingx = 0. ThisdoesnotcontradictTheorem1(i),
whichdealsonlywithcontinuousfunctions.
Provethat f (x) = |x| iscontinuousfor all x. Hint: Toprovecontinuityat x = 0, consider theone-sidedlimits.
83. Usetheresult of Exercise82toprovethat if g(x) iscontinuous, thenf (x) = |g(x)| isalsocontinuous.
solution Recall thatthecompositionof twocontinuousfunctionsiscontinuous. Now, f (x) = |g(x)| isacomposition
of thecontinuousfunctionsg(x) and|x|, soisalsocontinuous.
Whichof thefollowingquantitieswouldberepresentedby continuousfunctionsof timeandwhichwouldhave
oneor morediscontinuities?
(a) Velocityof anairplaneduringaight
(b) Temperatureinaroomunder ordinaryconditions
(c) Valueof abankaccount withinterest paidyearly
(d) Thesalaryof ateacher
(e) Thepopulationof theworld
85. In2009, thefederal incometaxT (x) onincomeof x dollars(upto$82,250) wasdeterminedbytheformula
T (x) =
0.10x for 0 x < 8350

0.15x 417.50 for 8350 x < 33,950
0.25x 3812.50 for 33,950 x < 82,250
Sketchthegraphof T (x). DoesT (x) haveanydiscontinuities? Explainwhy, if T (x) hadajumpdiscontinuity, it might
beadvantageousinsomesituationstoearnlessmoney.
solution T (x), theamount of federal incometaxowedonanincomeof x dollarsin2009, might beadiscontinuous
functiondependinguponhowthetaxtablesareconstructed(asdeterminedbythatyearsregulations). Hereisagraphof
T (x) for that particular year.
20,000
5000
10,000
15,000
40,000 60,000 80,000
x
y
If T (x) hadajumpdiscontinuity(sayat x = c), it might beadvantageoustoearnslightlylessincomethanc (sayc )
andbetaxedat alower ratethantoearnc or moreandbetaxedat ahigher rate. Your net earningsmayactuallybemore
intheformer casethaninthelatter one.
If f (x) hasaremovablediscontinuityat x = c, thenit ispossibletoredenef (c) sothat f (x) iscontinuous
at x = c. Canthisbedoneinmorethanoneway?
87. Giveanexampleof functionsf (x) andg(x) suchthat f (g(x)) iscontinuousbut g(x) hasat least onediscontinuity.
solution Answersmayvary. Thesimplestexamplesarethefunctionsf (g(x)) wheref (x) = C isaconstantfunction,
andg(x) isdenedfor all x. Inthesecases, f (g(x)) = C. For example, if f (x) = 3andg(x) = [x], g isdiscontinuous
at all integer valuesx = n, but f (g(x)) = 3iscontinuous.
ContinuousatOnlyOnePoint Showthat thefollowingfunctioniscontinuousonlyat x = 0:
f (x) =
_
x for x rational
x for x irrational
89. Showthat f (x) isadiscontinuousfunctionfor all x wheref (x) isdenedasfollows:
f (x) =
_
1 for x rational
1 for x irrational
Showthat f (x)
2
iscontinuousfor all x.
solution lim
xc
f (x) doesnotexistfor anyc. If c isirrational, thenthereisalwaysarational number r arbitrarilyclose
toc sothat |f (c) f (r)| = 2. If, ontheother hand, c isrational, thereisalwaysanirrational number z arbitrarilyclose
toc sothat |f (c) f (z)| = 2.
Ontheother hand, f (x)
2
isaconstant functionthat alwayshasvalue1, whichisobviouslycontinuous.
S E C T I ON 2.5 Evaluating Limits Algebraically 57
2.5 Evaluating Limits Algebraically
1. Whichof thefollowingisindeterminateat x = 1?
x
2
+1
x 1
,
x
2
1
x +2
,
x
2
1
x +32
,
x
2
+1
x +32
solution At x = 1,
x
2
1
x+32
isof theform
0
0
; hence, thisfunctionisindeterminate. Noneof theremainingfunctions
is indeterminateat x = 1:
x
2
+1
x1
and
x
2
+1
x+32
areundenedbecausethedenominator is zerobut thenumerator is not,
while
x
2
1
x+2
isequal to0.
2. Givecounterexamplestoshowthat thesestatementsarefalse:
(a) If f (c) isindeterminate, thentheright- andleft-handlimitsasx c arenot equal.
(b) If lim
xc
f (x) exists, thenf (c) isnot indeterminate.
(c) If f (x) isundenedat x = c, thenf (x) hasanindeterminateformat x = c.
solution
(a) Let f (x) =
x
2
1
x1
. At x = 1, f isindeterminateof theform
0
0
but
lim
x1
x
2
1
x 1
= lim
x1
(x +1) = 2= lim
x1+
(x +1) = lim
x1+
x
2
1
x 1
.
(b) Again, let f (x) =
x
2
1
x1
. Then
lim
x1
f (x) = lim
x1
x
2
1
x 1
= lim
x1
(x +1) = 2
but f (1) isindeterminateof theform
0
0
.
(c) Let f (x) =
1
x
. Thenf isundenedat x = 0but doesnot haveanindeterminateformat x = 0.
3. Themethodfor evaluatinglimitsdiscussedinthissectionissometimescalledsimplifyandplugin. Explainhowit
actuallyreliesonthepropertyof continuity.
solution If f is continuous at x = c, then, by denition, lim
xc
f (x) = f (c); in other words, the limit of a
continuous function at x = c is thevalueof thefunction at x = c. Thesimplify and plug-in" strategy is based on
simplifyingafunctionwhichisindeterminatetoacontinuousfunction. Oncethesimplicationhasbeenmade, thelimit
of theremainingcontinuousfunctionisobtainedbyevaluation.
Exercises
InExercises14, showthat thelimit leadstoanindeterminateform. Thencarryout thetwo-stepprocedure: Transform
thefunctionalgebraicallyandevaluateusingcontinuity.
1. lim
x6
x
2
36
x 6
solution Whenwesubstitutex = 6into
x
2
36
x6
, weobtaintheindeterminateform
0
0
. Uponfactoringthenumerator
andsimplifying, wend
lim
x6
x
2
36
x 6
= lim
x6
(x 6)(x +6)
x 6
= lim
x6
(x +6) = 12.
lim
h3
9h
2
h 3
3. lim
x1
x
2
+2x +1
x +1
solution Whenwesubstitutex = 1into
x
2
+2x+1
x+1
,weobtaintheindeterminateform
0
0
.Uponfactoringthenumerator
andsimplifying, wend
lim
x1
x
2
+2x +1
x +1
= lim
x1
(x +1)
2
x +1
= lim
x1
(x +1) = 0.
lim
t 9
2t 18
5t 45
InExercises534, evaluatethelimit, if it exists. If not, determinewhether theone-sidedlimitsexist (niteor innite).
5. lim
x7
x 7
x
2
49
solution lim
x7
x 7
x
2
49
= lim
x7
x 7
(x 7)(x +7)
= lim
x7
1
x +7
=
1
14
.
lim
x8
x
2
64
x 9
7. lim
x2
x
2
+3x +2
x +2
solution lim
x2
x
2
+3x +2
x +2
= lim
x2
(x +1)(x +2)
x +2
= lim
x2
(x +1) = 1.
lim
x8
x
3
64x
x 8
9. lim
x5
2x
2
9x 5
x
2
25
solution lim
x5
2x
2
9x 5
x
2
25
= lim
x5
(x 5)(2x +1)
(x 5)(x +5)
= lim
x5
2x +1
x +5
=
11
10
.
lim
h0
(1+h)
3
1
h
11. lim
x
1
2
2x +1
2x
2
+3x +1
solution lim
x
1
2
2x +1
2x
2
+3x +1
= lim
x
1
2
2x +1
(2x +1)(x +1)
= lim
x
1
2
1
x +1
= 2.
lim
x3
x
2
x
x
2
9
13. lim
x2
3x
2
4x 4
2x
2
8
solution lim
x2
3x
2
4x 4
2x
2
8
= lim
x2
(3x +2)(x 2)
2(x 2)(x +2)
= lim
x2
3x +2
2(x +2)
=
8
8
= 1.
lim
h0
(3+h)
3
27
h
15. lim
t 0
4
2t
1
4
t
1
solution lim
t t o0
4
2t
1
4
t
1
= lim
t t o0
(4
t
1)(4
t
+1)
4
t
1
= lim
t 0
(4
t
+1) = 2.
lim
h4
(h +2)
2
9h
h 4
17. lim
x16
x 4
x 16
solution lim
x16
x 4
x 16
= lim
x16
x 4
_
x +4
_ _
x 4
_ = lim
x16
1
x +4
=
1
8
.
lim
t 2
2t +4
123t
2
19. lim
y3
y
2
+y 12
y
3
10y +3
solution lim
y3
y
2
+y 12
y
3
10y +3
= lim
y3
(y 3)(y +4)
(y 3)(y
2
+3y 1)
= lim
y3
(y +4)
(y
2
+3y 1)
=
7
17
.
lim
h0
1
(h +2)
2

1
4
h
21. lim
h0
2+h 2
h
solution lim
h0
h +22
h
doesnot exist.
Ash 0+, wehave
h +22
h
=
_
h +22
_
(
h +2+2)
h(
h +2+2)
=
h 2
h(
h +2+2)
.
Ash 0, wehave
h +22
h
=
_
h +22
_
(
h +2+2)
h(
h +2+2)
=
h 2
h(
h +2+2)
.
lim
x8
x 42
x 8
23. lim
x4
x 4
8x
solution
lim
x4
x 4
8x
= lim
x4
(x 4)(
x +
8x)
(
8x)(
x +
8x)
= lim
x4
(x 4)(
x +
8x)
x (8x)
= lim
x4
(x 4)(
x +
8x)
2x 8
= lim
x4
(x 4)(
x +
8x)
2(x 4)
= lim
x4
(
x +
8x)
2
=
4+
4
2
= 2.
S E C T I ON 2.5 Evaluating Limits Algebraically 59
lim
x4
5x 1
2
x
25. lim
x4
_
1
x 2
4
x 4
_
solution lim
x4
_
1
x 2
4
x 4
_
= lim
x4
x +24
_
x 2
_ _
x +2
_ = lim
x4
x 2
_
x 2
_ _
x +2
_ =
1
4
.
lim
x0+
_
1
1
_
x
2
+x
_
27. lim
x0
cotx
cscx
solution lim
x0
cotx
cscx
= lim
x0
cosx
sinx
sinx = cos0= 1.
lim
2
cot
csc
29. lim
t 2
2
2t
+2
t
20
2
t
4
solution lim
t 2
2
2t
+2
t
20
2
t
4
= lim
t 2
(2
t
+5)(2
t
4)
2
t
4
= lim
t 2
(2
t
+5) = 9.
lim
x1
_
1
1x
2
1x
2
_
31. lim
x
4
sinx cosx
tanx 1
solution lim
x
4
sinx cosx
tanx 1

cosx
cosx
= lim
x
4
(sinx cosx) cosx
sinx cosx
= cos
4
=
2
2
.
lim
2
_
sec tan
_
33. lim
4
_
1
tan 1
2
tan
2
1
_
solution lim
4
_
1
tan 1
2
tan
2
1
_
= lim
4
(tan +1) 2
(tan +1)(tan 1)
= lim
4
1
tan +1
=
1
2
.
lim
x
3
2cos
2
x +3cosx 2
2cosx 1
35. Useaplot of f (x) =
x 4
8x
toestimate lim
x4
f (x) totwodecimal places. Comparewiththeanswer
obtainedalgebraicallyinExercise23.
solution Letf (x) =
x4
8x
. Fromtheplotof f (x) shownbelow, weestimate lim
x4
f (x) 2.00; totwodecimal
places, thismatchesthevalueof 2obtainedinExercise23.
1.996
1.997
1.998
1.999
2.000
2.001
1.995
3.6
y
x
3.8 4.0 4.2 4.4
Useaplotof f (x) =
1
x 2
4
x 4
toestimate lim
x4
f (x) numerically. Comparewiththeanswer obtained
algebraicallyinExercise25.
InExercises3742, evaluateusingtheidentity
a
3
b
3
= (a b)(a
2
+ab +b
2
)
37. lim
x2
x
3
8
x 2
solution lim
x2
x
3
8
x 2
= lim
x2
(x 2)
_
x
2
+2x +4
_
x 2
= lim
x2
_
x
2
+2x +4
_
= 12.
lim
x3
x
3
27
x
2
9
39. lim
x1
x
2
5x +4
x
3
1
solution lim
x1
x
2
5x +4
x
3
1
= lim
x1
(x 1)(x 4)
(x 1)
_
x
2
+x +1
_ = lim
x1
x 4
x
2
+x +1
= 1.
lim
x2
x
3
+8
x
2
+6x +8
41. lim
x1
x
4
1
x
3
1
solution
lim
x1
x
4
1
x
3
1
= lim
x1
(x
2
1)(x
2
+1)
(x 1)(x
2
+x +1)
= lim
x1
(x 1)(x +1)(x
2
+1)
(x 1)(x
2
+x +1)
= lim
x1
(x +1)(x
2
+1)
(x
2
+x +1)
=
4
3
.
lim
x27
x 27
x
1/3
3
43. Evaluate lim
h0
4
1+h 1
h
. Hint: Set x =
4
1+h andrewriteasalimit asx 1.

solution Let x =
4
1+h. Thenh = x
4
1= (x 1)(x +1)(x
2
+1), x 1ash 0and
lim
h0
4
1+h 1
h
= lim
x1
x 1
(x 1)(x +1)(x
2
+1)
= lim
x1
1
(x +1)(x
2
+1)
=
1
4
.
Evaluate lim
h0
3
1+h 1
2
1+h 1
. Hint: Set x =
6
1+h andrewriteasalimit asx 1.

InExercises4554, evaluateintermsof theconstant a.
45. lim
x0
(2a +x)
solution lim
x0
(2a +x) = 2a.
lim
h2
(4ah +7a)
47. lim
t 1
(4t 2at +3a)
solution lim
t 1
(4t 2at +3a) = 4+5a.
lim
h0
(3a +h)
2
9a
2
h
49. lim
h0
2(a +h)
2
2a
2
h
solution lim
h0
2(a +h)
2
2a
2
h
= lim
h0
4ha +2h
2
h
= lim
h0
(4a +2h) = 4a.
lim
xa
(x +a)
2
4x
2
x a
51. lim
xa
a
x a
solution lim
xa
a
x a
= lim
xa
a
_
x
a
_ _
x +
a
_ = lim
xa
1
x +
a
=
1
2
a
.
lim
h0
a +2h
a
h
53. lim
x0
(x +a)
3
a
3
x
solution lim
x0
(x +a)
3
a
3
x
= lim
x0
x
3
+3x
2
a +3xa
2
+a
3
a
3
x
= lim
x0
(x
2
+3xa +3a
2
) = 3a
2
.
lim
ha
1
h
1
a
h a
InExercises5558, ndall valuesof c suchthat thelimit exists.
55. lim
xc
x
2
5x 6
x c
solution lim
xc
x
2
5x 6
x c
will exist providedthat x c is afactor of thenumerator. (Otherwisetherewill bean
innitediscontinuityat x = c.) Sincex
2
5x 6= (x +1)(x 6), thisoccursfor c = 1andc = 6.
lim
x1
x
2
+3x +c
x 1
57. lim
x1
_
1
x 1
c
x
3
1
_
solution Simplifying, wend
1
x 1
c
x
3
1
=
x
2
+x +1c
(x 1)(x
2
+x +1)
.
Inorder for thelimit toexist asx 1, thenumerator must evaluateto0at x = 1. Thus, wemust have3c = 0, which
impliesc = 3.
lim
x0
1+cx
2
_
1+x
2
x
4
59. For whichsign doesthefollowinglimit exist?
lim
x0
_
1
x
1
x(x 1)
_
solution
Thelimit lim
x0
_
1
x
+
1
x(x 1)
_
= lim
x0
(x 1) +1
x(x 1)
= lim
x0
1
x 1
= 1.
Thelimit lim
x0
_
1
x
1
x(x 1)
_
doesnot exist.
Asx 0+, wehave
1
x
1
x(x 1)
=
(x 1) 1
x(x 1)
=
x 2
x(x 1)
.
Asx 0, wehave
1
x
1
x(x 1)
=
(x 1) 1
x(x 1)
=
x 2
x(x 1)
.
S E C T I ON 2.6 Trigonometric Limits 61
2.6 Trigonometric Limits
1. Assumethat x
4
f (x) x
2
. What is lim
x0
f (x)? Isthereenoughinformationtoevaluate lim
x
1
2
f (x)? Explain.
solution Since lim
x0
x
4
= lim
x0
x
2
= 0, the squeeze theoremguarantees that lim
x0
f (x) = 0. Since
lim
x
1
2
x
4
=
1
16
=
1
4
= lim
x
1
2
x
2
, wedonot haveenoughinformationtodeterminelim
x
1
2
f (x).
2. StatetheSqueezeTheoremcarefully.
solution Assumethat for x = c (insomeopeninterval containingc),
l(x) f (x) u(x)
andthat lim
xc
l(x) = lim
xc
u(x) = L. Then lim
xc
f (x) existsand
lim
xc
f (x) = L.
3. If youwant toevaluate lim
h0
sin5h
3h
, it isagoodideatorewritethelimit intermsof thevariable(chooseone):
(a) = 5h (b) = 3h (c) =
5h
3
solution Tomatchthegivenlimit tothepatternof
lim
0
sin
,
it isbest tosubstitutefor theargument of thesinefunction; thus, rewritethelimit intermsof (a): = 5h.
Exercises
1. Statepreciselythehypothesisandconclusionsof theSqueezeTheoremfor thesituationinFigure6.
1 2
2
u(x)
l(x)
f(x)
x
y
FIGURE 6
solution For all x = 1ontheopeninterval (0, 2) containingx = 1, (x) f (x) u(x). Moreover,
lim
x1
(x) = lim
x1
u(x) = 2.
Therefore, bytheSqueezeTheorem,
lim
x1
f (x) = 2.
InFigure7, isf (x) squeezedbyu(x) andl(x) at x = 3?At x = 2?
3. WhatdoestheSqueezeTheoremsayabout lim
x7
f (x) if lim
x7
l(x) = lim
x7
u(x) = 6andf (x),u(x),andl(x) arerelated
asinFigure8?Theinequalityf (x) u(x) isnot satisedfor all x. Doesthisaffect thevalidityof your conclusion?
7
6
x
u(x)
f(x)
l(x)
y
FIGURE 8
solution TheSqueezeTheoremdoesnotrequirethattheinequalitiesl(x) f (x) u(x) holdfor all x, onlythatthe
inequalitiesholdonsomeopeninterval containingx = c. InFigure8, it isclear that l(x) f (x) u(x) onsomeopen
interval containingx = 7. Because lim
x7
u(x) = lim
x7
l(x) = 6, theSqueezeTheoremguaranteesthat lim
x7
f (x) = 6.
Determine lim
x0
f (x) assumingthat cosx f (x) 1.
5. Statewhether theinequalityprovidessufcient informationtodetermine lim
x1
f (x), andif so, ndthelimit.
(a) 4x 5 f (x) x
2
(b) 2x 1 f (x) x
2
(c) 4x x
2
f (x) x
2
+2
solution
(a) Because lim
x1
(4x 5) = 1= 1= lim
x1
x
2
,thegiveninequalitydoesnotprovidesufcientinformationtodetermine
lim
x1
f (x).
(b) Because lim
x1
(2x 1) = 1= lim
x1
x
2
, it followsfromtheSqueezeTheoremthat lim
x1
f (x) = 1.
(c) Because lim
x1
(4x x
2
) = 3= lim
x1
(x
2
+2), it followsfromtheSqueezeTheoremthat lim
x1
f (x) = 3.
Plot thegraphs of u(x) = 1+

x

2
andl(x) = sinx onthesameset of axes. What canyousay about

lim
x
2
f (x) if f (x) issqueezedbyl(x) andu(x) at x =

2
?
InExercises716, evaluateusingtheSqueezeTheorem.
7. lim
x0
x
2
cos
1
x
solution Multiplyingtheinequality1 cos
1
x
1, whichholdsfor all x = 0, byx
2
yieldsx
2
x
2
cos
1
x
x
2
.
Because
lim
x0
x
2
= lim
x0
x
2
= 0,
it followsbytheSqueezeTheoremthat
lim
x0
x
2
cos
1
x
= 0.
lim
x0
x sin
1
x
2
9. lim
x1
(x 1) sin

x 1
solution Multiplyingtheinequality
sin

x1
1, whichholds for x = 1, by |x 1| yields
(x 1) sin

x1

|x 1| or |x 1| (x 1) sin

x1
|x 1|. Because
lim
x1
|x 1| = lim
x1
|x 1| = 0,
lim
x1
(x 1) sin

x 1
= 0.
lim
x3
(x
2
9)
x 3
|x 3|
11. lim
t 0
(2
t
1) cos
1
t
cos
1
t
1, whichholdsfor t = 0, by|2
t
1| yields
(2
t
1) cos
1
t
|2
t
1|
or |2
t
1| (2
t
1) cos
1
t
|2
t
1|. Because
lim
t 0
|2
t
1| = lim
t 0
|2
t
1| = 0,
lim
t 0
(2
t
1) cos
1
t
= 0.
lim
x0+
x 4
cos(/x) 13. lim
t 2
(t
2
4) cos
1
t 2
cos
1
t 2
1, whichholds for t = 2, by |t
2
4| yields
(t
2
4) cos
1
t 2

|t
2
4| or |t
2
4| (t
2
4) cos
1
t 2
|t
2
4|. Because
lim
t 2
|t
2
4| = lim
t 2
|t
2
4| = 0,
lim
t 2
(t
2
4) cos
1
t 2
= 0.
lim
x0
tanx cos
_
sin
1
x
_
15. lim
2
cos cos(tan)
solution Multiplyingtheinequality | cos(tan)| 1, whichholds for all near

2
but not equal to

2
, by | cos|
yields| cos cos(tan)| | cos| or | cos| cos cos(tan) | cos|. Because
lim
2
| cos| = lim
2
| cos| = 0,
it followsfromtheSqueezeTheoremthat
lim
2
cos cos(tan) = 0.
lim
t 0
sin
2
_
1
t
_
3
1/t
InExercises1726, evaluateusingTheorem2asnecessary.
17. lim
x0
tanx
x
solution lim
x0
tanx
x
= lim
x0
sinx
x
1
cosx
= lim
x0
sinx
x
lim
x0
1
cosx
= 1 1= 1.
lim
x0
sinx secx
x
19. lim
t 0
_
t
3
+9sint
t
solution lim
t 0
_
t
3
+9sint
t
= lim
t 0
_
t
3
+9 lim
t 0
sint
t
=
9 1= 3.
lim
t 0
sin
2
t
t
21. lim
x0
x
2
sin
2
x
solution lim
x0
x
2
sin
2
x
= lim
x0
1
sinx
x
sinx
x
= lim
x0
1
sinx
x
lim
x0
1
sinx
x
=
1
1
1
1
= 1.
lim
t
2
1cost
t
23. lim
0
sec 1
solution lim
0
sec 1
= lim
0
1cos
cos
= lim
0
1cos
lim
0
1
cos
= 0 1= 0.
lim
0
1cos
sin
25. lim
t
4
sint
t
solution
sint
t
iscontinuousat t =

4
. Hence, bysubstitution
lim
t
4
sint
t
=
2
2
4
=
2
.
lim
t 0
cost cos
2
t
t
27. Let L = lim
x0
sin14x
x
.
(a) Show, byletting = 14x, that L = lim
0
14
sin
.
(b) ComputeL.
solution
(a) Let = 14x. Thenx =

14
and 0asx 0, so
L = lim
x0
sin14x
x
= lim
0
sin
(/14)
= lim
0
14
sin
.
(b) Basedonpart (a),
L = 14 lim
0
sin
= 14.
Evaluate lim
h0
sin9h
sin7h
. Hint:
sin9h
sin7h
=
_
9
7
__
sin9h
9h
__
7h
sin7h
_
.
InExercises2948, evaluatethelimit.
29. lim
h0
sin9h
h
solution lim
h0
sin9h
h
= lim
h0
9
sin9h
9h
= 9.
lim
h0
sin4h
4h
31. lim
h0
sinh
5h
solution lim
h0
sinh
5h
= lim
h0
1
5
sinh
h
=
1
5
.
lim
x
6
x
sin3x
33. lim
0
sin7
sin3
solution Wehave
sin7
sin3
=
7
3
_
sin7
7
__
3
sin3
_
Therefore,
lim
0
sin7
3
=
7
3
_
lim
0
sin7
7
__
lim
0
3
sin3
_
=
7
3
(1)(1) =
7
3
lim
x0
tan4x
9x
35. lim
x0
x csc25x
solution Let h = 25x. Then
lim
x0
x csc25x = lim
h0
h
25
csch =
1
25
lim
h0
h
sinh
=
1
25
.
lim
t 0
tan4t
t sect
37. lim
h0
sin2h sin3h
h
2
solution
lim
h0
sin2h sin3h
h
2
= lim
h0
sin2h sin3h
h h
= lim
h0
sin2h
h
sin3h
h
= lim
h0
2
sin2h
2h
3
sin3h
3h
= lim
h0
2
sin2h
2h
lim
h0
3
sin3h
3h
= 2 3= 6.
lim
z0
sin(z/3)
sinz
39. lim
0
sin(3)
sin(4)
solution lim
0
sin(3)
sin(4)
= lim
0
sin(3)
3
3
4
4
sin(4)
=
3
4
.
lim
x0
tan4x
tan9x
41. lim
t 0
csc8t
csc4t
solution lim
t 0
csc8t
csc4t
= lim
t 0
sin4t
sin8t
8t
4t
1
2
=
1
2
.
lim
x0
sin5x sin2x
sin3x sin5x
43. lim
x0
sin3x sin2x
x sin5x
solution lim
x0
sin3x sin2x
x sin5x
= lim
x0
_
3
sin3x
3x
2
5
(sin2x) / (2x)
(sin5x) / (5x)
_
=
6
5
.
lim
h0
1cos2h
h
45. lim
h0
sin(2h)(1cosh)
h
2
solution lim
h0
sin(2h)(1cosh)
h
2
= lim
h0
sin(2h)
h
lim
h0
1cosh
h
= 1 0= 0.
lim
t 0
1cos2t
sin
2
3t
47. lim
0
cos2 cos
solution
lim
0
cos2 cos
= lim
0
(cos2 1) +(1cos)
= lim
0
cos2 1
+ lim
0
1cos
= 2 lim
0
1cos2
2
+ lim
0
1cos
= 2 0+0= 0.
lim
h
2
1cos3h
h
49. Calculate lim
x0
sinx
|x|
.
solution
lim
x0
sinx
|x|
= lim
x0
sinx
x
= 1
Usetheidentitysin3 = 3sin 4sin
3
toevaluatethelimit lim
0
sin3 3sin
3
.
51. Provethefollowingresult statedinTheorem2:
lim
0
1cos
= 0 7
Hint:
1cos
=
1
1+cos
1cos
2
.
solution
lim
0
1cos
= lim
0
1
1+cos
1cos
2
= lim
0
1
1+cos
sin
2
= lim
0
1
1+cos
lim
0
sin
2
= lim
0
1
1+cos
lim
0
sin
sin
= lim
0
1
1+cos
lim
0
sin lim
0
sin
=
1
2
0 1= 0.
Investigate lim
h0
1cosh
h
2
numerically (andgraphically if youhaveagraphingutility). Thenprovethat the
limit isequal to
1
2
. Hint: Seethehint for Exercise51.
InExercises5355, evaluateusingtheresult of Exercise52.
53. lim
h0
cos3h 1
h
2
solution Wemakethesubstitution = 3h. Thenh = /3, and
lim
h0
cos3h 1
h
2
= lim
0
cos 1
(/3)
2
= 9 lim
0
1cos
2
=
9
2
.
lim
h0
cos3h 1
cos2h 1
55. lim
t 0
1cost
t
solution lim
t 0+
1cost
t
=
_
lim
t 0+
1cost
t
2
=
_
1
2
=
2
2
; on the other hand, lim
t 0
1cost
t
=
_
lim
t 0
1cost
t
2
=
_
1
2
=
2
2
.
UsetheSqueezeTheoremtoprovethat if lim
xc
|f (x)| = 0, then lim
xc
f (x) = 0.
57. Usetheresult of Exercise52toprovethat for m = 0,
lim
x0
cosmx 1
x
2
=
m
2
2
solution Substituteu = mx into
cosmx 1
x
2
. Weobtainx =
u
m
. Asx 0, u 0; therefore,
lim
x0
cosmx 1
x
2
= lim
u0
cosu 1
(u/m)
2
= lim
u0
m
2
cosu 1
u
2
= m
2
_
1
2
_
=
m
2
2
.
Usingadiagramof theunit circleandthePythagoreanTheorem, showthat
sin
2
(1cos)
2
+sin
2

2
Concludethat sin
2
2(1 cos)
2
andusethistogiveanalternativeproof of Eq. (7) inExercise51. Then
giveanalternativeproof of theresult inExercise52.
59. (a) Investigate lim
xc
sinx sinc
x c
numericallyfor thevevaluesc = 0,

6
,

4
,

3
,

2
.
(b) Canyouguesstheanswer for general c?
(c) Checkthat your answer to(b) worksfor twoother valuesof c.
solution
(a)
x c 0.01 c 0.001 c +0.001 c +0.01
sinx sinc
x c
0.999983 0.99999983 0.99999983 0.999983
Herec = 0andcosc = 1.
x c 0.01 c 0.001 c +0.001 c +0.01
sinx sinc
x c
0.868511 0.866275 0.865775 0.863511
Herec =

6
andcosc =
3
2
0.866025.
x c 0.01 c 0.001 c +0.001 c +0.01
sinx sinc
x c
0.504322 0.500433 0.499567 0.495662
Herec =

3
andcosc =
1
2
.
x c 0.01 c 0.001 c +0.001 c +0.01
sinx sinc
x c
0.710631 0.707460 0.706753 0.703559
Herec =

4
andcosc =
2
2
0.707107.
x c 0.01 c 0.001 c +0.001 c +0.01
sinx sinc
x c
0.005000 0.000500 0.000500 0.005000
Herec =

2
andcosc = 0.
(b) lim
xc
sinx sinc
x c
= cosc.
(c)
x c 0.01 c 0.001 c +0.001 c +0.01
sinx sinc
x c
0.411593 0.415692 0.416601 0.420686
Herec = 2andcosc = cos2 0.416147.
x c 0.01 c 0.001 c +0.001 c +0.01
sinx sinc
x c
0.863511 0.865775 0.866275 0.868511
Herec =
6
andcosc =
3
2
0.866025.
2.7 Limits at Infinity
1. Assumethat
lim
x
f (x) = L and lim
xL
g(x) =
Whichof thefollowingstatementsarecorrect?
S E C T I ON 2.7 Limits at Infinity 67
(a) x = L isavertical asymptoteof g(x).
(b) y = L isahorizontal asymptoteof g(x).
(c) x = L isavertical asymptoteof f (x).
(d) y = L isahorizontal asymptoteof f (x).
solution
(a) Because lim
xL
g(x) = , x = L isavertical asymptoteof g(x). Thisstatement iscorrect.
(b) Thisstatement isnot correct.
(c) Thisstatement isnot correct.
(d) Because lim
x
f (x) = L, y = L isahorizontal asymptoteof f (x). Thisstatement iscorrect.
2. What arethefollowinglimits?
(a) lim
x
x
3
(b) lim
x
x
3
(c) lim
x
x
4
solution
(a) lim
x
x
3
=
(b) lim
x
x
3
=
(c) lim
x
x
4
=
3. Sketchthegraphof afunctionthatapproachesalimitasx butdoesnotapproachalimit(eitherniteorinnite)
asx .
solution
y
x
4. What isthesignof a if f (x) = ax
3
+x +1satises
lim
x
f (x) = ?
solution Because lim
x
x
3
= , a must benegativetohave lim
x
f (x) = .
5. What isthesignof theleadingcoefcient a
7
if f (x) isapolynomial of degree7suchthat lim
x
f (x) = ?
solution The behavior of f (x) as x is controlled by the leading term; that is, lim
x
f (x) =
lim
x
a
7
x
7
. Becausex
7
asx , a
7
must benegativetohavelim
x
f (x) = .
6. Explainwhy lim
x
sin
1
x
existsbut lim
x0
sin
1
x
doesnot exist. What is lim
x
sin
1
x
?
solution Asx ,
1
x
0, so
lim
x
sin
1
x
= sin0= 0.
Ontheother hand,
1
x
asx 0, andas
1
x
, sin
1
x
oscillatesinnitelyoften. Thus
lim
x0
sin
1
x
doesnot exist.
Exercises
1. What arethehorizontal asymptotesof thefunctioninFigure6?
20 20 40 60 80
x
1
2
y
y = f(x)
FIGURE 6
solution Because
lim
x
f (x) = 1 and lim
x
f (x) = 2,
thefunctionf (x) hashorizontal asymptotesof y = 1andy = 2.
Sketchthegraphof afunctionf (x) that hasbothy = 1andy = 5ashorizontal asymptotes.
3. Sketchthegraphof afunctionf (x) withasinglehorizontal asymptotey = 3.
solution
13
9
5
1
4 2 2
3
y
x
Sketchthegraphsof twofunctionsf (x) andg(x) thathavebothy = 2andy = 4ashorizontal asymptotesbut
lim
x
f (x) = lim
x
g(x).
5. Investigatetheasymptoticbehavior of f (x) =
x
3
x
3
+x
numericallyandgraphically:
(a) Makeatableof valuesof f (x) for x = 50, 100, 500, 1000.
(b) Plot thegraphof f (x).
(c) What arethehorizontal asymptotesof f (x)?
solution
(a) Fromthetablebelow, it appearsthat
lim
x
x
3
x
3
+x
= 1.
x 50 100 500 1000
f (x) 0.999600 0.999900 0.999996 0.999999
(b) Fromthegraphbelow, it alsoappearsthat
lim
x
x
3
x
3
+x
= 1.
5 5
0.2
0.4
0.6
0.8
1.0
y
x
(c) Thehorizontal asymptoteof f (x) isy = 1.
Investigate lim
x
12x +1
_
4x
2
+9
numericallyandgraphically:
(a) Makeatableof valuesof f (x) =
12x +1
_
4x
2
+9
for x = 100, 500, 1000, 10,000.
(b) Plot thegraphof f (x).
(c) What arethehorizontal asymptotesof f (x)?
7. lim
x
x
x +9
solution
lim
x
x
x +9
= lim
x
x
1
(x)
x
1
(x +9)
= lim
x
1
1+
9
x
=
1
1+0
= 1.
lim
x
3x
2
+20x
4x
2
+9
9. lim
x
3x
2
+20x
2x
4
+3x
3
29
solution
lim
x
3x
2
+20x
2x
4
+3x
3
29
= lim
x
x
4
(3x
2
+20x)
x
4
(2x
4
+3x
3
29)
= lim
x
3
x
2
+
20
x
3
2+
3
x

29
x
4
=
0
2
= 0.
lim
x
4
x +5
11. lim
x
7x 9
4x +3
solution
lim
x
7x 9
4x +3
= lim
x
x
1
(7x 9)
x
1
(4x +3)
= lim
x
7
9
x
4+
3
x
=
7
4
.
lim
x
9x
2
2
629x
13. lim
x
7x
2
9
4x +3
solution
lim
x
7x
2
9
4x +3
= lim
x
x
1
(7x
2
9)
x
1
(4x +3)
= lim
x
7x
9
x
4+
3
x
= .
lim
x
5x 9
4x
3
+2x +7
15. lim
x
3x
3
10
x +4
solution
lim
x
3x
3
10
x +4
= lim
x
x
1
(3x
3
10)
x
1
(x +4)
= lim
x
3x
2
10
x
1+
4
x
=

1
= .
lim
x
2x
5
+3x
4
31x
8x
4
31x
2
+12
InExercises1722, ndthehorizontal asymptotes.
17. f (x) =
2x
2
3x
8x
2
+8
solution First calculatethelimitsasx . For x ,
lim
x
2x
2
3x
8x
2
+8
= lim
x
2
3
x
8+
8
x
2
=
2
8
=
1
4
.
Similarly,
lim
x
2x
2
3x
8x
2
+8
= lim
x
2
3
x
8+
8
x
2
=
2
8
=
1
4
.
Thus, thehorizontal asymptoteof f (x) isy =
1
4
.
f (x) =
8x
3
x
2
7+11x 4x
4
19. f (x) =
_
36x
2
+7
9x +4
solution For x > 0, x
1
= |x
1
| =
x
2
, so
lim
x
_
36x
2
+7
9x +4
= lim
x
_
36+
7
x
2
9+
4
x
=
36
9
=
2
3
.
Ontheother hand, for x < 0, x
1
= |x
1
| =
x
2
, so
lim
x
_
36x
2
+7
9x +4
= lim
x
_
36+
7
x
2
9+
4
x
=

36
9
=
2
3
.
Thus, thehorizontal asymptotesof f (x) arey =
2
3
andy =
2
3
.
f (x) =
_
36x
4
+7
9x
2
+4
21. f (t ) =
3
t
1+3
t
solution With
lim
t
3
t
1+3
t
=

1
=
and
lim
t
3
t
1+3
t
= 0,
thefunctionf (t ) hasonehorizontal asymptote, y = 0.
f (t ) =
t
1/3
(64t
2
+9)
1/6
23. lim
x
_
9x
4
+3x +2
4x
3
+1
3
= |x
3
| =
x
6
, so
lim
x
_
9x
4
+3x +2
4x
3
+1
= lim
x
_
9
x
2
+
3
x
5
+
2
x
6
4+
1
x
3
= 0.
lim
x
_
x
3
+20x
10x 2
25. lim
x
8x
2
+7x
1/3
_
16x
4
+6
solution For x < 0, x
2
= |x
2
| =
x
4
, so
lim
x
8x
2
+7x
1/3
_
16x
4
+6
= lim
x
8+
7
x
5/3
_
16+
6
x
4
=
8
16
= 2.
lim
x
4x 3
_
25x
2
+4x
27. lim
t
t
4/3
+t
1/3
(4t
2/3
+1)
2
solution lim
t
t
4/3
+t
1/3
(4t
2/3
+1)
2
= lim
t
1+
1
t
(4+
1
t
2/3
)
2
=
1
16
.
lim
t
t
4/3
9t
1/3
(8t
4
+2)
1/3
29. lim
x
|x| +x
x +1
solution For x < 0, |x| = x. Therefore, for all x < 0,
|x| +x
x +1
=
x +x
x +1
= 0;
consequently,
lim
x
|x| +x
x +1
= 0.
lim
t
4+6 10
2t
59 10
3t
31. Determinethelimitsat innityof g(t ) = 5
1/t
2
.
t
1
t
2
=
1
= 0, it followsthat
lim
t
5
1/t
2
= 5
0
= 1
Showthat lim
x
(
_
x
2
+1x) = 0. Hint: Observethat
_
x
2
+1x =
1
_
x
2
+1+x
33. AccordingtotheMichaelisMentenequation(Figure7), whenanenzymeiscombinedwithasubstrateof concen-
trations (inmillimolars), thereactionrate(inmicromolars/min) is
R(s) =
As
K +s
(A, K constants)
(a) Show, bycomputing lim
s
R(s), that A isthelimitingreactionrateastheconcentrations approaches.
(b) Showthat thereactionrateR(s) attainsone-half of thelimitingvalueA whens = K.
(c) For acertainreaction, K = 1.25mM andA = 0.1. For whichconcentrations isR(s) equal to75%of itslimiting
value?
Leonor Michaelis
18751949
Maud Menten
18791960
FIGURE 7 Canadian-bornbiochemist MaudMentenisbest knownfor her fundamental workonenzymekineticswith
Germanscientist Leonor Michaelis. Shewasalsoanaccomplishedpainter, clarinetist, mountainclimber, andmaster of
numerouslanguages.
solution
(a) lim
s
R(s) = lim
s
As
K +s
= lim
s
A
1+
K
s
= A.
(b) Observethat
R(K) =
AK
K +K
=
AK
2K
=
A
2
,
haveof thelimitingvalue.
(c) Bypart (a), thelimitingvalueis0.1, soweneedtodeterminethevalueof s that satises
R(s) =
0.1s
1.25+s
= 0.075.
Solvingthisequationfor s yields
s =
(1.25)(0.075)
0.025
= 3.75mM.
Supposethattheaveragetemperatureof theearthisT (t ) =283+3(110
0.013t
) kelvins, wheret isthenumber
of yearssince2000.
(a) Calculatethelong-termaverageL = lim
t
T (t ).
(b) At what timeisT (t ) withinone-half adegreeof itslimitingvalue?
InExercises3542, calculatethelimit.
35. lim
x
_
_
4x
4
+9x 2x
2
_
solution Write
_
4x
4
+9x 2x
2
=
__
4x
4
+9x 2x
2
_
_
4x
4
+9x +2x
2
_
4x
4
+9x +2x
2
=
(4x
4
+9x) 4x
4
_
4x
4
+9x +2x
2
=
9x
_
4x
4
+9x +2x
2
.
Thus,
lim
x
(
_
4x
4
+9x 2x
2
) = lim
x
9x
_
4x
4
+9x +2x
2
= 0.
lim
x
(
_
9x
3
+x x
3/2
)
37. lim
x
_
2
x +2
_
solution Write
2
x +2=
_
2
x +2
_ 2
x +
x +2
2
x +
x +2
=
4x (x +2)
2
x +
x +2
=
3x 2
2
x +
x +2
.
Thus,
lim
x
(2
x +2) = lim
x
3x 2
2
x +
x +2
= .
lim
x
_
1
x
1
x +2
_
39. lim
t
tan
_
3
t
+1
43
t +1
_
solution Dividingnumerator anddenominator by3
t
gives
lim
t
tan
_
3
t
+1
43
t +1
_
= lim
t
tan
_
+3
t
4 3
t
3
_
= tan
_

3
_
= tan
_
3
_
=
3
lim
t
2
_
8t
t +1
10
t +1
_
41. Let P(n) betheperimeter of ann-goninscribedinaunit circle(Figure8).
(a) Explain, intuitively, whyP(n) approaches2 asn .
(b) Showthat P(n) = 2n sin
_
n
_
.
(c) Combine(a) and(b) toconcludethat lim
n
n
sin
_
n
_
= 1.
(d) Usethistogiveanother argument that lim
0
sin
= 1.
n = 6 n = 9 n = 12
FIGURE 8
solution
(a) Asn , then-gonapproachesacircleof radius1. Therefore, theperimeter of then-gonapproachesthecircum-
ferenceof theunit circleasn . That is, P(n) 2 asn .
(b) Eachsideof then-gonisthethirdsideof anisoscelestrianglewithequal lengthsidesof length1andangle =
2
n
betweentheequal lengthsides. Thelengthof eachsideof then-gonistherefore
_
1
2
+1
2
2cos
2
n
=
_
2(1cos
2
n
) =
_
4sin
2

n
= 2sin
n
.
Finally,
P(n) = 2n sin
n
.
(c) Combiningparts(a) and(b),
lim
n
P(n) = lim
n
2n sin
n
= 2.
Dividingbothsidesof thislast expressionby2 yields
lim
n
n
sin
n
= 1.
(d) Let =

n
. Then 0asn ,
n
sin
n
=
1
sin =
sin
,
and
lim
n
n
sin
n
= lim
0
sin
= 1.
PhysicistshaveobservedthatEinsteinstheoryof special relativityreducestoNewtonianmechanicsinthelimit
as c , wherec is thespeed of light. This is illustrated by astonetossed up vertically fromground level so
that it returnstoearthonesecondlater. UsingNewtonsLaws, wendthat thestonesmaximumheight ish = g/8
meters(g = 9.8m/s
2
). Accordingtospecial relativity, thestonesmassdependsonitsvelocitydividedbyc, andthe
maximumheight is
h(c) = c
_
c
2
/g
2
+1/4c
2
/g
Provethat lim
c
h(c) = g/8.
43. Everylimit asx canberewrittenasaone-sidedlimit ast 0+, wheret = x
1
. Settingg(t ) = f (t
1
), we
have
lim
x
f (x) = lim
t 0+
g(t )
Showthat lim
x
3x
2
x
2x
2
+5
= lim
t 0+
3t
2+5t
2
, andevaluateusingtheQuotient Law.
solution Let t = x
1
. Thenx = t
1
, t 0+ asx , and
3x
2
x
2x
2
+5
=
3t
2
t
1
2t
2
+5
=
3t
2+5t
2
.
Thus,
lim
x
3x
2
x
2x
2
+5
= lim
t 0+
3t
2+5t
2
=
3
2
.
Rewritethefollowingasone-sidedlimitsasinExercise43andevaluate.
(a) lim
x
312x
3
4x
3
+3x +1
(b) lim
x
2
1/x
(c) lim
x
x sin
1
x
(d) lim
x
cos
_
x +1
x 1
_
45. Let G(b) = lim
x
(1+b
x
)
1/x
for b 0. InvestigateG(b) numerically and graphically for b = 0.2, 0.8, 2, 3, 5
(and additional values if necessary). Then makeaconjecturefor thevalueof G(b) as afunction of b. Drawagraph
of y = G(b). Does G(b) appear to becontinuous? Wewill evaluateG(b) using LHpitals Rulein Section 7.7(see
Exercise65inSection7.7).
S E C T I ON 2.8 Intermediate Value Theorem 73
solution
b = 0.2:
x 5 10 50 100
f (x) 1.000064 1.000000 1.000000 1.000000
It appearsthat G(0.2) = 1.
b = 0.8:
x 5 10 50 100
f (x) 1.058324 1.010251 1.000000 1.000000
It appearsthat G(0.8) = 1.
b = 2:
x 5 10 50 100
f (x) 2.012347 2.000195 2.000000 2.000000
It appearsthat G(2) = 2.
b = 3:
x 5 10 50 100
f (x) 3.002465 3.000005 3.000000 3.000000
b = 5:
x 5 10 50 100
f (x) 5.000320 5.000000 5.000000 5.000000
Basedontheseobservationsweconjecturethat G(b) = 1if 0 b 1andG(b) = b for b > 1. Thegraphof y = G(b)
isshownbelow; thegraphdoesappear tobecontinuous.
1
0
0 1 2 3 4
2
3
4
y
x
2.8 Intermediate Value Theorem
1. Provethat f (x) = x
2
takesonthevalue0.5intheinterval [0, 1].
solution Observethat f (x) = x
2
iscontinuouson[0, 1] withf (0) = 0andf (1) = 1. Becausef (0) < 0.5< f (1),
theIntermediateValueTheoremguaranteesthereisac [0, 1] suchthat f (c) = 0.5.
2. ThetemperatureinVancouver was8
Cat6am androseto20
Catnoon. Whichassumptionabouttemperatureallows
ustoconcludethat thetemperaturewas15
C at somemoment of timebetween6am andnoon?

solution Wemust assumethat temperatureisacontinuousfunctionof time.
3. What isthegraphical interpretationof theIVT?
solution If f iscontinuouson[a, b], thenthehorizontal liney = k for everykbetweenf (a) andf (b) intersectsthe
graphof y = f (x) at least once.
4. Showthat thefollowingstatement isfalsebydrawingagraphthat providesacounterexample:
If f (x) iscontinuousandhasaroot in[a, b], thenf (a) andf (b) haveoppositesigns.
solution
f(a)
f(b)
a
y
x
b
5. Assumethat f (t ) iscontinuouson[1, 5] andthat f (1) = 20, f (5) = 100. Determinewhether eachof thefollowing
statementsisalwaystrue, never true, or sometimestrue.
(a) f (c) = 3hasasolutionwithc [1, 5].
(b) f (c) = 75hasasolutionwithc [1, 5].
(c) f (c) = 50hasnosolutionwithc [1, 5].
(d) f (c) = 30hasexactlyonesolutionwithc [1, 5].
solution
(a) Thisstatement issometimestrue.
(b) Thisstatement isalwaystrue.
(c) Thisstatement isnever true.
(d) Thisstatement issometimestrue.
Exercises
1. UsetheIVT toshowthat f (x) = x
3
+x takesonthevalue9for somex in[1, 2].
solution Observethat f (1) = 2andf (2) = 10. Sincef isapolynomial, it iscontinuouseverywhere; inparticular
on[1, 2]. Therefore, bytheIVT thereisac [1, 2] suchthat f (c) = 9.
Showthat g(t ) =
t
t +1
takesonthevalue0.499for somet in[0, 1].
3. Showthat g(t ) = t
2
tant takesonthevalue
1
2
for somet in
_
0,

4
_
.
solution g(0) = 0andg(
4
) =

2
16
. g(t ) iscontinuousfor all t between0and
4
, and0<
1
2
<

2
16
; therefore, bythe
IVT, thereisac [0,

4
] suchthat g(c) =
1
2
.
Showthat f (x) =
x
2
x
7
+1
takesonthevalue0.4.
5. Showthat cosx = x hasasolutionintheinterval [0, 1]. Hint: Showthat f (x) = x cosx hasazeroin[0, 1].
solution Let f (x) = x cosx. Observethat f is continuous with f (0) = 1 and f (1) = 1 cos1 0.46.
Therefore, by theIVT thereis ac [0, 1] such that f (c) = c cosc = 0. Thus c = cosc and hencetheequation
cosx = x hasasolutionc in[0, 1].
UsetheIVT tondaninterval of length
1
2
containingaroot of f (x) = x
3
+2x +1.
InExercises716, proveusingtheIVT.
7.

c +
c +2= 3hasasolution.
solution Let f (x) =

x +
x +2 3. Notethat f is continuous on
_
1
4
, 2
_
withf (
1
4
) =
_
1
4
+
_
9
4
3 = 1
andf (2) =

21 0.41. Therefore, bytheIVT thereisac
_
1
4
, 2
_
suchthat f (c) =

c +
c +23= 0. Thus
c +
c +2= 3andhencetheequation
x +
x +2= 3hasasolutionc in
_
1
4
, 2
_
.
For all integersn, sinnx = cosx for somex [0, ].
9.

2exists. Hint: Consider f (x) = x
2
.
solution Letf (x) = x
2
. Observethatf iscontinuouswithf (1) = 1andf (2) = 4. Therefore, bytheIVT thereisa
c [1, 2] suchthat f (c) = c
2
= 2. Thisprovestheexistenceof
2, anumber whosesquareis2.
A positivenumber c hasannthroot for all positiveintegersn.
11. For all positiveintegersk, cosx = x
k
hasasolution.
solution For eachpositiveinteger k, letf (x) = x
k
cosx. Observethatf iscontinuouson
_
0,

2
_
withf (0) = 1
andf (
2
) =
_
2
_
k
> 0. Therefore, bytheIVT thereisac
_
0,

2
_
suchthat f (c) = c
k
cos(c) = 0. Thuscosc = c
k
andhencetheequationcosx = x
k
hasasolutionc intheinterval
_
0,

2
_
.
2
x
= bx hasasolutionif b > 2.
13. 2
x
+3
x
= 4
x
hasasolution.
solution Let f (x) = 2
x
+3
x
4
x
. Observethat f iscontinuouson[0, 2] withf (0) = 1> 0andf (2) = 3< 0.
Therefore, bytheIVT, thereisac (0, 2) suchthat f (c) = 2
c
+3
c
4
c
= 0.
tanx = x hasinnitelymanysolutions.
S E C T I ON 2.8 Intermediate Value Theorem 75
15. 2
x
+
1
x
= 4hasasolution.
x
+
1
x
+ 4. Observethat f iscontinuousfor x < 0withf (1) = 2
1
+
1
1
+ 4=
7
2
> 0
and f
_
1
8
_
= 2
1/8
8+ 4 3.08 < 0. Therefore, by the IVT, there is a c
_
1,
1
8
_
such that f (c) =
2
c
1
c
+4= 0andthus2
c
1
c
= 4.
x = sinx +cosx hasasolution.
17. Carryout threestepsof theBisectionMethodfor f (x) = 2
x
x
3
asfollows:
(a) Showthat f (x) hasazeroin[1, 1.5].
(b) Showthat f (x) hasazeroin[1.25, 1.5].
(c) Determinewhether [1.25, 1.375] or [1.375, 1.5] containsazero.
solution Notethat f (x) iscontinuousfor all x.
(a) f (1) = 1, f (1.5) = 2
1.5
(1.5)
3
< 33.375< 0. Hence, f (x) = 0for somex between1and1.5.
(b) f (1.25) 0.4253> 0andf (1.5) < 0. Hence, f (x) = 0for somex between1.25and1.5.
(c) f (1.375) 0.0059. Hence, f (x) = 0for somex between1.25and1.375.
Figure4showsthat f (x) = x
3
8x 1hasaroot intheinterval [2.75, 3]. ApplytheBisectionMethodtwice
tondaninterval of length
1
16
containingthisroot.
19. Findaninterval of length
1
4
in[1, 2] containingaroot of theequationx
7
+3x 10= 0.
solution Let f (x) = x
7
+ 3x 10. Observethat f is continuous with f (1) = 6 and f (2) = 124. Therefore,
by the IVT there is a c [1, 2] such that f (c) = 0. f (1.5) 11.59 > 0, so f (c) = 0 for some c [1, 1.5].
f (1.25) 1.48< 0, andsof (c) = 0for somec [1.25, 1.5]. Thismeansthat[1.25, 1.5] isaninterval of length0.25
containingaroot of f (x).
Showthattan
3
8tan
2
+17tan 8= 0hasarootin[0.5, 0.6]. ApplytheBisectionMethodtwicetond
aninterval of length0.025containingthisroot.
InExercises2124, drawthegraphof afunctionf (x) on[0, 4] withthegivenproperty.
21. J umpdiscontinuityat x = 2anddoesnot satisfytheconclusionof theIVT.
solution Thefunctiongraphedbelowhas ajumpdiscontinuity at x = 2. Notethat whilef (0) = 2andf (4) = 4,
thereisnopoint c intheinterval [0, 4] suchthat f (c) = 3. Accordingly, theconclusionof theIVT isnot satised.
4 3 2 1
1
3
2
4
y
x
J umpdiscontinuityat x = 2andsatisestheconclusionof theIVT on[0, 4].
23. Inniteone-sidedlimitsat x = 2anddoesnot satisfytheconclusionof theIVT.
solution Thefunctiongraphedbelowhasinniteone-sidedlimitsat x = 2. Notethat whilef (0) = 2andf (4) = 4,
thereisnopoint c intheinterval [0, 4] suchthat f (c) = 3. Accordingly, theconclusionof theIVT isnot satised.
4 3 2 1
1
1
3
2
4
5
6
y
x
Inniteone-sidedlimitsat x = 2andsatisestheconclusionof theIVT on[0, 4].
25. CanCorollary2beappliedtof (x) = x
1
on[1, 1]? Doesf (x) haveanyroots?
solution No, becausef (x) = x
1
isnot continuouson[1, 1]. Eventhoughf (1) = 1< 0andf (1) = 1> 0,
thefunctionhasnorootsbetweenx = 1andx = 1. Infact, thisfunctionhasnorootsat all.
Takeany mapanddrawacircleonit anywhere(Figure5). Provethat at any moment intimethereexistsapair
of diametrically oppositepoints A andB onthat circlecorrespondingto locations wherethetemperatures at that
momentareequal. Hint: Let beanangularcoordinatealongthecircleandletf () bethedifferenceintemperatures
at thelocationscorrespondingto and +.
27. Showthat if f (x) is continuous and 0 f (x) 1 for 0 x 1, then f (c) = c for somec in [0, 1]
(Figure6).
1
1
y = f(x)
y = x
c
x
y
FIGURE 6 A functionsatisfying0 f (x) 1for 0 x 1.
solution If f (0) = 0, the proof is done with c = 0. We may assume that f (0) > 0. Let g(x) = f (x) x.
g(0) = f (0) 0 = f (0) > 0. Sincef (x) is continuous, theRuleof Differences dictates that g(x) is continuous. We
needtoprovethat g(c) = 0for somec [0, 1]. Sincef (1) 1, g(1) = f (1) 1 0. If g(1) = 0, theproof isdone
withc = 1, soletsassumethat g(1) < 0.
Wenowhaveacontinuousfunctiong(x) ontheinterval [0, 1] suchthat g(0) > 0andg(1) < 0. FromtheIVT, there
must besomec [0, 1] sothat g(c) = 0, sof (c) c = 0andsof (c) = c.
Thisisasimplecaseof averygeneral, useful, andbeautiful theoremcalledtheBrouwer xedpointtheorem.
UsetheIVTtoshowthatif f (x) iscontinuousandone-to-oneonaninterval [a, b], thenf (x) iseitheranincreasing
or adecreasingfunction.
29. HamSandwichTheoremFigure7(A) shows asliceof ham. Provethat for any angle (0 ), it
is possibleto cut theslicein half with acut of incline. Hint: Thelines of inclination aregiven by theequations
y = (tan)x +b, whereb variesfromto. Eachsuchlinedividesthesliceintotwopieces(oneof whichmaybe
empty). Let A(b) betheamount of hamtotheleft of thelineminustheamount totheright, andlet A bethetotal areaof
theham. Showthat A(b) = A if b is sufciently largeandA(b) = A if b is sufciently negative. ThenusetheIVT.
Thisworksif = 0or

2
. If = 0, deneA(b) astheamount of hamabovetheliney = b minustheamount below.
Howcanyoumodifytheargument toworkwhen =

2
(inwhichcasetan = )?
Cutting asliceof ham
at an angle .
L(0) = L( )
L( )
L( )
2
(A) (B) A sliceof hamon top
of asliceof bread.
x
y
x
y
FIGURE 7
solution Let besuchthat =

2
. For anyb, consider thelineL() drawnat angle tothex axisstartingat (0, b).
Thislinehasformulay = (tan)x +b. Let A(b) betheamount of hamabovethelineminusthat belowtheline.
Let A > 0betheareaof theham. Wehavetoaccept thefollowing(reasonable) assumptions:
For lowenoughb = b
0
, thelineL() liesentirelybelowtheham, sothat A(b
0
) = A 0= A.
For highenoughb
1
, thelineL() liesentirelyabovetheham, sothat A(b
1
) = 0A = A.
A(b) iscontinuousasafunctionof b.
Under theseassumptions, weseeA(b) isacontinuousfunctionsatisfyingA(b
0
) > 0andA(b
1
) < 0for someb
0
< b
1
.
BytheIVT, A(b) = 0for someb [b
0
, b
1
].
Supposethat =

2
. Let thelineL(c) bethevertical linethrough(c, 0) (x = c). Let A(c) betheareaof hamtothe
left of L(c) minus that to theright of L(c). SinceL(0) lies entirely to theleft of theham, A(0) = 0 A = A. For
somec = c
1
sufciently large, L(c) liesentirely totheright of theham, sothat A(c
1
) = A 0 = A. HenceA(c) isa
continuousfunctionof c suchthat A(0) < 0andA(c
1
) > 0. BytheIVT, thereissomec [0, c
1
] suchthat A(c) = 0.
Figure7(B) showsasliceof hamonapieceof bread. Provethatitispossibletoslicethisopen-facedsandwich
sothat eachpart hasequal amountsof hamandbread. Hint: ByExercise29, for all 0 thereisalineL()
of incline (whichweassumeisunique) that dividesthehamintotwoequal pieces. Let B() denotetheamount of
breadtotheleftof (or above) L() minustheamounttotheright(or below). NoticethatL() andL(0) arethesame
line, but B() = B(0) sinceleft andright get interchangedas theanglemoves from0to . Assumethat B()
is continuous andapply theIVT. (By afurther extensionof this argument, onecanprovethefull HamSandwich
Theorem, whichstatesthat if youallowtheknifetocut at aslant, thenit ispossibletocut asandwichconsistingof
asliceof hamandtwoslicesof breadsothat all threelayersaredividedinhalf.)
2.9 The Formal Definition of a Limit
1. Giventhat lim
x0
cosx = 1, whichof thefollowingstatementsistrue?
(a) If |cosx 1| isverysmall, thenx iscloseto0.
(b) Thereisan > 0suchthat |x| < 10
5
if 0< |cosx 1| < .
S E C T I ON 2.9 The Formal Definition of a Limit 77
(c) Thereisa > 0suchthat |cosx 1| < 10
5
if 0< |x| < .
(d) Thereisa > 0suchthat |cosx| < 10
5
if 0< |x 1| < .
solution Thetruestatement is(c): Thereisa > 0suchthat |cosx 1| < 10
5
if 0< |x| < .
2. Supposeit isknownthat for agiven and, |f (x) 2| < if 0< |x 3| < . Whichof thefollowingstatements
must alsobetrue?
(a) |f (x) 2| < if 0< |x 3| < 2
(b) |f (x) 2| < 2 if 0< |x 3| <
(c) |f (x) 2| <
2
if 0< |x 3| <
2
(d) |f (x) 2| < if 0< |x 3| <
2
solution Statements(b)and(d)aretrue.
Exercises
1. BasedontheinformationconveyedinFigure5(A), ndvaluesof L, , and > 0suchthat thefollowingstatement
holds: |f (x) L| < if 0< |x| < .
3 3.1 2.9
10
10.4
9.8
x
y
y = f(x)
y = f(x)
(A) (B)
0.1 0.1
4
4.8
3.5
x
y
FIGURE 5
solution Wesee0.1 < x < 0.1forces 3.5 < f (x) < 4.8. Rewritten, this means that |x 0| < 0.1implies that
|f (x) 4| < 0.8. Replacingnumberswhereappropriateinthedenitionof thelimit|x c| < implies|f (x) L| < ,
weget L = 4, = 0.8, c = 0, and = 0.1.
Based on theinformation conveyed in Figure5(B), nd values of c, L, , and > 0 such that thefollowing
statement holds: |f (x) L| < if 0< |x c| < .
3. Consider lim
x4
f (x), wheref (x) = 8x +3.
(a) Showthat |f (x) 35| = 8|x 4|.
(b) Showthat for any > 0, |f (x) 35| < if 0< |x 4| < , where =

8
. Explainhowthisprovesrigorouslythat
lim
x4
f (x) = 35.
solution
(a) |f (x) 35| = |8x +335| = |8x 32| = |8(x 4)| = 8|x 4|. (Rememberthatthelaststepisjustiedbecause
8> 0).
(b) Let > 0. Let = /8andsuppose|x 4| < . By part (a), |f (x) 35| = 8|x 4| < 8. Substituting = /8,
wesee|f (x) 35| < 8/8= . Weseethat, for any > 0, wefound an appropriate so that |x 4| < implies
|f (x) 35| < . Hence lim
x4
f (x) = 35.
Consider lim
x2
f (x), wheref (x) = 4x 1.
(a) Showthat |f (x) 7| < 4 if 0< |x 2| < .
(b) Finda suchthat
|f (x) 7| < 0.01 if 0< |x 2| <
(c) Proverigorouslythat lim
x2
f (x) = 7.
5. Consider lim
x2
x
2
= 4(refer toExample2).
(a) Showthat |x
2
4| < 0.05if 0< |x 2| < 0.01.
(b) Showthat |x
2
4| < 0.0009if 0< |x 2| < 0.0002.
(c) Findavalueof suchthat |x
2
4| islessthan10
4
if
0< |x 2| < .
solution
(a) If 0< |x 2| < = 0.01, then|x| < 3and
x
2
4
= |x 2||x +2| |x 2| (|x| +2) < 5|x 2| < 0.05.

(b) If 0< |x 2| < = 0.0002, then|x| < 2.0002and
x
2
4
= |x 2||x +2| |x 2| (|x| +2) < 4.0002|x 2| < 0.00080004< 0.0009.

(c) Notethat
x
2
4
= |(x +2)(x 2)| |x +2| |x 2|. Since|x 2| can get arbitrarily small, wecan require
|x 2| < 1so that 1 < x < 3. This ensures that |x +2| is at most 5. Nowweknowthat
x
2
4
5|x 2|. Let

= 10
5
. Then, if |x 2| < , weget
x
2
4
5|x 2| < 510

5
< 10
4
asdesired.
Withregardtothelimit lim
x5
x
2
= 25,
(a) Showthat |x
2
25| < 11|x 5| if 4< x < 6. Hint: Write|x
2
25| = |x +5| |x 5|.
(b) Finda suchthat |x
2
25| < 10
3
if 0< |x 5| < .
(c) Givearigorousproof of thelimitbyshowingthat|x
2
25| < if 0< |x 5| < , where isthesmaller of

11
and1.
7. Refer toExample3tondavalueof > 0suchthat
1
x
1
3
< 10
4
if 0< |x 3| <
solution TheExampleshowsthat for any > 0wehave
1
x
1
3
if |x 3| <
where isthesmaller of thenumbers6 and1. Inour case, wemaytake = 610
4
.
UseFigure6tondavalueof > 0suchthat thefollowingstatement holds:
1/x
2
1
4
< if 0< |x 2| <

for = 0.03. Thenndavalueof that worksfor = 0.01.
9. Plot f (x) =

2x 1together withthehorizontal linesy = 2.9andy = 3.1. Usethisplot tondavalueof
> 0suchthat |
2x 13| < 0.1if 0< |x 5| < .

solution Fromtheplot below, weseethat = 0.25will guaranteethat |
2x 13| < 0.1whenever |x 5| .

4.6 4.8 5 5.2 5.4
2.9
2.8
3
3.1
x
y
Plot f (x) = tanx together withthehorizontal lines y = 0.99andy = 1.01. Usethis plot tondavalueof
> 0suchthat |tanx 1| < 0.01if 0<
x

4
< .
11. The function f (x) = 2
x
2
satises lim
x0
f (x) = 1. Use a plot of f to nd a value of > 0 such that
|f (x) 1| < 0.001if 0< |x| < .
solution Fromtheplot below, weseethat = 0.03will guaranteethat
2
x
2
1
< 0.001
whenever 0< |x| < .
0.9985
0.04 0.02
0
0.02 0.04
x
y
0.9990
0.9995
1.0000
Let f (x) =
4
x
2
+1
and = 0.5. Usingaplot of f (x), ndavalueof > 0suchthat
f (x)
16
5
< for
0<
x
1
2
< . Repeat for = 0.2and0.1.

13. Consider lim
x2
1
x
.
(a) Showthat if |x 2| < 1, then
1
x
1
2
<
1
2
|x 2|
(b) Let bethesmaller of 1and2. Prove:
1
x
1
2
< if 0< |x 2| <

(c) Finda > 0suchthat
1
x

1
2
< 0.01if 0< |x 2| < .

(d) Proverigorouslythat lim
x2
1
x
=
1
2
.
solution
(a) Since|x 2| < 1, it followsthat 1< x < 3, inparticular that x > 1. Becausex > 1, then
1
x
< 1and
1
x
1
2
2x
2x
=
|x 2|
2x
<
1
2
|x 2|.
(b) Let = min{1, 2} andsupposethat |x 2| < . Thenbypart (a) wehave
1
x
1
2
<
1
2
|x 2| <
1
2
<
1
2
2 = .
(c) Choose = 0.02. Then
1
x
1
2
<
1
2
= 0.01bypart (b).
(d) Let > 0begiven. Thenwhenever 0< |x 2| < = min{1, 2}, wehave
1
x
1
2
<
1
2
.
Since wasarbitrary, weconcludethat lim
x2
1
x
=
1
2
.
Consider lim
x1
x +3.
(a) Showthat |
x +32| <
1
2
|x 1| if |x 1| < 4. Hint: Multiplytheinequalityby|
x +3+2| andobserve
that |
x +3+2| > 2.
(b) Find > 0suchthat |
x +32| < 10
4
for 0< |x 1| < .
(c) Proverigorouslythat thelimit isequal to2.
15. Let f (x) = sinx. Usingacalculator, wend:
f
_
4
0.1
_
0.633, f
_
4
_
0.707, f
_
4
+0.1
_
0.774
Usethesevaluesandthefact that f (x) isincreasingon
_
0,

2
_
tojustifythestatement
f (x) f
_
4
_
< 0.08 if 0<
x

4
< 0.1
ThendrawagurelikeFigure3toillustratethisstatement.
solution Sincef (x) isincreasingontheinterval, thethreef (x) valuestell usthat 0.633 f (x) 0.774for all x
between
4
0.1and
4
+0.1. Wemaysubtract f (
4
) fromtheinequalityfor f (x). Thisshowthat, for

4
0.1< x <
4
+ 0.1, 0.633 f (
4
) f (x) f (
4
) 0.774 f (
4
). Thismeansthat, if |x

4
| < 0.1, then0.633 0.707
f (x) f (
4
) 0.774 0.707, so 0.074 f (x) f (
4
) 0.067. Then0.08 < f (x) f (
4
) < 0.08follows
fromthis, so|x

4
| < 0.1implies|f (x) f (
4
)| < 0.08. Thegurebelowillustratesthis.
0.25 0.5 0.75 1 1.25 1.5
1
0.8
0.6
0.4
0.2
x
y
Adapt theargument inExample1toproverigorouslythat lim
xc
(ax +b) = ac +b, wherea, b, c arearbitrary.
17. Adapt theargument inExample2toproverigorouslythat lim
xc
x
2
= c
2
for all c.
solution Torelatethegapto|x c|, wetake
x
2
c
2
= |(x +c)(x c)| = |x +c| |x c| .

Wechoose intwosteps. First, sincewearerequiring|x c| tobesmall, werequire < |c|, sothat x liesbetween0
and2c. Thismeansthat |x +c| < 3|c|, so|x c||x +c| < 3|c|. Next, werequirethat <
3|c|
, so
|x c||x +c| <
3|c|
3|c| = ,
andwearedone.
Therefore, given > 0, welet
= min
_
|c|,
3|c|
_
.
Then, for |x c| < , wehave
|x
2
c
2
| = |x c| |x +c| < 3|c| < 3|c|

3|c|
= .
Adapt theargument inExample3toproverigorouslythat lim
xc
x
1
=
1
c
for all c = 0.
InExercises1924, usetheformal denitionof thelimit toprovethestatement rigorously.
19. lim
x4
x = 2
solution Let > 0begiven. Webound|
x 2| bymultiplying
x +2
x +2
.
|
x 2| =
x 2
_
x +2
x +2
_
x 4
x +2
= |x 4|
x +2
.
Wecanassume < 1, sothat |x 4| < 1, andhence
x +2>
3+2> 3. Thisgivesus
|
x 2| = |x 4|
x +2
< |x 4|
1
3
.
Let = min(1, 3). If |x 4| < ,
|
x 2| = |x 4|
x +2
< |x 4|
1
3
<
1
3
< 3
1
3
= ,
thusprovingthelimit rigorously.
lim
x1
(3x
2
+x) = 4
21. lim
x1
x
3
= 1
solution Let > 0begiven. Webound
x
3
1
byfactoringthedifferenceof cubes:
x
3
1
(x
2
+x +1)(x 1)
= |x 1|
x
2
+x +1
.
Let = min(1,

7
), andassume|x 1| < . Since < 1, 0 < x < 2. Sincex
2
+ x + 1increases as x increases for
x > 0, x
2
+x +1< 7for 0< x < 2, andso
x
3
1
= |x 1|
x
2
+x +1
< 7|x 1| < 7
7
=
andthelimit isrigorouslyproven.
lim
x0
(x
2
+x
3
) = 0
23. lim
x2
x
2
=
1
4
solution Let > 0begiven. First, weboundx
2
1
4
:
x
2
1
4
4x
2
4x
2
= |2x|
2+x
4x
2
.
Let = min(1,
4
5
), andsuppose|x 2| < . Since < 1, |x 2| < 1, so1 < x < 3. Thismeansthat 4x
2
> 4and
|2+x| < 5, sothat
2+x
4x
2
<
5
4
. Weget:
x
2
1
4
= |2x|
2+x
4x
2
<
5
4
|x 2| <
5
4
4
5
= .
andthelimit isrigorouslyproven.
lim
x0
x sin
1
x
= 0
25. Letf (x) =
x
|x|
. Proverigorouslythat lim
x0
f (x) doesnotexist. Hint: Showthatfor anyL, therealwaysexistssome
x suchthat |x| < but |f (x) L|
1
2
, nomatter howsmall istaken.
solution Let L beanyreal number. Let > 0beanysmall positivenumber. Let x =

2
, whichsatises|x| < , and
f (x) = 1. Weconsider twocases:
(|f (x) L|
1
2
) : wearedone.
(|f (x) L| <
1
2
): Thismeans
1
2
< L <
3
2
. Inthiscase, let x =
2
. f (x) = 1, andso
3
2
< L f (x).
Ineither case, thereexistsanx suchthat |x| <

2
, but |f (x) L|
1
2
.
Proverigorouslythat lim
x0
|x| = 0.
27. Let f (x) = min(x, x
2
), wheremin(a, b) istheminimumof a andb. Proverigorouslythat lim
x1
f (x) = 1.
solution Let > 0andlet = min(1,

2
). Then, whenever |x 1| < , it follows that 0 < x < 2. If 1 < x < 2,
thenmin(x, x
2
) = x and
|f (x) 1| = |x 1| < <
2
< .
Ontheother hand, if 0< x < 1, thenmin(x, x
2
) = x
2
, |x +1| < 2and
|f (x) 1| = |x
2
1| = |x 1| |x +1| < 2 < .
Thus, whenever |x 1| < , |f (x) 1| < .
x0
sin
1
x
doesnot exist.
29. First, usetheidentity
sinx +siny = 2sin
_
x +y
2
_
cos
_
x y
2
_
toverifytherelation
sin(a +h) sina = h
sin(h/2)
h/2
cos
_
a +
h
2
_
6
Thenusetheinequality
sinx
x
1for x = 0toshowthat |sin(a +h) sina| < |h| for all a. Finally, proverigorously
that lim
xa
sinx = sina.
solution Werst write
sin(a +h) sina = sin(a +h) +sin(a).
Applyingtheidentitywithx = a +h, y = a, yields:
sin(a +h) sina = sin(a +h) +sin(a) = 2sin
_
a +h a
2
_
cos
_
2a +h
2
_
= 2sin
_
h
2
_
cos
_
a +
h
2
_
= 2
_
h
h
_
sin
_
h
2
_
cos
_
a +
h
2
_
= h
sin(h/2)
h/2
cos
_
a +
h
2
_
.
Therefore,
|sin(a +h) sina| = |h|
sin(h/2)
h/2
cos
_
a +
h
2
_
.
Usingthefact that
sin
< 1andthat |cos| 1, andmakingthesubstitutionh = x a, weseethat thislast relation

isequivalent to
|sinx sina| < |x a|.
Now, toprovethedesiredlimit, let > 0, andtake = . If |x a| < , then
|sinx sina| < |x a| < = ,
Therefore, a wasfoundfor arbitrary, andtheproof iscomplete.
Uniquenessof theLimit Provethat afunctionconvergestoat most onelimitingvalue. Inother words, usethe
limit denitiontoprovethat if lim
xc
f (x) = L
1
and lim
xc
f (x) = L
2
, thenL
1
= L
2
.
InExercises3133, provethestatement usingtheformal limit denition.
31. TheConstant MultipleLaw[Theorem1, part (ii) inSection2.3, p. 58]
solution Supposethat lim
xc
f (x) = L. Wewishtoprovethat lim
xc
af (x) = aL.
Let > 0 begiven. /|a| is also a positivenumber. Since lim
xc
f (x) = L, weknow thereis a > 0 such that
|x c| < forces|f (x) L| < /|a|. Suppose|x c| < . |af (x) aL| = |a||f (x) aL| < |a|(/|a|) = , sothe
ruleisproven.
TheSqueezeTheorem. (Theorem1inSection2.6, p. 77)
33. TheProduct Law[Theorem1, part (iii) inSection2.3, p. 58]. Hint: Usetheidentity
f (x)g(x) LM = (f (x) L) g(x) +L(g(x) M)
solution Before we can prove the Product Law, we need to establish one preliminary result. We are given that
lim
xc
g(x) = M. Consequently, if weset = 1, thenthedenitionof alimit guarantees theexistenceof a
1
> 0
suchthat whenever 0< |x c| <
1
, |g(x) M| < 1. Applyingtheinequality|g(x)| |M| |g(x) M|, it follows
that |g(x)| < 1+ |M|. Inother words, becauselim
xc
g(x) = M, thereexists a
1
> 0suchthat |g(x)| < 1+ |M|
whenever 0< |x c| <
1
.
WecannowprovetheProduct Law. Let > 0. Asprovenabove, becauselim
xc
g(x) = M, thereexistsa
1
> 0
suchthat |g(x)| < 1+ |M| whenever 0 < |x c| <
1
. Furthermore, by thedenitionof alimit, lim
xc
g(x) = M
impliesthereexistsa
2
> 0suchthat |g(x) M| <

2(1+|L|)
whenever 0< |x c| <
2
. Wehaveincludedthe1+
inthedenominator toavoiddivisionbyzeroincaseL = 0. Thereasonfor includingthefactor of 2inthedenominator
will becomeclear shortly. Finally, becauselim
xc
f (x) = L, thereexists a
3
> 0suchthat |f (x) L| <

2(1+|M|)
whenever 0< |x c| <
3
. Now, let = min(
1
,
2
,
3
). Then, for all x satisfying0< |x c| < , wehave
|f (x)g(x) LM| = |(f (x) L)g(x) +L(g(x) M)|
|f (x) L| |g(x)| +|L| |g(x) M|
<
2(1+|M|)
(1+|M|) +|L|

2(1+|L|)
<
2
+

2
= .
Hence,
lim
xc
f (x)g(x) = LM = lim
xc
f (x) lim
xc
g(x).
Let f (x) = 1if x isrational andf (x) = 0if x isirrational. Provethat lim
xc
f (x) doesnot exist for anyc.
35. Hereisafunctionwithstrangecontinuityproperties:
f (x) =
1
q
if x istherational number p/q in
lowest terms
0 if x isanirrational number
(a) Showthat f (x) isdiscontinuousat c if c isrational. Hint: Thereexist irrational numbersarbitrarilycloseto c.
(b) Showthat f (x) iscontinuousat c if c isirrational. Hint: Let I betheinterval {x : |x c| < 1}. Showthat for any
Q > 0, I containsat most nitely many fractionsp/q withq < Q. Concludethat thereisa suchthat all fractionsin
{x : |x c| < } haveadenominator larger thanQ.
solution
(a) Let c beanyrational number andsupposethat, inlowest terms, c = p/q, wherep andq areintegers. Toprovethe
discontinuityof f at c, wemust showthereisan > 0suchthat for any > 0thereisanx for which|x c| < , but
that|f (x) f (c)| > . Let =
1
2q
and > 0. Sincethereisatleastoneirrational number betweenanytwodistinctreal
numbers, thereissomeirrational x betweenc andc +. Hence, |x c| < , but|f (x) f (c)| = |0
1
q
| =
1
q
>
1
2q
= .
(b) Let c beirrational, let > 0 begiven, and let N > 0 beaprimeinteger sufciently largeso that
1
N
< . Let
p
1
q
1
, . . . ,
p
m
q
m
beall rational numbers
p
q
inlowest terms suchthat |
p
q
c| < 1andq < N. SinceN is nite, this is a
nitelist; hence, onenumber
p
i
q
i
inthelist must beclosest toc. Let =
1
2
|
p
i
q
i
c|. Byconstruction, |
p
i
q
i
c| > for all
i = 1. . . m. Therefore, for anyrational number
p
q
suchthat |
p
q
c| < , q > N, so
1
q
<
1
N
< .
Therefore, for anyrational number x suchthat |x c| < , |f (x) f (c)| < . |f (x) f (c)| = 0for anyirrational
number x, so|x c| < impliesthat |f (x) f (c)| < for anynumber x.
1. Thepositionof aparticleat timet (s) iss(t ) =
_
t
2
+1m. Computeitsaveragevelocityover [2, 5] andestimateits
instantaneousvelocityat t = 2.
solution Let s(t ) =
_
t
2
+1. Theaveragevelocityover [2, 5] is
s(5) s(2)
52
=
26
5
3
0.954m/s.
Fromthedatainthetablebelow, weestimatethat theinstantaneousvelocityat t = 2isapproximately0.894m/s.
interval [1.9, 2] [1.99, 2] [1.999, 2] [2, 2.001] [2, 2.01] [2, 2.1]
averageROC 0.889769 0.893978 0.894382 0.894472 0.894873 0.898727
Thewellhead pricep of natural gasintheUnitedStates(indollarsper 1000ft
3
) ontherst dayof eachmonth
in2008islistedinthetablebelow.
J F M A M J
6.99 7.55 8.29 8.94 9.81 10.82
J A S O N D
10.62 8.32 7.27 6.36 5.97 5.87
Computetheaveragerateof changeof p (indollarsper1000ft
3
permonth)overthequarterlyperiodsJ anuaryMarch,
AprilJ une, andJ ulySeptember.
3. For awholenumber n, let P(n) bethenumber of partitionsof n, that is, thenumber of waysof writingn asasum
of oneor morewholenumbers. For example, P(4) = 5sincethenumber 4canbepartitionedinvedifferent ways: 4,
3+1, 2+2, 2+1+1, and1+1+1+1. TreatingP(n) asacontinuousfunction, useFigure1toestimatetherateof
changeof P(n) at n = 12.
n
P(n)
14 12 10 8 6 4 2 0
0
40
80
120
160
FIGURE 1 Graphof P(n).
solution Thetangentlinedrawninthegureappearstopassthroughthepoints(15, 140) and(10.5, 40). Wetherefore
estimatethat therateof changeof P(n) at n = 12is
14040
1510.5
=
100
4.5
=
200
9
.
Theaveragevelocity v (m/s) of anoxygenmoleculeintheair at temperatureT (
C) isv = 25.7
273.15+T .
What is theaveragespeedat T = 25
(roomtemperature)? Estimatetherateof changeof averagevelocity with

respect totemperatureat T = 25
. What aretheunitsof thisrate?

InExercises510, estimatethelimit numericallytotwodecimal placesor statethat thelimit doesnot exist.
5. lim
x0
1cos
3
(x)
x
2
1cos
3
x
x
2
. Thedatainthetablebelowsuggeststhat
lim
x0
1cos
3
x
x
2
1.50.
Inconstructingthetable, wetakeadvantageof thefact that f isanevenfunction.
x 0.001 0.01 0.1
f (x) 1.500000 1.499912 1.491275
(Theexact valueis
3
2
.)
lim
x1
x
1/(x1)
7. lim
x2
x
x
4
x
2
4
x
x
4
x
2
4
lim
x2
x
x
4
x
2
4
1.69.
x 1.9 1.99 1.999 2.001 2.01 2.1
f (x) 1.575461 1.680633 1.691888 1.694408 1.705836 1.828386
(Theexact valueis1+ln2.)
lim
x2
x 2
2
x
4
9. lim
x1
_
7
1x
7

3
1x
3
_
_
7
1x
7

3
1x
3
_
lim
x1
_
7
1x
7

3
1x
3
_
2.00.
x 0.9 0.99 0.999 1.001 1.01 1.1
f (x) 2.347483 2.033498 2.003335 1.996668 1.966835 1.685059
(Theexact valueis2.)
lim
x2
3
x
9
5
x
25
InExercises1150, evaluatethelimit if it exists. If not, determinewhether theone-sidedlimitsexist (niteor innite).
11. lim
x4
(3+x
1/2
)
solution lim
x4
(3+x
1/2
) = 3+
4= 5.
lim
x1
5x
2
4x +7
13. lim
x2
4
x
3
solution lim
x2
4
x
3
=
4
(2)
3
=
1
2
.
lim
x1
3x
2
+4x +1
x +1
15. lim
t 9
t 3
t 9
solution lim
t 9
t 3
t 9
= lim
t 9
t 3
(
t 3)(
t +3)
= lim
t 9
1
t +3
=
1
9+3
=
1
6
.
lim
x3
x +12
x 3
17. lim
x1
x
3
x
x 1
solution lim
x1
x
3
x
x 1
= lim
x1
x(x 1)(x +1)
x 1
= lim
x1
x(x +1) = 1(1+1) = 2.
lim
h0
2(a +h)
2
2a
2
h
19. lim
t 9
t 6
t 3
solution Becausetheone-sidedlimits
lim
t 9
t 6
t 3
= and lim
t 9+
t 6
t 3
= ,
arenot equal, thetwo-sidedlimit
lim
t 9
t 6
t 3
doesnot exist.
lim
s0
1
_
s
2
+1
s
2
21. lim
x1+
1
x +1
solution For x > 1, x +1> 0. Therefore,
lim
x1+
1
x +1
= .
lim
y
1
3
3y
2
+5y 2
6y
2
5y +1
23. lim
x1
x
3
2x
x 1
lim
x1
x
3
2x
x 1
= and lim
x1+
x
3
2x
x 1
= ,
lim
x1
x
3
2x
x 1
doesnot exist.
lim
ab
a
2
3ab +2b
2
a b
25. lim
x0
4
3x
4
x
4
x
1
solution
lim
x0
4
3x
4
x
4
x
1
= lim
x0
4
x
(4
x
1)(4
x
+1)
4
x
1
= lim
x0
4
x
(4
x
+1) = 1 2= 2.
lim
0
sin5
27. lim
x1.5
[x]
x
solution lim
x1.5
[x]
x
=
[1.5]
1.5
=
1
1.5
=
2
3
.
lim
4
sec 29. lim
z3
z +3
z
2
+4z +3
solution
lim
z3
z +3
z
2
+4z +3
= lim
z3
z +3
(z +3)(z +1)
= lim
z3
1
z +1
=
1
2
.
lim
x1
x
3
ax
2
+ax 1
x 1
31. lim
xb
x
3
b
3
x b
solution lim
xb
x
3
b
3
x b
= lim
xb
(x b)(x
2
+xb +b
2
)
x b
= lim
xb
(x
2
+xb +b
2
) = b
2
+b(b) +b
2
= 3b
2
.
lim
x0
sin4x
sin3x
33. lim
x0
_
1
3x
1
x(x +3)
_
solution lim
x0
_
1
3x
1
x(x +3)
_
= lim
x0
(x +3) 3
3x(x +3)
= lim
x0
1
3(x +3)
=
1
3(0+3)
=
1
9
.
lim
1
4
3
tan() 35. lim
x0
[x]
x
solution For x sufcientlyclosetozerobut negative, [x] = 1. Therefore,
lim
x0
[x]
x
= lim
x0
1
x
= .
lim
x0+
[x]
x
37. lim
2
sec
lim
sec = and lim
2
+
sec =
lim
2
sec doesnot exist.
lim
y3
_
sin
y
_
1/2 39. lim
0
cos 2
lim
0
cos 2
= and lim
0+
cos 2
=
lim
0
cos 2
doesnot exist.
lim
x4.3
1
x [x]
41. lim
x2
x 3
x 2
solution For x closeto2but lessthan2, x 3< 0andx 2< 0. Therefore,
lim
x2
x 3
x 2
= .
lim
t 0
sin
2
t
t
3
43. lim
x1+
_
1
x 1
1
_
x
2
1
_
solution lim
x1+
_
1
x 1
1
_
x
2
1
_
= lim
x1+
x +11
_
x
2
1
= .
lim
t
2t (cost 1)
45. lim
x
2
tanx
lim
x
tanx = and lim

x
2
+
tanx =
lim
x
2
tanx doesnot exist.
lim
t 0
cos
1
t
47. lim
t 0+
t cos
1
t
solution For t > 0,
1 cos
_
1
t
_
1,
so
t

t cos
_
1
t
_

t .
Because
lim
t 0+
t = lim
t 0+
t = 0,
lim
t 0+
t cos
_
1
t
_
= 0.
lim
x5+
x
2
24
x
2
25
49. lim
x0
cosx 1
sinx
solution
lim
x0
cosx 1
sinx
= lim
x0
cosx 1
sinx
cosx +1
cosx +1
= lim
x0
sin
2
x
sinx(cosx +1)
= lim
x0
sinx
cosx +1
=
0
1+1
= 0.
lim
0
tan sin
sin
3
51. Findtheleft- andright-handlimits of thefunctionf (x) inFigure2at x = 0, 2, 4. Statewhether f (x) is left- or

right-continuous(or both) at thesepoints.
x
y
1 3 5 2 4
1
2
FIGURE 2
solution Accordingtothegraphof f (x),
lim
x0
f (x) = lim
x0+
f (x) = 1
lim
x2
f (x) = lim
x2+
f (x) =
lim
x4
f (x) =
lim
x4+
f (x) = .
Thefunctionisbothleft- andright-continuousat x = 0andneither left- nor right-continuousat x = 2andx = 4.
Sketchthegraphof afunctionf (x) suchthat
(a) lim
x2
f (x) = 1, lim
x2+
f (x) = 3
(b) lim
x4
f (x) existsbut doesnot equal f (4).
53. Graphh(x) anddescribethediscontinuity:
h(x) =
_
2
x
for x 0
x
1/2
for x > 0
Ish(x) left- or right-continuous?
solution Thegraphof h(x) isshownbelow.Atx = 0, thefunctionhasaninnitediscontinuitybutisleft-continuous.
2
1
0
1
4 3 2 1 1 2 3 4
x
y
Sketchthegraphof afunctiong(x) suchthat
lim
x3
g(x) = , lim
x3+
g(x) = , lim
x4
g(x) =
55. Findthepointsof discontinuityof
g(x) =
cos
_
x
2
_
for |x| < 1
|x 1| for |x| 1
Determinethetypeof discontinuityandwhether g(x) isleft- or right-continuous.
solution First notethat cos
_
x
2
_
iscontinuousfor 1 < x < 1andthat |x 1| iscontinuousfor x 1andfor
x 1. Thus, theonlypointsat whichg(x) might bediscontinuousarex = 1. At x = 1, wehave
lim
x1
g(x) = lim
x1
cos
_
x
2
_
= cos
_
2
_
= 0
and
lim
x1+
g(x) = lim
x1+
|x 1| = |11| = 0,
sog(x) iscontinuousat x = 1. Ontheother hand, at x = 1,
lim
x1+
g(x) = lim
x1+
cos
_
x
2
_
= cos
_
2
_
= 0
and
lim
x1
g(x) = lim
x1
|x 1| = | 11| = 2,
sog(x) hasajumpdiscontinuityat x = 1. Sinceg(1) = 2, g(x) isleft-continuousat x = 1.
Showthat f (x) = x 2
sinx
iscontinuousonitsdomain.
57. Findaconstant b suchthat h(x) iscontinuousat x = 2, where
h(x) =
_
x +1 for |x| < 2
b x
2
for |x| 2
Withthischoiceof b, ndall pointsof discontinuity.
solution Tomakeh(x) continuousatx = 2, wemusthavethetwoone-sidedlimitsasx approaches2beequal. With
lim
x2
h(x) = lim
x2
(x +1) = 2+1= 3
and
lim
x2+
h(x) = lim
x2+
(b x
2
) = b 4,
itfollowsthatwemustchooseb = 7. Becausex +1iscontinuousfor 2< x < 2and7x
2
iscontinuousfor x 2
andfor x 2, theonlypossiblepoint of discontinuityisx = 2. At x = 2,
lim
x2+
h(x) = lim
x2+
(x +1) = 2+1= 1
and
lim
x2
h(x) = lim
x2
(7x
2
) = 7(2)
2
= 3,
soh(x) hasajumpdiscontinuityat x = 2.
InExercises5863, ndthehorizontal asymptotesof thefunctionbycomputingthelimitsat innity.
f (x) =
9x
2
4
2x
2
x
59. f (x) =
x
2
3x
4
x 1
solution Because
lim
x
x
2
3x
4
x 1
= lim
x
1/x
2
3
1/x
3
1/x
4
=
and
lim
x
x
2
3x
4
x 1
= lim
x
1/x
2
3
1/x
3
1/x
4
= ,
it followsthat thegraphof y =
x
2
3x
4
x 1
doesnot haveanyhorizontal asymptotes.
f (u) =
8u 3
_
16u
2
+6
61. f (u) =
2u
2
1
_
6+u
4
solution Because
lim
u
2u
2
1
_
6+u
4
= lim
u
21/u
2
_
6/u
4
+1
=
2
1
= 2
and
lim
u
2u
2
1
_
6+u
4
= lim
u
21/u
2
_
6/u
4
+1
=
2
1
= 2,
2u
2
1
_
6+u
4
hasahorizontal asymptoteof y = 2.
f (x) =
3x
2/3
+9x
3/7
7x
4/5
4x
1/3
63. f (t ) =
t
1/3
t
1/3
(t t
1
)
1/3
solution Because
lim
t
t
1/3
t
1/3
(t t
1
)
1/3
= lim
t
1t
2/3
(1t
2
)
1/3
=
1
1
1/3
= 1
and
lim
t
t
1/3
t
1/3
(t t
1
)
1/3
= lim
t
1t
2/3
(1t
2
)
1/3
=
1
1
1/3
= 1,
t
1/3
t
1/3
(t t
1
)
1/3
hasahorizontal asymptoteof y = 1.
Calculate(a)(d), assumingthat
lim
x3
f (x) = 6, lim
x3
g(x) = 4
(a) lim
x3
(f (x) 2g(x)) (b) lim
x3
x
2
f (x)
(c) lim
x3
f (x)
g(x) +x
(d) lim
x3
(2g(x)
3
g(x)
3/2
)
65. Assumethat thefollowinglimitsexist:
A = lim
xa
f (x), B = lim
xa
g(x), L = lim
xa
f (x)
g(x)
Provethat if L = 1, thenA = B. Hint: Youcannot usetheQuotient Lawif B = 0, soapplytheProduct LawtoL andB
instead.
solution SupposethelimitsA, B, andL all exist andL = 1. Then
B = B 1= B L = lim
xa
g(x) lim
xa
f (x)
g(x)
= lim
xa
g(x)
f (x)
g(x)
= lim
xa
f (x) = A.
Deneg(t ) = (1+2
1/t
)
1
for t = 0. Howshouldg(0) bedenedtomakeg(t ) left-continuousat t = 0?
67. Inthenotationof Exercise65, giveanexamplewhereL existsbut neither A nor B exists.
solution Suppose
f (x) =
1
(x a)
3
and g(x) =
1
(x a)
5
.
Then, neither A nor B exists, but
L = lim
xa
(x a)
3
(x a)
5
= lim
xa
(x a)
2
= 0.
Trueor false?
(a) If lim
x3
f (x) exists, then lim
x3
f (x) = f (3).
(b) If lim
x0
f (x)
x
= 1, thenf (0) = 0.
(c) If lim
x7
f (x) = 8, then lim
x7
1
f (x)
=
1
8
.
(d) If lim
x5+
f (x) = 4and lim
x5
f (x) = 8, then lim
x5
f (x) = 6.
(e) If lim
x0
f (x)
x
= 1, then lim
x0
f (x) = 0.
(f) If lim
x5
f (x) = 2, then lim
x5
f (x)
3
= 8.
69. Let f (x) = x
_
1
x
_
, where[x] isthegreatest integer function. Showthat for x = 0,
1
x
1<
_
1
x
_

1
x
ThenusetheSqueezeTheoremtoprovethat
lim
x0
x
_
1
x
_
= 1
Hint: Treat theone-sidedlimitsseparately.
solution Lety beanyreal number. Fromthedenitionof thegreatestintegerfunction, itfollowsthaty 1< [y] y,
withequalityholdingif andonlyif y isaninteger. If x = 0, then
1
x
isareal number, so
1
x
1<
_
1
x
_

1
x
.
Uponmultiplyingthisinequalitythroughbyx, wend
1x < x
_
1
x
_
1.
Because
lim
x0
(1x) = lim
x0
1= 1,
lim
x0
x
_
1
x
_
= 1.
Let r
1
and r
2
be the roots of f (x) = ax
2
2x + 20. Observe that f (x) approaches the linear function
L(x) = 2x +20asa 0. Becauser = 10istheuniqueroot of L(x), wemight expect oneof therootsof f (x)
toapproach10asa 0(Figure3). Provethat therootscanbelabeledsothat lim
a0
r
1
= 10and lim
a0
r
2
= .
71. UsetheIVT toprovethat thecurvesy = x
2
andy = cosx intersect.
2
cosx. Notethatanyrootof f (x) correspondstoapointof intersectionbetweenthecurves
y = x
2
and y = cosx. Now, f (x) is continuous over theinterval [0,

2
], f (0) = 1 < 0 and f (
2
) =

2
4
> 0.
Therefore, by theIntermediateValueTheorem, thereexists ac (0,

2
) suchthat f (c) = 0; consequently, thecurves
y = x
2
andy = cosx intersect.
UsetheIVT toprovethat f (x) = x
3
x
2
+2
cosx +2
hasaroot intheinterval [0, 2].
73. UsetheIVT toshowthat 2
x
2
= x hasasolutionon(0, 1).
x
2
x. Observethat f is continuous on[0, 1] withf (0) = 2
0
0 = 1 > 0andf (1) =
2
1
1< 0. Therefore, theIVT guaranteesthereexistsac (0, 1) suchthat f (c) = 2
c
2
c = 0.
UsetheBisectionMethodtolocateasolutionof x
2
7= 0totwodecimal places.
75. Giveanexampleof a(discontinuous) functionthat does not satisfy theconclusionof theIVT on[1, 1].
Thenshowthat thefunction
f (x) =
sin
1
x
x = 0
0 x = 0
satisestheconclusionof theIVT oneveryinterval [a, a], eventhoughf isdiscontinuousat x = 0.
solution Let g(x) = [x]. Thisfunctionisdiscontinuouson[1, 1] withg(1) = 1andg(1) = 1. For all c = 0,
thereisnox suchthatg(x) = c; thus, g(x) doesnotsatisfytheconclusionof theIntermediateValueTheoremon[1, 1].
Now, let
f (x) =
_
sin
_
1
x
_
for x = 0
0 for x = 0
andlet a > 0. Ontheinterval
x
_
a
2+2a
,
a
2
_
[a, a],
1
x
runsfrom
2
a
to
2
a
+2, sothesinefunctioncoversonefull periodandclearlytakesoneveryvaluefromsina through
sina.
Let f (x) =
1
x +2
.
(a) Showthat
f (x)
1
4
<
|x 2|
12
if |x 2| < 1. Hint: Observethat |4(x +2)| > 12if |x 2| < 1.
(b) Find > 0suchthat
f (x)
1
4
< 0.01for |x 2| < .

(c) Proverigorouslythat lim
x2
f (x) =
1
4
.
77. Plot the function f (x) = x
1/3
. Use the zoomfeature to nd a > 0 such that |x
1/3
2| < 0.05 for
|x 8| < .
solution Thegraphsof y = f (x) = x
1/3
andthehorizontal linesy = 1.95andy = 2.05areshownbelow. From
thisplot, weseethat = 0.55guaranteesthat |x
1/3
2| < 0.05whenever |x 8| < .
7 7.5 8 8.5
1.95
1.9
2
2.05
x
y
Usethefact that f (x) = 2
x
is increasingtondavalueof suchthat |2
x
8| < 0.001if |x 2| < . Hint:
Findc
1
andc
2
suchthat 7.999< f (c
1
) < f (c
2
) < 8.001.
79. Proverigorouslythat lim
x1
(4+8x) = 4.
solution Let > 0andtake = /8. Then, whenever |x (1)| = |x +1| < ,
|f (x) (4)| = |4+8x +4| = 8|x +1| < 8 = .
x3
(x
2
x) = 6.
3 DIFFERENTIATION
3.1 Definition of the Derivative
1. Whichof thelinesinFigure10aretangent tothecurve?
A
B
C
D
FIGURE 10
solution LinesB andD aretangent tothecurve.
2. What arethetwowaysof writingthedifferencequotient?
solution Thedifferencequotient maybewritteneither as
f (x) f (a)
x a
or as
f (a +h) f (a)
h
.
3. Finda andh suchthat
f (a +h) f (a)
h
isequal totheslopeof thesecant linebetween(3, f (3)) and(5, f (5)).
solution Witha = 3andh = 2,
f (a +h) f (a)
h
isequal totheslopeof thesecantlinebetweenthepoints(3, f (3))
and(5, f (5)) onthegraphof f (x).
4. Whichderivativeisapproximatedby
tan
_
4
+0.0001
_
1
0.0001
?
solution
tan(
4
+0.0001) 1
0.0001
isagoodapproximationtothederivativeof thefunctionf (x) = tanx at x =

4
.
5. What dothefollowingquantitiesrepresent intermsof thegraphof f (x) = sinx?
(a) sin1.3sin0.9 (b)
sin1.3sin0.9
0.4
(c) f
(0.9)
solution Consider thegraphof y = sinx.
(a) Thequantitysin1.3sin0.9representsthedifferenceinheight betweenthepoints(0.9, sin0.9) and(1.3, sin1.3).
(b) Thequantity
sin1.3sin0.9
0.4
representstheslopeof thesecantlinebetweenthepoints(0.9, sin0.9) and(1.3, sin1.3)
onthegraph.
(c) Thequantityf
(0.9) representstheslopeof thetangent linetothegraphat x = 0.9.

Exercises
1. Let f (x) = 5x
2
. Showthat f (3+h) = 5h
2
+30h +45. Thenshowthat
f (3+h) f (3)
h
= 5h +30
andcomputef
(3) bytakingthelimit ash 0.

91
92 C HA P T E R 3 DIFFERENTIATION
solution Withf (x) = 5x
2
, it followsthat
f (3+h) = 5(3+h)
2
= 5(9+6h +h
2
) = 45+30h +5h
2
.
Usingthisresult, wend
f (3+h) f (3)
h
=
45+30h +5h
2
5 9
h
=
45+30h +5h
2
45
h
=
30h +5h
2
h
= 30+5h.
Ash 0, 30+5h 30, sof
(3) = 30.
Let f (x) = 2x
2
3x 5. Showthat thesecant linethrough(2, f (2)) and(2+h, f (2+h)) hasslope2h +5.
Thenusethisformulatocomputetheslopeof:
(a) Thesecant linethrough(2, f (2)) and(3, f (3))
(b) Thetangent lineat x = 2(bytakingalimit)
In Exercises 36, compute f
(a) in two ways, using Eq. (1) and Eq. (2).

3. f (x) = x
2
+9x, a = 0
2
+9x. Then
f
(0) = lim
h0
f (0+h) f (0)
h
= lim
h0
(0+h)
2
+9(0+h) 0
h
= lim
h0
9h +h
2
h
= lim
h0
(9+h) = 9.
Alternately,
f
(0) = lim
x0
f (x) f (0)
x 0
= lim
x0
x
2
+9x 0
x
= lim
x0
(x +9) = 9.
f (x) = x
2
+9x, a = 2
5. f (x) = 3x
2
+4x +2, a = 1
solution Let f (x) = 3x
2
+4x +2. Then
f
(1) = lim
h0
f (1+h) f (1)
h
= lim
h0
3(1+h)
2
+4(1+h) +21
h
= lim
h0
3h
2
2h
h
= lim
h0
(3h 2) = 2.
Alternately,
f
(1) = lim
x1
f (x) f (1)
x (1)
= lim
x1
3x
2
+4x +21
x +1
= lim
x1
(3x +1)(x +1)
x +1
= lim
x1
(3x +1) = 2.
f (x) = x
3
, a = 2
In Exercises 710, refer to Figure 11.
0.5
1.0
1.5
2.0
2.5
3.0
f (x)
1.0 2.0 3.0 0.5 1.5 2.5
x
y
FIGURE 11
7. Find the slope of the secant line through (2, f (2)) and (2.5, f (2.5)). Is it larger or smaller than f
(2)?
Explain.
solution Fromthegraph, it appears that f (2.5) = 2.5 and f (2) = 2. Thus, theslopeof thesecant linethrough
(2, f (2)) and(2.5, f (2.5)) is
f (2.5) f (2)
2.52
=
2.52
2.52
= 1.
Fromthegraph, it isalsoclear that thesecant linethrough(2, f (2)) and(2.5, f (2.5)) hasalarger slopethanthetangent
lineat x = 2. Inother words, theslopeof thesecant linethrough(2, f (2)) and(2.5, f (2.5)) islarger thanf
(2).
Estimate
f (2+h) f (2)
h
for h = 0.5. Whatdoesthisquantityrepresent?Isitlarger or smaller thanf
(2)?
Explain.
9. Estimatef
(1) andf
(2).
solution Fromthegraph, it appearsthat thetangent lineat x = 1wouldbehorizontal. Thus, f
(1) 0. Thetangent
lineat x = 2appearstopassthroughthepoints(0.5, 0.8) and(2, 2). Thus
f
(2)
20.8
20.5
= 0.8.
Findavalueof h for which
f (2+h) f (2)
h
= 0.
S E C T I ON 3.1 Definition of the Derivative 93
In Exercises 1114, refer to Figure 12.
1
2
3
5
4
1 2 3 4 5 6 7 8 9
x
y
FIGURE 12 Graphof f (x).
11. Determinef
(a) for a = 1, 2, 4, 7.
solution Remember that thevalueof thederivativeof f at x = a canbeinterpretedastheslopeof thelinetangent
tothegraphof y = f (x) at x = a. FromFigure12, weseethat thegraphof y = f (x) isahorizontal line(that is, aline
withzeroslope) ontheinterval 0 x 3. Accordingly, f
(1) = f
(2) = 0. Ontheinterval 3 x 5, thegraphof

y = f (x) isalineof slope
1
2
; thus, f
(4) =
1
2
. Finally, thelinetangent tothegraphof y = f (x) at x = 7ishorizontal,
sof
(7) = 0.
For whichvaluesof x isf
(x) < 0?
13. Whichislarger, f
(5.5) or f
(6.5)?
solution Thelinetangent tothegraphof y = f (x) at x = 5.5hasalarger slopethanthelinetangent tothegraphof
y = f (x) at x = 6.5. Therefore, f
(5.5) islarger thanf
(6.5).
Showthat f
(3) doesnot exist.

In Exercises 1518, use the limit denition to calculate the derivative of the linear function.
15. f (x) = 7x 9
solution
lim
h0
f (a +h) f (a)
h
= lim
h0
7(a +h) 9(7a 9)
h
= lim
h0
7= 7.
f (x) = 12
17. g(t ) = 83t
solution
lim
h0
g(a +h) g(a)
h
= lim
h0
83(a +h) (83a)
h
= lim
h0
3h
h
= lim
h0
(3) = 3.
k(z) = 14z +12
19. Findanequationof thetangent lineat x = 3, assumingthat f (3) = 5andf
(3) = 2?
solution Bydenition, theequationof thetangent linetothegraphof f (x) at x = 3isy = f (3) +f
(3)(x 3) =
5+2(x 3) = 2x 1.
Findf (3) andf
(3), assumingthat thetangent linetoy = f (x) at a = 3hasequationy = 5x +2.

21. Describethetangent lineat anarbitrarypoint onthecurve y = 2x +8.
solution Sincey = 2x +8representsastraight line, thetangent lineat anypoint isthelineitself, y = 2x +8.
Supposethat f (2+h) f (2) = 3h
2
+5h. Calculate:
(a) Theslopeof thesecant linethrough(2, f (2)) and(6, f (6))
(b) f
(2)
23. Let f (x) =
1
x
. Does f (2+ h) equal
1
2+h
or
1
2
+
1
h
? Compute the difference quotient at a = 2 with
h = 0.5.
1
x
. Then
f (2+h) =
1
2+h
.
Witha = 2andh = 0.5, thedifferencequotient is
f (a +h) f (a)
h
=
f (1.5) f (2)
0.5
=
1
1.5
1
2
0.5
=
1
3
.
Let f (x) =

x. Does f (5+ h) equal

5+h or
5+

h? Computethedifferencequotient at a = 5with
h = 1.
25. Let f (x) = 1/
x. Computef
(5) byshowingthat
f (5+h) f (5)
h
=
1
5+h(
5+h +
5)
solution Let f (x) = 1/
x. Then
f (5+h) f (5)
h
=
1
5+h
5
h
=
5+h
h
5+h
=
5+h
h
5+h
_
5+
5+h
5+
5+h
_
=
5(5+h)
h
5+h(
5+h +
5)
=
1
5+h(
5+h +
5)
.
Thus,
f
(5) = lim
h0
f (5+h) f (5)
h
= lim
h0
5+h(
5+h +
5)
=
1
5(
5+
5)
=
1
10
5
.
Findanequationof thetangent linetothegraphof f (x) = 1/
x at x = 9.
In Exercises 2744, use the limit denition to compute f
(a) and nd an equation of the tangent line.

27. f (x) = 2x
2
+10x, a = 3
2
+10x. Then
f
(3) = lim
h0
f (3+h) f (3)
h
= lim
h0
2(3+h)
2
+10(3+h) 48
h
= lim
h0
18+12h +2h
2
+30+10h 48
h
= lim
h0
(22+2h) = 22.
At a = 3, thetangent lineis
y = f
(3)(x 3) +f (3) = 22(x 3) +48= 22x 18.

f (x) = 4x
2
, a = 1
29. f (t ) = t 2t
2
, a = 3
solution Let f (t ) = t 2t
2
. Then
f
(3) = lim
h0
f (3+h) f (3)
h
= lim
h0
(3+h) 2(3+h)
2
(15)
h
= lim
h0
3+h 1812h 2h
2
+15
h
= lim
h0
(112h) = 11.
y = f
(3)(t 3) +f (3) = 11(t 3) 15= 11t +18.

f (x) = 8x
3
, a = 1
31. f (x) = x
3
+x, a = 0
3
+x. Then
f
(0) = lim
h0
f (h) f (0)
h
= lim
h0
h
3
+h 0
h
= lim
h0
(h
2
+1) = 1.
y = f
(0)(x 0) +f (0) = x.
f (t ) = 2t
3
+4t , a = 4
33. f (x) = x
1
, a = 8
1
. Then
f
(8) = lim
h0
f (8+h) f (8)
h
= lim
h0
1
8+h

_
1
8
_
h
= lim
h0
88h
8(8+h)
h
= lim
h0
h
(64+8h)h
=
1
64
Thetangent at a = 8is
y = f
(8)(x 8) +f (8) =
1
64
(x 8) +
1
8
=
1
64
x +
1
4
.
f (x) = x +x
1
, a = 4
35. f (x) =
1
x +3
, a = 2
1
x+3
. Then
f
(2) = lim
h0
f (2+h) f (2)
h
= lim
h0
1
2+h+3
1
h
= lim
h0
1
1+h
1
h
= lim
h0
h
h(1+h)
= lim
h0
1
1+h
= 1.
Thetangent lineat a = 2is
y = f
(2)(x +2) +f (2) = 1(x +2) +1= x 1.

f (t ) =
2
1t
, a = 1
37. f (x) =

x +4, a = 1

x +4. Then
f
(1) = lim
h0
f (1+h) f (1)
h
= lim
h0
h +5
5
h
= lim
h0
h +5
5
h
h +5+
h +5+
5
= lim
h0
h
h(
h +5+
5)
= lim
h0
1
h +5+
5
=
1
2
5
.
y = f
(1)(x 1) +f (1) =
1
2
5
(x 1) +
5=
1
2
5
x +
9
2
5
.
f (t ) =

3t +5, a = 1
39. f (x) =
1
x
, a = 4
1
x
. Then
f
(4) = lim
h0
f (4+h) f (4)
h
= lim
h0
1
4+h
1
2
h
= lim
h0
2
4+h
2
4+h
2+
4+h
2+
4+h
h
= lim
h0
44h
4
4+h+2(4+h)
h
= lim
h0
1
4
4+h +2(4+h)
=
1
16
.
At a = 4thetangent lineis
y = f
(4)(x 4) +f (4) =
1
16
(x 4) +
1
2
=
1
16
x +
3
4
.
f (x) =
1
2x +1
, a = 4
41. f (t ) =
_
t
2
+1, a = 3
solution Let f (t ) =
_
t
2
+1. Then
f
(3) = lim
h0
f (3+h) f (3)
h
= lim
h0
_
10+6h +h
2
10
h
= lim
h0
_
10+6h +h
2
10
h
_
10+6h +h
2
+
10
_
10+6h +h
2
+
10
= lim
h0
6h +h
2
h(
_
10+6h +h
2
+
10)
= lim
h0
6+h
_
10+6h +h
2
+
10
=
3
10
.
y = f
(3)(t 3) +f (3) =
3
10
(t 3) +
10=
3
10
t +
1
10
.
f (x) = x
2
, a = 1
43. f (x) =
1
x
2
+1
, a = 0
1
x
2
+1
. Then
f
(0) = lim
h0
f (0+h) f (0)
h
= lim
h0
1
(0+h)
2
+1
1
h
= lim
h0
h
2
h
2
+1
h
= lim
h0
h
h
2
+1
= 0.
y = f (0) +f
(0)(x 0) = 1+0(x 1) = 1.
f (t ) = t
3
, a = 1
45. Figure13displaysdatacollectedbythebiologistJ ulianHuxley(18871975) ontheaverageantler weightW of male
reddeer asafunctionof aget . Estimatethederivativeatt = 4. For whichvaluesof t istheslopeof thetangentlineequal
tozero? For whichvaluesisit negative?
2 4 0 6 8 10 12 14
Age(years)
Antler
Weight
(kg)
0
1
2
3
4
5
6
7
8
FIGURE 13
solution LetW(t ) denotetheantlerweightasafunctionof age. Thetangentlinesketchedinthegurebelowpasses
throughthepoints(1, 1) and(6, 5.5). Therefore
W
(4)
5.51
61
= 0.9kg/year.
If theslopeof thetangentiszero, thetangentlineishorizontal. Thisappearstohappenatroughlyt = 10andatt = 11.6.
Theslopeof thetangent lineisnegativewhentheheight of thegraphdecreasesaswemovetotheright. For thegraphin
Figure13, thisoccursfor 10< t < 11.6.
2 4 0 6 8 10 12 14
0
1
2
3
4
5
6
7
8
y
x
Figure14(A) shows thegraph of f (x) =

x. Theclose-up in Figure14(B) shows that thegraph is nearly a
straight linenear x = 16. Estimatetheslopeof thislineandtakeit asanestimatefor f
(16). Thencomputef
(16)
andcomparewithyour estimate.
47. Let f (x) =
4
1+2
x
.
(a) Plot f (x) over [2, 2]. Thenzoominnear x = 0until thegraphappearsstraight, andestimatetheslopef
(0).
(b) Use(a) tondanapproximateequationtothetangent lineat x = 0. Plot thislineandf (x) onthesameset of axes.
solution
(a) Thegurebelowattheleftshowsthegraphof f (x) =
4
1+2
x
over [2, 2]. Thegurebelowattherightisaclose-up
near x = 0. Fromtheclose-up, weseethat thegraphisnearly straight andpassesthroughthepoints(0.22, 2.15) and
(0.22, 1.85). Wethereforeestimate
f
(0)
1.852.15
0.22(0.22)
=
0.3
0.44
= 0.68
y
x
2 1 1 2
y
x
0.2 0.1 0.1 0.2
0.5
1.8
2.0
2.2
2.4
1
1.5
2
3
2.5
(b) Usingtheestimatefor f
(0) obtainedinpart (a), theapproximateequationof thetangent lineis

y = f
(0)(x 0) +f (0) = 0.68x +2.

Thegurebelowshowsthegraphof f (x) andtheapproximatetangent line.
y
x
2 1 1 2
0.5
1
1.5
2
3
2.5
Let f (x) = cotx. Estimatef
2
_
graphicallybyzoominginonaplot of f (x) near x =

2
.
49. Determinetheintervalsalongthex-axisonwhichthederivativeinFigure15ispositive.
1.0 1.5 2.0 2.5 3.0 3.5 4.0 0.5
1.0
0.5
1.5
2.0
2.5
3.0
3.5
4.0
x
y
FIGURE 15
solution Thederivative(thatis, theslopeof thetangentline) ispositivewhentheheightof thegraphincreasesaswe
movetotheright. FromFigure15, thisappearstobetruefor 1< x < 2.5andfor x > 3.5.
Sketchthegraphof f (x) = sinx on[0, ] andguessthevalueof f
2
_
. Thencalculatethedifferencequotient
at x =

2
for twosmall positiveandnegativevaluesof h. Arethesecalculationsconsistent withyour guess?
In Exercises 5156, each limit represents a derivative f
(a). Find f (x) and a.

51. lim
h0
(5+h)
3
125
h
solution Thedifferencequotient
(5+h)
3
125
h
hastheform
f (a +h) f (a)
h
wheref (x) = x
3
anda = 5.
lim
x5
x
3
125
x 5
53. lim
h0
sin
_
6
+h
_
0.5
h
sin(
6
+h) 0.5
h
hastheform
f (a +h) f (a)
h
wheref (x) = sinx anda =

6
.
lim
x
1
4
x
1
4
x
1
4
55. lim
h0
5
2+h
25
h
5
(2+h)
25
h
hastheform
f (a +h) f (a)
h
wheref (x) = 5
x
anda = 2.
lim
h0
5
h
1
h
57. Applythemethodof Example6tof (x) = sinx todeterminef
4
_
accuratelytofour decimal places.
solution Weknowthat
f
(/4) = lim
h0
f (/4+h) f (/4)
h
= lim
h0
sin(/4+h)
2/2
h
.
Creatingatableof valuesof h closetozero:
h 0.001 0.0001 0.00001 0.00001 0.0001 0.001
sin(
4
+h) (
2/2)
h
0.7074602 0.7071421 0.7071103 0.7071033 0.7070714 0.7067531
Accurateuptofour decimal places, f
4
) 0.7071.
Applythemethodof Example6tof (x) = cosx todeterminef
5
_
accuratelytofour decimal places. Usea
graphof f (x) toexplainhowthemethodworksinthiscase.
59. For eachgraphinFigure16, determinewhether f
(1) islarger or smaller thantheslopeof thesecant line

betweenx = 1andx = 1+h for h > 0. Explain.
1 1
(A) (B)
y
x
y
x
y = f (x)
y = f (x)
FIGURE 16
solution
Oncurve(A),f
(1) islarger than

f (1+h) f (1)
h
;
thecurveisbendingdownwards, sothat thesecant linetotheright isat alower anglethanthetangent line. Wesay
suchacurveisconcavedown, andthat itsderivativeisdecreasing.
Oncurve(B), f
(1) issmaller than

f (1+h) f (1)
h
;
thecurveisbendingupwards, sothat thesecant linetotheright isat asteeper anglethanthetangent line. Wesay
suchacurveisconcaveup, andthat itsderivativeisincreasing.
Refer tothegraphof f (x) = 2
x
inFigure17.
(a) Explaingraphicallywhy, for h > 0,
f (h) f (0)
h
f
(0)
f (h) f (0)
h
(b) Use(a) toshowthat 0.69314 f
(0) 0.69315.
(c) Similarly, computef
(x) tofour decimal placesfor x = 1, 2, 3, 4.

(d) Nowcomputetheratiosf
(x)/f
(0) for x = 1, 2, 3, 4. Canyouguessanapproximateformulafor f
(x)?
61. Sketchthegraphof f (x) = x
5/2
on[0, 6].
(a) Usethesketchtojustifytheinequalitiesfor h > 0:
f (4) f (4h)
h
f
(4)
f (4+h) f (4)
h
(b) Use(a) tocomputef
(4) tofour decimal places.

(c) Useagraphingutilitytoplot f (x) andthetangent lineat x = 4, usingyour estimatefor f
(4).
solution
(a) Theslopeof thesecant linebetweenpoints(4, f (4)) and(4+h, f (4+h)) is
f (4+h) f (4)
h
.
x
5/2
is asmoothcurveincreasingat afaster rateas x . Therefore, if h > 0, thentheslopeof thesecant lineis
greater thantheslopeof thetangent lineat f (4), whichhappenstobef
(4). Likewise, if h < 0, theslopeof thesecant

lineislessthantheslopeof thetangent lineat f (4), whichhappenstobef
(4).
(b) Weknowthat
f
(4) = lim
h0
f (4+h) f (4)
h
= lim
h0
(4+h)
5/2
32
h
.
Creatingatablewithvaluesof h closetozero:
h 0.0001 0.00001 0.00001 0.0001
(4+h)
5/2
32
h
19.999625 19.99999 20.0000 20.0000375
Thus, f
(4) 20.0000.
(c) Usingtheestimatefor f
(4) obtainedinpart (b), theequationof thelinetangent tof (x) = x

5/2
at x = 4is
y = f
(4)(x 4) +f (4) = 20(x 4) +32= 20x 48.

y
x
1 2 3 4 5 6
20
40
60
20
40
60
80
Verify that P =
_
1,
1
2
_
lies onthegraphs of bothf (x) = 1/(1+x
2
) andL(x) =
1
2
+ m(x 1) for every
slopem. Plotf (x) andL(x) onthesameaxesfor several valuesof muntil youndavalueof mfor whichy = L(x)
appearstangent tothegraphof f (x). What isyour estimatefor f
(1)?
63. Useaplot of f (x) = x
x
toestimatethevaluec suchthat f
(c) = 0. Findc tosufcient accuracysothat
f (c +h) f (c)
h
0.006 for h = 0.001

solution Hereisagraphof f (x) = x
x
over theinterval [0, 1.5].
0.2 0.4 0.6 0.8 1 1.2 1.4
0.5
1
1.5
2
y
x
Thegraphshowsonelocationwithahorizontal tangentline. Thegurebelowattheleftshowsthegraphof f (x) together
withthehorizontal linesy = 0.6, y = 0.7andy = 0.8. Theliney = 0.7isvery closetobeingtangent tothegraphof
f (x). Thegurebelowattherightrenesthisestimatebygraphingf (x) andy = 0.69onthesamesetof axes. Thepoint
of tangencyhasanx-coordinateof roughly0.37, soc 0.37.
0.2 0.4 0.6 0.8 1 1.2 1.4
0.5
1
1.5
2
y
x
0.2 0.4 0.6 0.8 1 1.2 1.4
0.5
1
1.5
2
y
x
Wenotethat
f (0.37+0.001) f (0.37)
0.001
0.00491< 0.006
and
f (0.370.001) f (0.37)
0.001
0.00304< 0.006,
sowehavedeterminedc tothedesiredaccuracy.
Plot f (x) = x
x
andy = 2x +a onthesameset of axesfor several valuesof a until thelinebecomestangent
tothegraph. Thenestimatethevaluec suchthat f
(c) = 2.
In Exercises 6571, estimate derivatives using the symmetric difference quotient (SDQ), dened as the average of the
difference quotients at h and h:
1
2
_
f (a +h) f (a)
h
+
f (a h) f (a)
h
_
=
f (a +h) f (a h)
2h
4
The SDQ usually gives a better approximation to the derivative than the difference quotient.
65. Thevapor pressureof water at temperatureT (inkelvins) istheatmosphericpressureP at whichnonet evaporation
takesplace. UsethefollowingtabletoestimateP
(T ) for T = 303, 313, 323, 333, 343bycomputingtheSDQgivenby

Eq. (4) withh = 10.
T (K) 293 303 313 323 333 343 353
P (atm) 0.0278 0.0482 0.0808 0.1311 0.2067 0.3173 0.4754
solution Usingequation(4),
P
(303)
P(313) P(293)
20
=
0.08080.0278
20
= 0.00265atm/K;
P
(313)
P(323) P(303)
20
=
0.13110.0482
20
= 0.004145atm/K;
P
(323)
P(333) P(313)
20
=
0.20670.0808
20
= 0.006295atm/K;
P
(333)
P(343) P(323)
20
=
0.31730.1311
20
= 0.00931atm/K;
P
(343)
P(353) P(333)
20
=
0.47540.2067
20
= 0.013435atm/K
UsetheSDQ withh = 1year to estimateP
(T ) intheyears 2000, 2002, 2004, 2006, whereP(T ) is theU.S.

ethanol production(Figure18). Expressyour answer inthecorrect units.
In Exercises 67 and 68, trafc speed S along a certain road (in km/h) varies as a function of trafc density q (number of
cars per km of road). Use the following data to answer the questions:
q (density) 60 70 80 90 100
S (speed) 72.5 67.5 63.5 60 56
67. EstimateS
(80).
solution Let S(q) bethefunctiondeterminingS givenq. Usingequation(4) withh = 10,
S
(80)
S(90) S(70)
20
=
6067.5
20
= 0.375;
withh = 20,
S
(80)
S(100) S(60)
40
=
5672.5
40
= 0.4125;
Themeanof thesetwosymmetricdifferencequotientsis0.39375kphkm/car.
Explainwhy V = qS, calledtrafc volume, is equal tothenumber of cars passingapoint per hour. Usethe
datatoestimateV
(80).
Exercises 6971: The current (in amperes) at time t (in seconds) owing in the circuit in Figure 19 is given by Kirchhoff s
Law:
i(t ) = Cv
(t ) +R
1
v(t )
where v(t ) is the voltage (in volts), C the capacitance (in farads), and R the resistance (in ohms, ).
+
v
R
i
C
FIGURE 19
69. Calculatethecurrent at t = 3if
v(t ) = 0.5t +4V
whereC = 0.01F andR = 100.
solution Since v(t ) is a line with slope 0.5, v
(t ) = 0.5 volts/s for all t . Fromthe formula, i(3) = Cv
(3) +
(1/R)v(3) = 0.01(0.5) +(1/100)(5.5) = 0.005+0.055= 0.06amperes.
Usethefollowingdatatoestimatev
(10) (byanSDQ). Thenestimatei(10), assumingC = 0.03andR = 1000.

t 9.8 9.9 10 10.1 10.2
v(t ) 256.52 257.32 258.11 258.9 259.69
71. Assumethat R = 200 but C is unknown. Usethefollowingdatato estimatev
(4) (by anSDQ) anddeducean

approximatevaluefor thecapacitanceC.
t 3.8 3.9 4 4.1 4.2
v(t ) 388.8 404.2 420 436.2 452.8
i(t ) 32.34 33.22 34.1 34.98 35.86
S E C T I ON 3.2 The Derivative as a Function 101
solution Solvingi(4) = Cv
(4) +(1/R)v(4) for C yields

C =
i(4) (1/R)v(4)
v
(4)
=
34.1
420
200
v
(4)
.
TocomputeC, werst approximatev
(4). Takingh = 0.1, wend

v
(4)
v(4.1) v(3.9)
0.2
=
436.2404.2
0.2
= 160.
Pluggingthisintotheequationaboveyields
C
34.12.1
160
= 0.2farads.
TheSDQusuallyapproximatesthederivativemuchmorecloselythandoestheordinarydifferencequotient. Let
f (x) = 2
x
anda = 0. ComputetheSDQwithh = 0.001andtheordinary differencequotientswithh = 0.001.
Comparewiththeactual value, whichisf
(0) = ln2.
73. Explainhowthesymmetricdifferencequotient denedbyEq. (4) canbeinterpretedastheslopeof asecant line.
solution Thesymmetricdifferencequotient
f (a +h) f (a h)
2h
is theslopeof thesecant lineconnectingthepoints (a h, f (a h)) and(a + h, f (a + h)) onthegraphof f ; the
differenceinthefunctionvaluesisdividedbythedifferenceinthex-values.
Whichof thetwofunctionsinFigure20satisestheinequality
f (a +h) f (a h)
2h
f (a +h) f (a)
h
for h > 0? Explainintermsof secant lines.
75. Showthat if f (x) is aquadratic polynomial, then theSDQ at x = a (for any h = 0) is equal to f
(a).
Explainthegraphical meaningof thisresult.
solution Let f (x) = px
2
+qx +r beaquadraticpolynomial. WecomputetheSDQat x = a.
f (a +h) f (a h)
2h
=
p(a +h)
2
+q(a +h) +r (p(a h)
2
+q(a h) +r)
2h
=
pa
2
+2pah +ph
2
+qa +qh +r pa
2
+2pah ph
2
qa +qh r
2h
=
4pah +2qh
2h
=
2h(2pa +q)
2h
= 2pa +q
Sincethisdoesntdependonh, thelimit, whichisequal tof
(a), isalso2pa +q. Graphically, thisresulttellsusthatthe

secant linetoaparabolapassingthroughpointschosensymmetricallyabout x = a isalwaysparallel tothetangent line
at x = a.
Let f (x) = x
2
. Computef
(1) bytakingthelimit of theSDQs(witha = 1) ash 0.

3.2 The Derivative as a Function
1. What istheslopeof thetangent linethroughthepoint (2, f (2)) if f
(x) = x
3
?
solution Theslopeof thetangentlinethroughthepoint(2, f (2)) isgivenbyf
(2). Sincef
(x) = x
3
, itfollowsthat
f
(2) = 2
3
= 8.
2. Evaluate(f g)
(1) and(3f +2g)
(1) assumingthat f
(1) = 3andg
(1) = 5.
solution (f g)
(1) = f
(1) g
(1) = 35= 2and(3f +2g)
(1) = 3f
(1) +2g
(1) = 3(3) +2(5) = 19.

3. Towhichof thefollowingdoesthePower Ruleapply?
(a) f (x) = x
2
(b) f (x) = 2
(c) f (x) = x
(d) f (x) =
x
(e) f (x) = x
x
(f) f (x) = x
4/5
solution
(a) Yes. x
2
isapower function, sothePower Rulecanbeapplied.
(b) Yes. 2
isaconstant function, sothePower Rulecanbeapplied.

(c) Yes. x

(d) No.
x
isanexponential function(thebaseisconstant whiletheexponent isavariable), sothePower Ruledoesnot
apply.
(e) No. x
x
isnotapower functionbecauseboththebaseandtheexponentarevariable, sothePower Ruledoesnotapply.
(f) Yes. x
4/5
4. Choose(a) or (b). Thederivativedoesnot exist if thetangent lineis: (a) horizontal (b) vertical.
solution Thederivativedoesnot exist when: (b) thetangent lineisvertical. At ahorizontal tangent, thederivativeis
zero.
5. If f (x) isdifferentiableat x = c, isf (x) necessarilycontinuousat x = c? Dothereexist continuousfunctionsthat
arenot differentiable?
solution ByTheorem4, if f isdifferentiableat x = c, thenit iscontinuousat x = c. Theconversedoesnot hold,
however. For example, f (x) = |x| is continuous at x = 0but is not differentiabletheresincetheslopes fromtheleft
equal 1whilethosefromtheright equal 1.
Exercises
(x) using the limit denition.

1. f (x) = 3x 7
solution Let f (x) = 3x 7. Then,
f
(x) = lim
h0
f (x +h) f (x)
h
= lim
h0
3(x +h) 7(3x 7)
h
= lim
h0
3h
h
= 3.
f (x) = x
2
+3x
3. f (x) = x
3
3
. Then,
f
(x) = lim
h0
f (x +h) f (x)
h
= lim
h0
(x +h)
3
x
3
h
= lim
h0
x
3
+3x
2
h +3xh
2
+h
3
x
3
h
= lim
h0
3x
2
h +3xh
2
+h
3
h
= lim
h0
(3x
2
+3xh +h
2
) = 3x
2
.
f (x) = 1x
1
5. f (x) = x
x
x. Then,
f
(x) = lim
h0
f (x +h) f (x)
h
= lim
h0
x +h
x +h (x
x)
h
= 1 lim
h0
x +h
x
h
x +h +
x +h +
x
_
= 1 lim
h0
(x +h) x
h(
x +h +
x)
= 1 lim
h0
1
x +h +
x
= 1
1
2
x
.
f (x) = x
1/2
In Exercises 714, use the Power Rule to compute the derivative.
7.
d
dx
x
4
x=2
solution
d
dx
_
x
4
_
= 4x
3
so
d
dx
x
4
x=2
= 4(2)
3
= 32.
d
dt
t
3
t =4
9.
d
dt
t
2/3
t =8
solution
d
dt
_
t
2/3
_
=
2
3
t
1/3
so
d
dt
t
2/3
t =8
=
2
3
(8)
1/3
=
1
3
.
d
dt
t
2/5
t =1
11.
d
dx
x
0.35
solution
d
dx
_
x
0.35
_
= 0.35(x
0.351
) = 0.35x
0.65
.
d
dx
x
14/3
13.
d
dt
t
17
solution
d
dt
_
t
17
_
=
17t
171
d
dt
t
2
(x) and nd an equation of the tangent line to the graph at x = a.

15. f (x) = x
4
, a = 2
4
. Then, by thePower Rule, f
(x) = 4x
3
. Theequationof thetangent lineto thegraphof
f (x) at x = 2is
y = f
(2)(x 2) +f (2) = 32(x 2) +16= 32x 48.

f (x) = x
2
, a = 5
17. f (x) = 5x 32
x, a = 4
solution Letf (x) = 5x 32x
1/2
. Thenf
(x) = 516x
1/2
. Inparticular, f
(4) = 3. Thetangentlineatx = 4
is
y = f
(4)(x 4) +f (4) = 3(x 4) 44= 3x 32.

f (x) =
3
x, a = 8
19. Findanequationof thetangent linetoy =
1
x
at x = 9.
1
x
. Thenf (9) =
1
9
, f
(x) =
1
x
2
, andf
(9) =
1
81
. Theequationof thetangent lineis
y
1
9
=
1
81
(x 9), or y =
1
81
x +
2
9
Findapoint onthegraphof y =

x wherethetangent linehasslope10.
In Exercises 2132, calculate the derivative.
21. f (x) = 2x
3
3x
2
+5
solution
d
dx
_
2x
3
3x
2
+5
_
= 6x
2
6x.
f (x) = 2x
3
3x
2
+2x
23. f (x) = 4x
5/3
3x
2
12
solution
d
dx
_
4x
5/3
3x
2
12
_
=
20
3
x
2/3
+6x
3
.
f (x) = x
5/4
+4x
3/2
+11x
25. g(z) = 7z
5/14
+z
5
+9
solution
d
dz
_
7z
5/14
+z
5
+9
_
=
5
2
z
19/14
5z
6
.
h(t ) = 6
t +
1
t
27. f (s) =
4
s +
3
s
solution f (s) =
4
s +
3
s = s
1/4
+s
1/3
. Inthisform, wecanapplytheSumandPower Rules.
d
ds
_
s
1/4
+s
1/3
_
=
1
4
(s
(1/4)1
) +
1
3
(s
(1/3)1
) =
1
4
s
3/4
+
1
3
s
2/3
.
W(y) = 6y
4
+7y
2/3
29. g(x) =
2
solution Because
2
isaconstant,
d
dx
2
= 0.
f (x) = x
31. h(t ) =

2t
2
solution
d
dt
2t
2
= 2t
21
.
R(z) =
z
5/3
4z
3/2
z
Hint: simplify.
In Exercises 3336, calculate the derivative by expanding or simplifying the function.
33. P(s) = (4s 3)
2
solution P(s) = (4s 3)
2
= 16s
2
24s +9. Thus,
dP
ds
= 32s 24.
Q(r) = (12r)(3r +5)
35. g(x) =
x
2
+4x
1/2
x
2
solution g(x) =
x
2
+4x
1/2
x
2
= 1+4x
3/2
. Thus,
dg
dx
= 6x
5/2
.
s(t ) =
12t
t
1/2
In Exercises 3742, calculate the derivative indicated.
37.
dT
dC
C=8
, T = 3C
2/3
solution WithT (C) = 3C
2/3
, wehave
dT
dC
= 2C
1/3
. Therefore,
dT
dC
C=8
= 2(8)
1/3
= 1.
dP
dV
V=2
, P =
7
V
39.
ds
dz
z=2
, s = 4z 16z
2
solution Withs = 4z 16z
2
, wehave
ds
dz
= 432z. Therefore,
ds
dz
z=2
= 432(2) = 60.
dR
dW
W=1
, R = W
41.
dr
dt
t =4
, r =
t
2
+1
t
1/2
solution Wehave
dr
dt
=
d
dt
t
2
+1
t
1/2
=
d
dt
(t
3/2
+t
1/2
) =
3
2
t
1/2
1
2
t
3/2
Evaluatingat t = 4gives
dr
dt
t =4
=
3
2
4
1/2
1
2
4
3/2
=
47
16
dp
dh
h=32
, p = 16h
0.2
+8h
0.8
43. Matchthefunctions ingraphs (A)(D) withtheir derivatives (I)(III) inFigure11. Notethat two of thefunctions
havethesamederivative. Explainwhy.
y
x
x
(A)
y
(I)
x
y
(II)
x
y
(III)
y
x
(B)
y
x
(C)
y
x
(D)
FIGURE 11
solution
Consider thegraphin(A). Ontheleft sideof thegraph, theslopeof thetangent lineis positivebut ontheright
sidetheslopeof thetangent lineis negative. Thus thederivativeshouldtransitionfrompositivetonegativewith
increasingx. Thismatchesthegraphin(III).
Consider thegraphin(B). Thisisalinear function, soitsslopeisconstant. Thusthederivativeisconstant, which
matchesthegraphin(I).
Considerthegraphin(C). Movingfromlefttoright, theslopeof thetangentlinetransitionsfrompositivetonegative
thenback to positive. Thederivativeshouldthereforebenegativeinthemiddleandpositiveto either side. This
matchesthegraphin(II).
Consider thegraphin(D). Ontheleft sideof thegraph, theslopeof thetangent lineis positivebut ontheright
sidetheslopeof thetangent lineis negative. Thus thederivativeshouldtransitionfrompositivetonegativewith
increasingx. Thismatchesthegraphin(III).
Notethat thefunctionswhosegraphsareshownin(A) and(D) havethesamederivative. Thishappensbecausethe
graphin(D) is just avertical translationof thegraphin(A), whichmeans thetwo functions differ by aconstant. The
derivativeof aconstant iszero, sothetwofunctionsendupwiththesamederivative.
Of thetwofunctionsf andg inFigure12, whichisthederivativeof theother? J ustifyyour answer.
45. Assignthelabelsf (x), g(x), andh(x) tothegraphsinFigure13insuchawaythatf
(x) = g(x) andg
(x) = h(x).
y
x
y
x
y
x
(A) (B) (C)
FIGURE 13
solution Consider thegraphin(A). Movingfromleft toright, theslopeof thetangent lineispositiveover therst
quarter of thegraph, negativeinthemiddlehalf andpositiveagainover thenal quarter. Thederivativeof thisfunction
must thereforebenegativeinthemiddleandpositiveoneither side. Thismatchesthegraphin(C).
Nowfocus onthegraphin(C). Theslopeof thetangent lineis negativeover theleft half andpositiveontheright
half. Thederivativeof thisfunctionthereforeneedstobenegativeontheleft andpositiveontheright. Thisdescription
matchesthegraphin(B).
Weshould thereforelabel thegraph in (A) as f (x), thegraph in (B) as h(x), and thegraph in (C) as g(x). Then
f
(x) = g(x) andg
(x) = h(x).
Accordingtothepeak oil theory, rstproposedin1956bygeophysicistM. Hubbert, thetotal amountof crudeoil
Q(t ) producedworldwideuptotimet hasagraphlikethat inFigure14.
(a) SketchthederivativeQ
(t ) for 1900 t 2150. What doesQ
(t ) represent?
(b) Inwhichyear (approximately) doesQ
(t ) takeonitsmaximumvalue?
(c) What isL = lim
t
Q(t )?Andwhat isitsinterpretation?
(d) What isthevalueof lim
t
Q
(t )?
47. Usethetableof valuesof f (x) todeterminewhichof (A) or (B) inFigure15isthegraphof f
(x). Explain.
x 0 0.5 1 1.5 2 2.5 3 3.5 4
f (x) 10 55 98 139 177 210 237 257 268
x
y
x
y
(A) (B)
FIGURE 15 Whichisthegraphof f
(x)?
solution Theincrementbetweensuccessivex valuesinthetableisaconstant0.5buttheincrementbetweensuccessive
f (x) values decreases from45to 43to 41to 38andso on. Thus thedifferencequotients decreasewithincreasingx,
suggestingthat f
(x) decreasesasafunctionof x. Becausethegraphin(B) depictsadecreasingfunction, (B) might be

thegraphof thederivativeof f (x).
Let R beavariableandr aconstant. Computethederivatives:
(a)
d
dR
R (b)
d
dR
r (c)
d
dR
r
2
R
3
49. Computethederivatives, wherec isaconstant.
(a)
d
dt
ct
3
(b)
d
dy
(9c
2
y
3
24c) (c)
d
dz
(5z +4cz
2
)
solution
(a)
d
dt
ct
3
= 3ct
2
. (b)
d
dz
(5z +4cz
2
) = 5+8cz. (c)
d
dy
(9c
2
y
3
24c) = 27c
2
y
2
.
Findthepointsonthegraphof f (x) = 12x x
3
wherethetangent lineishorizontal.
51. Findthepointsonthegraphof y = x
2
+3x 7at whichtheslopeof thetangent lineisequal to4.
solution Let y = x
2
+3x 7. Solvingdy/dx = 2x +3= 4yieldsx =
1
2
.
Findthevaluesof x wherey = x
3
andy = x
2
+5x haveparallel tangent lines.
53. Determinea andb suchthat p(x) = x
2
+ax +b satisesp(1) = 0andp
(1) = 4.
solution Let p(x) = x
2
+ ax + b satisfy p(1) = 0andp
(1) = 4. Now, p
(x) = 2x + a. Therefore0 = p(1) =

1+a +b and4= p
(1) = 2+a; i.e., a = 2andb = 3.

Findall valuesof x suchthat thetangent linetoy = 4x
2
+11x +2issteeper thanthetangent linetoy = x
3
.
55. Letf (x) = x
3
3x +1. Showthatf
(x) 3for all x andthat, for everym > 3, therearepreciselytwopoints

wheref
(x) = m. Indicatethepositionof thesepointsandthecorrespondingtangentlinesfor onevalueof minasketch

of thegraphof f (x).
solution Let P = (a, b) beapoint onthegraphof f (x) = x
3
3x +1.
Thederivativesatisesf
(x) = 3x
2
3 3since3x
2
isnonnegative.
Supposetheslopem of thetangent lineisgreater than3. Thenf
(a) = 3a
2
3= m, whence
a
2
=
m+3
3
> 0 andthus a =
_
m+3
3
.
Thetwoparallel tangent lineswithslope2areshownwiththegraphof f (x) here.
2
1
2
2
4
1
2
x
y
Showthat thetangent linestoy =
1
3
x
3
x
2
at x = a andat x = b areparallel if a = b or a +b = 2.
57. Computethederivativeof f (x) = x
3/2
usingthelimit denition. Hint: Showthat
f (x +h) f (x)
h
=
(x +h)
3
x
3
h
_
1
_
(x +h)
3
+
x
3
_
solution Oncewehavethedifferenceof squareroots, wemultiplybytheconjugatetosolvetheproblem.
f
(x) = lim
h0
(x +h)
3/2
x
3/2
h
= lim
h0
_
(x +h)
3
x
3
h
__
(x +h)
3
+
x
3
_
(x +h)
3
+
x
3
_
= lim
h0
(x +h)
3
x
3
h
_
1
_
(x +h)
3
+
x
3
_
.
Therst factor of theexpressioninthelast lineisclearly thelimit denitionof thederivativeof x
3
, whichis3x
2
. The
secondfactor canbeevaluated, so
d
dx
x
3/2
= 3x
2
1
2
x
3
=
3
2
x
1/2
.
Sketchthegraphof acontinuousfunctionon(0, 5) that isdifferentiableexcept at x = 1andx = 4.
59. Show, usingthelimit denitionof thederivative, that f (x) = |x
2
4| isnot differentiableat x = 2.
solution Wehave
f
(2) = lim
h0
f (2+h) f (2)
h
= lim
h0
(2+h)
2
4
0
h
= lim
h0
h
2
+4h
h
= lim
h0
|h +4|
|h|
h
Ash approacheszero, h +4approaches4. However, thesecondfactor is1if h ispositiveandis1if h isnegative. Thus
lim
h0
|h +4|
|h|
h
= 4, and lim
h0+
|h +4|
|h|
h
= 4
Sincethetwoone-sidedlimitsareunequal, f (x) isnot differentiableat x = 2.
Theaveragespeed(inmetersper second) of agasmoleculeis
v
avg
=
_
8RT
M
whereT is thetemperature(in kelvins), M is themolar mass (in kilograms per mole), and R = 8.31. Calculate
dv
avg
/dT at T = 300K for oxygen, whichhasamolar massof 0.032kg/mol.
61. BiologistshaveobservedthatthepulserateP (inbeatsper minute) inanimalsisrelatedtobodymass(inkilograms)
bytheapproximateformulaP = 200m
1/4
.Thisisoneof manyallometric scaling laws prevalentinbiology. Is|dP/dm|
anincreasingor decreasingfunctionof m? Findanequationof thetangent lineat thepoints onthegraphinFigure16
that represent goat (m = 33) andman(m = 68).
Mass (kg)
500 400 300 200 100
Cattle
Pulse
(beats/min)
200
100
Guineapig
Goat
Man
FIGURE 16
solution dP/dm = 50m
5/4
. For m > 0, |dP/dm| = |50m
5/4
|. |dP/dm| 0asm getslarger; |dP/dm| gets
smaller asm getsbigger.
For eachm = c, theequationof thetangent linetothegraphof P at m is
y = P
(c)(mc) +P(c).
For agoat (m = 33kg), P(33) = 83.445beatsper minute(bpm) and
dP
dm
= 50(33)
5/4
0.63216bpm/kg.
Hence, y = 0.63216(m33) +83.445.
For aman(m = 68kg), wehaveP(68) = 69.647bpmand
dP
dm
= 50(68)
5/4
0.25606bpm/kg.
Hence, thetangent linehasformulay = 0.25606(m68) +69.647.
SomestudiessuggestthatkidneymassK inmammals(inkilograms) isrelatedtobodymassm(inkilograms) by
theapproximateformulaK = 0.007m
0.85
. CalculatedK/dm at m = 68. Thencalculatethederivativewithrespect
tom of therelativekidney-to-massratioK/m at m = 68.
63. TheClausiusClapeyronLawrelatesthevapor pressure of waterP (inatmospheres)tothetemperatureT (inkelvins):
dP
dT
= k
P
T
2
wherek isaconstant. EstimatedP/dT for T = 303, 313, 323, 333, 343usingthedataandtheapproximation
dP
dT
P(T +10) P(T 10)

20
T (K) 293 303 313 323 333 343 353
P (atm) 0.0278 0.0482 0.0808 0.1311 0.2067 0.3173 0.4754
Doyour estimatesseemtoconrmtheClausiusClapeyronLaw?What istheapproximatevalueof k?
solution Usingtheindicatedapproximationtotherst derivative, wecalculate
P
(303)
P(313) P(293)
20
=
0.08080.0278
20
= 0.00265atm/K;
P
(313)
P(323) P(303)
20
=
0.13110.0482
20
= 0.004145atm/K;
P
(323)
P(333) P(313)
20
=
0.20670.0808
20
= 0.006295atm/K;
P
(333)
P(343) P(323)
20
=
0.31730.1311
20
= 0.00931atm/K;
P
(343)
P(353) P(333)
20
=
0.47540.2067
20
= 0.013435atm/K
If theClausiusClapeyronlawis valid, then
T
2
P
dP
dT
shouldremainconstant as T varies. Usingthedatafor vapor
pressureandtemperatureandtheapproximatederivativevaluescalculatedabove, wend
T (K) 303 313 323 333 343
T
2
P
dP
dT
5047.59 5025.76 5009.54 4994.57 4981.45
Thesevaluesareroughlyconstant, suggestingthat theClausiusClapeyronlawisvalid, andthat k 5000.
Let L bethetangent lineto thehyperbolaxy = 1at x = a, wherea > 0. Showthat theareaof thetriangle
boundedbyL andthecoordinateaxesdoesnot dependona.
65. Inthesettingof Exercise64, showthat thepoint of tangency is themidpoint of thesegment of L lyingintherst
quadrant.
solution In theprevious exercise, wesawthat thetangent lineto thehyperbolaxy = 1 or y =
1
x
at x = a has
y-intercept P = (0,
2
a
) andx-intercept Q = (2a, 0). Themidpoint of thelinesegment connectingP andQisthus
_
0+2a
2
,
2
a
+0
2
_
=
_
a,
1
a
_
,
whichisthepoint of tangency.
Matchfunctions(A)(C) withtheir derivatives(I)(III) inFigure17.
67. Makearoughsketchof thegraphof thederivativeof thefunctioninFigure18(A).
(A) (B)
y = x
2
4 3
4 3 2 1
2 0 1 1
3
2
1
2
x
y
x
y
FIGURE 18
solution Thegraphhas atangent linewithnegativeslopeapproximately ontheinterval (1, 3.6), andhas atangent
linewithapositiveslopeelsewhere. Thisimpliesthatthederivativemustbenegativeontheinterval (1, 3.6) andpositive
elsewhere. Thegraphmaythereforelooklikethis:
y
x
1 2 3 4
Graphthederivativeof thefunctioninFigure18(B), omittingpointswherethederivativeisnot dened.
69. Sketchthegraphof f (x) = x |x|. Thenshowthat f
(0) exists.
solution For x < 0, f (x) = x
2
, andf
(x) = 2x. For x > 0, f (x) = x

2
, andf
(x) = 2x. At x = 0, wend

lim
h0+
f (0+h) f (0)
h
= lim
h0+
h
2
h
= 0
and
lim
h0
f (0+h) f (0)
h
= lim
h0
h
2
h
= 0.
Becausethetwoone-sidedlimitsexist andareequal, it followsthat f
(0) existsandisequal tozero. Hereisthegraph

of f (x) = x|x|.
y
x
1 2 1 2
2
4
4
2
Determinethevaluesof x at whichthefunctioninFigure19is: (a) discontinuous, and(b) nondifferentiable.
In Exercises 7176, nd the points c (if any) such that f
(c) does not exist.

71. f (x) = |x 1|
solution
y
x
1
0.5
1
1.5
2
2 3 1
Hereisthegraphof f (x) = |x 1|. Itsderivativedoesnot exist at x = 1. At that valueof x thereisasharpcorner.
f (x) = [x]
73. f (x) = x
2/3
solution Hereisthegraphof f (x) = x
2/3
. Itsderivativedoesnot exist at x = 0. At that valueof x, thereisasharp
corner or cusp.
y
x
1
1
1.5
2 1 2
f (x) = x
3/2
75. f (x) = |x
2
1|
solution Hereisthegraphof f (x) =
x
2
1
. Itsderivativedoesnot exist at x = 1or at x = 1. At thesevalues

of x, thegraphhassharpcorners.
y
x
1
2
1
3
2 1 2
f (x) = |x 1|
2
In Exercises 7782, zoom in on a plot of f (x) at the point (a, f (a)) and state whether or not f (x) appears to be
differentiable at x = a. If it is nondifferentiable, state whether the tangent line appears to be vertical or does not exist.
77. f (x) = (x 1)|x|, a = 0
solution Thegraphof f (x) = (x 1)|x| for x near 0isshownbelow. Becausethegraphhasasharpcorner atx = 0,
it appearsthat f isnot differentiableat x = 0. Moreover, thetangent linedoesnot exist at thispoint.
y
x
0.1 0.2 0.1 0.2
0.1
0.2
0.3
f (x) = (x 3)
5/3
, a = 3
79. f (x) = (x 3)
1/3
, a = 3
solution Thegraph of f (x) = (x 3)
1/3
for x near 3 is shown below. Fromthis graph, it appears that f is not
differentiableat x = 3. Moreover, thetangent lineappearstobevertical.
3.05 3.1 2.95 3 2.9
f (x) = sin(x
1/3
), a = 0
81. f (x) = | sinx|, a = 0
solution Thegraphof f (x) = | sinx| for x near 0isshownbelow. Becausethegraphhasasharpcorner at x = 0, it
appearsthat f isnot differentiableat x = 0. Moreover, thetangent linedoesnot exist at thispoint.
y
x
0.05
0.08
0.04
0.1
0.1 0.05 0.1
f (x) = |x sinx|, a = 0
83. Plotthederivativef
(x) of f (x) = 2x
3
10x
1
forx > 0(settheboundsof theviewingboxappropriately)
andobservethatf
(x) > 0. Whatdoesthepositivityof f
(x) tell usaboutthegraphof f (x) itself?Plotf (x) andconrm

thisconclusion.
3
10x
1
. Thenf
(x) = 6x
2
+10x
2
. Thegraphof f
(x) isshowninthegurebelowat
theleft andit isclear that f
(x) > 0for all x > 0. Thepositivityof f
(x) tellsusthat thegraphof f (x) isincreasingfor

x > 0. Thisisconrmedinthegurebelowat theright, whichshowsthegraphof f (x).
8 6 4 2
x
y
100
200
300
400
8 6 4 2
x
y
200
200
400
600
800
Findthecoordinatesof thepoint P inFigure20at whichthetangent linepassesthrough(5, 0).
Exercises 8588 refer to Figure 21. Length QR is called the subtangent at P, and length RT is called the subnormal.
x
y
P = (x, f (x))
T R
y = f (x)
Q
Tangent line
FIGURE 21
85. Calculatethesubtangent of
f (x) = x
2
+3x at x = 2
2
+3x. Thenf
(x) = 2x +3, andtheequationof thetangent lineat x = 2is

y = f
(2)(x 2) +f (2) = 7(x 2) +10= 7x 4.

Thislineintersectsthex-axisat x =
4
7
. ThusQhascoordinates(
4
7
, 0), R hascoordinates(2, 0) andthesubtangent is
2
4
7
=
10
7
.
Calculatethesubnormal of f (x) = x
2/3
at x = 8.
87. Proveingeneral that thesubnormal at P is|f
(x)f (x)|.
solution Theslopeof thetangent lineat P isf
(x). Theslopeof thelinenormal tothegraphat P isthen1/f
(x),
andthenormal lineintersectsthex-axisatthepointT withcoordinates(x +f (x)f
(x), 0). ThepointR hascoordinates

(x, 0), sothesubnormal is
|x +f (x)f
(x) x| = |f (x)f
(x)|.
Showthat PQhaslength|f (x)|
_
1+f
(x)
2
.
89. ProvethefollowingtheoremofApolloniusof Perga(theGreekmathematicianbornin262bce whogavetheparabola,
ellipse, andhyperbolatheir names): Thesubtangent of theparabolay = x
2
at x = a isequal toa/2.
2
. Thetangent linetof at x = a is
y = f
(a)(x a) +f (a) = 2a(x a) +a

2
= 2ax a
2
.
Thex-intercept of thisline(wherey = 0) is
a
2
asclaimed.
y
y = x
2
(a, a
2
)
x
(
, 0
)
a
2
Showthat thesubtangent toy = x
3
at x = a isequal to
1
3
a.
91. Formulateandproveageneralizationof Exercise90for y = x
n
.
n
. Thenf
(x) = nx
n1
, andtheequationof thetangent linet x = a is
y = f
(a)(x a) +f (a) = na
n1
(x a) +a
n
= na
n1
x (n 1)a
n
.
Thislineintersectsthex-axisat x = (n 1)a/n. Thus, Qhascoordinates((n 1)a/n, 0), R hascoordinates(a, 0) and
thesubtangent is
a
n 1
n
a =
1
n
a.
Twosmall archeshavetheshapeof parabolas. Therstisgivenbyf (x) = 1x
2
for1 x 1andthesecond
byg(x) = 4(x 4)
2
for 2 x 6. A boardisplacedontopof thesearchessoit restsonboth(Figure22). What
istheslopeof theboard? Hint: Findthetangent linetoy = f (x) that intersectsy = g(x) inexactlyonepoint.
FIGURE 22
93. A vaseisformedbyrotatingy = x
2
aroundthey-axis. If wedropinamarble, it will either touchthebottompoint
of thevaseor besuspendedabovethebottombytouchingthesides(Figure23). Howsmall must themarblebetotouch
thebottom?
FIGURE 23
solution Supposeacircleistangenttotheparabolay = x
2
atthepoint(t, t
2
). Theslopeof theparabolaatthispoint
is2t , sotheslopeof theradiusof thecircleat thispoint is
1
2t
(sinceit isperpendicular tothetangent lineof thecircle).
Thusthecenter of thecirclemust bewherethelinegivenby y =
1
2t
(x t ) + t
2
crossesthey-axis. Wecanndthe
y-coordinatebysettingx = 0: weget y =
1
2
+t
2
. Thus, theradiusextendsfrom(0,
1
2
+t
2
) to(t, t
2
) and
r =
_
_
1
2
+t
2
t
2
_
2
+t
2
=
_
1
4
+t
2
.
Thisradiusisgreater than
1
2
whenever t > 0; so, if amarblehasradius> 1/2it sitsontheedgeof thevase, but if it has
radius 1/2it rollsall thewaytothebottom.
Letf (x) beadifferentiablefunction, andsetg(x) = f (x +c), wherec isaconstant. Usethelimitdenitionto
showthat g
(x) = f
(x +c). Explainthisresult graphically, recallingthat thegraphof g(x) isobtainedbyshifting

thegraphof f (x) c unitstotheleft (if c > 0) or right (if c < 0).
95. NegativeExponents Letn beawholenumber. UsethePowerRuleforx
n
tocalculatethederivativeof f (x) = x
n
byshowingthat
f (x +h) f (x)
h
=
1
x
n
(x +h)
n
(x +h)
n
x
n
h
n
wheren isapositiveinteger.
Thedifferencequotient for f is
f (x +h) f (x)
h
=
(x +h)
n
x
n
h
=
1
(x+h)
n

1
x
n
h
=
x
n
(x+h)
n
x
n
(x+h)
n
h
=
1
x
n
(x +h)
n
(x +h)
n
x
n
h
.
Therefore,
f
(x) = lim
h0
f (x +h) f (x)
h
= lim
h0
1
x
n
(x +h)
n
(x +h)
n
x
n
h
= lim
h0
1
x
n
(x +h)
n
lim
h0
(x +h)
n
x
n
h
= x
2n
d
dx
_
x
n
_
.
Fromabove, wecontinue: f
(x) = x
2n
d
dx
_
x
n
_
= x
2n
nx
n1
= nx
n1
. Sincen isapositiveinteger,
k = n is anegativeinteger andwehave
d
dx
_
x
k
_
=
d
dx
_
x
n
_
= nx
n1
= kx
k1
; i.e.
d
dx
_
x
k
_
= kx
k1
for negativeintegersk.
VerifythePower Rulefor theexponent 1/n, wheren isapositiveinteger, usingthefollowingtrick: Rewritethe
differencequotient for y = x
1/n
at x = b intermsof u = (b +h)
1/n
anda = b
1/n
.
97. InnitelyRapidOscillations Dene
f (x) =
x sin
1
x
x = 0
0 x = 0
Showthat f (x) iscontinuousat x = 0but f
(0) doesnot exist (seeFigure22).

_
x sin
_
1
x
_
if x = 0
0 if x = 0
. Asx 0,
|f (x) f (0)| =
x sin
_
1
x
_
0
= |x|
sin
_
1
x
_
0
sincethevalues of thesineliebetween 1 and 1. Hence, by theSqueezeTheorem, lim
x0
f (x) = f (0) and thus f is
continuousat x = 0.
Asx 0, thedifferencequotient at x = 0,
f (x) f (0)
x 0
=
x sin
_
1
x
_
0
x 0
= sin
_
1
x
_
doesnot convergetoalimit sinceit oscillatesinnitelythrougheveryvaluebetween1and1. Accordingly, f
(0) does
not exist.
For whichvaluesof c doestheequationx
2
+4= cx haveauniquesolution? Hint: Drawagraph.
3.3 Product and Quotient Rules
1. Arethefollowingstatementstrueor false? If false, statethecorrect version.
(a) fg denotesthefunctionwhosevalueat x isf (g(x)).
(b) f/g denotesthefunctionwhosevalueat x isf (x)/g(x).
(c) Thederivativeof theproduct istheproduct of thederivatives.
(d)
d
dx
(fg)
x=4
= f (4)g
(4) g(4)f
(4)
(e)
d
dx
(fg)
x=0
= f (0)g
(0) +g(0)f
(0)
solution
(a) False. Thenotationfg denotesthefunctionwhosevalueat x isf (x)g(x).
(b) True.
(c) False. Thederivativeof aproduct fg isf
(x)g(x) +f (x)g
(x).
(d) False.
d
dx
(fg)
x=4
= f (4)g
(4) +g(4)f
(4).
(e) True.
2. Find(f/g)
(1) if f (1) = f
(1) = g(1) = 2andg
(1) = 4.
solution
d
dx
(f/g)
x=1
= [g(1)f
(1) f (1)g
(1)]/g(1)
2
= [2(2) 2(4)]/2
2
= 1.
3. Findg(1) if f (1) = 0, f
(1) = 2, and(fg)
(1) = 10.
solution (fg)
(1) = f (1)g
(1) +f
(1)g(1), so10= 0 g
(1) +2g(1) andg(1) = 5.

Exercises
In Exercises 16, use the Product Rule to calculate the derivative.
1. f (x) = x
3
(2x
2
+1)
3
(2x
2
+1). Then
f
(x) = x
3
d
dx
(2x
2
+1) +(2x
2
+1)
d
dx
x
3
= x
3
(4x) +(2x
2
+1)(3x
2
) = 10x
4
+3x
2
.
f (x) = (3x 5)(2x
2
3)
3. f (x) =

x(1x
3
)

x(1x
3
). Then
f
(x) =

x
d
dx
(1x
3
) +(1x
3
)
d
dx
x =

x(3x
2
) +(1x
3
)
_
1
2
x
1/2
_
= 3x
5/2
+
1
2
x
1/2
1
2
x
5/2
=
7
2
x
5/2
+
1
2x
1/2
S E C T I ON 3.3 Product and Quotient Rules 113
f (x) = (3x
4
+2x
6
)(x 2)
5.
dh
ds
s=4
, h(s) = (s
1/2
+2s)(7s
1
)
solution Let h(s) = (s
1/2
+2s)(7s
1
). Then
dh
ds
= (s
1/2
+2s)
d
dx
(7s
1
) +(7s
1
)
d
ds
_
s
1/2
+2s
_
= (s
1/2
+2s)(s
2
) +(7s
1
)
_
1
2
s
3/2
+2
_
=
7
2
s
3/2
+
3
2
s
5/2
+14.
Therefore,
dh
ds
s=4
=
7
2
(4)
3/2
+
3
2
(4)
5/2
+14=
871
64
.
y = (t 8t
1
)(t +t
2
)
In Exercises 712, use the Quotient Rule to calculate the derivative.
7. f (x) =
x
x 2
x
x2
. Then
f
(x) =
(x 2)
d
dx
x x
d
dx
(x 2)
(x 2)
2
=
(x 2) x
(x 2)
2
=
2
(x 2)
2
.
f (x) =
x +4
x
2
+x +1
9.
dg
dt
t =2
, g(t ) =
t
2
+1
t
2
1
solution Let g(t ) =
t
2
+1
t
2
1
. Then
dg
dt
=
(t
2
1)
d
dt
(t
2
+1) (t
2
+1)
d
dt
(t
2
1)
(t
2
1)
2
=
(t
2
1)(2t ) (t
2
+1)(2t )
(t
2
1)
2
=
4t
(t
2
1)
2
.
Therefore,
dg
dt
t =2
=
4(2)
((2)
2
1)
2
=
8
9
.
dw
dz
z=9
, w =
z
2
z +z
11. g(x) =
1
1+x
3/2
solution
g
(x) =
(1+x
3/2
)
d
dt
(1) 1
d
dt
(1+x
3/2
)
(1+x
3/2
)
2
=
3
2
x
1/2
(1+x
3/2
)
2
=
3
x
2(1+x
3/2
)
2
h(s) =
s
3/2
s
2
+1
In Exercises 1316, calculate the derivative in two ways. First use the Product or Quotient Rule; then rewrite the function
algebraically and apply the Power Rule directly.
13. f (t ) = (2t +1)(t
2
2)
solution Let f (t ) = (2t +1)(t
2
2). Then, usingtheProduct Rule,
f
(t ) = (2t +1)(2t ) +(t

2
2)(2) = 6t
2
+2t 4.
Multiplyingout rst, wendf (t ) = 2t
3
+t
2
4t 2. Therefore, f
(t ) = 6t
2
+2t 4.
f (x) = x
2
(3+x
1
) 15. h(t ) =
t
2
1
t 1
solution Let h(t ) =
t
2
1
t 1
. Usingthequotient rule,
f
(t ) =
(t 1)(2t ) (t
2
1)(1)
(t 1)
2
=
t
2
2t +1
(t 1)
2
= 1
for t = 1. Simplifyingrst, wendfor t = 1,
h(t ) =
(t 1)(t +1)
(t 1)
= t +1.
Henceh
(t ) = 1for t = 1.
g(x) =
x
3
+2x
2
+3x
1
x
17. f (x) = (x
3
+5)(x
3
+x +1)
solution Let f (x) = (x
3
+5)(x
3
+x +1). Then
f
(x) = (x
3
+5)(3x
2
+1) +(x
3
+x +1)(3x
2
) = 6x
5
+4x
3
+18x
2
+5.
f (x) =
_
1
x
x
2
_
(x
3
+1)
19.
dy
dx
x=3
, y =
1
x +10
solution Let y =
1
x+10
. Usingthequotient rule:
dy
dx
=
(x +10)(0) 1(1)
(x +10)
2
=
1
(x +10)
2
.
Therefore,
dy
dx
x=3
=
1
(3+10)
2
=
1
169
.
dz
dx
x=2
, z =
x
3x
2
+1
21. f (x) = (
x +1)(
x 1)
solution Let f (x) = (
x + 1)(
x 1). Multiplying through rst yields f (x) = x 1 for x 0. Therefore,

f
(x) = 1for x 0. If wecarryout theproduct ruleonf (x) = (x

1/2
+1)(x
1/2
1), weget
f
(x) = (x
1/2
+1)
_
1
2
(x
1/2
)
_
+(x
1/2
1)
_
1
2
x
1/2
_
=
1
2
+
1
2
x
1/2
+
1
2
1
2
x
1/2
= 1.
f (x) =
9x
5/2
2
x
23.
dy
dx
x=2
, y =
x
4
4
x
2
5
solution Let y =
x
4
4
x
2
5
. Then
dy
dx
=
_
x
2
5
_ _
4x
3
_
_
x
4
4
_
(2x)
_
x
2
5
_
2
=
2x
5
20x
3
+8x
_
x
2
5
_
2
.
Therefore,
dy
dx
x=2
=
2(2)
5
20(2)
3
+8(2)
(2
2
5)
2
= 80.
f (x) =
x
4
+x
1
x +1
25.
dz
dx
x=1
, z =
1
x
3
+1
solution Let z =
1
x
3
+1
dz
dx
=
(x
3
+1)(0) 1(3x
2
)
(x
3
+1)
2
=
3x
2
(x
3
+1)
2
.
Therefore,
dz
dx
x=1
=
3(1)
2
(1
3
+1)
2
=
3
4
.
f (x) =
3x
3
x
2
+2
x
27. h(t ) =
t
(t +1)(t
2
+1)
solution Let h(t ) =
t
(t +1)(t
2
+1)
=
t
t
3
+t
2
+t +1
. Then
h
(t ) =
_
t
3
+t
2
+t +1
_
(1) t
_
3t
2
+2t +1
_
_
t
3
+t
2
+t +1
_
2
=
2t
3
t
2
+1
_
t
3
+t
2
+t +1
_
2
.
f (x) = x
3/2
_
2x
4
3x +x
1/2
_ 29. f (t ) = 3
1/2
5
1/2

3
5. Thenf
(t ) = 0, sincef (t ) isaconstant function!

h(x) =
2
(x 1)
31. f (x) = (x +3)(x 1)(x 5)
solution Let f (x) = (x + 3)(x 1)(x 5). UsingtheProduct RuleinsidetheProduct Rulewitharst factor of
(x +3) andasecondfactor of (x 1)(x 5), wend
f
(x) = (x +3) ((x 1)(1) +(x 5)(1)) +(x 1)(x 5)(1) = 3x

2
6x 13.
Alternatively,
f (x) = (x +3)
_
x
2
6x +5
_
= x
3
3x
2
13x +15.
Therefore, f
(x) = 3x
2
6x 13.
h(s) = s(s +4)(s
2
+1) 33. f (x) =
x
3/2
(x
2
+1)
x +1
solution Usingthequotient rule, andthenusingtheproduct ruletodifferentiatethenumerator, wend
f
(x) =
(x +1)
_
3
2
x
1/2
(x
2
+1) +x
3/2
(2x)
_
x
3/2
(x
2
+1)(1)
(x +1)
2
=
5x
7/2
+7x
5/2
+x
3/2
+3x
1/2
2(x +1)
2
g(z) =
(z 2)(z
2
+1)
z
35. g(z) =
_
z
2
4
z 1
__
z
2
1
z +2
_
Hint: Simplifyrst.
solution Let
g(z) =
_
z
2
4
z 1
__
z
2
1
z +2
_
=
_
(z +2)(z 2)
z 1
__
(z +1)(z 1)
z +2
_
= (z 2)(z +1)
for z = 2andz = 1. Then,
g
(z) = (z +1)(1) +(z 2)(1) = 2z 1.

d
dx
_
(ax +b)(abx
2
+1)
_
(a, b constants)
37.
d
dt
_
xt 4
t
2
x
_
(x constant)
xt 4
t
2
x
f
(t ) =
(t
2
x)(x) (xt 4)(2t )
(t
2
x)
2
=
xt
2
x
2
2xt
2
+8t
(t
2
x)
2
=
xt
2
+8t x
2
(t
2
x)
2
.
d
dx
_
ax +b
cx +d
_
(a, b, c, d constants)
In Exercises 3942, calculate the derivative using the values:
f (4) f
(4) g(4) g
(4)
10 2 5 1
39. (fg)
(4) and(f/g)
(4).
solution Let h = fg andH = f/g. Thenh
= fg
+gf
andH
=
gf
fg
g
2
. Finally,
h
(4) = f (4)g
(4) +g(4)f
(4) = (10)(1) +(5)(2) = 20,

and
H
(4) =
g(4)f
(4) f (4)g
(4)
(g(4))
2
=
(5)(2) (10)(1)
(5)
2
= 0.
F
(4), whereF(x) = x
2
f (x).
41. G
(4), whereG(x) = g(x)

2
.
solution Let G(x) = g(x)
2
= g(x)g(x). ThenG
(x) = g(x)g
(x) +g(x)g
(x) = 2g(x)g
(x), and
G
(4) = 2g(4)g
(4) = 2(5)(1) = 10.

H
(4), whereH(x) =
x
g(x)f (x)
.
43. CalculateF
(0), where
F(x) =
x
9
+x
8
+4x
5
7x
x
4
3x
2
+2x +1
Hint: DonotcalculateF
(x). Instead, writeF(x) = f (x)/g(x) andexpressF
(0) directlyintermsof f (0), f
(0), g(0),
g
(0).
solution Takingthehint, let
f (x) = x
9
+x
8
+4x
5
7x
andlet
g(x) = x
4
3x
2
+2x +1.
ThenF(x) =
f (x)
g(x)
. Now,
f
(x) = 9x
8
+8x
7
+20x
4
7 and g
(x) = 4x
3
6x +2.
Moreover, f (0) = 0, f
(0) = 7, g(0) = 1, andg
(0) = 2.
Usingthequotient rule:
F
(0) =
g(0)f
(0) f (0)g
(0)
(g(0))
2
=
70
1
= 7.
ProceedasinExercise43tocalculateF
(0), where
F(x) =
_
1+x +x
4/3
+x
5/3
_ 3x
5
+5x
4
+5x +1
8x
9
7x
4
+1
45. Verifytheformula(x
3
)
= 3x
2
bywritingx
3
= x x x andapplyingtheProduct Rule.
solution Usingtheproduct rule, wehave
(x
3
)
= ((x x) x)
= (x x) x
+(x x)
x = (x x) +(x x
+x
x) x = (x x) +(2x) x = 3x
2
Plot the derivative of f (x) = x/(x
2
+1) over [4, 4]. Use the graph to determine the intervals on which
f
(x) > 0andf
(x) < 0. Thenplot f (x) anddescribehowthesignof f
(x) isreectedinthegraphof f (x).

47. Plot f (x) = x/(x
2
1) (in asuitably bounded viewing box). Usetheplot to determinewhether f
(x) is
positiveor negativeonitsdomain{x : x = 1}. Thencomputef
(x) andconrmyour conclusionalgebraically.

x
x
2
1
. Thegraphof f (x) isshownbelow. Fromthisplot, weseethat f (x) isdecreasingon
itsdomain{x : x = 1}. Consequently, f
(x) must benegative. Usingthequotient rule, wend

f
(x) =
(x
2
1)(1) x(2x)
(x
2
1)
2
=
x
2
+1
(x
2
1)
2
,
whichisnegativefor all x = 1.
4 3 2 1
x
y
5
5
1 2 3 4
Let P = V
2
R/(R +r)
2
asinExample7. CalculatedP/dr, assumingthat r isvariableandR isconstant.
49. Findall valuesof a suchthat thetangent lineto
f (x) =
x 1
x +8
at x = a
passesthroughtheorigin(Figure4).
8 4 4 8
y
x
FIGURE 4
solution Theslopeof thetangent lineisgivenbyf
(x), whichis
f
(x) =
(x +8)(1) (x 1)(1)
(x +8)
2
=
9
(x +8)
2
Whenx = a, then, wehavef (a) =
a 1
a +8
andf
(a) =
9
(a+8)
2
, sothat theequationof thetangent lineis
y = f
(a)(x a) +f (a) =
9
(a +8)
2
(x a) +
a 1
a +8
=
2
(a +8)
2
x +
9a +(a 1)(a +8)
(a +8)
2
=
2
(a +8)
2
x +
a
2
2a 8
(a +8)
2
Thetangent linepassesthroughtheoriginwhenitsconstant termiszero, i.e. whena
2
2a 8= 0, or for a = 4, 2.
CurrentI (amperes), voltageV (volts), andresistanceR (ohms) inacircuitarerelatedbyOhmsLaw, I = V/R.
(a) Calculate
dI
dR
R=6
if V isconstant withvalueV = 24.
(b) Calculate
dV
dR
R=6
if I isconstant withvalueI = 4.
51. Therevenueper monthearnedbytheCoutureclothingchainattimet isR(t ) = N(t )S(t ), whereN(t ) isthenumber
of stores andS(t ) is averagerevenueper storeper month. Coutureembarks onatwo-part campaign: (A) to buildnew
storesatarateof 5storesper month, and(B) touseadvertisingtoincreaseaveragerevenueper storeatarateof $10,000
per month. Assumethat N(0) = 50andS(0) = $150,000.
(a) Showthat total revenuewill increaseat therate
dR
dt
= 5S(t ) +10,000N(t )
Notethat thetwotermsintheProduct Rulecorrespondtotheseparateeffectsof increasingthenumber of storesonthe
onehand, andtheaveragerevenueper storeontheother.
(b) Calculate
dR
dt
t =0
.
(c) If Couturecanimplement only oneleg(A or B) of its expansionat t = 0, whichchoicewill growrevenuemost
rapidly?
solution
(a) GivenR(t ) = N(t )S(t ), it followsthat
dR
dt
= N(t )S
(t ) +S(t )N
(t ).
Wearetoldthat N
(t ) = 5storesper monthandS
(t ) = 10,000dollarsper month. Therefore,

dR
dt
= 5S(t ) +10,000N(t ).
(b) Usingpart (a) andthegivenvaluesof N(0) andS(0), wend
dR
dt
t =0
= 5(150,000) +10,000(50) = 1,250,000.
(c) Frompart (b), weseethat of thetwo terms contributingto total revenuegrowth, theterm5S(0) is larger thanthe
term10,000N(0). Thus, if onlyonelegof thecampaigncanbeimplemented, it shouldbepartA: increasethenumber of
storesby5per month.
Thetipspeedratioof aturbine(Figure5) istheratioR = T/W, whereT isthespeedof thetipof abladeand
W isthespeedof thewind. (Engineershavefoundempiricallythataturbinewithn bladesextractsmaximumpower
fromthewindwhenR = 2/n.) CalculatedR/dt (t inminutes) if W = 35km/handW decreases at arateof 4
km/hper minute, andthetipspeedhasconstant valueT = 150km/h.
53. The curve y = 1/(x
2
+1) is called the witch of Agnesi (Figure 6) after the Italian mathematician MariaAgnesi
(17181799), whowroteoneof therstbooksoncalculus. Thisstrangenameistheresultof amistranslationof theItalian
wordla versiera, meaningthat whichturns. Findequationsof thetangent linesat x = 1.
3 2 1 2 3 1
1
x
y
FIGURE 6 Thewitchof Agnesi.
1
x
2
+1
. Thenf
(x) =
(x
2
+1)(0) 1(2x)
(x
2
+1)
2
=
2x
_
x
2
+1
_
2
.
At x = 1, thetangent lineis
y = f
(1)(x +1) +f (1) =

1
2
(x +1) +
1
2
=
1
2
x +1.
At x = 1, thetangent lineis
y = f
(1)(x 1) +f (1) =
1
2
(x 1) +
1
2
=
1
2
x +1.
Let f (x) = g(x) = x. Showthat (f/g)
= f
/g
.
55. UsetheProduct Ruletoshowthat (f
2
)
= 2ff
.
solution Let g = f
2
= ff . Theng
=
_
f
2
_
= (ff )
= ff
+ff
= 2ff
.
Showthat (f
3
)
= 3f
2
f
.
57. Let f , g, h bedifferentiablefunctions. Showthat (fgh)
(x) isequal to
f (x)g(x)h
(x) +f (x)g
(x)h(x) +f
(x)g(x)h(x)
Hint: Writefgh asf (gh).
solution Let p = fgh. Then
p
= (fgh)
= f
_
gh
+hg
_
+ghf
= f
gh +fg
h +fgh
.
ProvetheQuotient Ruleusingthelimit denitionof thederivative.
59. Derivativeof theReciprocal Usethelimit denitiontoprove
d
dx
_
1
f (x)
_
=
f
(x)
f
2
(x)
7
Hint: Showthat thedifferencequotient for 1/f (x) isequal to
f (x) f (x +h)
hf (x)f (x +h)
solution Let g(x) =
1
f (x)
. Wethencomputethederivativeof g(x) usingthedifferencequotient:
g
(x) = lim
h0
g(x +h) g(x)
h
= lim
h0
1
h
_
1
f (x +h)
1
f (x)
_
= lim
h0
1
h
_
f (x) f (x +h)
f (x)f (x +h)
_
= lim
h0
_
f (x +h) f (x)
h
__
1
f (x)f (x +h)
_
.
Wecan apply theruleof products for limits. Therst parenthetical expression is thedifferencequotient denition of
f
(x). Thesecondcanbeevaluatedat h = 0togive

1
(f (x))
2
. Hence
g
(x) =
d
dx
_
1
f (x)
_
=
f
(x)
f
2
(x)
.
ProvetheQuotient RuleusingEq. (7) andtheProduct Rule.
61. Usethelimit denitionof thederivativetoprovethefollowingspecial caseof theProduct Rule:
d
dx
(xf (x)) = xf
(x) +f (x)
solution First notethat becausef (x) isdifferentiable, it isalsocontinuous. It followsthat
lim
h0
f (x +h) = f (x).
Nowwetacklethederivative:
d
dx
(xf (x)) = lim
h0
(x +h)f (x +h) f (x)
h
= lim
h0
_
x
f (x +h) f (x)
h
+f (x +h)
_
= x lim
h0
f (x +h) f (x)
h
+ lim
h0
f (x +h)
= xf
(x) +f (x).
Carryout MariaAgnesisproof of theQuotient Rulefromher book oncalculus, publishedin1748: Assumethat
f , g, andh = f/g aredifferentiable. Computethederivativeof hg = f usingtheProduct Rule, andsolvefor h
.
63. ThePower RuleRevisited If youarefamiliar withproof by induction, useinductiontoprovethePower Rulefor
all wholenumbersn. Showthat thePower Ruleholdsfor n = 1; thenwritex
n
asx x
n1
andusetheProduct Rule.
solution Let k beapositiveinteger. If k = 1, thenx
k
= x. Notethat
d
dx
_
x
1
_
=
d
dx
(x) = 1= 1x
0
.
S E C T I ON 3.4 Rates of Change 119
HencethePower Ruleholdsfor k = 1. Assumeit holdsfor k = n wheren 2. Thenfor k = n +1, wehave
d
dx
_
x
k
_
=
d
dx
_
x
n+1
_
=
d
dx
_
x x
n
_
= x
d
dx
_
x
n
_
+x
n
d
dx
(x)
= x nx
n1
+x
n
1= (n +1)x
n
= kx
k1
Accordingly, thePower Ruleholdsfor all positiveintegersbyinduction.
Exercises 64 and 65: A basic fact of algebra states that c is a root of a polynomial f (x) if and only if f (x) = (x c)g(x)
for some polynomial g(x). We say that c is a multiple root if f (x) = (x c)
2
h(x), where h(x) is a polynomial.
Showthat c isamultipleroot of f (x) if andonlyif c isaroot of bothf (x) andf
(x).
65. UseExercise64todeterminewhether c = 1isamultipleroot:
(a) x
5
+2x
4
4x
3
8x
2
x +2
(b) x
4
+x
3
5x
2
3x +2
solution
(a) Toshowthat 1isamultipleroot of
f (x) = x
5
+2x
4
4x
3
8x
2
x +2,
it sufcestocheckthat f (1) = f
(1) = 0. Wehavef (1) = 1+2+48+1+2= 0and

f
(x) = 5x
4
+8x
3
12x
2
16x 1
f
(1) = 5812+161= 0
(b) Let f (x) = x
4
+x
3
5x
2
3x +2. Thenf
(x) = 4x
3
+3x
2
10x 3. Because
f (1) = 115+3+2= 0
but
f
(1) = 4+3+103= 6= 0,
it followsthat x = 1isaroot of f , but not amultipleroot.
Figure7isthegraphof apolynomial withrootsatA, B, andC. Whichof theseisamultipleroot?Explainyour
reasoningusingExercise64.
3.4 Rates of Change
1. Whichunitsmight beusedfor eachrateof change?
(a) Pressure(inatmospheres) inawater tankwithrespect todepth
(b) Therateof achemical reaction(changeinconcentrationwithrespect totimewithconcentrationinmolesper liter)
solution
(a) Therateof changeof pressurewithrespect todepthmight bemeasuredinatmospheres/meter.
(b) Thereactionrateof achemical reactionmight bemeasuredinmoles/(literhour).
2. Twotrainstravel fromNewOrleanstoMemphisin4hours. Therst traintravelsat aconstant velocityof 90mph,
but thevelocityof thesecondtrainvaries. What wasthesecondtrainsaveragevelocityduringthetrip?
solution Sincebothtrainstravel thesamedistanceinthesameamount of time, theyhavethesameaveragevelocity:
90mph.
3. Estimatef (26), assumingthat f (25) = 43, f
(25) = 0.75.
solution f (x) f (25) +f
(25)(x 25), sof (26) 43+0.75(2625) = 43.75.

4. ThepopulationP(t ) of Freedoniain2009wasP(2009) = 5million.
(a) What isthemeaningof P
(2009)?
(b) EstimateP(2010) if P
(2009) = 0.2.
solution
(a) BecauseP(t ) measuresthepopulationof Freedoniaasafunctionof time, thederivativeP
(2009) measurestherate
of changeof thepopulationof Freedoniaintheyear 2009.
(b) P(2010) P(2009) +P
(2010). Thus, if P
(2009) = 0.2, thenP(2009) 5.2million.

Exercises
In Exercises 18, nd the rate of change.
1. Areaof asquarewithrespect toitssides whens = 5.
solution Let theareabeA = f (s) = s
2
. Thentherateof changeof A withrespect to s is d/ds(s
2
) = 2s. When
s = 5, theareachangesat arateof 10squareunitsper unit increase. (Drawa55squareongraphpaper andtracethe
areaaddedbyincreasingeachsidelengthby1, excludingthecorner, toseewhat thismeans.)
Volumeof acubewithrespect toitssides whens = 5.
3. Cuberoot
3
x withrespect tox whenx = 1, 8, 27.

3
x. Writingf (x) = x
1/3
, weseetherateof changeof f (x) withrespect to x is givenby
f
(x) =
1
3
x
2/3
. Therequestedratesof changearegiveninthetablethat follows:
c ROC of f (x) withrespect tox at x = c.
1 f
(1) =
1
3
(1) =
1
3
8 f
(8) =
1
3
(8
2/3
) =
1
3
(
1
4
) =
1
12
27 f
(27) =
1
3
(27
2/3
) =
1
3
(
1
9
) =
1
27
Thereciprocal 1/x withrespect tox whenx = 1, 2, 3.
5. Thediameter of acirclewithrespect toradius.
solution Therelationshipbetweenthediameter d of acircleandits radius r is d = 2r. Therateof changeof the
diameter withrespect totheradiusisthend
= 2.
SurfaceareaA of aspherewithrespect toradiusr (A = 4r
2
).
7. VolumeV of acylinder withrespect toradiusif theheight isequal totheradius.
solution Thevolumeof thecylinder isV = r
2
h = r
3
. ThusdV/dr = 3r
2
.
Speedof soundv (inm/s) withrespect toair temperatureT (inkelvins), wherev = 20
T .
In Exercises 911, refer to Figure 10, the graph of distance s(t ) from the origin as a function of time for a car trip.
t (h)
3.0 2.5 2.0 1.5 1.0 0.5
150
100
50
Distance(km)
FIGURE 10 Distancefromtheoriginversustimefor acar trip.
9. Findtheaveragevelocityover eachinterval.
(a) [0, 0.5] (b) [0.5, 1] (c) [1, 1.5] (d) [1, 2]
solution
(a) Theaveragevelocityover theinterval [0, 0.5] is
500
0.50
= 100km/hour.
(b) Theaveragevelocityover theinterval [0.5, 1] is
10050
10.5
= 100km/hour.
(c) Theaveragevelocityover theinterval [1, 1.5] is
100100
1.51
= 0km/hour.
(d) Theaveragevelocityover theinterval [1, 2] is
50100
21
= 50km/hour.
At what timeisvelocityat amaximum?
11. Matchthedescriptions(i)(iii) withtheintervals(a)(c).
(i) Velocityincreasing
(ii) Velocitydecreasing
(iii) Velocitynegative
(a) [0, 0.5]
(b) [2.5, 3]
(c) [1.5, 2]
solution
(a) (i): Thedistancecurveisincreasing, andisalsobending upward, sothat distanceisincreasingat anincreasingrate.
(b) (ii) : Over theinterval [2.5, 3], thedistancecurveis attening, showing that thecar is slowing down; that is, the
velocityisdecreasing.
(c) (iii): Thedistancecurveisdecreasing, sothetangent linehasnegativeslope; thismeansthevelocityisnegative.
UsethedatafromTable1inExample1tocalculatetheaveragerateof changeof MartiantemperatureT with
respect totimet over theinterval from8:36am to9:34am.
13. UseFigure3fromExample1to estimatetheinstantaneous rateof changeof Martiantemperaturewithrespect to
time(indegreesCelsiusper hour) at t = 4am.
solution Thesegmentof thetemperaturegrapharoundt = 4am appearstobeastraightlinepassingthroughroughly
(1:36, 70) and(4:48, 75). Theinstantaneousrateof changeof Martiantemperaturewithrespect totimeat t = 4am
isthereforeapproximately
dT
dt
=
75(70)
3.2
= 1.5625
C/hour.
Thetemperature(in
C) of anobject at timet (inminutes) isT (t ) =

3
8
t
2
15t + 180for 0 t 20. At what
rateistheobject coolingat t = 10? (Givecorrect units.)
15. Thevelocity(incm/s)of bloodmoleculesowingthroughacapillaryof radius0.008cmisv = 6.410
8
0.001r
2
,
wherer isthedistancefromthemoleculetothecenter of thecapillary. Findtherateof changeof velocity withrespect
tor whenr = 0.004cm.
solution Therateof changeof thevelocityof thebloodmoleculesisv
(r) = 0.002r. Whenr = 0.004cm, thisrate

is810
6
cm/s.
Figure11displays thevoltageV across acapacitor as afunctionof timewhilethecapacitor is beingcharged.
Estimatetherateof changeof voltageat t = 20s. Indicatethevaluesinyour calculationandincludeproper units.
Doesvoltagechangemorequicklyor moreslowlyastimegoeson? Explainintermsof tangent lines.
17. UseFigure12toestimatedT /dh at h = 30and70, whereT isatmospherictemperature(indegreesCelsius) andh
isaltitude(inkilometers). WhereisdT /dh equal tozero?
T
r
o
p
o
s
p
h
e
r
e
S
t
r
a
t
o
s
p
h
e
r
e
M
e
s
o
s
p
h
e
r
e
T
h
e
r
m
o
s
p
h
e
r
e
Altitude(km)
Temperature
(C)
250
200
150
100
50
0
50
100
10 50 100 150
FIGURE 12 Atmospherictemperatureversusaltitude.
solution Ath = 30km,thegraphof atmospherictemperatureappearstobelinearpassingthroughthepoints(23, 50)
and(40, 0). Theslopeof thissegment of thegraphisthen
0(50)
4023
=
50
17
= 2.94;
so
dT
dh
h=30
2.94
C/km.
Ath = 70km,thegraphof atmospherictemperatureappearstobelinearpassingthroughthepoints(58, 0) and(88, 100).
Theslopeof thissegment of thegraphisthen
1000
8858
=
100
30
= 3.33;
so
dT
dh
h=70
3.33
C/km.
dT
dh
= 0atthosepointswherethetangentlineonthegraphishorizontal. Thisappearstohappenovertheinterval [13, 23],
andnear thepointsh = 50andh = 90.
Theearthexerts agravitational forceof F(r) = (2.9910
16
)/r
2
newtons onanobject withamass of 75kg
locatedr metersfromthecenter of theearth. Findtherateof changeof forcewithrespecttodistancer atthesurface
of theearth.
19. Calculatetherateof changeof escapevelocity v
esc
= (2.82 10
7
)r
1/2
m/s withrespect to distancer fromthe
center of theearth.
solution Theratethat escapevelocitychangesisv
esc
(r) = 1.4110
7
r
3/2
.
Thepower deliveredbyabatterytoanapparatusof resistanceR (inohms) isP = 2.25R/(R +0.5)
2
watts. Find
therateof changeof power withrespect toresistancefor R = 3 andR = 5.
21. Thepositionof aparticlemovinginastraight lineduringa5-stripiss(t ) = t
2
t +10cm. Findatimet at which
theinstantaneousvelocityisequal totheaveragevelocityfor theentiretrip.
solution Let s(t ) = t
2
t + 10, 0 t 5, withs incentimeters (cm) andt inseconds (s). Theaveragevelocity
over thet -interval [0, 5] is
s(5) s(0)
50
=
3010
5
= 4cm/s.
The (instantaneous) velocity is v(t ) = s
(t ) = 2t 1. Solving 2t 1 = 4 yields t =
5
2
s, the time at which the
instantaneousvelocityequalsthecalculatedaveragevelocity.
Theheight (inmeters) of ahelicopter at timet (inminutes) iss(t ) = 600t 3t
3
for 0 t 12.
(a) Plot s(t ) andvelocityv(t ).
(b) Findthevelocityat t = 8andt = 10.
(c) Findthemaximumheight of thehelicopter.
23. A particlemovingalongalinehaspositions(t ) = t
4
18t
2
mat timet seconds. At whichtimesdoestheparticle
passthroughtheorigin?At whichtimesistheparticleinstantaneouslymotionless(that is, it haszerovelocity)?
solution Theparticlepassesthroughtheoriginwhens(t ) = t
4
18t
2
= t
2
(t
2
18) = 0. Thishappenswhent = 0
secondsandwhent = 3
2 4.24seconds.Withs(t ) = t
4
18t
2
,itfollowsthatv(t ) = s
(t ) = 4t
3
36t = 4t (t
2
9).
Theparticleisthereforeinstantaneouslymotionlesswhent = 0secondsandwhent = 3seconds.
Plot thepositionof theparticleinExercise23. What isthefarthest distancetotheleft of theoriginattainedby
theparticle?
25. A bullet is redintheair vertically fromgroundlevel withaninitial velocity 200m/s. Findthebullets maximum
velocityandmaximumheight.
solution Weemploy Galileosformula, s(t ) = s
0
+ v
0
t
1
2
gt
2
= 200t 4.9t
2
, wherethetimet isinseconds(s)
andtheheight s isinmeters(m). Thevelocityisv(t ) = 2009.8t . Themaximumvelocityof 200m/soccursat t = 0.
Thisistheinitial velocity. Thebullet reachesitsmaximumheight whenv(t ) = 2009.8t = 0; i.e., whent 20.41s.
At thispoint, theheight is2040.82m.
Findthevelocityof anobject droppedfromaheight of 300mat themoment it hitstheground.
27. A ball tossedintheair verticallyfromgroundlevel returnstoearth4slater. Findtheinitial velocityandmaximum
height of theball.
solution Galileosformulagivess(t ) = s
0
+ v
0
t
1
2
gt
2
= v
0
t 4.9t
2
, wherethetimet isinseconds(s) andthe
height s isinmeters(m). Whentheball hitsthegroundafter 4secondsitsheight is0. Solve0= s(4) = 4v
0
4.9(4)
2
toobtainv
0
= 19.6m/s. Theball reachesitsmaximumheight whens
(t ) = 0, that is, when19.6 9.8t = 0, or t = 2

s. At thistime, t = 2s,
s(2) = 0+19.6(2)
1
2
(9.8)(4) = 19.6m.
Oliviaisgazingout awindowfromthetenthoor of abuildingwhenabucket (droppedby awindowwasher)
passesby. Shenotesthatithitstheground1.5slater. Determinetheoor fromwhichthebucketwasdroppedif each
oor is5mhighandthewindowisinthemiddleof thetenthoor. Neglect air friction.
29. Showthat for anobject fallingaccordingtoGalileosformula, theaveragevelocityover anytimeinterval [t
1
, t
2
] is
equal totheaverageof theinstantaneousvelocitiesat t
1
andt
2
.
solution Thesimplest waytoproceedistocomputebothvaluesandshowthat theyareequal. Theaveragevelocity
over [t
1
, t
2
] is
s(t
2
) s(t
1
)
t
2
t
1
=
(s
0
+v
0
t
2
1
2
gt
2
2
) (s
0
+v
0
t
1
1
2
gt
2
1
)
t
2
t
1
=
v
0
(t
2
t
1
) +
g
2
(t
2
2
t
1
2
)
t
2
t
1
=
v
0
(t
2
t
1
)
t
2
t
1

g
2
(t
2
+t
1
) = v
0
g
2
(t
2
+t
1
)
Whereastheaverageof theinstantaneousvelocitiesat thebeginningandendof [t
1
, t
2
] is
s
(t
1
) +s
(t
2
)
2
=
1
2
_
(v
0
gt
1
) +(v
0
gt
2
)
_
=
1
2
(2v
0
)
g
2
(t
2
+t
1
) = v
0
g
2
(t
2
+t
1
).
Thetwoquantitiesarethesame.
An object falls under theinuenceof gravity near theearths surface. Which of thefollowingstatements is
true? Explain.
(a) Distancetraveledincreasesbyequal amountsinequal timeintervals.
(b) Velocityincreasesbyequal amountsinequal timeintervals.
(c) Thederivativeof velocityincreaseswithtime.
31. ByFaradaysLaw, if aconductingwireof length metersmovesat velocityv m/sperpendicular toamagneticeld
of strengthB (inteslas), avoltageof sizeV = Bv isinducedinthewire. Assumethat B = 2and = 0.5.
(a) CalculatedV/dv.
(b) Findtherateof changeof V withrespect totimet if v = 4t +9.
solution
(a) Assumingthat B = 2andl = 0.5, V = 2(0.5)v = v. Therefore,
dV
dv
= 1.
(b) If v = 4t +9, thenV = 2(0.5)(4t +9) = (4t +9). Therefore,
dV
dt
= 4.
ThevoltageV, current I, andresistanceR inacircuit arerelatedby OhmsLaw: V = IR, wheretheunitsare
volts, amperes, andohms. Assumethat voltageisconstant withV = 12volts. Calculate(specifyingunits):
(a) Theaveragerateof changeof I withrespect toR for theinterval fromR = 8toR = 8.1
(b) Therateof changeof I withrespect toR whenR = 8
(c) Therateof changeof R withrespect toI whenI = 1.5
33. Ethannds that withh hours of tutoring, heis abletoanswer correctly S(h) percent of theproblems ona
mathexam. Whichwouldyouexpect tobelarger: S
(3) or S
(30)? Explain.
solution Onepossiblegraphof S(h) is showninthegurebelowontheleft. This graphindicates that intheearly
hoursof workingwiththetutor, Ethanmakesrapidprogressinlearningthematerial but eventuallyapproacheseither the
limit of his ability to learnthematerial or themaximumpossiblescoreontheexam. Inthis scenario, S
(3) wouldbe
larger thanS
(30).
An alternativegraph of S(h) is shown belowon theright. Here, in theearly hours of working with thetutor little
progressismade(perhapsthetutor isassessinghowmuchEthanalreadyknows, hislearningstyle, hispersonality, etc.).
Thisisfollowedbyaperiodof rapidimprovement andnallyalevelingoff asEthanreacheshismaximumscore. Inthis
scenario, S
(3) andS
(30) might beroughlyequal.

Hours of tutoring
P
e
r
c
e
n
t
a
g
e

c
o
r
r
e
c
t
Hours of tutoring
P
e
r
c
e
n
t
a
g
e

c
o
r
r
e
c
t
Suppose(t ) measurestheanglebetweenaclocksminuteandhour hands. What is
(t ) at 3oclock?
35. Todeterminedrugdosages, doctorsestimateapersonsbodysurfacearea(BSA) (inmeterssquared) usingtheformula
BSA =

hm/60, whereh istheheight incentimetersandmthemassinkilograms. Calculatetherateof changeof BSA
withrespect to mass for apersonof constant height h = 180. What is this rateat m = 70andm = 80? Express your
result inthecorrect units. DoesBSA increasemorerapidlywithrespect tomassat lower or higher bodymass?
solution Assumingconstant height h = 180cm, let f (m) =

hm/60=
5
10
m betheformulafor bodysurfacearea
intermsof weight. Therateof changeof BSA withrespect tomassis
f
(m) =
5
10
_
1
2
m
1/2
_
=
5
20
m
.
If m = 70kg, thisis
f
(70) =
5
20
70
=
14
280
0.0133631
m
2
kg
.
If m = 80kg,
f
(80) =
5
20
80
=
1
20
16
=
1
80
m
2
kg
.
Becausetherateof changeof BSA dependson1/
m, it isclear that BSA increasesmorerapidlyat lower bodymass.

TheatmosphericCO
2
level A(t ) at MaunaLoa, Hawaii at timet (inpartsper millionbyvolume) isrecordedby
theScrippsInstitutionof Oceanography. Thevaluesfor themonthsJ anuaryDecember 2007were
382.45, 383.68, 384.23, 386.26, 386.39, 385.87,
384.39, 381.78, 380.73, 380.81, 382.33, 383.69
(a) Assumingthatthemeasurementsweremadeontherstof eachmonth, estimateA
(t ) onthe15thof themonths
J anuaryNovember.
(b) InwhichmonthsdidA
(t ) takeonitslargest andsmallest values?

(c) InwhichmonthwastheCO
2
level most nearlyconstant?
37. Thetangent linestothegraphof f (x) = x
2
growsteeper asx increases. At what ratedotheslopesof thetangent
linesincrease?
2
. Theslopes s of thetangent lines aregivenby s = f
(x) = 2x. Therateat whichthese

slopesareincreasingisds/dx = 2.
Figure13shows theheight y of amass oscillatingat theendof aspring. throughonecycleof theoscillation.
Sketchthegraphof velocityasafunctionof time.
In Exercises 3946, use Eq. (3) to estimate the unit change.
39. Estimate
1and
101
100. Compareyour estimateswiththeactual values.


x = x
1/2
. Thenf
(x) =
1
2
(x
1/2
). Weareusingthederivativetoestimatetheaveragerate
of change. That is,
x +h
x
h
f
(x),
sothat
x +h
x hf
(x).
Thus,
1 1f
(1) =
1
2
(1) =
1
2
. Theactual value, to six decimal places, is 0.414214. Also,
101
100
1f
(100) =
1
2
_
1
10
_
= 0.05. Theactual value, tosixdecimal places, is0.0498756.
Estimatef (4) f (3) if f
(x) = 2
x
. Thenestimatef (4), assumingthat f (3) = 12.
41. Let F(s) = 1.1s + 0.05s
2
bethestoppingdistanceas inExample3. CalculateF(65) andestimatetheincreasein
stoppingdistanceif speedisincreasedfrom65to66mph. Compareyour estimatewiththeactual increase.
solution Let F(s) = 1.1s +.05s
2
beasinExample3. F
(s) = 1.1+0.1s.
ThenF(65) = 282.75ft andF
(65) = 7.6ft/mph.
F
(65) F(66) F(65) is approximately equal tothechangeinstoppingdistanceper 1mphincreaseinspeed

whentravelingat 65mph. Increasingspeedfrom65to66thereforeincreasesstoppingdistanceby approximately
7.6ft.
The actual increase in stopping distance when speed increases from65 mph to 66 mph is F(66) F(65) =
290.4282.75= 7.65feet, whichdiffersbylessthanonepercent fromtheestimatefoundusingthederivative.
AccordingtoKleibersLaw, themetabolicrateP (inkilocaloriesper day) andbodymassm (inkilograms) of an
animal arerelatedbyathree-quarter-power law P = 73.3m
3/4
. Estimatetheincreaseinmetabolicratewhenbody
massincreasesfrom60to61kg.
43. Thedollarcostof producingx bagelsisC(x) = 300+0.25x 0.5(x/1000)
3
. Determinethecostof producing2000
bagelsandestimatethecost of the2001st bagel. Compareyour estimatewiththeactual cost of the2001st bagel.
solution Expandingthepower of 3yields
C(x) = 300+0.25x 510
10
x
3
.
Thisallowsustoget thederivativeC
(x) = 0.251.510
9
x
2
. Thecost of producing2000bagelsis
C(2000) = 300+0.25(2000) 0.5(2000/1000)
3
= 796
dollars. The cost of the 2001st bagel is, by denition, C(2001) C(2000). By the derivative estimate, C(2001)
C(2000) C
(2000)(1), sothecost of the2001st bagel isapproximately

C
(2000) = 0.251.510
9
(2000
2
) = $0.244.
C(2001) = 796.244, sotheexact cost of the2001st bagel isindistinguishablefromtheestimatedcost. Thefunctionis
verynearlylinear at thispoint.
Supposethedollar cost of producingx videocamerasisC(x) = 500x 0.003x
2
+10
8
x
3
.
(a) Estimatethemarginal costatproductionlevel x = 5000andcompareitwiththeactual costC(5001) C(5000).
(b) Comparethemarginal cost at x = 5000withtheaveragecost per camera, denedasC(x)/x.
45. Demandfor acommoditygenerallydecreasesasthepriceisraised. Supposethat thedemandfor oil (per capitaper
year) is D(p) = 900/p barrels, wherep is thedollar priceper barrel. Findthedemandwhenp = $40. Estimatethe
decreaseindemandif p risesto$41andtheincreaseif p declinesto$39.
solution D(p) = 900p
1
, soD
(p) = 900p
2
. Whenthepriceis$40abarrel, theper capitademandisD(40) =
22.5 barrels per year. With an increase in price from$40 to $41 a barrel, the change in demand D(41) D(40) is
approximatelyD
(40) = 900(40
2
) = 0.5625barrelsayear. Withadecreaseinpricefrom$40to$39abarrel, the
changeindemandD(39) D(40) isapproximatelyD
(40) = +0.5625. Anincreaseinoil pricesof adollar leadstoa

decreaseindemandof 0.5625barrelsayear, andadecreaseof adollar leadstoanincrease indemandof 0.5625barrels
ayear.
Thereproduction ratef of thefruit y Drosophila melanogaster, grown in bottles in a laboratory, decreases
withthenumber p of ies inthebottle. A researcher has foundthenumber of offspringper femaleper day to be
approximatelyf (p) = (340.612p)p
0.658
.
(a) Calculatef (15) andf
(15).
(b) Estimatethedecreaseindailyoffspringper femalewhenp isincreasedfrom15to16. Isthisestimatelarger or
smaller thantheactual valuef (16) f (15)?
(c) Plot f (p) for 5 p 25andverify that f (p) is adecreasingfunctionof p. Do youexpect f
(p) to be
positiveor negative? Plot f
(p) andconrmyour expectation.

47. AccordingtoStevens Lawinpsychology, theperceivedmagnitudeof astimulusisproportional (approxi-
mately) toapower of theactual intensityI of thestimulus. Experimentsshowthat theperceived brightness B of alight
satisesB = kI
2/3
, whereI isthelightintensity, whereastheperceived heaviness H of aweightW satisesH = kW
3/2
(k isaconstant that isdifferent inthetwocases). ComputedB/dI anddH/dW andstatewhether theyareincreasingor
decreasingfunctions. Thenexplainthefollowingstatements:
(a) A one-unit increaseinlight intensityisfelt morestronglywhenI issmall thanwhenI islarge.
(b) Addinganother poundtoaloadW isfelt morestronglywhenW islargethanwhenW issmall.
solution
(a) dB/dI =
2k
3
I
1/3
=
2k
3I
1/3
.
As I increases, dB/dI shrinks, so that the rate of change of perceived intensity decreases as the actual intensity
increases. Increasedlight intensityhasadiminished return inperceivedintensity. A sketchof B against I isshown: See
that theheight of thegraphincreasesmoreslowlyasyoumovetotheright.
(b) dH/dW =
3k
2
W
1/2
. As W increases, dH/dW increases as well, so that therateof changeof perceived weight
increasesasweight increases. A sketchof H against W isshown: Seethat thegraphbecomessteeper asyoumovetothe
right.
LetM(t ) bethemass(inkilograms) of aplantasafunctionof time(inyears). RecentstudiesbyNiklasandEnquist
havesuggestedthataremarkablywiderangeof plants(fromalgaeandgrasstopalmtrees)obeyathree-quarter-power
growth lawthat is, dM/dt = CM
3/4
for someconstant C.
(a) If atreehasagrowthrateof 6kg/yr whenM = 100kg, what isitsgrowthratewhenM = 125kg?
(b) If M = 0.5kg, howmuchmoremassmust theplant acquiretodoubleitsgrowthrate?
Exercises 4951: The Lorenz curve y = F(r) is used by economists to study income distribution in a given country (see
Figure 14). By denition, F(r) is the fraction of the total income that goes to the bottomrth part of the population, where
0 r 1. For example, if F(0.4) = 0.245, then the bottom 40%of households receive 24.5%of the total income. Note
that F(0) = 0and F(1) = 1.
0.2 0.4 0.6 0.8 1.0
0.2
0.4
0.6
F(r)
0.2 0.4 0.6 0.8 1.0
0.2
0.4
0.6
0.8
1.0
F(r)
P
(A) Lorenz curvefor Sweden in 2004 (B) Two Lorenz curves: Thetangent
lines at P and Q haveslope1.
Q
0.2 0.4 0.6 0.8 1.0
0.2
0.4
0.6
0.8
1.0
r
F(r)
0.2 0.4 0.6 0.8 1.0
0.2
0.4
0.6
0.8
1.0
r
F(r)
L
2
L
1
P
(A) Lorenz curvefor Sweden in 2004 (B) Two Lorenz curves: Thetangent
lines at P and Q haveslope1.
Q
FIGURE 14
49. Our goal is to ndaninterpretationfor F
(r). Theaverageincomefor agroupof households is thetotal

incomegoingtothegroupdividedbythenumber of householdsinthegroup. Thenational averageincomeisA = T/N,
whereN isthetotal number of householdsandT isthetotal incomeearnedbytheentirepopulation.
(a) Showthat theaverageincomeamonghouseholdsinthebottomrthpart isequal to(F(r)/r)A.
(b) Showmoregenerallythat theaverageincomeof householdsbelongingtoaninterval [r, r +r] isequal to
_
F(r +r) F(r)
r
_
A
(c) Let 0 r 1. A householdbelongstothe100rthpercentileif itsincomeisgreater thanor equal totheincomeof
100r %of all households. Passtothelimit asr 0in(b) toderivethefollowinginterpretation: A householdinthe
100rthpercentilehasincomeF
(r)A. Inparticular, ahouseholdinthe100rthpercentilereceivesmorethanthenational

averageif F
(r) > 1andlessif F
(r) < 1.
(d) For theLorenz curvesL
1
andL
2
inFigure14(B), what percentageof householdshaveabove-averageincome?
solution
(a) Thetotal incomeamonghouseholdsinthebottomrthpart isF(r)T andtherearerN householdsinthispart of the
population. Thus, theaverageincomeamonghouseholdsinthebottomrthpart isequal to
F(r)T
rN
=
F(r)
r
T
N
=
F(r)
r
A.
(b) Consider theinterval [r, r + r]. Thetotal incomeamonghouseholdsbetweenthebottomrthpart andthebottom
r +r-thpart isF(r +r)T F(r)T . Moreover, thenumber of householdscoveredbythisinterval is(r +r)N
rN = rN. Thus, theaverageincomeof householdsbelongingtoaninterval [r, r +r] isequal to
F(r +r)T F(r)T
rN
=
F(r +r) F(r)
r
T
N
=
F(r +r) F(r)
r
A.
(c) Taketheresult frompart (b) andlet r 0. Because
lim
r0
F(r +r) F(r)
r
= F
(r),
wendthat ahouseholdinthe100rthpercentilehasincomeF
(r)A.
(d) ThepointP inFigure14(B) hasanr-coordinateof 0.6, whilethepointQhasanr-coordinateof roughly0.75. Thus,
oncurveL
1
, 40%of householdshaveF
(r) > 1andthereforehaveabove-averageincome. OncurveL

2
, roughly25%
of householdshaveabove-averageincome.
Thefollowingtableprovidesvaluesof F(r) for Swedenin2004. Assumethat thenational averageincomewas
A = 30,000euros.
r 0 0.2 0.4 0.6 0.8 1
F(r) 0 0.01 0.245 0.423 0.642 1
(a) What wastheaverageincomeinthelowest 40%of households?
(b) Showthat theaverageincomeof thehouseholdsbelongingtotheinterval [0.4, 0.6] was26,700euros.
(c) Estimate F
(0.5). Estimate the income of households in the 50th percentile? Was it greater or less than the
ti l ?
51. UseExercise49(c) toprove:
(a) F
(r) isanincreasingfunctionof r.
(b) Incomeisdistributedequally(all householdshavethesameincome) if andonlyif F(r) = r for 0 r 1.
solution
(a) Recall fromExercise49(c) that F
(r)A istheincomeof ahouseholdinthe100r-thpercentile. Suppose0 r

1
<
r
2
1. Becauser
2
> r
1
, ahouseholdinthe100r
2
-thpercentilemust haveincomeat least aslargeasahouseholdinthe
100r
1
-thpercentile. Thus, F
(r
2
)A F
(r
1
)A, or F
(r
2
) F
(r
1
). ThisimpliesF
(r) isanincreasingfunctionof r.
(b) If F(r) = r for 0 r 1, then F
(r) = 1andhouseholds in all percentiles haveincomeequal to thenational

average; that is, incomeisdistributedequally. Alternately, if incomeisdistributedequally(all householdshavethesame
income), thenF
(r) = 1for 0 r 1. Thus, F must bealinear functioninr withslope1. Moreover, thecondition

F(0) = 0requirestheF intercept of thelinetobe0. Hence, F(r) = 1 r +0= r.
Studiesof Internetusageshowthatwebsitepopularityisdescribedquitewell byZipfsLaw, accordingtowhich
thenthmost popular websitereceivesroughlythefraction1/n of all visits. Supposethat onaparticular day, thenth
most popular sitehadapproximatelyV(n) = 10
6
/n visitors(for n 15,000).
(a) Verify that thetop50websites receivednearly 45%of thevisits. Hint: Let T (N) denotethesumof V(n) for
1 n N. Useacomputer algebrasystemtocomputeT (45) andT (15,000).
(b) Verify, by numerical experimentation, that whenEq. (3) is usedtoestimateV(n + 1) V(n), theerror inthe
estimatedecreases as n grows larger. Find (again, by experimentation) an N such that theerror is at most 10for
n N.
(c) UsingEq. (3), showthat for n 100, thenth websitereceivedat most 100morevisitors than the(n + 1)st
website.
In Exercises 53 and 54, the average cost per unit at production level x is dened as C
avg
(x) = C(x)/x, where C(x) is
the cost function. Average cost is a measure of the efciency of the production process.
53. Showthat C
avg
(x) isequal totheslopeof thelinethroughtheoriginandthepoint (x, C(x)) onthegraphof C(x).
Usingthisinterpretation, determinewhether averagecost or marginal cost isgreater at pointsA, B, C, D inFigure15.
C
x
A
B C
D
FIGURE 15 Graphof C(x).
solution Bydenition, theslopeof thelinethroughtheoriginand(x, C(x)), that is, between(0, 0) and(x, C(x)) is
C(x) 0
x 0
=
C(x)
x
= C
av
.
At point A, averagecost isgreater thanmarginal cost, asthelinefromtheorigintoA issteeper thanthecurveat this
point(weseethisbecausetheline, tracingfromtheorigin, crossesthecurvefrombelow). Atpoint B, theaveragecostis
still greater thanthemarginal cost. At thepoint C, theaveragecost andthemarginal cost arenearlythesame, sincethe
tangent lineandthelinefromtheoriginarenearly thesame. ThelinefromtheorigintoD crossesthecost curvefrom
above, andsoislesssteepthanthetangent linetothecurveat D; theaveragecost at thispoint islessthanthemarginal
cost.
Thecost indollarsof producingalarmclocksisC(x) = 50x
3
750x
2
+ 3740x + 3750wherex isinunitsof
1000.
(a) Calculatetheaveragecost at x = 4, 6, 8, and10.
(b) Usethegraphical interpretationof averagecost tondtheproductionlevel x
0
at whichaveragecost islowest.
What istherelationbetweenaveragecost andmarginal cost at x
0
(seeFigure16)?
3.5 Higher Derivatives
1. OnSeptember 4, 2003, theWall Street Journal printedtheheadlineStocks Go Higher, ThoughthePaceof Their
GainsSlows.Rephrasethisheadlineasastatementabouttherstandsecondtimederivativesof stockpricesandsketch
apossiblegraph.
solution Becausestocks aregoing higher, stock prices areincreasing and therst derivativeof stock prices must
thereforebepositive. Ontheother hand, becausethepaceof gainsisslowing, thesecondderivativeof stockpricesmust
benegative.
Stock
price
Time
2. Trueor false? Thethirdderivativeof positionwithrespect to timeis zero for anobject fallingto earthunder the
inuenceof gravity. Explain.
solution Thisstatementistrue. Theaccelerationof anobjectfallingtoearthundertheinuenceof gravityisconstant;
hence, thesecondderivativeof positionwithrespecttotimeisconstant. Becausethethirdderivativeisjustthederivative
of thesecondderivativeandthederivativeof aconstant iszero, it followsthat thethirdderivativeiszero.
3. Whichtypeof polynomial satisesf
(x) = 0for all x?

solution Thethirdderivativeof all quadraticpolynomials(polynomialsof theformax
2
+bx +c for someconstants
a, b andc) isequal to0for all x.
S E C T I ON 3.5 Higher Derivatives 127
4. What isthesixthderivativeof f (x) = x
6
?
solution Thesixthderivativeof f (x) = x
6
is6! = 720.
Exercises
In Exercises 116, calculate y
and y
.
1. y = 14x
2
solution Let y = 14x
2
. Theny
= 28x, y
= 28, andy
= 0.
y = 72x
3. y = x
4
25x
2
+2x
solution Let y = x
4
25x
2
+2x. Theny
= 4x
3
50x +2, y
= 12x
2
50, andy
= 24x.
y = 4t
3
9t
2
+7
5. y =
4
3
r
3
solution Let y =
4
3
r
3
. Theny
= 4r
2
, y
= 8r, andy
= 8.
y =

x
7. y = 20t
4/5
6t
2/3
solution Lety = 20t
4/5
6t
2/3
. Theny
= 16t
1/5
4t
1/3
, y
=
16
5
t
6/5
+
4
3
t
4/3
, andy
=
96
25
t
11/15
16
9
t
7/3
.
y = x
9/5 9. y = z
4
z
solution Let y = z 4z
1
. Theny
= 1+4z
2
, y
= 8z
3
, andy
= 24z
4
.
y = 5t
3
+7t
8/3
11. y =
2
(2 +7)
solution Let y =
2
(2 +7) = 2
3
+7
2
. Theny
= 6
2
+14, y
= 12 +14, andy
= 12.
y = (x
2
+x)(x
3
+1)
13. y =
x 4
x
solution Let y =
x4
x
= 14x
1
. Theny
= 4x
2
, y
= 8x
3
, andy
= 24x
4
.
y =
1
1x
15. y = s
1/2
(s +1)
solution Expandinggivesy = s
1/2
+s
1/2
. Then
y
=
1
2
s
1/2
1
2
s
3/2
y
=
1
4
s
3/2
+
3
4
s
5/2
y
=
3
8
s
5/2
15
8
s
7/2
y = (r
1/2
+r)(1r)
In Exercises 1725, calculate the derivative indicated.
17. f
(4)
(1), f (x) = x
4
4
. Thenf
(x) = 4x
3
, f
(x) = 12x
2
, f
(x) = 24x, andf

(4)
(x) = 24. Thusf
(4)
(1) = 24.
g
(1), g(t ) = 4t
5
19.
d
2
y
dt
2
t =1
, y = 4t
3
+3t
2
solution Let y = 4t
3
+3t
2
. Then
dy
dt
= 12t
4
+6t and
d
2
y
dt
2
= 48t
5
+6. Hence
d
2
y
dt
2
t =1
= 48(1)
5
+6= 54.
d
4
f
dt
4
t =1
, f (t ) = 6t
9
2t
5
21.
d
4
x
dt
4
t =16
, x = t
3/4
solution Let x(t ) = t
3/4
. Then
dx
dt
=
3
4
t
7/4
,
d
2
x
dt
2
=
21
16
t
11/4
,
d
3
x
dt
3
=
231
64
t
15/4
, and
d
4
x
dt
4
=
3465
256
t
19/4
.
Thus
d
4
x
dt
4
t =16
=
3465
256
16
19/4
=
3465
134,217,728
.
f
(4), f (t ) = 2t
2
t
23. f
(3), f (x) =
12
x
x
3
solution Differentiatinggives
f
(x) = 12x
2
3x
2
f
(x) = 24x
3
6x
f
(x) = 72x
4
6
Thus, f
(3) = 72 3
4
6=
62
9
.
f
(1), f (t ) =
t
t +1
25. h
(1), h(w) =
1
w +1
solution Differentiatinggives
h
(w) =

1
2
w
1/2
(
w +1)
2
=
1
2
w(
w +1)
2
h
(w) =
2
w
_
2(
w +1)
1
2
w
1/2
_
+2
1
2
w
1/2
(
w +1)
2
4w(
w +1)
4
=
2(
w +1) +w
1/2
(
w +1)
2
4w(
w +1)
4
sothat
h
(1) =
2(1+1) +1(1+1)
2
4 1(1+1)
4
=
8
64
=
1
8
g
(1), g(s) =
s
s +1
27. Calculatey
(k)
(0) for 0 k 5, wherey = x
4
+ax
3
+bx
2
+cx +d (witha, b, c, d theconstants).
solution Applyingthepower, constant multiple, andsumrulesat eachstage, weget (notey
(0)
isy byconvention):
k y
(k)
0 x
4
+ax
3
+bx
2
+cx +d
1 4x
3
+3ax
2
+2bx +c
2 12x
2
+6ax +2b
3 24x +6a
4 24
5 0
fromwhichweget y
(0)
(0) = d, y
(1)
(0) = c, y
(2)
(0) = 2b, y
(3)
(0) = 6a, y
(4)
(0) = 24, andy
(5)
(0) = 0.
Whichof thefollowingsatisfyf
(k)
(x) = 0for all k 6?
(a) f (x) = 7x
4
+4+x
1
(b) f (x) = x
3
2
(c) f (x) =

x (d) f (x) = 1x
6
(e) f (x) = x
9/5
(f) f (x) = 2x
2
+3x
5
29. Usetheresult inExample3tond
d
6
dx
6
x
1
.
solution TheequationinExample3indicatesthat
d
6
dx
6
x
1
= (1)
6
6!x
61
.
(1)
6
= 1and6! = 654321= 720, so
d
6
dx
6
x
1
= 720x
7
.
Calculatetherst vederivativesof f (x) =

x.
(a) Showthat f
(n)
(x) isamultipleof x
n+1/2
.
(b) Showthat f
(n)
(x) alternatesinsignas(1)
n1
for n 1.
(c) Findaformulafor f
(n)
(x) for n 2. Hint: Verifythat thecoefcient is1 3 5
2n 3
2
n
.
In Exercises 3136, nd a general formula for f
(n)
(x).
31. f (x) = x
2
solution f
(x) = 2x
3
, f
(x) = 6x
4
, f
(x) = 24x
5
, f
(4)
(x) = 5 24x
6
, . . . . Fromthiswecanconclude
that thenthderivativecanbewrittenasf
(n)
(x) = (1)
n
(n +1)!x
(n+2)
.
f (x) = (x +2)
1
33. f (x) = x
1/2
solution f
(x) =
1
2
x
3/2
. Wewill avoidsimplifyingnumeratorsanddenominatorstondthepattern:
f
(x) =
3
2
1
2
x
5/2
= (1)
2
31
2
2
x
5/2
f
(x) =
5
2
31
2
2
x
7/2
= (1)
3
531
2
3
x
7/2
.
.
.
f
(n)
(x) = (1)
n
(2n 1) (2n 3) . . . 1
2
n
x
(2n+1)/2
.
f (x) = x
3/2 35. f (x) =
x +1
x
2
x +1
x
2
= x
1
+x
2
. Takingsuccessivederivativesgives
f
(x) = x
2
2x
3
f
(x) = 2x
3
+6x
4
f
(x) = 6x
4
24x
5
f
(4)
(x) = 24x
5
+120x
6
.
.
.
f
(n)
(x) = (1)
n
(n!x
n1
+(n +1)!x
n2
)
f (x) =
x 1
x
37. (a) Findtheaccelerationat timet = 5minof ahelicopter whoseheight iss(t ) = 300t 4t
3
m.
(b) Plot theaccelerationh
(t ) for 0 t 6. Howdoesthisgraphshowthat thehelicopter isslowingdownduringthis

timeinterval?
solution
(a) Lets(t ) = 300t 4t
3
, witht inminutesands inmeters. Thevelocityisv(t ) = s
(t ) = 30012t
2
andacceleration
isa(t ) = s
(t ) = 24t . Thusa(5) = 120m/min

2
.
(b) Theaccelerationof thehelicopter for 0 t 6isshowninthegurebelow. Astheaccelerationof thehelicopter is
negative, thevelocityof thehelicopter must bedecreasing. Becausethevelocityispositivefor 0 t 6, thehelicopter
isslowingdown.
140
120
100
80
60
40
20
1 2 3 4 5 6
y
x
Findanequationof thetangent tothegraphof y = f
(x) at x = 3, wheref (x) = x

4
.
39. Figure5showsf , f
, andf
. Determinewhichiswhich.
(A) (B)
x
y
3 2 1
x
y
3 2 1
x
y
(C)
3 2 1
FIGURE 5
solution (a) f
(b) f
(c) f .
Thetangent lineto(c) ishorizontal at x = 1andx = 3, where(b) hasroots. Thetangent lineto(b) ishorizontal at
x = 2andx = 0, where(a) hasroots.
Thesecondderivativef
isshowninFigure6. Whichof (A) or (B) isthegraphof f andwhichisf
?
41. Figure7showsthegraphof thepositions of anobject asafunctionof timet . Determinetheintervalsonwhichthe
accelerationispositive.
Time
40 30 20 10
Position
FIGURE 7
solution Roughly fromtime10totime20andfromtime30totime40. Theaccelerationispositiveover thesame
intervalsover whichthegraphisbendingupward.
Findapolynomial f (x) that satisestheequationxf
(x) +f (x) = x
2
.
43. Findall valuesof n suchthat y = x
n
satises
x
2
y
2xy
= 4y
solution Wehavey
= nx
n1
, y
= n(n 1)x
n2
, sothat
x
2
y
2xy
= x
2
(n(n 1)x
n2
) 2xnx
n1
= (n
2
3n)x
n
= (n
2
3n)y
Thustheequationissatisedif andonlyif n
2
3n = 4, sothat n
2
3n 4= 0. Thishappensfor n = 1, 4.
Whichof thefollowingdescriptionscouldnot applytoFigure8? Explain.
(a) Graphof accelerationwhenvelocityisconstant
(b) Graphof velocitywhenaccelerationisconstant
(c) Graphof positionwhenaccelerationiszero
45. Accordingtoonemodel that takesintoaccount air resistance, theaccelerationa(t ) (inm/s
2
) of askydiver of mass
m infreefall satises
a(t ) = 9.8+
k
m
v(t )
2
wherev(t ) isvelocity(negativesincetheobjectisfalling) andk isaconstant. Supposethatm = 75kgandk = 14kg/m.
(a) What istheobjectsvelocitywhena(t ) = 4.9?
(b) What istheobjectsvelocitywhena(t ) = 0?Thisvelocityistheobjectsterminal velocity.
solution Solvinga(t ) = 9.8+
k
m
v(t )
2
for thevelocity andtakingintoaccount that thevelocity isnegativesince
theobject isfalling, wend
v(t ) =
_
m
k
(a(t ) +9.8) =
_
75
14
(a(t ) +9.8).
(a) Substitutinga(t ) = 4.9intotheaboveformulafor thevelocity, wend
v(t ) =
_
75
14
(4.9) =
26.25= 5.12m/s.
(b) Whena(t ) = 0,
v(t ) =
_
75
14
(9.8) =
52.5= 7.25m/s.
Accordingtoonemodel that attemptstoaccount for air resistance, thedistances(t ) (inmeters) traveledby a
fallingraindropsatises
d
2
s
dt
2
= g
0.0005
D
_
ds
dt
_
2
whereD istheraindropdiameter andg = 9.8m/s
2
. Terminal velocity v
term
isdenedasthevelocity at whichthe
drophaszeroacceleration(onecanshowthat velocityapproachesv
term
astimeproceeds).
(a) Showthat v
term
=

2000gD.
(b) Findv
term
for dropsof diameter 10
3
mand10
4
m.
(c) Inthismodel, doraindropsacceleratemorerapidlyat higher or lower velocities?
47. Aservomotorcontrolsthevertical movementof adrill bitthatwill drill apatternof holesinsheetmetal.Themaximum
vertical speedof thedrill bit is4in./s, andwhiledrillingthehole, it must movenomorethan2.6in./stoavoidwarping
themetal. Duringacycle, thebit beginsandendsat rest, quickly approachesthesheet metal, andquickly returnstoits
initial positionafter theholeisdrilled. Sketchpossiblegraphsof thedrill bitsvertical velocity andacceleration. Label
thepoint wherethebit entersthesheet metal.
solution Therewill bemultiplecycles, eachof whichwill bemoreorlessidentical. Letv(t ) bethedownward vertical
velocity of thedrill bit, andlet a(t ) bethevertical acceleration. Fromthenarrative, weseethat v(t ) canbeno greater
than4andnogreater than2.6whiledrillingistakingplace. Duringeachcycle, v(t ) = 0initially, v(t ) goesto4quickly.
Whenthebit hitsthesheet metal, v(t ) goesdownto2.6quickly, at whichit staysuntil thesheet metal isdrilledthrough.
Asthedrill pullsout, itreachesmaximumnon-drillingupwardspeed(v(t ) = 4) quickly, andmaintainsthisspeeduntil
it returnstorest. A possibleplot follows:
2
4
4
2
2 1.5 1 0.5
x
y
Metal
A graphof theaccelerationisextractedfromthisgraph:
20
40
40
20
2 1.5
1 0.5
x
y
Metal
In Exercises 48 and 49, refer to the following. In a 1997 study, Boardman and Lave related the trafc speed S on a
two-lane road to trafc density Q (number of cars per mile of road) by the formula
S = 2882Q
1
0.052Q+31.73
for 60 Q 400(Figure 9).
Density Q
400 300 200 100
Speed S
(mph)
10
20
30
40
50
60
70
FIGURE 9 Speedasafunctionof trafcdensity.
CalculatedS/dQandd
2
S/dQ
2
.
49. (a) Explainintuitivelywhyweshouldexpect that dS/dQ < 0.
(b) Showthat d
2
S/dQ
2
> 0. Thenusethefact that dS/dQ < 0andd
2
S/dQ
2
> 0tojustifythefollowingstatement:
A one-unit increase in trafc density slows down trafc more when Q is small than when Q is large.
(c) Plot dS/dQ. Whichpropertyof thisgraphshowsthat d
2
S/dQ
2
> 0?
solution
(a) Trafcspeedmust bereducedwhentheroadgetsmorecrowdedsoweexpect dS/dQtobenegative. Thisisindeed
thecasesincedS/dQ = 0.0522882/Q
2
< 0.
(b) Thedecreasein speed dueto aone-unit increasein density is approximately dS/dQ (anegativenumber). Since
d
2
S/dQ
2
= 5764Q
3
> 0ispositive, thistellsusthatdS/dQgetslargerasQincreasesandanegativenumberwhich
gets larger is gettingcloser tozero. Sothedecreaseinspeedis smaller whenQ is larger, that is, aone-unit increasein
trafcdensityhasasmaller effect whenQislarge.
(c) dS/dQisplottedbelow. Thefact that thisgraphisincreasingshowsthat d
2
S/dQ
2
> 0.
x
y
0.2
0.4
0.6
0.8
1
1.2
400 300 100 200
Useacomputer algebrasystemtocomputef
(k)
(x) for k = 1, 2, 3for thefollowingfunctions.
(a) f (x) = (1+x
3
)
5/3
(b) f (x) =
1x
4
15x 6x
2
51. Let f (x) =
x +2
x 1
. Useacomputer algebrasystemtocomputethef
(k)
(x) for 1 k 4. Canyounda
general formulafor f
(k)
(x)?
x +2
x 1
. Usingacomputer algebrasystem,
f
(x) =
3
(x 1)
2
= (1)
1
3 1
(x 1)
1+1
;
f
(x) =
6
(x 1)
3
= (1)
2
3 2 1
(x 1)
2+1
;
f
(x) =
18
(x 1)
4
= (1)
3
3 3!
(x 1)
3+1
; and
f
(4)
(x) =
72
(x 1)
5
= (1)
4
3 4!
(x 1)
4+1
.
Fromthepatternobservedabove, weconjecture
f
(k)
(x) = (1)
k
3 k!
(x 1)
k+1
.
Findthe100thderivativeof
p(x) = (x +x
5
+x
7
)
10
(1+x
2
)
11
(x
3
+x
5
+x
7
)
53. What isp
(99)
(x) for p(x) asinExercise52?
solution First notethat for anyinteger n 98,
d
99
dx
99
x
n
= 0.
Now, if weexpandp(x), wend
p(x) = x
99
+ termsof degreeat most 98;
therefore,
d
99
dx
99
p(x) =
d
99
dx
99
(x
99
+ termsof degreeat most 98) =
d
99
dx
99
x
99
Usinglogicsimilar tothat usedtocomputethederivativeinExample(3), wecompute:
d
99
dx
99
(x
99
) = 9998. . . 1,
sothat
d
99
dx
99
p(x) = 99!.
UsetheProductRuletwicetondaformulafor (fg)
intermsof f andg andtheir rstandsecondderivatives.

55. UsetheProduct Ruletondaformulafor (fg)
andcompareyour result withtheexpansionof (a +b)

3
. Thentry
toguessthegeneral formulafor (fg)
(n)
.
solution ContinuingfromExercise54, wehave
h
= f
+gf
+2(f
+g
) +fg
+g
= f
g +3f
+3f
+fg
Thebinomial theoremgives
(a +b)
3
= a
3
+3a
2
b +3ab
2
+b
3
= a
3
b
0
+3a
2
b
1
+3a
1
b
2
+a
0
b
3
andmoregenerally
(a +b)
n
=
n
k=0
_
n
k
_
a
nk
b
k
,
wherethebinomial coefcientsaregivenby
_
n
k
_
=
k(k 1) (k n +1)
n!
.
Accordingly, thegeneral formulafor (fg)
(n)
isgivenby
(fg)
(n)
=
n
k=0
_
n
k
_
f
(nk)
g
(k)
,
wherep
(k)
isthekthderivativeof p (or p itself whenk = 0).
Compute
f (x) = lim
h0
f (x +h) +f (x h) 2f (x)
h
2
for thefollowingfunctions:
(a) f (x) = x (b) f (x) = x
2
(c) f (x) = x
3
Basedontheseexamples, what doyouthinkthelimit f represents?
3.6 Trigonometric Functions
1. Determinethesign(+ or ) that yieldsthecorrect formulafor thefollowing:
(a)
d
dx
(sinx +cosx) = sinx cosx
(b)
d
dx
secx = secx tanx
(c)
d
dx
cotx = csc
2
x
solution Thecorrect formulasare
(a)
d
dx
(sinx +cosx) = sinx +cosx
(b)
d
dx
secx = secx tanx
(c)
d
dx
cotx = csc
2
x
2. Whichof thefollowingfunctionscanbedifferentiatedusingtheruleswehavecoveredsofar?
(a) y = 3cosx cotx (b) y = cos(x
2
) (c) y = 2
x
sinx
solution
(a) 3cosx cotx isaproductof functionswhosederivativesareknown. Thisfunctioncanthereforebedifferentiatedusing
theProduct Rule.
(b) cos(x
2
) is acompositionof thefunctions cosx andx
2
. Wehavenot yet discussedhowto differentiatecomposite
functions.
(c) 2
x
sinx is a product of functions, but we do not yet know how to differentiate 2
x
, so we do not know how to
differentiatetheproduct.
3. Compute
d
dx
(sin
2
x +cos
2
x) without usingthederivativeformulasfor sinx andcosx.
solution Recall that sin
2
x +cos
2
x = 1for all x. Thus,
d
dx
(sin
2
x +cos
2
x) =
d
dx
1= 0.
4. Howistheadditionformulausedinderivingtheformula(sinx)
= cosx?
solution Thedifferencequotient for thefunctionsinx involvestheexpressionsin(x +h). Theadditionformulafor
thesinefunctionisusedtoexpandthisexpressionassin(x +h) = sinx cosh +sinhcosx.
Exercises
In Exercises 14, nd an equation of the tangent line at the point indicated.
1. y = sinx, x =

4
solution Let f (x) = sinx. Thenf
(x) = cosx andtheequationof thetangent lineis

y = f
4
_ _
x

4
_
+f
_
4
_
=
2
2
_
x

4
_
+
2
2
=
2
2
x +
2
2
_
1

4
_
.
y = cosx, x =

3
3. y = tanx, x =

4
solution Let f (x) = tanx. Thenf
(x) = sec
2
x andtheequationof thetangent lineis
y = f
4
_ _
x

4
_
+f
_
4
_
= 2
_
x

4
_
+1= 2x +1

2
.
y = secx, x =

6
In Exercises 524, compute the derivative.
5. f (x) = sinx cosx
solution Let f (x) = sinx cosx. Then
f
(x) = sinx(sinx) +cosx(cosx) = sin

2
x +cos
2
x.
f (x) = x
2
cosx
7. f (x) = sin
2
x
solution Let f (x) = sin
2
x = sinx sinx. Then
f
(x) = sinx(cosx) +sinx(cosx) = 2sinx cosx.

f (x) = 9secx +12cotx
9. H(t ) = sint sec
2
t
solution Let H(t ) = sint sec
2
t . Then
H
(t ) = sint
d
dt
(sect sect ) +sec
2
t (cost )
= sint (sect sect tant +sect sect tant ) +sect
= 2sint sec
2
t tant +sect.
h(t ) = 9csct +t cott
11. f () = tan sec
solution Let f () = tan sec. Then
f
() = tan sec tan +sec sec

2
= sec tan
2
+sec
3
=
_
tan
2
+sec
2
_
sec.
k() =
2
sin
2
13. f (x) = (2x

4
4x
1
) secx
solution Let f (x) = (2x
4
4x
1
) secx. Then
f
(x) = (2x
4
4x
1
) secx tanx +secx(8x
3
+4x
2
).
f (z) = z tanz 15. y =
sec
solution Let y =
sec
. Then
y
=
sec tan sec
2
.
G(z) =
1
tanz cotz
17. R(y) =
3cosy 4
siny
solution Let R(y) =
3cosy 4
siny
. Then
R
(y) =
siny(3siny) (3cosy 4)(cosy)
sin
2
y
=
4cosy 3(sin
2
y +cos
2
y)
sin
2
y
=
4cosy 3
sin
2
y
.
f (x) =
x
sinx +2
19. f (x) =
1+tanx
1tanx
1+tanx
1tanx
. Then
f
(x) =
(1tanx) sec
2
x (1+tanx)
_
sec
2
x
_
(1tanx)
2
=
2sec
2
x
(1tanx)
2
.
f () = tan sec 21. f (x) =
sinx +1
sinx 1
solution Wehave
f
(x) =
(sinx 1)(cosx) (sinx +1)(cosx)
(sinx 1)
2
=
2cosx
(sinx 1)
2
f (x) =
csc
2
t
t
23. R() =
cos
4+cos
solution Differentiating, wehave
R
() =
(4+cos)(sin) (cos)(sin)
(4+cos)
2
=
4sin
(4+cos)
2
g(z) =
cotz
33sinz
In Exercises 2534, nd an equation of the tangent line at the point specied.
25. y = x
3
+cosx, x = 0
3
+cosx. Thenf
(x) = 3x
2
sinx andf
(0) = 0. Thetangent lineat x = 0is

y = f
(0)(x 0) +f (0) = 0(x) +1= 1.

y = tan, =

6
27. y = sinx +3cosx, x = 0
solution Let f (x) = sinx +3cosx. Thenf
(x) = cosx 3sinx andf
(0) = 1. Thetangent lineat x = 0is

y = f
(0)(x 0) +f (0) = x +3.

y =
sint
1+cost
, t =

3
29. y = 2(sin +cos), =

3
solution Let f () = 2(sin + cos). Thenf
() = 2(cos sin) andf
3
) = 1

3. Thetangent lineat
x =

3
is
y = f
3
_ _
x

3
_
+f
_
3
_
= (1
3)
_
x

3
_
+1+
3.
y = cscx cotx, x =

4
31. y = (cott )(cost ), t =

3
solution Let f (t ) = (cott )(cost ). Then
f
(t ) = (csc
2
t )(cost ) +(cott )(sint ) = cott csct cost
For t =

3
, wehave
f (t ) = cot
3
cos
3
=
3
6
, f
(t ) = cot
3
csc
3
cos
3
=
2
3
1
2
=
7
6
sothat theequationof thetangent lineis
y = f
3
_ _
x

3
_
+f
_
3
_
=
7
6
x +
7
18
+
3
6
y = x cos
2
x, x =

4
33. y = x
2
(1sinx), x =
3
2
2
(1sinx). Then
f
(x) = 2x(1sinx) x
2
cosx
For a =
3
2
, wehave
f (a) =
9
2
4
(1(1)) =
9
2
2
, f
(a) = 3(1(1))
9
2
4
(0) = 6
sothat theequationof thetangent lineis
y = f
(a)(x a) +f (a) = 6
_
x
3
2
_
+
9
2
2
= 6x
9
2
2
y =
sin cos
, =

4
In Exercises 3537, use Theorem 1 to verify the formula.
35.
d
dx
cotx = csc
2
x
solution cotx =
cosx
sinx
. Usingthequotient ruleandthederivativeformulas, wecompute:
d
dx
cotx =
d
dx
cosx
sinx
=
sinx(sinx) cosx(cosx)
sin
2
x
=
(sin
2
x +cos
2
x)
sin
2
x
=
1
sin
2
x
= csc
2
x.
d
dx
secx = secx tanx
37.
d
dx
cscx = cscx cotx
solution Sincecscx =
1
sinx
, wecanapplythequotient ruleandthetwoknownderivativestoget:
d
dx
cscx =
d
dx
1
sinx
=
sinx(0) 1(cosx)
sin
2
x
=
cosx
sin
2
x
=
cosx
sinx
1
sinx
= cotx cscx.
Showthat bothy = sinx andy = cosx satisfyy
= y.
In Exercises 3942, calculate the higher derivative.
39. f
(), f () = sin
solution Let f () = sin. Then
f
() = cos +sin
f
() = (sin) +cos +cos = sin +2cos.

d
2
dt
2
cos
2
t
41. y
, y
, y = tanx
solution Let y = tanx. Theny
= sec
2
x andbytheChainRule,
y
= =
d
dx
sec
2
x = 2(secx)(secx tanx) = 2sec
2
x tanx
y
= 2sec
2
x(sec
2
x) +(2sec
2
x tanx) tanx = 2sec
4
+4sec
4
x tan
2
x
y
, y
, y = t
2
sint
43. Calculatetherst vederivativesof f (x) = cosx. Thendeterminef
(8)
andf
(37)
.
solution Let f (x) = cosx.
Thenf
(x) = sinx, f
(x) = cosx, f
(x) = sinx, f
(4)
(x) = cosx, andf
(5)
(x) = sinx.
Accordingly, thesuccessivederivativesof f cycleamong
{sinx, cosx, sinx, cosx}
inthat order. Since8isamultipleof 4, wehavef
(8)
(x) = cosx.
Since36isamultipleof 4, wehavef
(36)
(x) = cosx. Therefore, f
(37)
(x) = sinx.
Findy
(157)
, wherey = sinx.
45. Findthevaluesof x between0and2 wherethetangent linetothegraphof y = sinx cosx ishorizontal.
solution Let y = sinx cosx. Then
y
= (sinx)(sinx) +(cosx)(cosx) = cos

2
x sin
2
x.
Wheny
= 0, wehavesinx = cosx. Intheinterval [0, 2], thisoccurswhenx =

4
,
3
4
,
5
4
,
7
4
.
Plot thegraphf () = sec + csc over [0, 2] anddeterminethenumber of solutionstof
() = 0inthis
interval graphically. Thencomputef
() andndthesolutions.
47. Let g(t ) = t sint .
(a) Plot thegraphof g withagraphingutilityfor 0 t 4.
(b) Showthat theslopeof thetangent lineisnonnegative. Verifythisonyour graph.
(c) For whichvaluesof t inthegivenrangeisthetangent linehorizontal?
solution Let g(t ) = t sint .
(a) Hereisagraphof g over theinterval [0, 4].
y
x
2 4 6 8 10 12
2
4
6
8
10
12
(b) Sinceg
(t ) = 1cost 0for all t , theslopeof thetangent linetog isalwaysnonnegative.

(c) Intheinterval [0, 4], thetangent lineishorizontal whent = 0, 2, 4.
Let f (x) = (sinx)/x for x = 0andf (0) = 1.
(a) Plot f (x) on[3, 3].
(b) Showthat f
(c) = 0if c = tanc. Usethenumerical root nder onacomputer algebrasystemto ndagood

approximationtothesmallest positive valuec
0
suchthat f
(c
0
) = 0.
(c) Verifythat thehorizontal liney = f (c
0
) istangent tothegraphof y = f (x) at x = c
0
byplottingthemonthe
sameset of axes.
49. Showthat notangent linetothegraphof f (x) = tanx haszeroslope. What istheleast slopeof atangent
line? J ustifybysketchingthegraphof (tanx)
.
solution Let f (x) = tanx. Thenf
(x) = sec
2
x =
1
cos
2
x
. Notethat f
(x) =
1
cos
2
x
hasnumerator 1; theequation
f
(x) = 0thereforehasnosolution. Becausethemaximumvalueof cos

2
x is1, theminimumvalueof f
(x) =
1
cos
2
x
is1. Hence, theleast slopefor atangent linetotanx is1. Hereisagraphof f
.
2 4 2 4
y
x
2
4
6
8
10
12
14
Theheight at timet (inseconds) of amass, oscillatingat theendof aspring, iss(t ) = 300+ 40sint cm. Find
thevelocityandaccelerationat t =

3
s.
51. The horizontal range R of a projectile launched fromground level at an angle and initial velocity v
0
m/s is
R = (v
2
0
/9.8) sin cos. CalculatedR/d. If = 7/24, will therangeincreaseor decreaseif theangleisincreased
slightly? Baseyour answer onthesignof thederivative.
solution Let R() = (v
2
0
/9.8) sin cos.
dR
d
= R
() = (v
2
0
/9.8)(sin
2
+cos
2
).
If = 7/24,

4
< <

2
, so| sin| > | cos|, anddR/d < 0(numerically, dR/d = 0.0264101v
2
0
). Atthispoint,
increasingtheanglewill decrease therange.
Showthat if

2
< < , then thedistancealong thex-axis between and thepoint wherethetangent line
intersectsthex-axisisequal to|tan| (Figure4).
53. Usethelimitdenitionof thederivativeandtheadditionlawfor thecosinefunctiontoprovethat(cosx)
= sinx.
solution Let f (x) = cosx. Then
f
(x) = lim
h0
cos(x +h) cosx
h
= lim
h0
cosx cosh sinx sinh cosx
h
= lim
h0
_
(sinx)
sinh
h
+(cosx)
cosh 1
h
_
= (sinx) 1+(cosx) 0= sinx.
Usetheadditionformulafor thetangent
tan(x +h) =
tanx +tanh
1+tanx tanh
tocompute(tanx)
directlyasalimit of thedifferencequotients. Youwill alsoneedtoshowthat lim

h0
tanh
h
= 1.
55. Verifythefollowingidentityanduseit togiveanother proof of theformula(sinx)
= cosx.
sin(x +h) sinx = 2cos
_
x +
1
2
h
_
sin
_
1
2
h
_
Hint: Usetheadditionformulatoprovethat sin(a +b) sin(a b) = 2cosa sinb.
solution Recall that
sin(a +b) = sina cosb +cosa sinb
and
sin(a b) = sina cosb cosa sinb.
Subtractingthesecondidentityfromtherst yields
sin(a +b) sin(a b) = 2cosa sinb.
If wenowset a = x +
h
2
andb =
h
2
, thenthepreviousequationbecomes
sin(x +h) sinx = 2cos
_
x +
h
2
_
sin
_
h
2
_
.
Finally, weusethelimit denitionof thederivativeof sinx toobtain
d
dx
sinx = lim
h0
sin(x +h) sinx
h
= lim
h0
2cos
_
x +
h
2
_
sin
_
h
2
_
h
= lim
h0
cos
_
x +
h
2
_
lim
h0
sin
_
h
2
_
_
h
2
_ = cosx 1= cosx.
Inother words,
d
dx
(sinx) = cosx.
Showthat anonzeropolynomial functiony = f (x) cannot satisfy theequationy
= y. Usethistoprove
that neither sinx nor cosx is apolynomial. Canyouthink of another way to reachthis conclusionby considering
limitsasx ?
57. Let f (x) = x sinx andg(x) = x cosx.
(a) Showthat f
(x) = g(x) +sinx andg
(x) = f (x) +cosx.

(b) Verifythat f
(x) = f (x) +2cosx and

g
(x) = g(x) 2sinx.

(c) Byfurtherexperimentation, trytondformulasforall higherderivativesof f andg. Hint: Thekthderivativedepends
onwhether k = 4n, 4n +1, 4n +2, or 4n +3.
solution Let f (x) = x sinx andg(x) = x cosx.
(a) Weexaminerst derivatives: f
(x) = x cosx +(sinx) 1= g(x) +sinx andg
(x) = (x)(sinx) +(cosx) 1=

f (x) +cosx; i.e., f
(x) = g(x) +sinx andg
(x) = f (x) +cosx.

(b) Now look at second derivatives: f
(x) = g
(x) + cosx = f (x) + 2cosx and g
(x) = f
(x) sinx =
g(x) 2sinx; i.e., f
(x) = f (x) +2cosx andg
(x) = g(x) 2sinx.

(c) Thethirdderivatives aref
(x) = f
(x) 2sinx = g(x) 3sinx andg
(x) = g
(x) 2cosx =
f (x) 3cosx; i.e., f
(x) = g(x) 3sinx andg
(x) = f (x) 3cosx.

The fourth derivatives are f
(4)
(x) = g
(x) 3cosx = f (x) 4cosx and g

(4)
(x) = f
(x) + 3sinx =
g(x) +4sinx; i.e., f
(4)
= f (x) 4cosx andg
(4)
(x) = g(x) +4sinx.
Wecannowseethepatternfor thederivatives, whicharesummarizedinthefollowingtable. Heren = 0, 1, 2, . . .
k 4n 4n +1 4n +2 4n +3
f
(k)
(x) f (x) k cosx g(x) +k sinx f (x) +k cosx g(x) k sinx
g
(k)
(x) g(x) +k sinx f (x) +k cosx g(x) k sinx f (x) k cosx
Figure5showsthegeometrybehindthederivativeformula(sin)
= cos. SegmentsBAandBD areparallel

tothex- andy-axes. Let sin = sin( +h) sin. Verifythefollowingstatements.
(a) sin = BC
(b)

BDA = Hint: OA AD.
(c) BD = (cos)AD
Nowexplainthefollowingintuitiveargument: If h issmall, thenBC BD andAD h, sosin (cos)h and
(sin)
= cos.
3.7 The Chain Rule
1. Identifytheoutsideandinsidefunctionsfor eachof thesecompositefunctions.
(a) y =
_
4x +9x
2
(b) y = tan(x
2
+1)
(c) y = sec
5
x (d) y = (1+x
12
)
4
solution
(a) Theouter functionis
x, andtheinner functionis4x +9x

2
.
(b) Theouter functionistanx, andtheinner functionisx
2
+1.
(c) Theouter functionisx
5
, andtheinner functionissecx.
(d) Theouter functionisx
4
, andtheinner functionis1+x
12
.
2. Whichof thefollowingcanbedifferentiatedeasilywithout usingtheChainRule?
(a) y = tan(7x
2
+2) (b) y =
x
x +1
(c) y =

x secx (d) y =

x cosx
(e) y = x sec
x (f) y = tan(4x)
x
x+1
canbedifferentiatedusingtheQuotient Rule, andthefunction
x secx canbediffer-
entiatedusingtheProductRule. Thefunctionstan(7x
2
+2),
x cosx, andtan(4x) requiretheChainRule. x sec
x can
bepartiallyevaluatedusingtheProductRule, butthentheChainruleisneededtodifferentiateoneof thefactors, sec
x.
3. Whichisthederivativeof f (5x)?
(a) 5f
(x) (b) 5f
(5x) (c) f
(5x)
solution Thecorrect answer is(b): 5f
(5x).
4. Supposethatf
(4) = g(4) = g
(4) = 1. DowehaveenoughinformationtocomputeF
(4), whereF(x) = f (g(x))?

If not, what ismissing?
solution If F(x) = f (g(x)), thenF
(x) = f
(g(x))g
(x) andF
(4) = f
(g(4))g
(4). Thus, wedonothaveenough

informationtocomputeF
(4). Wearemissingthevalueof f
(1).
Exercises
In Exercises 14, ll in a table of the following type:
f (g(x)) f
(u) f
(g(x)) g
(x) (f g)
1. f (u) = u
3/2
, g(x) = x
4
+1
solution
f (g(x)) f
(u) f
(g(x)) g
(x) (f g)
(x
4
+1)
3/2 3
2
u
1/2 3
2
(x
4
+1)
1/2
4x
3
6x
3
(x
4
+1)
1/2
S E C T I ON 3.7 The Chain Rule 139
f (u) = u
3
, g(x) = 3x +5
3. f (u) = tanu, g(x) = x
4
solution
f (g(x)) f
(u) f
(g(x)) g
(x) (f g)
tan(x
4
) sec
2
u sec
2
(x
4
) 4x
3
4x
3
sec
2
(x
4
)
f (u) = u
4
+u, g(x) = cosx
In Exercises 5 and 6, write the function as a composite f (g(x)) and compute the derivative using the Chain Rule.
5. y = (x +sinx)
4
4
, g(x) = x +sinx, andy = f (g(x)) = (x +sinx)
4
. Then
dy
dx
= f
(g(x))g
(x) = 4(x +sinx)

3
(1+cosx).
y = cos(x
3
)
7. Calculate
d
dx
cosu for thefollowingchoicesof u(x):
(a) u = 9x
2
(b) u = x
1
(c) u = tanx
solution
(a) cos(u(x)) = cos(9x
2
).
d
dx
cos(u(x)) = sin(u(x))u
(x) = sin(9x
2
)(2x) = 2x sin(9x
2
).
(b) cos(u(x)) = cos(x
1
).
d
dx
(x) = sin(x
1
)
_
1
x
2
_
=
sin(x
1
)
x
2
.
(c) cos(u(x)) = cos(tanx).
d
dx
(x) = sin(tanx)(sec
2
x) = sec
2
x sin(tanx).
Calculate
d
dx
f (x
2
+1) for thefollowingchoicesof f (u):
(a) f (u) = sinu (b) f (u) = 3u
3/2
(c) f (u) = u
2
u
9. Compute
df
dx
if
df
du
= 2and
du
dx
= 6.
solution Assumingf isafunctionof u, whichisinturnafunctionof x,
df
dx
=
df
du
du
dx
= 2(6) = 12.
Compute
df
dx
x=2
if f (u) = u
2
, u(2) = 5, andu
(2) = 5.
In Exercises 1122, use the General Power Rule or the Shifting and Scaling Rule to compute the derivative.
11. y = (x
4
+5)
3
solution UsingtheGeneral Power Rule,
d
dx
(x
4
+5)
3
= 3(x
4
+5)
2
d
dx
(x
4
+5) = 3(x
4
+5)
2
(4x
3
) = 12x
3
(x
4
+5)
2
.
y = (8x
4
+5)
3
13. y =

7x 3
solution UsingtheShiftingandScalingRule
d
dx
7x 3=
d
dx
(7x 3)
1/2
=
1
2
(7x 3)
1/2
(7) =
7
2
7x 3
.
y = (42x 3x
2
)
5
15. y = (x
2
+9x)
2
d
dx
(x
2
+9x)
2
= 2(x
2
+9x)
3
d
dx
(x
2
+9x) = 2(x
2
+9x)
3
(2x +9).
y = (x
3
+3x +9)
4/3
17. y = cos
4

d
d
cos
4
= 4cos
3
d
d
cos = 4cos
3
sin.
y = cos(9 +41)
19. y = (2cos +5sin)
9
d
d
(2cos +5sin)
9
= 9(2cos +5sin)
8
d
d
(2cos +5sin) = 9(2cos +5sin)
8
(5cos 2sin).
y =

9+x +sinx
21. y = sin
__
x
2
+2x +9
_
solution Usingthegeneral power rule,
d
dx
sin
__
x
2
+2x +9
_
= cos
__
x
2
+2x +9
_
1
2
(x
2
+2x +9)
1/2
(2x +2)
= (x +1)(x
2
+2x +9)
1/2
cos
__
x
2
+2x +9
_
y = tan(43x) sec(34x)
In Exercises 2326, compute the derivative of f g.
23. f (u) = sinu, g(x) = 2x +1
solution Let h(x) = f (g(x)) = sin(2x + 1). Then, applyingtheshiftingandscalingrule, h
(x) = 2cos(2x + 1).

Alternately,
d
dx
f (g(x)) = f
(g(x))g
(x) = cos(2x +1) 2= 2cos(2x +1).

f (u) = 2u +1, g(x) = sinx
25. f (u) = u +u
1
, g(x) = tanx
solution Let h(x) = f (g(x)) = tanx +cotx. Thenh
(x) = sec
2
x csc
2
x. Alternatively,
d
dx
f (g(x)) = f
(g(x))g
(x) = (1cot
2
x) sec
2
x = sec
2
x csc
2
x
f (u) =
u
u 1
, g(x) = cscx
In Exercises 27 and 28, nd the derivatives of f (g(x)) and g(f (x)).
27. f (u) = cosu, u = g(x) = x
2
+1
solution
d
dx
f (g(x)) = f
(g(x))g
(x) = sin(x
2
+1)(2x) = 2x sin(x
2
+1).
d
dx
g(f (x)) = g
(f (x))f
(x) = 2(cosx)(sinx) = 2sinx cosx.

f (u) = u
3
, u = g(x) =
1
x +1
In Exercises 2942, use the Chain Rule to nd the derivative.
29. y = sin(x
2
)
solution Let y = sin
_
x
2
_
. Theny
= cos
_
x
2
_
2x = 2x cos
_
x
2
_
.
y = sin
2
x
31. y =
_
t
2
+9
solution Let y =
_
t
2
+9= (t
2
+9)
1/2
. Then
y
=
1
2
(t
2
+9)
1/2
(2t ) =
t
_
t
2
+9
.
y = (t
2
+3t +1)
5/2
33. y = (x
4
x
3
1)
2/3
solution Let y =
_
x
4
x
3
1
_
2/3
. Then
y
=
2
3
_
x
4
x
3
1
_
1/3
_
4x
3
3x
2
_
.
y = (
x +11)
3/2
35. y =
_
x +1
x 1
_
4
solution Let y =
_
x +1
x 1
_
4
. Then
y
= 4
_
x +1
x 1
_
3
(x 1) 1(x +1) 1
(x 1)
2
=
8(x +1)
3
(x 1)
5
=
8(1+x)
3
(1x)
5
.
y = cos
3
(12)
37. y = sec
1
x
solution Let f (x) = sec
_
x
1
_
. Then
f
(x) = sec
_
x
1
_
tan
_
x
1
_
_
x
2
_
=
sec(1/x) tan(1/x)
x
2
.
y = tan(
2
4)
39. y = tan( +cos)
solution Let y = tan( +cos). Then
y
= sec
2
( +cos) (1sin) = (1sin) sec
2
( +cos) .
y =
_
cot
9
+1
41. y = csc(92
2
)
solution Wehave
y
= csc(92
2
) cot(92
2
) (4) = 4 csc(92
2
) cot(92
2
)
y = cot(
1)
In Exercises 4372, nd the derivative using the appropriate rule or combination of rules.
43. y = tan(x
2
+4x)
solution Let y = tan(x
2
+4x). Bythechainrule,
y
= sec
2
(x
2
+4x) (2x +4) = (2x +4) sec
2
(x
2
+4x).
y = sin(x
2
+4x)
45. y = x cos(13x)
solution Let y = x cos(13x). Applyingtheproduct ruleandthenthescalingandshiftingrule,
y
= x (sin(13x)) (3) +cos(13x) 1= 3x sin(13x) +cos(13x) .

y = sin(x
2
) cos(x
2
)
47. y = (4t +9)
1/2
solution Let y = (4t +9)
1/2
. Bytheshiftingandscalingrule,
dy
dt
= 4
_
1
2
_
(4t +9)
1/2
= 2(4t +9)
1/2
.
y = (z +1)
4
(2z 1)
3
49. y = (x
3
+cosx)
4
solution Let y = (x
3
+cosx)
4
. Bythegeneral power rule,
y
= 4(x
3
+cosx)
5
(3x
2
sinx) = 4(sinx 3x
2
)(x
3
+cosx)
5
.
y = sin(cos(sinx))
51. y =
sinx cosx
solution Westart byusingatrigidentitytorewrite
y =
sinx cosx =
_
1
2
sin2x =
1
2
(sin2x)
1/2
.
Then, after twoapplicationsof thechainrule,
y
=
1
1
2
(sin2x)
1/2
cos2x 2=
cos2x
2sin2x
.
y = (9(52x
4
)
7
)
3
53. y = (cos6x +sinx
2
)
1/2
solution Let y = (cos6x + sin(x
2
))
1/2
. Applyingthegeneral power rulefollowedby boththescalingandshifting
ruleandthechainrule,
y
=
1
2
_
cos6x +sin(x
2
)
_
1/2
_
sin6x 6+cos(x
2
) 2x
_
=
x cos(x
2
) 3sin6x
_
cos6x +sin(x
2
)
.
y =
(x +1)
1/2
x +2
55. y = tan
3
x +tan(x
3
)
solution Let y = tan
3
x + tan(x
3
) = (tanx)
3
+ tan(x
3
). Applyingthegeneral power ruletotherst termandthe
chainruletothesecondterm,
y
= 3(tanx)
2
sec
2
x +sec
2
(x
3
) 3x
2
= 3
_
x
2
sec
2
(x
3
) +sec
2
x tan
2
x
_
.
y =

43cosx 57. y =
_
z +1
z 1
solution Let y =
_
z +1
z 1
_
1/2
. Applyingthegeneral power rulefollowedbythequotient rule,
dy
dz
=
1
2
_
z +1
z 1
_
1/2
(z 1) 1(z +1) 1
(z 1)
2
=
1
z +1(z 1)
3/2
.
y = (cos
3
x +3cosx +7)
9 59. y =
cos(1+x)
1+cosx
solution Let
y =
cos(1+x)
1+cosx
.
Then, applyingthequotient ruleandtheshiftingandscalingrule,
dy
dx
=
(1+cosx) sin(1+x) +cos(1+x) sinx
(1+cosx)
2
=
cos(1+x) sinx cosx sin(1+x) sin(1+x)
(1+cosx)
2
=
sin(1) sin(1+x)
(1+cosx)
2
.
Thelast linefollowsfromtheidentity
sin(A B) = sinAcosB cosAsinB
withA = x andB = 1+x.
y = sec(
_
t
2
9)
61. y = cot
7
(x
5
)
solution Let y = cot
7
_
x
5
_
. Applyingthegeneral power rulefollowedbythechainrule,
dy
dx
= 7cot
6
_
x
5
_
_
csc
2
_
x
5
__
5x
4
= 35x
4
cot
6
_
x
5
_
csc
2
_
x
5
_
.
y =
cos(1/x)
1+x
2
63. y =
_
1+cot
5
(x
4
+1)
_
9
solution Lety =
_
1+cot
5
_
x
4
+1
__
9
. Applyingthegeneral power rule, thechainrule, andthegeneral power rule
insuccession,
dy
dx
= 9
_
1+cot
5
_
x
4
+1
__
8
5cot
4
_
x
4
+1
_
_
csc
2
_
x
4
+1
__
4x
3
= 180x
3
cot
4
_
x
4
+1
_
csc
2
_
x
4
+1
_ _
1+cot
5
_
x
4
+1
__
8
.
y =

cos2x +sin4x
65. y = (1csc
2
(1x
3
))
6
solution Usingthechainrulemultipletimes, wehave
d
dx
(1csc
2
(1x
3
))
6
= 6(1csc
2
(1x
3
))
5
d
dx
(1csc
2
(1x
3
))
= 6(1csc
2
(1x
3
))
5
(2csc(1x
3
))(csc(1x
3
) cot(1x
3
))(3x
2
)
= 36x
2
csc
2
(1x
3
) cot(1x
3
)(1csc
2
(1x
3
))
5
y = sin(
sin +1)
67. y =
_
x +
1
x
_
1/2
solution Applyingthechainrulegives
d
dx
_
x +
1
x
_
1/2
=
1
2
_
x +
1
x
_
3/2
(1x
2
) =
1
2
(x
2
1)(x +x
1
)
3/2
y = sec(1+(4+x)
3/2
)
69. y =
_
1+
_
1+
x
solution Let y =
_
1+
_
1+x
1/2
_
1/2
_
1/2
. Applyingthegeneral power ruletwice,
dy
dx
=
1
2
_
1+
_
1+x
1/2
_
1/2
_
1/2
1
2
_
1+x
1/2
_
1/2
1
2
x
1/2
=
1
8
x
_
1+
x
_
1+
_
1+
x
.
y =
_
x +1+1
71. y = (kx +b)
1/3
; k andb anyconstants
solution Let y = (kx +b)
1/3
, whereb andk areconstants. Bythescalingandshiftingrule,
y
=
1
3
(kx +b)
4/3
k =
k
3
(kx +b)
4/3
.
y =
1
_
kt
4
+b
; k, b constants, not bothzero
In Exercises 7376, compute the higher derivative.
73.
d
2
dx
2
sin(x
2
)
_
x
2
_
. Then, bythechainrule, f
(x) = 2x cos
_
x
2
_
and, bytheproduct ruleandthechain
rule,
f
(x) = 2x
_
sin
_
x
2
_
2x
_
+2cos
_
x
2
_
= 2cos
_
x
2
_
4x
2
sin
_
x
2
_
.
d
2
dx
2
(x
2
+9)
5
75.
d
3
dx
3
(9x)
8
solution Let f (x) = (9x)
8
. Then, byrepeateduseof thescalingandshiftingrule,
f
(x) = 8(9x)
7
(1) = 8(9x)
7
f
(x) = 56(9x)
6
(1) = 56(9x)
6
,
f
(x) = 336(9x)
5
(1) = 336(9x)
5
.
d
3
dx
3
sin(2x)
77. The average molecular velocity v of a gas in a certain container is given by v = 29
T m/s, where T is the

temperatureinkelvins. Thetemperatureisrelatedtothepressure(inatmospheres) byT = 200P. Find
dv
dP
P=1.5
.
solution First notethat whenP = 1.5atmospheres, T = 200(1.5) = 300K. Thus,
dv
dP
P=1.5
=
dv
dT
T =300
dT
dP
P=1.5
=
29
2
300
200=
290
3
3
m
s atmospheres
.
Alternately, substitutingT = 200P intotheequationfor v givesv = 290
2P. Therefore,
dv
dP
=
290
2
2
P
=
290
2P
,
so
dv
dP
P=1.5
=
290
3
=
290
3
3
m
s atmospheres
.
Thepower P inacircuit is P = Ri
2
, whereR is theresistanceandi is thecurrent. FinddP/dt at t =
1
3
if
R = 1000 andi variesaccordingtoi = sin(4t ) (timeinseconds).
79. Anexpandingspherehas radius r = 0.4t cmat timet (inseconds). Let V bethespheres volume. FinddV/dt
when(a) r = 3and(b) t = 3.
solution Let r = 0.4t , wheret isinseconds(s) andr isincentimeters(cm). WithV =
4
3
r
3
, wehave
dV
dr
= 4r
2
.
Thus
dV
dt
=
dV
dr
dr
dt
= 4r
2
(0.4) = 1.6r
2
.
(a) Whenr = 3,
dV
dt
= 1.6(3)
2
45.24cm/s.
(b) Whent = 3, wehaver = 1.2. Hence
dV
dt
= 1.6(1.2)
2
7.24cm/s.
A 2005study by theFisheriesResearchServicesinAberdeen, Scotland, suggeststhat theaveragelengthof the
speciesClupea harengus (Atlanticherring) asafunctionof aget (inyears) canbemodeledby
L(t ) = 32
_
1(1+0.37t +0.068t
2
+0.0085t
3
+0.0009t
4
)
1
_
for 0 t 13. SeeFigure2.
(a) Howfast istheaveragelengthchangingat aget = 6years?
(b) At what ageistheaveragelengthchangingat arateof 5cm/yr?
81. According to a1999 study by Starkey and Scarnecchia, theaverageweight (in kilograms) at aget (in years) of
channel catshintheLower YellowstoneRiver (Figure3) isapproximatedbythefunction:
W(t ) = (0.14+0.115t 0.002t
2
+0.000023t
3
)
3.4
Findtherateat whichaverageweight ischangingat aget = 10.
5 10 15 20
1
2
3
4
5
6
7
8
t (year)
W (kg)
Lower YellowstoneRiver
FIGURE 3 Averageweight of channel catshat aget
solution Let W(t ) = (0.14+0.115t 0.002t
2
+0.000023t
3
)
3.4
. Then
W
(t ) = 3.4(0.14+0.115t 0.002t
2
+0.000023t
3
)
2.4
(0.1150.004t +0.000069t
2
)
At aget = 10,
W
(10) 0.36kg/yr.
CalculateM
(0) intermsof theconstantsa, b, k, andm, where

M(t ) =
_
a +(b a)
_
1+kmt +
1
2
(kmt )
2
__
1/m
83. WithnotationasinExample7, calculate
(a)
d
d
sin
=60
(b)
d
d
_
+tan
_
=45
solution
(a)
d
d
sin
=60
=
d
d
sin
_

180
=60
=
_

180
_
cos
_

180
(60)
_
=

180
1
2
=

360
.
(b)
d
d
_
+tan
_
=45
=
d
d
_
+tan
_

180
__
=45
= 1+

180
sec
2
_
4
_
= 1+

90
.
Assumethat
f (0) = 2, f
(0) = 3, h(0) = 1, h
(0) = 7
Calculatethederivativesof thefollowingfunctionsat x = 0:
(a) (f (x))
3
(b) f (7x) (c) f (4x)h(5x)
85. Computethederivativeof h(sinx) at x =

6
, assumingthat h
(0.5) = 10.
solution Let u = sinx andsupposethat h
(0.5) = 10. Then

d
dx
(h(u)) =
dh
du
du
dx
=
dh
du
cosx.
Whenx =

6
, wehaveu = 0.5. Accordingly, thederivativeof h(sinx) at x =

6
is10cos
_
6
_
= 5
3.
Let F(x) = f (g(x)), wherethegraphs of f and g areshown in Figure4. Estimateg
(2) and f
(g(2)) and
computeF
(2).
In Exercises 8790, use the table of values to calculate the derivative of the function at the given point.
x 1 4 6
f (x) 4 0 6
f
(x) 5 7 4
g(x) 4 1 6
g
(x) 5
1
2
3
87. f (g(x)), x = 6
solution
d
dx
f (g(x))
x=6
= f
(g(6))g
(6) = f
(6)g
(6) = 43= 12.

sin(g(x)), x = 4
89. g(
x), x = 16
solution
d
dx
g(
x)
x=16
= g
(4)
_
1
2
_
(1/
16) =
_
1
2
__
1
2
__
1
4
_
=
1
16
.
f (2x +g(x)), x = 1
91. Theprice(indollars) of acomputer componentisP = 2C 18C
1
, whereC isthemanufacturerscosttoproduce
it. Assumethat cost at timet (inyears) isC = 9+ 3t
1
. Determinetherateof changeof pricewithrespect totimeat
t = 3.
solution
dC
dt
= 3t
2
. C(3) = 10andC
(3) =
1
3
, sowecompute:
dP
dt
t =3
= 2C
(3) +
18
(C(3))
2
C
(3) =
2
3
+
18
100
_
1
3
_
= 0.727
dollars
year
.
Plottheastroidy = (4x
2/3
)
3/2
for0 x 8. Showthatthepartof everytangentlineintherstquadrant
hasaconstant length8.
93. AccordingtotheU.S. standardatmospheric model, developedby theNational Oceanic andAtmosphericAdmin-
istrationfor useinaircraft androcket design, atmospherictemperatureT (indegreesCelsius), pressureP (kPa= 1000
pascals), andaltitudeh (inmeters) arerelatedbytheseformulas(validinthetroposphereh 11,000):
T = 15.040.000649h, P = 101.29+
_
T +273.1
288.08
_
5.256
UsetheChainRuletocalculatedP/dh. ThenestimatethechangeinP (inpascals, Pa) per additional meter of altitude
whenh = 3000.
solution
dP
dT
= 5.256
_
T +273.1
288.08
_
4.256
_
1
288.08
_
= 6.2151910
13
(273.1+T )
4.256
and
dT
dh
= 0.000649
C/m.
dP
dh
=
dP
dT
dT
dh
, so
dP
dh
=
_
6.2151910
13
(273.1+T )
4.256
_
(0.000649) = 4.0336610
16
(288.140.000649h)
4.256
.
Whenh = 3000,
dP
dh
= 4.0336610
16
(286.193)
4.256
= 1.1510
5
kPa/m;
therefore, for eachadditional meter of altitude,
P 1.1510
5
kPa= 1.1510
2
Pa.
ClimatescientistsusetheStefan-BoltzmannLawR = T
4
toestimatethechangeintheearthsaveragetemper-
atureT (inkelvins) causedbyachangeintheradiationR (injoulespersquaremeterpersecond) thattheearthreceives
fromthesun. Here = 5.6710
8
J s
1
m
2
K
4
. CalculatedR/dt , assumingthatT = 283and
dT
dt
= 0.05K/yr.
What aretheunitsof thederivative?
95. Inthesettingof Exercise94, calculatetheyearlyrateof changeof T if T = 283K andR increasesat arateof 0.5
J s
1
m
2
per year.
solution BytheChainRule,
dR
dt
=
dR
dT
dT
dt
= 4T
3
dT
dt
.
AssumingT = 283K and
dR
dt
= 0.5J s
1
m
2
per year, it followsthat author:
0.5= 4(283)
3
dT
dt
dT
dt
=
0.5
4(283)
3
0.0973 kelvins/yr
Useacomputer algebrasystemtocomputef
(k)
(x) for k = 1, 2, 3for thefollowingfunctions:
(a) f (x) = cot(x
2
) (b) f (x) =
_
x
3
+1
97. UsetheChainRuletoexpressthesecondderivativeof f g intermsof therstandsecondderivativesof f andg.
solution Let h(x) = f (g(x)). Then
h
(x) = f
(g(x))g
(x)
and
h
(x) = f
(g(x))g
(x) +g
(x)f
(g(x))g
(x) = f
(g(x))g
(x) +f
(g(x))
_
g
(x)
_
2
.
Computethesecondderivativeof sin(g(x)) at x = 2, assumingthat g(2) =

4
, g
(2) = 5, andg
(2) = 3.
99. Showthat if f , g, andh aredifferentiable, then
[f (g(h(x)))]
= f
(g(h(x)))g
(h(x))h
(x)
solution Let f , g, andh bedifferentiable. Let u = h(x), v = g(u), andw = f (v). Then
dw
dx
=
df
dv
dv
dx
=
df
dv
dg
du
du
dx
= f
(g(h(x))g
(h(x))h
(x)
Showthat differentiationreverses parity: If f is even, thenf
is odd, andif f is odd, thenf
is even. Hint:
Differentiatef (x).
101. (a) Sketchagraphof anyevenfunctionf (x) andexplaingraphicallywhyf
(x) isodd.
(b) Supposethat f
(x) iseven. Isf (x) necessarilyodd? Hint: Checkwhether thisistruefor linear functions.
solution
(a) Thegraphof anevenfunctionissymmetricwithrespect tothey-axis. Accordingly, itsimageintheleft half-planeis
amirror reectionof that intheright half-planethroughthey-axis. If at x = a 0, theslopeof f existsandisequal to
m, thenby reectionitsslopeat x = a 0ism. That is, f
(a) = f
(a). Note: Thismeansthat if f
(0) exists,
thenit equals0.
y
x
2 1 1 2
1
2
3
4
(b) Supposethat f
iseven. Thenf isnot necessarilyodd. Let f (x) = 4x +7. Thenf
(x) = 4, anevenfunction. But

f isnot odd. For example, f (2) = 15, f (2) = 1, but f (2) = f (2).
Power Rulefor Fractional Exponents Let f (u) = u
q
andg(x) = x
p/q
. Assumethat g(x) isdifferentiable.
(a) Showthat f (g(x)) = x
p
(recall thelawsof exponents).
(b) ApplytheChainRuleandthePower Rulefor whole-number exponentstoshowthat f
(g(x)) g
(x) = px
p1
.
(c) ThenderivethePower Rulefor x
p/q
.
103. Provethat for all wholenumbersn 1,
d
n
dx
n
sinx = sin
_
x +
n
2
_
Hint: Usetheidentitycosx = sin
_
x +

2
_
.
solution Wewill proceedbyinductiononn. For n = 1, wend
d
dx
sinx = cosx = sin
_
x +

2
_
,
asrequired. Now, supposethat for somepositiveinteger k,
d
k
dx
k
sinx = sin
_
x +
k
2
_
.
Then
d
k+1
dx
k+1
sinx =
d
dx
sin
_
x +
k
2
_
= cos
_
x +
k
2
_
= sin
_
x +
(k +1)
2
_
.
ADiscontinuousDerivative Usethelimit denitiontoshowthat g
(0) existsbut g
(0) = lim
x0
g
(x), where
g(x) =
x
2
sin
1
x
x = 0
0 x = 0
105. ChainRule ThisexerciseprovestheChainRulewithoutthespecial assumptionmadeinthetext. For anynumber
b, deneanewfunction
F(u) =
f (u) f (b)
u b
for all u = b
(a) Showthat if wedeneF(b) = f
(b), thenF(u) iscontinuousat u = b.

(b) Takeb = g(a). Showthat if x = a, thenfor all u,
f (u) f (g(a))
x a
= F(u)
u g(a)
x a
2
Notethat bothsidesarezeroif u = g(a).
(c) Substituteu = g(x) inEq. (2) toobtain
f (g(x)) f (g(a))
x a
= F(g(x))
g(x) g(a)
x a
DerivetheChainRulebycomputingthelimit of bothsidesasx a.
solution For anydifferentiablefunctionf andanynumber b, dene
F(u) =
f (u) f (b)
u b
for all u = b.
S E C T I ON 3.8 Implicit Differentiation 147
(a) DeneF(b) = f
(b). Then
lim
ub
F(u) = lim
ub
f (u) f (b)
u b
= f
(b) = F(b),
i.e., lim
ub
F(u) = F(b). Therefore, F iscontinuousat u = b.
(b) Let g beadifferentiablefunctionandtakeb = g(a). Let x beanumber distinct froma. If wesubstituteu = g(a)
intoEq. (2), bothsidesevaluateto0, soequalityissatised. Ontheother hand, if u = g(a), then
f (u) f (g(a))
x a
=
f (u) f (g(a))
u g(a)
u g(a)
x a
=
f (u) f (b)
u b
u g(a)
x a
= F(u)
u g(a)
x a
.
(c) Hencefor all u, wehave
f (u) f (g(a))
x a
= F(u)
u g(a)
x a
.
(d) Substitutingu = g(x) inEq. (2), wehave
f (g(x)) f (g(a))
x a
= F(g(x))
g(x) g(a)
x a
.
Lettingx a gives
lim
xa
f (g(x)) f (g(a))
x a
= lim
xa
_
F(g(x))
g(x) g(a)
x a
_
= F(g(a))g
(a) = F(b)g
(a) = f
(b)g
(a)
= f
(g(a))g
(a)
Therefore(f g)
(a) = f
(g(a))g
(a), whichistheChainRule.
3.8 Implicit Differentiation
1. Whichdifferentiationruleisusedtoshow
d
dx
siny = cosy
dy
dx
?
solution Thechainruleisusedtoshowthat
d
dx
siny = cosy
dy
dx
.
2. Oneof (a)(c) isincorrect. Findandcorrect themistake.
(a)
d
dy
sin(y
2
) = 2y cos(y
2
) (b)
d
dx
sin(x
2
) = 2x cos(x
2
) (c)
d
dx
sin(y
2
) = 2y cos(y
2
)
solution
(a) Thisiscorrect. Notethat thedifferentiationiswithrespect tothevariabley.
(b) Thisiscorrect. Notethat thedifferentiationiswithrespect tothevariablex.
(c) Thisisincorrect. Becausethedifferentiationiswithrespect tothevariablex, thechainruleisneededtoobtain
d
dx
sin(y
2
) = 2y cos(y
2
)
dy
dx
.
3. Onanexam, J asonwasaskedtodifferentiatetheequation
x
2
+2xy +y
3
= 7
FindtheerrorsinJ asonsanswer: 2x +2xy
+3y
2
= 0
solution Therearetwo mistakes inJ asons answer. First, J asonshouldhaveappliedtheproduct ruleto thesecond
termtoobtain
d
dx
(2xy) = 2x
dy
dx
+2y.
Second, heshouldhaveappliedthegeneral power ruletothethirdtermtoobtain
d
dx
y
3
= 3y
2
dy
dx
.
4. Whichof (a) or (b) isequal to
d
dx
(x sint )?
(a) (x cost )
dt
dx
(b) (x cost )
dt
dx
+sint
solution Usingtheproduct ruleandthechainruleweseethat
d
dx
(x sint ) = x cost
dt
dx
+sint,
sothecorrect answer is(b).
Exercises
1. Showthat if youdifferentiatebothsidesof x
2
+2y
3
= 6, theresult is2x +6y
2
dy
dx
= 0. Thensolvefor dy/dx and
evaluateit at thepoint (2, 1).
solution
d
dx
(x
2
+2y
3
) =
d
dx
6
2x +6y
2
dy
dx
= 0
2x +6y
2
dy
dx
= 0
6y
2
dy
dx
= 2x
dy
dx
=
2x
6y
2
.
At (2, 1),
dy
dx
=
4
6
=
2
3
.
Showthat if youdifferentiatebothsidesof xy +4x +2y = 1, theresult is(x +2)
dy
dx
+y +4= 0. Thensolve
for dy/dx andevaluateit at thepoint (1, 1).
In Exercises 38, differentiate the expression with respect to x, assuming that y = f (x).
3. x
2
y
3
solution Assumingthat y dependsonx, then
d
dx
_
x
2
y
3
_
= x
2
3y
2
y
+y
3
2x = 3x
2
y
2
y
+2xy
3
.
x
3
y
2
5. (x
2
+y
2
)
3/2
d
dx
_
_
x
2
+y
2
_
3/2
_
=
3
2
_
x
2
+y
2
_
1/2_
2x +2yy
_
= 3
_
x +yy
_
_
x
2
+y
2
.
tan(xy)
7.
y
y +1
d
dx
y
y +1
=
(y +1)y
yy
(y +1)
2
=
y
(y +1)
2
.
sin
y
x
In Exercises 926, calculate the derivative with respect to x.
9. 3y
3
+x
2
= 5
solution Let 3y
3
+x
2
= 5. Then9y
2
y
+2x = 0, andy
=
2x
9y
2
.
y
4
2y = 4x
3
+x
11. x
2
y +2x
3
y = x +y
solution Let x
2
y +2x
3
y = x +y. Then
x
2
y
+2xy +2x
3
y
+6x
2
y = 1+y
x
2
y
+2x
3
y
= 12xy 6x
2
y
y
=
12xy 6x
2
y
x
2
+2x
3
1
.
xy
2
+x
2
y
5
x
3
= 3
13. x
3
R
5
= 1
solution Let x
3
R
5
= 1. Thenx
3
5R
4
R
+R
5
3x
2
= 0, andR
=
3x
2
R
5
5x
3
R
4
=
3R
5x
.
x
4
+z
4
= 1
15.
y
x
+
x
y
= 2y
solution Let
y
x
+
x
y
= 2y.
Then
xy
y
x
2
+
y xy
y
2
= 2y
_
1
x
x
y
2
2
_
y
=
y
x
2

1
y
y
2
x
2
2xy
2
xy
2
y
=
y
2
x
2
x
2
y
y
=
y(y
2
x
2
)
x(y
2
x
2
2xy
2
)
.
x +s =
1
x
+
1
s
17. y
2/3
+x
3/2
= 1
solution Let y
2/3
+x
3/2
= 1. Then
2
3
y
5/3
y
+
3
2
x
1/2
= 0 or y
=
9
4
x
1/2
y
5/3
.
x
1/2
+y
2/3
= 4y
19. y +
1
y
= x
2
+x
solution Let y +
1
y
= x
2
+x. Then
y

1
y
2
y
= 2x +1 or y
=
2x +1
1y
2
=
(2x +1)y
2
y
2
1
.
sin(xt ) = t
21. sin(x +y) = x +cosy
solution Let sin(x +y) = x +cosy. Then
(1+y
) cos(x +y) = 1y
siny
cos(x +y) +y
cos(x +y) = 1y
siny
(cos(x +y) +siny) y
= 1cos(x +y)
y
=
1cos(x +y)
cos(x +y) +siny
.
tan(x
2
y) = (x +y)
3
23. tan(x +y) = tanx +tany
solution Implicitlydifferentiatinggives
sec
2
(x +y) +y
sec
2
(x +y) = sec
2
x +y
sec
2
y
y
(sec
2
(x +y) sec
2
y) = sec
2
x sec
2
(x +y)
y
=
sec
2
x sec
2
(x +y)
sec
2
(x +y) sec
2
y
x siny y cosx = 2
25. x +cos(3x y) = xy
solution Differentiateimplicitlytoget
13sin(3x y) +y
sin(3x y) = y +xy
(sin(3x y) x) = y 1+3sin(3x y)
y
=
y 1+3sin(3x y)
sin(3x y) x
2x
2
x y =
_
x
4
+y
4
27. Showthat x + yx
1
= 1andy = x x
2
denethesamecurve(except that (0, 0) is not asolution of therst
equation) andthat implicit differentiationyieldsy
= yx
1
x andy
= 12x. Explainwhytheseformulasproduce
thesamevaluesfor thederivative.
solution Multiplytherst equationbyx andthenisolatethey termtoobtain
x
2
+y = x y = x x
2
.
Implicit differentiationappliedtotherst equationyields
1yx
2
+x
1
y
= 0 or y
= yx
1
x.
Fromtherst equation, wendyx
1
= 1x; uponsubstitutingthisexpressionintothepreviousderivative, wend
y
= 1x x = 12x,
whichisthederivativeof thesecondequation.
Usethemethodof Example4tocompute
dy
dx
P
at P = (2, 1) onthecurvey
2
x
3
+y
3
x
4
10x +y = 5.
In Exercises 29 and 30, nd dy/dx at the given point.
29. (x +2)
2
6(2y +3)
2
= 3, (1, 1)
solution Bythescalingandshiftingrule,
2(x +2) 24(2y +3)y
= 0.
If x = 1andy = 1, then
2(3) 24(1)y
= 0.
sothat 24y
= 6, or y
=
1
4
.
sin
2
(3y) = x +y,
_
2
4
,
4
_
In Exercises 3138, nd an equation of the tangent line at the given point.
31. xy +x
2
y
2
= 5, (2, 1)
solution Takingthederivativeof bothsidesof xy +x
2
y
2
= 5yields
xy
+y +2xy
2
+2x
2
yy
= 0.
Substitutingx = 2, y = 1, wend
2y
+1+4+8y
= 0 or y
=
1
2
.
Hence, theequationof thetangent lineat (2, 1) isy 1=
1
2
(x 2) or y =
1
2
x +2.
x
2/3
+y
2/3
= 2, (1, 1)
33. x
2
+siny = xy
2
+1, (1, 0)
solution Takingthederivativeof bothsidesof x
2
+siny = xy
2
+1yields
2x +cosyy
= y
2
+2xyy
.
2+y
= 0 or y
= 2.
Hence, theequationof thetangent lineisy 0= 2(x 1) or y = 2x +2.
sin(x y) = x cos
_
y +

4
_
,
_
4
,

4
_
35. 2x
1/2
+4y
1/2
= xy, (1, 4)
solution Takingthederivativeof bothsidesof 2x
1/2
+4y
1/2
= xy yields
x
1/2
2y
3/2
y
= xy
+y.
12
_
1
8
_
y
= y
+4 or y
=
12
5
.
Hence, theequationof thetangent lineisy 4=
12
5
(x 1) or y =
12
5
x +
32
5
.
x
x +1
+
y
y +1
= 1, (1, 1)
37. sin(2x y) =
x
2
y
, (0, )
solution Usingimplicit differentiation,
2cos(2x y) y
cos(2x y) =
2x
y
x
2
y
y
2
At (0, ), this gives 2cos() y
cos() = 0, or 2+ y
= 0, so that y
= 2. Thetangent lineat (0, ) is thus

y = 2x, or y = 2x +.
x +
x = y
2
+y
4
, (1, 1)
39. Findthepointsonthegraphof y
2
= x
3
3x +1(Figure5) wherethetangent lineishorizontal.
(a) First showthat 2yy
= 3x
2
3, wherey
= dy/dx.
(b) Donot solvefor y
. Rather, set y
= 0andsolvefor x. Thisyieldstwovaluesof x wheretheslopemaybezero.

(c) Showthat thepositivevalueof x doesnot correspondtoapoint onthegraph.
(d) The negative value corresponds to the two points on the graph where the tangent line is horizontal. Find their
coordinates.
2
2
2 1 1 2
x
y
FIGURE 5 Graphof y
2
= x
3
3x +1.
solution
(a) Applyingimplicit differentiationtoy
2
= x
3
3x +1, wehave
2y
dy
dx
= 3x
2
3.
(b) Settingy
= 0wehave0= 3x
2
3, sox = 1or x = 1.
(c) If wereturntotheequationy
2
= x
3
3x +1andsubstitutex = 1, weobtaintheequationy
2
= 1, whichhasno
real solutions.
(d) Substitutingx = 1intoy
2
= x
3
3x +1yields
y
2
= (1)
3
3(1) +1= 1+3+1= 3,
soy =

3or
3. Thetangent ishorizontal at thepoints(1,
3) and(1,
3).
Show, bydifferentiatingtheequation, that if thetangent lineat apoint (x, y) onthecurvex
2
y 2x +8y = 2is
horizontal, thenxy = 1. Thensubstitutey = x
1
inx
2
y 2x +8y = 2toshowthat thetangent lineishorizontal
at thepoints
_
2,
1
2
_
and
_
4,
1
4
_
.
41. Findall pointsonthegraphof 3x
2
+4y
2
+3xy = 24wherethetangent lineishorizontal (Figure6).
x
y
FIGURE 6 Graphof 3x
2
+4y
2
+3xy = 24.
solution Differentiatingtheequation3x
2
+4y
2
+3xy = 24implicitlyyields
6x +8yy
+3xy
+3y = 0,
so
y
=
6x +3y
8y +3x
.
Settingy
= 0leadsto6x +3y = 0, or y = 2x. Substitutingy = 2x intotheequation3x

2
+4y
2
+3xy = 24yields
3x
2
+4(2x)
2
+3x(2x) = 24,
or 13x
2
= 24. Thus, x = 2
78/13, andthecoordinates of thetwo points onthegraphof 3x

2
+ 4y
2
+ 3xy = 24
wherethetangent lineishorizontal are
_
2
78
13
,
4
78
13
_
and
_
78
13
,
4
78
13
_
.
Showthat nopoint onthegraphof x
2
3xy +y
2
= 1hasahorizontal tangent line.
43. Figure1showsthegraphof y
4
+xy = x
3
x +2. Finddy/dx at thetwopointsonthegraphwithx-coordinate0
andndanequationof thetangent lineat (1, 1).
solution Consider theequationy
4
+xy = x
3
x +2. Then4y
3
y
+xy
+y = 3x
2
1, and
y
=
3x
2
y 1
x +4y
3
.
Substitutingx = 0into y
4
+ xy = x
3
x + 2gives y
4
= 2, whichhas two real solutions, y = 2
1/4
. When
y = 2
1/4
, wehave
y
=
2
1/4
1
4
_
2
3/4
_ =
2+
4
2
8
0.3254.
Wheny = 2
1/4
, wehave
y
=
2
1/4
1
4
_
2
3/4
_ =
2
4
2
8
0.02813.
At thepoint (1, 1), wehavey
=
1
5
. At thispoint thetangent lineisy 1=
1
5
(x 1) or y =
1
5
x +
4
5
.
FoliumofDescartesThecurvex
3
+y
3
= 3xy (Figure7) wasrstdiscussedin1638bytheFrenchphilosopher-
mathematicianRenDescartes, whocalledit thefolium(meaningleaf). DescartessscienticcolleagueGillesde
Roberval called it thejasmineower. Both men believed incorrectly that theleaf shapein therst quadrant was
repeatedineachquadrant, givingtheappearanceof petalsof aower. Findanequationof thetangentlineatthepoint
_
2
3
,
4
3
_
.
45. Findapoint onthefoliumx
3
+y
3
= 3xy other thantheoriginat whichthetangent lineishorizontal.
solution Usingimplicit differentiation, wend
d
dx
_
x
3
+y
3
_
=
d
dx
(3xy)
3x
2
+3y
2
y
= 3(xy
+y)
Settingy
= 0inthis equationyields 3x
2
= 3y or y = x
2
. If wesubstitutethis expressioninto theoriginal equation
x
3
+y
3
= 3xy, weobtain:
x
3
+x
6
= 3x(x
2
) = 3x
3
or x
3
(x
3
2) = 0.
Onesolutionof thisequationisx = 0andtheother isx = 2
1/3
. Thus, thetwopointsonthefoliumx
3
+ y
3
= 3xy at
whichthetangent lineishorizontal are(0, 0) and(2
1/3
, 2
2/3
).
Plotx
3
+y
3
= 3xy +b forseveral valuesof b anddescribehowthegraphchangesasb 0. Thencompute
dy/dx at thepoint (b
1/3
, 0). Howdoesthisvaluechangeasb ? Doyour plotsconrmthisconclusion?
47. Findthex-coordinatesof thepointswherethetangentlineishorizontal onthetrident curve xy = x
3
5x
2
+2x 1,
sonamedbyIsaacNewtoninhistreatiseoncurvespublishedin1710(Figure8).
Hint: 2x
3
5x
2
+1= (2x 1)(x
2
2x 1).
20
20
2 8 6
4
2
x
y
FIGURE 8 Trident curve: xy = x
3
5x
2
+2x 1.
solution Takethederivativeof theequationof atrident curve:
xy = x
3
5x
2
+2x 1
toobtain
xy
+y = 3x
2
10x +2.
Settingy
= 0givesy = 3x
2
10x +2. Substitutingthisintotheequationof thetrident, wehave
xy = x(3x
2
10x +2) = x
3
5x
2
+2x 1
or
3x
3
10x
2
+2x = x
3
5x
2
+2x 1
Collectingliketermsandsettingtozero, wehave
0= 2x
3
5x
2
+1= (2x 1)(x
2
2x 1).
Hence, x =
1
2
, 1
2.
Findanequationof thetangent lineat eachof thefour pointsonthecurve(x
2
+y
2
4x)
2
= 2(x
2
+y
2
) where
x = 1. Thiscurve(Figure9) isanexampleof alimaon of Pascal, namedafter thefather of theFrenchphilosopher
BlaisePascal, whorst describedit in1650.
49. Findthederivativeat thepointswherex = 1onthefolium(x
2
+y
2
)
2
=
25
4
xy
2
. SeeFigure10.
2
2
1
x
y
FIGURE 10 Foliumcurve: (x
2
+y
2
)
2
=
25
4
xy
2
solution First, ndthepoints(1, y) onthecurve. Settingx = 1intheequation(x
2
+y
2
)
2
=
25
4
xy
2
yields
(1+y
2
)
2
=
25
4
y
2
y
4
+2y
2
+1=
25
4
y
2
4y
4
+8y
2
+4= 25y
2
4y
4
17y
2
+4= 0
(4y
2
1)(y
2
4) = 0
y
2
=
1
4
or y
2
= 4
Hencey =
1
2
or y = 2. Taking
d
dx
of bothsidesof theoriginal equationyields
2(x
2
+y
2
)(2x +2yy
) =
25
4
y
2
+
25
2
xyy
4(x
2
+y
2
)x +4(x
2
+y
2
)yy
=
25
4
y
2
+
25
2
xyy
(4(x
2
+y
2
)
25
2
x)yy
=
25
4
y
2
4(x
2
+y
2
)x
y
=
25
4
y
2
4(x
2
+y
2
)x
y(4(x
2
+y
2
)
25
2
x)
At (1, 2), x
2
+y
2
= 5, and
y
=
25
4
2
2
4(5)(1)
2(4(5)
25
2
(1))
=
1
3
.
At (1, 2), x
2
+y
2
= 5aswell, and
y
=
25
4
(2)
2
4(5)(1)
2(4(5)
25
2
(1))
=
1
3
.
At (1,
1
2
), x
2
+y
2
=
5
4
, and
y
=
25
4
_
1
2
_
2
4
_
5
4
_
(1)
1
2
_
4
_
5
4
_
25
2
(1)
_ =
11
12
.
At (1,
1
2
), x
2
+y
2
=
5
4
, and
y
=
25
4
_
1
2
_
2
4
_
5
4
_
(1)
1
2
_
4
_
5
4
_
25
2
(1)
_ =
11
12
.
Thefoliumanditstangent linesareplottedbelow:
2
1
1
2
2 1.5 1 0.5
x
y
Plot (x
2
+y
2
)
2
= 12(x
2
y
2
) +2for 4 x 4, 4 y 4usingacomputer algebrasystem. Howmany
horizontal tangent linesdoesthecurveappear tohave? Findthepointswheretheseoccur.
Exercises 5153: If the derivative dx/dy (instead of dy/dx = 0) exists at a point and dx/dy = 0, then the tangent line
at that point is vertical.
51. Calculatedx/dy fortheequationy
4
+1= y
2
+x
2
andndthepointsonthegraphwherethetangentlineisvertical.
solution Let y
4
+1= y
2
+x
2
. Differentiatingthisequationwithrespect toy yields
4y
3
= 2y +2x
dx
dy
,
so
dx
dy
=
4y
3
2y
2x
=
y(2y
2
1)
x
.
Thus,
dx
dy
= 0 when y = 0 and when y =
2
2
. Substituting y = 0 into the equation y
4
+ 1 = y
2
+ x
2
gives
1 = x
2
, sox = 1. Substitutingy =
2
2
, givesx
2
= 3/4, sox =
3
2
. Thus, therearesix pointsonthegraphof
y
4
+1= y
2
+x
2
wherethetangent lineisvertical:
(1, 0), (1, 0),
_
3
2
,
2
2
_
,
_
3
2
,
2
2
_
,
_
3
2
,
2
2
_
,
_
3
2
,
2
2
_
.
Showthat thetangent linesat x = 1
2totheconchoid withequation(x 1)
2
(x
2
+y
2
) = 2x
2
arevertical
(Figure11).
53. Useacomputer algebrasystemtoplot y
2
= x
3
4x for 4 x 4, 4 y 4. Showthat if dx/dy = 0,
theny = 0. Concludethat thetangent lineisvertical at thepointswherethecurveintersectsthex-axis. Doesyour plot
conrmthisconclusion?
solution A plot of thecurvey
2
= x
3
4x isshownbelow.
1
2
1
2
1 2 3 2 1
x
y
Differentiatingtheequationy
2
= x
3
4x withrespect toy yields
2y = 3x
2
dx
dy
4
dx
dy
,
or
dx
dy
=
2y
3x
2
4
.
Fromhere, it followsthat
dx
dy
= 0wheny = 0, sothetangent linetothiscurveisvertical at thepointswherethecurve
intersectsthex-axis. Thisconclusionisconrmedbytheplot of thecurveshownabove.
Showthat for all pointsP onthegraphinFigure12, thesegmentsOP andPR haveequal length.
In Exercises 5558, use implicit differentiation to calculate higher derivatives.
55. Consider theequationy
3
3
2
x
2
= 1.
(a) Showthat y
= x/y
2
anddifferentiateagaintoshowthat
y
=
y
2
2xyy
y
4
(b) Expressy
intermsof x andy usingpart (a).

solution
(a) Let y
3
3
2
x
2
= 1. Then3y
2
y
3x = 0, andy
= x/y
2
. Therefore,
y
=
y
2
1x 2yy
y
4
=
y
2
2xyy
y
4
.
(b) Substitutingtheexpressionfor y
intotheresult for y
gives
y
=
y
2
2xy
_
x/y
2
_
y
4
=
y
3
2x
2
y
5
.
Usethemethodof thepreviousexercisetoshowthat y
= y
3
onthecirclex
2
+y
2
= 1.
57. Calculatey
at thepoint (1, 1) onthecurvexy

2
+y 2= 0bythefollowingsteps:
(a) Findy
byimplicit differentiationandcalculatey
at thepoint (1, 1).

(b) Differentiatetheexpressionfor y
foundin(a). Thencomputey
at (1, 1) by substitutingx = 1, y = 1, andthe

valueof y
foundin(a).
solution Let xy
2
+y 2= 0.
(a) Thenx 2yy
+y
2
1+y
= 0, andy
=
y
2
2xy +1
. At (x, y) = (1, 1), wehavey
=
1
3
.
(b) Therefore,
y
=
(2xy +1)
_
2yy
_
y
2
_
2xy
+2y
_
(2xy +1)
2
=
(3)
_
2
3
_
(1)
_
2
3
+2
_
3
2
=
6+26
27
=
10
27
giventhat (x, y) = (1, 1) andy
=
1
3
.
Usethemethodof thepreviousexercisetocomputey
at thepoint (1, 1) onthecurvex

3
+y
3
= 3x +y 2.
In Exercises 5961, x and y are functions of a variable t and use implicit differentiation to relate dy/dt and dx/dt .
59. Differentiatexy = 1withrespect tot andderivetherelation
dy
dt
=
y
x
dx
dt
.
solution Let xy = 1. Thenx
dy
dt
+y
dx
dt
= 0, and
dy
dt
=
y
x
dx
dt
.
Differentiatex
3
+3xy
2
= 1withrespect tot andexpressdy/dt intermsof dx/dt , asinExercise59.
61. Calculatedy/dt intermsof dx/dt .
(a) x
3
y
3
= 1 (b) y
4
+2xy +x
2
= 0
solution
(a) Takingthederivativeof bothsidesof theequationx
3
y
3
= 1withrespect tot yields
3x
2
dx
dt
3y
2
dy
dt
= 0 or
dy
dt
=
x
2
y
2
dx
dt
.
(b) Takingthederivativeof bothsidesof theequationy
4
+2xy +x
2
= 0withrespect tot yields
4y
3
dy
dt
+2x
dy
dt
+2y
dx
dt
+2x
dx
dt
= 0,
or
dy
dt
=
x +y
2y
3
+x
dx
dt
.
ThevolumeV and pressureP of gas in apiston (which vary in timet ) satisfy PV
3/2
= C, whereC is a
constant. Provethat
dP/dt
=
3 P
63. Showthat if P liesontheintersectionof thetwocurvesx
2
y
2
= c andxy = d (c, d constants), thenthetangents
tothecurvesat P areperpendicular.
solution Let C1bethecurvedescribedbyx
2
y
2
= c, andlet C2bethecurvedescribedbyxy = d. Supposethat
P = (x
0
, y
0
) liesontheintersectionof thetwocurvesx
2
y
2
= c andxy = d. Sincex
2
y
2
= c, thechainrulegives
us2x 2yy
= 0, sothat y
=
2x
2y
=
x
y
. Theslopetothetangent linetoC1is
x
0
y
0
. OnthecurveC2, sincexy = d, the
product ruleyields that xy
+ y = 0, so that y
=
y
x
. Thereforetheslopeto thetangent lineto C2is
y
0
x
0
. Thetwo
slopesarenegativereciprocalsof oneanother, hencethetangentstothetwocurvesareperpendicular.
Thelemniscate curve (x
2
+y
2
)
2
= 4(x
2
y
2
) wasdiscoveredbyJ acobBernoulli in1694, whonotedthat it is
shapedlikeagure8, or aknot, or thebowof aribbon.Findthecoordinatesof thefour pointsatwhichthetangent
lineishorizontal (Figure13).
65. DividethecurveinFigure14
y
5
y = x
2
y +x +1
intovebranches, eachof whichisthegraphof afunction. Sketchthebranches.
2
2
2 4 4 2
x
y
FIGURE 14 Graphof y
5
y = x
2
y +x +1.
solution Thebranchesare:
Upper branch:
2 4 4 2
x
2
2
y
Lower part of lower left curve:
x
y
4 3 2 1
2
1
1
Upper part of lower left curve:
x
y
4 3 2 1
1
1
2
Upper part of lower right curve:
y
1
2
1
1 2 3 4
x
S E C T I ON 3.9 Related Rates 157
Lower part of lower right curve:
y
1
2
1
1 2 3 4
x
3.9 Related Rates
1. Assignvariablesandrestatethefollowingproblemintermsof knownandunknownderivatives(butdonotsolveit):
Howfast isthevolumeof acubeincreasingif itssideincreasesat arateof 0.5cm/s?
solution Let s andV denotethelengthof thesideandthecorrespondingvolumeof acube, respectively. Determine
dV
dt
if
ds
dt
= 0.5cm/s.
2. What istherelationbetweendV/dt anddr/dt if V =
_
4
3
_
r
3
?
solution Applyingthegeneral power rule, wend
dV
dt
= 4r
2dr
dt
. Therefore, theratiois4r
2
.
In Questions 3 and 4, water pours into a cylindrical glass of radius 4 cm. Let V and h denote the volume and water level
respectively, at time t .
3. Restatethisquestionintermsof dV/dt anddh/dt : Howfast isthewater level risingif water poursinat arateof
2cm
3
/min?
solution Determine
dh
dt
if
dV
dt
= 2cm
3
/min.
4. Restatethisquestionintermsof dV/dt anddh/dt : At what rateiswater pouringinif thewater level risesat arate
of 1cm/min?
solution Determine
dV
dt
if
dh
dt
= 1cm/min.
Exercises
In Exercises 1 and 2, consider a rectangular bathtub whose base is 18 ft
2
.
1. Howfast isthewater level risingif water isllingthetubat arateof 0.7ft
3
/min?
solution Leth betheheightof thewaterinthetubandV bethevolumeof thewater.ThenV = 18h and
dV
dt
= 18
dh
dt
.
Thus
dh
dt
=
1
18
dV
dt
=
1
18
(0.7) 0.039ft/min.
At what rateiswater pouringintothetubif thewater level risesat arateof 0.8ft/min?
3. Theradiusof acircular oil slickexpandsat arateof 2m/min.
(a) Howfast istheareaof theoil slickincreasingwhentheradiusis25m?
(b) If theradiusis0at timet = 0, howfast istheareaincreasingafter 3min?
solution Let r betheradiusof theoil slickandA itsarea.
(a) ThenA = r
2
and
dA
dt
= 2r
dr
dt
. Substitutingr = 25and
dr
dt
= 2, wend
dA
dt
= 2 (25) (2) = 100 314.16m
2
/min.
(b) Since
dr
dt
= 2andr(0) = 0, it followsthat r(t ) = 2t . Thus, r(3) = 6and
dA
dt
= 2 (6) (2) = 24 75.40m
2
/min.
At what rateisthediagonal of acubeincreasingif itsedgesareincreasingat arateof 2cm/s?
In Exercises 58, assume that the radius r of a sphere is expanding at a rate of 30cm/min. The volume of a sphere is
V =
4
3
r
3
and its surface area is 4r
2
. Determine the given rate.
5. Volumewithrespect totimewhenr = 15cm.
solution Astheradiusisexpandingat 30centimetersper minute, weknowthat
dr
dt
= 30cm/min. Taking
d
dt
of the
equationV =
4
3
r
3
yields
dV
dt
=
4
3
_
3r
2
dr
dt
_
= 4r
2
dr
dt
.
Substitutingr = 15and
dr
dt
= 30yields
dV
dt
= 4(15)
2
(30) = 27,000 cm
3
/min.
Volumewithrespect totimeat t = 2min, assumingthat r = 0at t = 0.
7. Surfaceareawithrespect totimewhenr = 40cm.
solution Takingthederivativeof bothsidesof A = 4r
2
withrespect tot yields
dA
dt
= 8r
dr
dt
.
dr
dt
= 30, so
dA
dt
= 8(40)(30) = 9600 cm
2
/min.
Surfaceareawithrespect totimeat t = 2min, assumingthat r = 10at t = 0.
In Exercises 912, refer to a 5-meter ladder sliding down a wall, as in Figures 1 and 2. The variable h is the height of the
ladders top at time t , and x is the distance from the wall to the ladders bottom.
9. Assumethebottomslidesawayfromthewall at arateof 0.8m/s. Findthevelocityof thetopof theladder at t = 2s
if thebottomis1.5mfromthewall at t = 0s.
solution Let x denotethedistancefromthebaseof theladder tothewall, andh denotetheheight of thetopof the
ladder fromtheoor. Theladder is5mlong, soh
2
+x
2
= 5
2
. At anytimet , x = 1.5+0.8t . Therefore, at timet = 2,
thebaseisx = 1.5+0.8(2) = 3.1mfromthewall. Furthermore, wehave
2h
dh
dt
+2x
dx
dt
= 0 so
dh
dt
=
x
h
dx
dt
.
Substitutingx = 3.1, h =
_
5
2
3.1
2
and
dx
dt
= 0.8, weobtain
dh
dt
=
3.1
_
5
2
3.1
2
(0.8) 0.632m/s.
Supposethat thetopisslidingdownthewall at arateof 1.2m/s. Calculatedx/dt whenh = 3m.
11. Supposethat h(0) = 4andthetopslidesdownthewall at arateof 1.2m/s. Calculatex anddx/dt at t = 2s.
solution Lethandx betheheightof theladderstopandthedistancefromthewall of theladdersbottom,respectively.
After 2seconds, h = 4+2(1.2) = 1.6m. Sinceh
2
+x
2
= 5
2
,
x =
_
5
2
1.6
2
= 4.737m.
Furthermore, wehave2h
dh
dt
+2x
dx
dt
= 0, sothat
dx
dt
=
h
x
dh
dt
. Substitutingh = 1.6, x = 4.737, and
dh
dt
= 1.2, we
nd
dx
dt
=
1.6
4.737
(1.2) 0.405m/s.
What is therelationbetweenh andx at themoment whenthetopandbottomof theladder moveat thesame
speed?
13. A conical tank hasheight 3mandradius2mat thetop. Water owsinat arateof 2m
3
/min. Howfast isthewater
level risingwhenit is2m?
solution Consider theconeof water inthetank at acertaininstant. Let r betheradius of its (inverted) base, h its
height, andV itsvolume. Bysimilar triangles,
r
h
=
2
3
or r =
2
3
h andthusV =
1
3
r
2
h =
4
27
h
3
. Therefore,
dV
dt
=
4
9
h
2
dh
dt
,
and
dh
dt
=
9
4h
2
dV
dt
.
Substitutingh = 2and
dV
dt
= 2yields
dh
dt
=
9
4 (2)
2
2=
9
8
0.36m/min.
Followthesameset-upasExercise13, but assumethat thewater level isrisingat arateof 0.3m/minwhenit is
2m. At what rateiswater owingin?
15. Theradiusr andheighth of acircular conechangeatarateof 2cm/s. Howfastisthevolumeof theconeincreasing
whenr = 10andh = 20?
solution Let r betheradius, h betheheight, andV bethevolumeof aright circular cone. ThenV =
1
3
r
2
h, and
dV
dt
=
1
3
_
r
2
dh
dt
+2hr
dr
dt
_
.
Whenr = 10, h = 20, and
dr
dt
=
dh
dt
= 2, wend
dV
dt
=

3
_
10
2
2+2 20 10 2
_
=
1000
3
1047.20cm
3
/s.
Aroadperpendicular toahighwayleadstoafarmhouselocated2kmaway(Figure8).Anautomobiletravelspast
thefarmhouseat aspeedof 80km/h. Howfast isthedistancebetweentheautomobileandthefarmhouseincreasing
whentheautomobileis6kmpast theintersectionof thehighwayandtheroad?
17. A manof height 1.8meters walks away froma5-meter lamppost at aspeedof 1.2m/s (Figure9). Findtherateat
whichhisshadowisincreasinginlength.
x y
5
FIGURE 9
solution Sincethemanismovingat arateof 1.2m/s, hisdistancefromthelight post at anygiventimeisx = 1.2t .
Knowingthemanis1.8meterstall andthat thelengthof hisshadowisdenotedby y, weset upaproportionof similar
trianglesfromthediagram:
y
1.8
=
1.2t +y
5
.
Clearingfractionsandsolvingfor y yields
y = 0.675t.
Thus, dy/dt = 0.675metersper secondistherateat whichthelengthof theshadowisincreasing.
AsClaudiawalksawayfroma264-cmlamppost, thetipof her shadowmovestwiceasfast asshedoes. What is
Claudiasheight?
19. At agivenmoment, aplanepassesdirectlyabovearadar stationat analtitudeof 6km.
(a) Theplanes speedis 800km/h. Howfast is thedistancebetweentheplaneandthestationchanginghalf aminute
later?
(b) Howfastisthedistancebetweentheplaneandthestationchangingwhentheplanepassesdirectlyabovethestation?
solution Let x bethedistanceof theplanefromthestationalongthegroundandh thedistancethroughtheair.
(a) BythePythagoreanTheorem, wehave
h
2
= x
2
+6
2
= x
2
+36.
Thus2h
dh
dt
= 2x
dx
dt
, and
dh
dt
=
x
h
dx
dt
. After half aminute, x =
1
2
1
60
800=
20
3
kilometers. Withx =
20
3
,
h =
_
_
20
3
_
2
+36=
1
3
724=
2
3
181 8.969km,
and
dx
dt
= 800,
dh
dt
=
20
3
3
2
181
800=
8000
181
594.64km/h.
(b) Whentheplaneisdirectlyabovethestation, x = 0, sothedistancebetweentheplaneandthestationisnotchanging,
for at thisinstant wehave
dh
dt
=
0
6
800= 0km/h.
Inthesettingof Exercise19, let betheanglethatthelinethroughtheradar stationandtheplanemakeswiththe
horizontal. Howfast is changing12minafter theplanepassesover theradar station?
21. A hotair balloonrisingverticallyistrackedbyanobserver located4kmfromthelift-off point. Atacertainmoment,
theanglebetweentheobserverslineof sightandthehorizontal is
5
, anditischangingatarateof 0.2rad/min. Howfast
istheballoonrisingat thismoment?
solution Lety betheheightof theballoon(inmiles) and theanglebetweentheline-of-sightandthehorizontal. Via
trigonometry, wehavetan =
y
4
. Therefore,
sec
2

d
dt
=
1
4
dy
dt
,
and
dy
dt
= 4
d
dt
sec
2
.
Using
d
dt
= 0.2and =

5
yields
dy
dt
= 4(0.2)
1
cos
2
(/5)
1.22km/min.
A laser pointer isplacedonaplatformthatrotatesatarateof 20revolutionsper minute. Thebeamhitsawall 8m
away, producingadot of light that moveshorizontallyalongthewall. Let betheanglebetweenthebeamandthe
linethroughthesearchlight perpendicular tothewall (Figure10). Howfast isthisdot movingwhen =

6
?
23. A rocket travelsverticallyat aspeedof 1200km/h. Therocket istrackedthroughatelescopebyanobserver located
16kmfromthelaunchingpad. Findtherateat whichtheanglebetweenthetelescopeandthegroundisincreasing3min
after lift-off.
solution Let y betheheight of therocket and theanglebetweenthetelescopeandtheground. Usingtrigonometry,
wehavetan =
y
16
. Therefore,
sec
2

d
dt
=
1
16
dy
dt
,
and
d
dt
=
cos
2
16
dy
dt
.
Aftertherockethastraveledfor3minutes(or
1
20
hour), itsheightis
1
20
1200= 60km.Atthisinstant, tan = 60/16=
15/4andthus
cos =
4
_
15
2
+4
2
=
4
241
.
Finally,
d
dt
=
16/241
16
(1200) =
1200
241
4.98rad/hr.
Usingatelescope, youtrack arocket that waslaunched4kmaway, recordingtheangle betweenthetelescope
andthegroundat half-secondintervals. Estimatethevelocityof therocket if (10) = 0.205and(10.5) = 0.225.
25. A policecar travelingsouthtowardSioux Falls at 160km/hpursues atruck travelingeast away fromSioux Falls,
Iowa, at 140km/h(Figure11). At timet = 0, thepolicecar is20kmnorthandthetruck is30kmeast of Sioux Falls.
Calculatetherateat whichthedistancebetweenthevehiclesischanging:
(a) At timet = 0
(b) 5minuteslater
160 km/h
140 km/h
Sioux Falls
x
y
FIGURE 11
solution Let y denotethedistancethepolicecar isnorthof SiouxFallsandx thedistancethetruck iseast of Sioux
Falls. Theny = 20160t andx = 30+140t . If denotesthedistancebetweenthepolicecar andthetruck, then
2
= x
2
+y
2
= (30+140t )
2
+(20160t )
2
and
d
dt
= 140(30+140t ) 160(20160t ) = 1000+45,200t.
(a) At t = 0, =
_
30
2
+20
2
= 10
13, so
d
dt
=
1000
10
13
=
100
13
13
27.735km/h.
(b) At t = 5minutes=
1
12
hour,
=
_
_
30+140
1
12
_
2
+
_
20160
1
12
_
2
42.197km,
and
d
dt
=
1000+45,200
1
12
42.197
112.962km/h.
A car travelsdownahighwayat 25m/s. Anobserver stands150mfromthehighway.
(a) Howfast is thedistancefromtheobserver to thecar increasingwhenthecar passes infront of theobserver?
Explainyour answer without makinganycalculations.
(b) Howfast isthedistanceincreasing20slater?
27. Inthesettingof Example5, at acertainmoment, thetractorsspeedis3m/sandthebaleisrisingat 2m/s. Howfar
isthetractor fromthebaleat thismoment?
solution FromExample5, wehavetheequation
x
dx
dt
_
x
2
+4.5
2
=
dh
dt
,
wherex denotethedistancefromthetractor tothebaleandh denotestheheight of thebale. Given
dx
dt
= 3 and
dh
dt
= 2,
it followsthat
3x
_
4.5
2
+x
2
= 2,
whichyieldsx =

16.2 4.025m.
Placidopullsaropeattachedtoawagonthroughapulleyat arateof q m/s. WithdimensionsasinFigure12:
(a) Findaformulafor thespeedof thewagonintermsof q andthevariablex inthegure.
(b) Findthespeedof thewagonwhenx = 0.6if q = 0.5m/s.
29. J ulianisjoggingaroundacircular trackof radius50m. Inacoordinatesystemwithoriginat thecenter of thetrack,
J uliansx-coordinateischangingat arateof 1.25m/swhenhiscoordinatesare(40, 30). Finddy/dt at thismoment.
solution Wehavex
2
+y
2
= 50
2
, so
2x
dx
dt
+2y
dy
dt
= 0 or
dy
dt
=
x
y
dx
dt
.
Givenx = 40, y = 30anddx/dt = 1.25, wend
dy
dt
=
40
30
(1.25) =
5
3
m/s.
A particlemovescounterclockwisearoundtheellipsewithequation9x
2
+16y
2
= 25(Figure13).
(a) Inwhichof thefour quadrantsisdx/dt > 0? Explain.
(b) Findarelationbetweendx/dt anddy/dt .
(c) Atwhatrateisthex-coordinatechangingwhentheparticlepassesthepoint(1, 1) if itsy-coordinateisincreasing
at arateof 6m/s?
(d) Finddy/dt whentheparticleisat thetopandbottomof theellipse.
In Exercises 31 and 32, assume that the pressure P (in kilopascals) and volume V (in cubic centimeters) of an expanding
gas are related by PV
b
= C, where b and C are constants (this holds in an adiabatic expansion, without heat gain or
loss).
31. FinddP/dt if b = 1.2, P = 8kPa, V = 100cm
2
, anddV/dt = 20cm
3
/min.
solution Let PV
b
= C. Then
PbV
b1
dV
dt
+V
b
dP
dt
= 0,
and
dP
dt
=
Pb
V
dV
dt
.
Substitutingb = 1.2, P = 8, V = 100, and
dV
dt
= 20, wend
dP
dt
=
(8) (1.2)
100
(20) = 1.92kPa/min.
Findb if P = 25kPa, dP/dt = 12kPa/min, V = 100cm
2
, anddV/dt = 20cm
3
/min.
33. Thebasex of therighttriangleinFigure14increasesatarateof 5cm/s, whiletheheightremainsconstantath = 20.
Howfast istheangle changingwhenx = 20?
x
q
20
FIGURE 14
solution Wehavecot =
x
20
, fromwhich
csc
2

d
dt
=
1
20
dx
dt
andthus
d
dt
=
sin
2
20
dx
dt
.
Wearegiven
dx
dt
= 5andwhenx = h = 20, =

4
. Hence,
d
dt
=
sin
2
_
4
_
20
(5) =
1
8
rad/s.
Twoparallel paths15mapart runeast-west throughthewoods. Brookejogseast ononepathat 10km/h, while
J amail walkswest ontheother pathat 6km/h. If theypasseachother at timet = 0, howfar apart arethey3slater,
andhowfast isthedistancebetweenthemchangingat that moment?
35. A particletravelsalongacurvey = f (x) asinFigure15. Let L(t ) betheparticlesdistancefromtheorigin.
(a) Showthat
dL
dt
=
_
x +f (x)f
(x)
_
x
2
+f (x)
2
_
dx
dt
if theparticleslocationat timet isP = (x, f (x)).
(b) CalculateL
(t ) whenx = 1andx = 2if f (x) =

_
3x
2
8x +9anddx/dt = 4.
x
y
y = f (x)
O
P
1 2
2
FIGURE 15
solution
(a) If theparticleslocationat timet isP = (x, f (x)), then
L(t ) =
_
x
2
+f (x)
2
.
Thus,
dL
dt
=
1
2
(x
2
+f (x)
2
)
1/2
_
2x
dx
dt
+2f (x)f
(x)
dx
dt
_
=
_
x +f (x)f
(x)
_
x
2
+f (x)
2
_
dx
dt
.
(b) Givenf (x) =
_
3x
2
8x +9, it followsthat
f
(x) =
3x 4
_
3x
2
8x +9
.
Letsstart withx = 1. Thenf (1) = 2, f
(1) =
1
2
and
dL
dt
=
_
11
_
1
2
+2
2
_
(4) = 0.
Withx = 2, f (2) =

5, f
(2) = 2/
5and
dL
dt
=
2+2
_
2
2
+
5
2
(4) =
16
3
.
Let betheangleinFigure15, whereP = (x, f (x)). Inthesettingof thepreviousexercise, showthat
d
dt
=
_
xf
(x) f (x)
x
2
+f (x)
2
_
dx
dt
Hint: Differentiatetan = f (x)/x andobservethat cos = x/
_
x
2
+f (x)
2
.
Exercises 37 and 38 refer to the baseball diamond (a square of side 90ft) in Figure 16.
20 ft/s
15 ft/s
s
90 ft
First base
Second base
Homeplate
FIGURE 16
37. Abaseball player runsfromhomeplatetowardrstbaseat20ft/s. Howfastistheplayersdistancefromsecondbase
changingwhentheplayer ishalfwaytorst base?
solution Let x bethedistanceof theplayer fromhomeplateandh theplayersdistancefromsecondbase. Usingthe
Pythagoreantheorem, wehaveh
2
= 90
2
+(90x)
2
. Therefore,
2h
dh
dt
= 2(90x)
_
dx
dt
_
,
and
dh
dt
=
90x
h
dx
dt
.
Wearegiven
dx
dt
= 20. Whentheplayer ishalfwaytorst base, x = 45andh =
_
90
2
+45
2
, so
dh
dt
=
45
_
90
2
+45
2
(20) = 4
5 8.94ft/s.
Player 1runs torst baseat aspeedof 20ft/s whilePlayer 2runs fromsecondbasetothirdbaseat aspeedof
15ft/s. Lets bethedistancebetweenthetwoplayers. Howfastiss changingwhenPlayer 1is30ftfromhomeplate
andPlayer 2is60ft fromsecondbase?
39. Theconical wateringpail inFigure17hasagridof holes. Water owsout throughtheholesat arateof kA m
3
/min,
where k is a constant and A is the surface area of the part of the cone in contact with the water. This surface area
is A = r
_
h
2
+r
2
and the volume is V =
1
3
r
2
h. Calculate the rate dh/dt at which the water level changes at
h = 0.3m, assumingthat k = 0.25m.
0.45 m
0.15 m
h
r
FIGURE 17
solution Bysimilar triangles, wehave
r
h
=
0.15
0.45
=
1
3
so r =
1
3
h.
Substitutingthisexpressionfor r intotheformulafor V yields
V =
1
3
_
1
3
h
_
2
h =
1
27
h
3
.
Fromhereandtheproblemstatement, it followsthat
dV
dt
=
1
9
h
2
dh
dt
= kA = 0.25r
_
h
2
+r
2
.
Solvingfor dh/dt gives
dh
dt
=
9
4
r
h
2
_
h
2
+r
2
.
Whenh = 0.3, r = 0.1and
dh
dt
=
9
4
0.1
0.3
2
_
0.3
2
+0.1
2
= 0.79m/min.
A bowl contains water that evaporates at arateproportional to thesurfaceareaof water exposed to theair
(Figure18). Let A(h) bethecross-sectional areaof thebowl at height h.
(a) ExplainwhyV(h +h) V(h) A(h)h if h issmall.
(b) Use(a) toarguethat
dV
dh
= A(h).
(c) Showthat thewater level h decreasesat aconstant rate.
41. A roller coaster hastheshapeof thegraphinFigure19. Showthatwhentheroller coaster passesthepoint (x, f (x)),
thevertical velocityof theroller coaster isequal tof
(x) timesitshorizontal velocity.

(x, f (x))
FIGURE 19 Graphof f (x) asaroller coaster track.
solution Let theequationy = f (x) describetheshapeof theroller coaster track. Taking
d
dt
of bothsides of this
equationyields
dy
dt
= f
(x)
dx
dt
. Inother words, thevertical velocityof acar movingalongthetrack,
dy
dt
, isequal tof
(x)
timesthehorizontal velocity,
dx
dt
.
Twotrainsleaveastationat t = 0andtravel withconstant velocityv alongstraight tracksthat makeanangle.
(a) Showthat thetrainsareseparatingfromeachother at aratev
22cos.
(b) What doesthisformulagivefor = ?
43. As thewheel of radius r cminFigure20rotates, therodof lengthL attachedat point P drives apistonback and
forthinastraight line. Let x bethedistancefromtheorigintopoint Qat theendof therod, asshowninthegure.
(a) UsethePythagoreanTheoremtoshowthat
L
2
= (x r cos)
2
+r
2
sin
2
6
(b) DifferentiateEq. (6) withrespect tot toprovethat
2(x r cos)
_
dx
dt
+r sin
d
dt
_
+2r
2
sin cos
d
dt
= 0
(c) Calculatethespeedof thepistonwhen =

2
, assumingthat r = 10cm, L = 30cm, andthewheel rotates at 4
revolutionsper minute.
Piston moves
back and forth
x
L
q
P
Q
r
FIGURE 20
solution Fromthediagram, thecoordinatesof P are(r cos, r sin) andthoseof Qare(x, 0).
(a) Thedistanceformulagives
L =
_
(x r cos)
2
+(r sin)
2
.
Thus,
L
2
= (x r cos)
2
+r
2
sin
2
.
Notethat L (thelengthof thexedrod) andr (theradiusof thewheel) areconstants.
(b) From(a) wehave
0= 2(x r cos)
_
dx
dt
+r sin
d
dt
_
+2r
2
sin cos
d
dt
.
(c) Solvingfor dx/dt in(b) gives
dx
dt
=
r
2
sin cos
d
dt
r cos x
r sin
d
dt
=
rx sin
d
dt
r cos x
.
With =

2
, r = 10, L = 30, and
d
dt
= 8,
dx
dt
=
(10) (x)
_
sin
2
_
(8)
(10) (0) x
= 80 251.33cm/min
A spectator seated300mawayfromthecenter of acircular trackof radius100mwatchesanathleterunlapsat a
speedof 5m/s. Howfast isthedistancebetweenthespectator andathletechangingwhentherunner isapproaching
thespectator andthedistancebetweenthemis 250m? Hint: Thediagramfor this problemis similar toFigure20,
withr = 100andx = 300.
In Exercises 14, refer to the function f (x) whose graph is shown in Figure 1.
y
2.0 1.5 1.0 0.5
x
7
6
5
4
3
2
1
FIGURE 1
1. Computetheaveragerateof changeof f (x) over [0, 2]. What isthegraphical interpretationof thisaveragerate?
solution Theaveragerateof changeof f (x) over [0, 2] is
f (2) f (0)
20
=
71
20
= 3.
Graphically, thisaveragerateof changerepresentstheslopeof thesecant linethroughthepoints(2, 7) and(0, 1) onthe
graphof f (x).
For which valueof h is
f (0.7+h) f (0.7)
h
equal to theslopeof thesecant linebetween thepoints where
x = 0.7andx = 1.1?
3. Estimate
f (0.7+h) f (0.7)
h
for h = 0.3. Isthisnumber larger or smaller thanf
(0.7)?
solution For h = 0.3,
f (0.7+h) f (0.7)
h
=
f (1) f (0.7)
0.3

2.82
0.3
=
8
3
.
Becausethecurveisconcaveup, theslopeof thesecant lineislarger thantheslopeof thetangent line, sothevalueof
thedifferencequotient shouldbelarger thanthevalueof thederivative.
Estimatef
(0.7) andf
(1.1).
(a) using the limit denition and nd an equation of the tangent line to the graph of f (x)
at x = a.
5. f (x) = x
2
x, a = 1
2
x anda = 1. Then
f
(a) = lim
h0
f (a +h) f (a)
h
= lim
h0
(1+h)
2
(1+h) (1
2
1)
h
= lim
h0
1+2h +h
2
1h
h
= lim
h0
(1+h) = 1
andtheequationof thetangent linetothegraphof f (x) at x = a is
y = f
(a)(x a) +f (a) = 1(x 1) +0= x 1.

f (x) = 53x, a = 2
7. f (x) = x
1
, a = 4
1
anda = 4. Then
f
(a) = lim
h0
f (a +h) f (a)
h
= lim
h0
1
4+h

1
4
h
= lim
h0
4(4+h)
4h(4+h)
= lim
h0
1
4(4+h)
=
1
4(4+0)
=
1
16
andtheequationof thetangent linetothegraphof f (x) at x = a is
y = f
(a)(x a) +f (a) =
1
16
(x 4) +
1
4
=
1
16
x +
1
2
.
f (x) = x
3
, a = 2
In Exercises 912, compute dy/dx using the limit denition.
9. y = 4x
2
solution Let y = 4x
2
. Then
dy
dx
= lim
h0
4(x +h)
2
(4x
2
)
h
= lim
h0
4x
2
2xh h
2
4+x
2
h
= lim
h0
(2x h) = 2x 0= 2x.
y =

2x +1
11. y =
1
2x
solution Let y =
1
2x
. Then
dy
dx
= lim
h0
1
2(x+h)

1
2x
h
= lim
h0
(2x) (2x h)
h(2x h)(2x)
= lim
h0
1
(2x h)(2x)
=
1
(2x)
2
.
y =
1
(x 1)
2
In Exercises 1316, express the limit as a derivative.
13. lim
h0
1+h 1
h

x. Then
lim
h0
1+h 1
h
= lim
h0
f (1+h) f (1)
h
= f
(1).
lim
x1
x
3
+1
x +1
15. lim
t
sint cost
t
solution Let f (t ) = sint cost andnotethat f () = sin cos = 0. Then
lim
t
sint cost
t
= lim
t
f (t ) f ()
t
= f
().
lim
cos sin +1

17. Findf (4) andf
(4) if thetangent linetothegraphof f (x) at x = 4hasequationy = 3x 14.

solution Theequationof thetangent lineto thegraphof f (x) at x = 4is y = f
(4)(x 4) + f (4) = f
(4)x +
(f (4) 4f
(4)). Matchingthistoy = 3x 14, weseethat f
(4) = 3andf (4) 4(3) = 14, sof (4) = 2.

Each graph in Figure2 shows thegraph of afunction f (x) and its derivativef
(x). Determinewhich is the

functionandwhichisthederivative.
19. Is(A), (B), or (C) thegraphof thederivativeof thefunctionf (x) showninFigure3?
(A) (B)
y
(C)
y
x
2 2 1 1
x
2 2 1 1
y
y = f (x)
x
2 2 1 1
y
x
2 2 1 1
FIGURE 3
solution Thegraphof f (x) hasfour horizontal tangentlineson[2, 2], sothegraphof itsderivativemusthavefour
x-interceptson[2, 2]. Thiseliminates(B). Moreover, f (x) isincreasingat bothendsof theinterval, soitsderivative
must bepositiveat bothends. Thiseliminates(A) andidenties(C) asthegraphof f
(x).
LetN(t ) bethepercentageof astatepopulationinfectedwithauvirusonweekt of anepidemic.Whatpercentage
islikelytobeinfectedinweek4if N(3) = 8andN
(3) = 1.2?
21. A girlsheight h(t ) (incentimeters) ismeasuredat timet (inyears) for 0 t 14:
52, 75.1, 87.5, 96.7, 104.5, 111.8, 118.7, 125.2,
131.5, 137.5, 143.3, 149.2, 155.3, 160.8, 164.7
(a) What istheaveragegrowthrateover the14-year period?
(b) Istheaveragegrowthratelarger over therst half or thesecondhalf of thisperiod?
(c) Estimateh
(t ) (incentimetersper year) for t = 3, 8.

solution
(a) Theaveragegrowthrateover the14-year periodis
164.752
14
= 8.05cm/year.
(b) Over therst half of the14-year period, theaveragegrowthrateis
125.252
7
10.46cm/year,
whichislarger thantheaveragegrowthrateover thesecondhalf of the14-year period:
164.7125.2
7
5.64cm/year.
(c) For t = 3,
h
(3)
h(4) h(3)
43
=
104.596.7
1
= 7.8cm/year;
for t = 8,
h
(8)
h(9) h(8)
98
=
137.5131.5
1
= 6.0cm/year.
A planets periodP (number of days to completeonerevolutionaroundthesun) is approximately 0.199A
3/2
,
whereA istheaveragedistance(inmillionsof kilometers) fromtheplanet tothesun.
(a) CalculateP anddP/dA for EarthusingthevalueA = 150.
(b) EstimatetheincreaseinP if A isincreasedto152.
In Exercises 23 and 24, use the following table of values for the number A(t ) of automobiles (in millions) manufactured
in the United States in year t .
t 1970 1971 1972 1973 1974 1975 1976
A(t ) 6.55 8.58 8.83 9.67 7.32 6.72 8.50
23. What istheinterpretationof A
(t )? EstimateA
(1971). DoesA
(1974) appear tobepositiveor negative?

solution Because A(t ) measures the number of automobiles manufactured in the United States in year t , A
(t )
measurestherateof changeinautomobileproductionintheUnitedStates. For t = 1971,
A
(1971)
A(1972) A(1971)
19721971
=
8.838.58
1
= 0.25millionautomobiles/year.
BecauseA(t ) decreasesfrom1973to1974andfrom1974to1975, it appearsthat A
(1974) wouldbenegative.
Giventhedata, whichof (A)(C) inFigure4couldbethegraphof thederivativeA
(t )? Explain.
25. y = 3x
5
7x
2
+4
solution Let y = 3x
5
7x
2
+4. Then
dy
dx
= 15x
4
14x.
y = 4x
3/2
27. y = t
7.3
solution Let y = t
7.3
. Then
dy
dt
= 7.3t
8.3
.
y = 4x
2
x
2 29. y =
x +1
x
2
+1
solution Let y =
x +1
x
2
+1
. Then
dy
dx
=
(x
2
+1)(1) (x +1)(2x)
(x
2
+1)
2
=
12x x
2
(x
2
+1)
2
.
y =
3t 2
4t 9
31. y = (x
4
9x)
6
solution Let y = (x
4
9x)
6
. Then
dy
dx
= 6(x
4
9x)
5
d
dx
(x
4
9x) = 6(4x
3
9)(x
4
9x)
5
.
y = (3t
2
+20t
3
)
6
33. y = (2+9x
2
)
3/2
solution Let y = (2+9x
2
)
3/2
. Then
dy
dx
=
3
2
(2+9x
2
)
1/2
d
dx
(2+9x
2
) = 27x(2+9x
2
)
1/2
.
y = (x +1)
3
(x +4)
4
35. y =
z
1z
solution Let y =
z
1z
. Then
dy
dz
=
1z (
z
2
)
1
1z
1z
=
1z +
z
2
(1z)
3/2
=
2z
2(1z)
3/2
.
y =
_
1+
1
x
_
3
37. y =
x
4
+
x
x
2
solution Let
y =
x
4
+
x
x
2
= x
2
+x
3/2
.
Then
dy
dx
= 2x
3
2
x
5/2
.
y =
1
(1x)
2x
39. y =
_
x +
_
x +
x
solution Let y =
_
x +
_
x +
x. Then
dy
dx
=
1
2
_
x +
_
x +
x
_
1/2
d
dx
_
x +
_
x +
x
_
=
1
2
_
x +
_
x +
x
_
1/2
_
1+
1
2
_
x +
x
_
1/2
d
dx
_
x +
x
_
_
=
1
2
_
x +
_
x +
x
_
1/2
_
1+
1
2
_
x +
x
_
1/2
_
1+
1
2
x
1/2
__
.
h(z) =
_
z +(z +1)
1/2
_
3/2
41. y = tan(t
3
)
solution Let y = tan(t
3
). Then
dy
dt
= sec
2
(t
3
)
d
dt
t
3
= 3t
4
sec
2
(t
3
).
y = 4cos(23x)
43. y = sin(2x) cos
2
x
solution Let y = sin(2x) cos
2
x = 2sinx cos
3
x. Then
dy
dx
= 6sin
2
x cos
2
x +2cos
4
x.
y = sin
_
4
_
45. y =
t
1+sect
solution Let y =
t
1+sect
. Then
dy
dt
=
1+sect t sect tant
(1+sect )
2
.
y = z csc(9z +1) 47. y =
8
1+cot
solution Let y =
8
1+cot
= 8(1+cot)
1
. Then
dy
d
= 8(1+cot)
2
d
d
(1+cot) =
8csc
2
(1+cot)
2
.
y = tan(cosx)
49. y = tan(
1+csc)
solution
dy
dx
= sec
2
(
1+csc)
d
dx
1+csc
= sec
2
(
1+csc)
1
2
(1+csc)
1/2
d
dx
(1+csc)
=
sec
2
(
1+csc) csc cot

2(
1+csc)
.
y = cos(cos(cos()))
In Exercises 5156, use the following table of values to calculate the derivative of the given function at x = 2.
x f (x) g(x) f
(x) g
(x)
2 5 4 3 9
4 3 2 2 3
51. S(x) = 3f (x) 2g(x)
solution Let S(x) = 3f (x) 2g(x). ThenS
(x) = 3f
(x) 2g
(x) and
S
(2) = 3f
(2) 2g
(2) = 3(3) 2(9) = 27.

H(x) = f (x)g(x) 53. R(x) =
f (x)
g(x)
solution Let R(x) = f (x)/g(x). Then
R
(x) =
g(x)f
(x) f (x)g
(x)
g(x)
2
and
R
(2) =
g(2)f
(2) f (2)g
(2)
g(2)
2
=
4(3) 5(9)
4
2
=
57
16
.
G(x) = f (g(x))
55. F(x) = f (g(2x))
solution Let F(x) = f (g(2x)). ThenF
(x) = 2f
(g(2x))g
(2x) and
F
(2) = 2f
(g(4))g
(4) = 2f
(2)g
(4) = 2(3)(3) = 18.

K(x) = f (x
2
)
57. Findthepointsonthegraphof x
3
y
3
= 3xy 3wherethetangent lineishorizontal.
solution Useimplicit differentiation:
3x
2
3y
2
y
= 3y +3xy
3x
2
3y = y
(3x +3y
2
)
y
=
x
2
y
x +y
2
Thetangent lineis horizontal at thosepoints onthegraphwherey = x
2
. Substitutingx
2
for y intheequationof the
graphgives
x
3
x
6
= 3x
3
3, or x
6
+2x
3
3= 0
Treatingthis as aquadratic equation in x
3
gives x
3
= 1and x
3
= 3. Thus thex-values of thepoints on thecurve
wherethetangent ishorizontal arex = 1andx =
3
3. Sinceweknowthat at thesepointsy = x
2
, thecorresponding
y-coordinates are1and(3)
2/3
= 3
2/3
. Thus thecoordinates of thepoints at whichthetangent linetothis graphare
horizontal are(1, 1) and(3
1/3
, 3
2/3
).
Findthepointsonthegraphof x
2/3
+y
2/3
= 1wherethetangent linehasslope1.
59. Finda suchthat thetangent linesy = x
3
2x
2
+x +1at x = a andx = a +1areparallel.
3
2x
2
+ x + 1. Thenf
(x) = 3x
2
4x + 1andtheslopeof thetangent lineat x = a is
f
(a) = 3a
2
4a +1, whiletheslopeof thetangent lineat x = a +1is
f
(a +1) = 3(a +1)

2
4(a +1) +1= 3(a
2
+2a +1) 4a 4+1= 3a
2
+2a.
Inorder for thetangent linesat x = a andx = a +1tohavethesameslope, wemust havef
(a) = f
(a +1), or
3a
2
4a +1= 3a
2
+2a.
Theonlysolutiontothisequationisa =
1
6
. Theequationof thetangent lineat x =
1
6
is
y = f
_
1
6
__
x
1
6
_
+f
_
1
6
_
=
5
12
_
x
1
6
_
+
241
216
=
5
12
x +
113
108
,
andtheequationof thetangent lineat x =
7
6
is
y = f
_
7
6
__
x
7
6
_
+f
_
7
6
_
=
5
12
_
x
7
6
_
+
223
216
=
5
12
x +
59
108
.
Thegraphsof f (x) andthetwotangent linesappear below.
y
x
1
1
3
2
2 1.5 1 0.5
3
2
In Exercises 6063, let f (x) = x
3
3x
2
+x +4.
Findthepointsonthegraphof f (x) wherethetangent linehasslope10.
61. For whichvaluesof x arethetangent linestoy = f (x) horizontal?
solution Thetangent linesarehorizontal whenthederivativevanishes. Fromthepreviousproblem, f
(x) = 3x
2
6x +1, whichiszerowhenx = 1
6
3
.
Findall valuesof b suchthat y = 25x +b istangent tothegraphof f (x).
63. Findall valuesof k suchthat f (x) hasonlyonetangent lineof slopek.
solution Thetangent linehasslopek whenf
(x) = 3x
2
6x + 1 = k, sothat 3x
2
6x + (1 k) = 0. Thereis
exactlyonepointwherethetangentlinehasslopek if thisequationhasonlyoneroot, whichoccurswhenitsdiscriminant
iszero. Thediscriminant of thisquadraticis36 4 3 (1 k) = 24+ 12k, sothat k = 2. Thusk = 2istheonly
suchvalueof k.
Usethetableto computetheaveragerateof changeof CandidateAs percentageof votes over theintervals
fromday20today15, day15today10, andday10today5. If thistrendcontinuesover thelast 5daysbeforethe
election, will CandidateA win?
DaysBeforeElection 20 15 10 5
CandidateA 44.8% 46.8% 48.3% 49.3%
CandidateB 55.2% 53.2% 51.7% 50.7%
In Exercises 6570, calculate y
.
65. y = 12x
3
5x
2
+3x
solution Let y = 12x
3
5x
2
+3x. Then
y
= 36x
2
10x +3 and y
= 72x 10.
y = x
2/5
67. y =

2x +3
solution Let y =

2x +3= (2x +3)
1/2
. Then
y
=
1
2
(2x +3)
1/2
d
dx
(2x +3) = (2x +3)
1/2
and y
=
1
2
(2x +3)
3/2
d
dx
(2x +3) = (2x +3)
3/2
.
y =
4x
x +1
69. y = tan(x
2
)
solution Let y = tan(x
2
). Then
y
= 2x sec
2
(x
2
) and
y
= 2x
_
2sec(x
2
)
d
dx
sec(x
2
)
_
+2sec
2
(x
2
) = 8x
2
sec
2
(x
2
) tan(x
2
) +2sec
2
(x
2
).
y = sin
2
(4x +9) In Exercises 7176, compute
dy
dx
.
71. x
3
y
3
= 4
solution Consider theequationx
3
y
3
= 4. Differentiatingwithrespect tox yields
3x
2
3y
2
dy
dx
= 0.
Therefore,
dy
dx
=
x
2
y
2
.
4x
2
9y
2
= 36
73. y = xy
2
+2x
2
solution Consider theequationy = xy
2
+2x
2
. Differentiatingwithrespect tox yields
dy
dx
= 2xy
dy
dx
+y
2
+4x.
Therefore,
dy
dx
=
y
2
+4x
12xy
.
y
x
= x +y
75. y = sin(x +y)
solution Consider theequationy = sin(x +y). Differentiatingwithrespect tox yields
dy
dx
= cos(x +y)
_
1+
dy
dx
_
.
Therefore,
dy
dx
=
cos(x +y)
1cos(x +y)
.
tan(x +y) = xy
77. InFigure5, label thegraphsf , f
, andf
.
y
x
y
x
FIGURE 5
solution First consider theplot on theleft. Observethat thegreen curveis nonnegativewhereas thered curveis
increasing, suggestingthat thegreencurveis thederivativeof theredcurve. Moreover, thegreencurveis linear with
negativeslopefor x < 0andlinear withpositiveslopefor x > 0whilethebluecurveisanegativeconstant for x < 0
andapositiveconstant for x > 0, suggestingthebluecurveisthederivativeof thegreencurve. Thus, thered, greenand
bluecurves, respectively, arethegraphsof f , f
andf
.
Nowconsider theplotontheright. Becausetheredcurveisdecreasingwhenthebluecurveisnegativeandincreasing
whenthebluecurveispositiveandthegreencurveisdecreasingwhentheredcurveisnegativeandincreasingwhenthe
redcurveispositive, it followsthat thegreen, redandbluecurves, respectively, arethegraphsof f , f
andf
.
Let f (x) = x
2
sin(x
1
) for x = 0andf (0) = 0. Showthat f
(x) exists for all x (includingx = 0) but that

f
(x) isnot continuousat x = 0(Figure6).

Exercises 7981: Let q be the number of units of a product (cell phones, barrels of oil, etc.) that can be sold at the price p.
The price elasticity of demand E is dened as the percentage rate of change of q with respect to p. In terms of derivatives,
E =
p
q
dq
dp
= lim
p0
(100q)/q
(100p)/p
79. Showthat thetotal revenueR = pq satises
dR
dp
= q(1+E).
solution Let R = pq. Then
dR
dp
= p
dq
dp
+q = q
p
q
dq
dp
+q = q(E +1).
Acommercial bakerycansell q chocolatecakesperweekatprice$p, whereq = 50p(10p) for5< p < 10.
(a) Showthat E(p) =
2p 10
p 10
.
(b) Show, bycomputingE(8), that if p = $8, thena1%increaseinpricereducesdemandbyapproximately3%.
81. Themonthly demand(inthousands) for ights betweenChicago andSt. Louis at thepricep is q = 40 0.2p.
Calculatethepriceelasticity of demandwhenp = $150andestimatethepercentageincreaseinnumber of additional
passengersif theticket priceisloweredby1%.
solution Let q = 400.2p. Thenq
(p) = 0.2and
E(p) =
_
p
q
_
dq
dp
=
0.2p
0.2p 40
.
For p = 150,
E(150) =
0.2(150)
0.2(150) 40
= 3,
soa1%decreaseinpriceincreasesdemandby 3%. Thedemandwhenp = 150isq = 40 0.2(150) = 10, or 10,000
passengers. Therefore, a1%increaseindemandtranslatesto300additional passengers.
Howfast doesthewater level riseinthetank inFigure7whenthewater level ish = 4mandwater poursinat
20m
3
/min?
83. Theminutehandof aclockis8cmlong, andthehour handis5cmlong. Howfast isthedistancebetweenthetips
of thehandschangingat 3oclock?
solution Let S bethedistancebetweenthetipsof thetwohands. Bythelawof cosines
S
2
= 8
2
+5
2
2 8 5cos(),
where istheanglebetweenthehands. Thus
2S
dS
dt
= 80sin()
d
dt
.
At threeoclock = /2, S =

89, and
d
dt
=
_

360

30
_
rad/min=
11
360
rad/min,
so
dS
dt
=
1
2
89
(80)(1)
11
360
0.407cm/min.
ChloeandBaoareinmotorboatsat thecenter of alake. At timet = 0, Chloebeginstravelingsouthat aspeedof
50km/h. Oneminutelater, Baotakesoff, headingeast at aspeedof 40km/h. At what rateisthedistancebetween
themincreasingat t = 12min?
85. A beadslides downthecurvexy = 10. Findthebeads horizontal velocity at timet = 2s if its height at timet
secondsisy = 40016t
2
cm.
solution Let xy = 10. Thenx = 10/y and
dx
dt
=
10
y
2
dy
dt
.
If y = 40016t
2
, then
dy
dt
= 32t and
dx
dt
=
10
(40016t
2
)
2
(32t ) =
320t
(40016t
2
)
2
.
Thus, at t = 2,
dx
dt
=
640
(336)
2
0.00567cm/s.
InFigure8, x isincreasingat2cm/s, y isincreasingat3cm/s, and isdecreasingsuchthattheareaof thetriangle
hastheconstant value4cm
2
.
(a) Howfast is decreasingwhenx = 4, y = 4?
(b) Howfast isthedistancebetweenP andQchangingwhenx = 4, y = 4?
87. Alightmovingat0.8m/sapproachesamanstanding4mfromawall (Figure9). Thelightis1mabovetheground.
Howfast isthetipP of themansshadowmovingwhenthelight is7mfromthewall?
1.8 m
1 m
4 m 0.8 m/s
P
FIGURE 9
solution Let x denotethedistancebetweenthemanandthelight. Usingsimilar triangles, wend
0.8
x
=
P 1
4+x
or P =
3.2
x
+1.8.
Therefore,
dP
dt
=
3.2
x
2
dx
dt
.
Whenthelight is7feet fromthewall, x = 3. With
dx
dt
= 0.8, wehave
dP
dt
=
3.2
3
2
(0.8) = 0.284m/s.
4 APPLICATIONS OF
THE DERIVATIVE
4.1 Linear Approximation and Applications
1. Trueor False? TheLinear Approximationsays that thevertical changeinthegraphis approximately equal to the
vertical changeinthetangent line.
solution Thisstatementistrue.Thelinearapproximationdoessaythatthevertical changeinthegraphisapproximately
equal tothevertical changeinthetangent line.
2. Estimateg(1.2) g(1) if g
(1) = 4.
solution UsingtheLinearApproximation,
g(1.2) g(1) g
(1)(1.21) = 4(0.2) = 0.8.

3. Estimatef (2.1) if f (2) = 1andf
(2) = 3.
solution UsingtheLinearization,
f (2.1) f (2) +f
(2)(2.12) = 1+3(0.1) = 1.3

4. Completethesentence: TheLinearApproximationshowsthatuptoasmall error, thechangeinoutputf isdirectly
proportional to.
solution TheLinearApproximationtellsusthat uptoasmall error, thechangeinoutput f isdirectlyproportional
tothechangeininput x whenx issmall.
Exercises
In Exercises 16, use Eq. (1) to estimate f = f (3.02) f (3).
1. f (x) = x
2
2
. Thenf
(x) = 2x andf f
(3)x = 6(0.02) = 0.12.

f (x) = x
4
3. f (x) = x
1
1
. Thenf
(x) = x
2
andf f
(3)x =
1
9
(0.02) = 0.00222.
f (x) =
1
x +1
5. f (x) =
x +6
x +6. Thenf
(x) =
1
2
(x +6)
1/2
and
f f
(3)x =
1
2
9
1/2
(0.02) = 0.003333.
f (x) = tan
x
3
7. Thecuberoot of 27is3. Howmuchlarger isthecuberoot of 27.2? EstimateusingtheLinearApproximation.
1/3
, a = 27, andx = 0.2. Thenf
(x) =
1
3
x
2/3
andf
(a) = f
(27) =
1
27
. TheLinear
Approximationis
f f
(a)x =
1
27
(0.2) = 0.0074074
Estimatesin
_
2
+0.1
_
sin
2
usingdifferentials.
In Exercises 912, use Eq. (1) to estimate f . Use a calculator to compute both the error and the percentage error.
9. f (x) =
1+x, a = 3, x = 0.2
solution Let f (x) = (1+ x)
1/2
, a = 3, andx = 0.2. Thenf
(x) =
1
2
(1+ x)
1/2
, f
(a) = f
(3) =
1
4
and
f f
(a)x =
1
4
(0.2) = 0.05. Theactual changeis
f = f (a +x) f (a) = f (3.2) f (3) =
4.22 0.049390.
Theerror intheLinearApproximationistherefore|0.0493900.05| = 0.000610; inpercentageterms, theerror is
0.000610
0.049390
100% 1.24%.
174
S E C T I ON 4.1 Linear Approximation and Applications 175
f (x) = 2x
2
x, a = 5, x = 0.4
11. f (x) =
1
1+x
2
, a = 3, x = 0.5
1
1+x
2
, a = 3, and x = 0.5. Then f
(x) =
2x
(1+x
2
)
2
, f
(a) = f
(3) = 0.06 and

f f
(a)x = 0.06(0.5) = 0.03. Theactual changeis

f = f (a +x) f (a) = f (3.5) f (3) 0.0245283.
Theerror intheLinear Approximationis therefore| 0.0245283 (0.03)| = 0.0054717; inpercentageterms, the
error is
0.0054717
0.0245283
100% 22.31%
f (x) = tan
_
x
4
+

4
_
, a = 0, x = 0.01
In Exercises 1316, estimate y using differentials [Eq. (3)].
13. y = cosx, a =

6
, dx = 0.014
solution Let f (x) = cosx. Thenf
(x) = sinx and

y dy = f
(a)dx = sin
_
6
_
(0.014) = 0.007.
y = tan
2
x, a =

4
, dx = 0.02
15. y =
10x
2
2+x
2
, a = 1, dx = 0.01
10x
2
2+x
2
. Then
f
(x) =
(2+x
2
)(2x) (10x
2
)(2x)
(2+x
2
)
2
=
24x
(2+x
2
)
2
and
y dy = f
(a)dx =
24
9
(0.01) = 0.026667.
y =
3
x +3
, a = 1, dx = 0.1
In Exercises 1724, estimate using the Linear Approximation and nd the error using a calculator.
17.

26
25
x, a = 25, andx = 1. Thenf
(x) =
1
2
x
1/2
andf
(a) = f
(25) =
1
10
.
TheLinearApproximationisf f
(a)x =
1
10
(1) = 0.1.
Theactual changeisf = f (a +x) f (a) = f (26) f (25) 0.0990195.
Theerror inthisestimateis|0.09901950.1| = 0.000980486.
16.5
1/4
16
1/4 19.
1
101
1
10
1
x
, a = 100, andx = 1. Thenf
(x) =
d
dx
(x
1/2
) =
1
2
x
3/2
andf
(a) =
1
2
(
1
1000
) =
0.0005.
(a)x = 0.0005(1) = 0.0005.

Theactual changeis
f = f (a +x) f (a) =
1
101
1
10
= 0.000496281.
Theerror inthisestimateis|0.0005(0.000496281)| = 3.7190210
6
.
1
98
1
10
21. 9
1/3
2
1/3
, a = 8, andx = 1. Thenf
(x) =
1
3
x
2/3
andf
(a) = f
(8) =
1
12
.
(a)x =
1
12
(1) = 0.083333.
Theactual changeisf = f (a +x) f (a) = f (9) f (8) = 0.080084.
Theerror inthisestimateis|0.0800840.083333| 3.2510
3
.
176 C HA P T E R 4 APPLICATIONS OF THE DERIVATIVE
1
(27.5)
5/3
1
243
23. sin(0.023)
solution Let f (x) = sinx, a = 0, andx = 0.023. Thenf
(x) = cosx andf
(a) = f
(0) = 1.
(a)x = 1 0.023 0.023.

Theactual changeisf = f (a +x) f (a) = f (0.023) f (0) 0.02299797.
Theerror inthisestimateis|0.0230.02299797| 210
6
.
tan
_
4
+0.01
_
1
25. Estimatef (4.03) for f (x) asinFigure8.
(4, 2)
(10, 4)
x
y = f (x)
Tangent line
y
FIGURE 8
solution UsingtheLinear Approximation, f (4.03) f (4) + f
(4)(0.03). Fromthegure, wendthat f (4) = 2

and
f
(4) =
42
104
=
1
3
.
Thus,
f (4.03) 2+
1
3
(0.03) = 2.01.
At acertainmoment, anobject inlinear motionhasvelocity 100m/s. Estimatethedistancetraveledover the
next quarter-second, andexplainhowthisisanapplicationof theLinearApproximation.
27. Whichislarger:
2.1
2or
9.1
9? ExplainusingtheLinearApproximation.
x, andx = 0.1. Thenf
(x) =
1
2
x
1/2
andtheLinearApproximationat x = a gives
f =
a +0.1
a f
(a)(0.1) =
1
2
a
1/2
(0.1) =
0.05
a
Weseethat f decreasesasa increases. Inparticular
2.1
2
0.05
2
islarger than

9.1
9
0.05
3
Estimatesin61
sin60
usingtheLinearApproximation. Hint: Express inradians.

29. Boxofcerevenueat amultiplexcinemainParisisR(p) = 3600p 10p
3
eurosper showingwhentheticket price
isp euros. CalculateR(p) for p = 9andusetheLinear ApproximationtoestimateR if p israisedor loweredby0.5
euros.
solution Let R(p) = 3600p 10p
3
. ThenR(9) = 3600(9) 10(9)
3
= 25,110euros. Moreover, R
(p) = 3600
30p
2
, sobytheLinearApproximation,
R R
(9)p = 1170p.
If p israisedby0.5euros, thenR 585euros; ontheother hand, if p isloweredby0.5euros, thenR 585euros.
Thestopping distance for anautomobileisF(s) = 1.1s +0.054s
2
ft, wheres isthespeedinmph. UsetheLinear
Approximationtoestimatethechangeinstoppingdistanceper additional mphwhens = 35andwhens = 55.
31. Athinsilver wirehaslengthL = 18cmwhenthetemperatureisT = 30
C. EstimateL whenT decreasesto25
C
if thecoefcient of thermal expansionisk = 1.910
5
C
1
(seeExample3).
solution Wehave
dL
dT
= kL = (1.910
5
)(18) = 3.4210
4
cm/
C
ThechangeintemperatureisT = 5
C, sobytheLinearApproximation, thechangeinlengthisapproximately
L 3.4210
4
T = (3.4210
4
)(5) = 0.00171cm
At T = 25
C, thelengthof thewireisapproximately17.99829cm.
Atacertainmoment, thetemperatureinasnakecagesatisesdT/dt = 0.008
C/s. Estimatetheriseintemperature
over thenext 10seconds.
33. Theatmospheric pressureP at altitudeh = 20 kmis P = 5.5 kilopascals. EstimateP at altitudeh = 20.5 km
assumingthat
dP
dh
= 0.87.
solution Wehave
P P
(h)h = 0.87 0.5 0.435

sothat
P(20.5) P(20) +P = 5.50.435 5.065kilopascals.
TheresistanceR of acopper wireat temperatureT = 20
C isR = 15. Estimatetheresistanceat T = 22
C,
assumingthat dR/dT
T =20
= 0.06/
C.
35. Newtons Lawof Gravitationshows that if apersonweighs w pounds onthesurfaceof theearth, thenhis or her
weight at distancex fromthecenter of theearthis
W(x) =
wR
2
x
2
(for x R)
whereR = 3960milesistheradiusof theearth(Figure9).
(a) Showthat theweight lost at altitudeh milesabovetheearthssurfaceisapproximatelyW (0.0005w)h. Hint:
UsetheLinearApproximationwithdx = h.
(b) Estimatetheweight lost bya200-lbfootball player yinginajet at analtitudeof 7miles.
3
9
6
0
h
FIGURE 9 Thedistancetothecenter of theearthis3960+h miles.
solution
(a) UsingtheLinearApproximation
W W
(R)x =
2wR
2
R
3
h =
2wh
R
0.0005wh.
(b) Substitutew = 200andh = 7intotheresult frompart (a) toobtain
W 0.0005(200)(7) = 0.7pounds.
UsingExercise35(a), estimatethealtitudeat whicha130-lbpilot wouldweigh129.5lb.
37. A stonetossedverticallyintotheair withinitial velocityv cm/sreachesamaximumheight of h = v
2
/1960cm.
(a) Estimateh if v = 700cm/sandv = 1cm/s.
(b) Estimateh if v = 1000cm/sandv = 1cm/s.
(c) Ingeneral, doesa1cm/sincreaseinv leadtoagreater changeinh at lowor highinitial velocities? Explain.
solution A stonetossed vertically with initial velocity v cm/s attains amaximumheight of h(v) = v
2
/1960cm.
Thus, h
(v) = v/980.
(a) If v = 700andv = 1, thenh h
(v)v =
1
980
(700)(1) 0.71cm.
(b) If v = 1000andv = 1, thenh h
(v)v =
1
980
(1000)(1) = 1.02cm.
(c) Aonecentimeter per secondincreaseininitial velocityv increasesthemaximumheightbyapproximatelyv/980cm.
Accordingly, thereisabigger effect at higher velocities.
Thesides of asquarecarpet ismeasuredat 6m. Estimatethemaximumerror intheareaA of thecarpet if s is
accuratetowithin2centimeters.
In Exercises 39 and 40, use the following fact derived from Newtons Laws: An object released at an angle with initial
velocity v ft/stravels a horizontal distance
s =
1
32
v
2
sin2 ft (Figure10)
q
x
y
FIGURE 10 Trajectoryof anobject releasedat anangle.
39. A player located18.1ft fromthebasket launchesasuccessful jumpshot fromaheight of 10ft (level withtherimof
thebasket), at anangle = 34
andinitial velocityv = 25ft/s.)

(a) Showthat s 0.255 ft for asmall changeof .
(b) Isit likelythat theshot wouldhavebeensuccessful if theanglehadbeenoff by2
?
solution Using Newtons laws and the given initial velocity of v = 25ft/s, the shot travels s =
1
32
v
2
sin2t =
625
32
sin2t ft, wheret isinradians.
(a) If = 34
(i.e., t =
17
90
), then
s s
(t )t =
625
16
cos
_
17
45
_
t =
625
16
cos
_
17
45
_

180
0.255.
(b) If = 2
, thisgivess 0.51ft, inwhichcasetheshot wouldnot havebeensuccessful, havingbeenoff half a

foot.
Estimates if = 34
, v = 25ft/s, andv = 2.
41. Theradiusof aspherical ball ismeasuredat r = 25cm. Estimatethemaximumerror inthevolumeandsurfacearea
if r isaccuratetowithin0.5cm.
solution Thevolumeandsurfaceareaof thespherearegivenbyV =
4
3
r
3
andS = 4r
2
, respectively. If r = 25
andr = 0.5, then
V V
(25)r = 4(25)
2
(0.5) 3927cm
3
,
and
S S
(25)r = 8(25)(0.5) 314.2cm

2
.
ThedosageD of diphenhydraminefor adogof body mass w kgis D = 4.7w
2/3
mg. Estimatethemaximum
allowableerror inw for acocker spaniel of massw = 10kgif thepercentageerror inD must belessthan3%.
43. Thevolume(in liters) and pressureP (in atmospheres) of acertain gas satisfy PV = 24. A measurement yields
V = 4withapossibleerror of 0.3L. ComputeP andestimatethemaximumerror inthiscomputation.
solution GivenPV = 24andV = 4, itfollowsthatP = 6atmospheres. SolvingPV = 24forP yieldsP = 24V
1
.
Thus, P
= 24V
2
and
P P
(4)V = 24(4)
2
(0.3) = 0.45atmospheres.
Inthenotationof Exercise43, assumethat ameasurement yieldsV = 4. Estimatethemaximumallowableerror
inV if P must haveanerror of lessthan0.2atm.
In Exercises 4554, nd the linearization at x = a.
45. f (x) = x
4
, a = 1
4
. Thenf
(x) = 4x
3
. Thelinearizationat a = 1is
L(x) = f
(a)(x a) +f (a) = 4(x 1) +1= 4x 3.

f (x) =
1
x
, a = 2
47. f () = sin
2
, a =

4
solution Let f () = sin
2
. Thenf
() = 2sin cos = sin2. Thelinearizationat a =

4
is
L() = f
(a)( a) +f (a) = 1
_

4
_
+
1
2
=

4
+
1
2
.
g(x) =
x
2
x 3
, a = 4
49. y = (1+x)
1/2
, a = 0
solution Let f (x) = (1+x)
1/2
. Thenf
(x) =
1
2
(1+x)
3/2
. Thelinearizationat a = 0is
L(x) = f
(a)(x a) +f (a) =
1
2
x +1.
y = (1+x)
1/2
, a = 3
51. y = (1+x
2
)
1/2
, a = 0
solution Let f (x) = (1+ x
2
)
1/2
. Thenf
(x) = x(1+ x
2
)
3/2
, f (a) = 1andf
(a) = 0, sothelinearization
at a is
L(x) = f
(a)(x a) +f (a) = 1.
y =
1x
1+x
, a = 4
53. y =
sinx
x
, a =

2
sinx
x
. Then
f
(x) =
x cosx sinx
x
2
, f (a) =
2
, f
(a) =
4
2
sothelinearizationof f (x) at a is
L(x) = f
(a)(x a) +f (a) =
4
2
_
x

2
_
+
2
=
4
2
x +
4

y =
sinx
x
, a =

4
55. What isf (2) if thelinearizationof f (x) at a = 2isL(x) = 2x +4?
solution f (2) = L(2) = 2(2) +4= 8.
Computethelinearizationof f (x) = 3x 4at a = 0anda = 2. Provemoregenerally that alinear function
coincideswithitslinearizationat x = a for all a.
57. Estimate
16.2usingthelinearizationL(x) of f (x) =

x at a = 16. Plot f (x) andL(x) onthesameset of axes
anddeterminewhether theestimateistoolargeor toosmall.
1/2
, a = 16, andx = 0.2.Thenf
(x) =
1
2
x
1/2
andf
(a) = f
(16) =
1
8
.Thelinearization
tof (x) is
L(x) = f
(a)(x a) +f (a) =
1
8
(x 16) +4=
1
8
x +2.
Thus, wehave
16.2 L(16.2) = 4.025. Graphsof f (x) andL(x) areshownbelow. Becausethegraphof L(x) lies
abovethegraphof f (x), weexpect that theestimatefromtheLinearApproximationistoolarge.
y
x
1
2
3
4
0
5
5 10 15 25 20
f (x)
L(x)
Estimate1/
15usingasuitablelinearizationof f (x) = 1/
x. Plot f (x) andL(x) onthesameset of axes

anddeterminewhether theestimateistoolargeor toosmall. Useacalculator tocomputethepercentageerror.
In Exercises 5967, approximate using linearization and use a calculator to compute the percentage error.
59.
1
17
1/2
, a = 16, and x = 1. Then f
(x) =
1
2
x
3/2
, f
(a) = f
(16) =
1
128
and the
linearizationtof (x) is
L(x) = f
(a)(x a) +f (a) =
1
128
(x 16) +
1
4
=
1
128
x +
3
8
.
Thus, wehave
1
17
L(17) 0.24219. Thepercentageerror inthisestimateis
17
0.24219
1
17
100% 0.14%
1
101
61.
1
(10.03)
2
2
, a = 10 and x = 0.03. Then f
(x) = 2x
3
, f
(a) = f
(10) = 0.002 and the

linearizationtof (x) is
L(x) = f
(a)(x a) +f (a) = 0.002(x 10) +0.01= 0.002x +0.03.

Thus, wehave
1
(10.03)
2
L(10.03) = 0.002(10.03) +0.03= 0.00994.
Thepercentageerror inthisestimateis
1
(10.03)
2
0.00994
1
(10.03)
2
100% 0.0027%
(17)
1/4
63. (64.1)
1/3
1/3
,a = 64,andx = 0.1.Thenf
(x) =
1
3
x
2/3
,f
(a) = f
(64) =
1
48
andthelinearization
tof (x) is
L(x) = f
(a)(x a) +f (a) =
1
48
(x 64) +4=
1
48
x +
8
3
.
Thus, wehave(64.1)
1/3
L(64.1) 4.002083. Thepercentageerror inthisestimateis
(64.1)
1/3
4.002083
(64.1)
1/3
100% 0.000019%
(1.2)
5/3
65. tan(0.04)
solution Let f (x) = tanx and a = 0. Then f
(x) = sec
2
x, f (a) = tan0 = 0, and f
(a) = sec
2
0 = 1. The
linearizationof f (x) isthen
L(x) = f
(a)(x a) +f (a) = 1(x 0) +0= x

Thus, wehavetan(0.04) 0.04. Thepercentageerror inthisestimateis
tan(0.04) 0.04
tan(0.04)
100% 0.053%.
cos
_
3.1
4
_
67.
(3.1)/2
sin(3.1/2)
x
sinx
anda =

2
. Then
f
(x) =
sinx x cosx
sin
2
x
, f (a) =

2
, f
(a) = 1
Thelinearizationof f (x) isthen
L(x) = f
(a)(x a) +f (a) = x

2
+

2
= x
Thus, wehave
(3.1/2)
sin(3.1/2)
L
_
3.1
2
_
= 1.55. Thepercentageerror inthisestimateis
(3.1/2)
sin(3.1/2)
1.55
(3.1/2)
sin(3.1/2)
100% 0.022%
ComputethelinearizationL(x) of f (x) = x
2
x
3/2
at a = 4. Thenplot f (x) L(x) andndaninterval I
arounda = 4suchthat |f (x) L(x)| 0.1for x I.
69. Showthat theLinear Approximationtof (x) =

x at x = 9yieldstheestimate
9+h 3
1
6
h. Set K = 0.01
and showthat |f
(x)| K for x 9. Then verify numerically that theerror E satises Eq. (5) for h = 10
n
, for
1 n 4.
solution Letf (x) =
x. Thenf (9) = 3, f
(x) =
1
2
x
1/2
andf
(9) =
1
6
. Therefore, bytheLinearApproximation,
f (9+h) f (9) =
9+h 3
1
6
h.
Moreover, f
(x) =
1
4
x
3/2
, so|f
(x)| =
1
4
x
3/2
. Becausethisisadecreasingfunction, it followsthat for x 9,
K = max|f
(x)| |f
(9)| =
1
108
< 0.01.
Fromthefollowingtable, weseethat for h = 10
n
, 1 n 4, E
1
2
Kh
2
.
h E = |
9+h 3
1
6
h|
1
2
Kh
2
10
1
4.60410
5
5.0010
5
10
2
4.62710
7
5.0010
7
10
3
4.62910
9
5.0010
9
10
4
4.62710
11
5.0010
11
TheLinear Approximationtof (x) = tanx at x =

4
yieldstheestimatetan
_
4
+h
_
1 2h. Set K = 6.2
andshow, usingaplot, that |f
(x)| K for x [
4
,

4
+ 0.1]. Thenverify numerically that theerror E satises
Eq. (5) for h = 10
n
, for 1 n 4.
71. Compute dy/dx at the point P = (2, 1) on the curve y
3
+ 3xy = 7 and show that the linearization at P is
L(x) =
1
3
x +
5
3
. UseL(x) toestimatethey-coordinateof thepoint onthecurvewherex = 2.1.
solution Differentiatingbothsidesof theequationy
3
+3xy = 7withrespect tox yields
3y
2
dy
dx
+3x
dy
dx
+3y = 0,
so
dy
dx
=
y
y
2
+x
.
S E C T I ON 4.2 Extreme Values 181
Thus,
dy
dx
(2,1)
=
1
1
2
+2
=
1
3
,
andthelinearizationat P = (2, 1) is
L(x) = 1
1
3
(x 2) =
1
3
x +
5
3
.
Finally, whenx = 2.1, weestimatethat they-coordinateof thepoint onthecurveis
y L(2.1) =
1
3
(2.1) +
5
3
= 0.967.
Apply themethodof Exercise71toP = (0.5, 1) ony
5
+ y 2x = 1toestimatethey-coordinateof thepoint
onthecurvewherex = 0.55.
73. Apply themethodof Exercise71toP = (1, 2) ony
4
+ 7xy = 2toestimatethesolutionof y
4
7.7y = 2near
y = 2.
solution Differentiatingbothsidesof theequationy
4
+7xy = 2withrespect tox yields
4y
3
dy
dx
+7x
dy
dx
+7y = 0,
so
dy
dx
=
7y
4y
3
+7x
.
Thus,
dy
dx
(1,2)
=
7(2)
4(2)
3
+7(1)
=
14
25
,
andthelinearizationat P = (1, 2) is
L(x) = 2
14
25
(x +1) =
14
25
x +
36
25
.
Finally, theequationy
4
7.7y = 2correspondstox = 1.1, soweestimatethesolutionof thisequationnear y = 2is
y L(1.1) =
14
25
(1.1) +
36
25
= 2.056.
Showthat for anyreal number k, (1+x)
k
1+kx for small x. Estimate(1.02)
0.7
and(1.02)
0.3
.
75. Let f = f (5+h) f (5), wheref (x) = x
2
. Verifydirectlythat E = |f f
(5)h| satises(5) withK = 2.

2
. Then
f = f (5+h) f (5) = (5+h)
2
5
2
= h
2
+10h
and
E = |f f
(5)h| = |h
2
+10h 10h| = h
2
=
1
2
(2)h
2
=
1
2
Kh
2
.
Let f = f (1+h) f (1) wheref (x) = x
1
. Showdirectlythat E = |f f
(1)h| isequal toh

2
/(1+h).
Thenprovethat E 2h
2
if
1
2
h
1
2
. Hint: Inthiscase,
1
2
1+h
3
2
.
4.2 Extreme Values
1. What isthedenitionof acritical point?
solution Acritical pointisavalueof theindependentvariablex inthedomainof afunctionf atwhicheitherf
(x) = 0
or f
(x) doesnot exist.

In Questions 2 and 3, choose the correct conclusion.
2. If f (x) isnot continuouson[0, 1], then
(a) f (x) hasnoextremevalueson[0, 1].
(b) f (x) might not haveanyextremevalueson[0, 1].
solution Thecorrect responseis (b): f (x) might not haveany extremevalues on[0, 1]. Although[0, 1] is closed,
becausef isnot continuous, thefunctionisnot guaranteedtohaveanyextremevalueson[0, 1].
3. If f (x) iscontinuousbut hasnocritical pointsin[0, 1], then
(a) f (x) hasnominor maxon[0, 1].
(b) Either f (0) or f (1) istheminimumvalueon[0, 1].
solution Thecorrect responseis (b): either f (0) or f (1) is theminimumvalueon[0, 1]. Remember that extreme
valuesoccur either at critical pointsor endpoints. If acontinuousfunctiononaclosedinterval hasnocritical points, the
extremevaluesmust occur at theendpoints.
4. Fermats Theoremdoes not claimthat if f
(c) = 0, thenf (c) is alocal extremevalue(this is false). What does

FermatsTheoremassert?
solution FermatsTheoremclaims: If f (c) isalocal extremevalue, theneither f
(c) = 0or f
(c) doesnot exist.

Exercises
1. Thefollowingquestionsrefer toFigure15.
(a) Howmanycritical pointsdoesf (x) haveon[0, 8]?
(b) What isthemaximumvalueof f (x) on[0, 8]?
(c) What arethelocal maximumvaluesof f (x)?
(d) Findaclosedinterval onwhichboththeminimumandmaximumvaluesof f (x) occur at critical points.
(e) Findaninterval onwhichtheminimumvalueoccursat anendpoint.
8 3 4 5 6 7 2 1
2
3
4
5
6
1
f (x)
x
y
FIGURE 15
solution
(a) f (x) hasthreecritical pointsontheinterval [0, 8]: at x = 3, x = 5andx = 7. Twoof these, x = 3andx = 5, are
wherethederivativeiszeroandone, x = 7, iswherethederivativedoesnot exist.
(b) Themaximumvalueof f (x) on[0, 8] is6; thefunctiontakesthisvalueat x = 0.
(c) f (x) achievesalocal maximumof 5at x = 5.
(d) Answersmayvary. Oneexampleistheinterval [4, 8]. Another is[2, 6].
(e) Answersmayvary. Theeasiestwaytoensurethisistochooseaninterval onwhichthegraphtakesnolocal minimum.
Oneexampleis[0, 2].
Statewhether f (x) = x
1
(Figure16) hasaminimumor maximumvalueonthefollowingintervals:
(a) (0, 2) (b) (1, 2) (c) [1, 2]
In Exercises 320, nd all critical points of the function.
3. f (x) = x
2
2x +4
2
2x +4. Thenf
(x) = 2x 2= 0impliesthat x = 1isthelonecritical point of f .

f (x) = 7x 2
5. f (x) = x
3
9
2
x
2
54x +2
3
9
2
x
2
54x + 2. Then f
(x) = 3x
2
9x 54 = 3(x + 3)(x 6) = 0implies that
x = 3andx = 6arethecritical pointsof f .
f (t ) = 8t
3
t
2
7. f (x) = x
1
x
2
1
x
2
. Then
f
(x) = x
2
+2x
3
=
2x
x
3
= 0
impliesthat x = 2istheonly critical point of f . Thoughf
(x) doesnot exist at x = 0, thisisnot acritical point of f

becausex = 0isnot inthedomainof f .
g(z) =
1
z 1
1
z
9. f (x) =
x
x
2
+1
x
x
2
+1
. Thenf
(x) =
1x
2
(x
2
+1)
2
= 0impliesthat x = 1arethecritical pointsof f .
f (x) =
x
2
x
2
4x +8
11. f (t ) = t 4
t +1
solution Let f (t ) = t 4
t +1. Then
f
(t ) = 1
2
t +1
= 0
impliesthat t = 3isacritical point of f . Becausef
(t ) doesnot exist at t = 1, thisisanother critical point of f .

f (t ) = 4t
_
t
2
+1
13. f (x) = x
2
_
1x
2
2
_
1x
2
. Then
f
(x) =
x
3
_
1x
2
+2x
_
1x
2
=
2x 3x
3
_
1x
2
.
This derivativeis 0whenx = 0andwhenx =
2/3; thederivativedoes not exist whenx = 1. All veof these

valuesarecritical pointsof f
f (x) = x +|2x +1|
15. g() = sin
2
solution Let g() = sin

2
. Theng
() = 2sin cos = sin2 = 0impliesthat

=
n
2
isacritical valueof g for all integer valuesof n.
R() = cos +sin
2
17. Let f (x) = x

2
4x +1.
(a) Findthecritical point c of f (x) andcomputef (c).
(b) Computethevalueof f (x) at theendpointsof theinterval [0, 4].
(c) Determinetheminandmaxof f (x) on[0, 4].
(d) Findtheextremevaluesof f (x) on[0, 1].
2
4x +1.
(a) Thenf
(c) = 2c 4= 0impliesthat c = 2isthesolecritical point of f . Wehavef (2) = 3.

(b) f (0) = f (4) = 1.
(c) Usingtheresultsfrom(a) and(b), wendthemaximumvalueof f on[0, 4] is1andtheminimumvalueis3.
(d) Wehavef (1) = 2. Hencethemaximumvalueof f on[0, 1] is1andtheminimumvalueis2.
Findtheextremevaluesof f (x) = 2x
3
9x
2
+12x on[0, 3] and[0, 2].
19. Findthecritical pointsof f (x) = sinx +cosx anddeterminetheextremevalueson
_
0,

2
_
.
solution
Let f (x) = sinx + cosx. Thenontheinterval
_
0,

2
_
, wehavef
(x) = cosx sinx = 0at x =

4
, theonly
critical point of f inthisinterval.
Sincef (
4
) =

2andf (0) = f (
2
) = 1, themaximumvalueof f on
_
0,

2
_
is
2, whiletheminimumvalue
is1.
Computethecritical pointsof h(t ) = (t
2
1)
1/3
. Checkthatyour answer isconsistentwithFigure17. Thennd
theextremevaluesof h(t ) on[0, 1] and[0, 2].
21. Plot f (x) = 2
x x on [0, 4] and determinethemaximumvaluegraphically. Then verify your answer

usingcalculus.
solution Thegraphof y = 2
x x over theinterval [0, 4] isshownbelow. Fromthegraph, weseethat at x = 1,

thefunctionachievesitsmaximumvalueof 1.
y
x
0.2
0.4
0.6
0.8
0
1
1 2 3 4
To verify theinformationobtainedfromtheplot, let f (x) = 2
x x. Thenf
(x) = x
1/2
1. Solvingf
(x) = 0
yieldsthecritical pointsx = 0andx = 1. Becausef (0) = f (4) = 0andf (1) = 1, weseethat themaximumvalueof
f on[0, 4] is1.
Plot f (x) = 2x
3
9x
2
+ 12x on[0, 3] andlocatetheextremevalues graphically. Thenverify your answer
usingcalculus.
23. Approximatethecritical pointsof g(x) = x cosx onI = [0, 2] andestimatetheminimumvalueof g(x)
onI.
solution g
(x) = cosx x sinx, so g
(x) = 0whenx 0.860334andwhenx 3.425618. Evaluatingg at the

endpointsaswell asat thesecritical pointsgivesg(0) = 0, g(0.860334) 0.561096, g(3.425618) 3.288371, and
g(2) = 2 6.28. Hencetheminimumvalueof g(x) onI is 3.288371at x 3.425618.
Approximatethecritical pointsof g(x) = tan
2
x 5x onI =
_
2
,
2
_
andestimatetheminimumvalueof
g(x) onI.
In Exercises 2558, nd the min and max of the function on the given interval by comparing values at the critical points
and endpoints.
25. y = 2x
2
+4x +5, [2, 2]
2
+ 4x + 5. Thenf
(x) = 4x + 4= 0impliesthat x = 1istheonly critical point of f .

Theminimumof f ontheinterval [2, 2] isf (1) = 3, whereasitsmaximumisf (2) = 21. (Note: f (2) = 5.)
y = 2x
2
+4x +5, [0, 2]
27. y = 6t t
2
, [0, 5]
solution Letf (t ) = 6t t
2
. Thenf
(t ) = 62t = 0impliesthatt = 3istheonlycritical pointof f . Theminimum

of f ontheinterval [0, 5] isf (0) = 0, whereasthemaximumisf (3) = 9. (Note: f (5) = 5.)
y = 6t t
2
, [4, 6]
29. y = x
3
6x
2
+8, [1, 6]
3
6x
2
+ 8. Thenf
(x) = 3x
2
12x = 3x(x 4) = 0impliesthat x = 0andx = 4are
thecritical points of f . Theminimumof f ontheinterval [1, 6] is f (4) = 24, whereas themaximumis f (6) = 8.
(Note: f (1) = 3andthecritical point x = 0isnot intheinterval [1, 6].)
y = x
3
+x
2
x, [2, 2]
31. y = 2t
3
+3t
2
, [1, 2]
solution Let f (t ) = 2t
3
+ 3t
2
. Thenf
(t ) = 6t
2
+ 6t = 6t (t + 1) = 0implies that t = 0andt = 1arethe
critical pointsof f . Theminimumof f ontheinterval [1, 2] isf (1) = 5, whereasthemaximumisf (2) = 28. (Note:
Neither critical pointsareintheinterval [1, 2].)
y = x
3
12x
2
+21x, [0, 2]
33. y = z
5
80z, [3, 3]
solution Let f (z) = z
5
80z. Then f
(z) = 5z
4
80 = 5(z
4
16) = 5(z
2
+ 4)(z + 2)(z 2) = 0 implies
that z = 2arethecritical pointsof f . Theminimumvalueof f ontheinterval [3, 3] isf (2) = 128, whereasthe
maximumisf (2) = 128. (Note: f (3) = 3andf (3) = 3.)
y = 2x
5
+5x
2
, [2, 2] 35. y =
x
2
+1
x 4
, [5, 6]
x
2
+1
x 4
. Then
f
(x) =
(x 4) 2x (x
2
+1) 1
(x 4)
2
=
x
2
8x 1
(x 4)
2
= 0
impliesx = 4
17arecritical pointsof f . x = 4isnot acritical point becausex = 4isnot inthedomainof f . On

theinterval [5, 6], theminimumof f isf (6) =
37
2
= 18.5, whereasthemaximumof f isf (5) = 26. (Note: Thecritical
pointsx = 4
17arenot intheinterval [5, 6].)

y =
1x
x
2
+3x
, [1, 4]
37. y = x
4x
x +1
, [0, 3]
4x
x +1
. Then
f
(x) = 1
4
(x +1)
2
=
(x 1)(x +3)
(x +1)
2
= 0
impliesthatx = 1andx = 3arecritical pointsof f . x = 1isnotacritical pointbecausex = 1isnotinthedomain
of f . Theminimumof f ontheinterval [0, 3] is f (1) = 1, whereas themaximumis f (0) = f (3) = 0. (Note: The
critical point x = 3isnot intheinterval [0, 3].)
y = 2
_
x
2
+1x, [0, 2]
39. y = (2+x)
_
2+(2x)
2
, [0, 2]
solution Let f (x) = (2+x)
_
2+(2x)
2
. Then
f
(x) =
_
2+(2x)
2
(2+x)(2+(2x)
2
)
1/2
(2x) =
2(x 1)
2
_
2+(2x)
2
= 0
impliesthat x = 1isthecritical point of f . Ontheinterval [0, 2], theminimumisf (0) = 2
6 4.9andthemaximum
isf (2) = 4
2 5.66. (Note: f (1) = 3
3 5.2.)
y =
_
1+x
2
2x, [0, 1]
41. y =
_
x +x
2
2
x, [0, 4]
_
x +x
2
2
x. Then
f
(x) =
1
2
(x +x
2
)
1/2
(1+2x) x
1/2
=
1+2x 2
1+x
2
1+x
= 0
implies that x = 0andx =
3
2
arethecritical points of f . Neither x = 1nor x =
3
2
is acritical point because
neither isinthedomainof f . Ontheinterval [0, 4], theminimumof f isf
_
3
2
_
0.589980andthemaximumis
f (4) 0.472136. (Note: f (0) = 0.)
y = (t t
2
)
1/3
, [1, 2]
43. y = sinx cosx,
_
0,

2
_
solution Letf (x) = sinx cosx =
1
2
sin2x. Ontheinterval
_
0,

2
_
, f
(x) = cos2x = 0whenx =

4
. Theminimum
of f onthisinterval isf (0) = f (
2
) = 0, whereasthemaximumisf (
4
) =
1
2
.
y = x +sinx, [0, 2]
45. y =
2 sec,
_
0,

3
_
solution Let f () =

2 sec. Ontheinterval [0,

3
], f
() =

2 sec tan = 0at =

4
. Theminimum
valueof f on this interval is f (0) = 1, whereas themaximumvalueover this interval is f (
4
) =

2(
4
1)
0.303493. (Note: f (
3
) =
3
2 0.519039.)
y = cos +sin, [0, 2]
47. y = 2sin, [0, 2]
solution Let g() = 2sin. On theinterval [0, 2], g
() = 1 2cos = 0 at =

3
and =
5
3
. The
minimumof g onthisinterval isg(
3
) =

3

3 0.685andthemaximumisg(
5
3
) =
5
3
+
3 6.968. (Note:
g(0) = 0andg(2) = 2 6.283.)
y = 4sin
3
3cos
2
, [0, 2]
49. y = tanx 2x, [0, 1]
solution Let f (x) = tanx 2x. Thenontheinterval [0, 1], f
(x) = sec
2
x 2= 0at x =

4
. Theminimumof f
isf (
4
) = 1

2
0.570796andthemaximumisf (0) = 0. (Note: f (1) = tan12 0.442592.)
y = sec
2
x 2tanx,
_
6
,
3
_ 51. Let f () = 2sin2 +sin4.
(a) Showthat isacritical point if cos4 = cos2.
(b) Show, usingaunit circle, that cos
1
= cos
2
if andonlyif
1
=
2
+2k for aninteger k.
(c) Showthat cos4 = cos2 if andonlyif =

2
+k or =

6
+
_
3
_
k.
(d) Findthesixcritical pointsof f () on[0, 2] andndtheextremevaluesof f () onthisinterval.
(e) Checkyour resultsagainst agraphof f ().
solution f () = 2sin2 + sin4 isdifferentiableat all , sotheway tondthecritical pointsistondall points
suchthat f
() = 0.
(a) f
() = 4cos2 +4cos4. If f
() = 0, then4cos4 = 4cos2, socos4 = cos2.

(b) Giventhepoint(cos, sin) atangle ontheunitcircle, therearetwopointswithx coordinatecos. Thegraphic
showsthesetwopoints, whichare:
Thepoint (cos( +), sin( +)) ontheoppositeendof theunit circle.
Thepoint (cos( ), sin( )) obtainedbyreectingthroughthey axis.
If we include all angles representing these points on the circle, we nd that cos
1
= cos
2
if and only if
1
=
( +
2
) +2k or
1
= (
2
) +2k for integersk.
(c) Using(b), werecognizethat cos4 = cos2 if 4 = 2 + + 2k or 4 = 2 + 2k. Solvingfor , we
obtain =

2
+k or =

6
+

3
k.
(d) Tondall , 0 < 2 indicatedby(c), weusethefollowingtable:
k 0 1 2 3 4 5
2
+k

2
3
2
6
+

3
k

6

2
5
6
7
6
3
2
11
6
Thecritical pointsintherange[0, 2] are
6
,

2
,
5
6
,
7
6
,
3
2
, and
11
6
. Onthisinterval, themaximumvalueisf (
6
) =
f (
7
6
) =
3
3
2
andtheminimumvalueisf (
5
6
) = f (
11
6
) =
3
3
2
.
(e) Thegraphof f () = 2sin2 +sin4 isshownhere:
x
1
2 3
4
5
6
1
2
1
2
y
Wecanseethat therearesixat pointsonthegraphbetween0and2, aspredicted. Thereare4local extrema, andtwo
pointsat (
2
, 0) and(
3
2
, 0) wherethegraphhasneither alocal maximumnor alocal minimum.
Findthecritical pointsof f (x) = 2cos3x +3cos2x in[0, 2]. Checkyour answer against agraphof f (x).
In Exercises 5356, nd the critical points and the extreme values on [0, 4]. In Exercises 55 and 56, refer to Figure 18.
y = |x
2
+ 4x 12|
2 6
10
20
30
y = |cos x|
1
2
3
2
2
x x
y y
FIGURE 18
53. y = |x 2|
solution Let f (x) = |x 2|. For x < 2, wehavef
(x) = 1. For x > 2, wehavef
(x) = 1. Nowasx 2, we
have
f (x) f (2)
x 2
=
(2x) 0
x 2
1; whereasasx 2+, wehave
f (x) f (2)
x 2
=
(x 2) 0
x 2
1. Therefore,
f
(2) = lim
x2
f (x) f (2)
x 2
doesnot exist andthelonecritical point of f isx = 2. Alternately, weexaminethegraphof
f (x) = |x 2| shownbelow.
Tondtheextremum, wecheck thevaluesof f (x) at thecritical point andtheendpoints. f (0) = 2, f (4) = 2, and
f (2) = 0. f (x) takesitsminimumvalueof 0at x = 2, anditsmaximumof 2at x = 0andat x = 4.
y
x
0.5
1
1.5
2
0 3 4 2 1
y = |3x 9|
55. y = |x
2
+4x 12|
solution Let f (x) = |x
2
+4x 12| = |(x +6)(x 2)|. Fromthegraphof f inFigure18, weseethat f
(x) does
not exist at x = 6andat x = 2, so thesearecritical points of f . Thereis also acritical point betweenx = 6and
x = 2at whichf
(x) = 0. For 6 < x < 2, f (x) = x

2
4x + 12, so f
(x) = 2x 4 = 0whenx = 2. On
theinterval [0, 4] theminimumvalueof f isf (2) = 0andthemaximumvalueisf (4) = 20. (Note: f (0) = 12andthe
critical pointsx = 6andx = 2arenot intheinterval.)
y = | cosx|
In Exercises 5760, verify Rolles Theorem for the given interval.
57. f (x) = x +x
1
,
_
1
2
, 2
_
solution Becausef iscontinuouson[
1
2
, 2], differentiableon(
1
2
, 2) and
f
_
1
2
_
=
1
2
+
1
1
2
=
5
2
= 2+
1
2
= f (2),
wemayconcludefromRollesTheoremthatthereexistsac (
1
2
, 2) atwhichf
(c) = 0.Here,f
(x) = 1x
2
=
x
2
1
x
2
,
sowemaytakec = 1.
f (x) = sinx,
_
4
,
3
4
_
59. f (x) =
x
2
8x 15
, [3, 5]
solution Becausef is continuous on[3, 5], differentiableon(3, 5) andf (3) = f (5) = 1, wemay concludefrom
RollesTheoremthat thereexistsac (3, 5) at whichf
(c) = 0. Here,
f
(x) =
(8x 15)(2x) 8x
2
(8x 15)
2
=
2x(4x 15)
(8x 15)
2
,
sowemaytakec =
15
4
.
f (x) = sin
2
x cos
2
x,
_
4
,
3
4
_ 61. Provethat f (x) = x
5
+2x
3
+4x 12haspreciselyonereal root.
solution Letsrst establishthef (x) = x
5
+ 2x
3
+ 4x 12hasat least oneroot. Becausef isapolynomial, it is
continuousfor all x. Moreover, f (0) = 12 < 0andf (2) = 44 > 0. Therefore, by theIntermediateValueTheorem,
thereexistsac (0, 2) suchthat f (c) = 0.
Next, weprovethatthisistheonlyroot. Wewill useproof bycontradiction. Supposef (x) = x
5
+2x
3
+4x 12has
tworeal roots, x = a andx = b. Thenf (a) = f (b) = 0andRollesTheoremguaranteesthat thereexistsac (a, b) at
whichf
(c) = 0. However, f
(x) = 5x
4
+6x
2
+4 4for all x, sothereisnoc (a, b) atwhichf
(c) = 0. Basedon
thiscontradiction, weconcludethat f (x) = x
5
+ 2x
3
+ 4x 12cannot havemorethanonereal root. Finally, f must
havepreciselyonereal root.
Provethat f (x) = x
3
+3x
2
+6x haspreciselyonereal root.
63. Provethat f (x) = x
4
+5x
3
+4x hasnoroot c satisfyingc > 0. Hint: Notethat x = 0isaroot andapplyRolles
Theorem.
solution We will proceed by contradiction. Note that f (0) = 0 and suppose that there exists a c > 0 such that
f (c) = 0. Thenf (0) = f (c) = 0andRollesTheoremguarantees that thereexists ad (0, c) suchthat f
(d) = 0.
However, f
(x) = 4x
3
+ 15x
2
+ 4 > 4for all x > 0, so thereis no d (0, c) suchthat f
(d) = 0. Basedonthis
contradiction, weconcludethat f (x) = x
4
+5x
3
+4x hasnoroot c satisfyingc > 0.
Provethat c = 4isthelargest root of f (x) = x
4
8x
2
128.
65. Thepositionof amass oscillatingat theendof aspringis s(t ) = Asint , whereA is theamplitudeand is the
angular frequency. Showthat thespeed|v(t )| isat amaximumwhentheaccelerationa(t ) iszeroandthat |a(t )| isat a
maximumwhenv(t ) iszero.
solution Let s(t ) = Asint . Then
v(t ) =
ds
dt
= Acost
and
a(t ) =
dv
dt
= A
2
sint.
Thus, thespeed
|v(t )| = |Acost |
isamaximumwhen| cost | = 1, whichispreciselywhensint = 0; thatis, thespeed|v(t )| isatamaximumwhenthe
accelerationa(t ) iszero. Similarly,
|a(t )| = |A
2
sint |
isamaximumwhen| sint | = 1, whichispreciselywhencost = 0; that is, |a(t )| isat amaximumwhenv(t ) iszero.
TheconcentrationC(t ) (inmg/cm
3
) of adruginapatientsbloodstreamafter t hoursis
C(t ) =
0.016t
t
2
+4t +4
Findthemaximumconcentrationinthetimeinterval [0, 8] andthetimeat whichit occurs.
67. In1919, physicistAlfredBetzarguedthat themaximumefciencyof awindturbineisaround59%. If windentersa
turbinewithspeedv
1
andexitswithspeedv
2
, thenthepower extractedisthedifferenceinkineticenergyper unit time:
P =
1
2
mv
2
1
1
2
mv
2
2
watts
wheremisthemassof windowingthroughtherotor per unittime(Figure19). Betzassumedthatm = A(v
1
+v
2
)/2,
where isthedensity of air andA istheareaswept out by therotor. Windowingundisturbedthroughthesamearea
A would have mass per unit time Av
1
and power P
0
=
1
2
Av
3
1
. The fraction of power extracted by the turbine is
F = P/P
0
.
(a) Showthat F dependsonlyontheratior = v
2
/v
1
andisequal toF(r) =
1
2
(1r
2
)(1+r), where0 r 1.
(b) Showthat themaximumvalueof F(r), calledtheBetzLimit, is16/27 0.59.
(c) Explainwhy Betzs formulafor F(r) is not meaningful for r closeto zero. Hint: Howmuchwindwould
passthroughtheturbineif v
2
werezero? Isthisrealistic?
1
0.1
0.2
0.3
0.5
0.4
0.6
0.5
r
F
(A) Wind flowing through aturbine. (B) F is thefraction of energy
extracted by theturbineas a
function of r = v
2
/v
1.
v
1
v
2
FIGURE 19
solution
(a) Wenotethat
F =
P
P
0
=
1
2
A(v
1
+v
2
)
2
(v
2
1
v
2
2
)
1
2
Av
3
1
=
1
2
v
2
1
v
2
2
v
2
1

v
1
+v
2
v
1
=
1
2
_
1
v
2
2
v
2
1
_
_
1+
v
2
v
1
_
=
1
2
(1r
2
)(1+r).
(b) Basedonpart (a),
F
(r) =
1
2
(1r
2
) r(1+r) =
3
2
r
2
r +
1
2
.
Therootsof thisquadraticarer = 1andr =
1
3
. Now, F(0) =
1
2
, F(1) = 0and
F
_
1
3
_
=
1
2
8
9
4
3
=
16
27
0.59.
Thus, theBetz Limit is16/27 0.59.
(c) If v
2
were0, thennoair wouldbepassingthroughtheturbine, whichisnot realistic.
TheBohr radiusa
0
of thehydrogenatomisthevalueof r that minimizestheenergy
E(r) =

h
2
2mr
2
e
2
4
0
r
where
h, m, e, and
0
arephysical constants. Showthat a
0
= 4
0
h
2
/(me
2
). Assumethat theminimumoccursat a
critical point, assuggestedbyFigure20.
69. Theresponseof acircuitorotheroscillatorysystemtoaninputof frequency (omega) isdescribedbythefunction
() =
1
_
(
2
0
2
)
2
+4D
2
2
Both
0
(thenatural frequencyof thesystem) andD(thedampingfactor) arepositiveconstants. Thegraphof iscalleda
resonancecurve, andthepositivefrequency
r
> 0, where takesitsmaximumvalue, if itexists, iscalledtheresonant
frequency. Showthat
r
=
_
2
0
2D
2
if 0< D <
0
/
2andthatnoresonantfrequencyexistsotherwise(Figure21).
w
(A) D = 0.01 (B) D = 0.2
2 2 w
r
(C) D = 0.75 (no resonance)
50
f f f
w
2 w
r
1
2
3
w
3 1 2
1
0.5
FIGURE 21 Resonancecurveswith
0
= 1.
solution Let () = ((
2
0
2
)
2
+4D
2
2
)
1/2
. Then
() =
2((
2
0
2
) 2D
2
)
((
2
0
2
)
2
+4D
2
2
)
3/2
andthenon-negativecritical points are = 0and =
_
2
0
2D
2
. Thelatter critical point is positiveif andonly if
2
0
2D
2
> 0, andsincewearegivenD > 0, thisisequivalent to0< D <
0
/
2.
Dene
r
=
_
2
0
2D
2
. Now, (0) = 1/
2
0
and() 0as . Finally,
(
r
) =
1
2D
_
2
0
D
2
,
which, for 0< D <
0
/
2, islarger than1/
2
0
. Hence, thepoint =
_
2
0
2D
2
, if dened, isalocal maximum.
Beesbuildhoneycombstructuresoutof cellswithahexagonal baseandthreerhombus-shapedfacesontop, asin
Figure22. Wecanshowthat thesurfaceareaof thiscell is
A() = 6hs +
3
2
s
2
(
3 csc cot)
withh, s, and asindicatedinthegure. Remarkably, beesknow whichangle minimizesthesurfacearea(and
thereforerequirestheleast amount of wax).
(a) Showthat 54.7
(assumeh ands areconstant). Hint: Findthecritical point of A() for 0< < /2.
(b) Conrm, bygraphingf () =
3csc cot, that thecritical point indeedminimizesthesurfacearea.

71. Findthemaximumof y = x
a
x
b
on[0, 1] where0< a < b. Inparticular, ndthemaximumof y = x
5
x
10
on
[0, 1].
solution
Let f (x) = x
a
x
b
. Then f
(x) = ax
a1
bx
b1
. Sincea < b, f
(x) = x
a1
(a bx
ba
) = 0 implies
critical pointsx = 0andx = (
a
b
)
1/(ba)
, whichisintheinterval [0, 1] asa x
b
on theinterval [0, 1], which gives
f (x) > 0andthusthemaximumvalueof f on[0, 1] is
f
_
_
a
b
_
1/(ba)
_
=
_
a
b
_
a/(ba)
_
a
b
_
b/(ba)
.
Let f (x) = x
5
x
10
. Thenbypart (a), themaximumvalueof f on[0, 1] is
f
_
_
1
2
_
1/5
_
=
_
1
2
_
_
1
2
_
2
=
1
2
1
4
=
1
4
.
In Exercises 7274, plot the function using a graphing utility and nd its critical points and extreme values on [5, 5].
y =
1
1+|x 1|
73. y =
1
1+|x 1|
+
1
1+|x 4|
solution Let
f (x) =
1
1+|x 1|
+
1
1+|x 4|
.
Theplot follows:
5 4 3 2 1 1 2 3 4 5
0.2
0.4
0.6
0.8
1
1.2
Wecan seeon theplot that thecritical points of f (x) lieat thecusps at x = 1and x = 4and at thelocation of the
horizontal tangent lineat x =
5
2
. Withf (5) =
17
70
, f (1) = f (4) =
5
4
, f (
5
2
) =
4
5
andf (5) =
7
10
, it followsthat the
maximumvalueof f (x) on[5, 5] isf (1) = f (4) =
5
4
andtheminimumvalueisf (5) =
17
70
.
y =
x
|x
2
1| +|x
2
4|
75. (a) Useimplicit differentiationtondthecritical pointsonthecurve27x
2
= (x
2
+y
2
)
3
.
(b) Plot thecurveandthehorizontal tangent linesonthesameset of axes.
solution
(a) Differentiatingbothsidesof theequation27x
2
= (x
2
+y
2
)
3
withrespect tox yields
54x = 3(x
2
+y
2
)
2
_
2x +2y
dy
dx
_
.
Solvingfor dy/dx weobtain
dy
dx
=
27x 3x(x
2
+y
2
)
2
3y(x
2
+y
2
)
2
=
x(9(x
2
+y
2
)
2
)
y(x
2
+y
2
)
2
.
Thus, thederivativeiszerowhenx
2
+y
2
= 3. Substitutingintotheequationforthecurve, thisyieldsx
2
= 1, orx = 1.
Therearethereforefour pointsat whichthederivativeiszero:
(1,
2), (1,
2), (1,
2), (1,
2).
Therearealsocritical pointswherethederivativedoesnot exist. Thisoccurswheny = 0andgivesthefollowingpoints
withvertical tangents:
(0, 0), (
4
27, 0).
(b) Thecurve27x
2
= (x
2
+y
2
)
3
anditshorizontal tangentsareplottedbelow.
1
1
y
x
2 1 1 2
Sketchthegraphof acontinuousfunctionon(0, 4) withaminimumvaluebut nomaximumvalue.
77. Sketchthegraphof acontinuousfunctionon(0, 4) havingalocal minimumbut noabsoluteminimum.
solution Hereisthegraphof afunctionf on(0, 4) withalocal minimumvalue[betweenx = 2andx = 4] but no
absoluteminimum[sincef (x) asx 0+].
x
1 2 3
10
10
y
Sketchthegraphof afunctionon[0, 4] having
(a) Twolocal maximaandonelocal minimum.
(b) Anabsoluteminimumthat occursat anendpoint, andanabsolutemaximumthat occursat acritical point.
79. Sketchthegraphof afunctionf (x) on[0, 4] withadiscontinuity suchthat f (x) hasanabsoluteminimumbut no
absolutemaximum.
solution Hereis thegraphof afunctionf on[0, 4] that (a) has adiscontinuity [at x = 4] and(b) has anabsolute
minimum[at x = 0] but noabsolutemaximum[sincef (x) asx 4].
y
x
0
1
2
3
4
1 2 3 4
Arainbowisproducedbylightraysthatenteraraindrop(assumedspherical)andexitafterbeingreectedinternally
asinFigure23. Theanglebetweentheincomingandreectedraysis = 4r 2i, wheretheangleof incidencei
andrefractionr arerelatedbySnellsLawsini = n sinr withn 1.33(theindexof refractionfor air andwater).
(a) UseSnellsLawtoshowthat
dr
di
=
cosi
n cosr
.
(b) Showthat themaximumvalue
max
of occurswheni satisescosi =
_
n
2
1
3
. Hint: Showthat
d
di
= 0if
cosi =
n
2
cosr. ThenuseSnellsLawtoeliminater.
(c) Showthat
max
59.58
.
81. Showthat theextremevaluesof f (x) = a sinx +b cosx are
_
a
2
+b
2
.
solution If f (x) = a sinx +b cosx, thenf
(x) = a cosx b sinx, sothatf
(x) = 0impliesa cosx b sinx = 0.

Thisimpliestanx =
a
b
. Then,
sinx =
a
_
a
2
+b
2
and cosx =
b
_
a
2
+b
2
.
Therefore
f (x) = a sinx +b cosx = a
a
_
a
2
+b
2
+b
b
_
a
2
+b
2
=
a
2
+b
2
_
a
2
+b
2
=
_
a
2
+b
2
.
Show, byconsideringitsminimum, that f (x) = x
2
2x +3takesononlypositivevalues. Moregenerally, nd
theconditionsonr ands under whichthequadraticfunctionf (x) = x
2
+rx +s takesononlypositivevalues. Give
examplesof r ands for whichf takesonbothpositiveandnegativevalues.
83. Show that if thequadratic polynomial f (x) = x
2
+ rx + s takes on both positiveand negativevalues, then its
minimumvalueoccursat themidpoint betweenthetworoots.
2
+ rx + s and suppose that f (x) takes on both positive and negative values. This will
guaranteethat f hastworeal roots. Bythequadraticformula, therootsof f are
x =
r
_
r
2
4s
2
.
S E C T I ON 4.3 The Mean Value Theorem and Monotonicity 191
Observethat themidpoint betweentheserootsis
1
2
_
r +
_
r
2
4s
2
+
r
_
r
2
4s
2
_
=
r
2
.
Next, f
(x) = 2x +r = 0whenx =
r
2
and, becausethegraphof f (x) isanupwardopeningparabola, it followsthat
f (
r
2
) isaminimum. Thus, f takesonitsminimumvalueat themidpoint betweenthetworoots.
GeneralizeExercise83: Showthat if thehorizontal liney = c intersectsthegraphof f (x) = x
2
+rx +s at two
points(x
1
, f (x
1
)) and(x
2
, f (x
2
)), thenf (x) takesitsminimumvalueat themidpoint M =
x
1
+x
2
2
(Figure24).
85. A cubic polynomial may have a local min and max, or it may have neither (Figure 25). Find conditions on the
coefcientsa andb of
f (x) =
1
3
x
3
+
1
2
ax
2
+bx +c
that ensurethat f hasneither alocal minnor alocal max. Hint: ApplyExercise82tof
(x).
4 2 4 2
(A) (B)
2 4 2
20
20
60
30
x x
y y
FIGURE 25 Cubicpolynomials
1
3
x
3
+
1
2
ax
2
+bx +c. UsingExercise82, wehaveg(x) = f
(x) = x
2
+ax +b > 0for all
x providedb >
1
4
a
2
, inwhichcasef hasnocritical pointsandhencenolocal extrema. (Actuallyb
1
4
a
2
will sufce,
sinceinthiscase[aswell seeinalater section] f hasaninectionpoint but nolocal extrema.)
Findtheminandmaxof
f (x) = x
p
(1x)
q
on[0, 1],
wherep, q > 0.
87. Provethat if f iscontinuousandf (a) andf (b) arelocal minimawherea < b, thenthereexistsavaluec
betweena andb suchthat f (c) isalocal maximum. (Hint: ApplyTheorem1totheinterval [a, b].) Showthat continuity
isanecessaryhypothesisbysketchingthegraphof afunction(necessarilydiscontinuous) withtwolocal minimabut no
local maximum.
solution
Let f (x) beacontinuous functionwithf (a) andf (b) local minimaontheinterval [a, b]. By Theorem1, f (x)
musttakeonbothaminimumandamaximumon[a, b]. Sincelocal minimaoccur atf (a) andf (b), themaximum
must occur at someother point intheinterval, call it c, wheref (c) isalocal maximum.
Thefunctiongraphedhereisdiscontinuousat x = 0.
x
2 4 6 8 8 6 4 2
4
6
8
y
4.3 The Mean Value Theorem and Monotonicity
1. For whichvalueof m isthefollowingstatement correct?If f (2) = 3andf (4) = 9, andf (x) isdifferentiable, then
f hasatangent lineof slopem.
solution TheMeanValueTheoremguaranteesthat thefunctionhasatangent linewithslopeequal to
f (4) f (2)
42
=
93
42
= 3.
Hence, m = 3makesthestatement correct.
2. Assumef isdifferentiable. Whichof thefollowingstatementsdoesnot followfromtheMVT?
(a) If f hasasecant lineof slope0, thenf hasatangent lineof slope0.
(b) If f (5) < f (9), thenf
(c) > 0for somec (5, 9).

(c) If f hasatangent lineof slope0, thenf hasasecant lineof slope0.
(d) If f
(x) > 0for all x, theneverysecant linehaspositiveslope.

solution Conclusion(c)doesnot followfromtheMeanValueTheorem. Asacounterexample, consider thefunction
f (x) = x
3
. Notethat f
(0) = 0, but nosecant linehaszeroslope.

3. Canafunctionthat takesononlynegativevalueshaveapositivederivative? If so, sketchanexample.
solution Yes. Thegurebelowdisplaysafunctionthat takesononlynegativevaluesbut hasapositivederivative.
x
y
4. For f (x) withderivativeasinFigure12:
(a) Isf (c) alocal minimumor maximum?
(b) Isf (x) adecreasingfunction?
c
x
y
FIGURE 12 Graphof derivativef
(x).
solution
(a) Totheleft of x = c, thederivativeispositive, sof isincreasing; totheright of x = c, thederivativeisnegative, sof
isdecreasing. Consequently, f (c) must bealocal maximum.
(b) No. Thederivativeisadecreasingfunction, but asnotedinpart (a), f (x) isincreasingfor x < c anddecreasingfor
x > c.
Exercises
In Exercises 18, nd a point c satisfying the conclusion of the MVT for the given function and interval.
1. y = x
1
, [2, 8]
1
, a = 2, b = 8. Thenf
(x) = x
2
, andbytheMVT, thereexistsac (2, 8) suchthat
1
c
2
= f
(c) =
f (b) f (a)
b a
=
1
8
1
2
82
=
1
16
.
Thusc
2
= 16andc = 4. Choosec = 4 (2, 8).
y =
x, [9, 25]
3. y = cosx sinx, [0, 2]
solution Let f (x) = cosx sinx, a = 0, b = 2. Thenf
(x) = sinx cosx, andby theMVT, thereexistsa

c (0, 2) suchthat
sinc cosc = f
(c) =
f (b) f (a)
b a
=
11
2 0
= 0.
Thussinc = cosc. Chooseeither c =
3
4
or c =
7
4
(0, 2).
y =
x
x +2
, [1, 4]
5. y = x
3
, [4, 5]
3
, a = 4, b = 5. Thenf
(x) = 3x
2
, andbytheMVT, thereexistsac (4, 5) suchthat
3c
2
= f
(c) =
f (b) f (a)
b a
=
189
9
= 21.
Solvingfor c yieldsc
2
= 7, soc =
7. Bothof thesevaluesareintheinterval [4, 5], soeither valuecanbechosen.

y = (x 1)(x 3), [1, 3]
7. y = x sinx,
_
2
,
2
_
solution Let f (x) = x sinx, a =
2
, b =

2
. Then f
(x) = sinx + x cosx, and by theMVT, thereexists a

c
_
2
,
2
_
suchthat
sinc +c cosc = f
(c) =
f (b) f (a)
b a
=
2

2
(1)
= 0
Solvingfor c givesc = 0astheonlysolution.
y = x sin(x), [1, 1]
9. Let f (x) = x
5
+ x
2
. Thesecant linebetweenx = 0andx = 1has slope2(check this), so by theMVT,
f
(c) = 2for somec (0, 1). Plot f (x) andthesecant lineonthesameaxes. Thenplot y = 2x +b for different values
of b until thelinebecomestangent tothegraphof f . Zoominonthepoint of tangencytoestimatex-coordinatec of the
point of tangency.
5
+x
2
. Theslopeof thesecant linebetweenx = 0andx = 1is
f (1) f (0)
10
=
20
1
= 2.
A plot of f (x), thesecant linebetweenx = 0andx = 1, andtheliney = 2x 0.764isshownbelowat theleft. The
liney = 2x 0.764appearstobetangent tothegraphof y = f (x). Zoominginonthepoint of tangency(seebelowat
theright), it appearsthat thex-coordinateof thepoint of tangencyisapproximately0.62.
y = x
5
+ x
2
y = 2x .764
x
1
2
4
y
x
y
0.3
0.6
0.5
0.4
0.56 0.6 0.64 0.52
Plotthederivativeof f (x) = 3x
5
5x
3
. Describeitssignchangesandusethistodeterminethelocal extreme
valuesof f (x). Thengraphf (x) toconrmyour conclusions.
11. Determinetheintervalsonwhichf
(x) ispositiveandnegative, assumingthat Figure13isthegraphof f (x).

x
6 5 4 3 2 1
y
FIGURE 13
solution Thederivativeof f ispositiveontheintervals(0, 1) and(3, 5) wheref isincreasing; it isnegativeonthe
intervals(1, 3) and(5, 6) wheref isdecreasing.
Determinetheintervalsonwhichf (x) isincreasingor decreasing, assumingthatFigure13isthegraphof f
(x).
13. Statewhether f (2) andf (4) arelocal minimaor local maxima, assumingthat Figure13isthegraphof f
(x).
solution
f
(x) makesatransitionfrompositivetonegativeat x = 2, sof (2) isalocal maximum.

f
(x) makesatransitionfromnegativetopositiveat x = 4, sof (4) isalocal minimum.

Figure14showsthegraphof thederivativef
(x) of afunctionf (x). Findthecritical pointsof f (x) anddetermine

whether theyarelocal minima, local maxima, or neither.
In Exercises 1518, sketch the graph of a function f (x) whose derivative f
(x) has the given description.

15. f
(x) > 0for x > 3andf
(x) < 0for x < 3

solution Hereisthegraphof afunctionf for whichf
(x) > 0for x > 3andf
(x) < 0for x < 3.

y
x
0
2
4
6
8
10
1 2 3 4 5
f
(x) > 0for x < 1andf
(x) < 0for x > 1

17. f
(x) isnegativeon(1, 3) andpositiveeverywhereelse.

solution Hereisthegraphof afunctionf for whichf
(x) isnegativeon(1, 3) andpositiveelsewhere.

x
1 2 3 4
2
4
6
8
2
y
f
(x) makesthesigntransitions+, , +, .
In Exercises 1922, nd all critical points of f and use the First Derivative Test to determine whether they are local
minima or maxima.
19. f (x) = 4+6x x
2
solution Let f (x) = 4+6x x
2
. Thenf
(x) = 62x = 0impliesthat x = 3istheonlycritical point of f . Asx

increasesthrough3, f
(x) makesthesigntransition+, . Therefore, f (3) = 13isalocal maximum.

f (x) = x
3
12x 4 21. f (x) =
x
2
x +1
x
2
x +1
. Then
f
(x) =
x(x +2)
(x +1)
2
= 0
impliesthatx = 0andx = 2arecritical points. Notethatx = 1isnotacritical pointbecauseitisnotinthedomainof
f . Asx increasesthrough2, f
(x) makesthesigntransition+, sof (2) = 4isalocal maximum. Asx increases

through0, f
(x) makesthesigntransition, + sof (0) = 0isalocal minimum.

f (x) = x
3
+x
3
In Exercises 2344, nd the critical points and the intervals on which the function is increasing or decreasing. Use the
First Derivative Test to determine whether the critical point is a local min or max (or neither).
solution Here is a table legend for Exercises 2344.
SYMBOL MEANING
The entity is negative on the given interval.
0 The entity is zero at the specied point.
+ The entity is positive on the given interval.
U The entity is undened at the specied point.
f is increasing on the given interval.
f is decreasing on the given interval.
M f has a local maximum at the specied point.
m f has a local minimum at the specied point.
There is no local extremum here.
23. y = x
2
+7x 17
2
+7x 17. Thenf
(x) = 72x = 0yieldsthecritical point c =

7
2
.
x
_
,
7
2
_
7/2
_
7
2
,
_
f
+ 0
f M
y = 5x
2
+6x 4
25. y = x
3
12x
2
3
12x
2
. Thenf
(x) = 3x
2
24x = 3x(x 8) = 0yieldscritical pointsc = 0, 8.
x (, 0) 0 (0, 8) 8 (8, )
f
+ 0 0 +
f M m
y = x(x 2)
3
27. y = 3x
4
+8x
3
6x
2
24x
4
+8x
3
6x
2
24x. Then
f
(x) = 12x
3
+24x
2
12x 24
= 12x
2
(x +2) 12(x +2) = 12(x +2)(x
2
1)
= 12(x 1) (x +1) (x +2) = 0
yieldscritical pointsc = 2, 1, 1.
x (, 2) 2 (2, 1) 1 (1, 1) 1 (1, )
f
0 + 0 0 +
f m M m
y = x
2
+(10x)
2
29. y =
1
3
x
3
+
3
2
x
2
+2x +4
1
3
x
3
+
3
2
x
2
+2x +4. Thenf
(x) = x
2
+3x +2= (x +1) (x +2) = 0yieldscritical points
c = 2, 1.
x (, 2) 2 (2, 1) 1 (1, )
f
+ 0 0 +
f M m
y = x
4
+x
3
31. y = x
5
+x
3
+1
5
+x
3
+1. Thenf
(x) = 5x
4
+3x
2
= x
2
(5x
2
+3) yieldsasinglecritical point: c = 0.
x (, 0) 0 (0, )
f
+ 0 +
f
y = x
5
+x
3
+x
33. y = x
4
4x
3/2
(x > 0)
4
4x
3/2
for x > 0. Thenf
(x) = 4x
3
6x
1/2
= 2x
1/2
(2x
5/2
3) = 0, whichgivesus
thecritical point c = (
3
2
)
2/5
. (Note: c = 0isnot intheinterval under consideration.)
x
_
0,
_
3
2
_
2/5
_
3
2
2/5
_
_
3
2
_
2/5
,
_
f
0 +
f m
y = x
5/2
x
2
(x > 0)
35. y = x +x
1
(x > 0)
solution Let f (x) = x +x
1
for x > 0. Thenf
(x) = 1x
2
= 0yieldsthecritical point c = 1. (Note: c = 1
isnot intheinterval under consideration.)
x (0, 1) 1 (1, )
f
0 +
f m
y = x
2
4x
1
(x > 0)
37. y =
1
x
2
+1
_
x
2
+1
_
1
. Thenf
(x) = 2x
_
x
2
+1
_
2
= 0yieldscritical point c = 0.
x (, 0) 0 (0, )
f
+ 0
f M
y =
2x +1
x
2
+1
39. y =
x
3
x
2
+1
x
3
x
2
+1
. Then
f
(x) =
(x
2
+1)(3x
2
) x
3
(2x)
(x
2
+1)
2
=
x
2
(x
2
+3)
(x
2
+1)
2
= 0
yieldsthesinglecritical point c = 0.
x (, 0) 0 (0, )
f
+ 0 +
f
y =
x
3
x
2
3
41. y = +sin +cos
solution Let f () = + sin + cos. Thenf
() = 1+ cos sin = 0yields thecritical points c =

2
and
c = .
_
0,

2
_

2
_
2
,
_
(, 2)
f
+ 0 0 +
f M m
y = sin +
3cos
43. y = sin
2
+sin
solution Let f () = sin
2
+ sin. Thenf
() = 2sin cos + cos = cos(2sin + 1) = 0yieldsthecritical

pointsc =

2
,
7
6
,
3
2
, and
11
6
.
_
0,

2
_

2
_
2
,
7
6
_
7
6
_
7
6
,
3
2
_
3
2
_
3
2
,
11
6
_
11
6
_
11
6
, 2
_
f
+ 0 0 + 0 0 +
f M m M m
y = 2cos, [0, 2]
45. Findtheminimumvalueof f (x) = x
x
for x > 0.
x
. By logarithmic differentiation, weknowthat f
(x) = x
x
(1+ lnx). Thus, x =
1
e
is the
onlycritical point. Becausef
(x) < 0for 0< x <

1
e
andf
(x) > 0for x >

1
e
,
f
_
1
e
_
=
_
1
e
_
1/e
0.692201
istheminimumvalue.
Showthat f (x) = x
2
+bx +c isdecreasingon
_
,
b
2
_
andincreasingon
_
b
2
,
_
.
47. Showthat f (x) = x
3
2x
2
+2x isanincreasingfunction. Hint: Findtheminimumvalueof f
(x).
3
2x
2
+2x. For all x, wehave
f
(x) = 3x
2
4x +2= 3
_
x
2
3
_
2
+
2
3
2
3
> 0.
Sincef
(x) > 0for all x, thefunctionf iseverywhereincreasing.

Findconditionsona andb that ensurethat f (x) = x
3
+ax +b isincreasingon(, ).
49. Leth(x) =
x(x
2
1)
x
2
+1
andsupposethatf
(x) = h(x). Ploth(x) andusetheplottodescribethelocal extrema

andtheincreasing/decreasingbehavior of f (x). Sketchaplausiblegraphfor f (x) itself.
solution Thegraphof h(x) isshownbelowat theleft. Becauseh(x) isnegativefor x < 1andfor 0 < x < 1, it
follows that f (x) is decreasingfor x < 1andfor 0 < x < 1. Similarly, f (x) is increasingfor 1 < x < 0andfor
x > 1becauseh(x) is positiveontheseintervals. Moreover, f (x) has local minimaat x = 1andx = 1andalocal
maximumat x = 0. A plausiblegraphfor f (x) isshownbelowat theright.
x
0.3
0.2
0.1
2 1 1 2
0.2
x
1
0.5
1
0.5
h(x) f(x)
1 2 2 1
Sammadetwostatementsthat Deborahfounddubious.
(a) Theaveragevelocityfor mytripwas70mph; at nopoint intimedidmyspeedometer read70mph.
(b) A policemanclockedmegoing70mph, but myspeedometer never read65mph.
Ineachcase, whichtheoremdidDeborahapplytoproveSamsstatement false: theIntermediateValueTheoremor
theMeanValueTheorem? Explain.
51. Determine where f (x) = (1000 x)
2
+ x
2
is decreasing. Use this to decide which is larger: 800
2
+ 200
2
or
600
2
+400
2
.
solution If f (x) = (1000 x)
2
+ x
2
, then f
(x) = 2(1000 x) + 2x = 4x 2000. f
(x) < 0 as long as

x < 500. Therefore, 800
2
+200
2
= f (200) > f (400) = 600
2
+400
2
.
Showthat f (x) = 1|x| satisestheconclusionof theMVT on[a, b] if botha andb arepositiveor negative,
but not if a < 0andb > 0.
53. Whichvaluesof c satisfytheconclusionof theMVT ontheinterval [a, b] if f (x) isalinear function?
solution Letf (x) = px +q, wherep andq areconstants. Thentheslopeof every secantlineandtangentlineof f is
p.Accordingly, consideringtheinterval [a, b], every pointc (a, b) satisesf
(c) = p =
f (b) f (a)
b a
, theconclusion
of theMVT.
Showthat if f (x) isanyquadraticpolynomial, thenthemidpoint c =
a +b
2
satisestheconclusionof theMVT
on[a, b] for anya andb.
55. Supposethat f (0) = 2andf
(x) 3for x > 0. Apply theMVT to theinterval [0, 4] to provethat f (4) 14.
Provemoregenerallythat f (x) 2+3x for all x > 0.
solution TheMVT, appliedtotheinterval [0, 4], guaranteesthat thereexistsac (0, 4) suchthat
f
(c) =
f (4) f (0)
40
or f (4) f (0) = 4f
(c).
Becausec > 0, f
(c) 3, sof (4) f (0) 12. Finally, f (4) f (0) +12= 14.
Moregenerally, let x > 0. TheMVT, appliedtotheinterval [0, x], guaranteesthereexistsac (0, x) suchthat
f
(c) =
f (x) f (0)
x 0
or f (x) f (0) = f
(c)x.
Becausec > 0, f
(c) 3, sof (x) f (0) 3x. Finally, f (x) f (0) +3x = 3x +2.
Showthat if f (2) = 2andf
(x) 5for x > 2, thenf (4) 8.

57. Showthat if f (2) = 5andf
(x) 10for x > 2, thenf (x) 10x 15for all x > 2.

solution Let x > 2. TheMVT, appliedtotheinterval [2, x], guaranteesthereexistsac (2, x) suchthat
f
(c) =
f (x) f (2)
x 2
or f (x) f (2) = (x 2)f
(c).
Becausef
(x) 10, it followsthat f (x) f (2) 10(x 2), or f (x) f (2) +10(x 2) = 10x 15.
Showthat acubicfunctionf (x) = x
3
+ax
2
+bx +c isincreasingon(, ) if b > a
2
/3.
59. Provethatif f (0) = g(0) andf
(x) g
(x) forx 0, thenf (x) g(x) forall x 0. Hint: Showthatf (x) g(x)
isnonincreasing.
solution Let h(x) = f (x) g(x). By thesumrule, h
(x) = f
(x) g
(x). Sincef
(x) g
(x) for all x 0,

h
(x) 0for all x 0. Thisimpliesthat h isnonincreasing. Sinceh(0) = f (0) g(0) = 0, h(x) 0for all x 0(as
h isnonincreasing, it cannot climbabovezero). Hencef (x) g(x) 0for all x 0, andsof (x) g(x) for x 0.
UseExercise59toprovethat x tanx for 0 x <

2
.
61. UseExercise 59 and theinequality sinx x for x 0 (established inTheorem3 of Section 2.6) to provethe
followingassertionsfor all x 0(eachassertionfollowsfromthepreviousone).
(a) cosx 1
1
2
x
2
(b) sinx x
1
6
x
3
(c) cosx 1
1
2
x
2
+
1
24
x
4
(d) Canyouguessthenext inequalityintheseries?
solution
(a) Weprovethis using Exercise59: Let g(x) = cosx and f (x) = 1
1
2
x
2
. Then f (0) = g(0) = 1 and g
(x) =
sinx x = f
(x) for x 0byExercise60. NowapplyExercise59toconcludethat cosx 1

1
2
x
2
for x 0.
(b) Let g(x) = sinx andf (x) = x
1
6
x
3
. Thenf (0) = g(0) = 0andg
(x) = cosx 1
1
2
x
2
= f
(x) for x 0by

part (a). NowapplyExercise59toconcludethat sinx x
1
6
x
3
for x 0.
(c) Let g(x) = 1
1
2
x
2
+
1
24
x
4
andf (x) = cosx. Thenf (0) = g(0) = 1andg
(x) = x +
1
6
x
3
sinx = f
(x)
for x 0bypart (b). NowapplyExercise59toconcludethat cosx 1
1
2
x
2
+
1
24
x
4
for x 0.
(d) Thenext inequality intheseries is sinx x
1
6
x
3
+
1
120
x
5
, validfor x 0. Toconstruct (d) from(c), wenote
that thederivativeof sinx is cosx, andlook for apolynomial (whichwecurrently must do by educatedguess) whose
derivativeis 1
1
2
x
2
+
1
24
x
4
. Weknowthederivativeof x is 1, andthat atermwhosederivativeis
1
2
x
2
shouldbe
of theformCx
3
.
d
dx
Cx
3
= 3Cx
2
=
1
2
x
2
, soC =
1
6
. A termwhosederivativeis
1
24
x
4
shouldbeof theformDx
5
.
Fromthis,
d
dx
Dx
5
= 5Dx
4
=
1
24
x
4
, sothat 5D =
1
24
, or D =
1
120
.
Supposethat f (x) isafunctionsuchthat f (0) = 1andfor all x, f
(x) = f (x) andf (x) > 0(inChapter 7, we

will seethat f (x) istheexponential functione
x
). Provethat for all x 0(eachassertionfollowsfromtheprevious
one),
(a) f (x) 1
(b) f (x) 1+x
(c) f (x) 1+x +
1
2
x
2
Thenprovebyinductionthat for everywholenumber n andall x 0,
f (x) 1+x +
1
2!
x
2
+ +
1
n!
x
n
63. Assumethat f
existsandf
(x) = 0for all x. Provethat f (x) = mx +b, wherem = f
(0) andb = f (0).

solution
Let f
(x) = 0for all x. Thenf
(x) = constant for all x. Sincef
(0) = m, weconcludethat f
(x) = m for all x.

Let g(x) = f (x) mx. Theng
(x) = f
(x) m = mm = 0whichimpliesthat g(x) = constant for all x and

consequentlyf (x) mx = constant for all x. Rearrangingthestatement, f (x) = mx +constant. Sincef (0) = b,
weconcludethat f (x) = mx +b for all x.
Denef (x) = x
3
sin
_
1
x
_
for x = 0andf (0) = 0.
(a) Showthat f
(x) iscontinuousat x = 0andthat x = 0isacritical point of f .

(b) Examinethegraphsof f (x) andf
(x). CantheFirst DerivativeTest beapplied?

(c) Showthat f (0) isneither alocal minnor alocal max.
65. Supposethat f (x) satisesthefollowingequation(anexampleof adifferential equation):
f
(x) = f (x) 1
(a) Showthat f (x)
2
+f
(x)
2
= f (0)
2
+f
(0)
2
for all x. Hint: Showthat thefunctionontheleft haszeroderivative.
(b) Verifythat sinx andcosx satisfyEq. (1), anddeducethat sin
2
x +cos
2
x = 1.
solution
(a) Let g(x) = f (x)
2
+f
(x)
2
. Then
g
(x) = 2f (x)f
(x) +2f
(x)f
(x) = 2f (x)f
(x) +2f
(x)(f (x)) = 0,
wherewehaveusedthefactthatf
(x) = f (x). Becauseg
(0) = 0forall x, g(x) = f (x)

2
+f
(x)
2
mustbeaconstant
function. Inother words, f (x)
2
+f
(x)
2
= C for someconstant C. Todeterminethevalueof C, wecansubstituteany
number for x. Inparticular, for thisproblem, wewant tosubstitutex = 0andndC = f (0)
2
+f
(0)
2
. Hence,
f (x)
2
+f
(x)
2
= f (0)
2
+f
(0)
2
.
(b) Let f (x) = sinx. Thenf
(x) = cosx andf
(x) = sinx, so f
(x) = f (x). Next, let f (x) = cosx. Then

f
(x) = sinx, f
(x) = cosx, andweagainhavef
(x) = f (x). Finally, if wetakef (x) = sinx, theresultfrom

part (a) guaranteesthat
sin
2
x +cos
2
x = sin
2
0+cos
2
0= 0+1= 1.
Suppose that functions f and g satisfy Eq. (1) and have the same initial valuesthat is, f (0) = g(0) and
f
(0) = g
(0). Provethat f (x) = g(x) for all x. Hint: ApplyExercise65(a) tof g.

67. UseExercise66to prove: f (x) = sinx is theuniquesolutionof Eq. (1) suchthat f (0) = 0andf
(0) = 1; and
g(x) = cosx istheuniquesolutionsuchthatg(0) = 1andg
(0) = 0. Thisresultcanbeusedtodevelopall theproperties

of thetrigonometricfunctionsanalyticallythat is, without referencetotriangles.
solution Inpart (b) of Exercise65, it was shownthat f (x) = sinx satises Eq. (1), andwecandirectly calculate
that f (0) = sin0= 0andf
(0) = cos0= 1. Supposethereisanother function, call it F(x), that satisesEq. (1) with
thesameinitial conditions: F(0) = 0andF
(0) = 1. By Exercise66, it follows that F(x) = sinx for all x. Hence,

f (x) = sinx is theuniquesolutionof Eq. (1) satisfyingf (0) = 0andf
(0) = 1. Theproof that g(x) = cosx is the

uniquesolutionof Eq. (1) satisfyingg(0) = 1andg
(0) = 0iscarriedout inasimilar manner.

4.4 The Shape of a Graph
1. If f isconcaveup, thenf
is(chooseone):
(a) increasing (b) decreasing
solution Thecorrect responseis (a): increasing. If thefunctionis concaveup, thenf
is positive. Sincef
is the
derivativeof f
, it followsthat thederivativeof f
ispositiveandf
must thereforebeincreasing.
S E C T I ON 4.4 The Shape of a Graph 199
2. What conclusioncanyoudrawif f
(c) = 0andf
(c) < 0?
solution If f
(c) = 0andf
(c) < 0, thenf (c) isalocal maximum.

3. Trueor False? If f (c) isalocal min, thenf
(c) must bepositive.

solution False. f
(c) couldbezero.
4. Trueor False? If f
(x) changesfrom+ to at x = c, thenf hasapoint of inectionat x = c.

solution False. f will haveapoint of inectionat x = c onlyif x = c isinthedomainof f .
Exercises
1. MatchthegraphsinFigure13withthedescription:
(a) f
(x) < 0for all x. (b) f
(x) goesfrom+ to.

(c) f
(x) > 0for all x. (d) f
(x) goesfrom to+.

(A) (B) (C) (D)
FIGURE 13
solution
(a) InC, wehavef
(x) < 0for all x.

(b) InA, f
(x) goesfrom+ to.

(c) InB, wehavef
(x) > 0for all x.

(d) InD, f
(x) goesfrom to+.

Matcheachstatement withagraphinFigure14that representscompanyprotsasafunctionof time.
(a) Theoutlookisgreat: Thegrowthratekeepsincreasing.
(b) Werelosingmoney, but not asquicklyasbefore.
(c) Werelosingmoney, anditsgettingworseastimegoeson.
(d) Weredoingwell, but our growthrateislevelingoff.
(e) Businesshadbeencoolingoff, but nowitspickingup.
(f) Businesshadbeenpickingup, but nowitscoolingoff.
In Exercises 314, determine the intervals on which the function is concave up or down and nd the points of inection.
3. y = x
2
4x +3
2
4x + 3. Thenf
(x) = 2x 4andf
(x) = 2> 0for all x. Therefore, f isconcaveup

everywhere, andtherearenopointsof inection.
y = t
3
6t
2
+4
5. y = 10x
3
x
5
3
x
5
. Thenf
(x) = 30x
2
5x
4
andf
(x) = 60x 20x

3
= 20x(3 x
2
). Now, f is
concaveupfor x <
3andfor 0< x <
3sincef
(x) > 0there. Moreover, f isconcavedownfor
3< x < 0
andfor x >
3sincef
(x) < 0there. Finally, becausef
(x) changessignat x = 0andat x =
3, f (x) hasapoint
of inectionat x = 0andat x =
3.
y = 5x
2
+x
4
7. y = 2sin, [0, 2]
solution Letf () = 2sin.Thenf
() = 12cos andf
() = 2sin. Now, f isconcaveupfor0< <

sincef
() > 0there. Moreover, f isconcavedownfor < < 2 sincef
() < 0there. Finally, becausef
()
changessignat = , f () hasapoint of inectionat = .
y = +sin
2
, [0, ]
9. y = x(x 8
x) (x 0)
solution Let f (x) = x(x 8
x) = x
2
8x
3/2
. Thenf
(x) = 2x 12x
1/2
andf
(x) = 2 6x
1/2
. Now, f
is concavedownfor 0 < x < 9sincef
(x) < 0there. Moreover, f is concaveupfor x > 9sincef
(x) > 0there.

Finally, becausef
(x) changessignat x = 9, f (x) hasapoint of inectionat x = 9.

y = x
7/2
35x
2
11. y = (x 2)(1x
3
)
solution Let f (x) = (x 2)
_
1x
3
_
= x x
4
2+ 2x
3
. Thenf
(x) = 1 4x
3
+ 6x
2
andf
(x) = 12x
12x
2
= 12x(1 x) = 0at x = 0andx = 1. Now, f is concaveupon(0, 1) sincef
(x) > 0there. Moreover, f

isconcavedownon(, 0) (1, ) sincef
(x) < 0there. Finally, becausef
(x) changessignat bothx = 0and

x = 1, f (x) hasapoint of inectionat bothx = 0andx = 1.
y = x
7/5 13. y =
1
x
2
+3
1
x
2
+3
. Thenf
(x) =
2x
(x
2
+3)
2
and
f
(x) =
2(x
2
+3)
2
8x
2
(x
2
+3)
(x
2
+3)
4
=
6x
2
6
(x
2
+3)
3
.
Now, f is concaveupfor |x| > 1sincef
(x) > 0there. Moreover, f is concavedownfor |x| < 1sincef
(x) < 0
there. Finally, becausef
(x) changessignat bothx = 1andx = 1, f (x) hasapoint of inectionat bothx = 1and

x = 1.
y =
x
x
2
+9
15. Thegrowthof asunower duringtherst 100days after sproutingis modeledwell by thelogistic curve
y = h(t ) showninFigure15. Estimatethegrowthrateat thepoint of inectionandexplainitssignicance. Thenmake
aroughsketchof therst andsecondderivativesof h(t ).
20 40 60 80 100
50
100
150
200
300
250
t (days)
Height (cm)
FIGURE 15
solution Thepoint of inectioninFigure15appears to occur at t = 40days. Thegraphbelowshows thelogistic
curvewithanapproximatetangent linedrawnat t = 40. Theapproximatetangent linepassesroughlythroughthepoints
(20, 20) and(60, 240). Thegrowthrateat thepoint of inectionisthus
24020
6020
=
220
40
= 5.5cm/day.
Becausethelogistic curvechanges fromconcaveup to concavedown at t = 40, thegrowth rateat this point is the
maximumgrowthratefor thesunower plant.
20 40 60 80 100
50
100
150
200
300
250
t (days)
Height (cm)
Sketchesof therst andsecondderivativeof h(t ) areshownbelowat theleft andat theright, respectively.
20 40 60 80 100
1
2
3
4
6
5
t
100 80 60 40 20
t
h
0.1
0.1
h
AssumethatFigure16isthegraphof f (x). Wheredothepointsof inectionof f (x) occur, andonwhichinterval
isf (x) concavedown?
17. Repeat Exercise16but assumethat Figure16isthegraphof thederivative f
(x).
solution Points of inectionoccur whenf
(x) changes sign. Consequently, points of inectionoccur whenf
(x)
changesfromincreasingtodecreasingor fromdecreasingtoincreasing. InFigure16, thisoccursat x = b andat x = e;
therefore, f (x) has aninectionpoint at x = b andanother at x = e. Thefunctionf (x) will beconcavedownwhen
f
(x) < 0or whenf
(x) isdecreasing. Thus, f (x) isconcavedownfor b < x < e.

Repeat Exercise16but assumethat Figure16isthegraphof thesecond derivative f
(x).
19. Figure17shows thederivative f
(x) on[0, 1.2]. Locatethepoints of inectionof f (x) andthepoints wherethe

local minimaandmaximaoccur. Determinetheintervalsonwhichf (x) hasthefollowingproperties:
(a) Increasing (b) Decreasing
(c) Concaveup (d) Concavedown
1.2 1 0.17 0.64 0.4
x
y
y = f '(x)
FIGURE 17
solution Recall thatthegraphisthatof f
, not f . Theinectionpointsof f occur wheref
changesfromincreasing
todecreasingor viceversabecauseit isat thesepointsthat thesignof f
changes. Fromthegraphweconcludethat f
has points of inectionat x = 0.17, x = 0.64, andx = 1. Thelocal extremaof f occur wheref
changes sign. This

occursat x = 0.4. Becausethesignof f
changesfrom+ to, f (0.4) isalocal maximum. Therearenolocal minima.

(a) f isincreasingwhenf
ispositive. Hence, f isincreasingon(0, 0.4).

(b) f isdecreasingwhenf
isnegative. Hence, f isdecreasingon(0.4, 1) (1, 1.2).

(c) Nowf isconcaveupwheref
isincreasing. Thisoccurson(0, 0.17) (0.64, 1).

(d) Moreover, f isconcavedownwheref
isdecreasing. Thisoccurson(0.17, 0.64) (1, 1.2).

Leticiahasbeensellingsolar-poweredlaptopchargersthroughherwebsite, withmonthlysalesasrecordedbelow.
Inareporttoinvestors, shestates, Salesreachedapointof inectionwhenI startedusingpay-per-clickadvertising.
Inwhichmonthdidthat occur? Explain.
Month 1 2 3 4 5 6 7 8
Sales 2 30 50 60 90 150 230 340
In Exercises 2134, nd the critical points and apply the Second Derivative Test (or state that it fails).
21. f (x) = x
3
12x
2
+45x
3
12x
2
+ 45x. Thenf
(x) = 3x
2
24x + 45= 3(x 3)(x 5), andthecritical points
arex = 3andx = 5. Moreover, f
(x) = 6x 24, sof
(3) = 6< 0andf
(5) = 6> 0. Therefore, bytheSecond

DerivativeTest, f (3) = 54isalocal maximum, andf (5) = 50isalocal minimum.
f (x) = x
4
8x
2
+1
23. f (x) = 3x
4
8x
3
+6x
2
4
8x
3
+ 6x
2
. Thenf
(x) = 12x
3
24x
2
+ 12x = 12x(x 1)
2
= 0at x = 0, 1and
f
(x) = 36x
2
48x + 12. Thus, f
(0) > 0, whichimpliesf (0) isalocal minimum; however, f
(1) = 0, whichis
inconclusive.
f (x) = x
5
x
3
25. f (x) =
x
2
8x
x +1
x
2
8x
x +1
. Then
f
(x) =
x
2
+2x 8
(x +1)
2
and f
(x) =
2(x +1)
2
2(x
2
+2x 8)
(x +1)
3
.
Thus, thecritical points arex = 4 and x = 2. Moreover, f
(4) < 0 and f
(2) > 0. Therefore, by thesecond

derivativetest, f (4) = 16isalocal maximumandf (2) = 4isalocal minimum.
f (x) =
1
x
2
x +2
27. y = 6x
3/2
4x
1/2
3/2
4x
1/2
. Thenf
(x) = 9x
1/2
2x
1/2
= x
1/2
(9x 2), so therearetwo critical
points: x = 0andx =
2
9
. Now,
f
(x) =
9
2
x
1/2
+x
3/2
=
1
2
x
3/2
(9x +2).
Thus, f
_
2
9
_
> 0, whichimpliesf
_
2
9
_
isalocal minimum. f
(x) isundenedatx = 0, sotheSecondDerivativeTest

cannot beappliedthere.
y = 9x
7/3
21x
1/2 29. f (x) = x
3
+
48
x
solution Wehavef
(x) = 3x
2
48x
2
, sof
(x) = 0when3x
4
= 48, sox = 2. Now, f
(x) = 6x + 96x
3
,
andf
(2) = 12
96
8
= 24 < 0whilef
(2) = 12+
96
8
= 24 > 0, so that x = 2is alocal maximumand
x = 2isalocal minimum.
f (x) = x
4
+
128
x
2
31. f (x) = sin
2
x +cosx, [0, ]
2
x +cosx. Thenf
(x) = 2sinx cosx sinx = sinx(2cosx 1). Ontheinterval [0, ],

f
(x) = 0at x = 0, x =

3
andx = . Now,
f
(x) = 2cos
2
x 2sin
2
x cosx.
Thus, f
(0) > 0, so f (0) is alocal minimum. Ontheother hand, f
3
) < 0, so f (
3
) is alocal maximum. Finally,
f
() > 0, sof () isalocal minimum.

y =
1
sinx +4
, [0, 2]
33. f (x) = 2+tan
2
x,
_
2
,
2
_
solution Wehavef
(x) = 2tanx sec

2
x andf
(x) = 2sec
4
x + 4tan
2
x sec
2
x. Now, f
(x) = 0inthespecied
interval whentanx = 0, i.e. whenx = 0. Sincef
(0) = 2, f (0) = 2isalocal minimum.

f (x) = sinx cos
3
x, [0, ]
In Exercises 3552, nd the intervals on which f is concave up or down, the points of inection, the critical points, and
the local minima and maxima.
solution Hereisatablelegendfor Exercises3545.
SYMBOL MEANING
Theentityisnegativeonthegiveninterval.
0 Theentityiszeroat thespeciedpoint.
+ Theentityispositiveonthegiveninterval.
U Theentityisundenedat thespeciedpoint.
Thefunction(f , g, etc.) isincreasingonthegiveninterval.
Thefunction(f , g, etc.) isdecreasingonthegiveninterval.
Thefunction(f , g, etc.) isconcaveuponthegiveninterval.
Thefunction(f , g, etc.) isconcavedownonthegiveninterval.
M Thefunction(f , g, etc.) hasalocal maximumat thespeciedpoint.
m Thefunction(f , g, etc.) hasalocal minimumat thespeciedpoint.
I Thefunction(f , g, etc.) hasaninectionpoint here.
Thereisnolocal extremumor inectionpoint here.
35. f (x) = x
3
2x
2
+x
3
2x
2
+x.
Thenf
(x) = 3x
2
4x +1= (x 1)(3x 1) = 0yieldsx = 1andx =
1
3
ascandidatesfor extrema.
Moreover, f
(x) = 6x 4= 0givesacandidatefor apoint of inectionat x =

2
3
.
x
_
,
1
3
_
1
3
_
1
3
, 1
_
1
_
1,
_
f
+ 0 0 +
f M m
x
_
,
2
3
_
2
3
_
2
3
,
_
f
0 +
f I
f (x) = x
2
(x 4)
37. f (t ) = t
2
t
3
solution Let f (t ) = t
2
t
3
.
Thenf
(t ) = 2t 3t
2
= t (23t ) = 0yieldst = 0andt =
2
3
Moreover, f
(t ) = 26t = 0givesacandidatefor apoint of inectionat t =

1
3
.
t (, 0) 0
_
0,
2
3
_
2
3
_
2
3
,
_
f
0 + 0
f m M
t
_
,
1
3
_
1
3
_
1
3
,
_
f
+ 0
f I
f (x) = 2x
4
3x
2
+2
39. f (x) = x
2
8x
1/2
(x 0)
2
8x
1/2
. Notethat thedomainof f isx 0.
Thenf
(x) = 2x 4x
1/2
= x
1/2
_
2x
3/2
4
_
= 0yieldsx = 0andx = (2)
2/3
Moreover, f
(x) = 2+2x
3/2
> 0for all x 0, whichmeanstherearenoinectionpoints.
x 0
_
0, (2)
2/3
_
(2)
2/3
_
(2)
2/3
,
_
f
U 0 +
f M m
f (x) = x
3/2
4x
1/2
(x > 0)
41. f (x) =
x
x
2
+27
x
x
2
+27
.
Thenf
(x) =
27x
2
_
x
2
+27
_
2
= 0yieldsx = 3
3ascandidatesfor extrema.
Moreover, f
(x) =
2x
_
x
2
+27
_
2
(27x
2
)(2)
_
x
2
+27
_
(2x)
_
x
2
+27
_
4
=
2x
_
x
2
81
_
_
x
2
+27
_
3
= 0givescandidatesfor
apoint of inectionat x = 0andat x = 9.
x
_
, 3
3
_
3
3
_
3
3, 3
3
_
3
3
_
3
3,
_
f
0 + 0
f m M
x (, 9) 9 (9, 0) 0 (0, 9) 9 (9, )
f
0 + 0 0 +
f I I I
f (x) =
1
x
4
+1
43. f () = +sin, [0, 2]
solution Let f () = +sin on[0, 2].
Thenf
() = 1+cos = 0yields = asacandidatefor anextremum.

Moreover, f
() = sin = 0givescandidatesfor apoint of inectionat = 0, at = , andat = 2.

(0, ) (, 2)
f
+ 0 +
f
0 (0, ) (, 2) 2
f
0 0 + 0
f I
f (x) = cos
2
x, [0, ]
45. f (x) = tanx,
_
2
,

2
_
solution Let f (x) = tanx on
_
2
,

2
_
.
Thenf
(x) = sec
2
x 1> 0on
_
2
,

2
_
.
Moreover, f
(x) = 2secx secx tanx = 2sec

2
x tanx = 0givesacandidatefor apoint of inectionat x = 0.
x
_
2
,

2
_
f
+
f
x
_
2
, 0
_
0
_
0,

2
_
f
0 +
f I
f (x) =
x
x
6
+5
x (, 1) 1 (1, 1) 1 (1, )
f
0 + 0
f m M
x (,
6
7)
6
7 (
6
7, 0) 0 (0,
6
7)
6
7 (
6
7, )
f
0 + 0 0 +
f I I I
47. Sketchthegraphof anincreasingfunctionsuchthatf
(x) changesfrom+toatx = 2andfromto+atx = 4.

Dothesamefor adecreasingfunction.
solution Thegraph shown belowat theleft is an increasing function which changes fromconcaveup to concave
downat x = 2andfromconcavedownto concaveupat x = 4. Thegraphshownbelowat theright is adecreasing
functionwhichchangesfromconcaveuptoconcavedownat x = 2andfromconcavedowntoconcaveupat x = 4.
x
2 4
2
1
y
x
2 4
6
2
4
y
In Exercises 4850, sketch the graph of a function f (x) satisfying all of the given conditions.
f
(x) > 0andf
(x) < 0for all x.

49. (i) f
(x) > 0for all x, and

(ii) f
(x) < 0for x < 0andf
(x) > 0for x > 0.

solution Hereis thegraphof afunctionf (x) satisfying(i) f
(x) > 0for all x and (ii) f
(x) < 0for x < 0and

f
(x) > 0for x > 0.

x
10
5
10
5
y
1 2 2 1
(i) f
(x) < 0for x < 0andf
(x) > 0for x > 0, and

(ii) f
(x) < 0for |x| > 2, andf
(x) > 0for |x| < 2.

51. Aninfectious uspreads slowly at thebeginningof anepidemic. Theinfectionprocess accelerates until a
majorityof thesusceptibleindividualsareinfected, at whichpoint theprocessslowsdown.
(a) If R(t ) isthenumber of individualsinfectedat timet , describetheconcavity of thegraphof R near thebeginning
andendof theepidemic.
(b) Describethestatusof theepidemiconthedaythat R(t ) hasapoint of inection.
solution
(a) Near thebeginningof theepidemic, thegraphof R isconcaveup. Near theepidemicsend, R isconcavedown.
(b) Epidemicsubsiding: number of newcasesdeclining.
Water ispumpedintoasphereat aconstant rate(Figure18). Let h(t ) bethewater level at timet . Sketchthe
graphof h(t ) (approximately, but withthecorrect concavity). Wheredoesthepoint of inectionoccur?
53. Water is pumped into asphereof radius R at avariableratein such away that thewater level rises at a
constant rate(Figure18). Let V(t ) bethevolumeof water inthetank at timet . SketchthegraphV(t ) (approximately,
but withthecorrect concavity). Wheredoesthepoint of inectionoccur?
solution Becausewater isenteringthesphereinsuchawaythatthewater level risesataconstantrate, weexpectthe
volumetoincreasemoreslowlynear thebottomandtopof thespherewherethesphereisnot aswide andtoincrease
morerapidlynear themiddleof thesphere. Thegraphof V(t ) shouldthereforestart concaveupandchangetoconcave
downwhenthesphereishalf full; that is, thepoint of inectionshouldoccur whenthewater level isequal totheradius
of thesphere. A possiblegraphof V(t ) isshownbelow.
t
V
(Continuationof Exercise53) If thespherehasradiusR, thevolumeof water isV =
_
Rh
2
1
3
h
3
_
whereh is
thewater level. Assumethelevel risesat aconstant rateof 1(that is, h = t ).
(a) Findtheinectionpoint of V(t ). Doesthisagreewithyour conclusioninExercise53?
(b) Plot V(t ) for R = 1.
In Exercises 5557, assume that f (x) is differentiable.
55. Proof of theSecondDerivativeTest Let c beacritical point suchthat f
(c) > 0(thecasef
(c) < 0issimilar).

(a) Showthat f
(c) = lim
h0
f
(c +h)
h
.
(b) Use(a) toshowthatthereexistsanopeninterval (a, b) containingc suchthatf
(x) < 0if a < x < c andf
(x) > 0
if c < x < b. Concludethat f (c) isalocal minimum.
solution
(a) Becausec isacritical point, either f
(c) = 0or f
(c) doesnotexist; however, f
(c) exists, sof
(c) mustalsoexist.
Therefore, f
(c) = 0. Now, fromthedenitionof thederivative, wehave

f
(c) = lim
h0
f
(c +h) f
(c)
h
= lim
h0
f
(c +h)
h
.
(b) Wearegiventhat f
(c) > 0. Bypart (a), it followsthat

lim
h0
f
(c +h)
h
> 0;
inother words, for sufcientlysmall h,
f
(c +h)
h
> 0.
Now, if h issufciently small but negative, thenf
(c + h) must alsobenegative(sothat theratiof
(c + h)/h will be
positive) andc + h < c. Ontheother hand, if h issufciently small but positive, thenf
(c + h) must alsobepositive
andc +h > c. Thus, thereexistsanopeninterval (a, b) containingc suchthat f
(x) < 0for a < x < c andf
(c) > 0
for c < x < b. Finally, becausef
(x) changesfromnegativetopositiveat x = c, f (c) must bealocal minimum.

Provethat if f
(x) existsandf
(x) > 0for all x, thenthegraphof f (x) sitsabove itstangent lines.

(a) For anyc, setG(x) = f (x) f
(c)(x c) f (c). ItissufcienttoprovethatG(x) 0for all c. Explainwhy

withasketch.
(b) Showthat G(c) = G
(c) = 0andG
(x) > 0for all x. Concludethat G
(x) < 0for x < c andG
(x) > 0for

x > c. Thendeduce, usingtheMVT, that G(x) > G(c) for x = c.
57. Assumethat f
(x) existsandlet c beapoint of inectionof f (x).

(a) Usethemethodof Exercise56toprovethat thetangent lineat x = c crosses the graph (Figure19). Hint: Showthat
G(x) changessignat x = c.
(b) Verifythisconclusionfor f (x) =
x
3x
2
+1
bygraphingf (x) andthetangent lineat eachinectionpoint on
thesameset of axes.
FIGURE 19 Tangent linecrossesgraphat point of inection.
solution
(a) LetG(x) = f (x) f
(c)(x c) f (c). Then, asinExercise55, G(c) = G
(c) = 0andG
(x) = f
(x). If f
(x)
changes frompositiveto negativeat x = c, thenso does G
(x) andG
(x) is increasingfor x < c anddecreasingfor

x > c. Thismeansthat G
(x) < 0for x < c andG
(x) < 0for x > c. Thisinturnimpliesthat G(x) isdecreasing, so

G(x) > 0for x < c butG(x) < 0for x > c. Ontheother hand, if f
(x) changesfromnegativetopositiveatx = c, then

sodoesG
(x) andG
(x) isdecreasingfor x < c andincreasingfor x > c. Thus, G
(x) > 0for x < c andG
(x) > 0
for x > c. Thisinturnimpliesthat G(x) isincreasing, soG(x) < 0for x < c andG(x) > 0for x > c. Ineither case,
G(x) changessignat x = c, andthetangent lineat x = c crossesthegraphof thefunction.
(b) Let f (x) =
x
3x
2
+1
. Then
f
(x) =
13x
2
(3x
2
+1)
2
and f
(x) =
18x(1x
2
)
(3x
2
+1)
3
.
Thereforef (x) hasapoint of inectionat x = 0andat x = 1. Thegurebelowshowsthegraphof y = f (x) andits
tangentlinesateachof thepointsof inection. Itisclear thateachtangentlinecrossesthegraphof f (x) attheinection
point.
x
y
Let C(x) bethecost of producingx unitsof acertaingood. Assumethat thegraphof C(x) isconcaveup.
(a) ShowthattheaveragecostA(x) = C(x)/x isminimizedattheproductionlevel x
0
suchthataveragecostequals
marginal costthat is, A(x
0
) = C
(x
0
).
(b) Showthat thelinethrough(0, 0) and(x
0
, C(x
0
)) istangent tothegraphof C(x).
59. Let f (x) beapolynomial of degreen 2. Showthat f (x) hasat least onepoint of inectionif n isodd. Thengive
anexampletoshowthat f (x) neednot haveapoint of inectionif n iseven.
solution Let f (x) = a
n
x
n
+a
n1
x
n1
+ +a
1
x +a
0
beapolynomial of degreen. Thenf
(x) = na
n
x
n1
+
(n 1)a
n1
x
n2
+ +2a
2
x +a
1
andf
(x) = n(n 1)a

n
x
n2
+(n 1)(n 2)a
n1
x
n3
+ +6a
3
x +2a
2
.
If n 3andisodd, thenn 2isalsooddandf
(x) isapolynomial of odddegree. Thereforef
(x) must takeonboth

positiveandnegativevalues. It followsthat f
(x) hasat least oneroot c suchthat f
(x) changessignat c. Thefunction

f (x) will thenhaveapointof inectionatx = c. Ontheother hand, thefunctionsf (x) = x
2
, x
4
andx
8
arepolynomials
of evendegreethat donot haveanypointsof inection.
Critical andInectionPoints If f
(c) = 0andf (c) isneither alocal minnor alocal max, must x = c bea
point of inection? Thisistruefor reasonable functions(includingthefunctionsstudiedinthistext), but it isnot
trueingeneral. Let
f (x)
_
x
2
sin
1
x
for x = 0
4.5 Graph Sketching and Asymptotes
1. Sketchanarcwheref
andf
havethesigncombination++. Dothesamefor +.
solution Anarc withthesigncombination++ (increasing, concaveup) isshownbelowat theleft. Anarc withthe
signcombination+ (decreasing, concaveup) isshownbelowat theright.
x
y
x
y
2. If thesigncombinationof f
andf
changesfrom++ to+ at x = c, then(choosethecorrect answer):

(a) f (c) isalocal min (b) f (c) isalocal max
(c) c isapoint of inection
solution Becausethesign of thesecond derivativechanges at x = c, thecorrect responseis (c): c is a point of
inection.
3. Thesecondderivativeof thefunctionf (x) = (x 4)
1
isf
(x) = 2(x 4)
3
. Althoughf
(x) changessignat
x = 4, f (x) doesnot haveapoint of inectionat x = 4. Whynot?
solution Thefunctionf doesnot haveapoint of inectionat x = 4becausex = 4isnot inthedomainof f.
Exercises
1. Determinethesigncombinationsof f
andf
for eachinterval AG inFigure15.

C B D E F G A
x
y
y = f (x)
FIGURE 15
solution
InA, f isdecreasingandconcaveup, sof
< 0andf
> 0.
InB, f isincreasingandconcaveup, sof
> 0andf
> 0.
InC, f isincreasingandconcavedown, sof
> 0andf
< 0.
InD, f isdecreasingandconcavedown, sof
< 0andf
< 0.
InE, f isdecreasingandconcaveup, sof
< 0andf
> 0.
InF, f isincreasingandconcaveup, sof
> 0andf
> 0.
InG, f isincreasingandconcavedown, sof
> 0andf
< 0.
Statethesignchangeat eachtransitionpoint AG inFigure16. Example: f
(x) goesfrom+ to at A.
In Exercises 36, draw the graph of a function for which f
and f
take on the given sign combinations.

3. ++, +,
solution Thisfunctionchangesfromconcaveuptoconcavedownat x = 1andfromincreasingtodecreasingat
x = 0.
x
y
1
0 1 1
S E C T I ON 4.5 Graph Sketching and Asymptotes 207
+, , +
5. +, , +
solution Thefunctionisdecreasingeverywhereandchangesfromconcaveuptoconcavedownat x = 1andfrom
concavedowntoconcaveupat x =
1
2
.
x
y
0.05
1 0
+, ++, +
7. Sketchthegraphof y = x
2
5x +4.
2
5x + 4. Thenf
(x) = 2x 5andf
(x) = 2. Hencef is decreasingfor x < 5/2, is

increasingfor x > 5/2, hasalocal minimumat x = 5/2andisconcaveupeverywhere.
2
5
10
15
4 6
y
x
Sketchthegraphof y = 125x 2x
2
.
9. Sketchthegraphof f (x) = x
3
3x
2
+2. Includethezerosof f (x), whicharex = 1and1
3(approximately
0.73, 2.73).
3
3x
2
+2. Thenf
(x) = 3x
2
6x = 3x(x 2) = 0yieldsx = 0, 2andf
(x) = 6x 6.
Thusf isconcavedownfor x < 1, isconcaveupfor x > 1, hasaninectionpoint at x = 1, isincreasingfor x < 0and
for x > 2, isdecreasingfor 0< x < 2, hasalocal maximumat x = 0, andhasalocal minimumat x = 2.
1
x
1 2 3 1
y
1
2
2
Showthat f (x) = x
3
3x
2
+6x hasapoint of inectionbut nolocal extremevalues. Sketchthegraph.
11. Extendthesketchof thegraphof f (x) = cosx +
1
2
x inExample4totheinterval [0, 5].
solution Let f (x) = cosx +
1
2
x. Thenf
(x) = sinx +
1
2
= 0yields critical points at x =

6
,
5
6
,
13
6
,
17
6
,
25
6
, and
29
6
. Moreover, f
(x) = cosx sotherearepointsof inectionat x =

2
,
3
2
,
5
2
,
7
2
, and
9
2
.
2
x
2 4 6 8 10 12 14
y
4
6
0
Sketchthegraphsof y = x
2/3
andy = x
4/3
.
In Exercises 1334, nd the transition points, intervals of increase/decrease, concavity, and asymptotic behavior. Then
sketch the graph, with this information indicated.
13. y = x
3
+24x
2
3
+24x
2
. Thenf
(x) = 3x
2
+48x = 3x (x +16) andf
(x) = 6x +48. Thisshowsthat f

hascritical pointsat x = 0andx = 16andacandidatefor aninectionpoint at x = 8.
Interval (, 16) (16, 8) (8, 0) (0, )
Signsof f
andf
+ + ++
Thus, thereisalocal maximumat x = 16, alocal minimumat x = 0, andaninectionpoint at x = 8. Moreover,
lim
x
f (x) = and lim
x
f (x) = .
Hereisagraphof f withthesetransitionpointshighlightedasinthegraphsinthetextbook.
2015 10 5 5
1000
2000
3000
y
x
y = x
3
3x +5
15. y = x
2
4x
3
2
4x
3
. Thenf
(x) = 2x 12x
2
= 2x(1 6x) andf
(x) = 2 24x. Critical pointsare

at x = 0andx =
1
6
, andthesolecandidatepoint of inectionisat x =
1
12
.
Interval (, 0) (0,
1
12
) (
1
12
,
1
6
) (
1
6
, )
Signsof f
andf
+ ++ +
Thus, f (0) isalocal minimum, f (
1
6
) isalocal maximum, andthereisapoint of inectionat x =
1
12
. Moreover,
lim
x
f (x) = .
Hereisthegraphof f withtransitionpointshighlightedasinthetextbook:
0.04
x
0.2 0.2
y
0.04
y =
1
3
x
3
+x
2
+3x
17. y = 42x
2
+
1
6
x
4
1
6
x
4
2x
2
+ 4. Thenf
(x) =
2
3
x
3
4x =
2
3
x
_
x
2
6
_
andf
(x) = 2x
2
4. Thisshows
that f hascritical pointsat x = 0andx =
6andhascandidatesfor pointsof inectionat x =
2.
Interval (,
6) (
6,
2) (
2, 0) (0,
2) (
2,
6) (
6, )
Signsof f
andf
+ ++ + + ++
Thus, f haslocal minimaat x =
6, alocal maximumat x = 0, andinectionpointsat x =
2. Moreover,
lim
x
f (x) = .
Hereisagraphof f withtransitionpointshighlighted.
5
x
2 2
y
10
y = 7x
4
6x
2
+1
19. y = x
5
+5x
5
+ 5x. Thenf
(x) = 5x
4
+ 5= 5(x
4
+ 1) andf
(x) = 20x
3
. f
(x) > 0for all x, sothe

graphhasnocritical pointsandisalwaysincreasing. f
(x) = 0at x = 0. Signanalysesreveal that f
(x) changesfrom
negativetopositiveat x = 0, sothat thegraphof f (x) hasaninectionpoint at (0, 0). Moreover,
lim
x
f (x) = and lim
x
f (x) = .
20
x
2 1 2 1
y
40
40
20
y = x
5
15x
3
21. y = x
4
3x
3
+4x
4
3x
3
+ 4x. Then f
(x) = 4x
3
9x
2
+ 4 = (4x
2
x 2)(x 2) and f
(x) =
12x
2
18x = 6x(2x 3). This shows that f has critical points at x = 2andx =
1
33
8
andcandidatepoints of
inectionat x = 0andx =
3
2
. Signanalysesreveal that f
(x) changesfromnegativetopositiveat x =
1
33
8
, from
positiveto negativeat x =
1+
33
8
, andagainfromnegativeto positiveat x = 2. Therefore, f (
1
33
8
) andf (2) are
local minimaof f (x), andf (
1+
33
8
) isalocal maximum. Further signanalysesreveal thatf
(x) changesfrompositive
tonegativeatx = 0andfromnegativetopositiveatx =
3
2
, sothattherearepointsof inectionbothatx = 0andx =
3
2
.
Moreover,
lim
x
f (x) = .
Hereisagraphof f (x) withtransitionpointshighlighted.
4
x
2 1 1
y
6
2
2
y = x
2
(x 4)
2
23. y = x
7
14x
6
7
14x
6
. Then f
(x) = 7x
6
84x
5
= 7x
5
(x 12) and f
(x) = 42x
5
420x
4
=
42x
4
(x 10). Critical pointsareat x = 0andx = 12, andcandidateinectionpointsareat x = 0andx = 10. Sign
analysesreveal that f
(x) changesfrompositivetonegativeat x = 0andfromnegativetopositiveat x = 12. Therefore

f (0) is alocal maximumand f (12) is alocal minimum. Also, f
(x) changes fromnegativeto positiveat x = 10.

Therefore, thereisapoint of inectionat x = 10. Moreover,
lim
x
f (x) = and lim
x
f (x) = .
5 5 10
5 10
6
5 10
6
1 10
7
y
x
y = x
6
9x
4
25. y = x 4
x
solution Let f (x) = x 4
x = x 4x
1/2
. Thenf
(x) = 1 2x
1/2
. This shows that f has critical points at
x = 0(wherethederivativedoesnotexist) andatx = 4(wherethederivativeiszero). Becausef
(x) < 0for 0< x < 4

andf
(x) > 0for x > 4, f (4) is alocal minimum. Nowf
(x) = x
3/2
> 0for all x > 0, so thegraphis always
concaveup. Moreover,
lim
x
f (x) = .
2
2
4
5 10 15 20
y
x
y =
x +
16x
27. y = x(8x)
1/3
solution Let f (x) = x (8x)
1/3
. Then
f
(x) = x
1
3
(8x)
2/3
(1) +(8x)
1/3
1=
244x
3(8x)
2/3
andsimilarly
f
(x) =
4x 48
9(8x)
5/3
.
Critical pointsareat x = 8andx = 6, andcandidateinectionpointsarex = 8andx = 12. Signanalysesreveal that
f
(x) changesfrompositivetonegativeat x = 6andf
(x) remainsnegativeoneither sideof x = 8. Moreover, f
(x)
changesfromnegativetopositiveat x = 8andfrompositivetonegativeat x = 12. Therefore, f hasalocal maximum
at x = 6andinectionpointsat x = 8andx = 12. Moreover,
lim
x
f (x) = .
Hereisagraphof f withthetransitionpointshighlighted.
30
5 5 10 15
20
10
y
x
y = (x
2
4x)
1/3
29. y = (2x x
2
)
1/3
solution Wehave
f
(x) =
1
3
(2x x
2
)
2/3
(22x) =
22x
3(2x x
2
)
2/3
f
(x) =
2(x
2
2x +4)
9x(x 2)(2x x
2
)
2/3
Theonlycritical pointisatx = 1, andsincethenumerator of f
(x) isalwayspositive, therearenoinectionpoints. Sign

analysisshowsthat f
(x) changesfrompositivetonegativeat x = 1, sothat f hasalocal maximumat x = 1. Finally,

lim
x
f (x) =
sothat thegraphhasnohorizontal asymptotes. Hereisagraphof f withthetransitionpointshighlighted:
1
10 5 5 10
x
y
1
2
3
4
y = (x
3
3x)
1/3
31. y = x x
1
solution Wehavef
(x) = 1+ x
2
andf
(x) = 2x
3
. Sincef (x) never vanishes, therearenocritical points.
Sincef
(x) = 0onlyat x = 0, therearenoinectionpoints. Finally,

lim
x
f (x) = , lim
x0+
f (x) = , lim
x0
f (x) =
sothat thegraphhasavertical asymptoteat x = 0but nohorizontal asymptotes. Hereisagraphof f :
3 1 1 2 3
x
y
2
4
6
8
4
6
8
y = x
2
x
2 33. y = x
3
48
x
2
solution Wehavef
(x) = 3x
2
+ 96x
3
andf
(x) = 6x 288x
4
. Thecritical points occur for 0 = f
(x) =
3x
2
+ 96x
3
, sowhere3x
5
= 96. Thustheonly critical point isx = 2. f
(x) vanisheswhen6x
5
= 288, sothat
x
5
= 48andx =
5
48. Signanalysisshowsthat f
(x) changesfrompositivetonegativeat x = 2, sothat x = 2isa

local maximum, but doesnot changesign, remainingpositive, at x = 2. Thusx = 2isnot alocal extremum. Moreover,
f
(x) changesfromnegativetopositiveat x =
5
48, sothat thisisapoint of inection. Finally,

lim
x
f (x) = , lim
x0
f (x) =
sothat thegraphhasavertical asymptoteat x = 0but nohorizontal asymptotes. Hereisagraphof f withthetransition
pointshighlighted:
4 3 1 2 3 4
y
x
100
80
60
40
20
20
40
60
y = x
2
x +x
1
35. Sketchthegraphof f (x) = 18(x 3)(x 1)
2/3
usingtheformulas
f
(x) =
30
_
x
9
5
_
(x 1)
1/3
, f
(x) =
20
_
x
3
5
_
(x 1)
4/3
solution
f
(x) =
30(x
9
5
)
(x 1)
1/3
yieldscritical pointsat x =
9
5
, x = 1.
f
(x) =
20(x
3
5
)
(x 1)
4/3
yieldspotential inectionpointsat x =
3
5
, x = 1.
Interval signsof f
andf
(,
3
5
) +
(
3
5
, 1) ++
(1,
9
5
) +
(
9
5
, ) ++
Thegraph has an inection point at x =
3
5
, alocal maximumat x = 1 (at which thegraph has acusp), and alocal
minimumat x =
9
5
. Thesketchlookssomethinglikethis.
40
x
3 2 1 2 1
y
20
20
40
60
80
Sketchthegraphof f (x) =
x
x
2
+1
usingtheformulas
f
(x) =
1x
2
(1+x
2
)
2
, f
(x) =
2x(x
2
3)
(x
2
+1)
3
In Exercises 3740, sketch the graph of the function, indicating all transition points. If necessary, use a graphing
utility or computer algebra system to locate the transition points numerically.
37. y = x
3
4
x
2
+1
solution Wehave
f
(x) = 3x
2
+
8x
(x
2
+1)
2
, f
(x) = 6x
8(3x
2
1)
(x
2
+1)
3
Thecritical points aretheroots of f
(x), which arex = 0 and x = 0.8678. Candidates for inection points occur
when f
(x) vanishes; theonly root of f
(x) is 0.41119. Sign analysis reveals that f
(x) changes frompositiveto

negativeat x = 0.8678andfromnegativeto positiveat x = 0. Also, f
(x) changes signfromnegativeto positive

at x = 0.41119. Thus f has alocal maximumat x = 0.8678, apoint of inectionat x = 0.41119, andalocal
minimumat x = 0. Moreover, lim
x
= , sotherearenohorizontal asymptotes. Sincef (x) isdenedeverywhere,
therearenovertical asymptotes. Hereisagraphof f withthetransitionpointshighlighted:
2
2
2
6
1 1 2
x
y
y = 12
_
x
2
+2x +4x
2
39. y = x
4
4x
2
+x +1
4
4x
2
+ x + 1. Thenf
(x) = 4x
3
8x + 1andf
(x) = 12x
2
8. Thecritical points
arex = 1.473, x = 0.126andx = 1.347, whilethecandidates for points of inectionarex =
_
2
3
. Signanalysis
reveals that f
(x) changes fromnegativeto positiveat x = 1.473, frompositiveto negativeat x = 0.126andfrom

negativetopositiveat x = 1.347. For thesecondderivative, f
(x) changesfrompositivetonegativeat x =
_
2
3
and
fromnegativetopositiveat x =
_
2
3
. Therefore, f haslocal minimaat x = 1.473andx = 1.347, alocal maximumat
x = 0.126andpointsof inectionat x =
_
2
3
. Moreover,
lim
x
f (x) = .
Hereisagraphof f withthetransitionpointshighlighted.
2 1 1 2
5
5
10
15
20
y
x
y = 2
x sinx, 0 x 2
In Exercises 4146, sketch the graph over the given interval, with all transition points indicated.
41. y = x +sinx, [0, 2]
solution Let f (x) = x +sinx. Settingf
(x) = 1+cosx = 0yieldscosx = 1, sothat x = isthelonecritical

point ontheinterval [0, 2]. Settingf
(x) = sinx = 0yields potential points of inectionat x = 0, , 2 onthe

interval [0, 2].
Interval signsof f
andf
(0, ) +
(, 2) ++
Thegraphhasaninectionpoint at x = , andnolocal maximaor minima. Hereisasketchof thegraphof f (x):
y
x
1
2
3
4
5
6
0 6 5 4 3 2 1
y = sinx +cosx, [0, 2]
43. y = 2sinx cos
2
x, [0, 2]
solution Let f (x) = 2sinx cos
2
x. Thenf
(x) = 2cosx 2cosx (sinx) = sin2x + 2cosx andf
(x) =
2cos2x 2sinx. Settingf
(x) = 0yieldssin2x = 2cosx, sothat 2sinx cosx = 2cosx. Thisimpliescosx = 0

or sinx = 1, sothat x =

2
or
3
2
. Settingf
(x) = 0yields2cos2x = 2sinx, sothat 2sin(
2
2x) = 2sinx, or
2
2x = x 2n. Thisyields3x =

2
+2n, or x =

6
,
5
6
,
9
6
=
3
2
.
Interval signsof f
andf
_
0,

6
_
++
_
6
,

2
_
+
_
2
,
5
6
_

_
5
6
,
3
2
_
+
_
3
2
, 2
_
++
Thegraphhas alocal maximumat x =

2
, alocal minimumat x =
3
2
, andinectionpoints at x =

6
andx =
5
6
.
Hereisagraphof f without transitionpointshighlighted.
x
6 5 4
3
2 1
y
1
2
2
1
y = sinx +
1
2
x, [0, 2]
45. y = sinx +
3cosx, [0, ]
solution Let f (x) = sinx +
3cosx. Settingf
(x) = cosx
3sinx = 0yields tanx =

1
3
. Intheinterval
[0, ], thesolutionisx =

6
. Settingf
(x) = sinx
3cosx = 0yieldstanx =
3. Intheinterval [0, ], the

lonesolutionisx =
2
3
.
Interval signsof f
andf
(0, /6) +
(/6, 2/3)
(2/3, ) +
Thegraph has alocal maximumat x =

6
and apoint of inection at x =
2
3
. A plot without thetransition points
highlightedisgivenbelow:
x
3 2 1
y
1
2
2
1
y = sinx
1
2
sin2x, [0, ]
47. Areall signtransitionspossible? Explainwithasketchwhythetransitions++ + and + do
not occur if thefunctionisdifferentiable. (SeeExercise76for aproof.)
solution Inbothcases, thereisapoint wheref isnot differentiableat thetransitionfromincreasingtodecreasingor
decreasingtoincreasing.
y
x
y
x
Supposethat f istwicedifferentiablesatisfying(i) f (0) = 1, (ii) f
(x) > 0for all x = 0, and(iii) f
(x) < 0
for x < 0andf
(x) > 0for x > 0. Let g(x) = f (x

2
).
(a) Sketchapossiblegraphof f (x).
(b) Provethatg(x) hasnopointsof inectionandauniquelocal extremevalueatx = 0. Sketchapossiblegraphof
g(x).
49. Whichof thegraphsinFigure17cannot bethegraphof apolynomial? Explain.
(A) (B) (C)
x
x
x
y y y
FIGURE 17
solution Polynomials areeverywheredifferentiable. Accordingly, graph (B) cannot bethegraph of apolynomial,
sincethefunctionin(B) hasacusp(sharpcorner), signifyingnondifferentiabilityat that point.
WhichcurveinFigure18isthegraphof f (x) =
2x
4
1
1+x
4
? Explainonthebasisof horizontal asymptotes.
51. MatchthegraphsinFigure19withthetwofunctionsy =
3x
x
2
1
andy =
3x
2
x
2
1
. Explain.
(A) (B)
1 1 1 1
x x
y y
FIGURE 19
solution Since lim
x
3x
2
x
2
1
=
3
1
lim
x
1= 3, thegraphof y =
3x
2
x
2
1
hasahorizontal asymptoteof y = 3;
hence, theright curveisthegraphof f (x) =
3x
2
x
2
1
. Since
lim
x
3x
x
2
1
=
3
1
lim
x
x
1
= 0,
thegraphof y =
3x
x
2
1
hasahorizontal asymptoteof y = 0; hence, theleft curveisthegraphof f (x) =
3x
x
2
1
.
Matchthefunctionswiththeir graphsinFigure20.
(a) y =
1
x
2
1
(b) y =
x
2
x
2
+1
(c) y =
1
x
2
+1
(d) y =
x
x
2
1
In Exercises 5370, sketch the graph of the function. Indicate the transition points and asymptotes.
53. y =
1
3x 1
1
3x 1
. Thenf
(x) =
3
(3x 1)
2
, sothat f isdecreasingfor all x =
1
3
. Moreover, f
(x) =
18
(3x 1)
3
, so that f is concaveup for x >
1
3
and concavedown for x <
1
3
. Because lim
x
1
3x 1
= 0, f has a
horizontal asymptoteat y = 0. Finally, f hasavertical asymptoteat x =
1
3
with
lim
x
1
3
1
3x 1
= and lim
x
1
3
+
1
3x 1
= .
5
2 2
5
y
x
y =
x 2
x 3
55. y =
x +3
x 2
x +3
x 2
. Thenf
(x) =
5
(x 2)
2
, so that f is decreasingfor all x = 2. Moreover, f
(x) =
10
(x 2)
3
, sothatf isconcaveupfor x > 2andconcavedownfor x < 2. Because lim
x
x +3
x 2
= 1, f hasahorizontal
asymptoteat y = 1. Finally, f hasavertical asymptoteat x = 2with
lim
x2
x +3
x 2
= and lim
x2+
x +3
x 2
= .
x
10 5 10 5
y
10
5
10
5
y = x +
1
x
57. y =
1
x
+
1
x 1
solution Letf (x) =
1
x
+
1
x 1
. Thenf
(x) =
2x
2
2x +1
x
2
(x 1)
2
, sothatf isdecreasingfor all x = 0, 1. Moreover,
f
(x) =
2
_
2x
3
3x
2
+3x 1
_
x
3
(x 1)
3
, so that f is concaveupfor 0 < x <
1
2
andx > 1andconcavedownfor x < 0
and
1
2
< x < 1. Because lim
x
_
1
x
+
1
x 1
_
= 0, f has ahorizontal asymptoteat y = 0. Finally, f has vertical
asymptotesat x = 0andx = 1with
lim
x0
_
1
x
+
1
x 1
_
= and lim
x0+
_
1
x
+
1
x 1
_
=
and
lim
x1
_
1
x
+
1
x 1
_
= and lim
x1+
_
1
x
+
1
x 1
_
= .
x
1 2 1
y
5
5
y =
1
x
1
x 1
59. y =
1
x(x 2)
1
x(x 2)
. Thenf
(x) =
2(1x)
x
2
(x 2)
2
, so that f is increasingfor x < 0and0 < x < 1and
decreasingfor 1< x < 2andx > 2. Moreover, f
(x) =
2(3x
2
6x +4)
x
3
(x 2)
3
, sothatf isconcaveupfor x < 0andx > 2
andconcavedownfor 0< x < 2. Because lim
x
_
1
x(x 2)
_
= 0, f hasahorizontal asymptoteat y = 0. Finally, f
hasvertical asymptotesat x = 0andx = 2with
lim
x0
_
1
x(x 2)
_
= + and lim
x0+
_
1
x(x 2)
_
=
and
lim
x2
_
1
x(x 2)
_
= and lim
x2+
_
1
x(x 2)
_
= .
5
5
2 4
y
x
y =
x
x
2
9
61. y =
1
x
2
6x +8
1
x
2
6x +8
=
1
(x 2) (x 4)
. Then f
(x) =
62x
_
x
2
6x +8
_
2
, so that f is increasing for
x < 2andfor 2 < x < 3, is decreasingfor 3 < x < 4andfor x > 4, andhas alocal maximumat x = 3. Moreover,
f
(x) =
2
_
3x
2
18x +28
_
_
x
2
6x +8
_
3
, sothat f is concaveupfor x < 2andfor x > 4andis concavedownfor 2 < x < 4.
Because lim
x
1
x
2
6x +8
= 0, f hasahorizontal asymptoteat y = 0. Finally, f hasvertical asymptotesat x = 2
andx = 4, with
lim
x2
_
1
x
2
6x +8
_
= and lim
x2+
_
1
x
2
6x +8
_
=
and
lim
x4
_
1
x
2
6x +8
_
= and lim
x4+
_
1
x
2
6x +8
_
= .
x
6 5 4 2
3
1
y
5
5
y =
x
3
+1
x
63. y = 1
3
x
+
4
x
3
3
x
+
4
x
3
. Then
f
(x) =
3
x
2
12
x
4
=
3(x 2)(x +2)
x
4
,
sothat f isincreasingfor |x| > 2anddecreasingfor 2< x < 0andfor 0< x < 2. Moreover,
f
(x) =
6
x
3
+
48
x
5
=
6(8x
2
)
x
5
,
so that f is concave down for 2
2 < x < 0 and for x > 2
2, while f is concave up for x < 2
2 and for
0< x < 2
2. Because
lim
x
_
1
3
x
+
4
x
3
_
= 1,
f hasahorizontal asymptoteat y = 1. Finally, f hasavertical asymptoteat x = 0with
lim
x0
_
1
3
x
+
4
x
3
_
= and lim
x0+
_
1
3
x
+
4
x
3
_
= .
2
x
2 4 6 6 4 2
y
4
6
6
4
2
y =
1
x
2
+
1
(x 2)
2
65. y =
1
x
2
1
(x 2)
2
solution Letf (x) =
1
x
2
1
(x 2)
2
. Thenf
(x) = 2x
3
+2(x 2)
3
, sothatf isincreasingfor x < 0andfor
x > 2andisdecreasingfor 0< x < 2. Moreover,
f
(x) = 6x
4
6(x 2)
4
=
48(x 1)(x
2
2x +2)
x
4
(x 2)
4
,
sothat f is concaveupfor x < 0andfor 0 < x < 1, is concavedownfor 1 < x < 2andfor x > 2, andhas apoint
of inectionat x = 1. Because lim
x
_
1
x
2
1
(x 2)
2
_
= 0, f has ahorizontal asymptoteat y = 0. Finally, f has
vertical asymptotesat x = 0andx = 2with
lim
x0
_
1
x
2
1
(x 2)
2
_
= and lim
x0+
_
1
x
2
1
(x 2)
2
_
=
and
lim
x2
_
1
x
2
1
(x 2)
2
_
= and lim
x2+
_
1
x
2
1
(x 2)
2
_
= .
x
2
4
2
4
y
1 2 3 4
2 1
y =
4
x
2
9
67. y =
1
(x
2
+1)
2
solution Letf (x) =
1
(x
2
+1)
2
. Thenf
(x) =
4x
(x
2
+1)
3
, sothatf isincreasingfor x < 0, isdecreasingfor x > 0
andhasalocal maximumat x = 0. Moreover,
f
(x) =
4(x
2
+1)
3
+4x 3(x
2
+1)
2
2x
(x
2
+1)
6
=
20x
2
4
(x
2
+1)
4
,
sothat f isconcaveupfor |x| > 1/
5, isconcavedownfor |x| < 1/
5, andhaspointsof inectionat x = 1/
5.
Because lim
x
1
(x
2
+1)
2
= 0, f hasahorizontal asymptoteat y = 0. Finally, f hasnovertical asymptotes.
x
1
0.8
4 2 2 4
y
y =
x
2
(x
2
1)(x
2
+1)
69. y =
1
_
x
2
+1
1
x
2
+1
. Then
f
(x) =
x
_
(x
2
+1)
3
= x(x
2
+1)
3/2
,
sothat f isincreasingfor x < 0anddecreasingfor x > 0. Moreover,
f
(x) =
3
2
x(x
2
+1)
5/2
(2x) (x
2
+1)
3/2
= (2x
2
1)(x
2
+1)
5/2
,
sothat f isconcavedownfor |x| <
2
2
andconcaveupfor |x| >
2
2
. Because
lim
x
1
_
x
2
+1
= 0,
f hasahorizontal asymptoteat y = 0. Finally, f hasnovertical asymptotes.
x
1
0.8
0.2
10 5 5 10
y
y =
x
_
x
2
+1
In Exercises 7175, we explore functions whose graphs approach a nonhorizontal line as x . A line y = ax +b is
called a slant asymptoteif
lim
x
(f (x) (ax +b)) = 0
or
lim
x
(f (x) (ax +b)) = 0
71. Let f (x) =
x
2
x 1
(Figure21). Verifythefollowing:
(a) f (0) isalocal maxandf (2) alocal min.
(b) f isconcavedownon(, 1) andconcaveupon(1, ).
(c) lim
x1
f (x) = and lim
x1+
f (x) = .
(d) y = x +1isaslant asymptoteof f (x) asx .
(e) Theslant asymptoteliesabovethegraphof f (x) for x < 1andbelowthegraphfor x > 1.
y = x + 1
10 10
10
10
x
y
f (x) =
x
2
x 1
FIGURE 21
x
2
x 1
. Thenf
(x) =
x(x 2)
(x 1)
2
andf
(x) =
2
(x 1)
3
.
(a) Signanalysisof f
(x) revealsthat f
(x) < 0on(, 1) andf
(x) > 0on(1, ).

(b) Critical pointsof f
(x) occur at x = 0andx = 2. x = 1isnot acritical point becauseit isnot inthedomainof f .

Signanalysesreveal that x = 2isalocal minimumof f andx = 0isalocal maximum.
(c)
lim
x1
f (x) = 1 lim
x1
1
1x
= and lim
x1+
f (x) = 1 lim
x1+
1
x 1
= .
(d) Note that using polynomial division, f (x) =
x
2
x 1
= x + 1 +
1
x 1
. Then
lim
x
(f (x) (x +1)) = lim
x
x +1+
1
x 1
(x +1) = lim
x
1
x 1
= 0.
(e) For x > 1, f (x) (x +1) =
1
x 1
> 0, sof (x) approachesx +1fromabove. Similarly, for x < 1, f (x) (x +
1) =
1
x 1
< 0, sof (x) approachesx +1frombelow.
If f (x) = P(x)/Q(x), whereP andQ arepolynomials of degrees m + 1andm, thenby longdivision, we
canwrite
f (x) = (ax +b) +P
1
(x)/Q(x)
whereP
1
isapolynomial of degree< m. Showthat y = ax +b istheslant asymptoteof f (x). Usethisprocedure
tondtheslant asymptotesof thefollowingfunctions:
(a) y =
x
2
x +2
(b) y =
x
3
+x
x
2
+x +1
73. Sketchthegraphof
f (x) =
x
2
x +1
.
Proceedasinthepreviousexercisetondtheslant asymptote.
x
2
x +1
. Thenf
(x) =
x(x +2)
(x +1)
2
andf
(x) =
2
(x +1)
3
. Thus, f isincreasingfor x < 2and
for x > 0, isdecreasingfor 2< x < 1andfor 1< x < 0, hasalocal minimumat x = 0, hasalocal maximumat
x = 2, isconcavedownon(, 1) andconcaveupon(1, ). Limitanalysesgiveavertical asymptoteatx = 1,
with
lim
x1
x
2
x +1
= and lim
x1+
x
2
x +1
= .
Bypolynomial division, f (x) = x 1+
1
x +1
and
lim
x
_
x 1+
1
x +1
(x 1)
_
= 0,
whichimpliesthat theslant asymptoteisy = x 1. Noticethat f approachestheslant asymptoteasinexercise71.
x
4
2
4 2 2 4
4
6
2
y
Show that y = 3x is aslant asymptotefor f (x) = 3x +x
2
. Determinewhether f (x) approaches theslant
asymptotefromaboveor belowandmakeasketchof thegraph.
75. Sketchthegraphof f (x) =
1x
2
2x
.
1x
2
2x
. Usingpolynomial division, f (x) = x +2+
3
x 2
. Then
lim
x
(f (x) (x +2)) = lim
x
_
(x +2) +
3
x 2
(x +2)
_
= lim
x
3
x 2
=
3
1
lim
x
x
1
= 0
which implies that y = x + 2is theslant asymptoteof f (x). Sincef (x) (x + 2) =
3
x 2
> 0for x > 2, f (x)
approaches theslant asymptotefromabovefor x > 2; similarly,
3
x 2
< 0for x < 2so f (x) approaches theslant
asymptotefrombelowfor x < 2. Moreover, f
(x) =
x
2
4x +1
(2x)
2
andf
(x) =
6
(2x)
3
. Signanalysesreveal alocal
minimumat x = 2+
3, alocal maximumat x = 2
3andthat f isconcavedownon(, 2) andconcaveupon

(2, ). Limit analysesgiveavertical asymptoteat x = 2.
x
10
5
10 5 5 10
10
5
y
Assumethat f
(x) andf
(x) exist for all x andlet c beacritical point of f (x). Showthat f (x) cannot makea
transitionfrom++ to+ at x = c. Hint: ApplytheMVT tof
(x).
77. Assumethat f
(x) existsandf
(x) > 0for all x. Showthat f (x) cannot benegativefor all x. Hint: Show
that f
(b) = 0for someb andusetheresult of Exercise56inSection4.4.

solution Letf (x) beafunctionsuchthatf
(x) existsandf
(x) > 0for all x. Sincef
(x) > 0, thereisatleastone

point x = b suchthat f
(b) = 0. If not, f
(x) = 0for all x, sof
(x) = 0. Bytheresult of Exercise56inSection4.4,

f (x) f (b) +f
(b)(x b). Now, if f
(b) > 0, wendthat f (b) +f
(b)(x b) > 0whenever

x >
bf
(b) f (b)
f
(b)
,
aconditionthat must bemet for somex sufciently large. For suchx, f (x) > f (b) + f
(b)(x b) > 0. Ontheother

hand, if f
(b) < 0, wendthat f (b) +f
(b)(x b) > 0whenever

x <
bf
(b) f (b)
f
(b)
.
For suchanx, f (x) > f (b) +f
(b)(x b) > 0.
4.6 Applied Optimization
1. Theproblemis to ndtheright triangleof perimeter 10whoseareais as largeas possible. What is theconstraint
equationrelatingthebaseb andheight h of thetriangle?
solution Theperimeter of aright triangleisthesumof thelengthsof thebase, theheight andthehypotenuse. If the
basehaslengthb andtheheight ish, thenthelengthof thehypotenuseis
_
b
2
+h
2
andtheperimeter of thetriangleis
P = b +h +
_
b
2
+h
2
. Therequirement that theperimeter be10translatestotheconstraint equation
b +h +
_
b
2
+h
2
= 10.
2. Describeawayof showingthat acontinuousfunctiononanopeninterval (a, b) hasaminimumvalue.
solution If thefunctiontendstoinnity at theendpointsof theinterval, thenthefunctionmust takeonaminimum
valueat acritical point.
3. Istherearectangleof area100of largest perimeter? Explain.
solution No. Evenbyxingtheareaat 100, wecantakeoneof thedimensionsaslargeasweliketherebyallowing
theperimeter tobecomeaslargeaswelike.
Exercises
1. Findthedimensionsx andy of therectangleof maximumareathat canbeformedusing3metersof wire.
(a) What istheconstraint equationrelatingx andy?
(b) Findaformulafor theareaintermsof x alone.
(c) What istheinterval of optimization? Isit openor closed?
(d) Solvetheoptimizationproblem.
solution
(a) Theperimeter of therectangleis3meters, so3= 2x +2y, whichisequivalent toy =
3
2
x.
(b) Usingpart (a), A = xy = x(
3
2
x) =
3
2
x x
2
.
(c) Thisproblemrequiresoptimizationover theclosedinterval [0,
3
2
], sincebothx andy must benon-negative.
(d) A
(x) =
3
2
2x = 0, whichyields x =
3
4
andconsequently, y =
3
4
. BecauseA(0) = A(3/2) = 0andA(
3
4
) =
0.5625, themaximumarea0.5625m
2
isachievedwithx = y =
3
4
m.
Wireof length12misdividedintotwopiecesandeachpieceisbent intoasquare. Howshouldthisbedonein
order tominimizethesumof theareasof thetwosquares?
(a) Expressthesumof theareasof thesquaresintermsof thelengthsx andy of thetwopieces.
(b) What istheconstraint equationrelatingx andy?
(c) What istheinterval of optimization? Isit openor closed?
S E C T I ON 4.6 Applied Optimization 221
3. Wireof length12misdividedintotwopiecesandthepiecesarebent intoasquareandacircle. Howshouldthisbe
doneinorder tominimizethesumof their areas?
solution Supposethewireis divided into onepieceof length x mthat is bent into acircleand apieceof length
12x mthatisbentintoasquare. Becausethecirclehascircumferencex, itfollowsthattheradiusof thecircleisx/2;
therefore, theareaof thecircleis
_
x
2
_
2
=
x
2
4
.
Asfor thesquare, becausetheperimeter is12x, thelengthof eachsideis3x/4andtheareais(3x/4)
2
. Then
A(x) =
x
2
4
+
_
3
1
4
x
_
2
.
Now
A
(x) =
x
2
1
2
_
3
1
4
x
_
= 0
when
x =
12
4+
m 5.28m.
BecauseA(0) = 9m
2
, A(12) = 36/ 11.46m
2
, and
A
_
12
4+
_
5.04m
2
,
weseethat thesumof theareasisminimizedwhenapproximately5.28mof thewireisallottedtothecircle.
Findthepositivenumber x suchthat thesumof x andits reciprocal is as small as possible. Does this problem
requireoptimizationover anopeninterval or aclosedinterval?
5. A exibletubeof length 4 mis bent into an L-shape. Whereshould thebend bemadeto minimizethedistance
betweenthetwoends?
solution Let x, y > 0belengthsof thesideof theL. Sincex +y = 4or y = 4x, thedistancebetweentheends
of L ish(x) =
_
x
2
+y
2
=
_
x
2
+(4x)
2
. Wemayequivalentlyminimizethesquareof thedistance,
f (x) = x
2
+y
2
= x
2
+(4x)
2
This is easier computationally (whenworkingby hand). Solvef
(x) = 4x 8 = 0to obtainx = 2m. Nowf (0) =

f (4) = 16, whereasf (2) = 8. Hencethedistancebetweenthetwoendsof theL isminimizedwhenthebendismadeat
themiddleof thewire.
Findthedimensionsof theboxwithsquarebasewith:
(a) Volume12andtheminimal surfacearea.
(b) Surfacearea20andmaximal volume.
7. A rancher will use600mof fencingto buildacorral intheshapeof asemicircleontopof arectangle(Figure9).
Findthedimensionsthat maximizetheareaof thecorral.
FIGURE 9
solution Let x bethewidthof thecorral andthereforethediameter of thesemicircle, andlet y betheheight of the
rectangular section. Thentheperimeter of thecorral canbeexpressedbytheequation2y +x +

2
x = 2y +(1+

2
)x =
600 mor equivalently, y =
1
2
_
600(1+

2
)x
_
. Since x and y must both be nonnegative, it follows that x must
berestricted to theinterval [0,
600
1+/2
]. Theareaof thecorral is thesumof theareaof therectangleand semicircle,
A = xy +

8
x
2
. Makingthesubstitutionfor y fromtheconstraint equation,
A(x) =
1
2
x
_
600(1+

2
)x
_
+

8
x
2
= 300x
1
2
_
1+

2
_
x
2
+

8
x
2
.
Now, A
(x) = 300
_
1+

2
_
x +

4
x = 0impliesx =
300
_
1+
4
_
168.029746m. WithA(0) = 0m
2
,
A
_
300
1+/4
_
25204.5m
2
and A
_
600
1+/2
_
21390.8m
2
,
it followsthat thecorral of maximumareahasdimensions
x =
300
1+/4
m and y =
150
1+/4
m.
What isthemaximumareaof arectangleinscribedinaright trianglewith5and8asinFigure10. Thesidesof
therectangleareparallel tothelegsof thetriangle.
9. Findthedimensionsof therectangleof maximumareathat canbeinscribedinacircleof radiusr = 4(Figure11).
r
FIGURE 11
solution Placethecenter of thecircleat theoriginwiththesides of therectangle(of lengths 2x > 0and2y > 0)
parallel tothecoordinateaxes. BythePythagoreanTheorem, x
2
+y
2
= r
2
= 16, sothat y =
_
16x
2
. Thusthearea
of therectangleisA(x) = 2x 2y = 4x
_
16x
2
. Toguaranteebothx andy arereal andnonnegative, wemust restrict
x totheinterval [0, 4]. Solve
A
(x) = 4
_
16x
2
4x
2
_
16x
2
= 0
for x > 0toobtainx =
4
2
= 2
2. SinceA(0) = A(4) = 0andA(2
2) = 32, therectangleof maximumareahas

dimensions2x = 2y = 4
2.
Findthedimensionsx andy of therectangleinscribedinacircleof radiusr that maximizesthequantityxy
2
.
11. Findthepoint ontheliney = x closest tothepoint (1, 0). Hint: It isequivalent andeasier tominimizethesquare
of thedistance.
solution Withy = x, letsequivalentlyminimizethesquareof thedistance, f (x) = (x 1)
2
+y
2
= 2x
2
2x +1,
whichiscomputationallyeasier (whenworkingbyhand). Solvef
(x) = 4x 2= 0toobtainx =
1
2
. Sincef (x)
asx , (
1
2
,
1
2
) isthepoint ony = x closest to(1, 0).
Findthepoint P ontheparabolay = x
2
closest tothepoint (3, 0) (Figure12).
13. Findthecoordinatesof thepoint onthegraphof y = x +2x
1
closest totheoriginintheregionx > 0(Figure13).
1 2 3
2
4
6
8
y = x + 2x
1
y
x
FIGURE 13
solution The distance from the origin to the point (x, x + 2x
1
) on the graph of y = x + 2x
1
is d =
_
x
2
+(x +2x
1
)
2
. Asusual, wewill minimized
2
. Let d
2
= f (x) = x
2
+(x +2x
1
)
2
. Then
f
(x) = 2x +2(x +2x

1
)(12x
2
) = 4x 8x
3
.
Todeterminex, weneedtosolvef
(x) = 0. Multiplyingthroughby x
3
gives4x
4
8 = 0, sothat x =
4
2 1.189.
Sincef (1.189) 2.871, thepoint ony = x +2x
1
that isclosest totheoriginisapproximately(1.189, 2.871).
Problemof Tartaglia(15001557) Amongall positivenumbersa, b whosesumis8, ndthosefor whichthe
product of thetwonumbersandtheir differenceislargest.
15. Findtheangle that maximizestheareaof theisoscelestrianglewhoselegshavelength (Figure14).
q
FIGURE 14
solution Theareaof thetriangleis
A() =
1
2
2
sin,
where0 . Setting
A
() =
1
2
2
cos = 0
yields =

2
. SinceA(0) = A() = 0andA(
2
) =
1
2
2
, theanglethat maximizestheareaof theisoscelestriangleis
=

2
.
Arightcircularcone(Figure15)hasvolumeV =

3
r
2
h andsurfaceareaisS = r
_
r
2
+h
2
. Findthedimensions
of theconewithsurfacearea1andmaximal volume.
17. Findtheareaof thelargest isoscelestrianglethat canbeinscribedinacircleof radiusr.
solution Consider thefollowingdiagram:
q
p q p q
2q
r
r r
Theareaof theisoscelestriangleis
A() = 2
1
2
r
2
sin( ) +
1
2
r
2
sin(2) = r
2
sin +
1
2
r
2
sin(2),
where0 . Solve
A
() = r
2
cos +r
2
cos(2) = 0
toobtain =

3
, . SinceA(0) = A() = 0andA(
3
) =
3
3
4
r
2
, theareaof thelargest isoscelestrianglethat canbe
inscribedinacircleof radiusr is
3
3
4
r
2
.
Findtheradiusandheight of acylindrical canof total surfaceareaA whosevolumeisaslargeaspossible. Does
thereexist acylinder of surfaceareaA andminimal total volume?
19. A poster of area6000cm
2
hasblank marginsof width10cmonthetopandbottomand6cmonthesides. Findthe
dimensionsthat maximizetheprintedarea.
solution Letx bethewidthof theprintedregion, andlety betheheight. Thetotal printedareaisA = xy. Becausethe
total areaof theposter is6000cm
2
, wehavetheconstraint (x +12)(y +20) = 6000, sothat xy +12y +20x +240=
6000, or y =
576020x
x+12
. Therefore, A(x) = 20
288xx
2
x+12
, where0 x 288.
A(0) = A(288) = 0, sowearelookingfor acritical point ontheinterval [0, 288]. SettingA
(x) = 0yields
20
(x +12)(2882x) (288x x
2
)
(x +12)
2
= 0
x
2
24x +3456
(x +12)
2
= 0
x
2
+24x 3456= 0
(x 48)(x +72) = 0
Thereforex = 48or x = 72. x = 48istheonlycritical point of A(x) intheinterval [0, 288], soA(48) = 3840isthe
maximumvalueof A(x) intheinterval [0, 288]. Now, y = 20
28848
48+12
= 80cm, sotheposter withmaximumprintedarea
is48+12= 60cm. wideby80+20= 100cm. tall.
Accordingtopostal regulations, acartonisclassiedasoversized if thesumof itsheight andgirth( perimeter
of itsbase) exceeds108in. Findthedimensionsof acartonwithsquarebasethat isnot oversizedandhasmaximum
volume.
21. KeplersWineBarrel Problem In his work Nova stereometria doliorum vinariorum (NewSolid Geometry of
a Wine Barrel), published in 1615, astronomer J ohannes Kepler stated and solved the following problem: Find the
dimensionsof thecylinder of largest volumethat canbeinscribedinasphereof radiusR. Hint: Showthat aninscribed
cylinder hasvolume2x(R
2
x
2
), wherex isone-half theheight of thecylinder.
solution Placethecenter of thesphereat theorigininthree-dimensional space. Let thecylinder beof radiusy and
half-height x. ThePythagoreanTheoremstates, x
2
+ y
2
= R
2
, so that y
2
= R
2
x
2
. Thevolumeof thecylinder is
V(x) = y
2
(2x) = 2
_
R
2
x
2
_
x = 2R
2
x 2x
3
. Allowing for degeneratecylinders, wehave0 x R.
SolveV
(x) = 2R
2
6x
2
= 0 for x 0 to obtain x =
R
3
. SinceV(0) = V(R) = 0, thelargest volumeis
V(
R
3
) =
4
9
3R
3
whenx =
R
3
andy =
_
2
3
R.
Find theangle that maximizes theareaof thetrapezoid with abaseof length 4 and sides of length 2, as in
Figure16.
23. A landscapearchitect wishes toenclosearectangular gardenof area1,000m
2
ononesideby abrick wall costing
$90/mandontheother threesidesbyametal fencecosting$30/m. Whichdimensionsminimizethetotal cost?
solution Let x bethelengthof thebrickwall andy thelengthof anadjacent sidewithx, y > 0. Withxy = 1000or
y =
1000
x
, thetotal cost is
C(x) = 90x +30(x +2y) = 120x +60,000x
1
.
SolveC
(x) = 12060,000x
2
= 0for x > 0toobtainx = 10
5. SinceC(x) asx 0+ andasx , the

minimumcost isC(10
5) = 2400
5 $5366.56whenx = 10
5 22.36mandy = 20
5 44.72m.
Theamount of light reachingapoint at adistancer fromalight sourceA of intensityI
A
isI
A
/r
2
. Supposethat
asecondlight sourceB of intensityI
B
= 4I
A
islocated10mfromA. Findthepoint onthesegment joiningA and
B wherethetotal amount of light isat aminimum.
25. Find themaximumareaof arectangleinscribed in theregion bounded by thegraph of y =
4x
2+x
and theaxes
(Figure17).
2
4
y =
4 x
2 + x
x
y
FIGURE 17
solution Let s bethewidthof therectangle. Theheight of therectangleish =
4s
2+s
, sothat theareais
A(s) = s
4s
2+s
=
4s s
2
2+s
.
Wearemaximizingontheclosedinterval [0, 4]. It is obvious fromthepictures that A(0) = A(4) = 0, sowelook for
critical pointsof A.
A
(s) =
(2+s)(42s) (4s s
2
)
(2+s)
2
=
s
2
+4s 8
(s +2)
2
.
Theonlypoint whereA
(s) doesnt exist iss = 2whichisnt under consideration.

SettingA
(s) = 0gives, bythequadraticformula,

s =
4
48
2
= 22
3.
Of these, only2+2
3ispositive, sothisisour lonecritical point. A(2+2
3) 1.0718> 0. Sincewearending
themaximumoveraclosedinterval and2+2
3istheonlycritical point,themaximumareaisA(2+2
3) 1.0718.
Findthemaximumareaof atriangleformedby theaxesandatangent linetothegraphof y = (x + 1)
2
with
x > 0.
27. Findthemaximumareaof arectanglecircumscribedaroundarectangleof sidesL andH. Hint: Expresstheareain
termsof theangle (Figure18).
H
q
L
FIGURE 18
solution PositiontheL H rectangleintherst quadrant of thexy-planewithitsnorthwest corner at theorigin.
Let betheanglethebaseof thecircumscribedrectanglemakeswiththepositivex-axis, where0

2
. Thenthearea
of thecircumscribedrectangleisA = LH + 2
1
2
(H sin)(H cos) + 2
1
2
(Lsin)(Lcos) = LH +
1
2
(L
2
+ H
2
)
sin2, whichhas amaximumvalueof LH +
1
2
(L
2
+ H
2
) when =

4
becausesin2 achieves its maximumwhen
=

4
.
Acontractorisengagedtobuildstepsuptheslopeof ahill thathastheshapeof thegraphof y = x
2
(120x)/6400
for 0 x 80withx inmeters(Figure19). Whatisthemaximumvertical riseof astair if eachstair hasahorizontal
lengthof one-thirdmeter.
29. Findtheequationof thelinethroughP = (4, 12) suchthat thetriangleboundedbythislineandtheaxesintherst
quadrant hasminimal area.
solution Let P = (4, 12) be a point in the rst quadrant and y 12 = m(x 4), < m < 0, be a line
throughP that cutsthepositivex- andy-axes. Theny = L(x) = m(x 4) +12. ThelineL(x) intersectsthey-axisat
H (0, 124m) andthex-axisat W
_
4
12
m
, 0
_
. Hencetheareaof thetriangleis
A(m) =
1
2
(124m)
_
4
12
m
_
= 488m72m
1
.
Solve A
(m) = 72m
2
8 = 0 for m < 0 to obtain m = 3. Since A as m or m 0, we
concludethat theminimal triangular areais obtainedwhenm = 3. Theequationof thelinethroughP = (4, 12) is
y = 3(x 4) +12= 3x +24.
LetP = (a, b) lieintherstquadrant. Findtheslopeof thelinethroughP suchthatthetriangleboundedbythis
lineandtheaxesintherst quadrant hasminimal area. Thenshowthat P isthemidpoint of thehypotenuseof this
triangle.
31. Archimedes Problem A spherical cap(Figure20) of radius r andheight h has volumeV = h
2
_
r
1
3
h
_
and
surfaceareaS = 2rh. Provethat thehemisphereenclosesthelargest volumeamongall spherical capsof xedsurface
areaS.
r
h
FIGURE 20
solution Consider all spherical capsof xedsurfaceareaS. BecauseS = 2rh, it followsthat
r =
S
2h
and
V(h) = h
2
_
S
2h
1
3
h
_
=
S
2
h

3
h
3
.
Now
V
(h) =
S
2
h
2
= 0
when
h
2
=
S
2
or h =
S
2h
= r.
Hence, thehemisphereenclosesthelargest volumeamongall spherical capsof xedsurfaceareaS.
Find the isosceles triangle of smallest area (Figure 21) that circumscribes a circle of radius 1 (fromThomas
SimpsonsThe Doctrine and Application of Fluxions, acalculustext that appearedin1750).
33. A boxof volume72m
3
withsquarebottomandnotopisconstructedout of twodifferent materials. Thecost of the
bottomis$40/m
2
andthecost of thesidesis$30/m
2
. Findthedimensionsof theboxthat minimizetotal cost.
solution Let s denotethelengthof thesideof thesquarebottomof theboxandh denotetheheight of thebox. Then
V = s
2
h = 72 or h =
72
s
2
.
Thecost of theboxis
C = 40s
2
+120sh = 40s
2
+
8640
s
,
so
C
(s) = 80s
8640
s
2
= 0
whens = 3
3
4mandh = 2
3
4m. BecauseC as s 0 andas s , weconcludethat thecritical point

givestheminimumcost.
Findthedimensionsof acylinder of volume1m
3
of minimal costif thetopandbottomaremadeof material that
coststwiceasmuchasthematerial for theside.
35. Your task is to design a rectangular industrial warehouse consisting of three separate spaces of equal size as in
Figure22. Thewall materialscost $500per linear meter andyour companyallocates$2,400,000for theproject.
(a) Whichdimensionsmaximizetheareaof thewarehouse?
(b) What istheareaof eachcompartment inthiscase?
FIGURE 22
solution Letonecompartmenthavelengthx andwidthy.Thentotal lengthof thewall of thewarehouseisP = 4x +6y
andtheconstraint equationiscost = 2,400,000= 500(4x +6y), whichgivesy = 800
2
3
x.
(a) AreaisgivenbyA = 3xy = 3x
_
800
2
3
x
_
= 2400x 2x
2
, where0 x 1200. ThenA
(x) = 24004x = 0
yieldsx = 600andconsequentlyy = 400. SinceA(0) = A(1200) = 0andA(600) = 720, 000, theareaof thewarehouse
ismaximizedwheneachcompartment haslengthof 600 mandwidthof 400 m.
(b) Theareaof onecompartment is600 400= 240, 000squaremeters.
Suppose, inthepreviousexercise, that thewarehouseconsistsof n separatespacesof equal size. Findaformula
intermsof n for themaximumpossibleareaof thewarehouse.
37. Accordingtoamodel developedbyeconomistsE. HeadyandJ . Pesek, if fertilizer madefromN poundsof nitrogen
andP poundsof phosphateisusedonanacreof farmland, thentheyieldof corn(inbushelsper acre) is
Y = 7.5+0.6N +0.7P 0.001N
2
0.002P
2
+0.001NP
A farmer intendstospend$30per acreonfertilizer. If nitrogencosts25cents/lbandphosphatecosts20cents/lb, which
combinationof N andL producesthehighest yieldof corn?
solution Thefarmersbudget for fertilizer is$30per acre, sowehavetheconstraint equation
0.25N +0.2P = 30 or P = 1501.25N
Substitutingfor P intheequationfor Y, wend
Y(N) = 7.5+0.6N +0.7(1501.25N) 0.001N
2
0.002(1501.25N)
2
+0.001N(1501.25N)
= 67.5+0.625N 0.005375N
2
BothN andP must benonnegative. SinceP = 1501.25N 0, werequirethat 0 N 120. Next,
dY
dN
= 0.6250.01075N = 0 N =
0.625
0.01075
58.14pounds.
Now, Y(0) = 67.5, Y(120) = 65.1andY(58.14) 85.67, sothemaximumyieldof cornoccursfor N 58.14pounds
andP 77.33pounds.
Experimentsshowthat thequantitiesx of cornandy of soybeanrequiredtoproduceahogof weight Q satisfy
Q = 0.5x
1/2
y
1/4
. Theunit of x, y, andQ isthecwt, anagricultural unit equal to100lbs. Findthevaluesof x and
y that minimizethecost of ahogof weight Q = 2.5cwt if corncosts$3/cwt andsoycosts$7/cwt.
39. All unitsina100-unit apartment buildingarerentedout whenthemonthly rent isset at r = $900/month. Suppose
thatoneunitbecomesvacantwitheach$10increaseinrentandthateachoccupiedunitcosts$80/monthinmaintenance.
Whichrent r maximizesmonthlyprot?
solution Let n denotethenumber of $10increasesinrent. Thenthemonthlyprot isgivenby
P(n) = (100n)(900+10n 80) = 82000+180n 10n
2
,
and
P
(n) = 18020n = 0
whenn = 9. Weknowthisresultsinmaximumprot becausethisgivesthelocationof vertex of adownwardopening
parabola. Thus, monthlyprot ismaximizedwitharent of $990.
An8-billion-bushel corncropbringsapriceof $2.40/bu. A commoditybroker usestheruleof thumb: If thecrop
isreducedbyx percent, thenthepriceincreasesby10x cents. Whichcropsizeresultsinmaximumrevenueandwhat
isthepriceper bu? Hint: Revenueisequal topricetimescropsize.
41. Themonthly output of aSpanishlight bulbfactory isP = 2LK
2
(inmillions), whereL isthecost of labor andK
isthecostof equipment(inmillionsof euros). Thecompanyneedstoproduce1.7millionunitsper month. Whichvalues
of L andK wouldminimizethetotal cost L +K?
solution SinceP = 1.7andP = 2LK
2
, wehaveL =
0.85
K
2
. Accordingly, thecost of productionis
C(K) = L +K = K +
0.85
K
2
.
SolveC
(K) = 1
1.7
K
3
for K 0toobtainK =
3
1.7. SinceC(K) asK 0+ andasK , theminimum

cost of production is achieved for K =
3
1.7 1.2 and L = 0.6. Thecompany should invest 1.2 million euros in
equipment and600, 000eurosinlabor.
Therectangular plot inFigure23hassize100m 200m. PipeistobelaidfromA toapoint P onsideBC and
fromtheretoC. Thecost of layingpipealongthesideof theplot is $45/mandthecost throughtheplot is $80/m
(sinceit isunderground).
(a) Letf (x) bethetotal cost, wherex isthedistancefromP toB. Determinef (x), butnotethatf isdiscontinuous
at x = 0(whenx = 0, thecost of theentirepipeis$45/ft).
(b) What isthemost economical waytolaythepipe?What if thecost alongthesidesis$65/m?
43. Brandonis ononesideof ariver that is 50mwideandwants toreachapoint 200mdownstreamontheopposite
sideasquickly aspossibleby swimmingdiagonally acrosstheriver andthenrunningtherest of theway. Findthebest
routeif Brandoncanswimat 1.5m/sandrunat 4m/s.
solution Let lengthsbeinmeters, timesinseconds, andspeedsinm/s. Supposethat Brandonswimsdiagonally to
apoint locatedx metersdownstreamontheoppositeside. ThenBrandonthenswimsadistance
_
x
2
+50
2
andrunsa
distance200x. Thetotal timeof thetripis
f (x) =
_
x
2
+2500
1.5
+
200x
4
, 0 x 200.
Solve
f
(x) =
2x
3
_
x
2
+2500
1
4
= 0
toobtainx = 30
5
11
20.2andf (20.2) 80.9. Sincef (0) 83.3andf (200) 137.4, weconcludethat theminimal
timeis80.9s. ThisoccurswhenBrandonswimsdiagonallytoapoint located20.2mdownstreamandthenrunstherest
of theway.
SnellsLaw Whenalight beamtravels fromapoint A aboveaswimmingpool toapoint B belowthewater
(Figure24), it choosesthepaththat takestheleast time. Let v
1
bethevelocity of light inair andv
2
thevelocity in
water (it isknownthat v
1
> v
2
). ProveSnellsLawof Refraction:
sin
1
v
1
=
sin
2
v
2
In Exercises 4547, a box (with no top) is to be constructed from a piece of cardboard of sides A and B by cutting out
squares of length h from the corners and folding up the sides (Figure 26).
h
A
B
FIGURE 26
45. Findthevalueof h that maximizesthevolumeof thebox if A = 15andB = 24. What arethedimensionsof this
box?
solution Oncethesideshavebeenfoldedup, thebaseof theboxwill havedimensions(A2h) (B 2h) andthe
height of theboxwill beh. Thus
V(h) = h(A 2h)(B 2h) = 4h
3
2(A +B)h
2
+ABh.
WhenA = 15andB = 24, thisgives
V(h) = 4h
3
78h
2
+360h,
andweneedtomaximizeover 0 h
15
2
. Now,
V
(h) = 12h
2
156h +360= 0
yieldsh = 3andh = 10. Becauseh = 10isnotinthedomainof theproblemandV(0) = V(15/2) = 0andV(3) = 486,
volumeismaximizedwhenh = 3. Thecorrespondingdimensionsare9183.
Vascular Branching A small bloodvessel of radiusr branchesoff at anangle fromalarger vessel of radius
R to supply blood along apath fromA to B. According to Poiseuilles Law, thetotal resistanceto blood owis
proportional to
T =
_
a b cot
R
4
+
b csc
r
4
_
wherea andb areasinFigure25. Showthat thetotal resistanceisminimizedwhencos = (r/R)
4
.
47. Whichvaluesof A andB maximizethevolumeof theboxif h = 10cmandAB = 900cm.
solution Withh = 10andAB = 900(whichmeansthat B = 900/A), thevolumeof theboxis
V(A) = 10(A 20)
_
900
A
20
_
= 13,000200A
180,000
A
,
where20 A 45. Now, solving
V
(A) = 200+
180,000
A
2
= 0
yields A = 30. BecauseV(20) = V(45) = 0andV(30) = 1000cm
3
, maximumvolumeis achievedwithA = B =
30cm.
Givenn numbersx
1
, . . . , x
n
, ndthevalueof x minimizingthesumof thesquares:
(x x
1
)
2
+(x x
2
)
2
+ +(x x
n
)
2
First solvefor n = 2, 3andthentryit for arbitraryn.
49. A billboardof height b ismountedonthesideof abuildingwithitsbottomedgeat adistanceh fromthestreet asin
Figure27. At what distancex shouldanobserver standfromthewall tomaximizetheangleof observation?
h
b
x
P
q
P
A
R
B
C
Q
FIGURE 27
solution Fromtheupper diagraminFigure27andtheadditionformulafor thecotangent function, weseethat
cot =
1+
x
b+h
x
h
x
h

x
b+h
=
x
2
+h(b +h)
bx
,
whereb andh areconstant. Now, differentiatewithrespect tox andsolve
csc
2
d
dx
=
x
2
h(b +h)
bx
2
= 0
toobtainx =
_
bh +h
2
. Sincethisistheonlycritical point, andsince 0asx 0+ and 0asx , (x)
reachesitsmaximumat x =
_
bh +h
2
.
SolveExercise49againusinggeometry rather thancalculus. ThereisauniquecirclepassingthroughpointsB
andC whichistangent tothestreet. Let R bethepoint of tangency. Notethat thetwoangleslabeled inFigure27
areequal becausetheysubtendequal arcsonthecircle.
(a) Showthat themaximumvalueof is = . Hint: Showthat = +
PBA whereA istheintersectionof

thecirclewithPC.
(b) Provethat thisagreeswiththeanswer toExercise49.
(c) Showthat
QRB =
RCQfor themaximal angle.

51. Optimal DeliverySchedule A gas station sells Q gallons of gasolineper year, which is delivered N times per
year inequal shipments of Q/N gallons. Thecost of eachdelivery is d dollars andtheyearly storagecosts aresQT ,
whereT isthelengthof time(afractionof ayear) betweenshipmentsands isaconstant. Showthat costsareminimized
for N =

sQ/d. (Hint: T = 1/N.) Find theoptimal number of deliveries if Q = 2 million gal, d = $8000, and
s = 30cents/gal-yr. Your answer shouldbeawholenumber, socomparecostsfor thetwointeger valuesof N nearestthe
optimal value.
solution ThereareN shipmentsper year, sothetimeinterval betweenshipmentsisT = 1/N years. Hence, thetotal
storagecostsper year aresQ/N. TheyearlydeliverycostsaredN andthetotal costsisC(N) = dN +sQ/N. Solving,
C
(N) = d
sQ
N
2
= 0
for N yieldsN =
sQ/d. For thespeciccaseQ = 2,000,000, d = 8000ands = 0.30,

N =
_
0.30(2,000,000)
8000
= 8.66.
WithC(8) = $139,000andC(9) = $138,667, theoptimal number of deliveriesper year isN = 9.
VictorKleesEndpointMaximumProblemGiven40metersof straightfence, yourgoal istobuildarectangular
enclosureusing80additional metersof fencethatencompassesthegreatestarea. LetA(x) betheareaof theenclosure,
withx asinFigure28.
(a) Findthemaximumvalueof A(x).
(b) Whichinterval of x valuesisrelevant toour problem? Findthemaximumvalueof A(x) onthisinterval.
53. Let(a, b) beaxedpointintherstquadrantandletS(d) bethesumof thedistancesfrom(d, 0) tothepoints(0, 0),
(a, b), and(a, b).
(a) Findthevalueof d for whichS(d) isminimal. Theanswer dependsonwhether b <
3a or b
3a. Hint: Show

that d = 0whenb
3a.
(b) Let a = 1. Plot S(d) for b = 0.5,
3, 3anddescribethepositionof theminimum.
solution
(a) If d < 0, then thedistancefrom(d, 0) to theother threepoints can all bereduced by increasing thevalueof d.
Similarly, if d > a, thenthedistancefrom(d, 0) totheother threepointscanall bereducedby decreasingthevalueof
d. It followsthat theminimumof S(d) must occur for 0 d a. Restrictingattentiontothisinterval, wend
S(d) = d +2
_
(d a)
2
+b
2
.
Solving
S
(d) = 1+
2(d a)
_
(d a)
2
+b
2
= 0
yieldsthecritical point d = a b/
3. If b <
3a, thend = a b/
3> 0andtheminimumoccursatthisvalueof d.
Ontheother hand, if b
3a, thentheminimumoccursat theendpoint d = 0.

(b) Leta = 1. Plotsof S(d) for b = 0.5, b =
3andb = 3areshownbelow. For b = 0.5, theresultsof (a) indicatethe

minimumshouldoccur for d = 10.5/
3 0.711, andthisisconrmedintheplot. For bothb =
3andb = 3, the
resultsof (a) indicatethat theminimumshouldoccur at d = 0, andbothof theseconclusionsareconrmedintheplots.
1.6
x
0 0.2 0.4 0.6 0.8
b = 0.5
1
y
1.5
1.9
1.8
1.7
2
2.1
x
0 0.2 0.4 0.6 0.8 1
y
6.6
6.4
6.8
b = 3
4
x
0 0.2 0.4 0.6 0.8 1
y
4.2
4.1
4.3
4.4
b = 3
TheforceF (inNewtons)requiredtomoveaboxof massmkginmotionbypullingonanattachedrope(Figure29)
is
F() =
f mg
cos +f sin
where istheanglebetweentheropeandthehorizontal, f isthecoefcientof staticfriction, andg = 9.8m/s
2
. Find
theangle thatminimizestherequiredforceF, assumingf = 0.4. Hint: Findthemaximumvalueof cos +f sin.
55. Inthesettingof Exercise54, showthat for anyf theminimal forcerequiredisproportional to1/
_
1+f
2
.
solution WeminimizeF() by ndingthemaximumvalueg() = cos + f sin. Theangle isrestrictedtothe
interval [0,

2
]. Wesolvefor thecritical points:
g
() = sin +f cos = 0
Weobtain
f cos = sin tan = f
Fromthegurebelowwendthat cos = 1/
_
1+f
2
andsin = f/
_
1+f
2
. Hence
g() =
1
f
+
f
2
_
1+f
2
=
1+f
2
_
1+f
2
=
_
1+f
2
Thevaluesat theendpointsare
g(0) = 1, g
_
2
_
= f
Bothof thesevaluesarelessthan
_
1+f
2
. Thereforethemaximumvalueof g() is
_
1+f
2
andtheminimumvalue
of F() is
F =
f mg
g()
=
f mg
_
1+f
2
f
1
1 + f
2
BirdMigration Ornithologistshavefoundthat thepower (injoulesper second) consumedbyacertainpigeon
yingat velocityv m/sisdescribedwell bythefunctionP(v) = 17v
1
+10
3
v
3
J /s. Assumethat thepigeoncan
store510
4
J of usableenergyasbodyfat.
(a) Showthat at velocity v, apigeoncany atotal distanceof D(v) = (5 10
4
)v/P(v) if it usesall of itsstored
energy.
(b) Findthevelocityv
p
that minimizes P(v).
(c) Migratingbirdsaresmart enoughtoy at thevelocity that maximizesdistancetraveledrather thanminimizes
power consumption. Showthat thevelocity v
d
whichmaximizesD(v) satisesP
(v
d
) = P(v
d
)/v
d
. Showthat v
d
isobtainedgraphicallyasthevelocitycoordinateof thepoint wherealinethroughtheoriginistangent tothegraph
of P(v) (Figure30).
(d) Findv
d
andthemaximumdistanceD(v
d
).
57. Theproblemistoputaroofof sides onanatticroomof heighth andwidthb. Findthesmallestlengths for which
thisispossibleif b = 27andh = 8(Figure31).
s
h
b
FIGURE 31
solution Consider theright triangleformed by theright half of therectangleand its roof. This trianglehas hy-
potenuses.
h
b/2 x
s y
Asshown, let y betheheight of theroof, andlet x bethedistancefromtheright baseof therectangletothebaseof the
roof. Bysimilar trianglesappliedtothesmaller right trianglesat thetopandright of thelarger triangle, weget:
y 8
27/2
=
8
x
or y =
108
x
+8.
s, y, andx arerelatedbythePythagoreanTheorem:
s
2
=
_
27
2
+x
_
2
+y
2
=
_
27
2
+x
_
2
+
_
108
x
+8
_
2
.
Sinces > 0, s
2
isleastwhenever s isleast, sowecanminimizes
2
insteadof s. Settingthederivativeequal tozeroyields
2
_
27
2
+x
_
+2
_
108
x
+8
__
108
x
2
_
= 0
2
_
27
2
+x
_
+2
8
x
_
27
2
+x
__
108
x
2
_
= 0
2
_
27
2
+x
__
1
864
x
3
_
= 0
Thezerosarex =
27
2
(irrelevant) andx = 6
3
4. Sincethisistheonlycritical point of s withx > 0, andsinces

asx 0ands asx , thisisthepoint wheres attainsitsminimum. For thisvalueof x,
s
2
=
_
27
2
+6
3
4
_
2
+
_
9
3
2+8
_
2
904.13,
sothesmallest roof lengthis
s 30.07.
RedoExercise57for arbitraryb andh.
59. Findthemaximumlengthof apolethatcanbecarriedhorizontallyaroundacornerjoiningcorridorsof widthsa = 24
andb = 3(Figure32).
a
b
FIGURE 32
solution Inorder to ndthelengthof thelongest polethat canbecarriedaroundthecorridor, wehaveto ndthe
shortest lengthfromtheleft wall tothetopwall touchingthecorner of theinsidewall. Anypolethat doesnot t inthis
shortest spacecannot becarriedaroundthecorner, soanexact t representsthelongest possiblepole.
Let betheanglebetweenthepoleandahorizontal linetotheright. Let c
1
bethelengthof poleinthecorridor of
width24andlet c
2
bethelengthof poleinthecorridor of width3. Bythedenitionsof sineandcosine,
3
c
2
= sin and
24
c
1
= cos,
sothat c
1
=
24
cos
, c
2
=
3
sin
. What must beminimizedisthetotal length, givenby
f () =
24
cos
+
3
sin
.
Settingf
() = 0yields
24sin
cos
2
3cos
sin
2
= 0
24sin
cos
2
=
3cos
sin
2
24sin
3
= 3cos
3
As <

2
(thepoleisbeingturnedaroundacorner, after all), wecandividebothsidesby cos
3
, gettingtan
3
=
1
8
.
Thisimpliesthat tan =
1
2
(tan > 0astheangleisacute).
Sincef () as 0+ andas

2
, wecantell that theminimum is attainedat
0
wheretan
0
=
1
2
.
Because
tan
0
=
opposite
adjacent
=
1
2
,
wedrawatrianglewithoppositeside1andadjacent side2. ByPythagoras, c =
5, so
sin
0
=
1
5
and cos
0
=
2
5
.
Fromthis, weget
f (
0
) =
24
cos
0
+
3
sin
0
=
24
2
5+3
5= 15
5.
RedoExercise59for arbitrarywidthsa andb.
61. Findtheminimumlength of abeamthatcanclear afenceof heighth andtouchawall locatedb ftbehindthefence
(Figure33).
b x
h
FIGURE 33
solution Let y betheheight of thepoint wherethebeamtouchesthewall infeet. Bysimilar triangles,
y h
b
=
h
x
or y =
bh
x
+h
andbyPythagoras:
2
= (b +x)
2
+
_
bh
x
+h
_
2
.
Wecanminimize
2
rather than, sosettingthederivativeequal tozerogives:
2(b +x) +2
_
bh
x
+h
__
bh
x
2
_
= 2(b +x)
_
1
h
2
b
x
3
_
= 0.
Thezeroesareb = x (irrelevant) andx =
3
h
2
b. Since
2
asx 0+ andasx , x =
3
h
2
b corresponds
toaminimumfor
2
. For thisvalueof x, wehave
2
= (b +h
2/3
b
1/3
)
2
+(h +h
1/3
b
2/3
)
2
= b
2/3
(b
2/3
+h
2/3
)
2
+h
2/3
(h
2/3
+b
2/3
)
2
= (b
2/3
+h
2/3
)
3
andso
= (b
2/3
+h
2/3
)
3/2
.
A beamthat clears afenceof height h feet and touches awall b feet behind thefencemust havelength at least =
(b
2/3
+h
2/3
)
3/2
ft.
Whichvalueof h maximizesthevolumeof theboxif A = B?
63. A basketball player stands d feet fromthebasket. Let h and beas inFigure34. Usingphysics, onecan
showthat if theplayer releases theball at anangle, thentheinitial velocity requiredtomaketheball gothroughthe
basket satises
v
2
=
16d
cos
2
(tan tan)
(a) Explainwhythisformulaismeaningful onlyfor < <

2
. Whydoesv approachinnityat theendpointsof this
interval?
(b) Take =

6
andplot v
2
asafunctionof for

6
< <

2
. Verifythat theminimumoccursat =

3
.
(c) Set F() = cos
2
(tan tan). Explainwhyv isminimizedfor suchthat F() ismaximized.
(d) Verify that F
() = cos( 2) sec (you will need to usetheaddition formulafor cosine) and show that the
maximumvalueof F() on
_
,

2
_
occursat
0
=

2
+

4
.
(e) For agiven, theoptimal anglefor shootingthebasket is
0
becauseit minimizesv
2
andthereforeminimizesthe
energyrequiredtomaketheshot(energyisproportional tov
2
). Showthatthevelocityv
opt
attheoptimal angle
0
satises
v
2
opt
=
32d cos
1sin
=
32d
2
h +
_
d
2
+h
2
(f) Showwithagraphthatfor xedd (say, d = 15ft, thedistanceof afreethrow), v
2
opt
isanincreasingfunction
of h. Usethistoexplainwhytaller playershaveanadvantageandwhyit canhelptojumpwhileshooting.
q
a
h
d
FIGURE 34
solution
(a) = 0correspondstoshootingtheball directlyatthebasketwhile = /2correspondstoshootingtheball directly
upward. Inneither caseisit possiblefor theball togointothebasket.
If theangle is extremely closeto0, theball is shot almost directly at thebasket, sothat it must belaunchedwith
great speed, asit canonlyfall anextremelyshort distanceonthewaytothebasket.
Ontheother hand, if theangle is extremely closeto/2, theball is launchedalmost vertically. This requires the
ball totravel agreat distanceupwardinorder totravel thehorizontal distance. Ineither oneof thesecases, theball hasto
travel at anenormousspeed.
(b)
3
5
12
2
Theminimumclearlyoccurswhere = /3.
(c) If F() = cos
2
(tan tan),
v
2
=
16d
cos
2
(tan tan)
=
16d
F()
.
Since , F() > 0, hencev
2
issmallest whenever F() isgreatest.
(d) F
() = 2sin cos (tan tan) + cos

2
_
sec
2
_
= 2sin cos tan + 2sin cos tan + 1. Wewill
applyall thedoubleangleformulas:
cos(2) = cos
2
sin
2
= 12sin
2
; sin2 = 2sin cos,
getting:
F
() = 2sin cos tan 2sin cos tan +1

= 2sin cos
sin
cos
2sin cos
sin
cos
+1
= sec
_
2sin
2
cos +2sin cos sin +cos
_
= sec
_
cos
_
12sin
2
_
+sin (2sin cos)
_
= sec (cos(cos2) +sin(sin2))
= sec cos( 2)
A critical point of F() occurswherecos( 2) = 0, sothat 2 =
2
(negativebecause2 > > ), andthis
givesus = /2+/4. TheminimumvalueF(
0
) takesplaceat
0
= /2+/4.
(e) Plugin
0
= /2+/4. Tondv
2
opt
wemust simplify
cos
2
0
(tan
0
tan) =
cos
0
(sin
0
cos cos
0
sin)
cos
Bytheadditionlawfor sine:
sin
0
cos cos
0
sin = sin(
0
) = sin(/2+/4)
andso
cos
0
(sin
0
cos cos
0
sin) = cos(/2+/4) sin(/2+/4)
Nowusetheidentity(that followsfromtheadditionlaw):
sinx cosy =
1
2
(sin(x +y) +sin(x y))
toget
cos(/2+/4) sin(/2+/4) = (1/2)(1sin)
Sowenallyget
cos
2
0
(tan
0
tan) =
(1/2)(1sin)
cos
andtherefore
v
2
opt
=
32d cos
1sin
asclaimed. FromFigure34weseethat
cos =
d
_
d
2
+h
2
and sin =
h
_
d
2
+h
2
.
Substitutingthesevaluesintotheexpressionfor v
2
opt
yields
v
2
opt
=
32d
2
h +
_
d
2
+h
2
.
(f) A sketchof thegraphof v
2
opt
versus h for d = 15feet is givenbelow: v
2
opt
increases withrespect tobasket height
relativetotheshooter. Thisshowsthat theminimumvelocity requiredtolaunchtheball tothebasket dropsasshooter
heightincreases. Thisshowsoneof thewaysheightisanadvantageinfreethrows; ataller shooter neednotshoottheball
ashardtoreachthebasket.
100
200
300
400
500
600
4 5 0 3 2 1
ThreetownsA, B, andC aretobejoinedby anundergroundber cableasillustratedinFigure35(A). Assume
that C is locateddirectly belowthemidpoint of AB. Findthejunctionpoint P that minimizes thetotal amount of
cableused.
(a) FirstshowthatP mustliedirectlyaboveC. Hint: Usetheresultof Example6toshowthatif thejunctionisplaced
at point QinFigure35(B), thenwecanreducethecablelengthbymovingQhorizontallyover tothepoint P lying
aboveC.
(b) Withx asinFigure35(A), let f (x) bethetotal lengthof cableused. Showthat f (x) hasauniquecritical point
c. Computec andshowthat 0 c L if andonlyif D 2
3L.
(c) Findtheminimumof f (x) on[0, L] intwocases: D = 2, L = 4andD = 8, L = 2.
65. TomandAli drivealong ahighway represented by thegraph of f (x) in Figure36. During thetrip, Ali views a
billboardrepresentedbythesegmentBC alongthey-axis. LetQbethey-interceptof thetangentlinetoy = f (x). Show
that ismaximizedatthevalueof x for whichtheangles
QPB and
QCP areequal. ThisgeneralizesExercise50(c)

(whichcorrespondstothecasef (x) = 0). Hints:
(a) Showthat d/dx isequal to
(b c)
(x
2
+(xf
(x))
2
) (b (f (x) xf
(x)))(c (f (x) xf
(x)))
(x
2
+(b f (x))
2
)(x
2
+(c f (x))
2
)
(b) Showthat they-coordinateof Qisf (x) xf
(x).
(c) Showthat theconditiond/dx = 0isequivalent to
PQ
2
= BQ CQ
(d) Concludethat QPB andQCP aresimilar triangles.
x
x
y
billboard
highway
P = (x, f (x))
y = f (x)
B = (0, b)
C = (0, c)
FIGURE 36
solution
(a) Fromthegure, weseethat
(x) = tan
1
c f (x)
x
tan
1
b f (x)
x
.
Then
(x) =
b (f (x) xf
(x))
x
2
+(b f (x))
2

c (f (x) xf
(x))
x
2
+(c f (x))
2
= (b c)
x
2
bc +(b +c)(f (x) xf
(x)) (f (x))
2
+2xf (x)f
(x)
(x
2
+(b f (x))
2
)(x
2
+(c f (x))
2
)
= (b c)
(x
2
+(xf
(x))
2
(bc (b +c)(f (x) xf
(x)) +(f (x) xf
(x))
2
)
(x
2
+(b f (x))
2
)(x
2
+(c f (x))
2
)
= (b c)
(x
2
+(xf
(x))
2
(b (f (x) xf
(x)))(c (f (x) xf
(x)))
(x
2
+(b f (x))
2
)(x
2
+(c f (x))
2
)
.
(b) Thepoint Qisthey-intercept of thelinetangent tothegraphof f (x) at point P. Theequationof thistangent lineis
Y f (x) = f
(x)(X x).
They-coordinateof Qisthenf (x) xf
(x).
(c) Fromthegure, weseethat
BQ = b (f (x) xf
(x)),
CQ = c (f (x) xf
(x))
and
PQ =
_
x
2
+(f (x) (f (x) xf
(x)))
2
=
_
x
2
+(xf
(x))
2
.
Comparingtheseexpressionswiththenumerator of d/dx, it followsthat
d
dx
= 0isequivalent to
PQ
2
= BQ CQ.
(d) TheequationPQ
2
= BQ CQisequivalent to
PQ
BQ
=
CQ
PQ
.
Inother words, thesidesCQandPQfromthetriangleQCP areproportional inlengthtothesidesPQandBQfrom
thetriangleQPB. As
PQB =
CQP, it followsthat trianglesQCP andQPB aresimilar.

SeismicProspecting Exercises 6668 are concerned with determining the thickness d of a layer of soil that lies on top
of a rock formation. Geologists send two sound pulses from point A to point D separated by a distance s. The rst pulse
travels directly from A to D along the surface of the earth. The second pulse travels down to the rock formation, then
along its surface, and then back up to D (path ABCD), as in Figure 37. The pulse travels with velocity v
1
in the soil and
v
2
in the rock.
A
B C
s D
Soil
Rock
q q d
FIGURE 37
(a) Showthat thetimerequiredfor therst pulsetotravel fromA toD ist
1
= s/v
1
.
(b) Showthat thetimerequiredfor thesecondpulseis
t
2
=
2d
v
1
sec +
s 2d tan
v
2
providedthat
tan
s
2d
(Note: If thisinequalityisnot satised, thenpoint B doesnot lietotheleft of C.)
(c) Showthat t
2
isminimizedwhensin = v
1
/v
2
.
67. Inthisexercise, assumethat v
2
/v
1
_
1+4(d/s)
2
.
(a) Showthat inequality(2) holdsif sin = v
1
/v
2
.
(b) Showthat theminimal timefor thesecondpulseis
t
2
=
2d
v
1
(1k
2
)
1/2
+
s
v
2
wherek = v
1
/v
2
.
(c) Concludethat
t
2
t
1
=
2d(1k
2
)
1/2
s
+k.
solution
(a) If sin =
v
1
v
2
, then
tan =
v
1
_
v
2
2
v
2
1
=
1
_
_
v
2
v
1
_
2
1
.
Because
v
2
v
1

_
1+4(
d
s
)
2
, it followsthat
_
_
v
2
v
1
_
2
1
_
1+4
_
d
s
_
2
1=
2d
s
.
Hence, tan
s
2d
asrequired.
(b) For thetime-minimizingchoiceof , wehavesin =
v
1
v
2
fromwhichsec =
v
2
_
v
2
2
v
2
1
andtan =
v
1
_
v
2
2
v
2
1
.
Thus
t
2
=
2d
v
1
sec +
s 2d tan
v
2
=
2d
v
1
v
2
_
v
2
2
v
2
1
+
s 2d
v
1
_
v
2
2
v
2
1
v
2
=
2d
v
1
v
2
_
v
2
2
v
2
1
v
2
1
v
2
_
v
2
2
v
2
1
+
s
v
2
=
2d
v
1
v
2
2
v
2
1
v
2
_
v
2
2
v
2
1
+
s
v
2
=
2d
v
1
_
v
2
2
v
2
1
_
v
2
2
+
s
v
2
=
2d
v
1
_
1
_
v
1
v
2
_
2
+
s
v
2
=
2d
_
1k
2
_
1/2
v
1
+
s
v
2
.
(c) Recall that t
1
=
s
v
1
. Wethereforehave
t
2
t
1
=
2d
_
1k
2
_
1/2
v
1
+
s
v
2
s
v
1
=
2d
_
1k
2
_
1/2
s
+
v
1
v
2
=
2d
_
1k
2
_
1/2
s
+k.
Continuewiththeassumptionof thepreviousexercise.
(a) Findthethicknessof thesoil layer, assumingthat v
1
= 0.7v
2
, t
2
/t
1
= 1.3, ands = 400m.
(b) Thetimes t
1
and t
2
aremeasured experimentally. Theequation in Exercise67(c) shows that t
2
/t
1
is alinear
functionof 1/s.Whatmightyouconcludeif experimentswereformedforseveral valuesof s andthepoints(1/s, t
2
/t
1
)
didnot lieonastraight line?
69. InthisexerciseweuseFigure38toproveHeronsprincipleof Example6withoutcalculus. Bydenition, C
isthereectionof B acrossthelineMN (sothat BC isperpendicular toMN andBN = CN. Let P betheintersection
of AC andMN. Usegeometrytojustify:
(a) PNB andPNC arecongruent and
1
=
2
.
(b) ThepathsAPB andAPC haveequal length.
(c) SimilarlyAQB andAQC haveequal length.
(d) ThepathAPC isshorter thanAQC for all Q = P.
Concludethat theshortest pathAQB occursfor Q = P.
A
B
h
1
h
2
P
h
2
Q
C
M N
2
FIGURE 38
solution
(a) Bydenition, BC isorthogonal toQM, sotrianglesPNB andPNC arecongruentbysideangleside.Therefore
1
=
2
(b) BecausePNB andPNC arecongruent, it followsthat PB andPC areof equal length. Thus, pathsAPB and
APC haveequal length.
(c) Thesamereasoningusedinparts (a) and(b) leadus to concludethat QNB andQNC arecongruent andthat
PB andPC areof equal length. Thus, pathsAQB andAQC areof equal length.
(d) Consider triangleAQC. By thetriangleinequality, thelengthof sideAC is less thanor equal to thesumof the
lengthsof thesidesAQandQC. Thus, thepathAPC isshorter thanAQC for all Q = P.
Finally, theshortest pathAQB occursfor Q = P.
A jewelrydesigner planstoincorporateacomponent madeof goldintheshapeof afrustumof aconeof height
1cmandxedlower radiusr (Figure39). Theupper radiusx cantakeonanyvaluebetween0andr. Notethatx = 0
andx = r correspondtoaconeandcylinder, respectively. Asafunctionof x, thesurfacearea(not includingthetop
andbottom) isS(x) = s(r +x), wheres istheslant height asindicatedinthegure. Whichvalueof x yieldsthe
least expensivedesign[theminimumvalueof S(x) for 0 x r]?
(a) Showthat S(x) = (r +x)
_
1+(r x)
2
.
(b) Showthat if r <
2, thenS(x) isanincreasingfunction. Concludethat thecone(x = 0) hasminimal areain

thiscase.
(c) Assumethatr >
2. ShowthatS(x) hastwocritical pointsx

1
< x
2
in(0, r), andthatS(x
1
) isalocal maximum,
andS(x
2
) isalocal minimum.
(d) Concludethat theminimumoccursat x = 0or x
2
.
(e) Findtheminimuminthecasesr = 1.5andr = 2.
(f) Challenge: Let c =
_
(5+3
3)/4 1.597. Provethat theminimumoccursat x = 0(cone) if
2 < r < c,
but theminimumoccursat x = x
2
if r > c.
4.7 Newtons Method
1. Howmanyiterationsof NewtonsMethodarerequiredtocomputearoot if f (x) isalinear function?
solution Remember that NewtonsMethodusesthelinear approximationof afunctiontoestimatethelocationof a
root. If theoriginal functionislinear, thenonlyoneiterationof NewtonsMethodwill berequiredtocomputetheroot.
2. What happensinNewtonsMethodif your initial guesshappenstobeazeroof f ?
solution If x
0
happenstobeazeroof f, then
x
1
= x
0
f (x
0
)
f
(x
0
)
= x
0
0= x
0
;
inother words, everytermintheNewtonsMethodsequencewill remainx
0
.
3. What happensinNewtonsMethodif your initial guesshappenstobealocal minor maxof f ?
solution Assuming that the function is differentiable, then the derivative is zero at a local maximumor a local
minimum. If NewtonsMethodisstartedwithaninitial guesssuchthat f
(x
0
) = 0, thenNewtonsMethodwill fail in
thesensethat x
1
will not bedened. That is, thetangent linewill beparallel tothex-axisandwill never intersect it.
4. Isthefollowingareasonabledescriptionof NewtonsMethod: A root of theequationof thetangent linetof (x) is
usedasanapproximationtoaroot of f (x) itself? Explain.
solution Yes, thatisareasonabledescription. Theiterationformulafor NewtonsMethodwasderivedbysolvingthe
equationof thetangent linetoy = f (x) at x
0
for itsx-intercept.
Exercises
In this exercise set, all approximations should be carried out using Newtons Method.
In Exercises 16, apply Newtons Method to f (x) and initial guess x
0
to calculate x
1
, x
2
, x
3
.
1. f (x) = x
2
6, x
0
= 2
2
6anddene
x
n+1
= x
n
f (x
n
)
f
(x
n
)
= x
n
x
2
n
6
2x
n
.
Withx
0
= 2, wecompute
n 1 2 3
x
n
2.5 2.45 2.44948980
f (x) = x
2
3x +1, x
0
= 3
3. f (x) = x
3
10, x
0
= 2
3
10anddene
x
n+1
= x
n
f (x
n
)
f
(x
n
)
= x
n
x
3
n
10
3x
2
n
.
Withx
0
= 2wecompute
n 1 2 3
x
n
2.16666667 2.15450362 2.15443469
S E C T I ON 4.7 Newtons Method 237
f (x) = x
3
+x +1, x
0
= 1
5. f (x) = cosx 4x, x
0
= 1
solution Let f (x) = cosx 4x anddene
x
n+1
= x
n
f (x
n
)
f
(x
n
)
= x
n
cosx
n
4x
n
sinx
n
4
.
Withx
0
= 1wecompute
n 1 2 3
x
n
0.28540361 0.24288009 0.24267469
f (x) = 1x sinx, x
0
= 7
7. UseFigure6to chooseaninitial guess x
0
to theuniquereal root of x
3
+ 2x + 5 = 0andcomputetherst three
Newtoniterates.
2 1 2 1
x
y
FIGURE 6 Graphof y = x
3
+2x +5.
3
+2x +5anddene
x
n+1
= x
n
f (x
n
)
f
(x
n
)
= x
n
x
3
n
+2x
n
+5
3x
2
n
+2
.
Wetakex
0
= 1.4, basedonthegure, andthencalculate
n 1 2 3
x
n
1.330964467 1.328272820 1.328268856
Approximate a solution of sinx = cos2x in the interval
_
0,

2
_
to three decimal places. Then nd the exact
solutionandcomparewithyour approximation.
9. Approximatethepoint of intersectionof thegraphsy = x
2
+4+
1
x
andy =
2
x
2
tothreedecimal places(Figure7).
1 2
5
10
y
x
y = 2/x
2
y = x
2
+ 4 + 1/x
FIGURE 7 Graphsof x
2
+4+
1
x
and
2
x
2
.
solution Thepoint of intersectionisthesolutionof f (x) = x
2
+4+x
1
+2x
2
. Weuseaninitial guessof 0.5.
NewtonsMethod x
0
= 0.5(guess)
x
1
= 0.5
f (0.5)
f
(0.5)
x
1
0.560345
x
2
= 0.560345
f (0.560345)
f
(0.560345)
x
2
0.573460
x
3
= 0.573460
f (0.573460)
f
(0.573460)
x
3
0.573927
x
4
= 0.573927
f (0.573927)
f
(0.573927)
x
4
0.573928
Thepointof intersectionhasx-coordinate 0.574tothreedecimal places. Thecorrespondingy-coordinateis
2
0.574
2

6.070.
Therst positivesolutionof sinx = 0isx = . UseNewtonsMethodtocalculate tofour decimal places.
In Exercises 1114, approximate to three decimal places using Newtons Method and compare with the value from a
calculator.
11.

11
2
11, andlet x
0
= 3. NewtonsMethodyields:
n 1 2 3
x
n
3.33333333 3.31666667 3.31662479
A calculator yields3.31662479.
5
1/3
13. 2
7/3
solution Notethat 2
7/3
= 4 2
1/3
. Let f (x) = x
3
2, andlet x
0
= 1. NewtonsMethodyields:
n 1 2 3
x
n
1.33333333 1.26388889 1.25993349
Thus, 2
7/3
4 1.25993349= 5.03973397. A calculator yields5.0396842.
3
1/4
15. Approximatethelargest positiveroot of f (x) = x
4
6x
2
+ x + 5to within an error of at most 10
4
. Refer to
Figure5.
solution Figure5 fromthetext suggests thelargest positiveroot of f (x) = x
4
6x
2
+ x + 5 is near 2. So let
f (x) = x
4
6x
2
+x +5andtakex
0
= 2.
n 1 2 3 4
x
n
2.111111111 2.093568458 2.093064768 2.093064358
Thelargest positiveroot of x
4
6x
2
+x +5isapproximately2.093064358.
In Exercises 1619, approximate the root specied to three decimal places using Newtons Method. Use a plot to
choose an initial guess.
Largest positiveroot of f (x) = x
3
5x +1.
17. Negativeroot of f (x) = x
5
20x +10.
5
20x + 10. Thegraphof f (x) shownbelowsuggests takingx
0
= 2.2. Startingfrom
x
0
= 2.2, therst threeiteratesof NewtonsMethodare:
n 1 2 3
x
n
2.22536529 2.22468998 2.22468949
Thus, tothreedecimal places, thenegativeroot of f (x) = x
5
20x +10is2.225.
150
100
50
2.5 2.0 1.5 1.0 0.5
y
x
Positivesolutionof sin = 0.8.
19. Positivesolutionof 4cosx = x
2
.
solution Fromthe graph below, we see that the positive solution to the equation 4cosx = x
2
is approximately
x = 1.2. Choosingx
0
= 1.2, therst twoiteratesof NewtonsMethodappliedtof (x) = 4cosx x
2
are
n 1 2
x
n
1.20154 1.20154
Thus, tothreedecimal places, thepositivesolutionto4cosx = x
2
is1.20154.
1
0
2
0.5 1.0 1.5 2.0
3
4
5
6
x
y
4 cos x
x
2
Let x
1
, x
2
betheestimatestoaroot obtainedbyapplyingNewtonsMethodwithx
0
= 1tothefunctiongraphed
inFigure8. Estimatethenumerical valuesof x
1
andx
2
, anddrawthetangent linesusedtoobtainthem.
21. Findthesmallest positivevalueof x at whichy = x andy = tanx intersect. Hint: Drawaplot.
solution Hereisaplot of tanx andx onthesameaxes:
4 3 1 2
x
y
5
5
The rst intersection with x > 0 lies on the second branch of y = tanx, between x =
5
4
and x =
3
2
. Let
f (x) = tanx x. Thegraphsuggestsaninitial guessx
0
=
5
4
, fromwhichweget thefollowingtable:
n 1 2 3 4
x
n
6.85398 21.921 4480.8 7456.27
Thisisclearlyleadingnowhere, soweneedtotryabetter initial guess. Note: This happens with Newtons Methodit is
sometimes difcult to choose an initial guess. Wetrythepoint directlybetween
5
4
and
3
2
, x
0
=
11
8
:
n 1 2 3 4 5 6 7
x
n
4.64662 4.60091 4.54662 4.50658 4.49422 4.49341 4.49341
Therst point wherey = x andy = tanx crossisat approximatelyx = 4.49341, whichisapproximately1.4303.
In 1535, the mathematicianAntonio Fior challenged his rival Niccolo Tartaglia to solve this problem: A tree
stands12braccia high; it isbrokenintotwopartsat suchapoint that theheight of thepart left standingisthecube
root of thelengthof thepart cut away. What is theheight of thepart left standing? Showthat this is equivalent to
solvingx
3
+x = 12andndtheheighttothreedecimal places. Tartaglia, whohaddiscoveredthesecretof thecubic
equation, wasabletodeterminetheexact answer:
x =
_
3
_
2919+54
3
_
291954
__
3
9
23. Find(totwodecimal places) thecoordinatesof thepoint P inFigure9wherethetangent linetoy = cosx passes
throughtheorigin.
P
y = cos x
2
1
x
y
FIGURE 9
solution Let (x
r
, cos(x
r
)) bethecoordinates of thepoint P. Theslopeof thetangent lineis sin(x
r
), so weare
lookingfor atangent line:
y = sin(x
r
)(x x
r
) +cos(x
r
)
suchthat y = 0whenx = 0. Thisgivesustheequation:
sin(x
r
)(x
r
) +cos(x
r
) = 0.
Let f (x) = cosx + x sinx. Wearelookingfor therst point x = r wheref (r) = 0. Thesketchgivenindicates that
x
0
= 3/4wouldbeagoodinitial guess. ThefollowingtablegivessuccessiveNewtonsMethodapproximations:
n 1 2 3 4
x
n
2.931781309 2.803636974 2.798395826 2.798386046
Thepoint P hasapproximatecoordinates(2.7984, 0.941684).
Newtons Method is often used to determine interest rates in nancial calculations. In Exercises 2426, r denotes a yearly
interest rate expressed as a decimal (rather than as a percent).
If P dollarsaredepositedeverymonthinanaccount earninginterest at theyearlyrater, thenthevalueS of the
account after N yearsis
S = P
_
b
12N+1
b
b 1
_
whereb = 1+
r
12
Youhavedecidedtodeposit P = 100dollarsper month.
(a) DetermineS after 5yearsif r = 0.07(that is, 7%).
(b) Showthat to save$10,000after 5years, youmust earninterest at arater determinedbys theequationb
61
101b +100= 0. UseNewtonsMethodtosolvefor b. Thenndr. Notethat b = 1isaroot, but youwant theroot

satisfyingb > 1.
25. If youborrowL dollarsfor N yearsatayearlyinterestrater, your monthlypaymentof P dollarsiscalculatedusing
theequation
L = P
_
1b
12N
b 1
_
whereb = 1+
r
12
(a) FindP if L = $5000, N = 3, andr = 0.08(8%).
(b) Youareofferedaloanof L = $5000tobepaidbackover3yearswithmonthlypaymentsof P = $200. UseNewtons
Methodtocomputeb andndtheimpliedinterestrater of thisloan. Hint: Showthat(L/P)b
12N+1
(1+L/P)b
12N
+
1= 0.
solution
(a) b = (1+0.08/12) = 1.00667
P = L
_
b 1
1b
12N
_
= 5000
_
1.006671
11.00667
36
_
$156.69
(b) Startingfrom
L = P
_
1b
12N
b 1
_
,
dividebyP, multiplybyb 1, multiplybyb
12N
andcollect liketermstoarriveat
(L/P)b
12N+1
(1+L/P)b
12N
+1= 0.
SinceL/P = 5000/200= 25, wemust solve
25b
37
26b
36
+1= 0.
NewtonsMethodgivesb 1.02121and
r = 12(b 1) = 12(0.02121) 0.25452
Sotheinterest rateisaround25.45%.
If youdeposit P dollars inaretirement fundevery year for N years withtheintentionof thenwithdrawingQ
dollarsperyearforMyears, youmustearninterestatarater satisfyingP(b
N
1) = Q(1b
M
), whereb = 1+r.
Assumethat$2000isdepositedeachyear for 30yearsandthegoal istowithdraw$10,000per year for 25years. Use
NewtonsMethodtocomputeb andthenndr. Notethat b = 1isaroot, but youwant theroot satisfyingb > 1.
27. Thereisnosimpleformulafor thepositionat timet of aplanet P initsorbit (anellipse) aroundthesun. Introduce
theauxiliary circleandangle inFigure10(notethat P determines becauseit isthecentral angleof point B onthe
circle). Let a = OA ande = OS/OA (theeccentricityof theorbit).
(a) Showthat sector BSA hasarea(a
2
/2)( e sin).
(b) By KeplersSecondLaw, theareaof sector BSA isproportional tothetimet elapsedsincetheplanet passedpoint
A, andbecausethecirclehasareaa
2
, BSA hasarea(a
2
)(t /T ), whereT istheperiodof theorbit. DeduceKeplers
Equation:
2t
T
= e sin
(c) Theeccentricity of Mercurys orbit is approximately e = 0.2. UseNewtons Method to nd after aquarter of
Mercurys year has elapsed(t = T/4). Convert to degrees. Has Mercury coveredmorethanaquarter of its orbit at
t = T/4?
O
P
A
S
Auxiliary circle
Elliptical orbit
Sun
q
B
FIGURE 10
solution
(a) Thesector SAB is thesliceOAB withthetriangleOPS removed. OAB is acentral sector witharc andradius
OA = a, andthereforehasarea
a
2
2
. OPS isatrianglewithheight a sin andbaselengthOS = ea. Hence, theareaof
thesector is
a
2
2

1
2
ea
2
sin =
a
2
2
( e sin).
(b) SinceKeplers secondlawindicates that theareaof thesector is proportional tothetimet sincetheplanet passed
point A, weget
a
2
(t /T ) = a
2
/2( e sin)
2
t
T
= e sin.
(c) If t = T/4, thelast equationin(b) gives:
2
= e sin = 0.2sin.
Let f () = 0.2sin

2
. Wewill useNewtonsMethodtondthepoint wheref () = 0. Sinceaquarter of the
year onMercuryhaspassed, agoodrst estimate
0
wouldbe
2
.
n 1 2 3 4
x
n
1.7708 1.76696 1.76696 1.76696
Fromthepointof viewof theSun, Mercuryhastraversedanangleof approximately1.76696radians = 101.24
. Mercury
hasthereforetraveledmorethanonefourthof thewayaround(fromthepoint of viewof central angle) duringthistime.
Therootsof f (x) =
1
3
x
3
4x +1tothreedecimal placesare3.583, 0.251, and3.332(Figure11). Determine
theroottowhichNewtonsMethodconvergesfor theinitial choicesx
0
= 1.85, 1.7, and1.55. Theanswer showsthat
asmall changeinx
0
canhaveasignicant effect ontheoutcomeof NewtonsMethod.
29. What happenswhenyouapplyNewtonsMethodtondazeroof f (x) = x
1/3
? Notethat x = 0istheonlyzero.
1/3
. Dene
x
n+1
= x
n
f (x
n
)
f
(x
n
)
= x
n
x
1/3
n
1
3
x
2/3
n
= x
n
3x
n
= 2x
n
.
Takex
0
= 0.5. Thenthesequenceof iteratesis1, 2, 4, 8, 16, 32, 64, . . . That is, for any nonzerostartingvalue,
thesequenceof iteratesdivergesspectacularly, sincex
n
= (2)
n
x
0
. Thuslim
n
|x
n
| = lim
n
2
n
|x
0
| = .
What happens whenyouapply Newtons Methodto theequationx
3
20x = 0withtheunlucky initial guess
x
0
= 2?
31. NewtonsMethodcanbeusedtocomputereciprocalswithoutperformingdivision.Letc > 0andsetf (x) = x
1
c.
(a) Showthat x (f (x)/f
(x)) = 2x cx
2
.
(b) Calculatetherstthreeiteratesof NewtonsMethodwithc = 10.3andthetwoinitial guessesx
0
= 0.1andx
0
= 0.5.
(c) Explaingraphicallywhyx
0
= 0.5doesnot yieldasequenceconvergingto1/10.3.
solution
(a) Let f (x) =
1
x
c. Then
x
f (x)
f
(x)
= x
1
x
c
x
2
= 2x cx
2
.
(b) For c = 10.3, wehavef (x) =
1
x
10.3andthusx
n+1
= 2x
n
10.3x
2
n
.
Takex
0
= 0.1.
n 1 2 3
x
n
0.097 0.0970873 0.09708738
Takex
0
= 0.5.
n 1 2 3
x
n
1.575 28.7004375 8541.66654
(c) Thegraphisdisconnected. If x
0
= 0.5, (x
1
, f (x
1
)) isontheother portionof thegraph, whichwill never converge
toanypoint under NewtonsMethod.
In Exercises 32 and 33, consider a metal rod of length Lfastened at both ends. If you cut the rod and weld on an additional
segment of length m, leaving the ends xed, the rod will bow up into a circular arc of radius R (unknown), as indicated
in Figure 12.
R
h
q
L
FIGURE 12 Theboldcircular archaslengthL +m.
Let h bethemaximumvertical displacement of therod.
(a) Showthat L = 2Rsin andconcludethat
h =
L(1cos)
2sin
(b) Showthat L +m = 2R andthenprove
sin
=
L
L +m
33. Let L = 3andm = 1. ApplyNewtonsMethodtoEq. (2) toestimate, andusethistoestimateh.
solution Welet L = 3andm = 1. Wewant thesolutionof:
sin
=
L
L +m
sin
L
L +m
= 0
sin
3
4
= 0.
Let f () =
sin

3
4
.
1.5 0.5 1
x
y
0.2
0.2
0.2
0.1
Thegureabovesuggeststhat
0
= 1.5wouldbeagoodinitial guess. TheNewtonsMethodapproximationsfor the
solutionfollow:
n 1 2 3 4
n
1.2854388 1.2757223 1.2756981 1.2756981
Theanglewhere
sin
=
L
L+m
isapproximately1.2757. Hence
h = L
1cos
2sin
1.11181.
QuadraticConvergencetoSquareRoots Let f (x) = x
2
c andlet e
n
= x
n
c betheerror inx
n
.
(a) Showthat x
n+1
=
1
2
(x
n
+c/x
n
) ande
n+1
= e
2
n
/2x
n
.
(b) Showthat if x
0
>
c, thenx
n
>
c for all n. Explaingraphically.

(c) Showthat if x
0
>
c, thene
n+1
e
2
n
/(2
c).
4.8 Antiderivatives
1. Findanantiderivativeof thefunctionf (x) = 0.
solution Since the derivative of any constant is zero, any constant function is an antiderivative for the function
f (x) = 0.
2. Isthereadifferencebetweenndingthegeneral antiderivativeof afunctionf (x) andevaluating
_
f (x) dx?
solution Nodifference. Theindeniteintegral isthesymbol for denotingthegeneral antiderivative.
3. J acqueswastoldthat f (x) andg(x) havethesamederivative, andhewonderswhether f (x) = g(x). DoesJ acques
havesufcient informationtoanswer hisquestion?
solution No. Knowingthat thetwofunctionshavethesamederivativeisonly goodenoughtotell J acquesthat the
functionsmaydiffer byatmostanadditiveconstant. Todeterminewhether thefunctionsareequal for all x, J acquesneeds
toknowthevalueof eachfunctionfor asinglevalueof x. If thetwofunctionsproducethesameoutput valuefor asingle
input value, theymust takethesamevaluefor all input values.
4. Supposethat F
(x) = f (x) andG
(x) = g(x). Whichof thefollowingstatementsaretrue? Explain.

(a) If f = g, thenF = G.
(b) If F andG differ byaconstant, thenf = g.
(c) If f andg differ byaconstant, thenF = G.
S E C T I ON 4.8 Antiderivatives 243
solution
(a) False. Evenif f (x) = g(x), theantiderivativesF andG maydiffer byanadditiveconstant.
(b) True. Thisfollowsfromthefact that thederivativeof anyconstant is0.
(c) False. If thefunctionsf andg aredifferent, thentheantiderivativesFandGdifferbyalinearfunction: F(x) G(x) =
ax +b for someconstantsa andb.
5. Isy = x asolutionof thefollowingInitial ValueProblem?
dy
dx
= 1, y(0) = 1
solution Although
d
dx
x = 1, thefunctionf (x) = x takesthevalue0whenx = 0, soy = x isnot asolutionof the
indicatedinitial valueproblem.
Exercises
In Exercises 18, nd the general antiderivative of f (x) and check your answer by differentiating.
1. f (x) = 18x
2
solution
_
18x
2
dx = 18
_
x
2
dx = 18
1
3
x
3
+C = 6x
3
+C.
Asacheck, wehave
d
dx
(6x
3
+C) = 18x
2
asneeded.
f (x) = x
3/5
3. f (x) = 2x
4
24x
2
+12x
1
solution
_
(2x
4
24x
2
+12x
1
) dx = 2
_
x
4
dx 24
_
x
2
dx +12
_
1
x
dx
= 2
1
5
x
5
24
1
3
x
3
+12ln|x| +C
=
2
5
x
5
8x
3
+12ln|x| +C.
Asacheck, wehave
d
dx
_
2
5
x
5
8x
3
+12ln|x| +C
_
= 2x
4
24x
2
+12x
1
asneeded.
f (x) = 9x +15x
2
5. f (x) = 2cosx 9sinx
solution
_
(2cosx 9sinx) dx = 2
_
cosx dx 9
_
sinx dx
= 2sinx 9(cosx) +C = 2sinx +9cosx +C
Asacheck, wehave
d
dx
(2sinx +9cosx +C) = 2cosx +9(sinx) = 2cosx 9sinx
asneeded.
f (x) = 4x
7
3cosx
7. f (x) = sin2x +12cos3x
solution
_
(sin2x +12cos3x) dx =
_
sin2x dx +12
_
cos3x dx =
1
2
cos2x +
12
3
sin3x +C
= 4sin3x
1
2
cos2x +C
Asacheck, wehave
d
dx
_
4sin3x
1
2
cos2x +C
_
= sin2x +12cos3x
f (x) = sin(49x)
9. Matchfunctions(a)(d) withtheir antiderivatives(i)(iv).
(a) f (x) = sinx (i) F(x) = cos(1x)
(b) f (x) = x sin(x
2
) (ii) F(x) = cosx
(c) f (x) = sin(1x) (iii) F(x) =
1
2
cos(x
2
)
(d) f (x) = x sinx (iv) F(x) = sinx x cosx
solution
(a) Anantiderivativeof sinx iscosx, whichis(ii). Asacheck, wehave
d
dx
(cosx) = (sinx) = sinx.
(b) An antiderivative of x sin(x
2
) is
1
2
cos(x
2
), which is (iii). This is because, by the Chain Rule, we have
d
dx
_
1
2
cos(x
2
)
_
=
1
2
_
sin(x
2
)
_
2x = x sin(x
2
).
(c) Anantiderivativeof sin(1x) iscos(1x) or (i). Asacheck, wehave
d
dx
cos(1 x) = sin(1 x) (1) =
sin(1x).
(d) Anantiderivativeof x sinx issinx x cosx, whichis(iv). Thisisbecause
d
dx
(sinx x cosx) = cosx (x (sinx) +cosx 1) = x sinx
In Exercises 1039, evaluate the indenite integral.
_
(9x +2) dx
11.
_
(418x) dx
solution
_
(418x) dx = 4x 9x
2
+C.
_
x
3
dx
13.
_
t
6/11
dt
solution
_
t
6/11
dt =
t
5/11
5/11
+C =
11
5
t
5/11
+C.
_
(5t
3
t
3
) dt
15.
_
(18t
5
10t
4
28t ) dt
solution
_
(18t
5
10t
4
28t ) dt = 3t
6
2t
5
14t
2
+C.
_
14s
9/5
ds
17.
_
(z
4/5
z
2/3
+z
5/4
) dz
solution
_
((z
4/5
z
2/3
+z
5/4
) dz =
z
1/5
1/5

z
5/3
5/3
+
z
9/4
9/4
+C = 5z
1/5
3
5
z
5/3
+
4
9
z
9/4
+C.
_
3
2
dx
19.
_
1
3
x
dx
solution
_
1
3
x
dx =
_
x
1/3
dx =
x
2/3
2/3
+C =
3
2
x
2/3
+C.
_
dx
x
4/3
21.
_
36dt
t
3
solution
_
36
t
3
dt =
_
36t
3
dt = 36
t
2
2
+C =
18
t
2
+C.
_
x(x
2
4) dx
23.
_
(t
1/2
+1)(t +1) dt
solution
_
(t
1/2
+1)(t +1) dt =
_
(t
3/2
+t +t
1/2
+1) dt
=
t
5/2
5/2
+
1
2
t
2
+
t
3/2
3/2
+t +C
=
2
5
t
5/2
+
1
2
t
2
+
2
3
t
3/2
+t +C
_
12z
z
dz
25.
_
x
3
+3x
2
4
x
2
dx
solution
_
x
3
+3x
2
4
x
2
dx =
_
(x +34x
2
) dx
=
1
2
x
2
+3x +4x
1
+C
_ _
1
3
sinx
1
4
cosx
_
dx
27.
_
12secx tanx dx
solution
_
12secx tanx dx = 12secx +C.
_
( +sec
2
) d
29.
_
(csct cott ) dt
solution
_
(csct cott ) dt = csct +C.
_
sin(7x 5) dx
31.
_
sec
2
(73) d
solution
_
sec
2
(73) d =
1
3
tan(73) +C.
_
( cos(1)) d
33.
_
25sec
2
(3z +1) dz
solution
_
25sec
2
(3z +1) dz =
25
3
tan(3z +1) +C.
_
(12cos4x +9sin3x) dx
35.
_
sec12t tan12t dt
solution
_
sec12t tan12t dt =
1
12
sec12t +C
_
5tan(4 +3) sec(4 +3) d
37.
_
sec4x(3sec4x 5tan4x) dx
solution
_
sec4x(3sec4x 5tan4x) dx = 3
_
sec
2
4x dx 5
_
sec4x tan4x dx =
3
4
tan4x
5
4
sec4x +C
_
sec(x +5) tan(x +5) dx
39.
_ _
cos(3)
1
2
sec
2
_
4
__
d
solution
_ _
cos(3)
1
2
sec
2
_
4
__
d =
1
3
sin(3) 2tan
_
4
_
+C.
InFigure2, isgraph(A) or graph(B) thegraphof anantiderivativeof f (x)?
41. InFigure3, whichof graphs(A), (B), and(C) isnot thegraphof anantiderivativeof f (x)? Explain.
f (x)
(C) (B) (A)
x
x
y
x
y
x
y
y
FIGURE 3
solution Let F(x) bean antiderivativeof f (x). Noticethat f (x) = F
(x) changes sign from to + to to +.

Hence, F(x) must transitionfromdecreasingtoincreasingtodecreasingtoincreasing.
Bothgraph(A) andgraph(C) meet thecriteriadiscussedaboveandonlydiffer byanadditiveconstant. Thuseither
couldbeanantiderivativeof f (x).
Graph(B) doesnothavethesamelocal extremaasindicatedbyf (x) andthereforeisnot anantiderivativeof f (x).
Showthat F(x) =
1
3
(x +13)
3
isanantiderivativeof f (x) = (x +13)
2
.
In Exercises 4346, verify by differentiation.
43.
_
(x +13)
6
dx =
1
7
(x +13)
7
+C
solution
d
dx
_
1
7
(x +13)
7
+C
_
= (x +13)
6
asrequired.
_
(x +13)
5
dx =
1
4
(x +13)
4
+C
45.
_
(4x +13)
2
dx =
1
12
(4x +13)
3
+C
solution
d
dx
_
1
12
(4x +13)
3
+C
_
=
1
4
(4x +13)
2
(4) = (4x +13)
2
asrequired.
_
(ax +b)
n
dx =
1
a(n +1)
(ax +b)
n+1
+C (for n = 1)
In Exercises 4762, solve the initial value problem.
47.
dy
dx
= x
3
, y(0) = 4
solution Since
dy
dx
= x
3
, wehave
y =
_
x
3
dx =
1
4
x
4
+C.
Thus,
4= y(0) =
1
4
0
4
+C = C,
sothat C = 4. Therefore, y =
1
4
x
4
+4.
dy
dt
= 32t , y(0) = 5
49.
dy
dt
= 2t +9t
2
, y(1) = 2
solution Since
dy
dt
= 2t +9t
2
, wehave
y =
_
(2t +9t
2
) dt = t
2
+3t
3
+C.
Thus,
2= y(1) = 1
2
+3(1)
3
+C,
sothat C = 2. Thereforey = t
2
+3t
3
2.
dy
dx
= 8x
3
+3x
2
, y(2) = 0
51.
dy
dt
=
t , y(1) = 1
solution Since
dy
dt
=
t = t
1/2
, wehave
y =
_
t
1/2
dt =
2
3
t
3/2
+C.
Thus
1= y(1) =
2
3
+C,
sothat C =
1
3
. Therefore, y =
2
3
t
3/2
+
1
3
.
dz
dt
= t
3/2
, z(4) = 1
53.
dy
dx
= (3x +2)
3
, y(0) = 1
solution Since
dy
dx
= (3x +2)
3
, wehave
y =
_
(3x +2)
3
dx =
1
4
1
3
(3x +2)
4
+C =
1
12
(3x +2)
4
+C.
Thus,
1= y(0) =
1
12
(2)
4
+C,
sothat C = 1
4
3
=
1
3
. Therefore, y =
1
12
(3x +2)
4
1
3
.
dy
dt
= (4t +3)
2
, y(1) = 0
55.
dy
dx
= sinx, y
_
2
_
= 1
solution Since
dy
dx
= sinx, wehave
y =
_
sinx dx = cosx +C.
Thus
1= y
_
2
_
= 0+C,
sothat C = 1. Therefore, y = 1cosx.
dy
dz
= sin2z, y
_
4
_
= 4
57.
dy
dx
= cos5x, y() = 3
solution Since
dy
dx
= cos5x, wehave
y =
_
cos5x dx =
1
5
sin5x +C.
Thus3= y() = 0+C, sothat C = 3. Therefore, y = 3+
1
5
sin5x.
dy
dx
= sec
2
3x, y
_
4
_
= 2
59.
dy
d
= 6sec3 tan3, y
_

12
_
= 4
solution Wehave
y =
_
6sec3 tan3 d = 2sec3 +C
Thus,
4= y
_

12
_
= 2sec3

12
+C = 2sec
4
+C = 2
2+C
sothat C = 42
2. Therefore, y = 2sec3 42
2.
dy
dt
= 4t sin2t , y(0) = 2
61.
dy
d
= cos
_
3
1
2
_
, y(3) = 8.
solution Wehave
y =
_
cos
_
3
1
2
_
d = 2sin
_
3
1
2
_
+C
Thus,
8= y(3) = 2sin
_
3
3
2
_
+C = 2sin
_
3
2
_
+C = 2+C
sothat C = 6. Therefore, y = 2sin
_
3
1
2
_
+6
dy
dx
=
1
x
2
csc
2
x, y
_
2
_
= 0
In Exercises 6369, rst nd f
and then nd f .
63. f
(x) = 12x, f
(0) = 1, f (0) = 2
solution Let f
(x) = 12x. Thenf
(x) = 6x
2
+ C. Givenf
(0) = 1, it follows that 1 = 6(0)

2
+ C andC = 1.
Thus, f
(x) = 6x
2
+ 1. Next, f (x) = 2x
3
+ x + C. Givenf (0) = 2, it followsthat 2 = 2(0)
3
+ 0+ C andC = 2.
Finally, f (x) = 2x
3
+x +2.
f
(x) = x
3
2x, f
(1) = 0, f (1) = 2
65. f
(x) = x
3
2x +1, f
(0) = 1, f (0) = 0
solution Let g(x) = f
(x). Thestatement gives us g
(x) = x
3
2x + 1, g(0) = 1. Fromthis, weget g(x) =
1
4
x
4
x
2
+ x + C. g(0) = 1givesus1= C, sof
(x) = g(x) =
1
4
x
4
x
2
+ x + 1. f
(x) =
1
4
x
4
x
2
+ x + 1, so
f (x) =
1
20
x
5
1
3
x
3
+
1
2
x
2
+x +C. f (0) = 0givesC = 0, so
f (x) =
1
20
x
5
1
3
x
3
+
1
2
x
2
+x.
f
(x) = x
3
2x +1, f
(1) = 0, f (1) = 4
67. f
(t ) = t
3/2
, f
(4) = 1, f (4) = 4
solution Letg(t ) = f
(t ).Theproblemstatementisg
(t ) = t
3/2
, g(4) = 1.Fromg
(t ) wegetg(t ) =
1
1/2
t
1/2
+
C = 2t
1/2
+ C. Fromg(4) = 1weget 1+ C = 1so that C = 2. Hencef
(t ) = g(t ) = 2t
1/2
+ 2. From
f
(t ) weget f (t ) = 2
1
1/2
t
1/2
+2t +C = 4t
1/2
+2t +C. Fromf (4) = 4weget 8+8+C = 4, sothat C = 4.
Hence, f (t ) = 4t
1/2
+2t +4.
f
() = cos, f
2
_
= 1, f
_
2
_
= 6
69. f
(t ) = t cost , f
(0) = 2, f (0) = 2
solution Let g(t ) = f
(t ). Theproblemstatement gives
g
(t ) = t cost, g(0) = 2.
Fromg
(t ), wegetg(t ) =
1
2
t
2
sint +C. Fromg(0) = 2, wegetC = 2. Hencef
(t ) = g(t ) =
1
2
t
2
sint +2. From
f
(t ), weget f (t ) =
1
2
(
1
3
t
3
) +cost +2t +C. Fromf (0) = 2, weget 1+C = 2, henceC = 3, and
f (t ) =
1
6
t
3
+cost +2t 3.
ShowthatF(x) = tan
2
x andG(x) = sec
2
x havethesamederivative. Whatcanyouconcludeabouttherelation
betweenF andG?Verifythisconclusiondirectly.
71. A particlelocated at theorigin at t = 1 s moves along thex-axis with velocity v(t ) = (6t
2
t ) m/s. Statethe
differential equationwithinitial conditionsatisedbythepositions(t ) of theparticle, andnds(t ).
solution Thedifferential equationsatisedbys(t ) is
ds
dt
= v(t ) = 6t
2
t,
andtheassociatedinitial conditioniss(1) = 0. Fromthedifferential equation, wend
s(t ) =
_
(6t
2
t ) dt = 2t
3
1
2
t
2
+C.
Usingtheinitial condition, it followsthat
0= s(1) = 2
1
2
+C so C =
3
2
.
Finally,
s(t ) = 2t
3
1
2
t
2
3
2
.
A particlemovesalongthex-axiswithvelocityv(t ) = (6t
2
t ) m/s. Findtheparticlespositions(t ) assuming
that s(2) = 4.
73. Amassoscillatesattheendof aspring. Lets(t ) bethedisplacementof themassfromtheequilibriumpositionattime
t . Assumingthat themassislocatedat theoriginat t = 0andhasvelocity v(t ) = sin(t /2) m/s, statethedifferential
equationwithinitial conditionsatisedbys(t ), andnds(t ).
solution Thedifferential equationsatisedbys(t ) is
ds
dt
= v(t ) = sin(t /2),
andtheassociatedinitial conditioniss(0) = 0. Fromthedifferential equation, wend
s(t ) =
_
sin(t /2) dt =
2
cos(t /2) +C.

Usingtheinitial condition, it followsthat
0= s(0) =
2
+C so C =
2
.
Finally,
s(t ) =
2
(1cos(t /2)).
Beginningat t = 0withinitial velocity 4m/s, aparticlemovesinastraight linewithaccelerationa(t ) = 3t
1/2
m/s
2
. Findthedistancetraveledafter 25seconds.
75. A car traveling25m/sbeginstodecelerateat aconstant rateof 4m/s
2
. After howmanysecondsdoesthecar come
toastopandhowfar will thecar havetraveledbeforestopping?
solution Sincetheaccelerationof thecar isaconstant 4m/s
2
, v isgivenbythedifferential equation:
dv
dt
= 4, v(0) = 25.
From
dv
dt
, weget v(t ) =
_
4dt = 4t + C. Sincev(0)25, C = 25. Fromthis, v(t ) = 4t + 25
m
s
. Tondthetime
until thecar stops, wemust solvev(t ) = 0:
4t +25= 0
4t = 25
t = 25/4= 6.25s.
Nowwehaveadifferential equationfor s(t ). Sincewewant toknowhowfar thecar hastraveledfromthebeginningof
itsdecelerationat timet = 0, wehaves(0) = 0bydenition, so:
ds
dt
= v(t ) = 4t +25, s(0) = 0.
Fromthis, s(t ) =
_
(4t +25) dt = 2t
2
+25t +C. Sinces(0) = 0, wehaveC = 0, and
s(t ) = 2t
2
+25t.
At stoppingtimet = 0.25s, thecar hastraveled
s(6.25) = 2(6.25)
2
+25(6.25) = 78.125m.
At timet = 1s, aparticleistravelingat 72m/sandbeginstodecelerateat theratea(t ) = t
1/2
until it stops.
Howfar doestheparticletravel beforestopping?
77. A 900-kg rocket is released froma space station. As it burns fuel, the rockets mass decreases and its velocity
increases. Let v(m) bethevelocity (inmetersper second) asafunctionof massm. Findthevelocity whenm = 729if
dv/dm = 50m
1/2
. Assumethat v(900) = 0.
solution Since
dv
dm
= 50m
1/2
, we have v(m) =
_
50m
1/2
dm = 100m
1/2
+ C. Thus 0 = v(900) =
100
900+C = 3000+C, andC = 3000. Therefore, v(m) = 3000100
m. Accordingly,
v(729) = 3000100
729= 3000100(27) = 300meters/sec.

Aswater owsthroughatubeof radiusR = 10cm, thevelocityv of anindividual water particledependsonlyon
itsdistancer fromthecenterof thetube.Theparticlesatthewallsof thetubehavezerovelocityanddv/dr = 0.06r.
Determinev(r).
79. Verifythelinearitypropertiesof theindeniteintegral statedinTheorem4.
solution ToverifytheSumRule, let F(x) andG(x) beanyantiderivativesof f (x) andg(x), respectively. Because
d
dx
(F(x) +G(x)) =
d
dx
F(x) +
d
dx
G(x) = f (x) +g(x),
it followsthat F(x) +G(x) isanantiderivativeof f (x) +g(x); i.e.,
_
(f (x) +g(x)) dx =
_
f (x) dx +
_
g(x) dx.
ToverifytheMultiplesRule, againlet F(x) beanyantiderivativeof f (x) andlet c beaconstant. Because
d
dx
(cF(x)) = c
d
dx
F(x) = cf (x),
it followsthat cF(x) isandantiderivativeof cf (x); i.e.,
_
(cf (x)) dx = c
_
f (x) dx.
Findconstantsc
1
andc
2
suchthat F(x) = c
1
x sinx +c
2
cosx isanantiderivativeof f (x) = x cosx.
81. Findconstantsc
1
andc
2
suchthat F(x) = c
1
x cosx +c
2
sinx isanantiderivativeof f (x) = x sinx.
solution LetF(x) = c
1
x cosx +c
2
sinx. If F(x) istobeanantiderivativeof f (x), thenwemusthavef (x) = F
(x).
Therefore
x sinx = c
1
cosx c
1
x sinx +c
2
cosx = c
1
x sinx +(c
1
+c
2
) cosx
Equatingcoefcientsof x sinx yieldsc
1
= 1; equatingcoefcientsof cosx thengivesc
2
= 1.ThusF(x) = x cosx +
sinx.
SupposethatF
(x) = f (x) andG
(x) = g(x). IsittruethatF(x)G(x) isanantiderivativeof f (x)g(x)?Conrm

or provideacounterexample.
83. Supposethat F
(x) = f (x).
(a) Showthat
1
2
F(2x) isanantiderivativeof f (2x).
(b) Findthegeneral antiderivativeof f (kx) for k = 0.
solution Let F
(x) = f (x).
(a) BytheChainRule, wehave
d
dx
_
1
2
F(2x)
_
=
1
2
F
(2x) 2= F
(2x) = f (2x).
Thus
1
2
F(2x) isanantiderivativeof f (2x).
(b) For nonzeroconstant k, theChainRulesgives
d
dx
_
1
k
F (kx)
_
=
1
k
F
(kx) k = F
(kx) = f (kx).
Thus
1
k
F(kx) is an antiderivativeof f (kx). Hencethegeneral antiderivativeof f (kx) is
1
k
F(kx) + C, whereC is a
constant.
Findanantiderivativefor f (x) = |x|.
85. UsingTheorem1, provethat F
(x) = f (x) wheref (x) isapolynomial of degreen 1, thenF(x) isapolynomial

of degreen. Thenprovethat if g(x) isanyfunctionsuchthatg
(n)
(x) = 0, theng(x) isapolynomial of degreeat mostn.
solution SupposeF
(x) = f (x) wheref (x) isapolynomial of degreen 1. Now, weknowthat thederivativeof a

polynomial of degreen isapolynomial of degreen 1, andhenceanantiderivativeof apolynomial of degreen 1isa
polynomial of degreen. Thus, byTheorem1, F(x) candiffer fromapolynomial of degreen byat most aconstant term,
whichisstill apolynomial of degreen. Now, supposethat g(x) isanyfunctionsuchthat g
(n+1)
(x) = 0. Weknowthat
then +1-st derivativeof anypolynomial of degreeat most n iszero, sobyrepeatedapplicationof Theorem1, g(x) can
differ fromapolynomial of degreeatmostn byatmostaconstantterm. Hence, g(x) isapolynomial of degreeatmostn.
ThePowerRuleforantiderivativesdoesnotapplytof (x) = x
1
. Whichof thegraphsinFigure4couldplausibly
represent anantiderivativeof f (x) = x
1
?
In Exercises 16, estimate using the Linear Approximation or linearization, and use a calculator to estimate the error.
1. 8.1
1/3
2
1/3
, a = 8andx = 0.1. Thenf
(x) =
1
3
x
2/3
, f
(a) =
1
12
and, bytheLinearApproxima-
tion,
f = 8.1
1/3
2 f
(a)x =
1
12
(0.1) = 0.00833333.
Usingacalculator, 8.1
1/3
2= 0.00829885. Theerror intheLinearApproximationistherefore
|0.008298850.00833333| = 3.44510
5
.
1
4.1
1
2
3. 625
1/4
624
1/4
1/4
, a = 625 and x = 1. Then f
(x) =
1
4
x
3/4
, f
(a) =
1
500
and, by the Linear
Approximation,
f = 624
1/4
625
1/4
f
(a)x =
1
500
(1) = 0.002.
Thus625
1/4
624
1/4
0.002. Usingacalculator,
625
1/4
624
1/4
= 0.00200120.
Theerror intheLinearApproximationistherefore
|0.00200120(0.002)| = 1.20110
6
.
101
5.
1
1.02
1
anda = 1. Thenf (a) = 1, f
(x) = x
2
andf
(a) = 1. Thelinearizationof f (x) at

a = 1istherefore
L(x) = f (a) +f
(a)(x a) = 1(x 1) = 2x,

and
1
1.02
L(1.02) = 0.98. Usingacalculator,
1
1.02
= 0.980392, sotheerror intheLinearApproximationis
|0.9803920.98| = 3.92210
4
.
5
33
In Exercises 712, nd the linearization at the point indicated.
7. y =
x, a = 25
solution Let y =
x anda = 25. Theny(a) = 5, y
=
1
2
x
1/2
andy
(a) =
1
10
. Thelinearizationof y at a = 25is
therefore
L(x) = y(a) +y
(a)(x 25) = 5+
1
10
(x 25).
v(t ) = 32t 4t
2
, a = 2
9. A(r) =
4
3
r
3
, a = 3
solution Let A(r) =
4
3
r
3
anda = 3. ThenA(a) = 36, A
(r) = 4r
2
andA
(a) = 36. Thelinearizationof

A(r) at a = 3istherefore
L(r) = A(a) +A
(a)(r a) = 36 +36(r 3) = 36(r 2).

V(h) = 4h(2h)(42h), a = 1
11. P() = sin(3 +), a =

3
solution WehaveP(a) = sin(2) = 0, P
() = 3cos(3 + ), andP
(a) = 3cos2 = 3. Thelinearizationof

P() at a =

3
istherefore
L(h) = P
(a)(h a) +P(a) = 3
_
h

3
_
= 3h
R(t ) = tan
_
_
t
1
2
__
, a =
1
4
In Exercises 1318, use the Linear Approximation.
13. Thepositionof anobjectinlinear motionattimet iss(t ) = 0.4t
2
+(t +1)
1
. Estimatethedistancetraveledover
thetimeinterval [4, 4.2].
solution Lets(t ) = 0.4t
2
+(t +1)
1
, a = 4andt = 0.2. Thens
(t ) = 0.8t (t +1)
2
ands
(a) = 3.16. Using

theLinearApproximation, thedistancetraveledover thetimeinterval [4, 4.2] isapproximately
s = s(4.2) s(4) s
(a)t = 3.16(0.2) = 0.632.

A bondthat pays$10,000in6yearsisofferedfor saleat apriceP. ThepercentageyieldY of thebondis
Y = 100
_
_
10,000
P
_
1/6
1
_
Verifythat if P = $7500, thenY = 4.91%. Estimatethedropinyieldif thepricerisesto$7700.
15. WhenabuspassfromAlbuquerquetoLosAlamosispricedatp dollars, abuscompanytakesinamonthlyrevenue
of R(p) = 1.5p 0.01p
2
(inthousandsof dollars).
(a) EstimateR if thepricerisesfrom$50to$53.
(b) If p = 80, howwill revenuebeaffectedbyasmall increaseinprice? ExplainusingtheLinearApproximation.
solution
(a) If thepriceisraisedfrom$50to$53, thenp = 3and
R R
(50)p = (1.50.02(50))(3) = 1.5

Wethereforeestimateanincreaseof $1500inrevenue.
(b) BecauseR
(80) = 1.5 0.02(80) = 0.1, theLinear Approximationgives R 0.1p. A small increasein

pricewouldthusresult inadecreaseinrevenue.
A storesells80MP4playersper weekwhentheplayersarepricedat P = $75. Estimatethenumber N soldif P
israisedto$80, assumingthat dN/dP = 4. EstimateN if thepriceisloweredto$69.
17. Thecircumferenceof asphereismeasuredatC = 100cm. Estimatethemaximumpercentageerror inV if theerror
inC isat most 3cm.
solution Thevolumeof asphereisV =
4
3
r
3
andthecircumferenceisC = 2r, wherer istheradiusof thesphere.
Thus, r =
1
2
C and
V =
4
3
_
C
2
_
3
=
1
6
2
C
3
.
UsingtheLinearApproximation,
V
dV
dC
C =
1
2
2
C
2
C,
so
V
V
1
2
2
C
2
C
1
6
2
C
3
= 3
C
C
.
WithC = 100cmandC at most 3cm, weestimatethat themaximumpercentageerror inV is3
3
100
= 0.09, or 9%.
Showthat
_
a
2
+b a +
b
2a
if b issmall. Usethistoestimate
26andndtheerror usingacalculator.
19. VerifytheMVT for f (x) = x
1/3
on[1, 8].
solution Wehavef (1) = 1
1/3
= 1andf (8) = 8
1/3
=
1
2
, sotheMVT saysthat thereisc [1, 8] suchthat
f
(c) =
f (8) f (1)
81
=
1
14
Now, f
(x) =
1
3
x
4/3
. Solvingfor f
(x) =
1
14
gives
1
3
x
4/3
=
1
14
x
4/3
=
3
14
x
4/3
=
14
3
x =
_
14
3
_
3/4
3.175 [1, 8]
Showthat f (x) = 2x
3
+2x +sinx +1haspreciselyonereal root.
21. VerifytheMVT for f (x) = x +
1
x
on[2, 5].
solution Ontheinterval [2, 5], f (x) iscontinuousanddifferentiable, sotheMVT applies. Now, f
(x) = 1
1
x
2
, so
1
1
c
2
= f
(c) =
f (b) f (a)
b a
=
26
5

5
2
52
=
9
10
,
or
c
2
= 10 c =
10 [2, 5]
Supposethat f (1) = 5andf
(x) 2for x 1. UsetheMVT toshowthat f (8) 19.

23. UsetheMVT toprovethat if f
(x) 2for x > 0andf (0) = 4, thenf (x) 2x +4for all x 0.

solution Letx > 0.Becausef iscontinuouson[0, x] anddifferentiableon(0, x),theMeanValueTheoremguarantees
thereexistsac (0, x) suchthat
f
(c) =
f (x) f (0)
x 0
or f (x) = f (0) +xf
(c).
Now, wearegiventhat f (0) = 4andthat f
(x) 2for x > 0. Therefore, for all x 0,

f (x) 4+x(2) = 2x +4.
A functionf (x) has derivativef
(x) =
1
x
4
+1
. Whereontheinterval [1, 4] does f (x) takeonits maximum
value?
In Exercises 2530, nd the critical points and determine whether they are minima, maxima, or neither.
25. f (x) = x
3
4x
2
+4x
3
4x
2
+4x. Thenf
(x) = 3x
2
8x +4= (3x 2)(x 2), sothat x =
2
3
andx = 2are
critical points. Next, f
(x) = 6x 8, sof
(
2
3
) = 4 < 0andf
(2) = 4 > 0. Therefore, by theSecondDerivative

Test, f (
2
3
) isalocal maximumwhilef (2) isalocal minimum.
s(t ) = t
4
8t
2
27. f (x) = x
2
(x +2)
3
2
(x +2)
3
. Then
f
(x) = 3x
2
(x +2)
2
+2x(x +2)
3
= x(x +2)
2
(3x +2x +4) = x(x +2)
2
(5x +4),
sothat x = 0, x = 2andx =
4
5
arecritical points. Thesignof therst derivativeontheintervalssurroundingthe
critical pointsisindicatedinthetablebelow. Basedonthisinformation, f (2) isneither alocal maximumnor alocal
minimum, f (
4
5
) isalocal maximumandf (0) isalocal minimum.
Interval (, 2) (2,
4
5
) (
4
5
, 0) (0, )
Signof f
+ + +
f (x) = x
2/3
(1x)
29. g() = sin
2
+
2
+. Then
g
() = 2sin cos +1= 2sin2 +1,

sothecritical pointsare
=
3
4
+n
for all integers n. Becauseg
() 0 for all , it follows that g

_
3
4
+n
_
is neither alocal maximumnor alocal
minimumfor all integersn.
h() = 2cos2 +cos4
In Exercises 3138, nd the extreme values on the interval.
31. f (x) = x(10x), [1, 3]
solution Let f (x) = x(10x) = 10x x
2
. Thenf
(x) = 102x, sothat x = 5istheonlycritical point. Asthis

critical pointisnotintheinterval [1, 3], weonlyneedtocheckthevalueof f attheendpointstodeterminetheextreme
values. Becausef (1) = 11andf (3) = 21, themaximumvalueof f (x) = x(10 x) ontheinterval [1, 3] is21
whiletheminimumvalueis11.
f (x) = 6x
4
4x
6
, [2, 2]
33. g() = sin
2
cos, [0, 2]
2
cos. Then
g
() = 2sin cos +sin = sin(2cos +1) = 0

when = 0,
2
3
, ,
4
3
, 2. Thetablebelowlists thevalueof g at eachof thecritical points andtheendpoints of the
interval [0, 2]. Basedonthisinformation, theminimumvalueof g() ontheinterval [0, 2] is1andthemaximum
valueis
5
4
.
0 2/3 4/3 2
g() 1 5/4 1 5/4 1
R(t ) =
t
t
2
+t +1
, [0, 3]
35. f (x) = x
2/3
2x
1/3
, [1, 3]
2/3
2x
1/3
. Then f
(x) =
2
3
x
1/3
2
3
x
2/3
=
2
3
x
2/3
(x
1/3
1), so that thecritical
pointsarex = 0andx = 1. Withf (1) = 3, f (0) = 0, f (1) = 1andf (3) =
3
92
3
3 0.804, it followsthat
theminimumvalueof f (x) ontheinterval [1, 3] is1andthemaximumvalueis3.
f (x) = 4x tan
2
x,
_
4
,
3
_
37. f (x) = x x
3/2
, [0, 2]
solution Wehavef
(x) = 1
3
2
x
1/2
, sothat x =
4
9
istheonlycritical point. Then
f
_
4
9
_
=
4
27
0.148, f (0) = 0, f (2) 0.828
sothat themaximumvalueof f on[0, 2] is 0.148anditsminimumvalueis 0.828.
f (x) = secx cosx,
_
4
,
4
_ 39. Findthecritical pointsandextremevaluesof
f (x) = |x 1| +|2x 6| in[0, 8].
solution Let
f (x) = |x 1| +|2x 6| =
73x, x < 1
5x, 1 x < 3
3x 7, x 3
.
Thederivativeof f (x) isnever zerobut doesnot exist at thetransitionpointsx = 1andx = 3. Thus, thecritical points
of f arex = 1andx = 3. Withf (0) = 7, f (1) = 4, f (3) = 2andf (8) = 17, it followsthat theminimumvalueof
f (x) ontheinterval [0, 8] is2andthemaximumvalueis17.
Matchthedescriptionof f (x) withthegraphof itsderivative f
(x) inFigure1.
(a) f (x) isincreasingandconcaveup.
(b) f (x) isdecreasingandconcaveup.
(c) f (x) isincreasingandconcavedown.
In Exercises 4146, nd the points of inection.
41. y = x
3
4x
2
+4x
solution Let y = x
3
4x
2
+4x. Theny
= 3x
2
8x +4andy
= 6x 8. Thus, y
> 0andy isconcaveupfor

x >
4
3
, whiley
< 0andy isconcavedownfor x <

4
3
. Hence, thereisapoint of inectionat x =
4
3
.
y = x 2cosx
43. y =
x
2
x
2
+4
solution Let y =
x
2
x
2
+4
= 1
4
x
2
+4
. Theny
=
8x
(x
2
+4)
2
and
y
=
(x
2
+4)
2
(8) 8x(2)(2x)(x
2
+4)
(x
2
+4)
4
=
8(43x
2
)
(x
2
+4)
3
.
Thus, y
> 0andy isconcaveupfor
3
< x <
2
3
,
whiley
< 0andy isconcavedownfor

|x|
2
3
.
Hence, therearepointsof inectionat
x =
2
3
.
y =
x
(x
2
4)
1/3
45. f (x) =
x
3
x
x
2
+1
solution Wehave
f
(x) =
(x
2
+1)(3x
2
1) (x
3
x)(2x)
(x
2
+1)
=
x
4
+4x
2
1
(x
2
+1)
2
f
(x) =
(x
2
+1)
2
(4x
3
+8x) (x
4
+4x
2
1)(2(x
2
+1)(2x))
(x
2
+1)
4
=
4x(x
2
3)
(x
2
+1)
3
Sincethedenominatorof f
(x) isalwayspositive, f
(x) > 0andf (x) isconcaveupforx <
3andfor0< x <
3,
whilef
(x) < 0andf (x) isconcavedownfor
3 < x < 0andfor
3 < x. Thustherearepointsof inectionat

x = 0andat x =
3.
f (x) = sin2x 4cosx
In Exercises 4756, sketch the graph, noting the transition points and asymptotic behavior.
47. y = 12x 3x
2
solution Let y = 12x 3x
2
. Then y
= 12 6x and y
= 6. It follows that thegraph of y = 12x 3x

2
is
increasingfor x < 2, decreasingfor x > 2, hasalocal maximumat x = 2andisconcavedownfor all x. Because
lim
x
(12x 3x
2
) = ,
thegraphhasnohorizontal asymptotes. Therearealsonovertical asymptotes. Thegraphisshownbelow.
5
x
1 1 2 3 5 4
y
10
5
10
y = 8x
2
x
4
49. y = x
3
2x
2
+3
solution Lety = x
3
2x
2
+3. Theny
= 3x
2
4x andy
= 6x 4. Itfollowsthatthegraphof y = x
3
2x
2
+3
isincreasingfor x < 0andx >
4
3
, isdecreasingfor 0< x <
4
3
, hasalocal maximumat x = 0, hasalocal minimumat
x =
4
3
, isconcaveupfor x >
2
3
, isconcavedownfor x <
2
3
andhasapoint of inectionat x =
2
3
. Because
lim
x
(x
3
2x
2
+3) = and lim
x
(x
3
2x
2
+3) = ,
thegraphhasnohorizontal asymptotes. Therearealsonovertical asymptotes. Thegraphisshownbelow.
5
x
1 1 2
y
10
5
10
y = 4x x
3/2
51. y =
x
x
3
+1
solution Let y =
x
x
3
+1
. Then
y
=
x
3
+1x(3x
2
)
(x
3
+1)
2
=
12x
3
(x
3
+1)
2
and
y
=
(x
3
+1)
2
(6x
2
) (12x
3
)(2)(x
3
+1)(3x
2
)
(x
3
+1)
4
=
6x
2
(2x
3
)
(x
3
+1)
3
.
It followsthat thegraphof y =
x
x
3
+1
isincreasingfor x < 1and1< x <
3
_
1
2
, isdecreasingfor x >
3
_
1
2
, hasa
local maximumat x =
3
_
1
2
, isconcaveupfor x < 1andx >
3
2, isconcavedownfor 1< x < 0and0< x <

3
2
andhasapoint of inectionat x =
3
2. Notethat x = 1isnot aninectionpoint becausex = 1isnot inthedomain

of thefunction. Now,
lim
x
x
x
3
+1
= 0,
soy = 0isahorizontal asymptote. Moreover,
lim
x1
x
x
3
+1
= and lim
x1+
x
x
3
+1
= ,
sox = 1isavertical asymptote. Thegraphisshownbelow.
2
x
1 2 3 1 2 3
y
4
2
4
y =
x
(x
2
4)
2/3
53. y =
1
|x +2| +1
solution Let y =
1
|x +2| +1
. Because
lim
x
1
|x +2| +1
= 0,
thegraphof thisfunctionhasahorizontal asymptoteof y = 0. Thegraphhasnovertical asymptotesas|x +2| +1 1
for all x. Thegraphisshownbelow. Fromthisgraphweseethereisalocal maximumat x = 2.
0.8
x
4 2 6 8 2 4
y
1
0.4
0.2
0.6
y =
_
2x
3
55. y =
3sinx cosx on[0, 2]

solution Lety =
3sinx cosx.Theny
3cosx +sinx andy
3sinx +cosx. Itfollowsthatthegraph

of y =
3sinx cosx isincreasingfor 0< x < 5/6and11/6< x < 2, isdecreasingfor 5/6< x < 11/6, has
alocal maximumatx = 5/6, hasalocal minimumatx = 11/6, isconcaveupfor 0< x < /3and4/3< x < 2,
isconcavedownfor /3< x < 4/3andhaspointsof inectionat x = /3andx = 4/3. Thegraphisshownbelow.
x
1
1
4
5 6 2 3
y
1
y = 2x tanx on[0, 2]
57. Drawacurvey = f (x) for whichf
andf
havesignsasindicatedinFigure2.
x
2 0
- + + + + - - - - +
1 3 5
FIGURE 2
solution Thegurebelowdepictsacurvefor whichf
(x) andf
(x) havetherequiredsigns.
x
4 8 4
y
Findthedimensionsof acylindrical canwithabottombut notopof volume4m
3
that usestheleast amount of
metal.
59. A rectangular boxof height h withsquarebaseof sideb hasvolumeV = 4m
3
. Twoof thesidefacesaremadeof
material costing$40/m
2
. Theremainingsidescost $20/m
2
. Whichvaluesof b andh minimizethecost of thebox?
solution Becausethevolumeof theboxis
V = b
2
h = 4 it followsthat h =
4
b
2
.
Now, thecost of theboxis
C = 40(2bh) +20(2bh) +20b
2
= 120bh +20b
2
=
480
b
+20b
2
.
Thus,
C
(b) =
480
b
2
+40b = 0
whenb =
3
12meters. BecauseC(b) asb 0+andasb , itfollowsthatcostisminimizedwhenb =

3
12
metersandh =
1
3
3
12meters.
Thecornyieldonacertainfarmis
Y = 0.118x
2
+8.5x +12.9 (bushelsper acre)
wherex isthenumber of cornplantsper acre(inthousands). Assumethatcornseedcosts$1.25(per thousandseeds)
andthat corncanbesoldfor $1.50/bushel. Let P(x) betheprot (revenueminusthecost of seeds) at plantinglevel
x.
(a) ComputeP(x
0
) for thevaluex
0
that maximizesyieldY.
(b) Findthemaximumvalueof P(x). Doesmaximumyieldleadtomaximumprot?
61. A quantityN(T ) satises
dN
dt
=
2
t
8
t
2
for t 4(t indays). At whichtimeisN increasingmost rapidly?
solution N is increasingmost rapidly wheretheslopeis greatest, i.e. whereN
=
dN
dt
has alocal maximum. We
have
N
=
2
t
2
+
16
t
3
andN
(t ) = 0for t = 8. Signanalysisshowsthat N
(t ) changessignfrompositivetonegativeat t = 8, sothat infact

t = 8isalocal maximum.
A truck gets 10 miles per gallon of diesel fuel traveling along an interstatehighway at 50 mph. This mileage
decreasesby0.15mpgfor eachmileper hour increaseabove50mph.
(a) If thetruckdriver ispaid$30/hour anddiesel fuel costsP = $3/gal, whichspeedv between50and70mphwill
minimizethecost of atripalongthehighway? Noticethat theactual cost dependsonthelengthof thetrip, but the
optimal speeddoesnot.
(b) Plot cost asafunctionof v (choosethelengtharbitrarily) andverifyyour answer topart (a).
(c) Doyouexpecttheoptimal speedv toincreaseor decreaseif fuel costsgodowntoP = $2/gal?Plotthegraphs
of cost asafunctionof v for P = 2andP = 3onthesameaxisandverifyyour conclusion.
63. Findthemaximumvolumeof aright-circular coneplacedupside-downinaright-circular coneof radiusR = 3and
height H = 4asinFigure3. A coneof radiusr andheight h hasvolume
1
3
r
2
h.
R
H
FIGURE 3
solution Letr denotetheradiusandh theheightof theupsidedowncone. Bysimilar triangles, weobtaintherelation
4h
r
=
4
3
so h = 4
_
1
r
3
_
andthevolumeof theupsidedownconeis
V(r) =
1
3
r
2
h =
4
3
_
r
2
r
3
3
_
for 0 r 3. Thus,
dV
dr
=
4
3
_
2r r
2
_
,
andthecritical pointsarer = 0andr = 2. BecauseV(0) = V(3) = 0and
V (2) =
4
3
_
4
8
3
_
=
16
9
,
themaximumvolumeof aright-circular coneplacedupsidedowninaright-circular coneof radius3andheight 4is
16
9
.
RedoExercise63for arbitraryR andH.
65. Showthatthemaximumareaof aparallelogramADEF thatisinscribedinatriangleABC, asinFigure4, isequal
toone-half theareaof ABC.
D E
B
F
C A
FIGURE 4
solution Let denotethemeasureof angleBAC. Then theareaof theparallelogramis given by AD AF sin.
Now, supposethat
BE/BC = x.
Then, by similar triangles, AD = (1 x)AB, AF = DE = xAC, andtheareaof theparallelogrambecomes AB
ACx(1x) sin. Thefunctionx(1x) achievesitsmaximumvalueof
1
4
whenx =
1
2
. Thus, themaximumareaof a
parallelograminscribedinatriangleABC is
1
4
AB AC sin =
1
2
_
1
2
AB AC sin
_
=
1
2
(areaof ABC) .
A box of volume8m
3
withasquaretopandbottomisconstructedout of twotypesof metal. Themetal for the
topandbottomcosts$50/m
2
andthemetal for thesidescosts$30/m
2
. Findthedimensionsof theboxthat minimize
total cost.
67. Let f (x) beafunctionwhosegraphdoesnot passthroughthex-axisandlet Q = (a, 0). Let P = (x
0
, f (x
0
)) be
thepointonthegraphclosesttoQ(Figure5). ProvethatPQisperpendicular tothetangentlinetothegraphof x
0
. Hint:
Findtheminimumvalueof thesquare of thedistancefrom(x, f (x)) to(a, 0).
y
x
y = f (x)
P = (x
0
, f (x
0
))
Q = (a, 0)
FIGURE 5
solution Let P = (a, 0) andlet Q = (x
0
, f (x
0
)) bethepoint onthegraphof y = f (x) closest toP. Theslopeof
thesegment joiningP andQisthen
f (x
0
)
x
0
a
.
Now, let
q(x) =
_
(x a)
2
+(f (x))
2
,
thedistancefromthearbitrarypoint(x, f (x)) onthegraphof y = f (x) tothepointP.As(x
0
, f (x
0
)) isthepointclosest
toP, wemust have
q
(x
0
) =
2(x
0
a) +2f (x
0
)f
(x
0
)
_
(x
0
a)
2
+(f (x
0
))
2
= 0.
Thus,
f
(x
0
) =
x
0
a
f (x
0
)
=
_
f (x
0
)
x
0
a
_
1
.
Inother words, theslopeof thesegment joiningP andQisthenegativereciprocal of theslopeof thelinetangent tothe
graphof y = f (x) at x = x
0
; hence; thetwolinesareperpendicular.
Takeacircular pieceof paper of radius R, removeasector of angle (Figure6), andfoldtheremainingpiece
intoacone-shapedcup. Whichangle producesthecupof largest volume?
69. UseNewtonsMethodtoestimate
3
25tofour decimal places.

3
25anddene
x
n+1
= x
n
f (x
n
)
f
(x
n
)
= x
n
x
3
n
25
3x
2
n
.
Withx
0
= 3, wend
n 1 2 3
x
n
2.925925926 2.924018982 2.924017738
Thus, tofour decimal places
3
25= 2.9240.
UseNewtonsMethodtondaroot of f (x) = x
2
x 1tofour decimal places.
In Exercises 7184, calculate the indenite integral.
71.
_
_
4x
3
2x
2
_
dx
solution
_
(4x
3
2x
2
) dx = x
4
2
3
x
3
+C.
_
x
9/4
dx
73.
_
sin( 8) d
solution
_
sin( 8) d = cos( 8) +C.
_
cos(57) d
75.
_
(4t
3
12t
4
) dt
solution
_
(4t
3
12t
4
) dt = 2t
2
+4t
3
+C.
_
(9t
2/3
+4t
7/3
) dt
77.
_
sec
2
x dx
solution
_
sec
2
x dx = tanx +C.
_
tan3 sec3 d
79.
_
(y +2)
4
dy
solution
_
(y +2)
4
dy =
1
5
(y +2)
5
+C.
_
3x
3
9
x
2
dx
81.
_
(cos ) d
solution
_
(cos ) d = sin
1
2
2
+C.
_
sec
2
(1225) d
83.
_
8dx
x
3
solution
_
8dx
x
3
=
4
x
2
+C.
_
sin(4x 9) dx
In Exercises 8590, solve the differential equation with the given initial condition.
85.
dy
dx
= 4x
3
, y(1) = 4
solution Let
dy
dx
= 4x
3
. Then
y(x) =
_
4x
3
dx = x
4
+C.
Usingtheinitial conditiony(1) = 4, wendy(1) = 1
4
+C = 4, soC = 3. Thus, y(x) = x
4
+3.
dy
dt
= 3t
2
+cost , y(0) = 12
87.
dy
dx
= x
1/2
, y(1) = 1
solution Let
dy
dx
= x
1/2
. Then
y(x) =
_
x
1/2
dx = 2x
1/2
+C.
Usingtheinitial conditiony(1) = 1, wendy(1) = 2
1+C = 1, soC = 1. Thus, y(x) = 2x

1/2
1.
dy
dx
= sec
2
x, y
_
4
_
= 2
89.
dy
dt
= 1+ sin3t , y() =
solution Integratinggives
y(t ) =
_
(1+ sin3t ) dt = t

3
cos3t +C
Usingtheinitial condition, wend
= y() =

3
cos3 +C = +

3
+C
sothat C =
3
, andthusy(t ) = t

3
(1+cos3t )
dy
dt
= cos3t +sin4t , y
_
1
3
_
= 0
91. Findf (t ) if f
(t ) = 12t , f (0) = 2, andf
(0) = 1.
solution Supposef
(t ) = 12t . Then
f
(t ) =
_
f
(t ) dt =
_
(12t ) dt = t t
2
+C.
Usingtheinitial conditionf
(0) = 1, wendf
(0) = 00
2
+C = 1, soC = 1. Thus, f
(t ) = t t
2
1. Now,
f (t ) =
_
f
(t ) dt =
_
(t t
2
1) dt =
1
2
t
2
1
3
t
3
t +C.
Usingtheinitial conditionf (0) = 2, wendf (0) =
1
2
0
2
1
3
0
3
0+C = 2, soC = 2. Thus,
f (t ) =
1
2
t
2
1
3
t
3
t +2.
At timet = 0, adriver begins deceleratingat aconstant rateof 10m/s
2
andcomes to ahalt after traveling
500m. Findthevelocityat t = 0.
5 THE INTEGRAL
5.1 Approximating and Computing Area
1. What aretheright andleft endpointsif [2, 5] isdividedintosixsubintervals?
solution If theinterval [2, 5] is dividedintosix subintervals, thelengthof eachsubinterval is
52
6
=
1
2
. Theright
endpointsof thesubintervalsarethen
5
2
, 3,
7
2
, 4,
9
2
, 5, whiletheleft endpointsare2,
5
2
, 3,
7
2
, 4,
9
2
.
2. Theinterval [1, 5] isdividedintoeight subintervals.
(a) What istheleft endpoint of thelast subinterval?
(b) What aretheright endpointsof therst twosubintervals?
solution Notethat eachof the8subintervalshaslength
51
8
=
1
2
.
(a) Theleft endpoint of thelast subinterval is5
1
2
=
9
2
.
(b) Theright endpointsof therst twosubintervalsare1+
1
2
=
3
2
and1+2
_
1
2
_
= 2.
3. Whichof thefollowingpairsof sumsarenot equal?
(a)
4
i=1
i,
4
=1
(b)
4
j=1
j
2
,
5
k=2
k
2
(c)
4
j=1
j,
5
i=2
(i 1) (d)
4
i=1
i(i +1),
5
j=2
(j 1)j
solution
(a) Onlythenameof theindexvariablehasbeenchanged, sothesetwosumsare thesame.
(b) Thesetwosumsarenot thesame; thesecondsquaresthenumberstwothroughvewhiletherstsquaresthenumbers
onethroughfour.
(c) Thesetwosumsare thesame. Notethat wheni rangesfromtwothroughve, theexpressioni 1rangesfromone
throughfour.
(d) Thesetwosumsare thesame. Bothsumsare1 2+2 3+3 4+4 5.
4. Explain:
100
j=1
j =
100
j=0
j but
100
j=1
1isnot equal to
100
j=0
1.
solution Therst terminthesum
100
j=0
j isequal tozero, soit maybedropped. Morespecically,
100
j=0
j = 0+
100
j=1
j =
100
j=1
j.
Ontheother hand, therst termin
100
j=0
1isnot zero, sothistermcannot bedropped. Inparticular,
100
j=0
1= 1+
100
j=1
1=
100
j=1
1.
5. ExplainwhyL
100
R
100
for f (x) = x
2
on[3, 7].
solution On[3, 7], thefunctionf (x) = x
2
isadecreasingfunction; hence, foranysubinterval of [3, 7], thefunction
valueat theleft endpoint islarger thanthefunctionvalueat theright endpoint. Consequently, L
100
must belarger than
R
100
.
260
S E C T I ON 5.1 Approximating and Computing Area 261
Exercises
1. Figure15showsthevelocity of anobject over a3-mininterval. Determinethedistancetraveledover theintervals
[0, 3] and[1, 2.5] (remember toconvert fromkm/htokm/min).
3
min
km/h
2 1
20
30
10
FIGURE 15
solution Thedistancetraveledbytheobject canbedeterminedbycalculatingtheareaunderneaththevelocitygraph
over thespeciedinterval. Duringtheinterval [0, 3], theobject travels
_
10
60
__
1
2
_
+
_
25
60
_
(1) +
_
15
60
__
1
2
_
+
_
20
60
_
(1) =
23
24
0.96km.
Duringtheinterval [1, 2.5], it travels
_
25
60
__
1
2
_
+
_
15
60
__
1
2
_
+
_
20
60
__
1
2
_
=
1
2
= 0.5km.
Anostrich(Figure16) runswithvelocity20km/hfor 2minutes, 12km/hfor 3minutes, and40km/hfor another
minute. Computethetotal distancetraveledandindicatewithagraphhowthisquantitycanbeinterpretedasanarea.
3. A rainstormhit Portland, Maine, inOctober 1996, resultinginrecordrainfall. Therainfall rateR(t ) onOctober 21
isrecorded, incentimetersper hour, inthefollowingtable, wheret isthenumber of hourssincemidnight. Computethe
total rainfall duringthis24-hour periodandindicateonagraphhowthisquantitycanbeinterpretedasanarea.
t (h) 02 24 49 912 1220 2024
R(t ) (cm) 0.5 0.3 1.0 2.5 1.5 0.6
solution Over eachinterval, thetotal rainfall isthetimeinterval inhourstimestherainfall incentimetersper hour.
Thus
R = 2(0.5) +2(0.3) +5(1.0) +3(2.5) +8(1.5) +4(0.6) = 28.5cm.
Thegurebelowisagraphof therainfall asafunctionof time. Theareaof theshadedregionrepresentsthetotal rainfall.
0.5
1.0
1.5
2.0
2.5
5 10 15 20 25
y
x
Thevelocityof anobject isv(t ) = 12t m/s. UseEq. (2) andgeometrytondthedistancetraveledover thetime
intervals[0, 2] and[2, 5].
5. ComputeR
5
andL
5
over [0, 1] usingthefollowingvalues.
x 0 0.2 0.4 0.6 0.8 1
f (x) 50 48 46 44 42 40
solution x =
10
5
= 0.2. Thus,
L
5
= 0.2(50+48+46+44+42) = 0.2(230) = 46,
and
R
5
= 0.2(48+46+44+42+40) = 0.2(220) = 44.
Theaverageis
46+44
2
= 45.
ThisestimateisfrequentlyreferredtoastheTrapezoidal Approximation.
262 C HA P T E R 5 THE INTEGRAL
ComputeR
6
, L
6
, andM
3
toestimatethedistancetraveledover [0, 3] if thevelocityat half-secondintervalsisas
follows:
t (s) 0 0.5 1 1.5 2 2.5 3
v (m/s) 0 12 18 25 20 14 20
7. Let f (x) = 2x +3.
(a) ComputeR
6
andL
6
over [0, 3].
(b) Usegeometrytondtheexact areaA andcomputetheerrors|A R
6
| and|A L
6
| intheapproximations.
solution Let f (x) = 2x +3on[0, 3].
(a) Wepartition[0, 3] into 6equally-spacedsubintervals. Theleft endpoints of thesubintervals are
_
0,
1
2
, 1,
3
2
, 2,
5
2
_
whereastheright endpointsare
_
1
2
, 1,
3
2
, 2,
5
2
, 3
_
.
Let a = 0, b = 3, n = 6, x = (b a) /n =
1
2
, andx
k
= a +kx, k = 0, 1, . . . , 5(left endpoints). Then
L
6
=
5
k=0
f (x
k
)x = x
5
k=0
f (x
k
) =
1
2
(3+4+5+6+7+8) = 16.5.
Withx
k
= a +kx, k = 1, 2, . . . , 6(right endpoints), wehave
R
6
=
6
k=1
f (x
k
)x = x
6
k=1
f (x
k
) =
1
2
(4+5+6+7+8+9) = 19.5.
(b) Viageometry (seegurebelow), theexact areaisA =
1
2
(3) (6) + 3
2
= 18. Thus, L
6
underestimatesthetruearea
(L
6
A = 1.5), whileR
6
overestimatesthetruearea(R
6
A = +1.5).
0.5 1 1.5 2 2.5 3
3
6
9
x
y
Repeat Exercise7for f (x) = 203x over [2, 4].
9. CalculateR
3
andL
3
for f (x) = x
2
x +4 over [1, 4]
Thensketchthegraphof f andtherectangles that makeupeachapproximation. Is theareaunder thegraphlarger or
smaller thanR
3
? Isit larger or smaller thanL
3
?
2
x +4andset a = 1, b = 4, n = 3, x = (b a) /n = (41) /3= 1.
(a) Let x
k
= a +kx, k = 0, 1, 2, 3.
Selectingtheleft endpointsof thesubintervals, x
k
, k = 0, 1, 2, or {1, 2, 3}, wehave
L
3
=
2
k=0
f (x
k
)x = x
2
k=0
f (x
k
) = (1) (4+6+10) = 20.
Selectingtheright endpointsof thesubintervals, x
k
, k = 1, 2, 3, or {2, 3, 4}, wehave
R
3
=
3
k=1
f (x
k
)x = x
3
k=1
f (x
k
) = (1) (6+10+16) = 32.
(b) Hereareguresof thethreerectanglesthatapproximatetheareaunderthecurvef (x) overtheinterval [1, 4]. Clearly,
theareaunder thegraphislarger thanL
3
but smaller thanR
3
.
4
1.0 1.5 2.0 2.5 3.0 3.5
6
8
10
12
14
y
x
L
3
4
1.0 1.5 2.0 2.5 3.0 3.5
6
8
10
12
14
y
x
R
3
Let f (x) =
_
x
2
+1andx =
1
3
. Sketchthegraphof f (x) anddrawtheright-endpoint rectangleswhosearea
isrepresentedbythesum
6
i=1
f (1+ix)x.
11. EstimateR
3
, M
3
, andL
6
over [0, 1.5] for thefunctioninFigure17.
1
2
3
4
5
x
y
0.5 1 1.5
FIGURE 17
solution Let f (x) on [0,
3
2
] begiven by Figure17. For n = 3, x = (
3
2
0)/3 =
1
2
, {x
k
}
3
k=0
=
_
0,
1
2
, 1,
3
2
_
.
Therefore
R
3
=
1
2
3
k=1
f (x
k
) =
1
2
(2+1+2) = 2.5,
M
3
=
1
2
6
k=1
f
_
x
k

1
2
x
_
=
1
2
(3.25+1.25+1.25) = 2.875.
For n = 6, x = (
3
2
0)/6=
1
4
, {x
k
}
6
k=0
=
_
0,
1
4
,
1
2
,
3
4
, 1,
5
4
,
3
2
_
. Therefore
L
6
=
1
4
5
k=0
f (x
k
) =
1
4
(5+3.25+2+1.25+1+1.25) = 3.4375.
Calculatetheareaof theshadedrectanglesinFigure18. Whichapproximationdotheserectanglesrepresent?
In Exercises 1320, calculate the approximation for the given function and interval.
13. R
3
, f (x) = 7x, [3, 5]
solution Let f (x) = 7x on[3, 5]. For n = 3, x = (53)/3=
2
3
, and{x
k
}
3
k=0
=
_
3,
11
3
,
13
3
, 5
_
. Therefore
R
3
=
2
3
3
k=1
(7x
k
)
=
2
3
_
10
3
+
8
3
+2
_
=
2
3
(8) =
16
3
.
L
6
, f (x) =

6x +2, [1, 3]
15. M
6
, f (x) = 4x +3, [5, 8]
solution Let f (x) = 4x + 3on[5, 8]. For n = 6, x = (8 5)/6 =
1
2
, and{x
k
}
5
k=0
= {5.25, 5.75, 6.25, 6.75,
7.25, 7.75}. Therefore,
M
6
=
1
2
5
k=0
_
4x
k
+3
_
=
1
2
(24+26+28+30+32+34)
=
1
2
(174) = 87.
R
5
, f (x) = x
2
+x, [1, 1]
17. L
6
, f (x) = x
2
+3|x|, [2, 1]
2
+ 3|x| on[2, 1]. For n = 6, x = (1 (2))/6 =
1
2
, and{x
k
}
6
k=0
= {2, 1.5, 1,
0.5, 0, 0.5, 1}. Therefore
L
6
=
1
2
5
k=0
(x
2
k
+3|x
k
|) =
1
2
(10+6.75+4+1.75+0+1.75) = 12.125.
M
4
, f (x) =

x, [3, 5]
19. L
4
, f (x) = cos
2
x,
_
6
,

2
_
solution Let f (x) = cos
2
x on[
6
,

2
]. For n = 4,
x =
(/2/6)
4
=

12
and {x
k
}
4
k=0
=
_
6
,
4
,
3
,
5
12
,
2
_
.
Therefore
L
4
=

12
3
k=0
cos
2
x
k
0.410236.
M
4
, f (x) =
1
x
2
+1
, [1, 5]
In Exercises 2126, write the sum in summation notation.
21. 4
7
+5
7
+6
7
+7
7
+8
7
solution Therst termis4
7
, andthelast termis8
7
, soit seemsthekthtermisk
7
. Therefore, thesumis:
8
k=4
k
7
.
(2
2
+2) +(3
2
+3) +(4
2
+4) +(5
2
+5)
23. (2
2
+2) +(2
3
+2) +(2
4
+2) +(2
5
+2)
solution Therst termis2
2
+2, andthelast termis2
5
+2, soit seemsthesumlimitsare2and5, andthekthterm
is2
k
+2. Therefore, thesumis:
5
k=2
(2
k
+2).
_
1+1
3
+
_
2+2
3
+ +
_
n +n
3
25.
1
2 3
+
2
3 4
+ +
n
(n +1)(n +2)
solution Therst summandis
1
(1+1)(1+2)
. Thisshowsus
1
2 3
+
2
3 4
+ +
n
(n +1)(n +2)
=
n
i=1
i
(i +1)(i +2)
.
sin() +sin
_
2
_
+sin
_
3
_
+ +sin
_

n +1
_ 27. Calculatethesums:
(a)
5
i=1
9 (b)
5
i=0
4 (c)
4
k=2
k
3
solution
(a)
5
i=1
9= 9+9+9+9+9= 45. Alternatively,
5
i=1
9= 9
5
i=1
1= (9)(5) = 45.
(b)
5
i=0
4= 4+4+4+4+4+4= 24. Alternatively,
5
i=0
4= 4
5
i=0
= (4)(6) = 24.
(c)
4
k=2
k
3
= 2
3
+3
3
+4
3
= 99. Alternatively,
4
k=2
k
3
=
k=1
k
3
k=1
k
3
=
_
4
4
4
+
4
3
2
+
4
2
4
_
_
1
4
4
+
1
3
2
+
1
2
4
_
= 99.
Calculatethesums:
(a)
4
j=3
sin
_
j
2
_
(b)
5
k=3
1
k 1
(c)
2
j=0
3
j1
29. Let b
1
= 4, b
2
= 1, b
3
= 2, andb
4
= 4. Calculate:
(a)
4
i=2
b
i
(b)
2
j=1
(2
b
j
b
j
) (c)
3
k=1
kb
k
solution
(a)
4
i=2
b
i
= b
2
+b
3
+b
4
= 1+2+(4) = 1.
(b)
2
j=1
_
2
b
j
b
j
_
= (2
4
4) +(2
1
1) = 13.
(c)
3
k=1
kb
k
= 1(4) +2(1) +3(2) = 12.
Assumethat a
1
= 5,
10
i=1
a
i
= 20, and
10
i=1
b
i
= 7. Calculate:
(a)
10
i=1
(4a
i
+3) (b)
10
i=2
a
i
(c)
10
i=1
(2a
i
3b
i
)
31. Calculate
200
j=101
j. Hint: Writeasadifferenceof twosumsanduseformula(3).
solution
200
j=101
j =
200
j=1
j
100
j=1
j =
_
200
2
2
+
200
2
_
_
100
2
2
+
100
2
_
= 20,1005050= 15,050.
Calculate
30
j=1
(2j +1)
2
. Hint: Expandanduseformulas(3)(4).
In Exercises 3340, use linearity and formulas (3)(5) to rewrite and evaluate the sums.
33.
20
j=1
8j
3
solution
20
j=1
8j
3
= 8
20
j=1
j
3
= 8
_
20
4
4
+
20
3
2
+
20
2
4
_
= 8(44,100) = 352,800.
30
k=1
(4k 3)
35.
150
n=51
n
2
solution
150
n=51
n
2
=
150
n=1
n
2
50
n=1
n
2
=
_
150
3
3
+
150
2
2
+
150
6
_
_
50
3
3
+
50
2
2
+
50
6
_
= 1,136,27542,925= 1,093,350.
200
k=101
k
3
37.
50
j=0
j (j 1)
solution
50
j=0
j (j 1) =
50
j=0
(j
2
j) =
50
j=0
j
2
50
j=0
j
=
_
50
3
3
+
50
2
2
+
50
6
_
_
50
2
2
+
50
2
_
=
50
3
3

50
3
=
124,950
3
= 41,650.
Thepower sumformulaisusablebecause
50
j=0
j (j 1) =
50
j=1
j (j 1).
30
j=2
_
6j +
4j
2
3
_
39.
30
m=1
(4m)
3
solution
30
m=1
(4m)
3
=
30
m=1
(6448m+12m
2
m
3
)
= 64
30
m=1
148
30
m=1
m+12
30
m=1
m
2
30
m=1
m
3
= 64(30) 48
(30)(31)
2
+12
_
30
3
3
+
30
2
2
+
30
6
_
_
30
4
4
+
30
3
2
+
30
2
4
_
= 192022,320+113,460216,225= 123,165.
20
m=1
_
5+
3m
2
_
2
In Exercises 4144, use formulas (3)(5) to evaluate the limit.
41. lim
N
N
i=1
i
N
2
solution Let s
N
=
N
i=1
i
N
2
. Then,
s
N
=
N
i=1
i
N
2
=
1
N
2
N
i=1
i =
1
N
2
_
N
2
2
+
N
2
_
=
1
2
+
1
2N
.
Therefore, lim
N
s
N
=
1
2
.
lim
N
N
j=1
j
3
N
4
43. lim
N
N
i=1
i
2
i +1
N
3
solution Let s
N
=
N
i=1
i
2
i +1
N
3
. Then
s
N
=
N
i=1
i
2
i +1
N
3
=
1
N
3
i=1
i
2
i=1
i
i=1
1
=
1
N
3
__
N
3
3
+
N
2
2
+
N
6
_
_
N
2
2
+
N
2
_
+N
_
=
1
3
+
2
3N
2
.
Therefore, lim
N
s
N
=
1
3
.
lim
N
N
i=1
_
i
3
N
4

20
N
_
In Exercises 4550, calculate the limit for the given function and interval. Verify your answer by using geometry.
45. lim
N
R
N
, f (x) = 9x, [0, 2]
solution Let f (x) = 9x on [0, 2]. Let N beapositiveinteger and set a = 0, b = 2, and x = (b a)/N =
(2 0)/N = 2/N. Also, let x
k
= a + kx = 2k/N, k = 1, 2, . . . , N betheright endpointsof theN subintervalsof
[0, 2]. Then
R
N
= x
N
k=1
f (x
k
) =
2
N
N
k=1
9
_
2k
N
_
=
36
N
2
N
k=1
k =
36
N
2
_
N
2
2
+
N
2
_
= 18+
18
N
.
Theareaunder thegraphis
lim
N
R
N
= lim
N
_
18+
18
N
_
= 18.
Theregionunder thegraphisatrianglewithbase2andheight 18. Theareaof theregionisthen
1
2
(2)(18) = 18, which
agreeswiththevalueobtainedfromthelimit of theright-endpoint approximations.
lim
N
R
N
, f (x) = 3x +6, [1, 4]
47. lim
N
L
N
, f (x) =
1
2
x +2, [0, 4]
1
2
x +2on[0, 4]. Let N > 0beaninteger, andset a = 0, b = 4, andx = (40)/N =
4
N
.
Also, let x
k
= 0+kx =
4k
N
, k = 0, 1, . . . , N 1betheleft endpointsof theN subintervals. Then
L
N
= x
N1
k=0
f (x
k
) =
4
N
N1
k=0
_
1
2
_
4k
N
_
+2
_
=
8
N
N1
k=0
1+
8
N
2
N1
k=0
k
= 8+
8
N
2
_
(N 1)
2
2
+
N 1
2
_
= 12
4
N
.
Theareaunder thegraphis
lim
N
L
N
= 12.
Theregionunder thecurveover [0, 4] is atrapezoidwithbasewidth4andheights 2and4. Fromthis, weget that the
areais
1
2
(4)(2+4) = 12, whichagreeswiththeanswer obtainedfromthelimit of theleft-endpoint approximations.
lim
N
L
N
, f (x) = 4x 2, [1, 3]
49. lim
N
M
N
, f (x) = x, [0, 2]
solution Let f (x) = x on[0, 2]. Let N > 0beaninteger andset a = 0, b = 2, andx = (b a)/N =
2
N
. Also,
let x
k
= 0+(k
1
2
)x =
2k1
N
, k = 1, 2, . . . N, bethemidpointsof theN subintervalsof [0, 2]. Then
M
N
= x
N
k=1
f (x
k
) =
2
N
N
k=1
2k 1
N
=
2
N
2
N
k=1
(2k 1)
=
2
N
2
2
N
k=1
k N
=
4
N
2
_
N
2
2
+
N
2
_
2
N
= 2.
Theareaunder thecurveover [0, 2] is
lim
N
M
N
= 2.
Theregionunder thecurveover [0, 2] isatrianglewithbaseandheight 2, andthusarea2, whichagreeswiththeanswer
obtainedfromthelimit of themidpoint approximations.
lim
N
M
N
, f (x) = 124x, [2, 6]
51. Show, for f (x) = 3x
2
+4x over [0, 2], that
R
N
=
2
N
N
j=1
_
24j
2
N
2
+
16j
N
_
Thenevaluate lim
N
R
N
.
2
+4x on[0, 2]. Let N beapositiveinteger andset a = 0, b = 2, andx = (b a)/N =
(2 0)/N = 2/N. Also, let x
j
= a + jx = 2j/N, j = 1, 2, . . . , N betheright endpointsof theN subintervalsof
[0, 3]. Then
R
N
= x
N
j=1
f (x
j
) =
2
N
N
j=1
_
3
_
2j
N
_
2
+4
2j
N
_
=
2
N
N
j=1
_
12j
2
N
2
+
8j
N
_
Continuing, wend
R
N
=
24
N
3
N
j=1
j
2
+
16
N
2
N
j=1
j
=
24
N
3
_
N
3
3
+
N
2
2
+
N
6
_
+
16
N
2
_
N
2
2
+
N
2
_
= 16+
20
N
+
4
N
2
Thus,
lim
N
R
N
= lim
N
_
16+
20
N
+
4
N
2
_
= 16.
Show, for f (x) = 3x
3
x
2
over [1, 5], that
R
N
=
4
N
N
j=1
_
192j
3
N
3
+
128j
2
N
2
+
28j
N
+2
_
Thenevaluate lim
N
R
N
.
In Exercises 5360, nd a formula for R
N
and compute the area under the graph as a limit.
53. f (x) = x
2
, [0, 1]
2
ontheinterval [0, 1]. Thenx =
10
N
=
1
N
anda = 0. Hence,
R
N
= x
N
j=1
f (0+jx) =
1
N
N
j=1
j
2
1
N
2
=
1
N
3
_
N
3
3
+
N
2
2
+
N
6
_
=
1
3
+
1
2N
+
1
6N
2
and
lim
N
R
N
= lim
N
_
1
3
+
1
2N
+
1
6N
2
_
=
1
3
.
f (x) = x
2
, [1, 5]
55. f (x) = 6x
2
4, [2, 5]
2
4ontheinterval [2, 5]. Thenx =
52
N
=
3
N
anda = 2. Hence,
R
N
= x
N
j=1
f (2+jx) =
3
N
N
j=1
_
6
_
2+
3j
N
_
2
4
_
=
3
N
N
j=1
_
20+
72j
N
+
54j
2
N
2
_
= 60+
216
N
2
N
j=1
j +
162
N
3
N
j=1
j
2
= 60+
216
N
2
_
N
2
2
+
N
2
_
+
162
N
3
_
N
3
3
+
N
2
2
+
N
6
_
= 222+
189
N
+
27
N
2
and
lim
N
R
N
= lim
N
_
222+
189
N
+
27
N
2
_
= 222.
f (x) = x
2
+7x, [6, 11]
57. f (x) = x
3
x, [0, 2]
3
x ontheinterval [0, 2]. Thenx =
20
N
=
2
N
anda = 0. Hence,
R
N
= x
N
j=1
f (0+jx) =
2
N
N
j=1
_
_
2j
N
_
3
2j
N
_
=
2
N
N
j=1
_
8j
3
N
3

2j
N
_
=
16
N
4
N
j=1
j
3
4
N
2
N
j=1
j
=
16
N
4
_
N
4
4
+
N
3
2
+
N
2
2
_
4
N
2
_
N
2
2
+
N
2
_
= 2+
6
N
+
8
N
2
and
lim
N
R
N
= lim
N
_
2+
6
N
+
8
N
2
_
= 2.
f (x) = 2x
3
+x
2
, [2, 2]
59. f (x) = 2x +1, [a, b] (a, b constantswitha < b)
solution Let f (x) = 2x +1ontheinterval [a, b]. Thenx =
b a
N
. Hence,
R
N
= x
N
j=1
f (a +jx) =
(b a)
N
N
j=1
_
2
_
a +j
(b a)
N
_
+1
_
=
(b a)
N
(2a +1)
N
j=1
1+
2(b a)
2
N
2
N
j=1
j
=
(b a)
N
(2a +1)N +
2(b a)
2
N
2
_
N
2
2
+
N
2
_
= (b a)(2a +1) +(b a)
2
+
(b a)
2
N
and
lim
N
R
N
= lim
N
_
(b a)(2a +1) +(b a)
2
+
(b a)
2
N
_
= (b a)(2a +1) +(b a)
2
= (b
2
+b) (a
2
+a).
f (x) = x
2
, [a, b] (a, b constantswitha < b)
In Exercises 6164, describe the area represented by the limits.
61. lim
N
1
N
N
j=1
_
j
N
_
4
solution Thelimit
lim
N
R
N
= lim
N
1
N
N
j=1
_
j
N
_
4
representstheareabetweenthegraphof f (x) = x
4
andthex-axisover theinterval [0, 1].
lim
N
3
N
N
j=1
_
2+
3j
N
_
4
63. lim
N
5
N
N1
j=0
_
2+
5j
N
_
solution Thelimit
lim
N
L
N
= lim
N
5
N
N1
j=0
_
2+
5j
N
_
representstheareabetweenthegraphof y = x andthex-axisover theinterval [2, 3].
lim
N
2N
N
j=1
sin
_
3

4N
+
j
2N
_
In Exercises 6570, express the area under the graph as a limit using the approximation indicated (in summation notation),
but do not evaluate.
65. R
N
, f (x) = sinx over [0, ]
solution Let f (x) = sinx over [0, ] andset a = 0, b = , andx = (b a) /N = /N. Then
R
N
= x
N
k=1
f (x
k
) =

N
N
k=1
sin
_
k
N
_
.
Hence
lim
N
R
N
= lim
N
N
N
k=1
sin
_
k
N
_
istheareabetweenthegraphof f (x) = sinx andthex-axisover [0, ].
R
N
, f (x) = x
1
over [1, 7]
67. L
N
, f (x) =

2x +1over [7, 11]

2x +1over theinterval [7, 11]. Thenx =
117
N
=
4
N
anda = 7. Hence,
L
N
= x
N1
j=0
f (7+jx) =
4
N
N1
j=0
_
2(7+j
4
N
) +1
and
lim
N
L
N
= lim
N
4
N
N1
j=0
_
15+
8j
N
istheareabetweenthegraphof f (x) =

2x +1andthex-axisover [7, 11].
L
N
, f (x) = cosx over
_
8
,

4
_
69. M
N
, f (x) = tanx over
_
1
2
, 1
_
solution Let f (x) = tanx over theinterval [
1
2
, 1]. Thenx =
1
1
2
N
=
1
2N
anda =
1
2
. Hence
M
N
= x
N
j=1
f
_
1
2
+
_
j
1
2
_
x
_
=
1
2N
N
j=1
tan
_
1
2
+
1
2N
_
j
1
2
__
andso
lim
N
M
N
= lim
N
1
2N
N
j=1
tan
_
1
2
+
1
2N
_
j
1
2
__
istheareabetweenthegraphof f (x) = tanx andthex-axisover [
1
2
, 1].
M
N
, f (x) = x
2
over [3, 5]
71. Evaluate lim
N
1
N
N
j=1
_
1
_
j
N
_
2
byinterpretingit astheareaof part of afamiliar geometricgure.
solution Thelimit
lim
N
R
N
= lim
N
1
N
N
j=1
_
1
_
j
N
_
2
representstheareabetweenthegraphof y = f (x) =
_
1x
2
andthex-axisover theinterval [0, 1]. Thisistheportion
of thecircular diskx
2
+y
2
1that liesintherst quadrant. Accordingly, itsareais
1
4
(1)
2
=

4
.
2
and let R
N
, L
N
, and M
N
be the approximations for the interval [0, 1].
Showthat R
N
=
1
3
+
1
2N
+
1
6N
2
. Interpret thequantity
1
2N
+
1
6N
2
astheareaof aregion.
73. Showthat
L
N
=
1
3
1
2N
+
1
6N
2
, M
N
=
1
3
1
12N
2
ThenrankthethreeapproximationsR
N
, L
N
, andM
N
inorder of increasingaccuracy(useExercise72).
2
on[0, 1]. Let N beapositiveinteger andset a = 0, b = 1, andx = (b a) /N = 1/N.
Let x
k
= a +kx = k/N, k = 0, 1, . . . , N andlet x
k
= a +(k +
1
2
)x = (k +
1
2
)/N, k = 0, 1, . . . , N 1. Then
L
N
= x
N1
k=0
f (x
k
) =
1
N
N1
k=0
_
k
N
_
2
=
1
N
3
N1
k=1
k
2
=
1
N
3
_
(N 1)
3
3
+
(N 1)
2
2
+
N 1
6
_
=
1
3
1
2N
+
1
6N
2
M
N
= x
N1
k=0
f (x
k
) =
1
N
N1
k=0
_
k +
1
2
N
_
2
=
1
N
3
N1
k=0
_
k
2
+k +
1
4
_
=
1
N
3
N1
k=1
k
2
N1
k=1
k
+
1
4
N1
k=0
1
=
1
N
3
__
(N 1)
3
3
+
(N 1)
2
2
+
N 1
6
_
+
_
(N 1)
2
2
+
N 1
2
_
+
1
4
N
_
=
1
3
1
12N
2
Theerror of R
N
isgivenby
1
2N
+
1
6N
2
, theerror of L
N
isgivenby
1
2N
+
1
6N
2
andtheerror of M
N
isgivenby
1
12N
2
. Of thethreeapproximations, R
N
istheleast accurate, thenL
N
andnallyM
N
isthemost accurate.
For eachof R
N
, L
N
, andM
N
, ndthesmallest integer N for whichtheerror islessthan0.001.
In Exercises 7580, use the Graphical Insight on page 291 to obtain bounds on the area.
75. LetA betheareaunder f (x) =

x over [0, 1]. Provethat0.51 A 0.77bycomputingR
4
andL
4
. Explainyour
reasoning.
solution For n = 4, x =
10
4
=
1
4
and{x
i
}
4
i=0
= {0+ix} = {0,
1
4
,
1
2
,
3
4
, 1}. Therefore,
R
4
= x
4
i=1
f (x
i
) =
1
4
_
1
2
+
2
2
+
3
2
+1
_
0.768
L
4
= x
3
i=0
f (x
i
) =
1
4
_
0+
1
2
+
2
2
+
3
2
_
0.518.
Intheplot below, youcanseetherectangleswhoseareaisrepresentedbyL
4
under thegraphandthetopof thosewhose
areaisrepresentedbyR
4
abovethegraph. TheareaA under thecurveissomewherebetweenL
4
andR
4
, so
0.518 A 0.768.
L
4
, R
4
andthegraphof f (x).
UseR
5
andL
5
toshowthat theareaA under y = x
2
over [10, 13] satises0.0218 A 0.0244.
77. UseR
4
andL
4
toshowthat theareaA under thegraphof y = sinx over
_
0,

2
_
satises0.79 A 1.19.
solution Let f (x) = sinx. f (x) isincreasingover theinterval [0, /2], sotheInsight onpage291applies, which
indicatesthat L
4
A R
4
. For n = 4, x =
/20
4
=

8
and{x
i
}
4
i=0
= {0+ ix}
4
i=0
= {0,

8
,

4
,
3
8
,

2
}. From
this,
L
4
=

8
3
i=0
f (x
i
) 0.79, R
4
=

8
4
i=1
f (x
i
) 1.18.
HenceA isbetween0.79and1.19.
Left andRight endpoint approximationstoA.
Showthat theareaA under f (x) = x
1
over [1, 8] satises
1
2
+
1
3
+
1
4
+
1
5
+
1
6
+
1
7
+
1
8
A 1+
1
2
+
1
3
+
1
4
+
1
5
+
1
6
+
1
7
79. Showthat theareaA under y = x
1/4
over [0, 1] satisesL
N
A R
N
for all N. Useacomputer algebra
systemtocalculateL
N
andR
N
for N = 100and200, anddetermineA totwodecimal places.
solution On[0, 1], f (x) = x
1/4
isanincreasingfunction; therefore, L
N
A R
N
for all N. Wend
L
100
= 0.793988 and R
100
= 0.80399,
while
L
200
= 0.797074 and R
200
= 0.802075.
Thus, A = 0.80totwodecimal places.
Showthat theareaA under y = 4/(x
2
+ 1) over [0, 1] satisesR
N
A L
N
for all N. DetermineA toat
least threedecimal placesusingacomputer algebrasystem. Canyouguesstheexact valueof A?
81. Althoughtheaccuracyof R
N
generallyimprovesasN increases, thisneednot betruefor small valuesof N. Draw
thegraphof apositivecontinuousfunctionf (x) onaninterval suchthat R
1
iscloser thanR
2
totheexact areaunder the
graph. Cansuchafunctionbemonotonic?
solution Let beasmall positivenumber lessthan
1
4
. (Intheguresbelow, =
1
10
. But imagine beingvery tiny.)
Denef (x) on[0, 1] by
f (x) =
1 if 0 x <
1
2
1
2

x
if
1
2
x <
1
2
x

1
2
if
1
2
x <
1
2
+
1 if
1
2
+ x 1
Thenf iscontinuouson[0, 1]. (Again, just lookat thegures.)
Theexactareabetweenf andthex-axisisA = 1
1
2
bh = 1
1
2
(2)(1) = 1. (For =
1
10
, wehaveA =
9
10
.)
With R
1
= 1, theabsoluteerror is |E
1
| = |R
1
A| = |1(1)| = . (For =
1
10
, this absoluteerror is
|E
1
| =
1
10
.)
WithR
2
=
1
2
, theabsoluteerror is|E
2
| = |R
2
A| =

1
2
(1)

1
2
=
1
2
. (For =
1
10
, wehave
|E
2
| =
2
5
.)
Accordingly, R
1
iscloser totheexact areaA thanisR
2
. Indeed, thetinier is, themoredramatictheeffect.
For amonotonicfunction, thisphenomenoncannot occur. Successiveapproximationsfromeither sideget progres-
sivelymoreaccurate.
x
Right endpt approx, n = 1 Graph of f(x)
0.2 0.4 0.6 0.8 1
1
0.8
0.6
0.4
0.2
0
0.5 1
1
0.5
0
Right endpt approx, n = 2
0.5 1
1
0.5
0
Drawthegraphof apositivecontinuous functiononaninterval suchthat R
2
andL
2
arebothsmaller thanthe
exact areaunder thegraph. Cansuchafunctionbemonotonic?
83. Explaingraphically: The endpoint approximations are less accurate when f
(x) is large.
solution Whenf
islarge, thegraphof f issteeperandhencethereismoregapbetweenf andL

N
orR
N
. Recall that
thetoplinesegmentsof therectanglesinvolvedinanendpoint approximationconstituteapiecewiseconstant function.
If f
islarge, thenf isincreasingmorerapidlyandhenceislesslikeaconstant function.

1 2 4
1
2
3
0 x
y
Smaller f'
3 1 0 0 2 4
1
2
3
0 x
y
Larger f'
3
Provethat for anyfunctionf (x) on[a, b],
R
N
L
N
=
b a
N
(f (b) f (a))
85. Inthisexercise, weprovethat lim
N
R
N
and lim
N
L
N
exist andareequal if f (x) isincreasing[thecase
of f (x) decreasingissimilar]. Weusetheconcept of aleast upper bounddiscussedinAppendixB.
(a) ExplainwithagraphwhyL
N
R
M
for all N, M 1.
(b) By(a), thesequence{L
N
} isbounded, soit hasaleast upper boundL. Bydenition, L isthesmallest number such
that L
N
L for all N. Showthat L R
M
for all M.
(c) Accordingto(b), L
N
L R
N
for all N. UseEq. (8) toshowthat lim
N
L
N
= L and lim
N
R
N
= L.
solution
(a) Let f (x) bepositiveandincreasing, andlet N andM bepositiveintegers. Fromthegurebelowat theleft, wesee
that L
N
underestimatestheareaunder thegraphof y = f (x), whilefromthegurebelowat theright, weseethat R
M
overestimatestheareaunder thegraph. Thus, for all N, M 1, L
N
R
M
.
x
y
x
y
(b) Becausethesequence{L
N
} is bounded aboveby R
M
for any M, each R
M
is an upper bound for thesequence.
Furthermore, thesequence{L
N
} must havealeast upper bound, call it L. By denition, theleast upper boundmust be
nogreater thananyother upper bound; consequently, L R
M
for all M.
(c) SinceL
N
L R
N
, R
N
L R
N
L
N
, so|R
N
L| |R
N
L
N
|. Fromthis,
lim
N
|R
N
L| lim
N
|R
N
L
N
|.
ByEq. (8),
lim
N
|R
N
L
N
| = lim
N
1
N
|(b a)(f (b) f (a))| = 0,
so lim
N
|R
N
L| |R
N
L
N
| = 0, hence lim
N
R
N
= L.
Similarly, |L
N
L| = L L
N
R
N
L
N
, so
|L
N
L| |R
N
L
N
| =
(b a)
N
(f (b) f (a)).
Thisgivesusthat
lim
N
|L
N
L| lim
N
1
N
|(b a)(f (b) f (a))| = 0,
so lim
N
L
N
= L.
Thisproves lim
N
L
N
= lim
N
R
N
= L.
UseEq. (8) toshowthat if f (x) ispositiveandmonotonic, thentheareaA under itsgraphover [a, b] satises
|R
N
A|
b a
N
|f (b) f (a)|
In Exercises 87 and 88, use Eq. (9) to nd a value of N such that |R
N
A| < 10
4
for the given function and interval.
87. f (x) =

x, [1, 4]

x on[1, 4]. Thenb = 4, a = 1, and
|R
N
A|
41
N
(f (4) f (1)) =
3
N
(21) =
3
N
.
Weneed
3
N
< 10
4
, whichgivesN > 30,000. Thus|R
30,001
A| < 10
4
for f (x) =

x on[1, 4].
f (x) =
_
9x
2
, [0, 3]
89. Provethat if f (x) ispositiveandmonotonic, thenM
N
liesbetweenR
N
andL
N
andiscloser totheactual
areaunder thegraphthanbothR
N
andL
N
. Hint: Inthecasethat f (x) isincreasing, Figure19showsthatthepartof the
error inR
N
duetotheithrectangleis thesumof theareasA + B + D, andfor M
N
it is |B E|. Ontheother hand,
A E.
x
x
i 1
x
i midpoint
A
F
D
E
B
C
FIGURE 19
solution Supposef (x) ismonotonicincreasingontheinterval [a, b], x =
b a
N
,
{x
k
}
N
k=0
= {a, a +x, a +2x, . . . , a +(N 1)x, b}
and
_
x
k
_
N1
k=0
=
_
a +(a +x)
2
,
(a +x) +(a +2x)
2
, . . . ,
(a +(N 1)x) +b
2
_
.
Notethat x
i
< x
i
< x
i+1
impliesf (x
i
) < f (x
i
) < f (x
i+1
) for all 0 i < N becausef (x) ismonotoneincreasing.
Then
L
N
=
b a
N
N1
k=0
f (x
k
)
<
M
N
=
b a
N
N1
k=0
f (x
k
)
<
R
N
=
b a
N
N
k=1
f (x
k
)
Similarly, if f (x) ismonotonedecreasing,
L
N
=
b a
N
N1
k=0
f (x
k
)
>
M
N
=
b a
N
N1
k=0
f (x
k
)
>
R
N
=
b a
N
N
k=1
f (x
k
)
Thus, if f (x) ismonotonic, thenM

N
alwaysliesinbetweenR
N
andL
N
.
Now, as inFigure19, consider thetypical subinterval [x
i1
, x
i
] andits midpoint x
i
. Welet A, B, C, D, E, andF
be the areas as shown in Figure 19. Note that, by the fact that x
i
is the midpoint of the interval, A = D + E and
F = B +C. LetE
R
representtherightendpointapproximationerror ( = A+B +D), letE
L
representtheleftendpoint
approximationerror ( = C +F +E) andlet E
M
represent themidpoint approximationerror ( = |B E|).
If B > E, thenE
M
= B E. Inthiscase,
E
R
E
M
= A +B +D (B E) = A +D +E > 0,
soE
R
> E
M
, while
E
L
E
M
= C +F +E (B E) = C +(B +C) +E (B E) = 2C +2E > 0,
so E
L
> E
M
. Therefore, themidpoint approximationis moreaccuratethaneither theleft or theright endpoint
approximation.
If B < E, thenE
M
= E B. Inthiscase,
E
R
E
M
= A +B +D (E B) = D +E +D (E B) = 2D +B > 0,
sothat E
R
> E
M
while
E
L
E
M
= C +F +E (E B) = C +F +B > 0,
so E
L
> E
M
. Therefore, themidpoint approximationis moreaccuratethaneither theright or theleft endpoint
approximation.
If B = E, themidpoint approximationisexactlyequal tothearea.
Hence, for B < E, B > E, or B = E, themidpoint approximationismoreaccuratethaneither theleft endpoint or the
right endpoint approximation.
5.2 The Definite Integral
1. What is
_
5
3
dx [thefunctionisf (x) = 1]?
solution
_
5
3
dx =
_
5
3
1 dx = 1(53) = 2.
2. Let I =
_
7
2
f (x) dx, wheref (x) iscontinuous. Statewhether trueor false:
(a) I istheareabetweenthegraphandthex-axisover [2, 7].
(b) If f (x) 0, thenI istheareabetweenthegraphandthex-axisover [2, 7].
(c) If f (x) 0, thenI istheareabetweenthegraphof f (x) andthex-axisover [2, 7].
solution
(a) False.
_
b
a
f (x) dx isthesigned areabetweenthegraphandthex-axis.
(b) True.
(c) True.
3. Explaingraphically:
_

0
cosx dx = 0.
solution Becausecos( x) = cosx, thenegative areabetweenthegraphof y = cosx andthex-axis over
[
2
, ] exactlycancelsthepositive areabetweenthegraphandthex-axisover [0,

2
].
4. Whichisnegative,
_
5
1
8dx or
_
1
5
8dx?
solution Because5(1) = 4,
_
5
1
8dx isnegative.
Exercises
In Exercises 110, draw a graph of the signed area represented by the integral and compute it using geometry.
1.
_
3
3
2x dx
solution Theregionboundedby thegraphof y = 2x andthex-axis over theinterval [3, 3] consists of tworight
triangles. Onehasarea
1
2
(3)(6) = 9belowtheaxis, andtheother hasarea
1
2
(3)(6) = 9abovetheaxis. Hence,
_
3
3
2x dx = 99= 0.
3 2
2
4
6
1 1 2 3
2
4
6
x
y
S E C T I ON 5.2 The Definite Integral 275
_
3
2
(2x +4) dx
3.
_
1
2
(3x +4) dx
solution Theregionboundedby thegraphof y = 3x + 4andthex-axis over theinterval [2, 1] consists of two
right triangles. Onehasarea
1
2
(
2
3
)(2) =
2
3
belowtheaxis, andtheother hasarea
1
2
(
7
3
)(7) =
49
6
abovetheaxis. Hence,
_
1
2
(3x +4) dx =
49
6

2
3
=
15
2
.
2
2
1 1
2
4
8
6
x
y
_
1
2
4dx
5.
_
8
6
(7x) dx
solution Theregionboundedbythegraphof y = 7 x andthex-axisover theinterval [6, 8] consistsof tworight
triangles. Onetrianglehasarea
1
2
(1)(1) =
1
2
abovetheaxis, andtheother hasarea
1
2
(1)(1) =
1
2
belowtheaxis. Hence,
_
8
6
(7x) dx =
1
2
1
2
= 0.
1
8 6 4 2
0.5
0.5
1
x
y
_
3/2
/2
sinx dx
7.
_
5
0
_
25x
2
dx
solution Theregionboundedbythegraphof y =
_
25x
2
andthex-axisover theinterval [0, 5] isone-quarter of
acircleof radius5. Hence,
_
5
0
_
25x
2
dx =
1
4
(5)
2
=
25
4
.
5 4 3 2 1
3
4
5
1
2
x
y
_
3
2
|x| dx
9.
_
2
2
(2|x|) dx
solution Theregionboundedbythegraphof y = 2|x| andthex-axisover theinterval [2, 2] isatriangleabove
theaxiswithbase4andheight 2. Consequently,
_
2
2
(2|x|) dx =
1
2
(2)(4) = 4.
2 1 2 1
2
1
x
y
_
5
2
(3+x 2|x|) dx
11. Calculate
_
10
0
(8x) dx intwoways:
(a) Asthelimit lim
N
R
N
(b) Bysketchingtherelevant signedareaandusinggeometry
solution Let f (x) = 8x over [0, 10]. Consider theintegral
_
10
0
f (x) dx =
_
10
0
(8x) dx.
(a) Let N beapositiveinteger andset a = 0, b = 10, x = (b a) /N = 10/N. Also, let x
k
= a + kx = 10k/N,
k = 1, 2, . . . , N betheright endpointsof theN subintervalsof [0, 10]. Then
R
N
= x
N
k=1
f (x
k
) =
10
N
N
k=1
_
8
10k
N
_
=
10
N
k=1
1
10
N
k=1
k
=
10
N
_
8N
10
N
_
N
2
2
+
N
2
__
= 30
50
N
.
Hence lim
N
R
N
= lim
N
_
30
50
N
_
= 30.
(b) Theregionboundedbythegraphof y = 8x andthex-axisover theinterval [0, 10] consistsof tworighttriangles.
Onetrianglehasarea
1
2
(8)(8) = 32abovetheaxis, andtheother hasarea
1
2
(2)(2) = 2belowtheaxis. Hence,
_
10
0
(8x) dx = 322= 30.
2
2
4
6
8
4 6 8
10
y
x
Calculate
_
4
1
(4x 8) dx intwoways: Asthelimit lim
N
R
N
andusinggeometry.
In Exercises 13 and 14, refer to Figure 14.
y = f (x)
6 4 2
y
x
FIGURE 14 Thetwopartsof thegrapharesemicircles.
13. Evaluate: (a)
_
2
0
f (x) dx (b)
_
6
0
f (x) dx
solution Let f (x) begivenbyFigure14.
(a) Thedeniteintegral
_
2
0
f (x) dx isthesignedareaof asemicircleof radius1whichliesbelowthex-axis. Therefore,
_
2
0
f (x) dx =
1
2
(1)
2
=
2
.
(b) Thedeniteintegral
_
6
0
f (x) dx is thesigned areaof asemicircleof radius 1 which lies below thex-axis and a
semicircleof radius2whichliesabovethex-axis. Therefore,
_
6
0
f (x) dx =
1
2
(2)
2
1
2
(1)
2
=
3
2
.
Evaluate: (a)
_
4
1
f (x) dx (b)
_
6
1
|f (x)| dx
1 2 3 4 5
2
1
1
2
y = g(t)
t
y
FIGURE 15
15. Evaluate
_
3
0
g(t ) dt and
_
5
3
g(t ) dt .
solution
Theregionboundedbythecurvey = g(x) andthex-axisover theinterval [0, 3] iscomprisedof tworighttriangles,
one with area
1
2
below the axis, and one with area 2 above the axis. The denite integral is therefore equal to
2
1
2
=
3
2
.
Theregionboundedbythecurvey = g(x) andthex-axisover theinterval [3, 5] iscomprisedof another tworight
triangles, onewitharea1abovetheaxisandonewitharea1belowtheaxis. Thedeniteintegral isthereforeequal
to0.
Finda, b, andc suchthat
_
a
0
g(t ) dt and
_
c
b
g(t ) dt areaslargeaspossible.
17. DescribethepartitionP andthesetof samplepointsC for theRiemannsumshowninFigure16. Computethevalue
of theRiemannsum.
x
1 3 2.5 3.2 2 0.5 4.5 5
34.25
20
15
8
y
FIGURE 16
solution ThepartitionP isdenedby
x
0
= 0 < x
1
= 1 < x
2
= 2.5 < x
3
= 3.2 < x
4
= 5
Theset of samplepointsisgivenbyC = {c
1
= 0.5, c
2
= 2, c
3
= 3, c
4
= 4.5}. Finally, thevalueof theRiemannsumis
34.25(10) +20(2.51) +8(3.22.5) +15(53.2) = 96.85.
ComputeR(f, P, C) for f (x) = x
2
+x for thepartitionP andtheset of samplepointsC inFigure16.
In Exercises 1922, calculate the Riemann sum R(f, P, C) for the given function, partition, and choice of sample points.
Also, sketch the graph of f and the rectangles corresponding to R(f, P, C).
19. f (x) = x, P = {1, 1.2, 1.5, 2}, C = {1.1, 1.4, 1.9}
solution Let f (x) = x. With
P = {x
0
= 1, x
1
= 1.2, x
2
= 1.5, x
3
= 2} and C = {c
1
= 1.1, c
2
= 1.4, c
3
= 1.9},
weget
R(f, P, C) = x
1
f (c
1
) +x
2
f (c
2
) +x
3
f (c
3
)
= (1.21)(1.1) +(1.51.2)(1.4) +(21.5)(1.9) = 1.59.
Hereisasketchof thegraphof f andtherectangles.
0.5 1 1.5 2 2.5
0.5
1
2
1.5
x
y
f (x) = 2x +3, P = {4, 1, 1, 4, 8}, C = {3, 0, 2, 5}
21. f (x) = x
2
+x, P = {2, 3, 4.5, 5}, C = {2, 3.5, 5}
2
+x. With
P = {x
0
= 2, x
1
= 3, x
3
= 4.5, x
4
= 5} and C = {c
1
= 2, c
2
= 3.5, c
3
= 5},
weget
R(f, P, C) = x
1
f (c
1
) +x
2
f (c
2
) +x
3
f (c
3
)
= (32)(6) +(4.53)(15.75) +(54.5)(30) = 44.625.
Hereisasketchof thegraphof f andtherectangles.
5
10
15
20
25
30
y
x
5 1 4 2 3
f (x) = sinx, P =
_
0,

6
,

3
,

2
_
, C = {0.4, 0.7, 1.2}
In Exercises 2328, sketch the signed area represented by the integral. Indicate the regions of positive and negative area.
23.
_
5
0
(4x x
2
) dx
solution Hereisasketchof thesignedarearepresentedbytheintegral
_
5
0
(4x x
2
) dx.
1 2 3 4
5
4
2
2
4
y
x
_
/4
/4
tanx dx
25.
_
2
sinx dx
_
2
sinx dx.
0.4
0.8
1.2
7 5 3 1 6 4 2
0.4
x
y
_
3
0
sinx dx
27.
_
6
0
(|124x| 4) dx
_
6
0
(|124x| 4) dx:
+ +

1
2
2
0
4
6
8
2 3 4 5 6
x
y
_
2
2
(t
2
1)(t
2
4) dt
In Exercises 2932, determine the sign of the integral without calculating it. Draw a graph if necessary.
29.
_
1
2
x
4
dx
solution Theintegrand is always positive. Theintegral must thereforebepositive, sincethesigned areahas only
positivepart.
_
1
2
x
3
dx
31.
_
2
0
x sinx dx
solution Asyoucanseefromthegraphbelow, theareabelowtheaxisisgreater thantheareaabovetheaxis. Thus,
thedeniteintegral isnegative.
0.2
0.4
0.6
7 5 3 1 6 4 2
0.2
x
y
+
_
2
0
sinx
x
dx
In Exercises 3342, use properties of the integral and the formulas in the summary to calculate the integrals.
33.
_
4
0
(6t 3) dt
solution
_
4
0
(6t 3) dt = 6
_
4
0
t dt 3
_
4
0
1dt = 6
1
2
(4)
2
3(40) = 36.
_
2
3
(4x +7) dx
35.
_
9
0
x
2
dx
solution Byformula(5),
_
9
0
x
2
dx =
1
3
(9)
3
= 243.
_
5
2
x
2
dx
37.
_
1
0
(u
2
2u) du
solution
_
1
0
(u
2
2u) du =
_
1
0
u
2
du 2
_
1
0
u du =
1
3
(1)
3
2
_
1
2
_
(1)
2
=
1
3
1=
2
3
.
_
1/2
0
(12y
2
+6y) dy
39.
_
1
3
(7t
2
+t +1) dt
solution First, write
_
1
3
(7t
2
+t +1) dt =
_
0
3
(7t
2
+t +1) dt +
_
1
0
(7t
2
+t +1) dt
=
_
3
0
(7t
2
+t +1) dt +
_
1
0
(7t
2
+t +1) dt
Then,
_
1
3
(7t
2
+t +1) dt =
_
7
1
3
(3)
3
+
1
2
(3)
2
3
_
+
_
7
1
3
1
3
+
1
2
1
2
+1
_
=
_
63+
9
2
3
_
+
_
7
3
+
1
2
+1
_
=
196
3
.
_
3
3
(9x 4x
2
) dx
41.
_
1
a
(x
2
+x) dx
solution First,
_
b
0
(x
2
+x) dx =
_
b
0
x
2
dx +
_
b
0
x dx =
1
3
b
3
+
1
2
b
2
. Therefore
_
1
a
(x
2
+x) dx =
_
0
a
(x
2
+x) dx +
_
1
0
(x
2
+x) dx =
_
1
0
(x
2
+x) dx
_
a
0
(x
2
+x) dx
=
_
1
3
1
3
+
1
2
1
2
_
_
1
3
(a)
3
+
1
2
(a)
2
_
=
1
3
a
3
1
2
a
2
+
5
6
.
_
a
2
a
x
2
dx
In Exercises 4347, calculate the integral, assuming that
_
5
0
f (x) dx = 5,
_
5
0
g(x) dx = 12
43.
_
5
0
(f (x) +g(x)) dx
solution
_
5
0
(f (x) +g(x)) dx =
_
5
0
f (x) dx +
_
5
0
g(x) dx = 5+12= 17.
_
5
0
_
2f (x)
1
3
g(x)
_
dx
45.
_
0
5
g(x) dx
solution
_
0
5
g(x) dx =
_
5
0
g(x) dx = 12.
_
5
0
(f (x) x) dx
47. Isit possibletocalculate
_
5
0
g(x)f (x) dx fromtheinformationgiven?
solution It isnot possibletocalculate
_
5
0
g(x)f (x) dx fromtheinformationgiven.
Provebycomputingthelimit of right-endpoint approximations:
_
b
0
x
3
dx =
b
4
4
In Exercises 4954, evaluate the integral using the formulas in the summary and Eq. (9).
49.
_
3
0
x
3
dx
solution ByEq. (9),
_
3
0
x
3
dx =
3
4
4
=
81
4
.
_
3
1
x
3
dx
51.
_
3
0
(x x
3
) dx
solution
_
3
0
(x x
3
) dx =
_
3
0
x dx
_
3
0
x
3
dx =
1
2
3
2
1
4
3
4
=
63
4
.
_
1
0
(2x
3
x +4) dx
53.
_
1
0
(12x
3
+24x
2
8x) dx
solution
_
1
0
(12x
3
+24x
2
8x) dx = 12
_
1
0
x
3
dx +24
_
1
0
x
2
8
_
1
0
x dx
= 12
1
4
1
4
+24
1
3
1
3
8
1
2
1
2
= 3+84= 7
_
2
2
(2x
3
3x
2
) dx
In Exercises 5558, calculate the integral, assuming that
_
1
0
f (x) dx = 1,
_
2
0
f (x) dx = 4,
_
4
1
f (x) dx = 7
55.
_
4
0
f (x) dx
solution
_
4
0
f (x) dx =
_
1
0
f (x) dx +
_
4
1
f (x) dx = 1+7= 8.
_
2
1
f (x) dx
57.
_
1
4
f (x) dx
solution
_
1
4
f (x) dx =
_
4
1
f (x) dx = 7.
_
4
2
f (x) dx
In Exercises 5962, express each integral as a single integral.
59.
_
3
0
f (x) dx +
_
7
3
f (x) dx
solution
_
3
0
f (x) dx +
_
7
3
f (x) dx =
_
7
0
f (x) dx.
_
9
2
f (x) dx
_
9
4
f (x) dx
61.
_
9
2
f (x) dx
_
5
2
f (x) dx
solution
_
9
2
f (x) dx
_
5
2
f (x) dx =
_
_
5
2
f (x) dx +
_
9
5
f (x) dx
_
_
5
2
f (x) dx =
_
9
5
f (x) dx.
_
3
7
f (x) dx +
_
9
3
f (x) dx
In Exercises 6366, calculate the integral, assuming that f is integrable and
_
b
1
f (x) dx = 1b
1
for all b > 0.
63.
_
5
1
f (x) dx
solution
_
5
1
f (x) dx = 15
1
=
4
5
.
_
5
3
f (x) dx
65.
_
6
1
(3f (x) 4) dx
solution
_
6
1
(3f (x) 4) dx = 3
_
6
1
f (x) dx 4
_
6
1
1dx = 3(16
1
) 4(61) =
35
2
.
_
1
1/2
f (x) dx
67. Explainthedifferenceingraphical interpretationbetween
_
b
a
f (x) dx and
_
b
a
|f (x)| dx.
solution When f (x) takes on both positiveand negativevalues on [a, b],
_
b
a
f (x) dx represents thesigned area
betweenf (x) andthex-axis, whereas
_
b
a
|f (x)| dx represents thetotal (unsigned) areabetweenf (x) andthex-axis.
Anynegativelysignedareasthatwerepartof
_
b
a
f (x) dx areregardedaspositiveareasin
_
b
a
|f (x)| dx. Hereisagraphical
exampleof thisphenomenon.
20
2 4 4 2
10
30
10
x
Graph of f (x)
2 4 4 2
10
20
30
x
Graph of |f (x)|
Usethegraphical interpretationof thedeniteintegral toexplaintheinequality
_
b
a
f (x) dx

_
b
a
|f (x)| dx
wheref (x) iscontinuous. Explainalsowhy equality holdsif andonly if either f (x) 0for all x or f (x) 0for
all x.
69. Let f (x) = x. Findaninterval [a, b] suchthat
_
b
a
f (x) dx
=
1
2
and
_
b
a
|f (x)| dx =
3
2
solution If a > 0, thenf (x) 0for all x [a, b], so
_
b
a
f (x) dx
=
_
b
a
|f (x)| dx
bythepreviousexercise. Wendasimilar result if b < 0. Thus, wemust havea < 0andb > 0. Now,
_
b
a
|f (x)| dx =
1
2
a
2
+
1
2
b
2
.
Because
_
b
a
f (x) dx =
1
2
b
2
1
2
a
2
,
then
_
b
a
f (x) dx
=
1
2
|b
2
a
2
|.
If b
2
> a
2
, then
1
2
a
2
+
1
2
b
2
=
3
2
and
1
2
(b
2
a
2
) =
1
2
yielda = 1andb =

2. Ontheother hand, if b
2
< a
2
, then
1
2
a
2
+
1
2
b
2
=
3
2
and
1
2
(a
2
b
2
) =
1
2
yielda =
2andb = 1.
EvaluateI =
_
2
0
sin
2
x dx andJ =
_
2
0
cos
2
x dx as follows. First showwithagraphthat I = J. Then
provethat I +J = 2.
In Exercises 7174, calculate the integral.
71.
_
6
0
|3x| dx
solution Over theinterval, theregionbetweenthecurveandtheinterval [0, 6] consistsof twotrianglesabovethex
axis, eachof whichhasheight 3andwidth3, andsoarea
9
2
. Thetotal area, hencethedeniteintegral, is9.
6 5 4 3 2 1
1
2
3
x
y
Alternately,
_
6
0
|3x| dx =
_
3
0
(3x) dx +
_
6
3
(x 3) dx
= 3
_
3
0
dx
_
3
0
x dx +
_
_
6
0
x dx
_
3
0
x dx
_
3
_
6
3
dx
= 9
1
2
3
2
+
1
2
6
2
1
2
3
2
9= 9.
_
3
1
|2x 4| dx
73.
_
1
1
|x
3
| dx
solution
|x
3
| =
_
x
3
x 0
x
3
x < 0.
Therefore,
_
1
1
|x
3
| dx =
_
0
1
x
3
dx +
_
1
0
x
3
dx =
_
1
0
x
3
dx +
_
1
0
x
3
dx =
1
4
(1)
4
+
1
4
(1)
4
=
1
2
.
_
2
0
|x
2
1| dx
75. UsetheComparisonTheoremtoshowthat
_
1
0
x
5
dx
_
1
0
x
4
dx,
_
2
1
x
4
dx
_
2
1
x
5
dx
solution Ontheinterval [0, 1], x
5
x
4
, so, byTheorem5,
_
1
0
x
5
dx
_
1
0
x
4
dx.
Ontheother hand, x
4
x
5
for x [1, 2], so, bythesameTheorem,
_
2
1
x
4
dx
_
2
1
x
5
dx.
Provethat
1
3

_
6
4
1
x
dx
1
2
.
77. Provethat 0.0198
_
0.3
0.2
sinx dx 0.0296. Hint: Showthat 0.198 sinx 0.296for x in[0.2, 0.3].
solution For 0 x

6
0.52, we have
d
dx
(sinx) = cosx > 0. Hence sinx is increasing on [0.2, 0.3].
Accordingly, for 0.2 x 0.3, wehave
m = 0.198 0.19867 sin0.2 sinx sin0.3 0.29552 0.296= M
Therefore, bytheComparisonTheorem, wehave
0.0198= m(0.30.2) =
_
0.3
0.2
mdx
_
0.3
0.2
sinx dx
_
0.3
0.2
M dx = M(0.30.2) = 0.0296.
Provethat 0.277
_
/4
/8
cosx dx 0.363.
79. Provethat 0
_
/2
/4
sinx
x
dx
2
2
.
solution Let
f (x) =
sinx
x
.
Aswecanseeinthesketchbelow, f (x) isdecreasingontheinterval [/4, /2]. Thereforef (x) f (/4) for all x in
[/4, /2]. f (/4) =
2
, so:
_
/2
/4
sinx
x
dx
_
/2
/4
2
dx =

4
2
2
2
.
2
x
y
2/p
2/p
p/4 p/2
y =
sinx
x
Findupper andlower boundsfor
_
1
0
dx
_
5x
3
+4
.
81. Supposethat f (x) g(x) on[a, b]. BytheComparisonTheorem,
_
b
a
f (x) dx
_
b
a
g(x) dx. Isit alsotrue
that f
(x) g
(x) for x [a, b]? If not, giveacounterexample.

solution Theassertionf
(x) g
(x) isfalse. Consider a = 0, b = 1, f (x) = x, g(x) = 2. f (x) g(x) for all x in

theinterval [0, 1], but f
(x) = 1whileg
(x) = 0for all x.

Statewhether trueor false. If false, sketchthegraphof acounterexample.
(a) If f (x) > 0, then
_
b
a
f (x) dx > 0.
(b) If
_
b
a
f (x) dx > 0, thenf (x) > 0.
83. Explaingraphically: If f (x) isanoddfunction, then
_
a
a
f (x) dx = 0.
solution If f isanoddfunction, thenf (x) = f (x) for all x. Accordingly, for everypositivelysignedareainthe
right half-planewheref isabovethex-axis, thereisacorrespondingnegativelysignedareaintheleft half-planewhere
f is belowthex-axis. Similarly, for every negatively signedareaintheright half-planewheref is belowthex-axis,
thereisacorrespondingpositively signedareaintheleft half-planewheref isabovethex-axis. Weconcludethat the
net areabetweenthegraphof f andthex-axisover [a, a] is0, sincethepositivelysignedareasandnegativelysigned
areascancel eachother out exactly.
1 2 2
1
2
4
2
4
x
y
Compute
_
1
1
sin(sin(x))(sin
2
(x) +1) dx.
85. Let k andb bepositive. Show, bycomparingtheright-endpoint approximations, that
_
b
0
x
k
dx = b
k+1
_
1
0
x
k
dx
solution Let k and b beany positivenumbers. Let f (x) = x
k
on [0, b]. Sincef is continuous, both
_
b
0
f (x) dx
and
_
1
0
f (x) dx exist. Let N beapositiveinteger andset x = (b 0) /N = b/N. Let x
j
= a + jx = bj/N, j =
1, 2, . . . , N betherightendpointsof theN subintervalsof [0, b]. Thentheright-endpointapproximationto
_
b
0
f (x) dx =
_
b
0
x
k
dx is
R
N
= x
N
j=1
f (x
j
) =
b
N
N
j=1
_
bj
N
_
k
= b
k+1
1
N
k+1
N
j=1
j
k
.
Inparticular, if b = 1above, thentheright-endpoint approximationto
_
1
0
f (x) dx =
_
1
0
x
k
dx is
S
N
= x
N
j=1
f (x
j
) =
1
N
N
j=1
_
j
N
_
k
=
1
N
k+1
N
j=1
j
k
=
1
b
k+1
R
N
Inother words, R
N
= b
k+1
S
N
. Therefore,
_
b
0
x
k
dx = lim
N
R
N
= lim
N
b
k+1
S
N
= b
k+1
lim
N
S
N
= b
k+1
_
1
0
x
k
dx.
Supposethat f andg arecontinuousfunctionssuchthat, for all a,
_
a
a
f (x) dx =
_
a
a
g(x) dx
Giveanintuitive argument showingthat f (0) = g(0). Explainyour ideawithagraph.
87. Theorem4remainstruewithout theassumptiona b c. Verifythisfor thecasesb < a < c andc < a < b.
solution Theadditivity property of deniteintegrals states for a b c, wehave
_
c
a
f (x) dx =
_
b
a
f (x) dx +
_
c
b
f (x) dx.
Supposethat wehaveb < a < c. By theadditivity property, wehave
_
c
b
f (x) dx =
_
a
b
f (x) dx +
_
c
a
f (x) dx.
Therefore,
_
c
a
f (x) dx =
_
c
b
f (x) dx
_
a
b
f (x) dx =
_
b
a
f (x) dx +
_
c
b
f (x) dx.
Nowsupposethatwehavec < a < b. Bytheadditivityproperty, wehave
_
b
c
f (x) dx =
_
a
c
f (x) dx +
_
b
a
f (x) dx.
Therefore,
_
c
a
f (x) dx =
_
a
c
f (x) dx =
_
b
a
f (x) dx
_
b
c
f (x) dx =
_
b
a
f (x) dx +
_
c
b
f (x) dx.
Hencetheadditivity property holdsfor all real numbersa, b, andc, regardlessof their relationshipamongst each
other.
5.3 The Fundamental Theorem of Calculus, Part I
1. Supposethat F
(x) = f (x) andF(0) = 3, F(2) = 7.

(a) What istheareaunder y = f (x) over [0, 2] if f (x) 0?
(b) What isthegraphical interpretationof F(2) F(0) if f (x) takesonbothpositiveandnegativevalues?
solution
(a) If f (x) 0over [0, 2], thentheareaunder y = f (x) isF(2) F(0) = 73= 4.
(b) If f (x) takesonbothpositiveandnegativevalues, thenF(2) F(0) givesthesignedareabetweeny = f (x) and
thex-axis.
2. Supposethat f (x) isanegative functionwithantiderivativeF suchthat F(1) = 7andF(3) = 4. What isthearea
(apositivenumber) betweenthex-axisandthegraphof f (x) over [1, 3]?
solution
_
3
1
f (x) dx representsthesigned areaboundedbythecurveandtheinterval [1, 3]. Sincef (x) isnegative
on[1, 3],
_
3
1
f (x) dx isthenegativeof thearea. Therefore, if A istheareabetweenthex-axisandthegraphof f (x),
wehave:
A =
_
3
1
f (x) dx = (F(3) F(1)) = (47) = (3) = 3.
3. Arethefollowingstatementstrueor false? Explain.
(a) FTC I isvalidonlyfor positivefunctions.
(b) TouseFTC I, youhavetochoosetheright antiderivative.
(c) If youcannot ndanantiderivativeof f (x), thenthedeniteintegral doesnot exist.
solution
(a) False. TheFTC I isvalidfor continuousfunctions.
(b) False. TheFTC I worksfor anyantiderivativeof theintegrand.
(c) False. If youcannotndanantiderivativeof theintegrand, youcannotusetheFTCI toevaluatethedeniteintegral,
but thedeniteintegral maystill exist.
4. Evaluate
_
9
2
f
(x) dx wheref (x) isdifferentiableandf (2) = f (9) = 4.

solution Becausef isdifferentiable,
_
9
2
f
(x) dx = f (9) f (2) = 44= 0.

S E C T I ON 5.3 The Fundamental Theorem of Calculus, Part I 285
Exercises
In Exercises 14, sketch the region under the graph of the function and nd its area using FTC I.
1. f (x) = x
2
, [0, 1]
solution
0.2 0.4 0.6 0.8 1
0.2
0.4
0.8
0.6
1
x
y
Wehavethearea
A =
_
1
0
x
2
dx =
1
3
x
3
1
0
=
1
3
.
f (x) = 2x x
2
, [0, 2]
3. f (x) = x
2
, [1, 2]
solution
1.0
0.2
0.4
0.6
0.8
1.0
1.2 1.4 1.6 1.8 2.0
y
x
Wehavethearea
A =
_
2
1
x
2
dx =
x
1
1
2
1
=
1
2
+1=
1
2
.
f (x) = cosx,
_
0,

2
_
In Exercises 534, evaluate the integral using FTC I.
5.
_
6
3
x dx
solution
_
6
3
x dx =
1
2
x
2
6
3
=
1
2
(6)
2
1
2
(3)
2
=
27
2
.
_
9
0
2dx
7.
_
5
3
(3t 4) dt
solution
_
5
3
(3t 4) dt =
_
3
2
t
2
4t
_
5
3
=
3
2
5
2
4 5
3
2
(3)
2
+4 (3) = 8
_
4
2
(245u) du
9.
_
1
0
(4x 9x
2
) dx
solution
_
1
0
(4x 9x
2
) dx = (2x
2
3x
3
)
1
0
= (23) (00) = 1.
_
2
3
u
2
du
11.
_
2
0
(12x
5
+3x
2
4x) dx
solution
_
2
0
(12x
5
+3x
2
4x) dx = (2x
6
+x
3
2x
2
)
2
0
= (128+88) (0+00) = 128.
_
2
2
(10x
9
+3x
5
) dx
13.
_
0
3
(2t
3
6t
2
) dt
solution
_
0
3
(2t
3
6t
2
) dt =
_
1
2
t
4
2t
3
_
0
3
= (00)
_
81
2
54
_
=
27
2
.
_
1
1
(5u
4
+u
2
u) du
15.
_
4
0
y dy
solution
_
4
0
y dy =
_
4
0
y
1/2
dy =
2
3
y
3/2
4
0
=
2
3
(4)
3/2
2
3
(0)
3/2
=
16
3
.
_
8
1
x
4/3
dx
17.
_
1
1/16
t
1/4
dt
solution
_
1
1/16
t
1/4
dt =
4
5
t
5/4
1
1/16
=
4
5
1
40
=
31
40
.
_
1
4
t
5/2
dt
19.
_
3
1
dt
t
2
solution
_
3
1
dt
t
2
=
_
3
1
t
2
dt = t
1
3
1
=
1
3
+1=
2
3
.
_
4
1
x
4
dx
21.
_
1
1/2
8
x
3
dx
solution
_
1
1/2
8
x
3
dx =
_
1
1/2
8x
3
dx = 4x
2
1
1/2
= 4+16= 12.
_
1
2
1
x
3
dx
23.
_
2
1
(x
2
x
2
) dx
solution
_
2
1
(x
2
x
2
) dx =
_
1
3
x
3
+x
1
_
2
1
=
_
8
3
+
1
2
_
_
1
3
+1
_
=
11
6
.
_
9
1
t
1/2
dt
25.
_
27
1
t +1
t
dt
solution
_
27
1
t +1
t
dt =
_
27
1
(t
1/2
+t
1/2
) dt =
_
2
3
t
3/2
+2t
1/2
_
27
1
=
_
2
3
(81
3) +6
3
_
_
2
3
+2
_
= 60
3
8
3
.
_
1
8/27
10t
4/3
8t
1/3
t
2
dt
27.
_
3/4
/4
sin d
solution
_
3/4
/4
sin d = cos
3/4
/4
=
2
2
+
2
2
=
2.
_
4
2
sinx dx
29.
_
/2
0
cos
_
1
3
_
d
solution
_
/2
0
cos
_
1
3
_
d = 3sin
_
1
3
/2
0
=
3
2
.
_
5/8
/4
cos2x dx
31.
_
/6
0
sec
2
_
3t

6
_
dt
solution
_
/6
0
sec
2
_
3t

6
_
dt =
1
3
tan
_
3t

6
_
/6
0
=
1
3
_
3+
1
3
_
=
4
3
3
.
_
/6
0
sec tan d
33.
_
/10
/20
csc5x cot5x dx
solution
_
/10
/20
csc5x cot5x dx =
1
5
csc5x
/10
/20
=
1
5
_
1
2
_
=
1
5
(
21).
_
/14
/28
csc
2
7y dy
In Exercises 3540, write the integral as a sum of integrals without absolute values and evaluate.
35.
_
1
2
|x| dx
solution
_
1
2
|x| dx =
_
0
2
(x) dx +
_
1
0
x dx =
1
2
x
2
0
2
+
1
2
x
2
1
0
= 0
_
1
2
(4)
_
+
1
2
=
5
2
.
_
5
0
|3x| dx
37.
_
3
2
|x
3
| dx
solution
_
3
2
|x
3
| dx =
_
0
2
(x
3
) dx +
_
3
0
x
3
dx =
1
4
x
4
0
2
+
1
4
x
4
3
0
= 0+
1
4
(2)
4
+
1
4
3
4
0=
97
4
.
_
3
0
|x
2
1| dx
39.
_

0
|cosx| dx
solution
_

0
|cosx| dx =
_
/2
0
cosx dx +
_

/2
(cosx) dx = sinx
/2
0
sinx
/2
= 10(10) = 2.
_
5
0
|x
2
4x +3| dx
In Exercises 4144, evaluate the integral in terms of the constants.
41.
_
b
1
x
3
dx
solution
_
b
1
x
3
dx =
1
4
x
4
b
1
=
1
4
b
4
1
4
(1)
4
=
1
4
_
b
4
1
_
for anynumber b.
_
a
b
x
4
dx
43.
_
b
1
x
5
dx
solution
_
b
1
x
5
dx =
1
6
x
6
b
1
=
1
6
b
6
1
6
(1)
6
=
1
6
(b
6
1) for anynumber b.
_
x
x
(t
3
+t ) dt
45. Calculate
_
3
2
f (x) dx, where
f (x) =
_
12x
2
for x 2
x
3
for x > 2
solution
_
3
2
f (x) dx =
_
2
2
f (x) dx +
_
3
2
f (x) dx =
_
2
2
(12x
2
) dx +
_
3
2
x
3
dx
=
_
12x
1
3
x
3
_
2
2
+
1
4
x
4
3
2
=
_
12(2)
1
3
2
3
_
_
12(2)
1
3
(2)
3
_
+
1
4
3
4
1
4
2
4
=
128
3
+
65
4
=
707
12
.
Calculate
_
2
0
f (x) dx, where
f (x) =
_
cosx for x
cosx sin2x for x >
47. UseFTC I toshowthat
_
1
1
x
n
dx = 0if n isanoddwholenumber. Explaingraphically.
solution Wehave
_
1
1
x
n
dx =
x
n+1
n +1
1
1
=
(1)
n+1
n +1

(1)
n+1
n +1
.
Becausen isodd, n +1iseven, whichmeansthat (1)
n+1
= (1)
n+1
= 1. Hence
(1)
n+1
n +1

(1)
n+1
n +1
=
1
n +1
1
n +1
= 0.
Graphicallyspeaking, for anoddfunctionsuchasx
3
shownhere, thepositivelysignedareafromx = 0tox = 1cancels
thenegativelysignedareafromx = 1tox = 0.
0.5 1 1
0.5
0.5
1
0.5
1
x
y
Plot thefunctionf (x) = sin3x x. Findthepositiveroot of f (x) tothreeplacesanduseit tondthearea
under thegraphof f (x) intherst quadrant.
49. CalculateF(4) giventhat F(1) = 3andF
(x) = x
2
. Hint: ExpressF(4) F(1) asadeniteintegral.
solution ByFTC I,
F(4) F(1) =
_
4
1
x
2
dx =
4
3
1
3
3
= 21
ThereforeF(4) = F(1) +21= 3+21= 24.
CalculateG(16), wheredG/dt = t
1/2
andG(9) = 5.
51. Does
_
1
0
x
n
dx get larger or smaller asn increases? Explaingraphically.
solution Let n 0andconsider
_
1
0
x
n
dx. (Note: for n < 0theintegrandx
n
asx 0+, soweexcludethis
possibility.) Now
_
1
0
x
n
dx =
_
1
n +1
x
n+1
_
1
0
=
_
1
n +1
(1)
n+1
_
_
1
n +1
(0)
n+1
_
=
1
n +1
,
whichdecreasesasn increases. Recall that
_
1
0
x
n
dx representstheareabetweenthepositivecurvef (x) = x
n
andthe
x-axisover theinterval [0, 1]. Accordingly, thisareagetssmaller asn getslarger. Thisisreadilyevident inthefollowing
graph, whichshowscurvesfor several valuesof n.
1
y
1 0
1/4
1/2
1
2
4
8
x
Showthat theareaof theshadedparabolicarchinFigure7isequal tofour-thirdstheareaof thetriangleshown.
53. Proveafamousresultof Archimedes(generalizingExercise52): For r < s, theareaof theshadedregioninFigure8
isequal tofour-thirdstheareaof triangleACE, whereC isthepointontheparabolaatwhichthetangentlineisparallel
tosecant lineAE.
(a) Showthat C hasx-coordinate(r +s)/2.
(b) Showthat ABDE hasarea(s r)
3
/4byviewingit asaparallelogramof height s r andbaseof lengthCF.
(c) Showthat ACE hasarea(s r)
3
/8byobservingthat it hasthesamebaseandheight astheparallelogram.
(d) Computetheshadedareaastheareaunder thegraphminustheareaof atrapezoid, andproveArchimedes result.
r s
y
B
C
D
A
F
E
x
2
r + s
FIGURE 8 Graphof f (x) = (x a)(b x).
solution
(a) Theslopeof thesecant lineAE is
f (s) f (r)
s r
=
(s a)(b s) (r a)(b r)
s r
= a +b (r +s)
andtheslopeof thetangent linealongtheparabolais
f
(x) = a +b 2x.
If C is thepoint ontheparabolaat whichthetangent lineis parallel to thesecant lineAE, thenits x-coordinatemust
satisfy
a +b 2x = a +b (r +s) or x =
r +s
2
.
(b) ParallelogramABDE hasheight s r andbaseof lengthCF. Sincetheequationof thesecant lineAE is
y = [a +b (r +s)](x r) +(r a)(b r),
thelengthof thesegment CF is
_
r +s
2
a
__
b
r +s
2
_
[a +b (r +s)]
_
r +s
2
r
_
(r a)(b r) =
(s r)
2
4
.
Thus, theareaof ABDE is
(sr)
3
4
.
(c) TriangleACE iscomprisedof ACF andCEF. Eachof thesesmallertriangleshasheight
sr
2
andbaseof length
(sr)
2
4
. Thus, theareaof ACE is
1
2
s r
2

(s r)
2
4
+
1
2
s r
2

(s r)
2
4
=
(s r)
3
8
.
(d) Theareaunder thegraphof theparabolabetweenx = r andx = s is
_
s
r
(x a)(b x) dx =
_
abx +
1
2
(a +b)x
2
1
3
x
3
_
s
r
= abs +
1
2
(a +b)s
2
1
3
s
3
+abr
1
2
(a +b)r
2
+
1
3
r
3
= ab(r s) +
1
2
(a +b)(s r)(s +r) +
1
3
(r s)(r
2
+rs +s
2
),
whiletheareaof thetrapezoidunder theshadedregionis
1
2
(s r) [(s a)(b s) +(r a)(b r)]
=
1
2
(s r)
_
2ab +(a +b)(r +s) r
2
s
2
_
= ab(r s) +
1
2
(a +b)(s r)(r +s) +
1
2
(r s)(r
2
+s
2
).
Thus, theareaof theshadedregionis
(r s)
_
1
3
r
2
+
1
3
rs +
1
3
s
2
1
2
r
2
1
2
s
2
_
= (s r)
_
1
6
r
2
1
3
rs +
1
6
s
2
_
=
1
6
(s r)
3
,
whichisfour-thirdstheareaof thetriangleACE.
(a) ApplytheComparisonTheorem(Theorem5inSection5.2) totheinequalitysinx x (validfor x 0) toprove
that
1
x
2
2
cosx 1
(b) Applyit againtoprovethat
x
x
3
6
sinx x (for x 0)
(c) Verifytheseinequalitiesfor x = 0.3.
55. Usethemethodof Exercise54toprovethat
1
x
2
2
cosx 1
x
2
2
+
x
4
24
x
x
3
6
sinx x
x
3
6
+
x
5
120
(for x 0)
Verifytheseinequalitiesfor x = 0.1. Whyhavewespeciedx 0for sinx but not for cosx?
solution ByExercise54, t
1
6
t
3
sint t for t > 0. Integratingthisinequalityover theinterval [0, x], andthen
solvingfor cosx, yields:
1
2
x
2
1
24
x
4
1cosx
1
2
x
2
1
1
2
x
2
cosx 1
1
2
x
2
+
1
24
x
4
.
Theseinequalities apply for x 0. Sincecosx, 1
x
2
2
, and1
x
2
2
+
x
4
24
areall evenfunctions, they also apply for
x 0.
Havingestablishedthat
1
t
2
2
cost 1
t
2
2
+
t
4
24
,
for all t 0, weintegrateover theinterval [0, x], toobtain:
x
x
3
6
sinx x
x
3
6
+
x
5
120
.
Thefunctionssinx, x
1
6
x
3
andx
1
6
x
3
+
1
120
x
5
areall oddfunctions, sotheinequalitiesarereversedfor x < 0.
Evaluatingtheseinequalitiesat x = 0.1yields
0.995000000 0.995004165 0.995004167
0.0998333333 0.0998334166 0.0998334167,
bothof whicharetrue.
Calculatethenext pair of inequalitiesfor sinx andcosx byintegratingtheresultsof Exercise55. Canyouguess
thegeneral pattern?
57. UseFTC I toprovethat if |f
(x)| K for x [a, b], then|f (x) f (a)| K|x a| for x [a, b].
solution Let a > b bereal numbers, andlet f (x) besuchthat |f
(x)| K for x [a, b]. ByFTC,

_
x
a
f
(t ) dt = f (x) f (a).
Sincef
(x) K for all x [a, b], weget:

f (x) f (a) =
_
x
a
f
(t ) dt K(x a).
Sincef
(x) K for all x [a, b], weget:

f (x) f (a) =
_
x
a
f
(t ) dt K(x a).
Combiningthesetwoinequalitiesyields
K(x a) f (x) f (a) K(x a),
sothat, bydenition,
|f (x) f (a)| K|x a|.
(a) UseExercise57toprovethat | sina sinb| |a b| for all a, b.
(b) Let f (x) = sin(x + a) sinx. Usepart (a) to showthat thegraphof f (x) lies betweenthehorizontal lines
y = a.
(c) Plot f (x) andthelinesy = a toverify(b) for a = 0.5anda = 0.2.
5.4 The Fundamental Theorem of Calculus, Part II
1. Let G(x) =
_
x
4
_
t
3
+1dt .
(a) IstheFTC neededtocalculateG(4)?
(b) IstheFTC neededtocalculateG
(4)?
solution
(a) No. G(4) =
_
4
4
_
t
3
+1dt = 0.
(b) Yes. BytheFTC II, G
(x) =
_
x
3
+1, soG
(4) =

65.
2. Whichof thefollowingisanantiderivativeF(x) of f (x) = x
2
satisfyingF(2) = 0?
(a)
_
x
2
2t dt (b)
_
2
0
t
2
dt (c)
_
x
2
t
2
dt
solution Thecorrect answer is(c):
_
x
2
t
2
dt .
3. Doeseverycontinuousfunctionhaveanantiderivative? Explain.
solution Yes. All continuousfunctionshaveanantiderivative, namely
_
x
a
f (t ) dt .
4. Let G(x) =
_
x
3
4
sint dt . Whichof thefollowingstatementsarecorrect?
(a) G(x) isthecompositefunctionsin(x
3
).
(b) G(x) isthecompositefunctionA(x
3
), where
A(x) =
_
x
4
sin(t ) dt
S E C T I ON 5.4 The Fundamental Theorem of Calculus, Part II 291
(c) G(x) istoocomplicatedtodifferentiate.
(d) TheProduct RuleisusedtodifferentiateG(x).
(e) TheChainRuleisusedtodifferentiateG(x).
(f) G
(x) = 3x
2
sin(x
3
).
solution Statements(b), (e), and(f)arecorrect.
Exercises
1. Writetheareafunctionof f (x) = 2x +4withlower limit a = 2asanintegral andndaformulafor it.
solution Let f (x) = 2x +4. Theareafunctionwithlower limit a = 2is
A(x) =
_
x
a
f (t ) dt =
_
x
2
(2t +4) dt.
Carryingout theintegration, wend
_
x
2
(2t +4) dt = (t
2
+4t )
x
2
= (x
2
+4x) ((2)
2
+4(2)) = x
2
+4x +4
or (x +2)
2
. Therefore, A(x) = (x +2)
2
.
Findaformulafor theareafunctionof f (x) = 2x +4withlower limit a = 0.
3. Let G(x) =
_
x
1
(t
2
2) dt . CalculateG(1), G
(1) andG
(2). Thenndaformulafor G(x).

solution Let G(x) =
_
x
1
(t
2
2) dt . Then G(1) =
_
1
1
(t
2
2) dt = 0. Moreover, G
(x) = x
2
2, so that
G
(1) = 1andG
(2) = 2. Finally,
G(x) =
_
x
1
(t
2
2) dt =
_
1
3
t
3
2t
_
x
1
=
_
1
3
x
3
2x
_
_
1
3
(1)
3
2(1)
_
=
1
3
x
3
2x +
5
3
.
FindF(0), F
(0), andF
(3), whereF(x) =
_
x
0
_
t
2
+t dt .
5. FindG(1), G
(0), andG
(/4), whereG(x) =
_
x
1
tant dt .
solution By denition, G(1) =
_
1
1
tant dt = 0. By FTC, G
(x) = tanx, sothat G
(0) = tan0 = 0andG
4
) =
tan
4
= 1.
FindH(2) andH
(2), whereH(x) =
_
x
2
du
u
2
+1
.
In Exercises 716, nd formulas for the functions represented by the integrals.
7.
_
x
2
u
4
du
solution F(x) =
_
x
2
u
4
du =
1
5
u
5
x
2
=
1
5
x
5
32
5
.
_
x
2
(12t
2
8t ) dt
9.
_
x
0
sinu du
solution F(x) =
_
x
0
sinu du = (cosu)
x
0
= 1cosx.
_
x
/4
sec
2
d
11.
_

x
2
dt
t
2
solution F(x) =
_

x
2
dt
t
2
=
1
t
x
2
=
1
2
x
.
_
4
sin
(5t +9) dt
13.
_
x
2
1
t dt
solution F(x) =
_
x
2
1
t dt =
1
2
t
2
x
2
1
=
1
2
x
4
1
2
.
_
x/4
x/2
sec
2
u du
15.
_
x
3/2
3
x
t
3
dt
solution F(x) =
_
x
3/2
3
x
t
3
dt =
1
4
t
4
x
3/2
3
x
=
1
4
x
6
81
4
x
2
_
x
2x
sec
2
t dt
In Exercises 1720, express the antiderivative F(x) of f (x) satisfying the given initial condition as an integral.
17. f (x) =
_
x
3
+1, F(5) = 0
solution TheantiderivativeF(x) of
_
x
3
+1satisfyingF(5) = 0isF(x) =
_
x
5
_
t
3
+1dt .
f (x) =
x +1
x
2
+9
, F(7) = 0
19. f (x) = secx, F(0) = 0
solution TheantiderivativeF(x) of f (x) = secx satisfyingF(0) = 0isF(x) =
_
x
0
sect dt .
f () = sin(
3
), F() = 0
21.
d
dx
_
x
0
(t
5
9t
3
) dt
solution ByFTC II,
d
dx
_
x
0
(t
5
9t
3
) dt = x
5
9x
3
.
d
d
_

1
cotu du
23.
d
dt
_
t
100
sec(5x 9) dx
solution ByFTC II,
d
dt
_
t
100
sec(5x 9) dx = sec(5t 9).
d
ds
_
s
2
tan
_
1
1+u
2
_
du
25. Let A(x) =
_
x
0
f (t ) dt for f (x) inFigure8.
(a) CalculateA(2), A(3), A
(2), andA
(3).
(b) Findformulasfor A(x) on[0, 2] and[2, 4] andsketchthegraphof A(x).
4 3 2 1
2
3
4
1
x
y
y = f (x)
FIGURE 8
solution
(a) A(2) = 2 2 = 4, theareaunder f (x) fromx = 0tox = 2, whileA(3) = 2 3+
1
2
= 6.5, theareaunder f (x)
fromx = 0tox = 3. BytheFTC, A
(x) = f (x) soA
(2) = f (2) = 2andA
(3) = f (3) = 3.
(b) Foreachx [0, 2], theregionunderthegraphof y = f (x) isarectangleof lengthx andheight2; foreachx [2, 4],
theregioniscomprisedof asquareof sidelength2andatrapezoidof height x 2andbases2andx. Hence,
A(x) =
_
2x, 0 x < 2
1
2
x
2
+2, 2 x 4
A graphof theareafunctionA(x) isshownbelow.
4 3 2 1
x
AreaFunction
A(x)
2
4
8
6
10
y
Makearoughsketchof thegraphof A(x) =
_
x
0
g(t ) dt for g(x) inFigure9.
FIGURE 9
27. Verify:
_
x
0
|t | dt =
1
2
x|x|. Hint: Consider x 0andx 0separately.
solution Let f (t ) = |t | =
_
t for t 0
t for t < 0
. Then
F(x) =
_
x
0
f (t ) dt =
_
x
0
t dt for x 0
_
x
0
t dt for x < 0
=
1
2
t
2
x
0
=
1
2
x
2
for x 0
_
1
2
t
2
_
x
0
=
1
2
x
2
for x < 0
For x 0, wehaveF(x) =
1
2
x
2
=
1
2
x |x| since|x| = x, whilefor x < 0, wehaveF(x) =
1
2
x
2
=
1
2
x |x| since
|x| = x. Therefore, for all real x wehaveF(x) =
1
2
x |x|.
FindG
(1), whereG(x) =
_
x
2
0
_
t
3
+3dt .
29.
d
dx
_
x
2
0
t dt
t +1
solution BytheChainRuleandtheFTC,
d
dx
_
x
2
0
t dt
t +1
=
x
2
x
2
+1
2x =
2x
3
x
2
+1
.
d
dx
_
1/x
1
cos
3
t dt
31.
d
ds
_
coss
6
u
4
du
solution BytheChainRuleandtheFTC,
d
ds
_
coss
6
u
4
du = cos
4
s(sins) = cos
4
s sins.
d
dx
_
x
4
x
2
t dt
Hint for Exercise 32: F(x) = A(x
4
) A(x
2
).
33.
d
dx
_
x
2
x
tant dt
solution Let
G(x) =
_
x
2
x
tant dt =
_
x
2
0
tant dt
_

x
0
tant dt.
ApplyingtheChainRulecombinedwithFTC twice, wehave
G
(x) = tan(x
2
) 2x tan(
x)
1
2
x
1/2
= 2x tan(x
2
)
tan(
x)
2
x
.
d
du
_
3u
u
_
x
2
+1dx
In Exercises 3538, with f (x) as in Figure 10 let
A(x) =
_
x
0
f (t ) dt and B(x) =
_
x
2
f (t ) dt .
x
y
6 3 4 5 2 1
2
1
0
1
2
y = f (x)
FIGURE 10
35. Findtheminandmaxof A(x) on[0, 6].
solution Theminimumvalues of A(x) on [0, 6] occur whereA
(x) = f (x) goes fromnegativeto positive. This

occursat oneplace, wherex = 1.5. Theminimumvalueof A(x) isthereforeA(1.5) = 1.25. Themaximumvaluesof
A(x) on[0, 6] occur whereA
(x) = f (x) goesfrompositivetonegative. Thisoccursat oneplace, wherex = 4.5. The

maximumvalueof A(x) isthereforeA(4.5) = 1.25.
Findtheminandmaxof B(x) on[0, 6].
37. Findformulasfor A(x) andB(x) validon[2, 4].
solution Ontheinterval [2, 4], A
(x) = B
(x) = f (x) = 1. A(2) =

_
2
0
f (t ) dt = 1andB(2) =
_
2
2
f (t ) dt = 0.
HenceA(x) = (x 2) 1andB(x) = (x 2).
Findformulasfor A(x) andB(x) validon[4, 5]. 39. Let A(x) =
_
x
0
f (t ) dt , withf (x) asinFigure11.
(a) DoesA(x) havealocal maximumat P?
(b) WheredoesA(x) havealocal minimum?
(c) WheredoesA(x) havealocal maximum?
(d) Trueor false? A(x) < 0for all x intheinterval shown.
x
y
S R
Q
P
y = f (x)
FIGURE 11 Graphof f (x).
solution
(a) Inorder for A(x) tohavealocal maximum, A
(x) = f (x) musttransitionfrompositivetonegative. Asthisdoesnot

happenat P, A(x) doesnot havealocal maximumat P.
(b) A(x) will havealocal minimumwhenA
(x) = f (x) transitions fromnegativeto positive. This happens at R, so

A(x) hasalocal minimumat R.
(c) A(x) will havealocal maximumwhenA
(x) = f (x) transitions frompositiveto negative. This happens at S, so

A(x) hasalocal maximumat S.
(d) It istruethat A(x) < 0onI sincethesignedareafrom0tox isclearlyalwaysnegativefromthegure.
Determinef (x), assumingthat
_
x
0
f (t ) dt = x
2
+x.
41. Determinethefunctiong(x) andall valuesof c suchthat
_
x
c
g(t ) dt = x
2
+x 6
solution BytheFTC II wehave
g(x) =
d
dx
(x
2
+x 6) = 2x +1
andtherefore,
_
x
c
g(t ) dt = x
2
+x (c
2
+c)
Wemust choosec sothat c
2
+c = 6. Wecantakec = 2or c = 3.
Finda b suchthat
_
b
a
(x
2
9) dx hasminimal value.
In Exercises 43 and 44, let A(x) =
_
x
a
f (t ) dt .
43. AreaFunctionsandConcavity Explainwhythefollowingstatementsaretrue. Assumef (x) isdifferen-
tiable.
(a) If c isaninectionpoint of A(x), thenf
(c) = 0.
(b) A(x) isconcaveupif f (x) isincreasing.
(c) A(x) isconcavedownif f (x) isdecreasing.
solution
(a) If x = c isaninectionpoint of A(x), thenA
(c) = f
(c) = 0.
(b) If A(x) isconcaveup, thenA
(x) > 0. SinceA(x) istheareafunctionassociatedwithf (x), A
(x) = f (x) byFTC

II, soA
(x) = f
(x). Thereforef
(x) > 0, sof (x) isincreasing.

(c) If A(x) isconcavedown, thenA
(x) < 0. SinceA(x) istheareafunctionassociatedwithf (x), A
(x) = f (x) by
FTC II, soA
(x) = f
(x). Therefore, f
(x) < 0andsof (x) isdecreasing.

Matchthepropertyof A(x) withthecorrespondingpropertyof thegraphof f (x).Assumef (x) isdifferentiable.
AreafunctionA(x)
(a) A(x) isdecreasing.
(b) A(x) hasalocal maximum.
(c) A(x) isconcaveup.
(d) A(x) goesfromconcaveuptoconcavedown.
Graphof f(x)
(i) Liesbelowthex-axis.
(ii) Crossesthex-axisfrompositivetonegative.
(iii) Hasalocal maximum.
(iv) f (x) isincreasing.
45. Let A(x) =
_
x
0
f (t ) dt , withf (x) asinFigure12. Determine:
(a) TheintervalsonwhichA(x) isincreasinganddecreasing
(b) Thevaluesx whereA(x) hasalocal minor max
(c) Theinectionpointsof A(x)
(d) TheintervalswhereA(x) isconcaveupor concavedown
2 4 6 8 10 12
x
y
y = f (x)
FIGURE 12
solution
(a) A(x) isincreasingwhenA
(x) = f (x) > 0, whichcorrespondstotheintervals(0, 4) and(8, 12). A(x) isdecreasing

whenA
(x) = f (x) < 0, whichcorrespondstotheintervals(4, 8) and(12, ).

(b) A(x) has alocal minimumwhenA
(x) = f (x) changes from to +, correspondingto x = 8. A(x) has alocal

maximumwhenA
(x) = f (x) changesfrom+ to, correspondingtox = 4andx = 12.

(c) Inectionpointsof A(x) occurwhereA
(x) = f
(x) changessign, orwheref changesfromincreasingtodecreasing

or viceversa. Consequently, A(x) hasinectionpointsat x = 2, x = 6, andx = 10.
(d) A(x) isconcaveupwhenA
(x) = f
(x) ispositiveor f (x) isincreasing, whichcorrespondstotheintervals(0, 2)

and (6, 10). Similarly, A(x) is concavedown when f (x) is decreasing, which corresponds to theintervals (2, 6) and
(10, ).
Let f (x) = x
2
5x 6andF(x) =
_
x
0
f (t ) dt .
(a) Findthecritical pointsof F(x) anddeterminewhether theyarelocal minimaor local maxima.
(b) Findthepoints of inectionof F(x) anddeterminewhether theconcavity changes fromupto downor from
downtoup.
(c) Plot f (x) andF(x) onthesameset of axesandconrmyour answersto(a) and(b).
47. Sketchthegraphof anincreasingfunctionf (x) suchthat bothf
(x) andA(x) =
_
x
0
f (t ) dt aredecreasing.
solution If f
(x) isdecreasing, thenf
(x) mustbenegative. Furthermore, if A(x) =

_
x
0
f (t ) dt isdecreasing, then
A
(x) = f (x) must alsobenegative. Thus, weneedafunctionwhichisnegativebut increasingandconcavedown. The

graphof onesuchfunctionisshownbelow.
x
y
Figure13showsthegraphof f (x) = x sinx. Let F(x) =
_
x
0
t sint dt .
(a) Locatethelocal maxandabsolutemaxof F(x) on[0, 3].
(b) J ustifygraphically: F(x) haspreciselyonezeroin[, 2].
(c) HowmanyzerosdoesF(x) havein[0, 3]?
(d) Findtheinectionpointsof F(x) on[0, 3]. For eachone, statewhether theconcavitychangesfromuptodown
or fromdowntoup.
49. Findthesmallest positivecritical point of
F(x) =
_
x
0
cos(t
3/2
) dt
anddeterminewhether it isalocal minor max. Thenndthesmallest positiveinectionpoint of F(x) anduseagraph
of y = cos(x
3/2
) todeterminewhether theconcavitychangesfromuptodownor fromdowntoup.
solution A critical point of F(x) occurswhereF
(x) = cos(x
3/2
) = 0. Thesmallest positivecritical pointsoccurs
wherex
3/2
= /2, sothatx = (/2)
2/3
. F
(x) goesfrompositivetonegativeatthispoint, sox = (/2)

2/3
corresponds
toalocal maximum..
Candidate inection points of F(x) occur where F
(x) = 0. By FTC, F
(x) = cos(x
3/2
), so F
(x) =
(3/2)x
1/2
sin(x
3/2
). Findingthesmallest positivesolutionof F
(x) = 0, weget:
(3/2)x
1/2
sin(x
3/2
) = 0
sin(x
3/2
) = 0 (sincex > 0)
x
3/2
=
x =
2/3
2.14503.
Fromtheplot below, weseethat F
(x) = cos(x
3/2
) changes fromdecreasingto increasingat
2/3
, so F(x) changes
fromconcavedowntoconcaveupat that point.
x
y
3
1
0.5
0.5
1
2 1
Proof of FTC II Theproof inthetext assumesthat f (x) isincreasing. Toproveit for all continuousfunctions,
let m(h) andM(h) denotetheminimum andmaximum of f (t ) on[x, x + h] (Figure14). Thecontinuity of f (x)
impliesthat lim
h0
m(h) = lim
h0
M(h) = f (x). Showthat for h > 0,
hm(h) A(x +h) A(x) hM(h)
For h < 0, theinequalitiesarereversed. Provethat A
(x) = f (x).
51. Proof of FTC I FTCI assertsthat
_
b
a
f (t ) dt = F(b) F(a) if F
(x) = f (x). UseFTCII togiveanewproof of

FTC I asfollows. Set A(x) =
_
x
a
f (t ) dt .
(a) Showthat F(x) = A(x) +C for someconstant.
(b) Showthat F(b) F(a) = A(b) A(a) =
_
b
a
f (t ) dt .
solution Let F
(x) = f (x) andA(x) =

_
x
a
f (t ) dt .
(a) Thenby theFTC, Part II, A
(x) = f (x) andthus A(x) andF(x) arebothantiderivatives of f (x). HenceF(x) =

A(x) +C for someconstant C.
(b)
F(b) F(a) = (A(b) +C) (A(a) +C) = A(b) A(a)
=
_
b
a
f (t ) dt
_
a
a
f (t ) dt =
_
b
a
f (t ) dt 0=
_
b
a
f (t ) dt
whichprovestheFTC, Part I.
CanEveryAntiderivativeBeExpressedasanIntegral? Theareafunction
_
x
a
f (t ) dt isanantiderivativeof
f (x) for every valueof a. However, not all antiderivatives areobtainedinthis way. Thegeneral antiderivativeof
f (x) = x isF(x) =
1
2
x
2
+C. Showthat F(x) isanareafunctionif C 0but not if C > 0.
53. Provetheformula
d
dx
_
v(x)
u(x)
f (t ) dt = f (v(x))v
(x) f (u(x))u
(x)
solution Write
_
v(x)
u(x)
f (x) dx =
_
0
u(x)
f (x) dx +
_
v(x)
0
f (x) dx =
_
v(x)
0
f (x) dx
_
u(x)
0
f (x) dx.
Then, bytheChainRuleandtheFTC,
d
dx
_
v(x)
u(x)
f (x) dx =
d
dx
_
v(x)
0
f (x) dx
d
dx
_
u(x)
0
f (x) dx
= f (v(x))v
(x) f (u(x))u
(x).
Usetheresult of Exercise53tocalculate
d
dx
_

x
1/
x
sint dt
5.5 Net Change as the Integral of a Rate
1. Ahotmetal objectissubmergedincoldwater. Therateatwhichtheobjectcools(indegreesper minute) isafunction
f (t ) of time. Whichquantityisrepresentedbytheintegral
_
T
0
f (t ) dt ?
solution Thedeniteintegral
_
T
0
f (t ) dt represents thetotal dropintemperatureof themetal object intherst T
minutesafter beingsubmergedinthecoldwater.
2. A planetravels560kmfromLosAngelestoSanFranciscoin1hour. If theplanesvelocity at timet isv(t ) km/h,
what isthevalueof
_
1
0
v(t ) dt ?
_
1
0
v(t ) dt represents thetotal distancetraveled by theairplaneduring theonehour
ight fromLosAngelestoSanFrancisco. Thereforethevalueof
_
1
0
v(t ) dt is560km.
3. Whichof thefollowingquantitieswouldbenaturallyrepresentedasderivativesandwhichasintegrals?
(a) Velocityof atrain
(b) Rainfall duringa6-monthperiod
(c) Mileageper gallonof anautomobile
(d) IncreaseintheU.S. populationfrom1990to2010
solution Quantities (a) and (c) involve rates of change, so these would naturally be represented as derivatives.
Quantities(b)and(d)involveanaccumulation, sothesewouldnaturallyberepresentedasintegrals.
Exercises
1. Water ows into an empty reservoir at a rateof 3000+ 20t litersper hour. What is thequantity of water in the
reservoir after 5hours?
solution Thequantityof water inthereservoir after vehoursis
_
5
0
(3000+20t ) dt =
_
3000t +10t
2
_
5
0
= 15,250gallons.
A populationof insectsincreasesat arateof 200+10t +0.25t
2
insectsper day. Findtheinsect populationafter
3days, assumingthat thereare35insectsat t = 0.
3. A survey shows that amayoral candidateis gainingvotes at arateof 2000t + 1000votes per day, wheret is the
numberof dayssincesheannouncedhercandidacy. Howmanysupporterswill thecandidatehaveafter60days, assuming
that shehadnosupportersat t = 0?
solution Thenumber of supportersthecandidatehasafter 60daysis
_
60
0
(2000t +1000) dt = (1000t
2
+1000t )
60
0
= 3,660,000.
A factoryproducesbicyclesat arateof 95+3t
2
t bicyclesper week. Howmanybicycleswereproducedfrom
thebeginningof week2totheendof week3?
5. Findthedisplacement of aparticlemovinginastraight linewithvelocityv(t ) = 4t 3m/sover thetimeinterval
[2, 5].
solution Thedisplacement isgivenby
_
5
2
(4t 3) dt = (2t
2
3t )
5
2
= (5015) (86) = 33m.
Findthedisplacement over thetimeinterval [1, 6] of ahelicopter whose(vertical) velocity at timet is v(t ) =
0.02t
2
+t m/s.
S E C T I ON 5.5 Net Change as the Integral of a Rate 297
7. Acatfallsfromatree(withzeroinitial velocity) attimet = 0. Howfar doesthecatfall betweent = 0.5andt = 1s?
UseGalileosformulav(t ) = 9.8t m/s.
solution Givenv(t ) = 9.8t m/s, thetotal distancethecat fallsduringtheinterval [
1
2
, 1] is
_
1
1/2
|v(t )| dt =
_
1
1/2
9.8t dt = 4.9t
2
1
1/2
= 4.91.225= 3.675m.
A projectileis released with an initial (vertical) velocity of 100 m/s. Usetheformula v(t ) = 100 9.8t for
velocitytodeterminethedistancetraveledduringtherst 15seconds.
In Exercises 912, a particle moves in a straight line with the given velocity (in m/s). Find the displacement and distance
traveled over the time interval, and draw a motion diagram like Figure 3 (with distance and time labels).
9. v(t ) = 124t , [0, 5]
solution Displacement isgivenby
_
5
0
(124t ) dt = (12t 2t
2
)
5
0
= 10ft, whiletotal distanceisgivenby
_
5
0
|124t | dt =
_
3
0
(124t ) dt +
_
5
3
(4t 12) dt = (12t 2t
2
)
3
0
+(2t
2
12t )
5
3
= 26ft.
Thedisplacement diagramisgivenhere.
0 18
t = 0
t = 5
t = 3
10
Distance
v(t ) = 3624t +3t
2
, [0, 10]
11. v(t ) = t
2
1, [0.5, 2]
solution Displacement isgivenby
_
2
0.5
(t
2
1) dt = (t
1
t )
2
0.5
= 0m, whiletotal distanceisgivenby
_
2
0.5
t
2
1
dt =
_
1
0.5
(t
2
1) dt +
_
2
1
(1t
2
) dt = (t
1
t )
1
0.5
+(t +t
1
)
2
1
= 1m.
Thedisplacement diagramisgivenhere.
0 0.5
t = 0
t = 2
t = 1
Distance
v(t ) = cost , [0, 3]
13. Findthenet changeinvelocityover [1, 4] of anobject witha(t ) = 8t t
2
m/s
2
.
solution Thenet changeinvelocityis
_
4
1
a(t ) dt =
_
4
1
(8t t
2
) dt =
_
4t
2
1
3
t
3
_
4
1
= 39m/s.
Showthatif accelerationisconstant, thenthechangeinvelocityisproportional tothelengthof thetimeinterval.
15. Thetrafcowratepast acertainpoint onahighwayisq(t ) = 3000+2000t 300t
2
(t inhours), wheret = 0is
8am. Howmanycarspassbyinthetimeinterval from8to10am?
solution Thenumber of carsisgivenby
_
2
0
q(t ) dt =
_
2
0
(3000+2000t 300t
2
) dt =
_
3000t +1000t
2
100t
3
_
2
0
= 3000(2) +1000(4) 100(8) = 9200cars.
Themarginal cost of producingx tablet computers is C
(x) = 120 0.06x + 0.00001x

2
What is thecost of
producing3000unitsif thesetupcostis$90,000?If productionissetat3000units, whatisthecostof producing200
additional units?
17. A small boutiqueproduces wool sweaters at amarginal cost of 40 5[[x/5]] for 0 x 20, where[[x]] is the
greatest integer function. Findthecost of producing20sweaters. Thencomputetheaveragecost of therst 10sweaters
andthelast 10sweaters.
solution Thetotal cost of producing20sweatersis
_
20
0
(405[[x/5]]) dx =
_
5
0
40dx +
_
10
5
35dx +
_
15
10
30dx +
_
20
15
25dx
= 40(5) +35(5) +30(5) +25(5) = 650dollars.
Fromthiscalculation, weseethat thecost of therst 10sweatersis$375andthecost of thelast tensweatersis$275;
thus, theaveragecost of therst tensweatersis$37.50andtheaveragecost of thelast tensweatersis$27.50.
Therate(inlitersper minute) atwhichwater drainsfromatankisrecordedathalf-minuteintervals. Computethe
averageof theleft- andright-endpoint approximationstoestimatethetotal amount of water drainedduringtherst
3minutes.
t (min) 0 0.5 1 1.5 2 2.5 3
r (l/min) 50 48 46 44 42 40 38
19. The velocity of a car is recorded at half-second intervals (in feet per second). Use the average of the left- and
right-endpoint approximationstoestimatethetotal distancetraveledduringtherst 4seconds.
t 0 0.5 1 1.5 2 2.5 3 3.5 4
v(t ) 0 12 20 29 38 44 32 35 30
solution Let t = 0.5. Then
R
N
= 0.5 (12+20+29+38+44+32+35+30) = 120ft.
L
N
= 0.5 (0+12+20+29+38+44+32+35) = 105ft.
Theaverageof R
N
andL
N
is112.5ft.
Tomodel theeffectsof acarbontaxonCO
2
emissions, policymakersstudythemarginal cost of abatement B(x),
denedasthecostof increasingCO
2
reductionfromx tox +1tons(inunitsof tenthousandtonsFigure4). Which
quantityisrepresentedbytheareaunder thecurveover [0, 3] inFigure4?
21. A megawatt of power is 10
6
W, or 3.6 10
9
J /hour. Whichquantity is representedby theareaunder thegraphin
Figure5? Estimatetheenergy(injoules) consumedduringtheperiod4pm to8pm.
18
19
20
21
22
23
24
25
26
27
28
00 02 04 06 08 10 12 14 16 18 20 22 24
Megawatts (in thousands)
Hour of theday
FIGURE 5 Power consumptionover 1-dayperiodinCalifornia(February2010).
solution Theareaunder thegraphinFigure5represents thetotal power consumptionover oneday inCalifornia.
Assumingt = 0corresponds tomidnight, theperiod4pm to8pm corresponds tot = 16tot = 20. Theleft andright
endpoint approximationsare
L = 1(22.8+23.5+26.1+26.7) = 99.1megawatt hours
R = 1(23.5+26.1+26.7+26.1) = 102.4megawatt hours
Theaverageof thesevaluesis
100.75megawatt hours= 3.62710
11
joules.
Figure6showsthemigrationrateM(t ) of Irelandintheperiod19881998. Thisistherateatwhichpeople(in
thousandsper year) moveintoor out of thecountry.
(a) Isthefollowingintegral positiveor negative?What doesthisquantityrepresent?
_
1998
1988
M(t ) dt
(b) Didmigrationintheperiod19881998result inanet inux of peopleinto Irelandor anet outowof people
fromIreland?
(c) During which two years could theIrish primeminister announce, Wevehit an inection point. Wearestill
losingpopulation, but thetrendisnowimproving.
23. Let N(d) bethenumber of asteroids of diameter d kilometers. Datasuggest that thediameters aredistributed
accordingtoapiecewisepower law:
N
(d) =
_
1.910
9
d
2.3
for d < 70
2.610
12
d
4
for d 70
(a) Computethenumber of asteroidswithdiameter between0.1and100km.
(b) UsingtheapproximationN(d +1) N(d) N
(d), estimatethenumber of asteroidsof diameter 50km.

solution
(a) Thenumber of asteroidswithdiameter between0.1and100kmis
_
100
0.1
N
(d) dd =
_
70
0.1
1.910
9
d
2.3
dd +
_
100
70
2.610
12
d
4
dd
=
1.910
9
1.3
d
1.3
70
0.1
2.610
12
3
d
3
100
70
= 2.91610
10
+1.6610
6
2.91610
10
.
(b) Takingd = 49.5,
N(50.5) N(49.5) N
(49.5) = 1.910
9
49.5
2.3
= 240,525.79.
Thus, thereareapproximately240,526asteroidsof diameter 50km.
S E C T I ON 5.5 Net Change as the Integral of a Rate 299
HeatCapacity Theheat capacityC(T ) of asubstanceistheamount of energy(injoules) requiredtoraisethe
temperatureof 1gby1
C at temperatureT .
(a) ExplainwhytheenergyrequiredtoraisethetemperaturefromT
1
toT
2
istheareaunder thegraphof C(T ) over
[T
1
, T
2
].
(b) Howmuchenergyisrequiredtoraisethetemperaturefrom50to100
C if C(T ) = 6+0.2
T ?
25. Figure7showstherateR(t ) of natural gasconsumption(inbillionsof cubicfeet per day) inthemid-Atlanticstates
(NewYork, NewJ ersey, Pennsylvania). Expressthetotal quantity of natural gasconsumedin2009asanintegral (with
respect totimet indays). Thenestimatethisquantity, giventhefollowingmonthlyvaluesof R(t ):
3.18, 2.86, 2.39, 1.49, 1.08, 0.80,
1.01, 0.89, 0.89, 1.20, 1.64, 2.52
Keepinmindthat thenumber of daysinamonthvarieswiththemonth.
1
2
3
J A S O N D J F M A M J
Natural gas consumption (10
9
cubic ft/day)
FIGURE 7 Natural gasconsumptionin2009inthemid-Atlanticstates
solution Thetotal quantityof natural gasconsumedisgivenby
_
365
0
R(t ) dt.
Withthegivendata, wend
_
365
0
R(t ) dt 31(3.18) +28(2.86) +31(2.39) +30(1.49) +31(1.08) +30(0.80)
+31(1.01) +31(0.89) +30(0.89) +31(1.20) +30(1.64) +31(2.52)
= 605.05billioncubicfeet.
CardiacoutputistherateR of volumeof bloodpumpedbytheheartperunittime(inlitersperminute). Doctors
measureR byinjectingA mgof dyeintoaveinleadingintotheheart at t = 0andrecordingtheconcentrationc(t )
of dye(inmilligramsper liter) pumpedout at short regular timeintervals(Figure8).
(a) Explain: Thequantityof dyepumpedout inasmall timeinterval [t, t +t ] isapproximatelyRc(t )t .
(b) Showthat A = R
_
T
0
c(t ) dt , whereT islargeenoughthat all of thedyeispumpedthroughtheheart but not so
largethat thedyereturnsbyrecirculation.
(c) AssumeA = 5mg. EstimateR usingthefollowingvaluesof c(t ) recordedat 1-secondintervalsfromt = 0to
t = 10:
0, 0.4, 2.8, 6.5, 9.8, 8.9,
6.1, 4, 2.3, 1.1, 0
Exercises 27 and 28: A study suggests that the extinction rate r(t ) of marine animal families during the Phanerozoic Eon
can be modeled by the function r(t ) = 3130/(t + 262) for 0 t 544, where t is time elapsed (in millions of years)
since the beginning of the eon 544million years ago. Thus, t = 544refers to the present time, t = 540is 4million years
ago, and so on.
27. Computetheaverageof R
N
andL
N
withN = 5toestimatethetotal number of familiesthat becameextinct inthe
periods100 t 150and350 t 400.
solution
(100 t 150) For N = 5,
t =
150100
5
= 10.
Thetableof values{r(t
i
)}
i=0...5
isgivenbelow:
t
i
100 110 120 130 140 150
r(t
i
) 8.64641 8.41398 8.19372 7.98469 7.78607 7.59709
Theendpoint approximationsare:
R
N
= 10(8.41398+8.19372+7.98469+7.78607+7.59709) 399.756families
L
N
= 10(8.64641+8.41398+8.19372+7.98469+7.78607) 410.249families
Theright endpoint approximation estimates 399.756 families becameextinct in theperiod 100 t 150, the
left endpoint approximationestimates410.249familiesbecameextinct duringthistime. Theaverageof thetwois
405.362families.
(350 t 400) For N = 10,
t =
400350
5
= 19.
Thetableof values{r(t
i
)}
i=0...5
isgivenbelow:
t
i
350 360 370 380 390 400
r(t
i
) 5.11438 5.03215 4.95253 4.87539 4.80061 4.72810
Theendpoint approximationsare:
R
N
= 10(5.03215+4.95253+4.87539+4.80061+4.72810) 243.888families
L
N
= 10(5.11438+5.03215+4.95253+4.87539+4.80061) 247.751families
Theright endpoint approximation estimates 243.888 families becameextinct in theperiod 350 t 400, the
left endpoint approximationestimates247.751familiesbecameextinct duringthistime. Theaverageof thetwois
245.820families.
Estimatethetotal number of extinct familiesfromt = 0tothepresent, usingM
N
withN = 544.
29. Showthat aparticle, locatedat theoriginat t = 1andmovingalongthex-axiswithvelocityv(t ) = t
2
, will never
passthepoint x = 2.
solution Theparticles velocity is v(t ) = s
(t ) = t
2
, an antiderivativefor which is F(t ) = t
1
. Hence, the
particlespositionat timet is
s(t ) =
_
t
1
s
(u) du = F(u)
t
1
= F(t ) F(1) = 1
1
t
< 1
for all t 1. Thus, theparticlewill never passx = 1, whichimpliesit will never passx = 2either.
Showthataparticle, locatedattheoriginatt = 1andmovingalongthex-axiswithvelocityv(t ) = t
1/2
moves
arbitrarilyfar fromtheoriginafter sufcient timehaselapsed.
5.6 Substitution Method
1. Whichof thefollowingintegralsisacandidatefor theSubstitutionMethod?
(a)
_
5x
4
sin(x
5
) dx (b)
_
sin
5
x cosx dx (c)
_
x
5
sinx dx
solution Thefunctionin(c): x
5
sinx isnotof theformg(u(x))u
(x). Thefunctionin(a)meetstheprescribedpattern
with g(u) = sinu and u(x) = x
5
. Similarly, the function in (b) meets the prescribed pattern with g(u) = u
5
and
u(x) = sinx.
2. Findanappropriatechoiceof u for evaluatingthefollowingintegralsbysubstitution:
(a)
_
x(x
2
+9)
4
dx (b)
_
x
2
sin(x
3
) dx (c)
_
sinx cos
2
x dx
solution
(a) x(x
2
+9)
4
=
1
2
(2x)(x
2
+9)
4
; hence, c =
1
2
, f (u) = u
4
, andu(x) = x
2
+9.
(b) x
2
sin(x
3
) =
1
3
(3x
2
) sin(x
3
); hence, c =
1
3
, f (u) = sinu, andu(x) = x
3
.
(c) sinx cos
2
x = (sinx) cos
2
x; hence, c = 1, f (u) = u
2
, andu(x) = cosx.
3. Whichof thefollowingisequal to
_
2
0
x
2
(x
3
+1) dx for asuitablesubstitution?
(a)
1
3
_
2
0
u du (b)
_
9
0
u du (c)
1
3
_
9
1
u du
solution Withthesubstitutionu = x
3
+ 1, thedeniteintegral
_
2
0
x
2
(x
3
+ 1) dx becomes
1
3
_
9
1
u du. Thecorrect
answer is(c).
Exercises
In Exercises 16, calculate du.
1. u = x
3
x
2
solution Let u = x
3
x
2
. Thendu = (3x
2
2x) dx.
u = 2x
4
+8x
1
3. u = cos(x
2
)
solution Let u = cos(x
2
). Thendu = sin(x
2
) 2x dx = 2x sin(x
2
) dx.
u = tanx
5. u = sin
4
solution Let u = sin

4
. Thendu = 4sin
3
cos d.
u =
t
t +1
S E C T I ON 5.6 Substitution Method 301
In Exercises 720, write the integral in terms of u and du. Then evaluate.
7.
_
(x 7)
3
dx, u = x 7
solution Let u = x 7. Thendu = dx. Hence
_
(x 7)
3
dx =
_
u
3
du =
1
4
u
4
+C =
1
4
(x 7)
4
+C.
_
(x +25)
2
dx, u = x +25
9.
_
t
_
t
2
+1dt , u = t
2
+1
solution Let u = t
2
+1. Thendu = 2t dt . Hence,
_
t
_
t
2
+1dt =
1
2
_
u
1/2
du =
1
3
u
3/2
+C =
1
3
(t
2
+1)
3/2
+C.
_
(x
3
+1) cos(x
4
+4x) dx, u = x
4
+4x
11.
_
t
3
(42t
4
)
11
dt , u = 42t
4
solution Let u = 42t
4
. Thendu = 8t
3
dt . Hence,
_
t
3
(42t
4
)
11
dt =
1
8
_
u
11
du =
1
80
u
10
+C =
1
80
(42t
4
)
10
+C.
_

4x 1dx, u = 4x 1
13.
_
x(x +1)
9
dx, u = x +1
solution Let u = x +1. Thenx = u 1anddu = dx. Hence
_
x(x +1)
9
dx =
_
(u 1)u
9
du =
_
(u
10
u
9
) du
=
1
11
u
11
1
10
u
10
+C =
1
11
(x +1)
11
1
10
(x +1)
10
+C.
_
x
4x 1dx, u = 4x 1
15.
_
x
2
x +1dx, u = x +1
solution Let u = x +1. Thenx = u 1anddu = dx. Hence
_
x
2
x +1dx =
_
(u 1)
2
u
1/2
du =
_
(u
5/2
2u
3/2
+u
1/2
) du
=
2
7
u
7/2
4
5
u
5/2
+
2
3
u
3/2
+C
=
2
7
(x +1)
7/2
4
5
(x +1)
5/2
+
2
3
(x +1)
3/2
+C.
_
sin(4 7) d, u = 4 7
17.
_
sin
2
cos d, u = sin
solution Let u = sin. Thendu = cos d. Hence,
_
sin
2
cos d =
_
u
2
du =
1
3
u
3
+C =
1
3
sin
3
+C.
_
sec
2
x tanx dx, u = tanx
19.
_
x sec
2
(x
2
) dx, u = x
2
solution Let u = x
2
. Thendu = 2x dx, sothat
1
2
du = x dx. Hence,
_
x sec
2
(x
2
) dx =
1
2
_
sec
2
u du =
1
2
tanu +C =
1
2
tan(x
2
) +C
_
sec
2
(cosx) sinx dx, u = cosx
In Exercises 2124, evaluate the integral in the form a sin(u(x)) +C for an appropriate choice of u(x) and constant a.
21.
_
x
3
cos(x
4
) dx
solution Let u = x
4
. Thendu = 4x
3
dx or
1
4
du = x
3
dx. Hence
_
x
3
cos(x
4
) dx =
1
4
_
cosu du =
1
4
sinu +C =
1
4
sin(x
4
) +C.
_
x
2
cos(x
3
+1) dx
23.
_
x
1/2
cos(x
3/2
) dx
solution Let u = x
3/2
. Thendu =
3
2
x
1/2
dx or
2
3
du = x
1/2
dx. Hence
_
x
1/2
cos(x
3/2
) dx =
2
3
_
cosu du =
2
3
sinu +C =
2
3
sin(x
3/2
) +C.
_
cosx cos(sinx) dx
In Exercises 2560, evaluate the indenite integral.
25.
_
(4x +5)
9
dx
solution Let u = 4x +5. Thendu = 4dx and
_
(4x +5)
9
dx =
1
4
_
u
9
du =
1
40
u
10
+C =
1
40
(4x +5)
10
+C.
_
dx
(x 9)
5
27.
_
dt
t +12
solution Let u = t +12. Thendu = dt and
_
dt
t +12
=
_
u
1/2
du = 2u
1/2
+C = 2
t +12+C.
_
(9t +2)
2/3
dt
29.
_
x +1
(x
2
+2x)
3
dx
solution Let u = x
2
+2x. Thendu = (2x +2) dx or
1
2
du = (x +1) dx. Hence
_
x +1
(x
2
+2x)
3
dx =
1
2
_
1
u
3
du =
1
2
_
1
2
u
2
_
+C =
1
4
(x
2
+2x)
2
+C =
1
4(x
2
+2x)
2
+C.
_
(x +1)(x
2
+2x)
3/4
dx
31.
_
x
_
x
2
+9
dx
solution Let u = x
2
+9. Thendu = 2x dx or
1
2
du = x dx. Hence
_
x
_
x
2
+9
dx =
1
2
_
1
u
du =
1
2
u
1
2
+C =
_
x
2
+9+C.
_
2x
2
+x
(4x
3
+3x
2
)
2
dx
33.
_
(3x
2
+1)(x
3
+x)
2
dx
solution Let u = x
3
+x. Thendu = (3x
2
+1) dx. Hence
_
(3x
2
+1)(x
3
+x)
2
dx =
_
u
2
du =
1
3
u
3
+C =
1
3
(x
3
+x)
3
+C.
_
5x
4
+2x
(x
5
+x
2
)
3
dx
35.
_
(3x +8)
11
dx
solution Let u = 3x +8. Thendu = 3dx and
_
(3x +8)
11
dx =
1
3
_
u
11
du =
1
36
u
12
+C =
1
36
(3x +8)
12
+C.
_
x(3x +8)
11
dx
37.
_
x
2
_
x
3
+1dx
solution Let u = x
3
+1. Thendu = 3x
2
dx and
_
x
2
_
x
3
+1dx =
1
3
_
u
1/2
du =
2
9
u
3/2
+C =
2
9
(x
3
+1)
3/2
+C.
_
x
5
_
x
3
+1dx
39.
_
dx
(x +5)
3
solution Let u = x +5. Thendu = dx and
_
dx
(x +5)
3
=
_
u
3
du =
1
2
u
2
+C =
1
2
(x +5)
2
+C.
_
x dx
(x +5)
3/2
41.
_
z
2
(z
3
+1)
12
dz
solution Let u = z
3
+1. Thendu = 3z
2
dz and
_
z
2
(z
3
+1)
12
dz =
1
3
_
u
12
du =
1
39
u
13
+C =
1
39
(z
3
+1)
13
+C.
_
(z
5
+4z
2
)(z
3
+1)
12
dz
43.
_
(x +2)(x +1)
1/4
dx
solution Let u = x +1. Thenx = u 1, du = dx and
_
(x +2)(x +1)
1/4
dx =
_
(u +1)u
1/4
du =
_
(u
5/4
+u
1/4
) du
=
4
9
u
9/4
+
4
5
u
5/4
+C
=
4
9
(x +1)
9/4
+
4
5
(x +1)
5/4
+C.
_
x
3
(x
2
1)
3/2
dx
45.
_
sin(83) d
solution Let u = 83. Thendu = 3d and
_
sin(83) d =
1
3
_
sinu du =
1
3
cosu +C =
1
3
cos(83) +C.
_
sin(
2
) d
47.
_
cos
t
dt
solution Let u =

t = t
1/2
. Thendu =
1
2
t
1/2
dt and
_
cos
t
dt = 2
_
cosu du = 2sinu +C = 2sin
t +C.
_
x
2
sin(x
3
+1) dx
49.
_
sinx cosx
sinx +1
dx
solution Let u = sinx +1. Thendu = cosx dx andsinx = u 1. Hence
_
sinx cosx
sinx +1
dx =
_
u 1
u
du =
_
_
u
1/2
u
1/2
_
du =
2
3
u
3/2
2u
1/2
+C
=
2
3
(sinx +1)
3/2
2(sinx +1)
1/2
+C
_
sin
8
cos d
51.
_
sec
2
x(12tan
3
x 6tan
2
x) dx
solution Let u = tanx. Thendu = sec
2
x dx and
_
sec
2
x(12tan
3
x 6tan
2
x) dx =
_
_
12u
3
6u
2
_
du = 3u
4
2u
3
+C = 3tan
4
x 2tan
3
x +C
_
x
1/5
sec
_
x
4/5
_
tan
_
x
4/5
_
dx
53.
_
sec
2
(4x +9) dx
solution Let u = 4x +9. Thendu = 4dx or
1
4
du = dx. Hence
_
sec
2
(4x +9) dx =
1
4
_
sec
2
u du =
1
4
tanu +C =
1
4
tan(4x +9) +C.
_
sec
2
x tan
4
x dx
55.
_
sec
2
(
x) dx
x
solution Let u =

x. Thendu =
1
2
x
dx or 2du =
1
x
dx. Hence,
_
sec
2
(
x) dx
x
= 2
_
sec
2
u dx = 2tanu +C = 2tan(
x) +C.
_
cos2x
(1+sin2x)
2
dx
57.
_
sin4x
cos4x +1dx
solution Let u = cos4x +1. Thendu = 4sin4x or
1
4
du = sin4x. Hence
_
sin4x
cos4x +1dx =
1
4
_
u
1/2
du =
1
4
_
2
3
u
3/2
_
+C =
1
6
(cos4x +1)
3/2
+C.
_
cosx(3sinx 1) dx
59.
_
sec tan(sec 1) d
solution Let u = sec 1. Thendu = sec tan d and
_
sec tan(sec 1) d =
_
u du =
1
2
u
2
+C =
1
2
(sec 1)
2
+C.
_
cost cos(sint ) dt
61. Evaluate
_
dx
(1+
x)
3
usingu = 1+
x. Hint: Showthat dx = 2(u 1)du.

solution Let u = 1+
x. Then
du =
1
2
x
dx or dx = 2
x du = 2(u 1) du.
Hence,
_
dx
(1+
x)
3
= 2
_
u 1
u
3
du = 2
_
(u
2
u
3
) du
= 2u
1
+u
2
+C =
2
1+
x
+
1
(1+
x)
2
+C.
CanThey BothBeRight? Hannah uses the substitution u = tanx andAkiva uses u = secx to evaluate
_
tanx sec
2
x dx. Showthat theyobtaindifferent answers, andexplaintheapparent contradiction.
63. Evaluate
_
sinx cosx dx usingsubstitutionintwodifferent ways: rst usingu = sinx andthenusingu = cosx.
Reconcilethetwodifferent answers.
solution First, let u = sinx. Thendu = cosx dx and
_
sinx cosx dx =
_
u du =
1
2
u
2
+C
1
=
1
2
sin
2
x +C
1
.
Next, let u = cosx. Thendu = sinx dx or du = sinx dx. Hence,
_
sinx cosx dx =
_
u du =
1
2
u
2
+C
2
=
1
2
cos
2
x +C
2
.
Toreconcilethesetwoseeminglydifferent answers, recall that anytwoantiderivativesof aspeciedfunctiondiffer bya
constant. Toshowthat thisistruehere, notethat (
1
2
sin
2
x +C
1
) (
1
2
cos
2
x +C
2
) =
1
2
+C
1
C
2
, aconstant. Here
weusedthetrigonometricidentitysin
2
x +cos
2
x = 1.
SomeChoicesAreBetter ThanOthers Evaluate
_
sinx cos
2
x dx
twice. First useu = sinx toshowthat
_
sinx cos
2
x dx =
_
u
_
1u
2
du
andevaluatetheintegral ontheright byafurther substitution. Thenshowthat u = cosx isabetter choice.
65. What arethenewlimitsof integrationif weapplythesubstitutionu = 3x + totheintegral
_
0
sin(3x +) dx?
solution Thenewlimitsof integrationareu(0) = 3 0+ = andu() = 3 + = 4.
Whichof thefollowingistheresult of applyingthesubstitutionu = 4x 9totheintegral
_
8
2
(4x 9)
20
dx?
(a)
_
8
2
u
20
du (b)
1
4
_
8
2
u
20
du
(c) 4
_
23
1
u
20
du (d)
1
4
_
23
1
u
20
du
In Exercises 6778, use the Change-of-Variables Formula to evaluate the denite integral.
67.
_
3
1
(x +2)
3
dx
solution Let u = x +2. Thendu = dx. Hence
_
3
1
(x +2)
3
dx =
_
5
3
u
3
du =
1
4
u
4
5
3
=
5
4
4

3
4
4
= 136.
_
6
1
x +3dx
69.
_
1
0
x
(x
2
+1)
3
dx
solution Let u = x
2
1
2
du = x dx. Hence
_
1
0
x
(x
2
+1)
3
dx =
1
2
_
2
1
1
u
3
du =
1
2
_
1
2
u
2
_
2
1
=
1
16
+
1
4
=
3
16
= 0.1875.
_
2
1
5x +6dx
71.
_
4
0
x
_
x
2
+9dx
solution Let u = x
2
1
2
du = x dx. Hence
_
4
0
_
x
2
+9dx =
1
2
_
25
9
u du =
1
2
_
2
3
u
3/2
_
25
9
=
1
3
(12527) =
98
3
.
_
2
1
4x +12
(x
2
+6x +1)
2
dx
73.
_
1
0
(x +1)(x
2
+2x)
5
dx
solution Let u = x
2
+2x. Thendu = (2x +2) dx = 2(x +1) dx, and
_
1
0
(x +1)(x
2
+2x)
5
dx =
1
2
_
3
0
u
5
du =
1
12
u
6
3
0
=
729
12
=
243
4
.
_
17
10
(x 9)
2/3
dx
75.
_
/2
/2
cosx dx
sinx +1
solution Let u = sinx +1. Thendu = cosx dx, and
_
/2
/2
cosx dx
sinx +1
=
_
sin(/2)+1
sin(/2)+1
1
u
du = 2
2
0
= 2
2
_
/6
0
sec
2
_
2x

6
_
dx
77.
_
/2
0
cos
3
x sinx dx
solution Let u = cosx. Thendu = sinx dx. Hence
_
/2
0
cos
3
x sinx dx =
_
0
1
u
3
du =
_
1
0
u
3
du =
1
4
u
4
1
0
=
1
4
0=
1
4
.
_
/2
/3
cot
2
x
2
csc
2
x
2
dx
79. Evaluate
_
2
0
r
_
5
_
4r
2
dr.
solution Let u = 5
_
4r
2
. Then
du =
r dr
_
4r
2
=
r dr
5u
sothat
r dr = (5u) du.
Hence, theintegral becomes:
_
2
0
r
_
5
_
4r
2
dr =
_
5
3
u(5u) du =
_
5
3
_
5u
1/2
u
3/2
_
du =
_
10
3
u
3/2
2
5
u
5/2
_
5
3
=
_
50
3
510
5
_
_
10
3
18
5
3
_
=
20
3
5
32
5
3.
Findnumbersa andb suchthat
_
b
a
(u
2
+1) du =
_
/4
/4
sec
4
d
andevaluate. Hint: Usetheidentitysec
2
= tan
2
+1.
In Exercises 81 and 82, use substitution to evaluate the integral in terms of f (x).
81.
_
f (x)
3
f
(x) dx
solution Let u = f (x). Thendu = f
(x) dx. Hence

_
f (x)
3
f
(x) dx =
_
u
3
du =
1
4
u
4
+C =
1
4
f (x)
4
+C.
_
f
(x)
f (x)
2
dx
83. Showthat
_
/6
0
f (sin) d =
_
1/2
0
f (u)
1
_
1u
2
du.
solution Let u = sin. Thenu(/6) = 1/2andu(0) = 0, asrequired. Furthermore, du = cos d, sothat
d =
du
cos
.
If sin = u, thenu
2
+cos
2
= 1, sothat cos =
_
1u
2
. Therefored = du/
_
1u
2
. Thisgives
_
/6
0
f (sin) d =
_
1/2
0
f (u)
1
_
1u
2
du.
Evaluate
_
/2
0
sin
n
x cosx dx for n 0.
85. EvaluateI =
_
/2
0
d
1+tan
6000
. Hint: Usesubstitutiontoshowthat I isequal toJ =

_
/2
0
d
1+cot
6000
and
thencheckthat I +J =
_
/2
0
d.
solution Toevaluate
I =
_
/2
0
dx
1+tan
6000
x
,
wesubstitutet = /2x. Thendt = dx, x = /2t , t (0) = /2, andt (/2) = 0. Hence,
I =
_
/2
0
dx
1+tan
6000
x
=
_
0
/2
dt
1+tan
6000
(/2t )
=
_
/2
0
dt
1+cot
6000
t
.
Let J =
_
/2
0
dt
1+cot
6000
(t )
. WeknowI = J, soI +J = 2I. Ontheother hand, bythedenitionof I andJ andthe
linearityof theintegral,
I +J =
_
/2
0
dx
1+tan
6000
x
+
dx
1+cot
6000
x
=
_
/2
0
_
1
1+tan
6000
x
+
1
1+cot
6000
x
_
dx
=
_
/2
0
_
1
1+tan
6000
x
+
1
1+(1/ tan
6000
x)
_
dx
=
_
/2
0
_
1
1+tan
6000
x
+
1
(tan
6000
x +1)/ tan
6000
x
_
dx
=
_
/2
0
_
1
1+tan
6000
x
+
tan
6000
x
1+tan
6000
x
_
dx
=
_
/2
0
_
1+tan
6000
x
1+tan
6000
x
_
dx =
_
/2
0
1dx = /2.
Hence, I +J = 2I = /2, soI = /4.
Usethesubstitutionu = 1+x
1/n
toshowthat
_
_
1+x
1/n
dx = n
_
u
1/2
(u 1)
n1
du
Evaluatefor n = 2, 3.
87. Usesubstitutiontoprovethat
_
a
a
f (x) dx = 0if f isanoddfunction.
solution Weassumethat f is continuous. If f (x) is an odd function, then f (x) = f (x). Let u = x. Then
x = u anddu = dx or du = dx. Accordingly,
_
a
a
f (x) dx =
_
0
a
f (x) dx +
_
a
0
f (x) dx =
_
0
a
f (u) du +
_
a
0
f (x) dx
=
_
a
0
f (x) dx
_
a
0
f (u) du = 0.
Provethat
_
b
a
1
x
dx =
_
b/a
1
1
x
dx fora, b > 0. Thenshowthattheregionsunderthehyperbolaovertheintervals
[1, 2], [2, 4], [4, 8], . . . all havethesamearea(Figure4).
89. ShowthatthetworegionsinFigure5havethesamearea. Thenusetheidentitycos
2
u =
1
2
(1+cos2u) tocompute
thesecondarea.
(A) (B)
x
1 1
1 1
u
y = cos
2
u
y y
y = 1 x
2
FIGURE 5
solution Theareaof theregioninFigure5(A) isgivenby
_
1
0
_
1x
2
dx. Let x = sinu. Thendx = cosu du and
_
1x
2
=
_
1sin
2
u = cosu. Hence,
_
1
0
_
1x
2
dx =
_
/2
0
cosu cosu du =
_
/2
0
cos
2
u du.
This last integral represents theareaof theregioninFigure5(B). Thetworegions inFigure5thereforehavethesame
area.
Letsnowfocusonthedeniteintegral
_
/2
0
cos
2
u du. Usingthetrigonometricidentitycos
2
u =
1
2
(1+cos2u), we
have
_
/2
0
cos
2
u du =
1
2
_
/2
0
1+cos2u du =
1
2
_
u +
1
2
sin2u
_
/2
0
=
1
2

2
0=

4
.
Areaof anEllipse ProvetheformulaA = ab for theareaof theellipsewithequation(Figure6)
x
2
a
2
+
y
2
b
2
= 1
Hint: Useachangeof variablestoshowthat A isequal toab timestheareaof theunit circle.
In Exercises 14, refer to the function f (x) whose graph is shown in Figure 1.
1
2
3
1 2 3 4
y
x
FIGURE 1
1. EstimateL
4
andM
4
on[0, 4].
solution Withn = 4andaninterval of [0, 4], x =
40
4
= 1. Then,
L
4
= x(f (0) +f (1) +f (2) +f (3)) = 1
_
1
4
+1+
5
2
+2
_
=
23
4
and
M
4
= x
_
f
_
1
2
_
+f
_
3
2
_
+f
_
5
2
_
+f
_
7
2
__
= 1
_
1
2
+2+
9
4
+
9
4
_
= 7.
EstimateR
4
, L
4
, andM
4
on[1, 3].
3. Findaninterval [a, b] onwhichR
4
islarger than
_
b
a
f (x) dx. Dothesamefor L
4
.
solution Ingeneral, R
N
is larger than
_
b
a
f (x) dx onany interval [a, b] over whichf (x) is increasing. Giventhe
graphof f (x), wemaytake[a, b] = [0, 2]. Inorder for L
4
tobelarger than
_
b
a
f (x) dx, f (x) must bedecreasingover
theinterval [a, b]. Wemaythereforetake[a, b] = [2, 3].
J ustify
3
2

_
2
1
f (x) dx
9
4
.
2
+3x.
5. CalculateR
6
, M
6
, andL
6
for f (x) ontheinterval [2, 5]. Sketchthegraphof f (x) andthecorrespondingrectangles
for eachapproximation.
2
+3x. A uniformpartitionof [2, 5] withN = 6subintervalshas
x =
52
6
=
1
2
, x
j
= a +jx = 2+
j
2
,
and
x
j
= a +
_
j
1
2
_
x =
7
4
+
j
2
.
Now,
R
6
= x
6
j=1
f (x
j
) =
1
2
_
f
_
5
2
_
+f (3) +f
_
7
2
_
+f (4) +f
_
9
2
_
+f (5)
_
=
1
2
_
55
4
+18+
91
4
+28+
135
4
+40
_
=
625
8
.
Therectanglescorrespondingtothisapproximationareshownbelow.
10
2.0 2.5 3.0 3.5 4.0 4.5
15
20
25
30
35
y
x
Next,
M
6
= x
6
j=1
f (x
j
) =
1
2
_
f
_
9
4
_
+f
_
11
4
_
+f
_
13
4
_
+f
_
15
4
_
+f
_
17
4
_
+f
_
19
4
__
=
1
2
_
189
16
+
253
16
+
325
16
+
405
16
+
493
16
+
589
16
_
=
2254
32
=
1127
16
.
10
2.0 2.5 3.0 3.5 4.0 4.5
15
20
25
30
35
y
x
Finally,
L
6
= x
5
j=0
f (x
j
) =
1
2
_
f (2) +f
_
5
2
_
+f (3) +f
_
7
2
_
+f (4) +f
_
9
2
__
=
1
2
_
10+
55
4
+18+
91
4
+28+
135
4
_
=
505
8
.
10
2.0 2.5 3.0 3.5 4.0 4.5
15
20
25
30
35
y
x
UseFTC I toevaluateA(x) =
_
x
2
f (t ) dt .
7. Findaformulafor R
N
for f (x) on[2, 5] andcompute
_
5
2
f (x) dx bytakingthelimit.
2
+3x ontheinterval [2, 5]. Thenx =
52
N
=
3
N
anda = 2. Hence,
R
N
= x
N
j=1
f (2+jx) =
3
N
N
j=1
_
_
2+
3j
N
_
2
+3
_
2+
3j
N
_
_
=
3
N
N
j=1
_
10+
21j
N
+
9j
2
N
2
_
= 30+
63
N
2
N
j=1
j +
27
N
3
N
j=1
j
2
= 30+
63
N
2
_
N
2
2
+
N
2
_
+
27
N
3
_
N
3
3
+
N
2
2
+
N
6
_
=
141
2
+
45
N
+
9
2N
2
and
lim
N
R
N
= lim
N
_
141
2
+
45
N
+
9
2N
2
_
=
141
2
.
Findaformulafor L
N
for f (x) on[0, 2] andcompute
_
2
0
f (x) dx bytakingthelimit.
9. CalculateR
5
, M
5
, andL
5
for f (x) = (x
2
+1)
1
ontheinterval [0, 1].
solution Let f (x) = (x
2
+1)
1
. A uniformpartitionof [0, 1] withN = 5subintervalshas
x =
10
5
=
1
5
, x
j
= a +jx =
j
5
,
and
x
j
= a +
_
j
1
2
_
x =
2j 1
10
.
Now,
R
5
= x
5
j=1
f (x
j
) =
1
5
_
f
_
1
5
_
+f
_
2
5
_
+f
_
3
5
_
+f
_
4
5
_
+f (1)
_
=
1
5
_
25
26
+
25
29
+
25
34
+
25
41
+
1
2
_
0.733732.
Next,
M
5
= x
5
j=1
f (x
j
) =
1
5
_
f
_
1
10
_
+f
_
3
10
_
+f
_
1
2
_
+f
_
7
10
_
+f
_
9
10
__
=
1
5
_
100
101
+
100
109
+
4
5
+
100
149
+
100
181
_
0.786231.
Finally,
L
5
= x
4
j=0
f (x
j
) =
1
5
_
f (0) +f
_
1
5
_
+f
_
2
5
_
+f
_
3
5
_
+f
_
4
5
__
=
1
5
_
1+
25
26
+
25
29
+
25
34
+
25
41
_
0.833732.
Let R
N
betheNthright-endpoint approximationfor f (x) = x
3
on[0, 4] (Figure2).
(a) Provethat R
N
=
64(N +1)
2
N
2
.
(b) Provethat theareaof theregionwithintheright-endpoint rectanglesabovethegraphisequal to
64(2N +1)
N
2
11. WhichapproximationtotheareaisrepresentedbytheshadedrectanglesinFigure3? ComputeR
5
andL
5
.
x
y
30
18
6
1 2 3 4 5
FIGURE 3
solution Thereareverectangles andtheheight of eachis givenby thefunctionvalueat theright endpoint of the
subinterval. Thus, thearearepresentedbytheshadedrectanglesisR
5
.
Fromthegure, weseethat x = 1. Then
R
5
= 1(30+18+6+6+30) = 90 and L
5
= 1(30+30+18+6+6) = 90.
Calculateany two Riemannsums for f (x) = x
2
ontheinterval [2, 5], but choosepartitions withat least ve
subintervalsof unequal widthsandintermediatepointsthat areneither endpointsnor midpoints.
In Exercises 1316, express the limit as an integral (or multiple of an integral) and evaluate.
13. lim
N
6N
N
j=1
sin
_
3
+
j
6N
_
solution Let f (x) = sinx and N be a positive integer. A uniformpartition of the interval [/3, /2] with N
subintervalshas
x =

6N
and x
j
=

3
+
j
6N
for 0 j N. Then
6N
N
j=1
sin
_
3
+
j
6N
_
= x
N
j=1
f (x
j
) = R
N
;
consequently,
lim
N
6N
N
j=1
sin
_
3
+
j
6N
_
=
_
/2
/3
sinx dx = cosx
/2
/3
= 0+
1
2
=
1
2
.
lim
N
3
N
N1
k=0
_
10+
3k
N
_
15. lim
N
5
N
N
j=1
_
4+5j/N

x andN beapositiveinteger. A uniformpartitionof theinterval [4, 9] withN subintervals
has
x =
5
N
and x
j
= 4+
5j
N
for 0 j N. Then
5
N
N
j=1
_
4+5j/N = x
N
j=1
f (x
j
) = R
N
;
consequently,
lim
N
5
N
N
j=1
_
4+5j/N =
_
9
4
x dx =
2
3
x
3/2
9
4
=
54
3

16
3
=
38
3
.
lim
N
1
k
+2
k
+ +N
k
N
k+1
(k > 0)
In Exercises 1720, use the given substitution to evaluate the integral.
17.
_
2
0
dt
(4t +12)
2
, u = 4t +12
solution Let u = 4t +12. Thendu = 4dt , andthenewlimitsof integrationareu = 12andu = 20. Thus,
_
2
0
dt
(4t +12)
2
=
1
4
_
20
12
du
u
2
=
1
4
u
1
20
12
=
1
4
_
1
12
1
20
_
=
1
120
_
(x
2
+1) dx
(x
3
+3x)
4
, u = x
3
+3x
19.
_
/6
0
sinx cos
4
x dx, u = cosx
solution Let u = cosx. Thendu = sinx dx andthenewlimitsof integrationareu = 1andu =

3/2. Thus,
_
/6
0
sinx cos
4
x dx =
_

3/2
1
u
4
du
=
1
5
u
5
3/2
1
=
1
5
_
1
9
3
32
_
.
_
sec
2
(2) tan(2) d, u = tan(2)
In Exercises 2148, evaluate the integral.
21.
_
(20x
4
9x
3
2x) dx
solution
_
(20x
4
9x
3
2x) dx = 4x
5
9
4
x
4
x
2
+C.
_
2
0
(12x
3
3x
2
) dx
23.
_
(2x
2
3x)
2
dx
solution
_
(2x
2
3x)
2
dx =
_
(4x
4
12x
3
+9x
2
) dx =
4
5
x
5
3x
4
+3x
3
+C.
_
1
0
(x
7/3
2x
1/4
) dx
25.
_
x
5
+3x
4
x
2
dx
solution
_
x
5
+3x
4
x
2
dx =
_
(x
3
+3x
2
) dx =
1
4
x
4
+x
3
+C.
_
3
1
r
4
dr
27.
_
3
3
|x
2
4| dx
solution
_
3
3
|x
2
4| dx =
_
2
3
(x
2
4) dx +
_
2
2
(4x
2
) dx +
_
3
2
(x
2
4) dx
=
_
1
3
x
3
4x
_
2
3
+
_
4x
1
3
x
3
_
2
2
+
_
1
3
x
3
4x
_
3
2
=
_
16
3
3
_
+
_
16
3
+
16
3
_
+
_
3+
16
3
_
=
46
3
.
_
4
2
|(x 1)(x 3)| dx
29.
_
3
1
[t ] dt
solution
_
3
1
[t ] dt =
_
2
1
[t ] dt +
_
3
2
[t ] dt =
_
2
1
dt +
_
3
2
2dt = t
2
1
+2t
3
2
= (21) +(64) = 3.
_
2
0
(t [t ])
2
dt
31.
_
(10t 7)
14
dt
solution Let u = 10t 7. Thendu = 10dt and
_
(10t 7)
14
dt =
1
10
_
u
14
du =
1
150
u
15
+C =
1
150
(10t 7)
15
+C.
_
3
2
_
7y 5dy
33.
_
(2x
3
+3x) dx
(3x
4
+9x
2
)
5
solution Let u = 3x
4
+9x
2
. Thendu = (12x
3
+18x) dx = 6(2x
3
+3x) dx and
_
(2x
3
+3x) dx
(3x
4
+9x
2
)
5
=
1
6
_
u
5
du =
1
24
u
4
+C =
1
24
(3x
4
+9x
2
)
4
+C.
_
1
3
x dx
(x
2
+5)
2
35.
_
5
0
15x
x +4dx
solution Let u = x +4. Thenx = u 4, du = dx andthenewlimitsof integrationareu = 4andu = 9. Thus,
_
5
0
15x
x +4dx =
_
9
4
15(u 4)
u du
= 15
_
9
4
(u
3/2
4u
1/2
) du
= 15
_
2
5
u
5/2
8
3
u
3/2
_
9
4
= 15
__
486
5
72
_
_
64
5

64
3
__
= 506.
_
t
2
t +8dt
37.
_
1
0
cos
_
3
(t +2)
_
dt
solution
_
1
0
cos
_
3
(t +2)
_
dt =
3
sin
_
3
(t +2)
_
1
0
=
3
3
2
.
_

/2
sin
_
5
6
_
d
39.
_
t
2
sec
2
(9t
3
+1) dt
solution Let u = 9t
3
+1. Thendu = 27t
2
dt and
_
t
2
sec
2
(9t
3
+1) dt =
1
27
_
sec
2
u du =
1
27
tanu +C =
1
27
tan(9t
3
+1) +C.
_
sin
2
(3) cos(3) d
41.
_
csc
2
(92) d
solution Let u = 92. Thendu = 2d and
_
csc
2
(92) d =
1
2
_
csc
2
u du =
1
2
cotu +C =
1
2
cot(92) +C.
_
sin
4cos d
43.
_
/3
0
sin
cos
2/3
d
solution Let u = cos. Thendu = sin d andwhen = 0, u = 1andwhen =

3
, u =
1
2
. Finally,
_
/3
0
sin
cos
2/3
d =
_
1/2
1
u
2/3
du = 3u
1/3
1/2
1
= 3(2
1/3
1) = 3
3
3
4
2
.
_
sec
2
t dt
(tant 1)
2
45.
_
y
_
2y +3dy
solution Let u = 2y +3; thendu = 2dy sothat dy =
1
2
du, andy =
1
2
(u 3). Then
_
y
_
2y +3dy =
1
4
_
(u 3)
u du =
1
4
_
_
u
3/2
3u
1/2
_
du =
1
10
u
5/2
1
2
u
3/2
+C
=
1
10
(2y +3)
5/2
1
2
(2y +3)
3/2
+C
_
8
1
t
2
t +8dt
47.
_
/2
0
sec
2
(cos) sin d
solution Let u = cos; thendu = sin d, andthenewboundsof integrationarecos0= 1tocos
2
= 0. Thus
_
/2
0
sec
2
(cos) sin d =
_
0
1
sec
2
u du = tanu
1
0
= tan1
_
2
4
12x dx
(x
2
+2)
3
49. Combinetowriteasasingleintegral:
_
8
0
f (x) dx +
_
0
2
f (x) dx +
_
6
8
f (x) dx
solution First, rewrite
_
8
0
f (x) dx =
_
6
0
f (x) dx +
_
8
6
f (x) dx
andobservethat
_
6
8
f (x) dx =
_
8
6
f (x) dx.
Thus,
_
8
0
f (x) dx +
_
6
8
f (x) dx =
_
6
0
f (x) dx.
Finally,
_
8
0
f (x) dx +
_
0
2
f (x) dx +
_
6
8
f (x) dx =
_
6
0
f (x) dx +
_
0
2
f (x) dx =
_
6
2
f (x) dx.
LetA(x) =
_
x
0
f (x) dx, wheref (x) isthefunctionshowninFigure4. Identifythelocationof thelocal minima,
thelocal maxima, and points of inection of A(x) on theinterval [0, E], as well as theintervals whereA(x) is
increasing, decreasing, concaveup, or concavedown. Wheredoestheabsolutemaxof A(x) occur?
51. Findtheinectionpointsof A(x) =
_
x
3
t dt
t
2
+1
. However, donot evaluateA(x) explicitly.
solution Let
A(x) =
_
x
3
t dt
t
2
+1
.
Then
A
(x) =
x
x
2
+1
and
A
(x) =
(x
2
+1)(1) x(2x)
(x
2
+1)
2
=
1x
2
(x
2
+1)
2
.
ClearlyA
(x) < 0for |x| > 1andA
(x) > 0for |x| < 1. ThusA(x) isconcavedownfor |x| > 1andconcaveupfor
|x| < 1. A(x) thereforehasinectionpointsat x = 1.
A particlestartsat theoriginat timet = 0andmoveswithvelocityv(t ) asshowninFigure5.
(a) Howmanytimesdoestheparticlereturntotheoriginintherst 12seconds?
(b) What istheparticlesmaximumdistancefromtheorigin?
(c) What isparticlesmaximumdistancetotheleft of theorigin?
53. Onatypical day, acity consumeswater at therateof r(t ) = 100+ 72t 3t
2
(inthousandsof gallonsper hour),
wheret is thenumber of hours past midnight. What is thedaily water consumption? How much water is consumed
between6pm andmidnight?
solution Withaconsumptionrateof r(t ) = 100+ 72t 3t
2
thousandgallons per hour, thedaily consumptionof
water is
_
24
0
(100+72t 3t
2
) dt =
_
100t +36t
2
t
3
_
24
0
= 100(24) +36(24)
2
(24)
3
= 9312,
or 9.312milliongallons. From6PM tomidnight, thewater consumptionis
_
24
18
(100+72t 3t
2
) dt =
_
100t +36t
2
t
3
_
24
18
= 100(24) +36(24)
2
(24)
3
_
100(18) +36(18)
2
(18)
3
_
= 93127632= 1680,
or 1.68milliongallons.
Thelearningcurveinacertainbicyclefactory is L(x) = 12x
1/5
(inhours per bicycle), whichmeans that it
takes abikemechanic L(n) hours to assemblethenthbicycle. If amechanic has produced24bicycles, howlong
doesit takeher or himtoproducethesecondbatchof 12?
55. Cost engineersat NASA havethetask of projectingthecost P of major spaceprojects. It hasbeenfoundthat the
cost C of developingaprojectionincreaseswithP at theratedC/dP 21P
0.65
, whereC isinthousandsof dollars
andP inmillionsof dollars. Whatisthecostof developingaprojectionfor aprojectwhosecostturnsouttobeP = $35
million?
solution Assumingit costs nothingtodevelopaprojectionfor aproject withacost of $0, thecost of developinga
projectionfor aproject whosecost turnsout tobe$35millionis
_
35
0
21P
0.65
dP = 60P
0.35
35
0
= 60(35)
0.35
208.245,
or $208,245.
Anastronomer estimatesthat inacertainconstellation, thenumber of starsper magnitudem, per degree-squared
of sky, is equal toA(m) = 2.4 10
6
m
7.4
(fainter stars havehigher magnitudes). Determinethetotal number of
starsof magnitudebetween6and15inaone-degree-squaredregionof sky.
57. Evaluate
_
8
8
x
15
dx
3+cos
2
x
, usingthepropertiesof oddfunctions.
x
15
3+cos
2
x
andnotethat
f (x) =
(x)
15
3+cos
2
(x)
=
x
15
cos
2
x
= f (x).
Becausef (x) isanoddfunctionandtheinterval 8 x 8issymmetricabout x = 0, it followsthat
_
8
8
x
15
dx
3+cos
2
x
= 0.
Evaluate
_
1
0
f (x) dx, assumingthat f (x) isanevencontinuousfunctionsuchthat
_
2
1
f (x) dx = 5,
_
1
2
f (x) dx = 8
59. Plot thegraphof f (x) = sinmx sinnx on[0, ] for thepairs(m, n) = (2, 4), (3, 5) andineachcaseguess
thevalueof I =
_
0
f (x) dx. Experiment withafewmorevalues (includingtwo cases withm = n) andformulatea
conjecturefor whenI iszero.
solution Thegraphsof f (x) = sinmx sinnx with(m, n) = (2, 4) and(m, n) = (3, 5) areshownbelow. It appears
asif thepositiveareasbalancethenegativeareas, soweexpect that
I =
_

0
f (x) dx = 0
inthesecases.
0.5
3 2.5 2 1.5 1 0.5
0.5
(2, 4)
x
y
0.5
3 2.5 2 1.5 1 0.5
0.5
(3, 5)
x
y
Wearriveat thesameconclusionfor thecases(m, n) = (4, 1) and(m, n) = (5, 2).
0.5
3 2.5 2 1.5 1 0.5
0.5
(4, 1)
x
y
0.5
3 2.5 2 1.5 1 0.5
0.5
(5, 2)
x
y
However, when(m, n) = (3, 3) andwhen(m, n) = (5, 5), thevalueof
I =
_

0
f (x) dx
isclearlynot zeroasthereisnonegativearea.
0.5
3 2.5 2 1.5 1 0.5
0.5
(3, 3)
x
y
0.5
3 2.5 2 1.5 1 0.5
0.5
(5, 5)
x
y
Wethereforeconjecturethat I iszerowhenever m = n.
Showthat
_
x f (x) dx = xF(x) G(x)
whereF
(x) = f (x) andG
(x) = F(x). Usethistoevaluate

_
x cosx dx.
61. Prove
2
_
2
1
2
x
dx 4 and
1
9

_
2
1
3
x
dx
1
3
solution Thefunctionf (x) = 2
x
isincreasing, so1 x 2impliesthat 2= 2
1
2
x
2
2
= 4. Consequently,
2=
_
2
1
2dx
_
2
1
2
x
dx
_
2
1
4dx = 4.
Ontheother hand, thefunctionf (x) = 3
x
isdecreasing, so1 x 2impliesthat
1
9
= 3
2
3
x
3
1
=
1
3
.
It thenfollowsthat
1
9
=
_
2
1
1
9
dx
_
2
1
3
x
dx
_
2
1
1
3
dx =
1
3
.
Plot thegraphof f (x) = x
2
sinx, andshowthat 0.2
_
2
1
f (x) dx 0.9.
63. Findupper andlower boundsfor
_
1
0
f (x) dx, for f (x) inFigure6.
1
1
2
y
x
f (x)
y = x
2
+ 1
y = x
1/2
+ 1
FIGURE 6
solution Fromthegure, weseethat theinequalitiesx
2
+1 f (x)

x +1holdfor 0 x 1. Because
_
1
0
(x
2
+1) dx =
_
1
3
x
3
+x
_
1
0
=
4
3
and
_
1
0
(
x +1) dx =
_
2
3
x
3/2
+x
_
1
0
=
5
3
,
it followsthat
4
3

_
1
0
f (x) dx
5
3
.
In Exercises 6469, nd the derivative.
A
(x), whereA(x) =
_
x
3
sin(t
3
) dt
65. A
(), whereA(x) =
_
x
2
cost
1+t
dt
solution Let A(x) =
_
x
2
cost
1+t
dt . ThenA
(x) =
cosx
1+x
and
A
() =
cos
1+
=
1
1+
.
d
dy
_
y
2
3
x
dx
67. G
(x), whereG(x) =
_
sinx
2
t
3
dt
solution Let G(x) =
_
sinx
2
t
3
dt . Then
G
(x) = sin
3
x
d
dx
sinx = sin
3
x cosx.
G
(2), whereG(x) =
_
x
3
0
t +1dt
69. H
(1), whereH(x) =
_
9
4x
2
1
t
dt
solution Let H(x) =
_
9
4x
2
1
t
dt =
_
4x
2
9
1
t
dt . Then
H
(x) =
1
4x
2
d
dx
4x
2
=
8x
4x
2
=
2
x
andH
(1) = 2.
Explainwithagraph: If f (x) isincreasingandconcaveupon[a, b], thenL
N
ismoreaccuratethanR
N
. Which
ismoreaccurateif f (x) isincreasingandconcavedown?
71. Explainwithagraph: If f (x) islinear on[a, b], thenthe
_
b
a
f (x) dx =
1
2
(R
N
+L
N
) for all N.
solution Consider thegurebelow, whichdisplaysaportionof thegraphof alinear function.
x
y
The shaded rectangles represent the differences between the right-endpoint approximation R
N
and the left-endpoint
approximationL
N
. Inparticular, theportionof eachrectanglethat lies belowthegraphof y = f (x) is theamount by
whichL
N
underestimates theareaunder thegraph, whereas theportionof eachrectanglethat lies abovethegraphof
y = f (x) istheamount bywhichR
N
overestimatesthearea. Becausethegraphof y = f (x) isaline, thelower portion
of eachshadedrectangleisexactlythesamesizeastheupper portion. Therefore, if weaverageL
N
andR
N
, theerror in
thetwoapproximationswill exactlycancel, leaving
1
2
(R
N
+L
N
) =
_
b
a
f (x) dx.
Letf (x) beapositiveincreasingcontinuousfunctionon[a, b], where0 a < b asinFigure7. Showthatthe
shadedregionhasarea
I = bf (b) af (a)
_
b
a
f (x) dx
73. Howcanweinterpret thequantityI inEq. (1) if a < b 0? Explainwithagraph.
solution Wewill consider eachtermontheright-handsideof (1) separately. For convenience, let I, II, III andIV
denotetheareaof thesimilarlylabeledregioninthediagrambelow.
y
x
b a
I
III
II
IV
f (b)
f (a)
Becauseb < 0, theexpressionbf (b) istheoppositeof theareaof therectanglealongtheright; that is,
bf (b) = II IV.
Similarly,
af (a) = III +IV and
_
b
a
f (x) dx = I III.
Therefore,
bf (b) af (a)
_
b
a
f (x) dx = I II;
that is, theoppositeof theareaof theshadedregionshownbelow.
y
x
b a
f (b)
f (a)
6 APPLICATIONS OF THE
INTEGRAL
6.1 Area Between Two Curves
1. What istheareainterpretationof
_
b
a
_
f (x) g(x)
_
dx if f (x) g(x)?
solution Becausef (x) g(x),
_
b
a
(f (x) g(x)) dx representstheareaof theregionboundedbetweenthegraphs
of y = f (x) andy = g(x), boundedontheleft bythevertical linex = a andontheright bythevertical linex = b.
2. Is
_
b
a
_
f (x) g(x)
_
dx still equal totheareabetweenthegraphsof f andg if f (x) 0but g(x) 0?
solution Yes. Sincef (x) 0andg(x) 0, it followsthat f (x) g(x) 0.
3. Supposethat f (x) g(x) on[0, 3] andg(x) f (x) on[3, 5]. Expresstheareabetweenthegraphsover [0, 5] asa
sumof integrals.
solution Remember that tocalculateanareabetweentwocurves, onemust subtract theequationfor thelower curve
fromtheequationfor theupper curve. Over theinterval [0, 3], y = f (x) istheupper curve. Ontheother hand, over the
interval [3, 5], y = g(x) istheupper curve. Theareabetweenthegraphsover theinterval [0, 5] isthereforegivenby
_
3
0
(f (x) g(x)) dx +
_
5
3
(g(x) f (x)) dx.
4. Supposethat thegraphof x = f (y) liestotheleft of they-axis. Is
_
b
a
f (y) dy positiveor negative?
solution If thegraphof x = f (y) liestotheleft of they-axis, thenfor eachvalueof y, thecorrespondingvalueof x
islessthanzero. Hence, thevalueof
_
b
a
f (y) dy isnegative.
Exercises
1. Findtheareaof theregionbetweeny = 3x
2
+12andy = 4x +4over [3, 3] (Figure9).
50
25
y
x
y = 3x
2
+ 12
y = 4x + 4
3 1
3 1 2
FIGURE 9
solution Asthegraphof y = 3x
2
+12liesabovethegraphof y = 4x +4over theinterval [3, 3], theareabetween
thegraphsis
_
3
3
_
(3x
2
+12) (4x +4)
_
dx =
_
3
3
(3x
2
4x +8) dx =
_
x
3
2x
2
+8x
_
3
3
= 102.
317
318 C HA P T E R 6 APPLICATIONS OF THE INTEGRAL
Findtheareaof theregionbetweenthegraphsof f (x) = 3x +8andg(x) = x
2
+2x +2over [0, 2].
3. Findtheareaof theregionenclosedbythegraphsof f (x) = x
2
+2andg(x) = 2x +5(Figure10).
g(x) = 2x + 5
f (x) = x
2
+ 2
1 1 2 3
10
y
x
FIGURE 10
solution Fromthegure, weseethat thegraphof g(x) = 2x + 5liesabovethegraphof f (x) = x
2
+ 2over the
interval [1, 3]. Thus, theareabetweenthegraphsis
_
3
1
_
(2x +5)
_
x
2
+2
__
dx =
_
3
1
_
x
2
+2x +3
_
dx
=
_
1
3
x
3
+x
2
+3x
_
3
1
= 9
_
5
3
_
=
32
3
.
Findtheareaof theregionenclosedbythegraphsof f (x) = x
3
10x andg(x) = 6x (Figure11).
In Exercises 5 and 6, sketch the region between y = sinx and y = cosx over the interval and nd its area.
5.
_
4
,
2
_
solution Over theinterval [
4
,

2
], thegraphof y = cosx liesbelowthatof y = sinx (seethesketchbelow). Hence,
theareabetweenthetwocurvesis
_
/2
/4
(sinx cosx) dx = (cosx sinx)
/2
/4
= (01)
_
2
2

2
2
_
=
21.
y
x
0.2
0.5 1.0 1.5
0.4
0.6
0.8
1.0
y = sin x
y = cos x
[0, ]
In Exercises 7 and 8, let f (x) = 20+x x
2
and g(x) = x
2
5x.
7. Sketchtheregionenclosedbythegraphsof f (x) andg(x) andcomputeitsarea.
solution Settingf (x) = g(x) gives20+x x
2
= x
2
5x, whichsimpliesto
0= 2x
2
6x 20= 2(x 5)(x +2).
Thus, thecurvesintersectatx = 2andx = 5. Withy = 20+x x
2
beingtheupper curve(seethesketchbelow), the
areabetweenthetwocurvesis
_
5
2
_
(20+x x
2
) (x
2
5x)
_
dx =
_
5
2
_
20+6x 2x
2
_
dx =
_
20x +3x
2
2
3
x
3
_
5
2
=
343
3
.
y
x
5
5
10
15
2 2 4
y = 20 + x x
2
y = x
2
5x
S E C T I ON 6.1 Area Between Two Curves 319
Sketchtheregionbetweenthegraphsof f (x) andg(x) over [4, 8] andcomputeitsareaasasumof twointegrals.
9. Findthepointsof intersectionof y = x(x
2
1) andy = 1x
2
. Sketchtheregionenclosedbythesecurves
over [1, 1] andcomputeitsarea.
solution A sketchof theregionboundedbyy = x(x
2
1) andy = 1x
2
isshownbelow; theregionextendsfrom
x = 1tox = 1.
3
2
1
2
3
2 1
0
2
x
y
x(x
2
1)
1 x
2
Asthegraphof y = 1x
2
liesabovethegraphof y = x(x
2
1), theareabetweenthegraphsis
_
1
1
_
(1x
2
) x(x
2
1)
_
dx =
_
1
1
_
x
3
x
2
+x +1
_
dx =
_
1
4
x
4
1
3
x
3
+
1
2
x
2
+x
_
1
1
=
4
3
Findthepointsof intersectionof y = x(4x) andy = x
2
(4x). Sketchtheregionenclosedbythesecurves
over [0, 4] andcomputeitsarea.
11. Sketchtheregionboundedbytheliney = 2andthegraphof y = sec
2
x for
2
< x <

2
andnditsarea.
solution A sketch of theregion bounded by y = sec
2
x and y = 2 is shown below. Notetheregion extends from
x =
4
ontheleft tox =

4
ontheright. Asthegraphof y = 2liesabovethegraphof y = sec
2
x, theareabetween
thegraphsis
_
/4
/4
(2sec
2
x) dx = (2x tanx)
/4
/4
=
_
2
1
_
2
+1
_
= 2.
0.5 0.5
0.5
1
1.5
2
y =sec
2
x
Sketchtheregionboundedby
y =
x
_
1x
2
and y =
x
_
1x
2
for 0 x 0.8andnditsarea.
In Exercises 1316, nd the area of the shaded region in Figures 1215.
13. y
x
2
y = 3x
2
+ 4x 10
y = x
3
2x
2
+ 10
2
FIGURE 12
solution As thegraphof y = x
3
2x
2
+ 10lies abovethegraphof y = 3x
2
+ 4x 10, theareaof theshaded
regionis
_
2
2
_
(x
3
2x
2
+10) (3x
2
+4x 10)
_
dx =
_
2
2
_
x
3
5x
2
4x +20
_
dx
=
_
1
4
x
4
5
3
x
3
2x
2
+20x
_
2
2
=
160
3
.
15.
6
3
2
3
1
2
x
y
y = cosx
( )
,
( ) ,
6

3

2
FIGURE 14
solution Thelineonthetop-left has equationy =
3
x, andthelineonthebottom-right has equationy =

3
2
x.
Thus, theareatotheleft of x =

6
is
_
/6
0
_
3
x
3
2
x
_
dx =
_
3
3
2
x
2
3
4
x
2
_
/6
0
=
3
3
2
2
36

3
4
2
36
=
(2
31)
48
.
Theareatotheright of x =

6
is
_
/3
/6
_
cosx
3
2
x
_
dx =
_
sinx
3
4
x
2
_
/3
/6
=
8
38
16
.
Theentireareaisthen
(2
31)
48
+
8
38
16
=
12
312+(
32)
24
.
In Exercises 17 and 18, nd the area between the graphs of x = siny and x = 1 cosy over the given interval
(Figure 16).
x = 1 cosy
x = siny
x
y
2
2
2
FIGURE 16
17. 0 y

2
solution As shown in thegure, thegraph on theright is x = siny and thegraph on theleft is x = 1 cosy.
Therefore, theareabetweenthetwocurvesisgivenby
_
/2
0
(siny (1cosy)) dy = (cosy y +siny)
/2
0
=
_
2
+1
_
(1) = 2

2
.
2
y

2
19. Findtheareaof theregionlyingtotheright of x = y
2
+4y 22andtotheleft of x = 3y +8.
solution Settingy
2
+4y 22= 3y +8yields
0= y
2
+y 30= (y +6)(y 5),
sothetwocurvesintersect at y = 6andy = 5. Theareainquestionisthengivenby
_
5
6
_
(3y +8) (y
2
+4y 22)
_
dy =
_
5
6
_
y
2
y +30
_
dy =
_
y
3
3

y
2
2
+30y
_
5
6
=
1331
6
.
Findtheareaof theregionlyingtotheright of x = y
2
5andtotheleft of x = 3y
2
.
21. Figure17showstheregionenclosedbyx = y
3
26y +10andx = 406y
2
y
3
. Matchtheequationswiththe
curvesandcomputetheareaof theregion.
x
y
3
1
5
FIGURE 17
solution Substitutingy = 0intotheequationsfor bothcurvesindicatesthat thegraphof x = y
3
26y +10passes
throughthepoint (10, 0) whilethegraphof x = 40 6y
2
y
3
passes throughthepoint (40, 0). Therefore, over the
y-interval [1, 3], thegraphof x = 406y
2
y
3
liestotheright of thegraphof x = y
3
26y +10. Theorientation
of thetwographsisreversedover they-interval [5, 1]. Hence, theareaof theshadedregionis
_
1
5
_
(y
3
26y +10) (406y
2
y
3
)
_
dy +
_
3
1
_
(406y
2
y
3
) (y
3
26y +10)
_
dy
=
_
1
5
_
2y
3
+6y
2
26y 30
_
dy +
_
3
1
_
2y
3
6y
2
+26y +30
_
dy
=
_
1
2
y
4
+2y
3
13y
2
30y
_
1
5
+
_
1
2
y
4
2y
3
+13y
2
+30y
_
3
1
= 256.
Figure18showstheregionenclosedbyy = x
3
6x andy = 83x
2
. Matchtheequationswiththecurvesand
computetheareaof theregion.
In Exercises 23 and 24, nd the area enclosed by the graphs in two ways: by integrating along the x-axis and by integrating
along the y-axis.
23. x = 9y
2
, x = 5
solution Alongthey-axis, wehavepointsof intersectionat y = 2. Therefore, theareaenclosedbythetwocurves
is
_
2
2
_
9y
2
5
_
dy =
_
2
2
_
4y
2
_
dy =
_
4y
1
3
y
3
_
2
2
=
32
3
.
Alongthex-axis, wehaveintegrationlimitsof x = 5andx = 9. Therefore, theareaenclosedbythetwocurvesis
_
9
5
2
9x dx =
4
3
(9x)
3/2
9
5
= 0
_
32
3
_
=
32
3
.
Thesemicubical parabola y
2
= x
3
andthelinex = 1.
In Exercises 25 and 26, nd the area of the region using the method (integration along either the x- or the y-axis) that
requires you to evaluate just one integral.
25. Regionbetweeny
2
= x +5andy
2
= 3x
solution Fromthegurebelow, weseethat integrationalongthex-axiswouldrequiretwointegrals, but integration
alongthey-axisrequiresonly oneintegral. Settingy
2
5= 3 y
2
yieldspointsof intersectionat y = 2. Thus, the
areaisgivenby
_
2
2
_
(3y
2
) (y
2
+5)
_
dy =
_
2
2
_
82y
2
_
dy =
_
8y
2
3
y
3
_
2
2
=
64
3
.
2
1
1
2
y
x
2 4 2
x = y
2
5
x = 3 y
2
Regionbetweeny = x andx +y = 8over [2, 3]
In Exercises 2744, sketch the region enclosed by the curves and compute its area as an integral along the x- or y-axis.
27. y = 4x
2
, y = x
2
4
solution Setting4x
2
= x
2
4yields2x
2
= 8or x
2
= 4. Thus, thecurvesy = 4x
2
andy = x
2
4intersect
at x = 2. Fromthegurebelow, weseethat y = 4 x
2
liesabovey = x
2
4over theinterval [2, 2]; hence, the
areaof theregionenclosedbythecurvesis
_
2
2
_
(4x
2
) (x
2
4)
_
dx =
_
2
2
(82x
2
) dx =
_
8x
2
3
x
3
_
2
2
=
64
3
.
2
4
2
4
y
x
2 1 2 1
y = x
2
4
y = 4 x
2
y = x
2
6, y = 6x
3
, y-axis
29. x +y = 4, x y = 0, y +3x = 4
solution Fromthegraphbelow, weseethat thetopof theregionenclosedby thethreelines is always boundedby
x +y = 4. Ontheother hand, thebottomof theregionisboundedbyy +3x = 4for 0 x 1andbyx y = 0for
1 x 2. Thetotal areaof theregionisthen
_
1
0
((4x) (43x)) dx +
_
2
1
((4x) x) dx =
_
1
0
2x dx +
_
2
1
(42x) dx
= x
2
1
0
+ (4x x
2
)
2
1
= 1+(84) (41) = 2.
2 1 0.5 1.5
x
y = x
y + 3x = 4
x + y = 4
1
4
3
2
y
y = 83x, y = 6x, y = 2
31. y = 8
x, y =
x, x = 0
solution Setting8
x =

x yields
x = 4or x = 16. Usingthegraphshownbelow, weseethat y = 8
x
liesabovey =
x over theinterval [0, 16]. Theareaof theregionenclosedbythesetwocurvesandthey-axisisthen

_
16
0
_
8
x
_
dx =
_
16
0
_
82
x
_
dx =
_
8x
4
3
x
3/2
_
16
0
=
128
3
.
y
x
2
2 4 6 8 10 12 14 16
4
6
8
y = 8 x
1/2
y = x
1/2
y = |x
2
4|, y = 5
33. x = |y|, x = 1|y|
solution Fromthegraphbelow, weseethat theregionenclosedbythecurvesx = |y| andx = 1|y| issymmetric
withrespect tothex-axis. Wecanthereforedeterminethetotal areabydoublingtheareaintherst quadrant. For y > 0,
settingy = 1y yieldsy =
1
2
asthepointof intersection. Moreover, x = 1|y| = 1y liestotherightof x = |y| = y,
sothetotal areaof theregionis
2
_
1/2
0
_
(1y) y
_
dy = 2
_
y y
2
_
1/2
0
= 2
_
1
2
1
4
_
=
1
2
.
1 0.2 0.6 0.8 0.4
x
0.4
0.2
y
0.2
0.4
x = y
x = 1 y
y = |x|, y = x
2
6
35. x = y
3
18y, y +2x = 0
solution Settingy
3
18y =
y
2
yields
0= y
3
35
2
y = y
_
y
2
35
2
_
,
sothepointsof intersectionoccur aty = 0andy =
70
2
. Fromthegraphbelow, weseethatbothcurvesaresymmetric
withrespecttotheorigin. Itfollowsthattheportionof theregionenclosedbythecurvesinthesecondquadrantisidentical
to theregionenclosedinthefourthquadrant. Wecanthereforedeterminethetotal areaenclosedby thetwo curves by
doublingtheareaenclosedinthesecondquadrant. Inthesecondquadrant, y +2x = 0liestotheright of x = y
3
18y,
sothetotal areaenclosedbythetwocurvesis
2
_

70/2
0
_
y
2
(y
3
18y)
_
dy = 2
_
35
4
y
2
1
4
y
4
_
70/2
0
= 2
_
1225
8

1225
16
_
=
1225
8
.
20 10
x
2
y
2
10 20
y + 2x = 0
x = y
3
18y
y = x
x 2, y = x
x 2, x = 4
37. x = 2y, x +1= (y 1)
2
solution Setting2y = (y 1)
2
1yields
0= y
2
4y = y(y 4),
so the two curves intersect at y = 0 and at y = 4. Fromthe graph below, we see that x = 2y lies to the right of
x +1= (y 1)
2
over theinterval [0, 4] alongthey-axis. Thus, theareaof theregionenclosedbythetwocurvesis
_
4
0
_
2y ((y 1)
2
1)
_
dy =
_
4
0
_
4y y
2
_
dy =
_
2y
2
1
3
y
3
_
4
0
=
32
3
.
x
2 4 6 8
x + 1= (y 1)
2
x = 2y
1
2
3
4
y
x +y = 1, x
1/2
+y
1/2
= 1
39. y = cosx, y = cos2x, x = 0, x =
2
3
solution Fromthegraphbelow, weseethat y = cosx liesabovey = cos2x over theinterval [0,
2
3
]. Theareaof
theregionenclosedbythetwocurvesistherefore
_
2/3
0
(cosx cos2x) dx =
_
sinx
1
2
sin2x
_
2/3
0
=
3
3
4
.
0.5 1 1.5 2
x
1
0.5
y
0.5
1
y = cosx
y = cos 2x
y = tanx, y = tanx, x =

4
41. y = sinx, y = csc
2
x, x =

4
solution Over theinterval [
4
,

2
], y = csc
2
x liesabovey = sinx. Theareaof theregionenclosedbythetwocurves
isthen
_
/2
/4
_
csc
2
x sinx
_
dx =
_
cotx +cosx
_
/2
/4
= (00)
_
1+
2
2
_
= 1
2
2
.
y
x
0.2
0.5
1.0
1.5
2.0
0.4 0.6 0.8 1.0 1.2 1.4
y = csc
2
x
y = sin x
x = siny, x =
2
y
43. y = sinx, y = x sin(x
2
), 0 x 1
solution A sketchof theregionisshownbelow:
0.1
0.2 0.4 0.6 0.8 1.0
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 x
y
sin(x)
x sin(x
2
)
Theareaof theregionisthen
A =
_
1
0
_
sinx x sin(x
2
)
_
dx =
_
1
0
sinx dx
_
1
0
x sin(x
2
) dx
= cosx
1
0
_
1
0
x sin(x
2
) dx = 1cos1
_
1
0
x sin(x
2
) dx
For theremainingintegral, usethesubstitutionu = x
2
; thendu = 2x dx andthenewboundof integrationareu = 0to
u = 1. Thus
A = 1cos1
_
1
0
x sin(x
2
) dx = 1cos1
1
2
_
1
0
sinu du = 1cos1
1
2
(cosx)
1
0
=
1
2
(1cos1)
y =
sin(
x)
x
, y = 0,
2
x 9
2
45. Plot
y =
x
_
x
2
+1
and y = (x 1)
2
onthesameset of axes. Useacomputer algebrasystemtondthepoints of intersectionnumerically andcomputethe
areabetweenthecurves.
solution Usingacomputer algebrasystem, wendthat thecurves
y =
x
_
x
2
+1
and y = (x 1)
2
intersect at x = 0.3943285581andat x = 1.942944418. Fromthegraph below, weseethat y =
x
x
2
+1
lies above
y = (x 1)
2
, sotheareaof theregionenclosedbythetwocurvesis
_
1.942944418
0.3943285581
_
x
_
x
2
+1
(x 1)
2
_
dx = 0.7567130951
Thevalueof thedeniteintegral wasalsoobtainedusingacomputer algebrasystem.
y = (x 1)
2
2 1.5 1 0.5
x
1
0.8
0.6
0.4
0.2
y
y =
x
2
+ 1
x
Sketcharegionwhoseareaisrepresentedby
_

2/2
2/2
_
_
1x
2
|x|
_
dx
andevaluateusinggeometry.
47. Athletes1and2runalongastraight trackwithvelocitiesv
1
(t ) andv
2
(t ) (inm/s) asshowninFigure19.
(a) Whichof thefollowingisrepresentedbytheareaof theshadedregionover [0, 10]?
i. Thedistancebetweenathletes1and2at timet = 10s.
ii. Thedifferenceinthedistancetraveledbytheathletesover thetimeinterval [0, 10].
(b) DoesFigure19giveusenoughinformationtodeterminewhoisaheadat timet = 10s?
(c) If theathletesbeginat thesametimeandplace, whoisaheadat t = 10s?At t = 25s?
v
1
v
2
5 10 15 20 25 30
1
2
3
4
5
6
7
(m/s)
t (s)
FIGURE 19
solution
(a) Theareaof theshadedregionover [0, 10] represents(ii): thedifferenceinthedistancetraveledbytheathletesover
thetimeinterval [0, 10].
(b) No, Figure19doesnotgiveusenoughinformationtodeterminewhoisaheadattimet = 10s. Wewouldadditionally
needtoknowtherelativepositionof therunnersat t = 0s.
(c) If theathletes beginat thesametimeandplace, thenathlete1is aheadat t = 10s becausethevelocity graphfor
athlete1liesabovethevelocitygraphfor athlete2over theinterval [0, 10]. Over theinterval [10, 25], thevelocitygraph
for athlete2liesabovethevelocitygraphfor athlete1andappearstohavealarger areathantheareabetweenthegraphs
over [0, 10]. Thus, it appearsthat athlete2isaheadat t = 25s.
Express thearea(not signed) of theshadedregioninFigure20as asumof threeintegrals involvingf (x) and
g(x).
49. Findtheareaenclosedby thecurvesy = c x
2
andy = x
2
c asafunctionof c. Findthevalueof c for which
thisareaisequal to1.
solution Thecurves intersect at x =
c, withy = c x
2
abovey = x
2
c over theinterval [
c,
c]. The
areaof theregionenclosedbythetwocurvesisthen
_

c
c
_
c x
2
) (x
2
c)
_
dx =
_

c
c
_
2c 2x
2
_
dx =
_
2cx
2
3
x
3
_
c
=
8
3
c
3/2
.
Inorder for theareatoequal 1, wemust have
8
3
c
3/2
= 1, whichgives
c =
9
1/3
4
0.520021.
Set up (but do not evaluate) an integral that expresses the area between the circles x
2
+ y
2
= 2 and x
2
+
(y 1)
2
= 1.
51. Set up(but donot evaluate) anintegral that expressestheareabetweenthegraphsof y = (1+x
2
)
1
andy = x
2
.
solution Setting(1+ x
2
)
1
= x
2
yieldsx
4
+ x
2
1 = 0. Thisisaquadratic equationinthevariablex
2
. By the
quadraticformula,
x
2
=
1
14(1)
2
=
1
5
2
.
Asx
2
must benonnegative, wediscard
1
5
2
. Finally, wendthetwocurvesintersect at x =
_
1+
5
2
. Fromthe
graphbelow, weseethat y = (1+x
2
)
1
liesabovey = x
2
. Theareaenclosedbythetwocurvesisthen
_
_
1+
5
2
_
1+
5
2
_
(1+x
2
)
1
x
2
_
dx.
y = x
2
y = (1+ x
2
)
1
0.5 1
1
0.5
0.5 1
x
y
Find anumerical approximation to theareaabovey = 1 (x/) and belowy = sinx (nd thepoints of
intersectionnumerically).
53. Findanumerical approximationtotheareaabovey = |x| andbelowy = cosx.
solution Theregioninquestionisshowninthegurebelow. Weseethat theregionissymmetricwithrespect tothe
y-axis, sowecandeterminethetotal areaof theregionby doublingtheareaof theportionintherst quadrant. Using
acomputer algebrasystem, wendthat y = cosx andy = |x| intersect at x = 0.7390851332. Theareaof theregion
betweenthetwocurvesisthen
2
_
0.7390851332
0
(cosx x) dx = 0.8009772242,
wherethedeniteintegral wasevaluatedusingacomputer algebrasystem.
y = cosx
y = |x|
0.5 1
1
0.5
0.5 1
x
y
Useacomputer algebrasystemto ndanumerical approximationto thenumber c (besides zero) in
_
0,

2
_
,
wherethecurvesy = sinx andy = tan
2
x intersect. Thenndtheareaenclosedbythegraphsover [0, c].
55. Thebackof J onsguitar(Figure21) is19incheslong. J onmeasuredthewidthat1-in. intervals, beginningandending
1
2
in. fromtheends, obtainingtheresults
6, 9, 10.25, 10.75, 10.75, 10.25, 9.75, 9.5, 10, 11.25,
12.75, 13.75, 14.25, 14.5, 14.5, 14, 13.25, 11.25, 9
Usethemidpoint ruletoestimatetheareaof theback.
1
0
.
7
5
1
1
.
2
5
9
1
0
.
2
5
9 6
FIGURE 21 Backof guitar.
solution Notethat themeasurementsweretakenat themidpoint of eachone-inchsectionof theguitar. For example,
inthe0to1inchsection, themidpoint wouldbeat
1
2
inch, andthustheapproximateareaof therst rectanglewouldbe
1 6inches
2
. Anapproximationfor theentireareaisthen
A = 1(6+9+10.25+10.75+10.75+10.25+9.75+9.5+10+11.25
+12.75+13.75+14.25+14.5+14.5+14+13.25+11.25+9)
= 214.75in
2
.
ReferringtoFigure1atthebeginningof thissection, estimatetheprojectednumber of additional joulesproduced
intheyears20092030asaresult of government stimulusspendingin20092010. Note: Onewatt isequal toone
jouleper second, andonegigawatt is10
9
watts.
Exercises 57 and 58 use the notation and results of Exercises 4951 of Section 3.4. For a given country, F(r) is the
fraction of total income that goes to the bottom rth fraction of households. The graph of y = F(r) is called the Lorenz
curve.
57. Let A betheareabetweeny = r andy = F(r) over theinterval [0, 1] (Figure22). TheGini indexisthe
ratioG = A/B, whereB istheareaunder y = r over [0, 1].
(a) Showthat G = 2
_
1
0
(r F(r)) dr.
(b) CalculateG if
F(r) =
_
1
3
r for 0 r
1
2
5
3
r
2
3
for
1
2
r 1
(c) TheGini indexisameasureof incomedistribution, withalower valueindicatingamoreequal distribution. Calculate
G if F(r) = r (inthiscase, all householdshavethesameincomebyExercise51(b) of Section3.4).
(d) What isG if all of theincomegoestoonehousehold? Hint: Inthisextremecase, F(r) = 0for 0 r < 1.
0.8
1.0
0.4 0.6 1.0 0.2
0.8
0.4
0.6
0.2
y
x
y = F(r)
y = r
FIGURE 22 Lorenz Curvefor U.S. in2001.
solution
(a) Becausethegraphof y = r liesabovethegraphof y = FinFigure22,
A =
_
1
0
(r F(r)) dr.
Moreover,
B =
_
1
0
r dr =
1
2
r
2
1
0
=
1
2
.
Thus,
G =
A
B
= 2
_
1
0
(r F(r)) dr.
(b) WiththegivenF(r),
G = 2
_
1/2
0
_
r
1
3
r
_
dr +2
_
1
1/2
_
r
_
5
3
r
2
3
__
dr
=
4
3
_
1/2
0
r dr
4
3
_
1
1/2
(r 1) dr
=
2
3
r
2
1/2
0

4
3
_
1
2
r
2
r
_
1
1/2
=
1
6
4
3
_
1
2
_
+
4
3
_
3
8
_
=
1
3
.
(c) If F(r) = r, then
G = 2
_
1
0
(r r) dr = 0.
(d) If F(r) = 0for 0 r < 1, then
G = 2
_
1
0
(r 0) dr = 2
_
1
2
r
2
_
1
0
= 2
_
1
2
_
= 1.
CalculatetheGini indexof theUnitedStatesintheyear 2001fromtheLorenzcurveinFigure22, whichconsists
of segmentsjoiningthedatapointsinthefollowingtable.
r 0 0.2 0.4 0.6 0.8 1
F(r) 0 0.035 0.123 0.269 0.499 1
59. Findtheliney = mx that dividestheareaunder thecurvey = x(1x) over [0, 1] intotworegionsof equal area.
solution First notethat
_
1
0
x(1x) dx =
_
1
0
_
x x
2
_
dx =
_
1
2
x
2
1
3
x
3
_
1
0
=
1
6
.
Now, theliney = mx andthecurvey = x(1 x) intersect whenmx = x(1 x), or at x = 0andat x = 1 m. The
areaof theregionenclosedbythetwocurvesisthen
_
1m
0
(x(1x) mx) dx =
_
1m
0
_
(1m)x x
2
_
dx =
_
(1m)
x
2
2

1
3
x
3
_
1m
0
=
1
6
(1m)
3
.
Tohave
1
6
(1m)
3
=
1
2
1
6
requires
m = 1
_
1
2
_
1/3
0.206299.
Letc bethenumber suchthattheareaunder y = sinx over [0, ] isdividedinhalf bytheliney = cx (Figure
23). Findanequationfor c andsolvethisequationnumerically usingacomputer algebrasystem.
61. Explaingeometrically(without calculation):
_
1
0
x
n
dx +
_
1
0
x
1/n
dx = 1 (for n > 0)
solution LetA
1
denotetheareaof region1inthegurebelow. DeneA
2
andA
3
similarly. Itisclear fromthegure
that
A
1
+A
2
+A
3
= 1.
Now, notethat x
n
andx
1/n
areinverses of eachother. Therefore, thegraphs of y = x
n
andy = x
1/n
aresymmetric
about theliney = x, soregions1and3arealsosymmetricabout y = x. Thisguaranteesthat A
1
= A
3
. Finally,
_
1
0
x
n
dx +
_
1
0
x
1/n
dx = A
3
+(A
2
+A
3
) = A
1
+A
2
+A
3
= 1.
y
x
1
1
3
2
0 1
Let f (x) beanincreasingfunctionwithinverseg(x). Explaingeometrically:
_
a
0
f (x) dx +
_
f (a)
f (0)
g(x) dx = af (a)
6.2 Setting Up Integrals: Volume, Density, Average Value
1. What istheaveragevalueof f (x) on[0, 4] if theareabetweenthegraphof f (x) andthex-axisisequal to12?
solution Assumingthat f (x) 0over theinterval [1, 4], thefact that theareabetweenthegraphof f andthex-axis
isequal to9indicatesthat
_
4
1
f (x) dx = 9. Theaveragevalueof f over theinterval [1, 4] isthen
_
4
1
f (x) dx
41
=
9
3
= 3.
2. Findthevolumeof asolidextendingfromy = 2toy = 5if everycrosssectionhasareaA(y) = 5.
solution Becausethecross-sectional areaof thesolidisconstant, thevolumeissimplythecross-sectional areatimes
thelength, or 53= 15.
3. What isthedenitionof owrate?
solution Theowrateof auidisthevolumeof uidthat passesthroughacross-sectional areaat agivenpoint per
unit time.
4. Whichassumptionabout uidvelocitydidweusetocomputetheowrateasanintegral?
solution Toexpressowrateasanintegral, weassumedthat theuidvelocitydependedonlyontheradial distance
fromthecenter of thetube.
5. Theaveragevalueof f (x) on[1, 4] is5. Find
_
4
1
f (x) dx.
solution
_
4
1
f (x) dx = averagevalueon[1, 4] lengthof [1, 4]
= 53= 15.
S E C T I ON 6.2 Setting Up Integrals: Volume, Density, Average Value 329
Exercises
1. Let V bethevolumeof apyramidof height 20whosebaseisasquareof side8.
(a) Usesimilar trianglesasinExample1tondtheareaof thehorizontal crosssectionat aheight y.
(b) CalculateV byintegratingthecross-sectional area.
solution
(a) Wecanusesimilar trianglestodeterminethesidelength, s, of thesquarecrosssectionatheighty. Usingthediagram
below, wend
8
20
=
s
20y
or s =
2
5
(20y).
Theareaof thecrosssectionat height y isthengivenby
4
25
(20y)
2
.
s
20
8
20 y
(b) Thevolumeof thepyramidis
_
20
0
4
25
(20y)
2
dy =
4
75
(20y)
3
20
0
=
1280
3
.
Let V bethevolumeof aright circular coneof height 10whosebaseisacircleof radius4[Figure17(A)].
(a) Usesimilar trianglestondtheareaof ahorizontal crosssectionat aheight y.
(b) CalculateV byintegratingthecross-sectional area.
3. Usethemethodof Exercise2tondtheformulafor thevolumeof aright circular coneof height h whosebaseisa
circleof radiusR [Figure17(B)].
solution
(a) Fromsimilar triangles(seeFigure17),
h
h y
=
R
r
0
,
wherer
0
istheradiusof theconeat aheight of y. Thus, r
0
= R
Ry
h
.
(b) Thevolumeof theconeis
_
h
0
_
R
Ry
h
_
2
dy =
h
R
_
R
Ry
h
_
3
3
h
0
=
h
R
R
3
3
=
R
2
h
3
.
Calculatethevolumeof therampinFigure18inthreewaysbyintegratingtheareaof thecrosssections:
(a) Perpendicular tothex-axis(rectangles).
(b) Perpendicular tothey-axis(triangles).
(c) Perpendicular tothez-axis(rectangles).
5. Findthevolumeof liquidneededtoll asphereof radiusR toheight h (Figure19).
R
y
h
FIGURE 19 Spherelledwithliquidtoheight h.
solution Theradiusr at anyheight y isgivenbyr =
_
R
2
(R y)
2
. Thus, thevolumeof thelledportionof the
sphereis
_
h
0
r
2
dy =
_
h
0
_
R
2
(R y)
2
_
dy =
_
h
0
(2Ry y
2
) dy =
_
Ry
2
y
3
3
_
h
0
=
_
Rh
2
h
3
3
_
.
Findthevolumeof thewedgeinFigure20(A) byintegratingtheareaof vertical crosssections.
7. Deriveaformulafor thevolumeof thewedgeinFigure20(B) intermsof theconstantsa, b, andc.
solution The line fromc to a is given by the equation (z/c) + (x/a) = 1 and the line fromb to a is given by
(y/b) + (x/a) = 1. Thecross sections perpendicular tothex-axis areright triangles withheight c(1 x/a) andbase
b(1x/a). Thuswehave
_
a
0
1
2
bc (1x/a)
2
dx =
1
6
abc
_
1
x
a
_
3
a
0
=
1
6
abc.
Let B bethesolidwhosebaseistheunit circlex
2
+ y
2
= 1andwhosevertical crosssectionsperpendicular to
thex-axisareequilateral triangles. Showthatthevertical crosssectionshaveareaA(x) =
3(1x
2
) andcompute
thevolumeof B.
In Exercises 914, nd the volume of the solid with the given base and cross sections.
9. Thebaseistheunit circlex
2
+y
2
= 1, andthecrosssectionsperpendicular tothex-axisaretriangleswhoseheight
andbaseareequal.
solution At eachlocationx, thesideof thetriangular crosssectionthat liesinthebaseof thesolidextendsfromthe
tophalf of theunit circle(withy =
_
1x
2
) tothebottomhalf (withy =
_
1x
2
). Thetrianglethereforehasbase
andheight equal to2
_
1x
2
andarea2(1x
2
). Thevolumeof thesolidisthen
_
1
1
2(1x
2
) dx = 2
_
x
1
3
x
3
_
1
1
=
8
3
.
Thebaseisthetriangleenclosedbyx +y = 1, thex-axis, andthey-axis. Thecrosssectionsperpendicular tothe
y-axisaresemicircles.
11. Thebaseis thesemicircley =
_
9x
2
, where3 x 3. Thecross sections perpendicular to thex-axis are
squares.
solution For eachx, thebaseof thesquarecrosssectionextendsfromthesemicircley =
_
9x
2
tothex-axis. The
squarethereforehasabasewithlength
_
9x
2
andanareaof
__
9x
2
_
2
= 9x
2
. Thevolumeof thesolidisthen
_
3
3
_
9x
2
_
dx =
_
9x
1
3
x
3
_
3
3
= 36.
Thebaseisasquare, oneof whosesidesistheinterval [0, ] alongthex-axis. Thecrosssectionsperpendicular
tothex-axisarerectanglesof height f (x) = x
2
.
13. Thebaseistheregionenclosedbyy = x
2
andy = 3. Thecrosssectionsperpendicular tothey-axisaresquares.
solution At anylocationy, thedistancetotheparabolafromthey-axisis
y. Thusthebaseof thesquarewill have

length2
y. Thereforethevolumeis
_
3
0
_
2
y
_ _
2
y
_
dy =
_
3
0
4y dy = 2y
2
3
0
= 18.
Thebaseistheregionenclosedbyy = x
2
andy = 3. Thecrosssectionsperpendiculartothey-axisarerectangles
of height y
3
.
15. Findthevolumeof thesolidwhosebaseistheregion|x| +|y| 1andwhosevertical crosssectionsperpendicular
tothey-axisaresemicircles(withdiameter alongthebase).
solution TheregionR inquestionisadiamondshapeconnectingthepoints(1, 0), (0, 1), (1, 0), and(0, 1). Thus,
inthelower half of thexy-plane, theradiusof thecirclesisy + 1andintheupper half, theradiusis1 y. Therefore,
thevolumeis
2
_
0
1
(y +1)
2
dy +

2
_
1
0
(1y)
2
dy =

2
_
1
3
+
1
3
_
=

3
.
Showthat apyramidof height h whosebaseisanequilateral triangleof sides hasvolume
3
12
hs
2
.
17. Theareaof an ellipseis ab, wherea and b arethelengths of thesemimajor and semiminor axes (Figure21).
Computethevolumeof aconeof height12whosebaseisanellipsewithsemimajoraxisa = 6andsemiminoraxisb = 4.
Ellipse
12
6 4
FIGURE 21
solution At each height y, the elliptical cross section has major axis
1
2
(12 y) and minor axis
1
3
(12 y). The
cross-sectional areaisthen

6
(12y)
2
, andthevolumeis
_
12
0
6
(12y)
2
dy =
18
(12y)
3
12
0
= 96.
Find the volume V of a regular tetrahedron (Figure 22) whose face is an equilateral triangle of side s. The
tetrahedronhasheight h =
2/3s.
19. A frustumof apyramidisapyramidwithitstopcut off [Figure23(A)]. Let V bethevolumeof afrustumof height
h whosebaseisasquareof sidea andwhosetopisasquareof sideb witha > b 0.
(a) Showthat if thefrustumwerecontinuedtoafull pyramid, it wouldhaveheight ha/(a b) [Figure23(B)].
(b) Showthat thecrosssectionat height x isasquareof side(1/h)(a(h x) +bx).
(c) Showthat V =
1
3
h(a
2
+ ab + b
2
). A papyrusdatingtotheyear 1850bce indicatesthat Egyptianmathematicians
haddiscoveredthisformulaalmost 4000yearsago.
(B) (A)
h
a
b
FIGURE 23
solution
(a) Let H betheheight of thefull pyramid. Usingsimilar triangles, wehavetheproportion
H
a
=
H h
b
whichgives
H =
ha
a b
.
(b) Let w denotethesidelengthof thesquarecrosssectionat height x. Bysimilar triangles, wehave
a
H
=
w
H x
.
Substitutingthevaluefor H frompart (a) gives
w =
a(h x) +bx
h
.
(c) Thevolumeof thefrustrumis
_
h
0
_
1
h
(a(h x) +bx)
_
2
dx =
1
h
2
_
h
0
_
a
2
(h x)
2
+2ab(h x)x +b
2
x
2
_
dx
=
1
h
2
_
a
2
3
(h x)
3
+abhx
2
2
3
abx
3
+
1
3
b
2
x
3
_
h
0
=
h
3
_
a
2
+ab +b
2
_
.
Aplaneinclinedatanangleof 45
passesthroughadiameter of thebaseof acylinder of radiusr. Findthevolume

of theregionwithinthecylinder andbelowtheplane(Figure24).
21. ThesolidS inFigure25istheintersectionof twocylindersof radiusr whoseaxesareperpendicular.
(a) Thehorizontal crosssectionof eachcylinder at distancey fromthecentral axisisarectangular strip. Findthestrips
width.
(b) Findtheareaof thehorizontal crosssectionof S at distancey.
(c) Findthevolumeof S asafunctionof r.
S
y
FIGURE 25 Twocylindersintersectingat right angles.
solution
(a) Thehorizontal crosssectionatdistancey fromthecentral axis(forr y r) isasquareof widthw = 2
_
r
2
y
2
.
(b) Theareaof thehorizontal crosssectionof S at distancey fromthecentral axisisw
2
= 4(r
2
y
2
).
(c) Thevolumeof thesolidS isthen
4
_
r
r
_
r
2
y
2
_
dy = 4
_
r
2
y
1
3
y
3
_
r
r
=
16
3
r
3
.
Let S betheintersectionof twocylindersof radiusr whoseaxesintersect at anangle. Findthevolumeof S as
afunctionof r and.
23. Calculatethevolumeof acylinder inclinedat anangle = 30
withheight 10andbaseof radius4(Figure26).

30
4
10
FIGURE 26 Cylinder inclinedat anangle = 30
.
solution Theareaof eachcircular crosssectionis(4)
2
= 16, hencethevolumeof thecylinder is
_
10
0
16 dx = (16x)
10
0
= 160
Theareasof crosssectionsof LakeNogebowat 5-meter intervalsaregiveninthetablebelow. Figure27shows
acontour mapof thelake. EstimatethevolumeV of thelakeby takingtheaverageof theright- andleft-endpoint
approximationstotheintegral of cross-sectional area.
Depth(m) 0 5 10 15 20
Area(millionm
2
) 2.1 1.5 1.1 0.835 0.217
25. Findthetotal massof a1-mrodwhoselinear densityfunctionis(x) = 10(x +1)
2
kg/mfor 0 x 1.
solution Thetotal massof therodis
_
1
0
(x) dx =
_
1
0
_
10(x +1)
2
_
dx =
_
10(x +1)
1
_
1
0
= 5kg.
Findthetotal massof a2-mrodwhoselinear densityfunctionis(x) = 1+0.5sin(x) kg/mfor 0 x 2.
27. A mineral deposit alongastripof length6cmhasdensity s(x) = 0.01x(6 x) g/cmfor 0 x 6. Calculatethe
total massof thedeposit.
solution Thetotal massof thedeposit is
_
6
0
s(x) dx =
_
6
0
0.01x(6x) dx =
_
0.03x
2
0.01
3
x
3
_
6
0
= 0.36g.
Chargeisdistributedalongaglasstubeof length10cmwithlinear chargedensity(x) = x(x
2
+1)
2
10
4
coulombsper centimeter for 0 x 10. Calculatethetotal charge.
29. Calculatethepopulationwithina10-mileradiusof thecity center if theradial populationdensity is(r) = 4(1+
r
2
)
1/3
(inthousandsper squaremile).
solution Thetotal populationis
2
_
10
0
r (r) dr = 2
_
10
0
4r(1+r
2
)
1/3
dr = 3(1+r
2
)
4/3
10
0
4423.59thousand 4.4million.
OdzalaNational ParkintheRepublicof theCongohasahighdensityof gorillas. Supposethattheradial population
densityis(r) = 52(1+r
2
)
2
gorillasper squarekilometer, wherer isthedistancefromagrassyclearingwitha
sourceof water. Calculatethenumber of gorillaswithina5-kmradiusof theclearing.
31. Table1liststhepopulationdensity(inpeopleper squarekilometer) asafunctionof distancer (inkilometers) from
thecenter of arural town. Estimatethetotal populationwithina1.2-kmradiusof thecenter bytakingtheaverageof the
left- andright-endpoint approximations.
TABLE 1 Population Density
r (r) r (r)
0.0 125.0 0.8 56.2
0.2 102.3 1.0 46.0
0.4 83.8 1.2 37.6
0.6 68.6
solution Thetotal populationisgivenby
2
_
1.2
0
r (r) dr.
Withr = 0.2, theleft- andright-endpoint approximationstotherequireddeniteintegral are
L
6
= 0.2(2)[0(125) +(0.2)(102.3) +(0.4)(83.8) +(0.6)(68.6) +(0.8)(56.2) +(1)(46)]
= 233.86;
R
10
= 0.2(2)[(0.2)(102.3) +(0.4)(83.8) +(0.6)(68.6) +(0.8)(56.2) +(1)(46) +(1.2)(37.6)]
= 290.56.
Thisgivesanaverageof 262.21. Thus, thereareroughly262peoplewithina1.2-kmradiusof thetowncenter.
Findthetotal massof acircular plateof radius20cmwhosemassdensity istheradial function(r) = 0.03+
0.01cos(r
2
) g/cm
2
.
33. Thedensityof deer inaforest istheradial function(r) = 150(r
2
+2)
2
deer per squarekilometer, wherer isthe
distance(inkilometers) toasmall meadow. Calculatethenumber of deer intheregion2 r 5km.
solution Thenumber of deer intheregion2 r 5kmis
2
_
5
2
r (150)
_
r
2
+2
_
2
dr = 150
_
1
r
2
+2
_
5
2
= 150
_
1
27
1
6
_
61deer.
Showthat acircular plateof radius 2cmwithradial mass density (r) =
4
r
g/cm
2
has nitetotal mass, even
thoughthedensitybecomesinniteat theorigin.
35. Findtheowratethroughatubeof radius4cm, assumingthat thevelocityof uidparticlesat adistancer cmfrom
thecenter isv(r) = (16r
2
) cm/s.
solution Theowrateis
2
_
R
0
rv(r) dr = 2
_
4
0
r
_
16r
2
_
dr = 2
_
8r
2
1
4
r
4
_
4
0
= 128
cm
3
s
.
Thevelocityof uidparticlesowingthroughatubeof radius5cmisv(r) = (100.3r 0.34r
2
) cm/s, where
r cmisthedistancefromthecenter. What quantity per secondof uidowsthroughtheportionof thetubewhere
0 r 2?
37. A solid rod of radius 1 cmis placed in apipeof radius 3 cmso that their axes arealigned. Water ows through
the pipe and around the rod. Find the ow rate if the velocity of the water is given by the radial function v(r) =
0.5(r 1)(3r) cm/s.
solution Theowrateis
2
_
3
1
r(0.5)(r 1)(3r) dr =
_
3
1
_
r
3
+4r
2
3r
_
dr =
_
1
4
r
4
+
4
3
r
3
3
2
r
2
_
3
1
=
8
3
cm
3
s
.
Letv(r) bethevelocityof bloodinanarterial capillaryof radiusR = 410
5
m. UsePoiseuillesLaw(Exam-
ple6) withk = 10
6
(m-s)
1
to determinethevelocity at thecenter of thecapillary andtheowrate(usecorrect
units).
In Exercises 3948, calculate the average over the given interval.
39. f (x) = x
3
, [0, 4]
solution Theaverageis
1
40
_
4
0
x
3
dx =
1
4
_
4
0
x
3
dx =
1
16
x
4
4
0
= 16.
f (x) = x
3
, [1, 1]
41. f (x) = cosx,
_
0,

6
_
1
/60
_
/6
0
cosx dx =
6
_
/6
0
cosx dx =
6
sinx
/6
0
=
3
.
f (x) = sec
2
x,
_
6
,

3
_ 43. f (s) = s
2
, [2, 5]
1
52
_
5
2
s
2
ds =
1
3
s
1
5
2
=
1
10
.
f (x) =
sin(/x)
x
2
, [1, 2]
45. f (x) = 2x
3
6x
2
, [1, 3]
1
3(1)
_
3
1
(2x
3
6x
2
) dx =
1
4
_
3
1
(2x
3
6x
2
) dx =
1
4
_
1
2
x
4
2x
3
_
3
1
=
1
4
_
27
2

5
2
_
= 4.
f (x) =
x
(x
2
+16)
3/2
, [0, 3]
47. f (x) = x
n
for n 0, [0, 1]
solution For n > 1, theaverageis
1
10
_
1
0
x
n
dx =
_
1
0
x
n
dx =
1
n +1
x
n+1
1
0
=
1
n +1
.
f (x) = sin(nx), [0, ]
49. Thetemperature(in
C) at timet (inhours) inanart museumvariesaccordingtoT (t ) = 20+5cos

_

12
t
_
. Findthe
averageover thetimeperiods[0, 24] and[2, 6].
solution
Theaveragetemperatureover the24-hour periodis
1
240
_
24
0
_
20+5cos
_

12
t
__
dt =
1
24
_
20t +
60
sin
_

12
t
_
_
24
0
= 20
C.
Theaveragetemperatureover the4-hour periodis
1
62
_
6
2
_
20+5cos
_

12
t
__
dt =
1
4
_
20t +
60
sin
_

12
t
_
_
6
2
= 22.4
C.
A ball thrownintheair verticallyfromgroundlevel withinitial velocity18m/shasheight h(t ) = 18t 9.8t
2
at
timet (inseconds). Findtheaverageheight andtheaveragespeedover thetimeinterval extendingfromtheballs
releasetoitsreturntogroundlevel.
51. Findtheaveragespeedover thetimeinterval [1, 5] of aparticlewhosepositionat timet iss(t ) = t
3
6t
2
m/s.
solution Theaveragespeedover thetimeinterval [1, 5] is
1
51
_
5
1
|s
(t )| dt.
Becauses
(t ) = 3t
2
12t = 3t (t 4), it followsthat
_
5
1
|s
(t )| dt =
_
4
1
(12t 3t
2
) dt +
_
5
4
(3t
2
12t ) dt
= (6t
2
t
3
)
4
1
+(t
3
6t
2
)
5
4
= (9664) (61) +(125150) (6496)
= 34.
Thus, theaveragespeedis
34
4
=
17
2
m/s.
Anobject withzeroinitial velocityacceleratesat aconstant rateof 10m/s
2
. Finditsaveragevelocityduringthe
rst 15seconds.
53. Theaccelerationof aparticleis a(t ) = 60t 4t
3
m/s
2
. Computetheaverageaccelerationandtheaveragespeed
over thetimeinterval [2, 6], assumingthat theparticlesinitial velocityiszero.
solution Theaverageaccelerationover thetimeinterval [2, 6] is
1
62
_
6
2
(60t 4t
3
) dt =
1
4
(30t
2
t
4
)
6
2
=
1
4
[(10801296) (12016)]
=
320
4
= 80m/s
2
.
Givena(t ) = 60t 4t
3
andv(0) = 0, it followsthat v(t ) = 30t
2
t
4
. Now, averagespeedisgivenby
1
62
_
6
2
|v(t )| dt.
Basedontheformulafor v(t ),
_
6
2
|v(t )| dt =
_

30
2
(30t
2
t
4
) dt +
_
6
30
(t
4
30t
2
) dt
=
_
10t
3
1
5
t
5
_
30
2
+
_
1
5
t
5
10t
3
_
30
= 120
30
368
5

3024
5
+120
30
= 240
30
3392
5
.
Finally, theaveragespeedis
1
4
_
240
30
3392
5
_
= 60
30
848
5
159.03m/s.
What istheaverageareaof thecircleswhoseradii varyfrom0toR?
55. Let M betheaveragevalueof f (x) = x
4
on[0, 3]. Findavalueof c in[0, 3] suchthat f (c) = M.
solution Wehave
M =
1
30
_
3
0
x
4
dx =
1
3
_
3
0
x
4
dx =
1
15
x
5
3
0
=
81
5
.
ThenM = f (c) = c
4
=
81
5
impliesc =
3
5
1/4
= 2.006221.
Let f (x) =
x. Findavalueof c in[4, 9] suchthat f (c) isequal totheaverageof f on[4, 9].

57. Let M betheaveragevalueof f (x) = x
3
on[0, A], whereA > 0. Whichtheoremguaranteesthat f (c) = M hasa
solutionc in[0, A]? Findc.
solution TheMeanValueTheoremforIntegralsguaranteesthatf (c) = M hasasolutionc in[0, A]. Withf (x) = x
3
on[0, A],
M =
1
A 0
_
A
0
x
3
dx =
1
A
1
4
x
4
A
0
=
A
3
4
.
Solvingf (c) = c
3
=
A
3
4
for c yields
c =
A
3
4
.
Let f (x) = 2sinx x. Useacomputer algebrasystemtoplot f (x) andestimate:
(a) Thepositiveroot of f (x).
(b) TheaveragevalueM of f (x) on[0, ].
(c) A valuec [0, ] suchthat f (c) = M.
59. Whichof f (x) = x sin
2
x andg(x) = x
2
sin
2
x hasalarger averagevalueover [0, 1]? Over [1, 2]?
solution Thefunctions f and g differ only in thepower of x multiplying sin
2
x. It is also important to notethat
sin
2
x 0for all x. Now, for eachx (0, 1), x > x
2
so
f (x) = x sin
2
x > x
2
sin
2
x = g(x).
Thus, over [0, 1], f (x) will havealarger averagevaluethang(x). Ontheother hand, for eachx (1, 2), x
2
> x, so
g(x) = x
2
sin
2
x > x sin
2
x = f (x).
Thus, over [1, 2], g(x) will havethelarger averagevalue.
Findtheaverageof f (x) = ax +b over theinterval [M, M], wherea, b, andM arearbitraryconstants.
61. Sketchthegraphof afunctionf (x) suchthat f (x) 0on[0, 1] andf (x) 0on[1, 2], whoseaverageon
[0, 2] isnegative.
solution Manysolutionswill exist. Onecouldbe
1
2
1
y
x
1 2
Giveanexampleof afunction(necessarily discontinuous) that does not satisfy theconclusionof theMVT for
Integrals.
63. Anobjectistossedintotheair verticallyfromgroundlevel withinitial velocityv
0
ft/sattimet = 0. Findtheaverage
speedof theobject over thetimeinterval [0, T ], whereT isthetimetheobject returnstoearth.
solution Theheight is givenby h(t ) = v
0
t 16t
2
. Theball is at groundlevel at timet = 0andT = v
0
/16. The
velocityisgivenbyv(t ) = v
0
32t andthusthespeedisgivenbys(t ) = |v
0
32t |. Theaveragespeedis
1
v
0
/160
_
v
0
/16
0
|v
0
32t | dt =
16
v
0
_
v
0
/32
0
(v
0
32t ) dt +
16
v
0
_
v
0
/16
v
0
/32
(32t v
0
) dt
=
16
v
0
_
v
0
t 16t
2
_
v
0
/32
0
+
16
v
0
_
16t
2
v
0
t
_
v
0
/16
v
0
/32
= v
0
/2.
ReviewtheMVT statedinSection4.3(Theorem1, p. 194) andshowhowit canbeused, together withthe
Fundamental Theoremof Calculus, toprovetheMVT for Integrals. 6.3 Volumes of Revolution
1. Whichof thefollowingisasolidof revolution?
(a) Sphere (b) Pyramid (c) Cylinder (d) Cube
solution Thesphereandthecylinder havecircular crosssections; hence, thesearesolidsof revolution. Thepyramid
andcubedonot havecircular crosssections, sothesearenot solidsof revolution.
2. True or false? When the region under a single graph is rotated about the x-axis, the cross sections of the solid
perpendicular tothex-axisarecircular disks.
solution True. Thecrosssectionswill bediskswithradiusequal tothevalueof thefunction.
3. True or false? When the region between two graphs is rotated about the x-axis, the cross sections to the solid
perpendicular tothex-axisarecircular disks.
solution False. Thecrosssectionsmaybewashers.
4. Whichof thefollowingintegralsexpressesthevolumeobtainedbyrotatingtheareabetweeny = f (x) andy = g(x)
over [a, b] aroundthex-axis? [Assumef (x) g(x) 0.]
(a)
_
b
a
_
f (x) g(x)
_
2
dx (b)
_
b
a
_
f (x)
2
g(x)
2
_
dx
solution Thecorrectanswer is(b). Crosssectionsof thesolidwill bewasherswithouter radiusf (x) andinner radius
g(x). Theareaof thewasher isthenf (x)
2
g(x)
2
= (f (x)
2
g(x)
2
).
Exercises
In Exercises 14, (a) sketch the solid obtained by revolving the region under the graph of f (x) about the x-axis over the
given interval, (b) describe the cross section perpendicular to the x-axis located at x, and (c) calculate the volume of the
solid.
1. f (x) = x +1, [0, 3]
solution
(a) A sketchof thesolidof revolutionisshownbelow:
2
2
y
x
1 2 3
S E C T I ON 6.3 Volumes of Revolution 337
(b) Eachcrosssectionisadiskwithradiusx +1.
(c) Thevolumeof thesolidof revolutionis
_
3
0
(x +1)
2
dx =
_
3
0
(x
2
+2x +1) dx =
_
1
3
x
3
+x
2
+x
_
3
0
= 21.
f (x) = x
2
, [1, 3]
3. f (x) =
x +1, [1, 4]
solution
(a) A sketchof thesolidof revolutionisshownbelow:
2
1
2
1
y
x
1 2 3 4
(b) Eachcrosssectionisadiskwithradius
x +1.
_
4
1
(
x +1)
2
dx =
_
4
1
(x +1) dx =
_
1
2
x
2
+x
_
4
1
=
21
2
.
f (x) = x
1
, [1, 4]
In Exercises 512, nd the volume of revolution about the x-axis for the given function and interval.
5. f (x) = 3x x
2
, [0, 3]
solution Thevolumeof thesolidof revolutionis
_
3
0
(3x x
2
)
2
dx =
_
3
0
(9x
2
6x
3
+x
4
) dx =
_
3x
3
3
2
x
4
+
1
5
x
5
_
3
0
=
81
10
.
f (x) =
1
x
2
, [1, 4]
7. f (x) = x
5/3
, [1, 8]
_
8
1
(x
5/3
)
2
dx =
_
8
1
x
10/3
dx =
3
13
x
13/3
8
1
=
3
13
(2
13
1) =
24573
13
.
f (x) = 4x
2
, [0, 2]
9. f (x) =
2
x +1
, [1, 3]
_
3
1
_
2
x +1
_
2
dx = 4
_
3
1
(x +1)
2
dx = 4 (x +1)
1
3
1
= .
f (x) =
_
x
4
+1, [1, 3]
11. f (x) = cscx,
_
4
,
3
4
_
_
3/4
/4
csc
2
x dx = cotx
3/4
/4
= 2
f (x) =
cosx sinx,
_
0,

2
_
In Exercises 13 and 14, R is the shaded region in Figure 11.
x
y
a b
y = f (x)
9
2
y = g(x)
R
FIGURE 11
13. Whichof theintegrands(i)(iv) isusedtocomputethevolumeobtainedbyrotatingregionR about y = 2?
(i) (f (x)
2
+2
2
) (g(x)
2
+2
2
)
(ii) (f (x) +2)
2
(g(x) +2)
2
(iii) (f (x)
2
2
2
) (g(x)
2
2
2
)
(iv) (f (x) 2)
2
(g(x) 2)
2
solution whentheregionR is rotatedabout y = 2, theouter radius is f (x) (2) = f (x) + 2andtheinner
radiusisg(x) (2) = g(x) +2. Thus, theappropriateintegrandis(ii): (f (x) +2)
2
(g(x) +2)
2
.
Whichof theintegrands(i)(iv) isusedtocomputethevolumeobtainedbyrotatingR about y = 9?
(i) (9+f (x))
2
(9+g(x))
2
(ii) (9+g(x))
2
(9+f (x))
2
(iii) (9f (x))
2
(9g(x))
2
(iv) (9g(x))
2
(9f (x))
2
In Exercises 1520, (a) sketch the region enclosed by the curves, (b) describe the cross section perpendicular to the x-axis
located at x, and (c) nd the volume of the solid obtained by rotating the region about the x-axis.
15. y = x
2
+2, y = 10x
2
solution
(a) Settingx
2
+ 2 = 10 x
2
yields 2x
2
= 8, or x
2
= 4. Thetwo curves thereforeintersect at x = 2. Theregion
enclosedbythetwocurvesisshowninthegurebelow.
4
8
y
2 1
x
1 2
y = 10 x
2
y = x
2
+ 2
(b) Whentheregionisrotatedabout thex-axis, eachcrosssectionisawasher withouter radiusR = 10x
2
andinner
radiusr = x
2
+2.
_
2
2
_
(10x
2
)
2
(x
2
+2)
2
_
dx =
_
2
2
(9624x
2
) dx =
_
96x 8x
3
_
2
2
= 256.
y = x
2
, y = 2x +3
17. y = 16x, y = 3x +12, x = 1
solution
(a) Setting16 x = 3x + 12, wendthat thetwolines intersect at x = 1. Theregionenclosedby thetwocurves is
showninthegurebelow.
10
y
1 0.5
x
0.5 1
y = 16 x
y = 3x + 12
(b) Whentheregionisrotatedabout thex-axis, eachcrosssectionisawasher withouter radiusR = 16 x andinner
radiusr = 3x +12.
_
1
1
_
(16x)
2
(3x +12)
2
_
dx =
_
1
1
(112104x 8x
2
) dx =
_
112x 52x
2
8
3
x
3
_
1
1
=
656
3
.
y =
1
x
, y =
5
2
x
19. y = secx, y = 0, x =
4
, x =

4
solution
(a) Theregioninquestionisshowninthegurebelow.
0.8
1.2
0.4
y
y = sec x
0.4
x
0.4
(b) Whentheregionisrotatedabout thex-axis, eachcrosssectionisacircular diskwithradiusR = secx.
_
/4
/4
(secx)
2
dx = (tanx)
/4
/4
= 2.
y = secx, y = 0, x = 0, x =

4
In Exercises 2124, nd the volume of the solid obtained by rotating the region enclosed by the graphs about the y-axis
over the given interval.
21. x =
y, x = 0; 1 y 4
solution Whentheregioninquestion(showninthegurebelow) isrotatedabout they-axis, eachcrosssectionisa
diskwithradius
y. Thevolumeof thesolidof revolutionis
_
4
1
_
y
_
2
dy =
y
2
2
4
1
=
15
2
.
y
x
2
1
0
4
3
2 1 1.5 0.5
x = y
x =
_
siny, x = 0; 0 y
23. x = y
2
, x =
y
solution Settingy
2
=
y andthensquaringbothsidesyields
y
4
= y or y
4
y = y(y
3
1) = 0,
sothetwocurvesintersect at y = 0andy = 1. Whentheregioninquestion(showninthegurebelow) isrotatedabout
they-axis, eachcrosssectionisawasher withouter radiusR =
y andinner radiusr = y
2
. Thevolumeof thesolidof
revolutionis
_
1
0
_
(
y)
2
(y
2
)
2
_
dy =
_
y
2
2

y
5
5
_
1
0
=
3
10
.
x = y
2
x = y
y
x
1
0 1
x = 4y, x = 16y
2
25. Rotation of theregion in Figure12 about they-axis produces asolid with two types of different cross sections.
Computethevolumeasasumof twointegrals, onefor 12 y 4andonefor 4 y 12.
y
2
12
12
4
x
y
y = 12 4x
y = 8x 12
FIGURE 12
solution For 12 y 4, thecrosssectionisadisk withradius
1
8
(y +12); for 4 y 12, thecrosssectionisa
diskwithradius
1
4
(12y). Therefore, thevolumeof thesolidof revolutionis
V =

8
_
4
12
(y +12)
2
dy +

4
_
12
4
(12y)
2
dy
=

24
(y +12)
3
4
12

12
(12y)
3
12
4
=
512
3
+
128
3
=
640
3
.
Let R betheregionenclosedbyy = x
2
+2, y = (x 2)
2
andtheaxesx = 0andy = 0. Computethevolume
V obtainedbyrotatingR about thex-axis. Hint: ExpressV asasumof twointegrals.
In Exercises 2732, nd the volume of the solid obtained by rotating region A in Figure 13 about the given axis.
x
y
1 2
6
2
y = x
2
+ 2
A
B
FIGURE 13
27. x-axis
solution Rotating region A about thex-axis produces asolid whosecross sections arewashers with outer radius
R = 6andinner radiusr = x
2
+2. Thevolumeof thesolidof revolutionis
_
2
0
_
(6)
2
(x
2
+2)
2
_
dx =
_
2
0
(324x
2
x
4
) dx =
_
32x
4
3
x
3
1
5
x
5
_
2
0
=
704
15
.
y = 2
29. y = 2
solution Rotating theregion A about y = 2 produces asolid whosecross sections arewashers with outer radius
R = 62= 4andinner radiusr = x
2
+22= x
2
. Thevolumeof thesolidof revolutionis
_
2
0
_
4
2
(x
2
)
2
_
dx =
_
16x
1
5
x
5
_
2
0
=
128
5
.
y-axis
31. x = 3
solution Rotating region A about x = 3 produces a solid whose cross sections are washers with outer radius
R =
y 2(3) =
y 2+3andinner radiusr = 0(3) = 3. Thevolumeof thesolidof revolutionis
_
6
2
_
(3+
_
y 2)
2
(3)
2
_
dy =
_
6
2
(6
_
y 2+y 2) dy =
_
4(y 2)
3/2
+
1
2
y
2
2y
_
6
2
= 40.
x = 2
In Exercises 3338, nd the volume of the solid obtained by rotating region B in Figure 13 about the given axis.
33. x-axis
solution RotatingregionB aboutthex-axisproducesasolidwhosecrosssectionsarediskswithradiusR = x
2
+2.
Thevolumeof thesolidof revolutionis
_
2
0
(x
2
+2)
2
dx =
_
2
0
(x
4
+4x
2
+4) dx =
_
1
5
x
5
+
4
3
x
3
+4x
_
2
0
=
376
15
.
y = 2
35. y = 6
solution RotatingregionB about y = 6produces asolidwhosecross sections arewashers withouter radius R =
60= 6andinner radiusr = 6(x
2
+2) = 4x
2
_
2
0
_
6
2
(4x
2
)
2
_
dy =
_
2
0
_
20+8x
2
x
4
_
dy =
_
20x +
8
3
x
3
1
5
x
5
_
2
0
=
824
15
.
y-axis
Hint for Exercise 36: Expressthevolumeasasumof twointegralsalongthey-axisor useExercise30.
37. x = 2
solution RotatingregionB about x = 2producesasolidwithtwodifferent crosssections. For eachy [0, 2], the
crosssectionisadisk withradiusR = 2; for eachy [2, 6], thecrosssectionisadisk withradiusR = 2
y 2.
Thevolumeof thesolidof revolutionis
_
2
0
(2)
2
dy +
_
6
2
(2
_
y 2)
2
dy =
_
2
0
4dy +
_
6
2
(2+y 4
_
y 2) dy
= (4y)
2
0
+
_
2y +
1
2
y
2
8
3
(y 2)
3/2
_
6
2
=
32
3
.
x = 3
In Exercises 3952, nd the volume of the solid obtained by rotating the region enclosed by the graphs about the given
axis.
39. y = x
2
, y = 12x, x = 0, about y = 2
solution Rotating theregion enclosed by y = x
2
, y = 12 x and they-axis (shown in thegurebelow) about
y = 2producesasolidwhosecrosssectionsarewasherswithouter radius R = 12 x (2) = 14 x andinner
radiusr = x
2
(2) = x
2
_
3
0
_
(14x)
2
(x
2
+2)
2
_
dx =
_
3
0
(19228x 3x
2
x
4
) dx
=
_
192x 14x
2
x
3
1
5
x
5
_
3
0
=
1872
5
.
y = x
2
y = 12 x
y
x
8
4
0
12
2 3 1
y = x
2
, y = 12x, x = 0, about y = 15
41. y = 162x, y = 6, x = 0, about x-axis
solution Rotatingtheregionenclosedbyy = 16 2x, y = 6andthey-axis(showninthegurebelow) about the
x-axis produces asolidwhosecross sections arewashers withouter radius R = 16 2x andinner radius r = 6. The
volumeof thesolidof revolutionis
_
5
0
_
(162x)
2
6
2
_
dx =
_
5
0
(22064x +4x
2
) dx
=
_
220x 32x
2
+
4
3
x
3
_
5
0
=
1400
3
.
y
x
1
2
4
6
8
10
12
14
16
2 3 4 5
y = 16 2x
y = 6
y = 322x, y = 2+4x, x = 0, about y-axis 43. y = secx, y = 1+
3
x, about x-axis
solution Werst notethat y = secx and y = 1+ (3/)x intersect at x = 0 and x = /3. Rotating theregion
enclosedbyy = secx andy = 1+(3/)x (showninthegurebelow) about thex-axisproducesacrosssectionthat is
awasher withouter radiusR = 1+(3/)x andinner radiusr = secx. Thevolumeof thesolidof revolutionis
V =
_
/3
0
_
_
1+
3
x
_
2
sec
2
x
_
dx
=
_
/3
0
_
1+
6
x +
9
2
x
2
sec
2
x
_
dx
=
_
x +
3
x
2
+
3
2
x
3
tanx
_
/3
0
=
_
3
+

3
+

9

3
_
=
7
2
9

3.
y
x
0.5
0.2 0.4 0.6 0.8 1.0
1.0
1.5
2.0
y = 1 + (3/)x
y = sec x
x = 2, x = 3, y = 16x
4
, y = 0, about y-axis
45. y = 2
x, y = x, about x = 2
solution Setting2
x = x andsquaringbothsidesyields
4x = x
2
or x(x 4) = 0,
so thetwo curves intersect at x = 0andx = 4. Rotatingtheregionenclosedby y = 2
x andy = x (seethegure
below) about x = 2producesasolidwhosecrosssectionsarewasherswithouter radiusR = y (2) = y + 2and
inner radiusr =
1
4
y
2
(2) =
1
4
y
2
V =
_
4
0
_
(y +2)
2
_
1
4
y
2
+2
_
2
_
dy
=
_
4
0
_
4y
1
16
y
4
_
dy
=
_
2y
2
1
80
y
5
_
4
0
=
_
32
64
5
_
=
96
5
.
y
x
1
1 2 3 4
2
3
4
y = 2x
1/2
y = x
y = 2
x, y = x, about y = 4
47. y = x
3
, y = x
1/3
, for x 0, about y-axis
solution Rotatingtheregionenclosedbyy = x
3
andy = x
1/3
(showninthegurebelow) aboutthey-axisproduces
asolidwhosecrosssectionsarewasherswithouter radiusR = y
1/3
andinner radiusr = y
3
. Thevolumeof thesolidof
revolutionis
_
1
0
_
(y
1/3
)
2
(y
3
)
2
_
dy =
_
1
0
(y
2/3
y
6
) dy =
_
3
5
y
5/3
1
7
y
7
_
1
0
=
16
35
.
y
x
0.2
0.2 0.4 0.6 0.8 1.0
0.4
0.6
0.8
1.0
y = x
1/3
y = x
3
y = x
2
, y = x
1/2
, about x = 2
49. y =
9
x
2
, y = 10x
2
, x 0, about y = 12
solution Theregion enclosed by thetwo curves is shown in thegurebelow. Rotating this region about y = 12
producesasolidwhosecrosssectionsarewasherswithouterradiusR = 129x
2
andinnerradiusr = 12(10x
2
) =
2+x
2
_
3
1
_
(129x
2
)
2
(x
2
+2)
2
_
dx =
_
3
1
_
1404x
2
x
4
216x
2
+81x
4
_
dx
=
_
140x
4
3
x
3
1
5
x
5
+216x
1
27x
3
_
3
1
=
1184
15
.
y
x
2
3
4
5
6
7
8
9
0.5 1.0 1.5 2.0 2.5 3.0
y = 10 x
2
y =
9
x
2
y =
9
x
2
, y = 10x
2
, x 0, about x = 1
51. y =
1
x
, y =
5
2
x, about y-axis
solution Wewill rotateabout they-axis, so solvingfor x gives thecurves x =
1
y
andx =
5
2
y. Thesecurves
intersect at y =
1
2
andat y = 2. Rotatingtheregionenclosedbythesecurves(seegurebelow) producesasolidwhose
crosssectionsarewasherswithouter radiusR =
5
2
y andinner radiusr =
1
y
then
_
2
1/2
_
_
5
2
y
_
2
1
y
2
_
dy =
_
1
3
_
5
2
y
_
3
+y
1
_
2
1/2
=
9
8
1 2 3
1
1
0
2
3
4
x
y
1
x
5
2
x
y
2
= 4x, y = x, about y = 8
53. Thebowl inFigure14(A) is 21cmhigh, obtainedby rotatingthecurveinFigure14(B) as indicated. Estimatethe
volumecapacityof thebowl shownbytakingtheaverageof right- andleft-endpoint approximationstotheintegral with
N = 7. Theinner radii (incm) startingfromthetopare0, 4, 7, 8, 10, 13, 14, 20.
20
(A) (B)
21 cm
y
x
19
25
16
21
12
9
30
FIGURE 14
solution Using thegiven values for theinner radii and thevalues in Figure14(B), which indicatethedifference
betweentheinner andouter radii, wend
R
7
= 3
_
(23
2
14
2
) +(25
2
13
2
) +(26
2
10
2
) +(27
2
8
2
) +(28
2
7
2
) +(29
2
4
2
) +(30
2
0
2
)
_
= 3(4490) = 13470
and
L
7
= 3
_
(20
2
20
2
) +(23
2
14
2
) +(25
2
13
2
) +(26
2
10
2
) +(27
2
8
2
) +(28
2
7
2
) +(29
2
4
2
)
_
= 3(3590) = 10770
Averagingthesetwovalues, weestimatethat thevolumecapacityof thebowl is
V = 12120 38076.1cm
3
.
Theregionbetweenthegraphsof f (x) andg(x) over [0, 1] isrevolvedabout theliney = 3. Usethemidpoint
approximationwithvaluesfromthefollowingtabletoestimatethevolumeV of theresultingsolid.
x 0.1 0.3 0.5 0.7 0.9
f (x) 8 7 6 7 8
g(x) 2 3.5 4 3.5 2
55. Findthevolumeof theconeobtainedby rotatingtheregionunder thesegment joining(0, h) and(r, 0) about the
y-axis.
solution Thesegment joining(0, h) and(r, 0) hastheequation
y =
h
r
x +h or x =
r
h
(h y).
Rotatingtheregionunder thissegment about they-axisproducesaconewithvolume
r
2
h
2
_
h
0
(h y)
2
dx =
r
2
3h
2
(h y)
3
h
0
=
1
3
r
2
h.
Thetorus(doughnut-shapedsolid) inFigure15isobtainedbyrotatingthecircle(x a)
2
+y
2
= b
2
aroundthe
y-axis(assumethata > b). Showthatithasvolume2
2
ab
2
. Hint: Evaluatetheintegral byinterpretingitasthearea
of acircle.
57. Sketchthehypocycloidx
2/3
+y
2/3
= 1andndthevolumeof thesolidobtainedbyrevolvingit about the
x-axis.
solution A sketchof thehypocycloidisshownbelow.
1
1
y
x
1 1
Forthehypocycloid, y =
_
1x
2/3
_
3/2
. Rotatingthisregionaboutthex-axiswill produceasolidwhosecrosssections
arediskswithradiusR =
_
1x
2/3
_
3/2
. Thusthevolumeof thesolidof revolutionwill be
_
1
1
_
(1x
2/3
)
3/2
_
2
dx =
_
x
3
3
+
9
7
x
7/3
9
5
x
5/3
+x
_
1
1
=
32
105
.
Thesolidgeneratedby rotatingtheregionbetweenthebranchesof thehyperbolay
2
x
2
= 1about thex-axis
iscalledahyperboloid(Figure16). Findthevolumeof thehyperboloidfor a x a.
59. A bead isformedbyremovingacylinder of radiusr fromthecenter of asphereof radiusR (Figure17). Findthe
volumeof thebeadwithr = 1andR = 2.
y
x
h
r
y
x
R
FIGURE 17 A beadisaspherewithacylinder removed.
solution Theequationof theoutercircleisx
2
+y
2
= 2
2
, andtheinnercylinderintersectsthespherewheny =
3.
Eachcrosssectionof thebeadisawasher withouter radius
_
4y
2
andinner radius1, sothevolumeisgivenby
_

3
3
_
__
4y
2
_
2
1
2
_
dy =
_

3
3
_
3y
2
_
dy = 4
3.
FindthevolumeV of thebead(Figure17) intermsof r andR. ThenshowthatV =

6
h
3
, whereh istheheight
of thebead. Thisformulahasasurprisingconsequence: SinceV canbeexpressedintermsof h alone, itfollowsthat
twobeadsof height 1cm, oneformedfromaspherethesizeof anorangeandtheother fromaspherethesizeof the
earth, wouldhavethesamevolume! Canyouexplainintuitivelyhowthisispossible?
61. Thesolidgeneratedby rotatingtheregioninsidetheellipsewithequation
_
x
a
_
2
+
_
y
b
_
2
= 1aroundthex-axis is
calledanellipsoid. Showthat theellipsoidhas volume
4
3
ab
2
. What is thevolumeif theellipseis rotatedaroundthe
y-axis?
solution
Rotating the ellipse about the x-axis produces an ellipsoid whose cross sections are disks with radius R =
b
_
1(x/a)
2
. Thevolumeof theellipsoidisthen
_
a
a
_
b
_
1(x/a)
2
_
2
dx = b
2
_
a
a
_
1
1
a
2
x
2
_
dx = b
2
_
x
1
3a
2
x
3
_
a
a
=
4
3
ab
2
.
Rotating the ellipse about the y-axis produces an ellipsoid whose cross sections are disks with radius R =
a
_
1(y/b)
2
. Thevolumeof theellipsoidisthen
_
b
b
_
a
_
1(y/b)
2
_
2
dy = a
2
_
b
b
_
1
1
b
2
y
2
_
dy = a
2
_
y
1
3b
2
y
3
_
b
b
=
4
3
a
2
b.
Thecurvey = f (x) inFigure18, calledatractrix, has thefollowingproperty: thetangent lineat eachpoint
(x, y) onthecurvehasslope
dy
dx
=
y
_
1y
2
Let R betheshadedregionunder thegraphof 0 x a inFigure18. ComputethevolumeV of thesolidobtained
by revolvingR aroundthex-axisintermsof theconstant c = f (a). Hint: Usethesubstitutionu = f (x) toshow
that
V =
_
1
c
u
_
1u
2
du
63. Verifytheformula
_
x
2
x
1
(x x
1
)(x x
2
) dx =
1
6
(x
1
x
2
)
3
3
Thenprovethat thesolidobtainedbyrotatingtheshadedregioninFigure19about thex-axishasvolumeV =

6
BH
2
,
withB andH asinthegure. Hint: Let x
1
andx
2
betherootsof f (x) = ax + b (mx + c)
2
, wherex
1
< x
2
. Show
that
V =
_
x
2
x
1
f (x) dx
anduseEq. (3).
x
y
B
y = mx + c
y
2
= ax + b
H
FIGURE 19 Theliney = mx +c intersectstheparabolay
2
= ax +b at twopointsabovethex-axis.
solution First, wecalculate
_
x
2
x
1
(x x
1
)(x x
2
) dx =
_
1
3
x
3
1
2
(x
1
+x
2
)x
2
+x
1
x
2
x
_
x
2
x
1
=
1
6
x
3
1
1
2
x
2
1
x
2
+
1
2
x
1
x
2
2
1
6
x
3
2
=
1
6
_
x
3
1
3x
2
1
x
2
+3x
1
x
2
2
x
3
2
_
=
1
6
(x
1
x
2
)
3
.
Now, consider theregionenclosedbytheparabolay
2
= ax + b andtheliney = mx + c, andlet x
1
andx
2
denotethe
x-coordinates of thepoints of intersectionbetweenthetwo curves withx
1
< x
2
. Rotatingtheregionabout they-axis
produces asolidwhosecross sections arewashers withouter radius R =

ax +b andinner radius r = mx + c. The
volumeof thesolidof revolutionisthen
V =
_
x
2
x
1
_
ax +b (mx +c)
2
_
dx
Becausex
1
andx
2
arerootsof theequationax +b (mx +c)
2
= 0andax +b (mx +c)
2
isaquadraticpolynomial
inx withleadingcoefcient m
2
, it followsthat ax +b (mx +c)
2
= m
2
(x x
1
)(x x
2
). Therefore,
V = m
2
_
x
2
x
1
(x x
1
)(x x
2
) dx =

6
m
2
(x
2
x
1
)
3
,
wherewehaveusedEq. (3). Fromthediagram, weseethat
B = x
2
x
1
and H = mB,
so
V =

6
m
2
B
3
=

6
B (mB)
2
=

6
BH
2
.
LetR betheregionintheunitcirclelyingabovethecutwiththeliney = mx +b (Figure20). Assumethepoints
wherethelineintersectsthecirclelieabovethex-axis. Usethemethodof Exercise63toshowthatthesolidobtained
byrotatingR about thex-axishasvolumeV =

6
hd
2
, withh andd asinthegure.
6.4 The Method of Cylindrical Shells
1. Consider theregionR under thegraphof theconstant functionf (x) = h over theinterval [0, r]. Givetheheight
andtheradiusof thecylinder generatedwhenRisrotatedabout:
(a) thex-axis (b) they-axis
solution
(a) Whentheregionisrotatedabout thex-axis, eachshell will haveradiush andheight r.
(b) Whentheregionisrotatedabout they-axis, eachshell will haveradiusr andheight h.
2. Let V bethevolumeof asolidof revolutionabout they-axis.
(a) DoestheShell Methodfor computingV leadtoanintegral withrespect tox or y?
(b) DoestheDiskor Washer Methodfor computingV leadtoanintegral withrespect tox or y?
solution
(a) TheShell methodrequiresslicingthesolidparallel totheaxisof rotation. Inthiscase, thatwill meanslicingthesolid
inthevertical direction, sointegrationwill bewithrespect tox.
(b) TheDisk or Washer methodrequiresslicingthesolidperpendicular totheaxisof rotation. Inthiscase, that means
slicingthesolidinthehorizontal direction, sointegrationwill bewithrespect toy.
Exercises
In Exercises 16, sketch the solid obtained by rotating the region underneath the graph of the function over the given
interval about the y-axis, and nd its volume.
1. f (x) = x
3
, [0, 1]
solution A sketchof thesolidisshownbelow. Eachshell hasradiusx andheight x
3
, sothevolumeof thesolidis
2
_
1
0
x x
3
dx = 2
_
1
0
x
4
dx = 2
_
1
5
x
5
_
1
0
=
2
5
.
1
y
x
1 1
S E C T I ON 6.4 The Method of Cylindrical Shells 347
f (x) =
x, [0, 4]
3. f (x) = x
1
, [1, 3]
solution A sketchof thesolidisshownbelow. Eachshell hasradiusx andheight x
1
2
_
3
1
x(x
1
) dx = 2
_
3
1
1dx = 2 (x)
3
1
= 4.
3 2 1
0.2
0.6
0.8
3 2 1
x
y
f (x) = 4x
2
, [0, 2]
5. f (x) =
_
x
2
+9, [0, 3]
solution Asketchof thesolidisshownbelow. Eachshell hasradiusx andheight
_
x
2
+9, sothevolumeof thesolid
is
2
_
3
0
x
_
x
2
+9dx.
Let u = x
2
+9. Thendu = 2x dx and
2
_
3
0
x
_
x
2
+9dx =
_
18
9
u du =
_
2
3
u
3/2
_
18
9
= 18(2
21).
3 2 1
1
2
4
3 2 1
x
y
f (x) =
x
_
1+x
3
, [1, 4]
In Exercises 712, use the Shell Method to compute the volume obtained by rotating the region enclosed by the graphs as
indicated, about the y-axis.
7. y = 3x 2, y = 6x, x = 0
solution Theregionenclosedbyy = 3x 2, y = 6x andx = 0isshownbelow. Whenrotatingthisregionabout
they-axis, eachshell hasradiusx andheight 6x (3x 2) = 84x. Thevolumeof theresultingsolidis
2
_
2
0
x(84x) dx = 2
_
2
0
(8x 4x
2
) dx = 2
_
4x
2
4
3
x
3
_
2
0
=
32
3
.
y
x
2
2
4
6
0.5 1.0 1.5 2.0
y = 6 x
y = 3x 2
y =
x, y = x
2
9. y = x
2
, y = 8x
2
, x = 0, for x 0
solution Theregionenclosedbyy = x
2
, y = 8x
2
andthey-axisisshownbelow. Whenrotatingthisregionabout
they-axis, eachshell hasradiusx andheight 8x
2
x
2
= 82x
2
. Thevolumeof theresultingsolidis
2
_
2
0
x(82x
2
) dx = 2
_
2
0
(8x 2x
3
) dx = 2
_
4x
2
1
2
x
4
_
2
0
= 16.
y = 8 x
2
y = x
2
y
x
4
2
0
8
6
2 1 1.5 0.5
y = 8x
3
, y = 84x, for x 0
11. y = (x
2
+1)
2
, y = 2(x
2
+1)
2
, x = 2
solution Theregionenclosedby y = (x
2
+ 1)
2
, y = 2 (x
2
+ 1)
2
andx = 2isshownbelow. Whenrotating
thisregionabout they-axis, eachshell hasradiusx andheight 2(x
2
+1)
2
(x
2
+1)
2
= 22(x
2
+1)
2
. The
volumeof theresultingsolidis
2
_
2
0
x(22(x
2
+1)
2
) dx = 2
_
2
0
_
2x
2x
(x
2
+1)
2
_
dx = 2
_
x
2
+
1
x
2
+1
_
2
0
=
32
5
.
y
x
0.5
1.0
1.5
2.0
0.5 1.0 1.5 2.0
y = 2 (x
2
+ 1)
2
y = (x
2
+ 1)
2
y = 1|x 1|, y = 0
In Exercises 13 and 14, use a graphing utility to nd the points of intersection of the curves numerically and then compute
the volume of rotation of the enclosed region about the y-axis.
13. y =
1
2
x
2
, y = sin(x
2
), x 0
solution Theregionenclosedbyy =
1
2
x
2
andy = sinx
2
isshownbelow. Whenrotatingthisregionaboutthey-axis,
eachshell hasradiusx andheight sinx
2
1
2
x
2
. Usingacomputer algebrasystem, wendthat thex-coordinateof the
point of intersectionontheright isx = 1.376769504. Thus, thevolumeof theresultingsolidof revolutionis
2
_
1.376769504
0
x
_
sinx
2
1
2
x
2
_
dx = 1.321975576.
y = sinx
2
y
x
0
1
1
x
2
2
y =
y = 1x
4
, y = x, x 0
In Exercises 1520, sketch the solid obtained by rotating the region underneath the graph of f (x) over the interval about
the given axis, and calculate its volume using the Shell Method.
15. f (x) = x
3
, [0, 1], about x = 2
solution Asketchof thesolidisshownbelow. Eachshell hasradius2x andheightx
3
2
_
1
0
(2x)
_
x
3
_
dx = 2
_
1
0
(2x
3
x
4
) dx = 2
_
x
4
2

x
5
5
_
1
0
=
3
5
.
y
x
0
1
4
f (x) = x
3
, [0, 1], about x = 2
17. f (x) = x
4
, [3, 1], about x = 4
solution A sketchof thesolidisshownbelow. Eachshell hasradius4x andheightx
4
, sothevolumeof thesolid
is
2
_
1
3
(4x)
_
x
4
_
dx = 2
_
1
3
(4x
4
x
3
) dx = 2
_
1
2
x
2
4
3
x
3
_
1
3
=
280
81
.
0.8
0.4
y
2
x
10
f (x) =
1
_
x
2
+1
, [0, 2], about x = 0
19. f (x) = a x witha > 0, [0, a], about x = 1
solution A sketchof thesolidis shownbelow. Eachshell has radius x (1) = x + 1andheight a x, so the
volumeof thesolidis
2
_
a
0
(x +1) (a x) dx = 2
_
a
0
_
a +(a 1)x x
2
_
dx
= 2
_
ax +
a 1
2
x
2
1
3
x
3
_
a
0
= 2
_
a
2
+
a
2
(a 1)
2

a
3
3
_
=
a
2
(a +3)
3
.
2 a
1
2
a
a
f (x) = 1x
2
, [1, 1], x = c withc > 1
In Exercises 2126, sketch the enclosed region and use the Shell Method to calculate the volume of rotation about the
x-axis.
21. x = y, y = 0, x = 1
solution Whentheregionshownbelowis rotatedabout thex-axis, eachshell has radius y andheight 1 y. The
2
_
1
0
y(1y) dy = 2
_
1
0
(y y
2
) dy = 2
_
1
2
y
2
1
3
y
3
_
1
0
=

3
.
y = x
y
x
0
0.2
0.4
0.6
0.8
1
0.2 0.4 0.6 0.8 1
x =
1
4
y +1, x = 3
1
4
y, y = 0
23. x = y(4y), y = 0
solution Whentheregionshownbelowisrotatedabout thex-axis, eachshell hasradiusy andheight y(4y). The
2
_
4
0
y
2
(4y) dy = 2
_
4
0
(4y
2
y
3
) dy = 2
_
4
3
y
3
1
4
y
4
_
4
0
=
128
3
.
y
x
1
2
3
4
1 2 3 4
x = y(4 y)
x = y(4y), x = (y 2)
2
25. y = 4x
2
, x = 0, y = 0
solution Whentheregionshownbelowisrotatedabout thex-axis, eachshell hasradiusy andheight
4y. The
2
_
4
0
y
_
4y dy.
Let u = 4y. Thendu = dy, y = 4u, and
2
_
4
0
y
_
4y dy = 2
_
0
4
(4u)
u du = 2
_
4
0
_
4
u u
3/2
_
du
= 2
_
8
3
u
3/2
2
5
u
5/2
_
4
0
=
256
15
.
x
1
2
3
4
y
2 0 1.5 1 0.5
y = 4 x
2
y = x
1/3
2, y = 0, x = 27
27. Useboth theShell and Disk Methods to calculatethevolumeobtained by rotating theregion under thegraph of
f (x) = 8x
3
for 0 x 2about:
(a) thex-axis (b) they-axis
solution
(a) x-axis: Usingthediskmethod, thecrosssectionsarediskswithradiusR = 8x
3
; hencethevolumeof thesolidis
_
2
0
(8x
3
)
2
dx =
_
64x 4x
4
+
1
7
x
7
_
2
0
=
576
7
.
Withtheshell method, eachshell hasradiusy andheight (8y)
1/3
. Thevolumeof thesolidis
2
_
8
0
y (8y)
1/3
dy
Let u = 8y. Thendy = du, y = 8u and
2
_
8
0
y (8y)
1/3
dy = 2
_
8
0
(8u) u
1/3
du = 2
_
8
0
(8u
1/3
u
4/3
) du
= 2
_
6u
4/3
3
7
u
7/3
_
8
0
=
576
7
.
(b) y-axis: Withtheshell method, eachshell hasradiusx andheight 8x
3
. Thevolumeof thesolidis
2
_
2
0
x(8x
3
) dx = 2
_
4x
2
1
5
x
5
_
2
0
=
96
5
.
Usingthediskmethod, thecrosssectionsarediskswithradiusR = (8y)
1/3
. Thevolumeisthengivenby
_
8
0
(8y)
2/3
dy =
3
5
(8y)
5/3
8
0
=
96
5
.
Sketchthesolidof rotationabout they-axis for theregionunder thegraphof theconstant functionf (x) = c
(wherec > 0) for 0 x r.
(a) Findthevolumewithout usingintegration.
(b) UsetheShell Methodtocomputethevolume.
29. ThegraphinFigure11(A) canbedescribedbybothy = f (x) andx = h(y), whereh istheinverseof f . Let V be
thevolumeobtainedbyrotatingtheregionunder thegraphabout they-axis.
(a) DescribetheguresgeneratedbyrotatingsegmentsAB andCB about they-axis.
(b) Set upintegralsthat computeV bytheShell andDiskMethods.
x
y
x
y
1.3
A
B
A
B
C C
(B) (A)
y = g(x) y = f (x)
x = h(y)
2 2
FIGURE 11
solution
(a) Whenrotatedaboutthey-axis, thesegmentAB generatesadiskwithradiusR = h(y) andthesegmentCB generates
ashell withradiusx andheight f (x).
(b) BasedonFigure11(A) andtheinformationfrompart (a), whenusingtheShell Method,
V = 2
_
2
0
xf (x) dx;
whenusingtheDiskMethod,
V =
_
1.3
0
(h(y))
2
dy.
Let W bethevolumeof thesolidobtainedby rotatingtheregionunder thegraphinFigure11(B) about the
y-axis.
(a) DescribetheguresgeneratedbyrotatingsegmentsA
andA
about they-axis.
(b) Set upanintegral that computesW bytheShell Method.
(c) ExplainthedifcultyincomputingW bytheWasher Method.
31. Let R betheregionunder thegraphof y = 9x
2
for 0 x 2. UsetheShell Methodtocomputethevolumeof
rotationof R about thex-axisasasumof twointegralsalongthey-axis. Hint: Theshellsgenerateddependonwhether
y [0, 5] or y [5, 9].
solution TheregionR is sketchedbelow. Whenrotatingthis regionabout thex-axis, weproduceasolidwithtwo
different shell structures. For 0 y 5, theshell hasradiusy andheight 2; for 5 y 9, theshell hasradiusy and
height
9y. Thevolumeof thesolidistherefore

V = 2
_
5
0
2y dy +2
_
9
5
y
_
9y dy
For therst integral, wecalculate
2
_
5
0
2y dy = 2y
2
5
0
= 50.
For thesecondintegral, wemakethesubstitutionu = 9y, du = dy andnd
2
_
9
5
y
_
9y dy = 2
_
0
4
(9u)
u du
= 2
_
4
0
(9u
1/2
u
3/2
) du
= 2
_
6u
3/2
2
5
u
5/2
_
4
0
= 2
_
48
64
5
_
=
352
5
.
Thus, thetotal volumeis
V = 50 +
352
5
=
602
5
.
y
x
2
0.5 1.0 1.5 2.0
3
4
5
6
7
8
9
Let R betheregionunder thegraphof y = 4x
1
for 1 y 4. UsetheShell Methodtocomputethevolume
of rotationof R about they-axisasasumof twointegralsalongthex-axis.
In Exercises 3338, use the Shell Method to nd the volume obtained by rotating region A in Figure 12 about the given
axis.
x
y
6
2
y = x
2
+ 2
A
B
1 2
FIGURE 12
33. y-axis
solution WhenrotatingregionA about they-axis, eachshell has radius x andheight 6 (x
2
+ 2) = 4 x
2
. The
2
_
2
0
x(4x
2
) dx = 2
_
2
0
(4x x
3
) dx = 2
_
2x
2
1
4
x
4
_
2
0
= 8.
x = 3
35. x = 2
solution WhenrotatingregionA about x = 2, eachshell hasradius2 x andheight 6 (x
2
+ 2) = 4 x
2
. The
2
_
2
0
(2x)
_
4x
2
_
dx = 2
_
2
0
_
82x
2
4x +x
3
_
dx = 2
_
8x
2
3
x
3
2x
2
+
1
4
x
4
_
2
0
=
40
3
.
x-axis
37. y = 2
solution WhenrotatingregionA about y = 2, eachshell has radius y (2) = y + 2andheight
y 2. The
2
_
6
2
(y +2)
_
y 2dy
Let u = y 2. Thendu = dy, y +2= u +4and
2
_
6
2
(y +2)
_
y 2dy = 2
_
4
0
(u +4)
u du = 2
_
2
5
u
5/2
+
8
3
u
3/2
_
4
0
=
1024
15
.
y = 6
In Exercises 3944, use the most convenient method (Disk or Shell Method) to nd the volume obtained by rotating region
B in Figure 12 about the given axis.
39. y-axis
solution Becauseavertical sliceof regionB will produceasolidwithasinglecrosssectionwhileahorizontal slice
will produceasolidwithtwodifferentcrosssections, wewill useavertical slice. Now, becauseavertical sliceisparallel
totheaxisof rotation, wewill usetheShell Method. Eachshell hasradiusx andheightx
2
+2. Thevolumeof theresulting
solidis
2
_
2
0
x(x
2
+2) dx = 2
_
2
0
(x
3
+2x) dx = 2
_
1
4
x
4
+x
2
_
2
0
= 16.
x = 3
41. x = 2
solution Becauseavertical sliceof regionB will produceasolidwithasinglecrosssectionwhileahorizontal slice
will produceasolidwithtwodifferentcrosssections, wewill useavertical slice. Now, becauseavertical sliceisparallel
totheaxisof rotation, wewill usetheShell Method. Eachshell hasradius2 x andheight x
2
+ 2. Thevolumeof the
resultingsolidis
2
_
2
0
(2x)
_
x
2
+2
_
dx = 2
_
2
0
_
2x
2
x
3
+42x
_
dx = 2
_
2
3
x
3
1
4
x
4
+4x x
2
_
2
0
=
32
3
.
x-axis
43. y = 2
solution Becauseavertical sliceof regionB will produceasolidwithasinglecrosssectionwhileahorizontal slicewill
produceasolidwithtwodifferentcrosssections, wewill useavertical slice. Now, becauseavertical sliceisperpendicular
totheaxisof rotation, wewill usetheDiskMethod. Eachdiskhasouter radiusR = x
2
+2(2) = x
2
+4andinner
radiusr = 0(2) = 2. Thevolumeof thesolidisthen
_
2
0
_
(x
2
+4)
2
2
2
_
dx =
_
2
0
(x
4
+8x
2
+12) dx
=
_
1
5
x
5
+
8
3
x
3
+12x
_
2
0
=
_
32
5
+
64
3
+24
_
=
776
15
.
y = 8
In Exercises 4550, use the most convenient method (Disk or Shell Method) to nd the given volume of rotation.
45. Regionbetweenx = y(5y) andx = 0, rotatedabout they-axis
solution Examinethegurebelow, whichshows theregionboundedby x = y(5 y) andx = 0. If theindicated
regionis slicedvertically, thenthetopof theslicelies alongonebranchof theparabolax = y(5 y) andthebottom
lies alongtheother branch. On theother hand, if theregion is slicedhorizontally, then theright endpoint of theslice
alwaysliesalongtheparabolaandleft endpoint alwaysliesalongthey-axis. Clearly, it will beeasier toslicetheregion
horizontally.
Now, supposetheregionisrotatedabout they-axis. Becauseahorizontal sliceisperpendicular tothey-axis, wewill
calculatethevolumeof theresultingsolidusingthediskmethod. Eachcrosssectionisadiskof radiusR = y(5y), so
thevolumeis
_
5
0
y
2
(5y)
2
dy =
_
5
0
(25y
2
10y
3
+y
4
) dy =
_
25
3
y
3
5
2
y
4
+
1
5
y
5
_
5
0
=
625
6
.
y
x
1
1 2 3 4 5 6
2
3
4
5 x = y(5 y)
Regionbetweenx = y(5y) andx = 0, rotatedabout thex-axis
47. RegioninFigure13, rotatedabout thex-axis
x
y
y = x x
12
1
FIGURE 13
solution ExamineFigure13. If theindicatedregionisslicedvertically, thenthetopof thesliceliesalongthecurve
y = x x
12
andthebottomliesalongthecurvey = 0(thex-axis). Ontheother hand, if theregionisslicedhorizontally,
theequationy = x x
12
must besolvedfor x inorder todeterminetheendpoint locations. Clearly, it will beeasier to
slicetheregionvertically.
Now, supposetheregion in Figure13 is rotated about the x-axis. Becausea vertical sliceis perpendicular to the
x-axis, wewill calculatethevolumeof theresultingsolidusingthedisk method. Eachcrosssectionisadisk of radius
R = x x
12
, sothevolumeis
_
1
0
_
x x
12
_
2
dx =
_
1
3
x
3
1
7
x
14
+
1
25
x
25
_
1
0
=
121
525
.
RegioninFigure13, rotatedabout they-axis
49. RegioninFigure14, rotatedabout x = 4
x
y
y = x
3
+ 2
y = 4 x
2
1 2
FIGURE 14
solution ExamineFigure14. If theindicatedregionisslicedvertically, thenthetopof thesliceliesalongthecurve
y = x
3
+2andthebottomliesalongthecurvey = 4x
2
. Ontheother hand, theleftendof ahorizontal sliceswitches
fromy = 4x
2
toy = x
3
+2at y = 3. Here, vertical sliceswill bemoreconvenient.
Now, supposetheregioninFigure14is rotatedabout x = 4. Becauseavertical sliceis parallel to x = 4, wewill
calculatethevolumeof theresultingsolidusingtheshell method. Eachshell hasradius4x andheight x
3
+2(4
x
2
) = x
3
+x
2
2, sothevolumeis
2
_
2
1
(4x)(x
3
+x
2
2) dx = 2
_
1
5
x
5
+
3
4
x
4
+
4
3
x
3
+x
2
8x
_
2
1
=
563
30
.
RegioninFigure14, rotatedabout y = 2
In Exercises 5154, use the Shell Method to nd the given volume of rotation.
51. A sphereof radiusr
solution A sphereof radiusr canbegeneratedby rotatingtheregionunder thesemicircley =
_
r
2
x
2
about the
x-axis. Eachshell hasradiusy andheight
_
r
2
y
2
_
r
2
y
2
_
= 2
_
r
2
y
2
.
Thus, thevolumeof thesphereis
2
_
r
0
2y
_
r
2
y
2
dy.
Let u = r
2
y
2
. Thendu = 2y dy and
2
_
r
0
2y
_
r
2
y
2
dy = 2
_
r
2
0
u du = 2
_
2
3
u
3/2
_
r
2
0
=
4
3
r
3
.
Thebead formedby removingacylinder of radius r fromthecenter of asphereof radius R (comparewith
Exercise59inSection6.3)
53. Thetorusobtainedbyrotatingthecircle(x a)
2
+y
2
= b
2
aboutthey-axis, wherea > b (comparewithExercise56
inSection6.3). Hint: Evaluatetheintegral byinterpretingpart of it astheareaof acircle.
solution Whenrotatingtheregionenclosedbythecircle(x a)
2
+y
2
= b
2
about they-axiseachshell hasradius
x andheight
_
b
2
(x a)
2
_
b
2
(x a)
2
_
= 2
_
b
2
(x a)
2
.
Thevolumeof theresultingtorusisthen
2
_
a+b
ab
2x
_
b
2
(x a)
2
dx.
Let u = x a. Thendu = dx, x = u +a and
2
_
a+b
ab
2x
_
b
2
(x a)
2
dx = 2
_
b
b
2(u +a)
_
b
2
u
2
du
= 4
_
b
b
u
_
b
2
u
2
du +4a
_
b
b
_
b
2
u
2
du.
Now,
_
b
b
u
_
b
2
u
2
du = 0
becausetheintegrandis anoddfunctionandtheintegrationinterval is symmetric withrespect to zero. Moreover, the
other integral isone-half theareaof acircleof radiusb; thus,
_
b
b
_
b
2
u
2
du =
1
2
b
2
.
Finally, thevolumeof thetorusis
4(0) +4a
_
1
2
b
2
_
= 2
2
ab
2
.
Theparaboloid obtainedbyrotatingtheregionbetweeny = x
2
andy = c (c > 0) about they-axis
S E C T I ON 6.5 Work and Energy 355
55. Thesurfaceareaof asphereof radiusr is4r
2
. Usethistoderivetheformulafor thevolumeV of asphere
of radiusR inanewway.
(a) Showthat thevolumeof athinspherical shell of inner radiusr andthicknessr isapproximately4r
2
r.
(b) ApproximateV bydecomposingthesphereof radiusR intoN thinspherical shellsof thicknessr = R/N.
(c) Showthat theapproximationisaRiemannsumthat convergestoanintegral. Evaluatetheintegral.
solution
(a) Thevolumeof athinspherical shell of inner radiusr andthicknessx isgivenbytheproduct of thesurfaceareaof
theshell, 4r
2
andthethickness. Thus, wehave4r
2
x.
(b) Thevolumeof thesphereisapproximatedby
R
N
= 4
_
R
N
_
N
k=1
(x
k
)
2
wherex
k
= k
R
N
.
(c) V = 4 lim
N
_
R
N
_
N
k=1
(x
k
)
2
= 4
_
R
0
x
2
dx = 4
_
1
3
x
3
_
R
0
=
4
3
R
3
.
Showthat thesolid(an ellipsoid) obtainedby rotatingtheregionR inFigure15about they-axis has volume
4
3
a
2
b.
57. Thebell-shapedcurvey = f (x) inFigure16satises dy/dx = xy. UsetheShell Methodandthesubstitution
u = f (x) toshowthat thesolidobtainedby rotatingtheregionR about they-axishasvolumeV = 2(1 c), where
c = f (a). Observethat asc 0, theregionR becomesinnitebut thevolumeV approaches2.
1
y = f (x)
R
c
y
x
a
FIGURE 16 Thebell-shapedcurve.
solution Let y = f (x) betheexponential function depicted in Figure16. When rotating theregion R about the
y-axis, eachshell intheresultingsolidhasradiusx andheight f (x). Thevolumeof thesolidisthen
V = 2
_
a
0
xf (x) dx.
Now, let u = f (x). Thendu = f
(x) dx = xf (x) dx; hence, xf (x)dx = du, and

V = 2
_
c
1
(du) = 2
_
1
c
du = 2(1c).
6.5 Work and Energy
1. Whyisintegrationneededtocomputetheworkperformedinstretchingaspring?
solution Recall that theforceneededtoextendor compressaspringdependsontheamount bywhichthespringhas
alreadybeenextendedor compressedfromitsequilibriumposition. Inother words, theforceneededtomoveaspringis
variable. Whenever theforceisvariable, workneedstobecomputedwithanintegral.
2. Why is integrationneededto computethework performedinpumpingwater out of atank but not to computethe
workperformedinliftingupthetank?
solution Tolift atank throughavertical distanced, theforceneededtomovethetank remainsconstant; hence, no
integral isneededtocalculatethework doneinliftingthetank. Ontheother hand, pumpingwater fromatank requires
thatdifferentlayersof thewaterbeliftedthroughdifferentdistances, and, dependingontheshapeof thetank, mayrequire
different forces. Thus, pumpingwater fromatankrequiresthat anintegral beevaluated.
3. Whichof thefollowingrepresentstheworkrequiredtostretchaspring(withspringconstant k) adistancex beyond
itsequilibriumposition: kx, kx,
1
2
mk
2
,
1
2
kx
2
, or
1
2
mx
2
?
solution Theworkrequiredtostretchaspringwithspringconstant k adistancex beyonditsequilibriumpositionis
_
x
0
ky dy =
1
2
ky
2
x
0
=
1
2
kx
2
.
Exercises
1. Howmuchworkisdoneraisinga4-kgmasstoaheight of 16maboveground?
solution Theforceneededtolift a4-kgobject isaconstant
(4kg)(9.8m/s
2
) = 39.2N.
Theworkdoneinliftingtheobject toaheight of 16misthen
(39.2N)(16m) = 627.2J .
Howmuchworkisdoneraisinga4-lbmasstoaheight of 16ft aboveground?
In Exercises 36, compute the work (in joules) required to stretch or compress a spring as indicated, assuming a spring
constant of k = 800N/m.
3. Stretchingfromequilibriumto12cmpast equilibrium
solution Theworkrequiredtostretchthespring12cmpast equilibriumis
_
0.12
0
800x dx = 400x
2
0.12
0
= 5.76J .
Compressingfromequilibriumto4cmpast equilibrium
5. Stretchingfrom5cmto15cmpast equilibrium
solution Theworkrequiredtostretchthespringfrom5cmto15cmpast equilibriumis
_
0.15
0.05
800x dx = 400x
2
0.15
0.05
= 8J .
Compressing4cmmorewhenit isalreadycompressed5cm
7. If 5J of workareneededtostretchaspring10cmbeyondequilibrium, howmuchworkisrequiredtostretchit15cm
beyondequilibrium?
solution First, wedeterminethevalueof thespringconstant asfollows:
_
0.1
0
kx dx =
1
2
kx
2
0.1
0
= 0.005k = 5J .
Thus, k = 1000N/m. Next, wecalculatetheworkrequiredtostretchthespring15cmbeyondequilibrium:
_
0.15
0
1000x dx = 500x
2
0.15
0
= 11.25J .
Tocreateimagesof samplesat themolecular level, atomic forcemicroscopesusesiliconmicro-cantileversthat
obeyHookesLawF(x) = kx, wherex isthedistancethroughwhichthetipisdeected(Figure6). Supposethat
10
17
J of workarerequiredtodeect thetipadistance10
8
m. Findthedeectionif aforceof 10
9
Nisapplied
tothetip.
9. A springobeysaforcelawF(x) = kx
1.1
withk = 100N/m
1.1
. Findtheworkrequiredtostretchthespring0.3m
past equilibrium.
solution Theworkrequiredtostretchthisspring0.3mpast equilibriumis
_
0.3
0
100x
1.1
dx =
100
1.1
x
2.1
0.3
0
7.25J .
Showthat thework requiredtostretchaspringfrompositiona topositionb is
1
2
k(b
2
a
2
), wherek is the
springconstant. Howdoyouinterpret thenegativeworkobtainedwhen|b| < |a|?
In Exercises 1114, use the method of Examples 2 and 3 to calculate the work against gravity required to build the
structure out of a lightweight material of density 600kg/m
3
.
11. Boxof height 3mandsquarebaseof side2m
solution Thevolumeof onelayer is 4y m
3
andso theweight of onelayer is 23520y N. Thus, thework done
against gravitytobuildthetower is
W =
_
3
0
23520y dy = 11760y
2
3
0
= 105840J .
Cylindrical columnof height 4mandradius0.8m
13. Right circular coneof height 4mandbaseof radius1.2m
solution By similar triangles, thelayer of theconeat aheight y abovethebasehas radius r = 0.3(4 y) meters.
Thus, thevolumeof thesmall layer at this height is 0.09(4 y)
2
y m
3
, and theweight is 529.2(4 y)
2
y N.
Finally, thetotal workdoneagainst gravitytobuildthetower is
_
4
0
529.2(4y)
2
y dy = 11289.6 J 35467.3J .
Hemisphereof radius0.8m
15. Built around2600bce, theGreat Pyramidof GizainEgypt (Figure7) is146mhighandhasasquarebaseof side
230m. Findthework(againstgravity) requiredtobuildthepyramidif thedensityof thestoneisestimatedat2000kg/m
3
.
FIGURE 7 TheGreat PyramidinGiza, Egypt.
solution Fromsimilar triangles, theareaof onelayer is
_
230
230
146
y
_
2
m
2
,
sothevolumeof eachsmall layer is
_
230
230
146
y
_
2
y m
3
.
Theweight of onelayer isthen
19600
_
230
230
146
y
_
2
y N.
Finally, thetotal workneededtobuildthepyramidwas
_
146
0
19600
_
230
230
146
y
_
2
y dy 1.8410
12
J .
Calculatethework (against gravity) requiredto buildabox of height 3mandsquarebaseof side2mout of
material of variabledensity, assumingthat thedensityat height y isf (y) = 1000100y kg/m
3
.
In Exercises 1722, calculate the work (in joules) required to pump all of the water out of a full tank. Distances are in
meters, and the density of water is 1000kg/m
3
.
17. Rectangular tankinFigure8; water exitsfromasmall holeat thetop.
8
4
5
Water exits here.
1
FIGURE 8
solution Placetheoriginonthetopof thebox, andlet thepositivey-axispoint downward. Thevolumeof onelayer
of water is32y m
3
, sotheforceneededtolift eachlayer is
(9.8)(1000)32y = 313600y N.
Eachlayer must beliftedy meters, sothetotal workneededtoemptythetankis
_
5
0
313600y dy = 156800y
2
5
0
= 3.9210
6
J .
Rectangular tankinFigure8; water exitsthroughthespout.
19. HemisphereinFigure9; water exitsthroughthespout.
10
2
FIGURE 9
solution Placetheoriginat thecenter of thehemisphere, andlet thepositivey-axispoint downward. Theradiusof a
layer of water at depthy is
_
100y
2
m, sothevolumeof thelayer is(100y
2
)y m
3
, andtheforceneededtolift
thelayer is9800(100y
2
)y N. Thelayer must beliftedy +2meters, sothetotal workneededtoemptythetankis
_
10
0
9800(100y
2
)(y +2) dy =
112700000
3
J 1.1810
8
J .
Conical tankinFigure10; water exitsthroughthespout.
21. Horizontal cylinder inFigure11; water exitsfromasmall holeat thetop. Hint: Evaluatetheintegral byinterpreting
part of it astheareaof acircle.
r
Water exits here.
FIGURE 11
solution Placetheoriginalongtheaxisof thecylinder.Atlocationy, thelayer of water isarectangular slabof length
, width2
_
r
2
y
2
andthicknessy. Thus, thevolumeof thelayer is2
_
r
2
y
2
y, andtheforceneededtolift the
layer is 19,600
_
r
2
y
2
y. Thelayer must beliftedadistancer y, so thetotal work neededto empty thetank is
givenby
_
r
r
19,600
_
r
2
y
2
(r y) dy = 19,600r
_
r
r
_
r
2
y
2
dy 19,600
_
r
r
y
_
r
2
y
2
dy.
Now,
_
r
r
y
_
r
2
y
2
du = 0
becausetheintegrandis anoddfunctionandtheintegrationinterval is symmetric withrespect to zero. Moreover, the
other integral isone-half theareaof acircleof radiusr; thus,
_
r
r
_
r
2
y
2
dy =
1
2
r
2
.
Finally, thetotal workneededtoemptythetankis
19,600r
_
1
2
r
2
_
19,600(0) = 9800r
3
J .
TroughinFigure12; water exitsbypouringover thesides.
23. FindtheworkW requiredtoemptythetankinFigure8throughtheholeat thetopif thetankishalf full of water.
solution Place the origin on the top of the box, and let the positive y-axis point downward. Note that with this
coordinatesystem, thebottomhalf of theboxcorrespondstoy valuesfrom2.5to5. Thevolumeof onelayer of water is
32y m
3
, sotheforceneededtolift eachlayer is
(9.8)(1000)32y = 313,600y N.
Eachlayer must beliftedy meters, sothetotal workneededtoemptythetankis
_
5
2.5
313,600y dy = 156,800y
2
5
2.5
= 2.9410
6
J .
Assumethetank in Figure8 is full of water and let W bethework required to pump out half of thewater
throughtheholeatthetop. Doyouexpect W toequal theworkcomputedinExercise23?Explainandthencompute
W.
25. AssumethetankinFigure10isfull. Findtheworkrequiredtopumpout half of thewater. Hint: First, determinethe
level H at whichthewater remaininginthetankisequal toone-half thetotal capacityof thetank.
solution Our rst stepistodeterminethelevel H at whichthewater remaininginthetank isequal toone-half the
total capacity of thetank. FromFigure10andsimilar triangles, weseethat theradiusof theconeat level H isH/2so
thevolumeof water is
V =
1
3
r
2
H =
1
3
_
H
2
_
2
H =
1
12
H
3
.
Thetotal capacityof thetankis250/3m
3
, sothewater level whenthewater remaininginthetankisequal toone-half
thetotal capacityof thetanksatises
1
12
H
3
=
125
3
or H =
10
2
1/3
m.
Placetheoriginat thevertexof theinvertedcone, andlet thepositivey-axispoint upward. Now, consider alayer of
water at aheight of y meters. Fromsimilar triangles, theareaof thelayer is
_
y
2
_
2
m
2
,
sothevolumeis
_
y
2
_
2
y m
3
.
Thustheweight of onelayer is
9800
_
y
2
_
2
y N.
Thelayer must belifted12y meters, sothetotal workneededtoemptythehalf-full tankis
_
10
10/2
1/3
9800
_
y
2
_
2
(12y) dy 3.7910
6
J .
Assumethat thetankinFigure10isfull.
(a) CalculatetheworkF(y) requiredtopumpout water until thewater level hasreachedlevel y.
(b) Plot F(y).
(c) What isthesignicanceof F
(y) asarateof change?

(d) If your goal istopumpout all of thewater, at whichwater level y
0
will half of theworkbedone?
27. Calculatethework requiredtolift a10-mchainover thesideof abuilding(Figure13) Assumethat thechainhasa
density of 8kg/m. Hint: Break upthechainintoN segments, estimatethework performedonasegment, andcompute
thelimit asN asanintegral.
Segment of
length y
y
FIGURE 13 Thesmall segment of thechainof lengthy locatedy metersfromthetopisliftedthroughavertical
distancey.
solution In this example, each part of thechain is lifted adifferent distance. Therefore, wedividethechain into
N small segments of length y = 10/N. Supposethat theith segment is located a distancey
i
fromthetop of the
building. Thissegment weighs8(9.8)y kilogramsandit must beliftedapproximatelyy
i
meters(not exactlyy
i
meters,
becauseeachpoint alongthesegment isaslightlydifferent distancefromthetop). TheworkW
i
doneonthissegment is
approximatelyW
i
78.4y
i
y N. Thetotal workW isthesumof theW
i
andwehave
W =
N
j=1
W
i

N
j=1
78.4y
j
y.
Passingtothelimit asN , weobtain
W =
_
10
0
78.4y dy = 39.2y
2
10
0
= 3920 J .
Howmuchworkisdoneliftinga3-mchainover thesideof abuildingif thechainhasmassdensity4kg/m?
29. A 6-mchainhasmass18kg. Findtheworkrequiredtolift thechainover thesideof abuilding.
solution First, notethat thechainhasamassdensityof 3kg/m. Now, consider asegment of thechainof lengthy
locatedadistancey
j
feet fromthetopof thebuilding. Theworkneededtolift thissegment of thechaintothetopof the
buildingisapproximately
W
j
(3y)9.8y
j
ft-lb.
Summingover all segmentsof thechainandpassingtothelimit asy 0, it followsthat thetotal workis
_
6
0
29.4y dy = 14.7y
2
6
0
= 529.2J .
A 10-mchainwithmassdensity4kg/misinitiallycoiledontheground. Howmuchworkisperformedinlifting
thechainsothat it isfullyextended(andoneendtouchestheground)?
31. Howmuchworkisdoneliftinga12-mchainthat hasmassdensity3kg/m(initiallycoiledontheground) sothat its
topendis10mabovetheground?
solution Consider asegmentof thechainof lengthy thatmustbeliftedy
j
feetoff theground. Theworkneededto
lift thissegment of thechainisapproximately
W
j
(3y)9.8y
j
J .
Summingover all segmentsof thechainandpassingtothelimit asy 0, it followsthat thetotal workis
_
10
0
29.4y dy = 14.7y
2
10
0
= 1470J .
A 500-kgwreckingball hangsfroma12-mcableof density15kg/mattachedtoacrane. Calculatetheworkdone
if thecraneliftstheball fromgroundlevel to12mintheair bydrawinginthecable.
33. Calculate the work required to lift a 3-mchain over the side of a building if the chain has variable density of
(x) = x
2
3x +10kg/mfor 0 x 3.
solution Considerasegmentof thechainof lengthx thatmustbeliftedx
j
feet. Theworkneededtoliftthissegment
isapproximately
W
j

_
(x
j
)x
_
9.8x
j
J .
Summingover all segmentsof thechainandpassingtothelimit asx 0, it followsthat thetotal workis
_
3
0
9.8(x)x dx = 9.8
_
3
0
_
x
3
3x
2
+10x
_
dx
= 9.8
_
1
4
x
4
x
3
+5x
2
_
3
0
= 374.85J .
A 3-mchainwithlinear massdensity(x) = 2x(4x) kg/mliesontheground. Calculatetheworkrequiredto
lift thechainsothat itsbottomis2maboveground.
Exercises 3537: The gravitational force between two objects of mass mand M, separated by a distance r, has magnitude
GMm/r
2
, where G = 6.6710
11
m
3
kg
1
s
1
.
35. Showthat if two objects of mass M andm areseparatedby adistancer
1
, thenthework requiredto increasethe
separationtoadistancer
2
isequal toW = GMm(r
1
1
r
1
2
).
solution Theworkrequiredtoincreasetheseparationfromadistancer
1
toadistancer
2
is
_
r
2
r
1
GMm
r
2
dr =
GMm
r
r
2
r
1
= GMm(r
1
1
r
1
2
).
Usetheresultof Exercise35tocalculatetheworkrequiredtoplacea2000-kgsatelliteinanorbit1200kmabove
thesurfaceof theearth.Assumethattheearthisasphereof radiusR
e
= 6.3710
6
mandmassM
e
= 5.9810
24
kg.
Treat thesatelliteasapoint mass.
37. Usetheresultof Exercise35tocomputetheworkrequiredtomovea1500-kgsatellitefromanorbit1000toanorbit
1500kmabovethesurfaceof theearth.
solution Thesatellitewill movefromadistancer
1
= R
e
+ 1,000,000to adistancer
2
= R
e
+ 1,500,000. Thus,
fromExercise35,
W = (6.6710
11
)(5.9810
24
)(1500)
_
1
6.3710
6
+1,000,000
1
6.3710
6
+1,500,000
_
5.1610
9
J .
ThepressureP andvolumeV of thegasinacylinder of length0.8metersandradius0.2meters, withamovable
piston, arerelated by PV
1.4
= k, wherek is aconstant (Figure14). When thepiston is fully extended, thegas
pressureis2000kilopascals(onekilopascal is10
3
newtonsper squaremeter).
(a) Calculatek.
(b) TheforceonthepistonisPA, whereA isthepistonsarea. Calculatetheforceasafunctionof thelengthx of
thecolumnof gas.
(c) Calculatetheworkrequiredtocompressthegascolumnfrom0.8mto0.5m.
39. Work-EnergyTheorem Anobject of massm movesfromx
1
tox
2
duringthetimeinterval [t
1
, t
2
] duetoaforce
F(x) actinginthedirectionof motion. Let x(t ), v(t ), anda(t ) betheposition, velocity, andaccelerationat timet . The
objectskineticenergyisKE =
1
2
mv
2
.
(a) Usethechange-of-variablesformulatoshowthat theworkperformedisequal to
W =
_
x
2
x
1
F(x) dx =
_
t
2
t
1
F(x(t ))v(t ) dt
(b) UseNewtonsSecondLaw, F(x(t )) = ma(t ), toshowthat
d
dt
_
1
2
mv(t )
2
_
= F(x(t ))v(t )
(c) UsetheFTC to provetheWork-Energy Theorem: Thechangeinkinetic energy duringthetimeinterval [t
1
, t
2
] is
equal totheworkperformed.
solution
(a) Let x
1
= x(t
1
) andx
2
= x(t
2
), thenx = x(t ) givesdx = v(t ) dt . Bysubstitutionwehave
W =
_
x
2
x
1
F(x) dx =
_
t
2
t
1
F(x(t ))v(t ) dt.
(b) KnowingF(x(t )) = m a(t ), wehave
d
dt
_
1
2
m v(t )
2
_
= m v(t ) v
(t ) (ChainRule)
= m v(t ) a(t )
= v(t ) F(x(t )) (Newtons2ndlaw)
(c) FromtheFTC,
1
2
m v(t )
2
=
_
F(x(t )) v(t ) dt.
SinceKE =
1
2
mv
2
,
KE = KE(t
2
) KE(t
1
) =
1
2
mv(t
2
)
2
1
2
mv(t
1
)
2
=
_
t
2
t
1
F(x(t )) v(t ) dt.
(d)
W =
_
x
2
x
1
F(x) dx =
_
t
2
t
1
F(x(t )) v(t ) dt (Part (a))
= KE(t
2
) KE(t
1
)
= KE (asrequired)
A model trainof mass 0.5kgis placedat oneendof astraight 3-melectric track. Assumethat aforceF(x) =
(3x x
2
) Nactsonthetrainatdistancex alongthetrack. UsetheWork-EnergyTheorem(Exercise39) todetermine
thevelocityof thetrainwhenit reachestheendof thetrack.
41. Withwhat initial velocity v
0
must werearocket soit attainsamaximumheight r abovetheearth? Hint: Usethe
resultsof Exercises35and39. Astherocket reachesitsmaximumheight, itsKE decreasesfrom
1
2
mv
2
0
tozero.
solution Theworkrequiredtomovetherocket adistancer fromthesurfaceof theearthis
W(r) = GM
e
m
_
1
R
e
1
r +R
e
_
.
As therocket climbs toaheight r, its kinetic energy is reducedby theamount W(r). Therocket reaches its maximum
height whenitskineticenergyisreducedtozero, that is, when
1
2
mv
2
0
= GM
e
m
_
1
R
e
1
r +R
e
_
.
Therefore, itsinitial velocitymust be
v
0
=
_
2GM
e
_
1
R
e
1
r +R
e
_
.
Withwhat initial velocity must werearocket soit attainsamaximumheight of r = 20kmabovethesurface
of theearth?
43. Calculateescapevelocity,theminimuminitial velocityof anobjecttoensurethatitwill continuetravelingintospace
andneverfall backtoearth(assumingthatnoforceisappliedaftertakeoff). Hint: Takethelimitasr inExercise41.
solution Theresultof Exercise41leadstoaninterestingconclusion. Theinitial velocityv
0
requiredtoreachaheight
r doesnot increasebeyondall boundsasr tendstoinnity; rather, it approachesanitelimit, calledtheescapevelocity:
v
esc
= lim
r
_
2GM
e
_
1
R
e
1
r +R
e
_
=
_
2GM
e
R
e
Inother words, v
esc
islargeenoughtoinsurethat therocket reachesaheight r for everyvalueof r! Therefore, arocket
redwithinitial velocityv
esc
never returnstoearth. It continuestravelingindenitelyintoouter space.
Now, letsseehowlargeescapevelocityactuallyis:
v
esc
=
_
2 6.6710
11
5.98910
24
6.3710
6
_
1/2
11,190m/sec.
Sinceonemeter per secondisequal to2.236milesper hour, escapevelocityisapproximately11,190(2.236) = 25,020
milesper hour.
1. Computetheareaof theregioninFigure1(A) enclosedbyy = 2x
2
andy = 2.
y
x
2 2 2
2
2
y
x
1
y = 2 x
2
y = 2 x
2
y = x
(A) (B)
y = 2
FIGURE 1
solution Thegraphsof y = 2 x
2
andy = 2intersect where2 x
2
= 2, or x = 2. Therefore, theenclosed
arealiesover theinterval [2, 2]. Theregionenclosedby thegraphsliesbelowy = 2 x
2
andabovey = 2, sothe
areais
_
2
2
_
(2x
2
) (2)
_
dx =
_
2
2
(4x
2
) dx =
_
4x
1
3
x
3
_
2
2
=
32
3
.
Computetheareaof theregioninFigure1(B) enclosedbyy = 2x
2
andy = x.
In Exercises 312, nd the area of the region enclosed by the graphs of the functions.
3. y = x
3
2x
2
+x, y = x
2
x
solution Theregionboundedby thegraphsof y = x
3
2x
2
+ x andy = x
2
x over theinterval [0, 2] isshown
below. For x [0, 1], thegraphof y = x
3
2x
2
+ x liesabovethegraphof y = x
2
x, whereas, for x [1, 2], the
graphof y = x
2
x liesabovethegraphof y = x
3
2x
2
+x. Theareaof theregionisthereforegivenby
_
1
0
_
(x
3
2x
2
+x) (x
2
x)
_
dx +
_
2
1
_
(x
2
x) (x
3
2x
2
+x)
_
dx
=
_
1
4
x
4
x
3
+x
2
_
1
0
+
_
x
3
x
2
1
4
x
4
_
2
1
=
1
4
1+1+(844)
_
11
1
4
_
=
1
2
.
x
y
y = x
3
2x
2
+ x
y = x
2
x
0.5
1
1.5
2
0.5 1 1.5 2
y = x
2
+2x, y = x
2
1, h(x) = x
2
+x 2
5. x = 4y, x = 248y, y = 0
solution Theregionboundedby thegraphsx = 4y, x = 24 8y andy = 0isshownbelow. For each0 y 2,
thegraphof x = 248y liestotheright of x = 4y. Theareaof theregionistherefore
A =
_
2
0
(248y 4y) dy =
_
2
0
(2412y) dy
= (24y 6y
2
)
2
0
= 24.
2.0
y
x
1.5
1.0
0.5
5 10 15 20 25
x = 4y
x = 24 8y
x = y
2
9, x = 152y
7. y = 4x
2
, y = 3x, y = 4
solution Theregionboundedby thegraphsof y = 4 x
2
, y = 3x andy = 4isshownbelow. For x [0, 1], the
graphof y = 4liesabovethegraphof y = 4 x
2
, whereas, for x [1,
4
3
], thegraphof y = 4liesabovethegraphof
y = 3x. Theareaof theregionisthereforegivenby
_
1
0
(4(4x
2
)) dx +
_
4/3
1
(43x) dx =
1
3
x
3
1
0
+
_
4x
3
2
x
2
_
4/3
1
=
1
3
+
_
16
3

8
3
_
_
4
3
2
_
=
1
2
.
y = 4
y = 4 x
2
y = 3x
y
x
4
2
3
1
0 0.8 1.2 0.4 0.6 1 0.2
x =
1
2
y, x = y
_
1y
2
, 0 y 1
9. y = sinx, y = cosx, 0 x
5
4
solution Theregionboundedbythegraphsof y = sinx andy = cosx over theinterval [0,
5
4
] isshownbelow. For
x [0,

4
], thegraphof y = cosx liesabovethegraphof y = sinx, whereas, for x [
4
,
5
4
], thegraphof y = sinx
liesabovethegraphof y = cosx. Theareaof theregionisthereforegivenby
_
/4
0
(cosx sinx) dx +
_
5/4
/4
(sinx cosx) dx
= (sinx +cosx)
/4
0
+ (cosx sinx)
5/4
/4
=
2
2
+
2
2
(0+1) +
_
2
2
+
2
2
_
2
2

2
2
_
= 3
21.
x
y = sin x
y = cos x
y
0.5
0.5
1
1
4 3 2 1
f (x) = sinx, g(x) = sin2x,

3
x
11. y = sec
2
_
x
4
_
, y = sec
2
_
x
8
_
, 0 x 1
solution Theregionboundedbythesecurvesfor 0 x 1isshownbelow. Asthegraphof sec
2
_
x
4
_
liesabove
thegraphof sec
2
_
x
8
_
, theareaof theregionisgivenby
_
1
0
_
sec
2
_
x
4
_
sec
2
_
x
8
__
dx =
_
4
tan
_
x
4
_
tan
_
x
8
_
_
1
0
=
4
tan
_
8
_
0.5
0.2 0.4 0.6 0.8 1.0
1.0
1.5
2.0
0 x
y
sec
(
x
)
2
1
4
sec
(
x
)
2
1
8
y =
x
_
x
2
+1
, y =
x
_
x
2
+4
, 1 x 1
13. Useagraphingutilitytolocatethepointsof intersectionof y = x
2
andy = cosx, andndtheareabetween
thetwocurves(approximately).
solution Usingacomputer algebrasystem, thepointsof intersectionarex = 0.8241323123. Theregionbounded
by thetwo curves between thesepoints is shown below. By symmetry, wecan computetheareafromx = 0 to x =
0.8241323123anddoubleit. Sincecosx liesabovex
2
inthisrange, theareaof theregionisgivenby
2
_
0.8241323123
0
(cosx x
2
) dx = 2
_
sinx
1
3
x
3
_
0.8241323123
0
1.094753609
0.8 0.4
0.2
0.4
0.6
0.8
1.0
0
0.4 0.8
x
y
cos(x)
x
2
Figure2showsasolidwhosehorizontal crosssectionat height y isacircleof radius(1+y)
2
for 0 y H.
Findthevolumeof thesolid.
15. Thebaseof asolidistheunit circlex
2
+y
2
= 1, anditscrosssectionsperpendicular tothex-axisarerectanglesof
height 4. Finditsvolume.
solution Becausethecrosssectionsarerectanglesof constant height 4, thegureisacylinder of radius1andheight
4. Thevolumeisthereforer
2
h = 4.
The base of a solid is the triangle bounded by the axes and the line 2x + 3y = 12, and its cross sections
perpendicular tothey-axishaveareaA(y) = (y +2). Finditsvolume.
17. Findthetotal massof arodof length1.2mwithlinear density(x) = (1+2x +
2
9
x
3
) kg/m.
solution Thetotal weight of therodis
_
1.2
0
(x) dx =
_
x +x
2
+
1
18
x
4
_
1.2
0
= 2.7552kg.
Findtheowrate(inthecorrectunits) throughapipeof diameter6cmif thevelocityof uidparticlesatadistance
r fromthecenter of thepipeisv(r) = (3r) cm/s.
In Exercises 1924, nd the average value of the function over the interval.
19. f (x) = x
3
2x +2, [1, 2]
solution Theaveragevalueis
1
2(1)
_
2
1
_
x
3
2x +2
_
dx =
1
3
_
1
4
x
4
x
2
+2x
_
2
1
=
1
3
_
(44+4)
_
1
4
12
__
=
9
4
.
f (x) = |x|, [4, 4]
21. f (x) = (x +1)(x
2
+2x +1)
4/5
, [0, 4]
solution Usethesubstitutionu = x
2
+ 2x + 1, sothat du = 2(x + 1) dx. Thenewboundsof integrationarefrom
u = 1tou = 25. Thentheaveragevalueis
1
40
_
4
0
(x +1)(x
2
+2x +1)
4/5
dx =
1
8
_
25
1
u
4/5
du =
5
72
u
9/5
25
1
=
625
72
5
3/5
5
72
f (x) = |x
2
1|, [0, 4]
23. f (x) =
_
9x
2
, [0, 3] Hint: Usegeometrytoevaluatetheintegral.
solution Theregionbelowthegraphof y =
_
9x
2
but abovethex-axisover theinterval [0, 3] isone-quarter of a
circleof radius3; consequently,
_
3
0
_
9x
2
dx =
1
4
(3)
2
=
9
4
.
Theaveragevalueisthen
1
30
_
3
0
_
9x
2
dx =
1
3
_
9
4
_
=
3
4
.
f (x) = x[x], [0, 3], where[x] isthegreatest integer function.
25. Find
_
5
2
g(t ) dt if theaveragevalueof g(t ) on[2, 5] is9.
solution Theaveragevalueof thefunctiong(t ) on[2, 5] isgivenby
1
52
_
5
2
g(t ) dt =
1
3
_
5
2
g(t ) dt.
Therefore,
_
5
2
g(t ) dt = 3(averagevalue) = 3(9) = 27.
Theaveragevalueof R(x) over [0, x] isequal tox for all x. UsetheFTC todetermineR(x).
27. UsetheWasher MethodtondthevolumeobtainedbyrotatingtheregioninFigure3about thex-axis.
y = x
2
y = mx
y
x
FIGURE 3
solution Settingx
2
= mx yieldsx(x m) = 0, sothetwocurvesintersect at (0, 0) and(m, m
2
). Tousethewasher
method, wemust slicethesolidperpendicular totheaxisof rotation; aswearerevolvingabout they-axis, thisimpliesa
horizontal sliceandintegrationiny. For eachy [0, m
2
], thecrosssectionisawasher withouter radiusR =

y and
inner radiusr =
y
m
. Thevolumeof thesolidisthereforegivenby
_
m
2
0
_
(
y)
2
_
y
m
_
2
_
dy =
_
1
2
y
2
y
3
3m
2
_
m
2
0
=
_
m
4
2

m
4
3
_
=

6
m
4
.
UsetheShell MethodtondthevolumeobtainedbyrotatingtheregioninFigure3about thex-axis.
In Exercises 2940, use any method to nd the volume of the solid obtained by rotating the region enclosed by the curves
about the given axis.
29. y = x
2
+2, y = x +4, x-axis
solution Letschoosetoslicetheregionboundedbythegraphsof y = x
2
+2andy = x +4(seethegurebelow)
vertically. Becauseavertical sliceisperpendicular totheaxisof rotation, wewill usethewasher methodtocalculatethe
volumeof thesolidof revolution. For eachx [1, 2], thewasher hasouter radiusx +4andinner radiusx
2
+2. The
volumeof thesolidisthereforegivenby
_
2
1
((x +4)
2
(x
2
+2)
2
) dx =
_
2
1
(x
4
3x
2
+8x +12) dx
=
_
1
5
x
5
x
3
+4x
2
+12x
_
2
1
=
_
128
5
+
34
5
_
=
162
5
.
y
x
1.00.5 0.5 1.0 1.5 2.0
1
3
4
5
6
y = x + 4
y = x
2
+ 2
y = x
2
+6, y = 8x 1, y-axis
31. x = y
2
3, x = 2y, axisy = 4
solution Lets choosetoslicetheregionboundedby thegraphs of x = y
2
3andx = 2y (seethegurebelow)
horizontally. Becauseahorizontal sliceis parallel to theaxis of rotation, wewill usetheshell methodto calculatethe
volumeof thesolidof revolution.Foreachy [1, 3],theshell hasradius4y andheight2y (y
2
3) = 3+2y y
2
.
Thevolumeof thesolidisthereforegivenby
2
_
3
1
(4y)(3+2y y
2
) dy = 2
_
3
1
(12+5y 6y
2
+y
3
) dy
= 2
_
12y +
5
2
y
2
2y
3
+
1
4
y
4
_
3
1
= 2
_
99
4
+
29
4
_
= 64.
y
x
1
2 3 1 2 3 4 5 6
1
2
3 x = y
2
3
x = 2y
y = 2x, y = 0, x = 8, axisx = 3
33. y = x
2
1, y = 2x 1, axisx = 2
solution The region bounded by the graphs of y = x
2
1 and y = 2x 1 is shown below. Lets choose to
slicetheregion vertically. Becausea vertical sliceis parallel to theaxis of rotation, wewill usetheshell method to
calculatethevolumeof thesolidof revolution. For eachx [0, 2], theshell has radius x (2) = x + 2andheight
(2x 1) (x
2
1) = 2x x
2
2
_
2
0
(x +2)(2x x
2
) dx = 2
_
2x
2
1
4
x
4
_
2
0
= 2(84) = 8.
y = x
2
1
x
y = 2x 1
y
2
1
3
1
2 1
y = x
2
1, y = 2x 1, axisy = 4
35. y = x
2
+4x 3, y = 0, axisy = 1
solution Theregionboundedby thegraphof y = x
2
+ 4x 3andthex-axis is shownbelow. Lets chooseto
slicetheregionvertically. Becauseavertical sliceisperpendicular totheaxisof rotation, wewill usethewasher method
to calculatethevolumeof thesolidof revolution. For eachx [1, 3], thecross sectionis awasher withouter radius
R = x
2
+4x 3(1) = x
2
+4x 2andinner radiusr = 0(1) = 1. Thevolumeof thesolidistherefore
givenby
_
3
1
_
(x
2
+4x 2)
2
1
_
dx =
_
1
5
x
5
2x
4
+
20
3
x
3
8x
2
+3x
_
3
1
=
__
243
5
162+18072+9
_
_
1
5
2+
20
3
8+3
__
=
56
15
.
y = x
2
+ 4x 3
y
x
0.8
1
0.4
0.6
0.2
0 2 3 1 1.5 2.5 0.5
y = x
2
+4x 3, y = 0, axisx = 4
37. x = 4y y
3
, x = 0, y 0, x-axis
solution Theregionboundedby thegraphs of x = 4y y
3
andx = 0for y 0is shownbelow. Lets chooseto
slicethisregionhorizontally. Becauseahorizontal sliceisparallel totheaxisof rotation, wewill usetheshell methodto
calculatethevolumeof thesolidof revolution. For eachy [0, 2], theshell hasradiusy andheight4y y
3
. Thevolume
of thesolidisthereforegivenby
2
_
2
0
y(4y y
3
) dy = 2
_
2
0
(4y
2
y
4
) dy
= 2
_
4
3
y
3
1
5
y
5
_
2
0
= 2
_
32
3

32
5
_
=
128
15
.
x = 4y y
3
0.5 1 1.5 2 2.5 3 0
x
0.5
1
1.5
2
y
y
2
= x
1
, x = 1, x = 3, axisy = 3
39. y = cos(x
2
), y = 0, 0 x
_
2
, y-axis
solution Theregionisshownbelow.
0.2
0.2 0.4 0.6 0.8 1.0 1.2
0.4
0.6
0.8
1.0
0 x
y
Wewill slicetheregionvertically andusetheshell method. For eachx
_
0,
_
2
_
, theshell has radius x andheight
cos(x
2
). Thevolumeof thesolidisthusgivenby
V = 2
_

/2
0
x cos(x
2
) dx
Usingthesubstitutionu = x
2
, wehavedu = 2x dx; thenewboundsof integrationareu = 0tou =

2
, sowehave
V =
1
2
2
_
/2
0
cosu du = sinu
/2
0
=
y = secx, y = cscx, y = 0, x = 0, x =

2
, x-axis
In Exercises 4144, nd the volume obtained by rotating the region about the given axis. The regions refer to the graph
of the hyperbola y
2
x
2
= 1in Figure 4.
x
y
c c
3
2
1
1
2
3
y = x
y
2
x
2
= 1
FIGURE 4
41. Theshadedregionbetweentheupper branchof thehyperbolaandthex-axisfor c x c, about thex-axis.
solution Lets choosetoslicetheregionvertically. Becauseavertical sliceis perpendicular totheaxis of rotation,
wewill usethewasher methodtocalculatethevolumeof thesolidof revolution. For eachx [c, c], crosssectionsare
circular diskswithradiusR =
_
1+x
2
_
c
c
(1+x
2
) dx =
_
x +
1
3
x
3
_
c
c
=
__
c +
c
3
3
_
_
c
c
3
3
__
= 2
_
c +
c
3
3
_
.
Theregionbetweentheupper branchof thehyperbolaandthex-axisfor 0 x c, about they-axis.
43. Theregionbetweentheupper branchof thehyperbolaandtheliney = x for 0 x c, about thex-axis.
solution Lets choosetoslicetheregionvertically. Becauseavertical sliceis perpendicular totheaxis of rotation,
wewill usethewasher methodtocalculatethevolumeof thesolidof revolution. For eachx [0, c], crosssectionsare
washerswithouter radiusR =
_
1+x
2
andinner radiusr = x. Thevolumeof thesolidisthereforegivenby
_
c
0
_
(1+x
2
) x
2
_
dx = x
c
0
= c.
Theregionbetweentheupper branchof thehyperbolaandy = 2, about they-axis.
45. Let R betheintersectionof thecircles of radius 1centeredat (1, 0) and(0, 1). Express as anintegral (but do not
evaluate): (a)theareaof R and(b)thevolumeof revolutionof R about thex-axis.
solution TheregionR isshownbelow.
x
2
+ (y 1)
2
= 1
(x 1)
2
+ y
2
= 1
y
x
0
0.2
0.4
0.6
0.8
1
0.2 0.4 0.6 0.8 1
(a) Avertical sliceof R hasitstopalongtheupper leftarcof thecircle(x 1)
2
+y
2
= 1anditsbottomalongthelower
right arcof thecirclex
2
+(y 1)
2
= 1. Theareaof R isthereforegivenby
_
1
0
__
1(x 1)
2
(1
_
1x
2
)
_
dx.
(b) If we revolve R about the x-axis and use the washer method, each cross section is a washer with outer radius
_
1(x 1)
2
andinner radius1
_
1x
2
_
1
0
_
(1(x 1)
2
) (1
_
1x
2
)
2
_
dx.
Let a > 0. Showthat thevolumeobtainedwhentheregionbetweeny = a
_
x ax
2
andthex-axis is rotated
about thex-axisisindependent of theconstant a.
47. If 12J of work areneededtostretchaspring20cmbeyondequilibrium, howmuchwork isrequiredtocompressit
6cmbeyondequilibrium?
solution First, wedeterminethevalueof thespringconstant k asfollows:
1
2
k(0.2)
2
= 12 so k = 600N/m.
Now, theworkneededtocompressthespring6cmbeyondequilibriumis
W =
_
0.06
0
600x dx = 300x
2
0.06
0
= 1.08J .
A springwhoseequilibriumlengthis15cmexertsaforceof 50N whenit isstretchedto20cm. Findthework
requiredtostretchthespringfrom22to24cm.
49. If 18ft-lbof work areneededtostretchaspring1.5ft beyondequilibrium, howfar will thespringstretchif a12-lb
weight isattachedtoitsend?
solution First, wedeterminethevalueof thespringconstant asfollows:
1
2
k(1.5)
2
= 18 so k = 16lb/ft.
Now, if a12-lbweight isattachedtotheendof thespring, balancingtheforcesactingontheweight, wehave12= 16d,
whichimpliesd = 0.75ft. A 12-lbweight will thereforestretchthespring9inches.
LetW bethework(againstthesunsgravitational force) requiredtotransportan80-kgpersonfromEarthtoMars
whenthetwoplanetsarealignedwiththesunat their minimal distanceof 55.710
6
km. UseNewtonsUniversal
Lawof Gravity (seeExercises3537inSection6.5) toexpressW asanintegral andevaluateit. Thesunhasmass
M
s
= 1.9910
30
kg, andthedistancefromthesuntotheearthis149.610
6
km.
In Exercises 51 and 52, water is pumped into a spherical tank of radius 2mfrom a source located 1mbelow a hole at
the bottom (Figure 5). The density of water is 1000kg/m
3
.
2
1
Water source
FIGURE 5
51. Calculatetheworkrequiredtoll thetank.
solution Placetheorigin at thebaseof thespherewith thepositivey-axis pointing upward. Theequation for the
great circleof thesphereis then x
2
+ (y 2)
2
= 4. At location y, thehorizontal cross section is acircleof radius
_
4(y 2)
2
=
_
4y y
2
; thevolumeof thelayer isthen(4y y
2
)y m
3
, andtheforceneededtolift thelayer is
1000(9.8)(4y y
2
)y N. Thelayer of water must beliftedy +1meters, sotheworkrequiredtoll thetankisgiven
by
9800
_
4
0
(y +1)(4y y
2
) dy = 9800
_
4
0
(3y
2
+4y y
3
) dy
= 9800
_
y
3
+2y
2
1
4
y
4
_
4
0
= 313,600 985,203.5J .
CalculatetheworkF(h) requiredtoll thetanktolevel h metersinthesphere.
53. A tank of mass 20kgcontaining100kgof water (density 1000kg/m
3
) is raisedvertically at aconstant speedof
100m/minfor oneminute, duringwhichtimeit leakswater at arateof 40kg/min. Calculatethetotal workperformedin
raisingthecontainer.
solution Lett denotetheelapsedtimeinminutesandlety denotetheheightof thecontainer. Giventhatthespeedof
ascent is100m/min, y = 100t ; moreover, themassof water inthecontainer is
10040t = 1000.4ykg.
Theforceneededtolift thecontainer anditscontentsisthen
9.8(20+(1000.4y)) = 11763.92y N,
andtheworkrequiredtolift thecontainer anditscontentsis
_
100
0
(11763.92y) dy = (1176y 1.96y
2
)
100
0
= 98,000J .
June 9, 2011 LTSV SSM Rough
7 EXPONENTIAL FUNCTIONS
7.1 Derivative of f(x) = b
x
and the Number e
1. Whichof thefollowingequationsisincorrect?
(a) 3
2
3
5
= 3
7
(b) (
5)
4/3
= 5
2/3
(c) 3
2
2
3
= 1 (d) (2
2
)
2
= 16
solution
(a) Thisequationiscorrect: 3
2
3
5
= 3
2+5
= 3
7
.
(b) Thisequationiscorrect: (
5)
4/3
= (5
1/2
)
4/3
= 5
(1/2)(4/3)
= 5
2/3
.
(c) Thisequationisincorrect: 3
2
2
3
= 9 8= 72= 1.
(d) thisequationiscorrect: (2
2
)
2
= 2
(2)(2)
= 2
4
= 16.
2. What arethedomainandrangeof lnx?Whenislnx negative?
solution Thedomainof lnx isx > 0andtherangeisall real numbers. lnx isnegativefor 0< x < 1.
3. Towhichof thefollowingdoesthePower Ruleapply?
(a) f (x) = x
2
(b) f (x) = 2
e
(c) f (x) = x
e
(d) f (x) = e
x
(e) f (x) = x
x
(f) f (x) = x
4/5
solution ThePowerRuleapplieswhenthefunctionhasavariablebaseandaconstantexponent. Therefore, thePower
Ruleappliesto(a)x
2
,(c)x
e
and(f)x
4/5
.
4. For whichvaluesof b doesb
x
haveanegativederivative?
solution Thefunctionb
x
hasanegativederivativewhen0 0except b = 1.
6. Whichpoint liesonthegraphof y = b
x
for all b?
solution Thepoint (0, 1) liesonthegraphof y = b
x
for all b.
7. Whichof thefollowingstatementsisnot true?
(a) (e
x
)
= e
x
(b) lim
h0
e
h
1
h
= 1
(c) Thetangent linetoy = e
x
at x = 0hasslopee.
(d) Thetangent linetoy = e
x
at x = 0hasslope1.
solution
(a) Thisstatement istrue: (e
x
)
= e
x
.
(b) Thisstatement istrue:
lim
h0
e
h
1
h
=
d
dx
e
x
x=0
= e
0
= 1.
(c) Thisstatement isfalse: thetangent linetoy = e
x
at x = 0hasslopee
0
= 1.
(d) Thisstatement istrue: thetangent linetoy = e
x
at x = 0hasslopee
0
= 1.
370
S E C T I ON 7.1 Derivative of f(x) = b
x
Exercises
1. Rewriteasawholenumber (without usingacalculator):
(a) 7
0
(b) 10
2
(2
2
+5
2
)
(c)
(4
3
)
5
(4
5
)
3
(d) 27
4/3
(e) 8
1/3
8
5/3
(f) 3 4
1/4
12 2
3/2
solution
(a) 7
0
= 1.
(b) 10
2
(2
2
+5
2
) = 100(1/4+1/25) = 25+4= 29.
(c) (4
3
)
5
/(4
5
)
3
= 4
15
/4
15
= 1.
(d) (27)
4/3
= (27
1/3
)
4
= 3
4
= 81.
(e) 8
1/3
8
5/3
= (8
1/3
)
5
/8
1/3
= 2
5
/2= 2
4
= 16.
(f) 3 4
1/4
12 2
3/2
= 3 2
1/2
3 2
2
2
3/2
= 0.
Compute(16
1/16
)
4
.
In Exercises 310, solve for the unknown variable.
3. 9
2x
= 9
8
solution If 9
2x
= 9
8
, then2x = 8, andx = 4.
e
t
2
= e
4t 3
5. 3
x
=
_
1
3
_
x+1
solution Rewrite(
1
3
)
x+1
as(3
1
)
x+1
= 3
x1
. Then3
x
= 3
x1
, whichrequiresx = x 1. Thus, x = 1/2.
(
5)
x
= 125
7. 4
x
= 2
x+1
solution Rewrite4
x
as (2
2
)
x
= 2
2x
. Then2
2x
= 2
x+1
, whichrequires 2x = x + 1. Solvingfor x gives
x = 1/3.
b
4
= 10
12
9. k
3/2
= 27
solution Raisebothsidesof theequationtothetwo-thirdspower. Thisgivesk = (27)
2/3
= (27
1/3
)
2
= 3
2
= 9.
(b
2
)
x+1
= b
6
In Exercises 1114, determine the limit.
11. lim
x
4
x
solution lim
x
4
x
= .
lim
x
4
x
13. lim
x
_
1
4
_
x
solution lim
x
_
1
4
_
x
= lim
x
4
x
= .
lim
x
e
xx
2 In Exercises 1518, nd the equation of the tangent line at the point indicated.
15. y = 4e
x
, x
0
= 0
solution Let f (x) = 4e
x
. Thenf
(x) = 4e
x
andf
(0) = 4. At x
0
= 0, f (0) = 4, sotheequationof thetangent
lineisy = 4(x 0) +4= 4x +4.
y = e
4x
, x
0
= 0
17. y = e
x+2
, x
0
= 1
solution Let f (x) = e
x+2
. Thenf
(x) = e
x+2
andf
(1) = e
1
. At x
0
= 1, f (1) = e, sotheequationof the
tangent lineisy = e(x +1) +e = ex +2e.
y = e
x
2
, x
0
= 1
19. f (x) = 7e
2x
+3e
4x
solution
d
dx
(7e
2x
+3e
4x
) = 14e
2x
+12e
4x
.
f (x) = e
5x
21. f (x) = e
x
solution
d
dx
e
x
= e
x
.
f (x) = e
3
23. f (x) = e
4x+9
solution
d
dx
e
4x+9
= 4e
4x+9
.
f (x) = 4e
x
+7e
2x
372 C HA P T E R 7 EXPONENTIAL FUNCTIONS
25. f (x) =
e
x
2
x
solution
d
dx
_
e
x
2
x
_
=
x
_
e
x
2
(2x)
_
e
x
2
(1)
x
2
=
2x
2
e
x
2
e
x
2
x
2
.
f (x) = x
2
e
2x
27. f (x) = (1+e
x
)
4
solution
d
dx
(1+e
x
)
4
= 4(1+e
x
)
3
e
x
.
f (x) = (2e
3x
+2e
2x
)
4
29. f (x) = e
x
2
+2x3
solution
d
dx
e
x
2
+2x3
= (2x +2)e
x
2
+2x3
.
f (x) = e
1/x
31. f (x) = e
sinx
solution
d
dx
e
sinx
= cosxe
sinx
.
f (x) = e
(x
2
+2x+3)
2
33. f () = sin(e
)
solution
d
d
sin(e
) = e
cos(e
).
f (t ) = e
t 35. f (t ) =
1
1e
3t
solution
d
dt
_
1
1e
3t
_
=
d
dt
(1e
3t
)
1
= (1e
3t
)
2
(3e
3t
).
f (t ) = cos(t e
2t
)
37. f (x) =
e
x
3x +1
solution
d
dx
_
e
x
3x +1
_
=
(3x +1) e
x
3e
x
(3x +1)
2
=
3xe
x
2e
x
(3x +1)
2
.
f (x) = tan(e
56x
) 39. f (x) =
e
x+1
+x
2e
x
1
solution
d
dx
_
e
x+1
+x
2e
x
1
_
=
(2e
x
1)(e
x+1
+1) (e
x+1
+x)(2e
x
)
(2e
x
1)
2
=
2e
2x+1
+2e
x
e
x+1
12e
2x+1
2xe
x
(2e
x
1)
2
=
2e
x
e
x+1
2xe
x
1
(2e
x
1)
2
.
f (x) = e
e
x In Exercises 4146, calculate the derivative indicated.
41. f
(x); f (x) = e
4x3
4x3
. Thenf
(x) = 4e
4x3
andf
(x) = 16e
4x3
.
f
(x); f (x) = e
123x
43.
d
2
y
dt
2
; y = e
t
sint
solution Let y = e
t
sint . Then
dy
dt
= e
t
cost +e
t
sint = e
t
(cost +sint ),
and
d
2
y
dt
2
= e
t
(sint +cost ) +e
t
(cost +sint ) = 2e
t
cost.
d
2
y
dt
2
; y = e
2t
sin3t
45.
d
2
dt
2
e
t t
2
solution
d
dt
e
t t
2
= (12t )e
t t
2
and
d
2
dt
2
e
t t
2
= (12t )
2
e
t t
2
2e
t t
2
= (4t
2
4t 1)e
t t
2
.
x
d
3
d
3
cos(e
)
In Exercises 4752, nd the critical points and determine whether they are local minima, maxima, or neither.
47. f (x) = e
x
x
solution Settingf
(x) = e
x
1equal tozeroandsolvingfor x givese
x
= 1, whichistrueif andonly if x = 0.
f
(x) = e
x
, sof
(0) = e
0
= 1> 0. Therefore, x = 0correspondstoalocal minimum.
f (x) = x +e
x
49. f (x) =
e
x
x
for x > 0
solution Setting
f
(x) =
e
x
x e
x
x
2
equal tozeroandsolvingfor x givese
x
(x 1) = 0whichistrueif andonlyif x = 1. Now,
f
(x) =
(e
x
(x 1) +e
x
)x
2
2x(e
x
(x 1))
x
4
,
so
f
(1) =
(e(0) +e) 2(e(0))
1
= e > 0.
Therefore, x = 1correspondstoalocal minimum.
f (x) = x
2
e
x
51. g(t ) =
e
t
t
2
+1
solution
g
(t ) =
(t
2
+1)e
t
e
t
(2t )
(t
2
+1)
2
.
Theonlycritical point iswhene
t
(t
2
2t +1) = e
t
(t 1)
2
= 0. Thus, thecritical point ist = 1. Noticethat g
(t ) > 0
for all t = 1. Thus, g
(t ) doesnot changesignandthereforet = 1correspondstoneither amaximumnor aminimum.

g(t ) = (t
3
2t )e
t
In Exercises 5358, nd the critical points and points of inection. Then sketch the graph.
53. y = xe
x
solution Let f (x) = xe
x
. Then
f
(x) = e
x
xe
x
= (1x)e
x
,
sox = 1isacritical point. Further, f
(x) > 0for x < 1andf
(x) < 0for x > 1. Hence, f (x) isincreasingfor x < 1

anddecreasingfor x > 1. With
f
(x) = e
x
e
x
+xe
x
= (x 2)e
x
,
it followsthat f (x) isconcavedownfor x < 2, isconcaveupfor x > 2andhasapoint of inectionat x = 2. A graph
of y = f (x) isshownbelow.
2
1
1
1
y
x
4 3 2 1
y = e
x
+e
x
55. y = e
x
cosx on
_

2
,

2
_
x
cosx. Then
f
(x) = e
x
(sinx) +(e
x
) cosx = e
x
(sinx +cosx).
Ontheinterval [
2
,

2
], thereis only onecritical point, at x =
4
, andf (x) is increasingfor
2
< x <
4
and
decreasingfor
4
< x <

2
. Now,
f
(x) = e
x
(cosx sinx) +e
x
(sinx +cosx) = 2e
x
sinx,
sof (x) isconcavedownfor
2
< x < 0, isconcaveupfor 0< x <

2
, andhasaninectionpoint at x = 0. A graph
of y = f (x) isshownbelow.
1 0.5 1.5 1.5 1 0.5
x
0.4
0.8
1.6
1.2
y
y = e
x
2
57. y = e
x
x
x
x. Thenf
(x) = e
x
1, andx = 0istheonlycritical point. Further, f (x) isdecreasing
for x < 0andincreasingfor x > 0. Observethat f
(x) = e
x
> 0for all x, sof (x) isconcaveupfor all x, andthereare
nopointsof inection. A graphof y = f (x) isshownbelow.
5
2
1
4
3
y
x
1 2 1 2
y = x
2
e
x
on[0, 10]
59. Finda > 0suchthat thetangent linetothegraphof f (x) = x
2
e
x
at x = a passesthroughtheorigin(Figure6).
y
x
a
f (x) = x
2
e
x
FIGURE 6
2
e
x
. Thenf (a) = a
2
e
a
,
f
(x) = x
2
e
x
+2xe
x
= e
x
(2x x
2
),
f
(a) = (2a a
2
)e
a
, andtheequationof thetangent linetof at x = a is
y = f
(a)(x a) +f (a) = (2a a

2
)e
a
(x a) +a
2
e
a
.
For thislinetopassthroughtheorigin, wemust have
0= (2a a
2
)e
a
(a) +a
2
e
a
= e
a
_
a
2
2a
2
+a
3
_
= a
2
e
a
(a 1).
Thus, a = 0or a = 1. Theonly valuea > 0suchthat thetangent lineto f (x) = x
2
e
x
passes throughtheoriginis
thereforea = 1.
UseNewtonsMethodtondthetwosolutionsof e
x
= 5x tothreedecimal places(Figure7).
61. Computethelinearizationof f (x) = e
2x
sinx at a = 0.
2x
sinx. Then
f
(x) = e
2x
cosx 2e
2x
sinx, f (a) = 0, andf
(a) = 1,
L(x) = f
(a)(x a) +f (a) = x.
Computethelinearizationof f (x) = xe
63x
at a = 2.
63. Findthelinearizationof f (x) = e
x
at a = 0anduseit toestimatee
0.1
.
x
. Then
f
(x) = e
x
, f (a) = 1, andf
(a) = 1,
x
L(x) = f
(a)(x a) +f (a) = (x 0) +1= x +1.

Usingthelinearization, wend
e
0.1
= f (0.1) L(0.1) = 0.1+1= 0.9.
Usethelinear approximationtoestimatef (1.03) f (1) wherey = x
1/3
e
x1
.
65. A 2005studybytheFisheriesResearchServicesinAberdeenScotlandshowedthattheaveragelengthof thespecies
ClupeaHarengus(Atlantic herring) asafunctionof aget (inyears) canbemodeledby L(t ) = 32(1 e
0.37t
) cmfor
0 t 13.
(a) Howfast istheaveragelengthchangingat aget = 6yrs?
(b) At what ageistheaveragelengthchangingat arateof 5cm/yr?
(c) CalculateL = lim
t
L(t ).
solution GivenL(t ) = 32(1e
0.37t
), it followsthat L
(t ) = 11.84e
0.37t
cm/year.
(a) At aget = 6years,
L
(6) = 11.84e
0.37(6)
= 1.28cm/year.
(b) SettingL
(t ) = 5andsolvingfor t yields
t =
1
0.37
ln
5
11.84
2.33years.
(c) L = lim
t
(3232e
0.37t
) = 3232 lim
t
e
0.37t
= 320= 32cm.
Accordingtoa1999studybyStarkeyandScarnecchia, theaverageweight(kg) ataget (years) of channel catsh
intheLower YellowstoneRiver canbemodeledby
W(t ) =
_
3.462933.32173e
0.03456t
_
3.4026
Findtherateat whichweight ischangingat aget = 10.
67. ThefunctionsinExercises65and66areexamplesof theVonBertalanffygrowthfunction
M(t ) =
_
a +(b a)e
kmt
_
1/m
introducedinthe1930sbyAustrian-bornbiologistKarl LudwigVonBertalanffy.CalculateM
(0) intermsof theconstants

a, b, k, andm.
solution GivenM(t ) =
_
a +(b a)e
kmt
_
1/m
, wend
M
(t ) =
1
m
_
a +(b a)e
kmt
_
1/m1
(km(b a)e
kmt
) = k(b a)e
kmt
_
a +(b a)e
kmt
_
1/m1
andM
(0) = kb
1/m1
(b a).
Findanapproximationtom(4) usingthelimit denitionandestimatetheslopeof thetangent linetoy = 4
x
at
x = 0andx = 2.
In Exercises 6986, evaluate the integral.
69.
_
(e
x
+2) dx
solution
_
(e
x
+2) dx = e
x
+2x +C.
_
e
4x
dx
71.
_
1
0
e
3x
dx
solution
_
1
0
e
3x
dx =
1
3
e
3x
1
0
=
1
3
e
3
+
1
3
=
1
3
(1e
3
).
_
6
2
e
x/2
dx
73.
_
3
0
e
16t
dt
solution
_
3
0
e
16t
dt =
1
6
e
16t
3
0
=
1
6
e
17
+
1
6
e =
1
6
(e e
17
).
_
3
2
e
4t 3
dt
75.
_
(e
4x
+1) dx
solution Usethesubstitutionu = 4x, du = 4dx. Then
_
(e
4x
+1) dx =
1
4
_
(e
u
+1) du =
1
4
(e
u
+u) +C =
1
4
e
4x
+x +C.
_
(e
x
+e
x
) dx
77.
_
1
0
xe
x
2
/2
dx
solution Let u = x
2
/2. Thendu = x dx and
_
1
0
xe
x
2
/2
dx =
_
1/2
0
e
u
du = e
u
1/2
0
= e
1/2
+1.
_
2
0
ye
3y
2
dy
79.
_
e
t
_
e
t
+1dt
solution Usethesubstitutionu = e
t
+1, du = e
t
dt . Then
_
e
t
_
e
t
+1dt =
_

u du =
2
3
u
3/2
+C =
2
3
(e
t
+1)
3/2
+C.
_
(e
x
4x) dx
81.
_
e
2x
e
4x
e
x
dx
solution
_
_
e
2x
e
4x
e
x
_
dx =
_
(e
x
e
3x
) dx = e
x
e
3x
3
+C.
_
e
x
cos(e
x
) dx
83.
_
e
x
e
x
+1
dx
x
+1, du = e
x
dx. Then
_
e
x
e
x
+1
dx =
_
du
u
= 2
u +C = 2
e
x
+1+C.
_
e
x
(e
2x
+1)
3
dx
85.
_
e
x
dx
x
solution Let u =

x. Then
du =
1
2
x
dx
and
_
e
x
dx
x
= 2
_
e
u
du = 2e
u
+C = 2e
x
+C.
_
x
2/3
e
x
1/3
dx
87. Findtheareabetweeny = e
x
andy = e
2x
over [0, 1].
solution Over [0, 1], thegraphof y = e
2x
liesabovethegraphof y = e
x
. Hence, theareabetweenthegraphsis
_
1
0
(e
2x
e
x
) dx =
_
e
2x
2
e
x
_
1
0
=
e
2
2
e
_
1
2
1
_
=
e
2
2
e +
1
2
.
Findtheareabetweeny = e
x
andy = e
x
over [0, 2].
89. Findtheareaboundedbyy = e
2
, y = e
x
, andx = 0.
solution Thegraphsof y = e
2
andy = e
x
intersectatx = 2. Over theinterval [0, 2], thegraphof y = e
2
liesabove
thegraphof y = e
x
, sotheareabetweenthegraphsis
_
2
0
(e
2
e
x
) dx = (xe
2
e
x
)
2
0
= 2e
2
e
2
(01) = e
2
+1.
Findthevolumeobtainedbyrevolvingy = e
x
about thex-axisfor 0 x 1.
91. Windengineershavefoundthatwindspeedv (inm/s) atagivenlocationfollowsaRayleighdistributionof thetype
W(v) =
1
32
ve
v
2
/64
Thismeansthat theprobabilitythat v liesbetweena andb isequal totheshadedareainFigure8.
(a) Showthat theprobabilitythat v [0, b] is1e
b
2
/64
.
(b) Calculatetheprobabilitythat v [2, 5].
20
0.05
0.1
a b
y = W(v)
v (m/s)
y
FIGURE 8 Theshadedareaistheprobabilitythat v liesbetweena andb.
x
solution
(a) Theprobabilitythat v [0, b] is
_
b
0
1
32
ve
v
2
/64
dv.
Let u = v
2
/64. Thendu =
v
32
dv and
_
b
0
1
32
ve
v
2
/64
dv =
_
b
2
/64
0
e
u
du = 1e
b
2
/64
.
(b) Theprobabilitythat v [2, 5] istheprobabilitythat v [0, 5] minustheprobabilitythat v [0, 2]. Usingpart (a),
it followsthat theprobabilitythat v [2, 5] is
(1e
25/64
) (1e
4/64
) = e
1/16
e
25/64
0.263.
Thefunctionf (x) = e
x
satisesf
(x) = f (x). Showthat if g(x) isanother functionsatisfyingg
(x) = g(x),
theng(x) = Ce
x
for someconstant C. Hint: Computethederivativeof g(x)e
x
.
93. Provethat f (x) = e
x
isnot apolynomial function. Hint: Differentiationlowersthedegreeof apolynomial by1.
solution Assumef (x) = e
x
is apolynomial function of degreen. Then f
(n+1)
(x) = 0. But weknow that any
derivativeof e
x
ise
x
ande
x
= 0. Hence, e
x
cannot beapolynomial function.
Recall thefollowingpropertyof integrals: If f (t ) g(t ) for all t 0, thenfor all x 0,
_
x
0
f (t ) dt
_
x
0
g(t ) dt
Theinequality e
t
1holdsfor t 0becausee > 1. UseEq. (4) toprovethat e
x
1+ x for x 0. Thenprove,
bysuccessiveintegration, thefollowinginequalities(for x 0):
e
x
1+x +
1
2
x
2
, e
x
1+x +
1
2
x
2
+
1
6
x
3
95. GeneralizeExercise94; that is, useinduction (if you arefamiliar with this method of proof) to provethat for all
n 0,
e
x
1+x +
1
2
x
2
+
1
6
x
3
+ +
1
n!
x
n
(x 0)
solution For n = 1, e
x
1+ x by Exercise94. Assumethestatement is truefor n = k. Weneed to provethe
statement istruefor n = k +1. BytheInductionHypothesis,
e
x
1+x +x
2
/2+ +x
k
/k!.
Integratingbothsidesof thisinequalityyields
_
x
0
e
t
dt = e
x
1 x +x
2
/2+ +x
k+1
/(k +1)!
or
e
x
1+x +x
2
/2+ +x
k+1
/(k +1)!
asrequired.
UseExercise94to showthat
e
x
x
2

x
6
andconcludethat lim
x
e
x
x
2
= . ThenuseExercise95to provemore
generallythat lim
x
e
x
x
n
= for all n.
97. Calculatetherst threederivativesof f (x) = xe
x
. Thenguesstheformulafor f
(n)
(x) (useinductiontoproveit if
youarefamiliar withthismethodof proof).
solution f
(x) = e
x
+ xe
x
, f
(x) = e
x
+ e
x
+ xe
x
= 2e
x
+ xe
x
, f
(x) = 2e
x
+ e
x
+ xe
x
= 3e
x
+ xe
x
. So
onewouldguessthatf
(n)
(x) = ne
x
+xe
x
. Assumingthisistruefor f
(n)
(x), weverifythatf
(n+1)
(x) = (f
(n)
(x))
=
ne
x
+e
x
+xe
x
= (n +1)e
x
+xe
x
.
Consider theequatione
x
= x, where isaconstant.
(a) For which doesit haveauniquesolution? For intuition, drawagraphof y = e
x
andtheliney = x.
(b) For which doesit haveat least onesolution?
99. Proveintwowaysthat thenumbersm(a) satisfy
m(ab) = m(a) +m(b)
(a) First method: Usethelimit denitionof m
b
and
(ab)
h
1
h
= b
h
_
a
h
1
h
_
+
b
h
1
h
(b) Secondmethod: ApplytheProduct Ruletoa
x
b
x
= (ab)
x
.
solution
(a) m(ab) = lim
h0
(ab)
h
1
h
= lim
h0
b
h
(a
h
1)
h
+
b
h
1
h
= lim
h0
b
h
lim
h0
a
h
1
h
+ lim
h0
b
h
1
h
= lim
h0
a
h
1
h
+ lim
h0
b
h
1
h
= m(a) +m(b).
So, m(ab) = m(a) +m(b).
(b) m(ab)(ab)
x
= ((ab)
x
)
= (a
x
b
x
)
= (a
x
)
b
x
+(b
x
)
a
x
= m(a)a
x
b
x
+m(b)a
x
b
x
= a
x
b
x
(m(a) +m(b)) = (ab)
x
(m(a) +m(b)).
Therefore, wehavem(ab)(ab)
x
= (ab)
x
(m(a) + m(b)). Dividingbothsides by (ab)
x
, weseethat m(ab) = m(a) +
m(b).
7.2 Inverse Functions
1. Whichof thefollowingsatisfyf
1
(x) = f (x)?
(a) f (x) = x (b) f (x) = 1x
(c) f (x) = 1 (d) f (x) =

x
(e) f (x) = |x| (f) f (x) = x
1
solution Thefunctions(a)f (x) = x, (b)f (x) = 1x and(f)f (x) = x
1
satisfyf
1
(x) = f (x).
2. Thegraphof afunctionlookslikethetrackof aroller coaster. Isthefunctionone-to-one?
solution Becausethegraphlookslikethetrack of aroller coaster, therewill beseveral locationsat whichthegraph
hasthesameheight. Thegraphwill thereforefail thehorizontal linetest, meaningthat thefunctionisnot one-to-one.
3. Thefunctionf mapsteenagersintheUnitedStatestotheir last names. Explainwhytheinversefunctionf
1
does
not exist.
solution Manydifferentteenagerswill havethesamelastname, sothisfunctionwill notbeone-to-one. Consequently,
thefunctiondoesnot haveaninverse.
4. Thefollowingfragment of atrainschedulefor theNewJ ersey Transit Systemdenes afunctionf fromtowns to
times. Isf one-to-one?What isf
1
(6:27)?
Trenton 6:21
HamiltonTownship 6:27
PrincetonJ unction 6:34
NewBrunswick 6:38
solution Thisfunctionisone-to-one, andf
1
(6:27) = HamiltonTownship.
5. Ahomeworkproblemasksfor asketchof thegraphof theinverse of f (x) = x +cosx. Frank, after tryingbutfailing
tondaformulafor f
1
(x), saysitsimpossibletographtheinverse. Biancahandsinanaccuratesketchwithoutsolving
for f
1
. HowdidBiancacompletetheproblem?
solution Thegraphof theinversefunctionisthereectionof thegraphof y = f (x) throughtheliney = x.
6. What istheslopeof thelineobtainedbyreectingtheliney =
x
2
throughtheliney = x?
solution Thelineobtainedbyreectingtheliney = x/2throughtheliney = x hasslope2.
7. SupposethatP = (2, 4) liesonthegraphof f (x) andthattheslopeof thetangentlinethroughP ism = 3.Assuming
that f
1
(x) exists, what istheslopeof thetangent linetothegraphof f
1
(x) at thepoint Q = (4, 2)?
solution Thetangent linetothegraphof f
1
(x) at thepoint Q = (4, 2) hasslope
1
3
.
Exercises
1. Showthat f (x) = 7x 4isinvertibleandnditsinverse.
solution Solvingy = 7x 4for x yieldsx =
y +4
7
. Thus, f
1
(x) =
x +4
7
.
Isf (x) = x
2
+2one-to-one? If not, describeadomainonwhichit isone-to-one.
3. What isthelargest interval containingzeroonwhichf (x) = sinx isone-to-one?
solution Lookingat thegraphof sinx, thefunctionisone-to-oneontheinterval [/2, /2].
Showthat f (x) =
x 2
x +3
isinvertibleandnditsinverse.
(a) What isthedomainof f (x)?Therangeof f
1
(x)?
(b) What isthedomainof f
1
(x)?Therangeof f (x)?
5. Verifythat f (x) = x
3
+3andg(x) = (x 3)
1/3
areinversesbyshowingthat f (g(x)) = x andg(f (x)) = x.
solution
f (g(x)) =
_
(x 3)
1/3
_
3
+3= x 3+3= x.
g(f (x)) =
_
x
3
+33
_
1/3
=
_
x
3
_
1/3
= x.
S E C T I ON 7.2 Inverse Functions 379
Repeat Exercise5for f (t ) =
t +1
t 1
andg(t ) =
t +1
t 1
.
7. Theescapevelocity fromaplanet of radius R is v(R) =
_
2GM
R
, whereG is theuniversal gravitational constant
andM isthemass. Findtheinverseof v(R) expressingR intermsof v.
solution Tondtheinverse, wesolve
y =
_
2GM
R
for R. Thisyields
R =
2GM
y
2
.
Therefore,
v
1
(R) =
2GM
R
2
.
In Exercises 815, nd a domain on which f is one-to-one and a formula for the inverse of f restricted to this domain.
Sketch the graphs of f and f
1
.
f (x) = 3x 2
9. f (x) = 4x
solution The linear function f (x) = 4 x is one-to-one for all real numbers. Solving y = x 4 for x gives
x = 4y. Thus, f
1
(x) = 4x.
1 2 3 4
x
2
1
3
4
y
f(x) = f
1
(x) = 4 x
f (x) =
1
x +1
11. f (x) =
1
7x 3
solution Thegraphof f (x) = 1/(7x 3) givenbelowshowsthat f passesthehorizontal linetest, andistherefore
one-to-one, onitsentiredomain{x : x =
3
7
}. Solvingy = 1/(7x 3) for x gives
x =
1
7y
+
3
7
; thus, f
1
(x) =
1
7x
+
3
7
.
4
2
4
2
2 4 2 4
x
y
4
2
4
2
2 4 2 4
x
y
y = f(x) y = f
1
(x)
f (s) =
1
s
2
13. f (x) =
1
_
x
2
+1
solution To makethefunctionf (x) =
1
_
x
2
+1
one-to-one, wemust restrict thedomainto either {x : x 0} or
{x : x 0}. If wechoosethedomain{x : x 0}, thensolvingy =
1
_
x
2
+1
for x yields
x =
_
1y
2
y
; hence, f
1
(x) =
_
1x
2
x
.
Hadwechosenthedomain{x : x 0}, theinversewouldhavebeen
f
1
(x) =
_
1x
2
x
.
x
1 2 1 2
1
0.5
1.5
y
y = f
1
(x)
y = f(x)
f (z) = z
3
15. f (x) =
_
x
3
+9
solution The graph of f (x) =
_
x
3
+9 given below shows that f passes the horizontal line test, and therefore
is one-to-one, on its entire domain {x : x 9
1/3
}. Solving y =
_
x
3
+9 for x yields x = (y
2
9)
1/3
. Thus,
f
1
(x) = (x
2
9)
1/3
.
2
2
4
2
6
8
y
x
8 4 2 6
y = f
1
(x)
y = f(x)
ForeachfunctionshowninFigure15, sketchthegraphof theinverse(restrictthefunctionsdomainif necessary).
17. Whichof thegraphsinFigure16isthegraphof afunctionsatisfyingf
1
= f ?
(A)
x
(B)
y
x
(D)
y
(C)
x
y
x
y
FIGURE 16
solution Figures (B) and(C) wouldnot changewhenreectedaroundtheliney = x. Therefore, thesetwosatisfy
f
1
= f .
Let n beanonzerointeger. Findadomainonwhichf (x) = (1 x
n
)
1/n
coincides withits inverse. Hint: The
answer dependsonwhether n isevenor odd.
19. Let f (x) = x
7
+x +1.
(a) Showthat f
1
exists(but donot attempt tondit). Hint: Showthat f isincreasing.
(b) What isthedomainof f
1
?
(c) Findf
1
(3).
solution
(a) Thegraphof f (x) = x
7
+x +1isshownbelow. Fromthisgraph, weseethat f (x) isastrictlyincreasingfunction;
byExample3, it isthereforeone-to-one. Becausef isone-to-one, byTheorem3, f
1
exists.
1 1
20
10
10
20
y
x
(b) Thedomainof f
1
(x) istherangeof f (x) : (, ).
(c) Notethat f (1) = 1
7
+1+1= 3; therefore, f
1
(3) = 1.
S E C T I ON 7.2 Inverse Functions 381
Showthat f (x) = (x
2
+1)
1
isone-to-oneon(, 0], andndaformulafor f
1
for thisdomainof f .
21. Let f (x) = x
2
2x. Determineadomainonwhichf
1
exists, andndaformulafor f
1
for thisdomainof f .
solution Fromthegraphof y = x
2
2x shownbelow, weseethat if thedomainof f isrestrictedtoeither x 1or
x 1, thenf isone-to-oneandf
1
exists. Tondaformulafor f
1
, wesolvey = x
2
2x for x asfollows:
y +1= x
2
2x +1= (x 1)
2
x 1=
_
y +1
x = 1
_
y +1
If thedomainof f isrestrictedtox 1,thenwechoosethenegativesigninfrontof theradical andf
1
(x) = 1
x +1.
If thedomainof f isrestrictedtox 1, wechoosethepositivesigninfront of theradical andf
1
(x) = 1+
x +1.
y = x
2
2x
x
1 2 3
4
2
6
y
Showthat theinverseof f (x) = e
x
exists(without ndingit explicitly). What isthedomainof f
1
?
23. Findtheinverseg(x) of f (x) =
_
x
2
+9withdomainx 0andcalculateg
(x) intwoways: usingTheorem2and

bydirect calculation.
solution Tondaformulafor g(x) = f
1
(x), solvey =
_
x
2
+9for x. Thisyieldsx =
_
y
2
9. Becausethe
domainof f wasrestrictedtox 0, wemust choosethepositivesigninfront of theradical. Thus
g(x) = f
1
(x) =
_
x
2
9.
Becausex
2
+9 9for all x, itfollowsthatf (x) 3for all x. Thus, thedomainof g(x) = f
1
(x) isx 3. Therange
of g istherestricteddomainof f : y 0.
ByTheorem2,
g
(x) =
1
f
(g(x))
.
With
f
(x) =
x
_
x
2
+9
,
it followsthat
f
(g(x)) =
_
x
2
9
_
__
x
2
9
_
2
+9
=
_
x
2
9
x
2
=
_
x
2
9
x
sincethedomainof g isx 3. Thus,
g
(x) =
1
f
(g(x))
=
x
_
x
2
9
.
Thisagreeswiththeanswer weobtainbydifferentiatingdirectly:
g
(x) =
2x
2
_
x
2
9
=
x
_
x
2
9
.
Letg(x) betheinverseof f (x) = x
3
+1. Findaformulaforg(x) andcalculateg
(x) intwoways: usingTheorem

2andthenbydirect calculation.
In Exercises 2530, use Theorem 2 to calculate g
(x), where g(x) is the inverse of f (x).

25. f (x) = 7x +6
solution Letf (x) = 7x +6thenf
(x) = 7. Solvingy = 7x +6forx andswitchingvariables, weobtaintheinverse

g(x) = (x 6)/7. Thus,
g
(x) =
1
f
(g(x))
=
1
7
.
f (x) =

3x
27. f (x) = x
5
5
, then f
(x) = 5x
6
. Solvingy = x
5
for x andswitchingvariables, weobtain the
inverseg(x) = x
1/5
. Thus,
g
(x) =
1
5(x
1/5
)
6
=
1
5
x
6/5
.
f (x) = 4x
3
1
29. f (x) =
x
x +1
x
x+1
, then
f
(x) =
(x +1) x
(x +1)
2
=
1
(x +1)
2
.
Solvingy =
x
x+1
for x andswitchingvariables, weobtaintheinverseg(x) =
x
1x
. Thus
g
(x) = 1
_
1
(x/(1x) +1)
2
=
1
(1x)
2
.
f (x) = 2+x
1
31. Let g(x) betheinverseof f (x) = x
3
+ 2x + 4. Calculateg(7) [without nding a formula for g(x)], and then
calculateg
(7).
solution Let g(x) betheinverseof f (x) = x
3
+2x +4. Because
f (1) = 1
3
+2(1) +4= 7,
it followsthat g(7) = 1. Moreover, f
(x) = 3x
2
+2, and
g
(7) =
1
f
(g(7))
=
1
f
(1)
=
1
5
.
Findg
1
2
_
, whereg(x) istheinverseof f (x) =
x
3
x
2
+1
.
In Exercises 3338, calculate g(b) and g
(b), where g is the inverse of f (in the given domain, if indicated).

33. f (x) = x +cosx, b = 1
solution f (0) = 1, sog(1) = 0. f
(x) = 1sinx sof
(g(1)) = f
(0) = 1sin0= 1. Thus, g
(1) = 1/1= 1.
f (x) = 4x
3
2x, b = 2
35. f (x) =
_
x
2
+6x for x 0, b = 4
solution Todetermineg(4), wesolvef (x) =
_
x
2
+6x = 4for x. Thisyields:
x
2
+6x = 16
x
2
+6x 16= 0
(x +8)(x 2) = 0
or x = 8, 2. Becausethedomainof f hasbeenrestrictedtox 0, wehaveg(4) = 2. With
f
(x) =
x +3
_
x
2
+6x
,
it thenfollowsthat
g
(4) =
1
f
(g(4))
=
1
f
(2)
=
4
5
.
f (x) =
_
x
2
+6x for x 6, b = 4
37. f (x) =
1
x +1
, b =
1
4
solution f (3) = 1/4, sog(1/4) = 3. f
(x) =
1
(x+1)
2
sof
(g(1/4)) = f
(3) =
1
(3+1)
2
= 1/16. Thus, g
(1/4) =
16.
f (x) = e
x
, b = e
39. Let f (x) = x
n
andg(x) = x
1/n
. Computeg
(x) usingTheorem2andcheckyour answer usingthePower Rule.

solution Notethat g(x) = f
1
(x). Therefore,
g
(x) =
1
f
(g(x))
=
1
n(g(x))
n1
=
1
n(x
1/n
)
n1
=
1
n(x
11/n
)
=
x
1/n1
n
=
1
n
(x
1/n1
)
whichagreeswiththePower Rule.
S E C T I ON 7.3 Logarithms and Their Derivatives 383
Showthat f (x) =
1
1+x
andg(x) =
1x
x
areinverses. Thencomputeg
(x) directly andverify that g
(x) =
1/f
(g(x)).
41. Usegraphical reasoningtodetermineif thefollowingstatementsaretrueorfalse.If false,modifythestatement
tomakeit correct.
(a) If f (x) isincreasing, thenf
1
(x) isincreasing.
(b) If f (x) isdecreasing, thenf
1
(x) isdecreasing.
(c) If f (x) isconcaveup, thenf
1
(x) isconcaveup.
(d) If f (x) isconcavedown, thenf
1
(x) isconcavedown.
(e) Linear functionsf (x) = ax +b (a = 0) arealwaysone-to-one.
(f) Quadraticpolynomialsf (x) = ax
2
+bx +c (a = 0) arealwaysone-to-one.
(g) sinx isnot one-to-one.
solution
(a) Thisstatementistrue. Reectingthegraphof anincreasingfunctionacrosstheliney = x producesanotherincreasing
function.
(b) Thisstatementistrue.Reectingthegraphof andecreasingfunctionacrosstheliney = x producesanotherdecreasing
function.
(c) Thisstatement isfalse. Reectingthegraphof aconcaveupfunctionacrosstheliney = x producesagraphthat is
concavedown. Thecorrect statement is: If f (x) isconcaveup, thenf
1
(x) isconcavedown.
(d) Thisstatement isfalse. Reectingthegraphof aconcavedownfunctionacrosstheliney = x producesagraphthat
isconcaveup. Thecorrect statement is: If f (x) isconcavedown, thenf
1
(x) isconcaveup.
(e) Thisstatementistrue.Anylinear functionf (x) = ax +b witha = 0iseither alwaysincreasing(if a > 0) or always
decreasing(if a < 0) andisthusone-to-one.
(f) Thisstatement isfalse. Every quadraticpolynomial f (x) = ax
2
+ bx + c witha = 0failsthehorizontal linetest.
Thecorrect statement is: Quadraticpolynomialsf (x) = ax
2
+bx +c (a = 0) arenever one-to-one.
(g) Thisstatement istrue. Thegraphof sinx failsthehorizontal linetest.
Showthat if f (x) isoddandf
1
(x) exists, thenf
1
(x) isodd. Show, ontheother hand, that anevenfunction
doesnot haveaninverse.
43. Let g betheinverseof afunctionf satisfyingf
(x) = f (x). Showthat g
(x) = x
1
. Wewill applythisinthenext
sectiontoshowthat theinverseof f (x) = e
x
(thenatural logarithm) isanantiderivativeof x
1
.
solution
g
(x) =
1
f
(g(x))
=
1
f
(f
1
(x))
=
1
f (f
1
(x))
=
1
x
.
7.3 Logarithms and Their Derivatives
1. Computelog
b
2(b
4
).
solution Becauseb
4
= (b
2
)
2
, log
b
2(b
4
) = 2.
2. Whenislnx negative?
solution lnx isnegativefor 0< x < 1.
3. What isln(3)? Explain.
solution ln(3) isnot dened.
4. ExplainthephraseThelogarithmconvertsmultiplicationintoaddition.
solution Thisphraseisaverbal descriptionof thegeneral propertyof logarithmsthat states
log(ab) = loga +logb.
5. What arethedomainandrangeof lnx?
solution Thedomainof lnx isx > 0andtherangeisall real numbers.
6. Doesx
1
haveanantiderivativefor x < 0? If so, describeone.
solution Yes, ln(x) isanantiderivativeof f (x) = x
1
for x < 0.
7. What istheslopeof thetangent linetoy = 4
x
at x = 0?
solution Theslopeof thetangent linetoy = 4
x
at x = 0is
d
dx
4
x
x=0
= 4
x
ln4
x=0
= ln4.
8. What istherateof changeof y = lnx at x = 10?
solution Therateof changeof y = lnx at x = 10is
d
dx
lnx
x=10
=
1
x
x=10
=
1
10
.
Exercises
In Exercises 116, calculate without using a calculator.
1. log
3
27
solution log
3
27= log
3
3
3
= 3log
3
3= 3.
log
5
1
25
3. ln1
solution ln1= 0
log
5
(5
4
)
5. log
2
(2
5/3
)
solution log
2
2
5/3
=
5
3
log
2
2=
5
3
.
log
2
(8
5/3
)
7. log
64
4
solution log
64
4= log
64
64
1/3
=
1
3
log
64
64=
1
3
.
log
7
(49
2
)
9. log
8
2+log
4
2
solution log
8
2+log
4
2= log
8
8
1/3
+log
4
4
1/2
=
1
3
+
1
2
=
5
6
.
log
25
30+log
25
5
6
11. log
4
48log
4
12
solution log
4
48log
4
12= log
4
48
12
= log
4
4= 1.
ln(
e e
7/5
)
13. ln(e
3
) +ln(e
4
)
solution ln(e
3
) +ln(e
4
) = 3+4= 7.
log
2
4
3
+log
2
24
15. 7
log
7
(29)
solution 7
log
7
(29)
= 29.
8
3log
8
(2)
17. Writeasthenatural logof asingleexpression:
(a) 2ln5+3ln4 (b) 5ln(x
1/2
) +ln(9x)
solution
(a) 2ln5+3ln4= ln5
2
+ln4
3
= ln25+ln64= ln(25 64) = ln1600.
(b) 5lnx
1/2
+ln9x = lnx
5/2
+ln9x = ln(x
5/2
9x) = ln(9x
7/2
).
Solvefor x: ln(x
2
+1) 3lnx = ln(2).
In Exercises 1924, solve for the unknown.
19. 7e
5t
= 100
solution Dividetheequationby7andthentakethenatural logarithmof bothsides. Thisgives
5t = ln
_
100
7
_
or t =
1
5
ln
_
100
7
_
.
6e
4t
= 2
21. 2
x
2
2x
= 8
solution Since8= 2
3
, wehavex
2
2x 3= 0or (x 3)(x +1) = 0. Thus, x = 1or x = 3.
e
2t +1
= 9e
1t
23. ln(x
4
) ln(x
2
) = 2
solution ln(x
4
) ln(x
2
) = ln
_
x
4
x
2
_
= ln(x
2
) = 2lnx. Thus, 2lnx = 2or lnx = 1. Hence, x = e.
log
3
y +3log
3
(y
2
) = 14
25. Show, byproducingacounterexample, that ln(ab) isnot equal to(lna)(lnb).
solution Let a = e andb = e
2
. Then
ln(ab) = lne
3
= 3lne = 3;
but
(lna)(lnb) = (lne)(lne
2
) = 1(2) = 2.
What isb if (log
b
x)
=
1
3x
?
27. Thepopulationof acity(inmillions) attimet (years) isP(t ) = 2.4e
0.06t
, wheret = 0istheyear 2000. Whenwill
thepopulationdoublefromitssizeat t = 0?
solution Populationdoubleswhen4.8= 2.4e
0.06t
. Thus, 0.06t = ln2or t =
ln2
0.06
11.55years.
TheGutenbergRichter Lawstatesthatthenumber N of earthquakesper year worldwideof Richter magnitude
at least M satises anapproximaterelationlog
10
N = a M for someconstant a. Finda, assumingthat thereis
oneearthquakeof magnitudeM 8per year. Howmanyearthquakesof magnitudeM 5occur per year?
29. y = x lnx
solution
d
dx
x lnx = lnx +
x
x
= lnx +1.
y = t lnt t
31. y = (lnx)
2
solution
d
dx
(lnx)
2
= (2lnx)
1
x
=
2
x
lnx.
y = ln(x
5
)
33. y = ln(9x
2
8)
solution
d
dx
ln(9x
2
8) =
1
9x
2
8
(18x) =
18x
9x
2
8
.
y = ln(t 5
t
)
35. y = ln(sint +1)
solution
d
dt
ln(sint +1) =
cost
sint +1
.
y = x
2
lnx
37. y =
lnx
x
solution
d
dx
lnx
x
=
1
x
(x) lnx
x
2
=
1lnx
x
2
.
y = e
(lnx)
2
39. y = ln(lnx)
solution
d
dx
ln(lnx) =
1
x lnx
.
y = ln(cotx)
41. y =
_
ln(lnx)
_
3
solution
d
dx
(ln(lnx))
3
= 3(ln(lnx))
2
_
1
lnx
__
1
x
_
=
3(ln(lnx))
2
x lnx
.
y = ln
_
(lnx)
3
_
43. y = ln
_
(x +1)(2x +9)
_
solution
d
dx
ln((x +1)(2x +9)) =
1
(x +1)(2x +9)
((x +1)2+(2x +9)) =
4x +11
(x +1)(2x +9)
.
Alternately, becauseln((x +1)(2x +9)) = ln(x +1) +ln(2x +9),
d
dx
ln((x +1)(2x +9)) =
1
x +1
+
2
2x +9
=
4x +11
(x +1)(2x +9)
.
y = ln
_
x +1
x
3
+1
_ 45. y = 11
x
solution
d
dx
11
x
= ln11 11
x
.
y = 7
4xx
2
47. y =
2
x
3
x
x
solution
d
dx
2
x
3
x
x
=
x(2
x
ln2+3
x
ln3) (2
x
3
x
)
x
2
.
y = 16
sinx
49. f
(x), f (x) = log

2
x
solution f (x) = log
2
x =
lnx
ln2
. Thus, f
(x) =
1
x
1
ln2
.
f
(3), f (x) = log

5
x
51.
d
dt
log
3
(sint )
solution
d
dt
log
3
(sint ) =
d
dt
_
ln(sint )
ln3
_
=
1
ln3
1
sint
cost =
cott
ln3
.
d
dt
log
10
(t +2
t
)
In Exercises 5364, nd an equation of the tangent line at the point indicated.
53. f (x) = 6
x
, x = 2
x
. Thenf (2) = 36, f
(x) = 6
x
ln6andf
(2) = 36ln6. Theequationof thetangent lineis

thereforey = 36ln6(x 2) +36.
y = (
2)
x
, x = 8
55. s(t ) = 3
9t
, t = 2
solution Let s(t ) = 3
9t
. Thens(2) = 3
18
, s
(t ) = 3
9t
9ln3, ands
(2) = 3
18
9ln3= 3
20
ln3. Theequationof the
tangent lineisthereforey = 3
20
ln3(t 2) +3
18
.
y =
5x2
, x = 1
57. f (x) = 5
x
2
2x
, x = 1
x
2
2x+9
. Then f (1) = 5
8
. f
(x) = ln5 5
x
2
2x+9
(2x 2), so f
(1) = ln5(0) = 0.
Therefore, theequationof thetangent lineisy = 5
8
.
s(t ) = lnt , t = 5
59. s(t ) = ln(84t ), t = 1
solution Let s(t ) = ln(84t ). Thens(1) = ln(84) = ln4. s
(t ) =
4
84t
, sos
(1) = 4/4= 1. Thereforethe

equationof thetangent lineisy = 1(t 1) +ln4.
f (x) = ln(x
2
), x = 4
61. R(z) = log
5
(2z
2
+7), z = 3
solution Let R(z) = log
5
(2z
2
+7). ThenR(3) = log
5
(25) = 2,
R
(z) =
4z
(2z
2
+7) ln5
, and R
(3) =
12
25ln5
.
Theequationof thetangent lineistherefore
y =
12
25ln5
(z 3) +2.
y = ln(sinx), x =

4
63. f (w) = log
2
w, w =
1
8
solution Let f (w) = log
2
w. Then
f
_
1
8
_
= log
2
1
8
= log
2
2
3
= 3,
f
(w) =
1
wln2
, and
f
_
1
8
_
=
8
ln2
.
Theequationof thetangent lineistherefore
y =
8
ln2
_
w
1
8
_
3.
y = log
2
(1+4x
1
), x = 4
In Exercises 6572, nd the derivative using logarithmic differentiation as in Example 8.
65. y = (x +5)(x +9)
solution Let y = (x + 5)(x + 9). Then lny = ln((x + 5)(x + 9)) = ln(x + 5) + ln(x + 9). By logarithmic
differentiation
y
y
=
1
x +5
+
1
x +9
or
y
= (x +5)(x +9)
_
1
x +5
+
1
x +9
_
= (x +9) +(x +5) = 2x +14.
y = (3x +5)(4x +9)
67. y = (x 1)(x 12)(x +7)
solution Let y = (x 1)(x 12)(x + 7). Then lny = ln(x 1) + ln(x 12) + ln(x + 7). By logarithmic
differentiation,
y
y
=
1
x 1
+
1
x 12
+
1
x +7
or
y
= (x 12)(x +7) +(x 1)(x +7) +(x 1)(x 12) = 3x

2
12x 79.
y =
x(x +1)
3
(3x 1)
2
69. y =
x(x
2
+1)
x +1
solution Let y =
x(x
2
+1)
x+1
. Thenlny = lnx +ln(x
2
+1)
1
2
ln(x +1). Bylogarithmicdifferentiation
y
y
=
1
x
+
2x
x
2
+1
1
2(x +1)
,
so
y
=
x(x
2
+1)
x +1
_
1
x
+
2x
x
2
+1
1
2(x +1)
_
.
y = (2x +1)(4x
2
)
x 9
71. y =
_
x(x +2)
(2x +1)(3x +2)
solution Let y =
_
x(x+2)
(2x+1)(2x+2)
. Thenlny =
1
2
[ln(x) + ln(x + 2) ln(2x + 1) ln(2x + 2)]. By logarithmic
differentiation
y
y
=
1
2
_
1
x
+
1
x +2
2
2x +1
2
2x +2
_
,
so
y
=
1
2
_
x(x +2)
(2x +1)(2x +2)
_
1
x
+
1
x +2
2
2x +1
1
x +1
_
.
y = (x
3
+1)(x
4
+2)(x
5
+3)
2
In Exercises 7378, nd the derivative using either method of Example 9.
73. f (x) = x
3x
solution Method1: x
3x
= e
3x lnx
, so
d
dx
x
3x
= e
3x lnx
(3+3lnx) = x
3x
(3+3lnx).
Method2: Let y = x
3x
. Then, lny = 3x lnx. Bylogarithmicdifferentiation
y
y
= 3x
1
x
+3lnx,
so
y
= y(3+3lnx) = x
3x
(3+3lnx) .
f (x) = x
cosx 75. f (x) = x
e
x
solution Method1: x
e
x
= e
e
x
lnx
, so
d
dx
x
e
x
= e
e
x
lnx
_
e
x
x
+e
x
lnx
_
= x
e
x
_
e
x
x
+e
x
lnx
_
.
Method2: Let y = x
e
x
. Thenlny = e
x
lnx. Bylogarithmicdifferentiation
y
y
= e
x
1
x
+e
x
lnx,
so
y
= y
_
e
x
x
+e
x
lnx
_
= x
e
x
_
e
x
x
+e
x
lnx
_
.
f (x) = x
x
2
77. f (x) = x
3
x
solution Method1: x
3
x
= e
3
x
lnx
, so
d
dx
x
3
x
= e
3
x
lnx
_
3
x
x
+(lnx)(ln3)3
x
_
= x
3
x
_
3
x
x
+(lnx)(ln3)3
x
_
.
Method2: Let y = x
3
x
. Thenlny = 3
x
lnx. Bylogarithmicdifferentiation
y
y
= 3
x
1
x
+(lnx)(ln3)3
x
,
so
y
= x
3
x
_
3
x
x
+(lnx)(ln3)3
x
_
.
f (x) = e
x
x In Exercises 7982, nd the local extreme values in the domain {x : x > 0} and use the Second Derivative Test to
determine whether these values are local minima or maxima.
79. g(x) =
lnx
x
solution Let g(x) =
lnx
x
. Then
g
(x) =
x(1/x) lnx
x
2
=
1lnx
x
2
.
Weknowg
(x) = 0when1lnx = 0, or whenx = e.

g
(x) =
x
2
(1/x) (1lnx)(2x)
x
4
=
3+2lnx
x
3
sog
(e) =
1
e
3
< 0. Thus, g(e) isalocal maximum.
g(x) = x lnx 81. g(x) =
lnx
x
3
solution Let g(x) =
lnx
x
3
. Then
g
(x) =
x
3
(1/x) 3x
2
lnx
x
6
=
13lnx
x
4
.
Weknowg
(x) = 0when13lnx = 0, or whenx = e

1/3
.
g
(x) =
x
4
(3/x) (13lnx)(4x
3
)
x
8
=
7+12lnx
x
5
sog
(e
1/3
) =
3
e
5/3
< 0. Thus, g(e) isalocal maximum.
g(x) = x lnx
In Exercises 83 and 84, nd the local extreme values and points of inection, and sketch the graph of y = f (x) over the
interval [1, 4].
83. f (x) =
10lnx
x
2
10lnx
x
2
. Then
f
(x) =
x
2
(10/x) 20x lnx
x
4
=
10(12lnx)
x
3
and
f
(x) =
x
3
(20/x) 30x
2
(12lnx)
x
6
=
10(6lnx 5)
x
4
.
Thus, f isincreasingfor 1 x <
e, isdecreasingfor
e < x 4andhasalocal maximumvalueof 5/e at x =

e.
Moreover, f isconcavedownfor 1 x < e
5/6
, isconcaveupfor e
5/6
< x 4andhasapointof inectionatx = e
5/6
.
A graphof y = f (x) isshownbelow.
2.0
1.5
1.0
0.5
1 2 3 4
x
y
f (x) = x
2
8lnx
In Exercises 85105, evaluate the indenite integral, using substitution if necessary.
85.
_
7dx
x
solution
_
7dx
x
= 7
_
dx
x
= 7ln|x| +C.
_
dx
x +7
87.
_
dx
2x +4
solution Let u = 2x +4. Thendu = 2dx, and
_
dx
2x +4
=
1
2
_
1
u
du =
1
2
ln|2x +4| +C.
_
dx
9x 3
89.
_
t dt
t
2
+4
solution Let u = t
2
+4. Thendu = 2t dt or
1
2
du = t dt , and
_
t
t
2
+4
dt =
1
2
_
1
u
du =
1
2
ln
_
t
2
+4
_
+C,
wherewehaveusedthefact that t
2
+4 4todroptheabsolutevalue.
_
x
2
dx
x
3
+2
91.
_
(3x 1) dx
92x +3x
2
solution Let u = 92x +3x
2
. Thendu = (2+6x) dx = 2(3x 1) dx, and
_
(3x 1)dx
92x +3x
2
=
1
2
_
du
u
=
1
2
ln(92x +3x
2
) +C,
wherewehaveusedthefact that 92x +3x
2
> 0for all x todroptheabsolutevalue.
_
tan(4x +1) dx
93.
_
cotx dx
solution Werewrite
_
cotx dx as
_
cosx
sinx
dx. Let u = sinx. Thendu = cosx dx, and
_
cosx
sinx
dx =
_
du
u
= ln| sinx| +C.
_
cosx
2sinx +3
dx
95.
_
lnx
x
dx
solution Let u = lnx. Thendu = (1/x) dx, and
_
lnx
x
dx =
_
u du =
u
2
2
+C =
(lnx)
2
2
+C.
_
4lnx +5
x
dx
97.
_
(lnx)
2
x
dx
solution Let u = lnx. Thendu = (1/x)dx, and
_
(lnx)
2
x
dx =
_
u
2
du =
1
3
u
3
+C =
(lnx)
3
3
+C.
_
dx
x lnx
99.
_
dx
(4x 1) ln(8x 2)
solution Let u = ln(8x 2). Thendu =
8
8x 2
dx =
4
4x 1
dx, and
_
dx
(4x 1) ln(8x 2)
=
1
4
_
du
u
=
1
4
ln|u| +C =
1
4
ln| ln(8x 2)| +C.
_
ln(lnx)
x lnx
dx
101.
_
cotx ln(sinx) dx
solution Let u = ln(sinx). Then
du =
1
sinx
cosx dx = cotx dx,
and
_
cotx ln(sinx) dx =
_
u du =
u
2
2
+C =
(ln(sinx))
2
2
+C.
_
3
x
dx
103.
_
x3
x
2
dx
solution Let u = x
2
. Thendu = 2x dx, and
_
x3
x
2
dx =
1
2
_
3
u
du =
1
2
3
u
ln3
+C =
3
x
2
2ln3
+C.
_
cosx 3
sinx
dx
105.
_ _
1
2
_
3x+2
dx
solution Let u = 3x +2. Thendu = 3dx, and
_ _
1
2
_
3x+2
dx =
1
3
_ _
1
2
_
u
du =
1
3
(1/2)
u
ln1/2
+C =
(1/2)
3x+2
3ln(1/2)
+C.
In Exercises 106111, evaluate the denite integral.
_
2
1
1
x
dx
107.
_
12
4
1
x
dx
solution
_
12
4
1
x
dx = ln|x|
12
4
= ln12ln4= ln(12/4) = ln3.
_
e
1
1
x
dx
109.
_
4
2
dt
3t +4
solution Let u = 3t +4. Thendu = 3dt and
_
4
2
dt
3t +4
=
1
3
_
16
10
du
u
=
1
3
ln|u|
16
10
=
1
3
(ln16ln10) .
_
e
e
2
1
t
dt
111.
_
e
2
e
1
t lnt
dt
solution Let u = lnt . Thendu = (1/t )dt and
_
e
2
e
1
t lnt
dt =
_
2
1
du
u
= ln|u|
2
1
= ln2ln1= ln2.
Findagoodnumerical approximationtothecoordinatesof thepoint onthegraphof y = lnx x closest to
theorigin(Figure9).
113. Findtheminimumvalueof f (x) = x
x
for x > 0.
x
. ByExample9fromthetext, weknowthat f
(x) = x
x
(1+lnx). Thus, x =
1
e
istheonly
critical point. Becausef
(x) < 0for 0< x <

1
e
andf
(x) > 0for x >

1
e
,
f
_
1
e
_
=
_
1
e
_
1/e
0.692201
istheminimumvalue.
Usetheformula(lnf (x))
= f
(x)/f (x) to showthat lnx andln(2x) havethesamederivative. Is therea

simpler explanationof thisresult?
115. According to one simplied model, the purchasing power of a dollar in the year 2000+ t is equal to P(t ) =
0.68(1.04)
t
(in1983dollars). Calculatethepredictedrateof declineinpurchasingpower (incentsper year) intheyear
2020.
solution First, notethat
P
(t ) = 0.68(1.04)
t
ln1.04;
thus, therateof changeintheyear 2020is
P
(20) = 0.68(1.04)
20
ln1.04= 0.0122.
That is, therateof declineis1.22centsper year.
TheenergyE (injoules) radiatedasseismicwavesbyanearthquakeof Richter magnitudeM satiseslog
10
E =
4.8+1.5M.
(a) Showthat whenM increasesby1, theenergyincreasesbyafactor of approximately31.5.
(b) CalculatedE/dM.
117. ThePalermoTechnical ImpactHazardScaleP isusedtoquantifytheriskassociatedwiththeimpactof anasteroid
collidingwiththeearth:
P = log
10
_
p
i
E
0.8
0.03T
_
wherep
i
istheprobabilityof impact, T isthenumber of yearsuntil impact, andE istheenergyof impact (inmegatons
of TNT). Theriskisgreater thanarandomevent of similar magnitudeif P > 0.
(a) CalculatedP/dT , assumingthat p
i
= 210
5
andE = 2megatons.
(b) UsethederivativetoestimatethechangeinP if T increasesfrom8to9years.
solution
(a) Observethat
P = log
10
_
p
i
E
0.8
0.03T
_
= log
10
_
p
i
E
0.8
0.03
_
log
10
T,
so
dP
dT
=
1
T ln10
.
(b) If T increasesto26yearsfrom25years, then
P
dP
dT
T =25
T =
1
(25yr) ln10
(1yr) = 0.017
(a) Showthat if f andg aredifferentiable, then
d
dx
ln(f (x)g(x)) =
f
(x)
f (x)
+
g
(x)
g(x)
(b) Giveanewproof of theProduct Rulebyobservingthat theleft-handsideof Eq. (8) isequal to
(f (x)g(x))
f (x)g(x)
.
119. Provetheformula
log
b
x =
log
a
x
log
a
b
for all positivenumbersa, b witha = 1andb = 1.
solution Let y = log
b
x. Thenx = b
y
andlog
a
x = log
a
b
y
= y log
a
b. Thus, y =
log
a
x
log
a
b
.
Provetheformulalog
a
b log
b
a = 1for all positivenumbersa, b witha = 1andb = 1.
Exercises 121123 develop an elegant approach to the exponential and logarithm functions. Dene a function G(x) for
x > 0:
G(x) =
_
x
1
1
t
dt
121. Deninglnx asanIntegral This exerciseproceeds as if wedidnt knowthat G(x) = lnx andshows directly
that G(x) hasall thebasicpropertiesof thelogarithm. Provethefollowingstatements.
(a)
_
ab
a
1
t
dt =
_
b
1
1
t
dt for all a, b > 0. Hint: Usethesubstitutionu = t /a.
(b) G(ab) = G(a) +G(b). Hint: Breakuptheintegral from1toab intotwointegralsanduse(a).
(c) G(1) = 0andG(a
1
) = G(a) for a > 0.
(d) G(a
n
) = nG(a) for all a > 0andintegersn.
(e) G(a
1/n
) =
1
n
G(a) for all a > 0andintegersn = 0.
(f) G(a
r
) = rG(a) for all a > 0andrational numbersr.
(g) G(x) isincreasing. Hint: UseFTC II.
(h) Thereexistsanumber a suchthat G(a) > 1. Hint: Showthat G(2) > 0andtakea = 2
m
for m > 1/G(2).
(i) lim
x
G(x) = and lim
x0+
G(x) =
(j) Thereexistsauniquenumber E suchthat G(E) = 1.
(k) G(E
r
) = r for everyrational number r.
solution
(a) Let u = t /a. Thendu = dt /a, u(a) = 1, u(ab) = b and
_
ab
a
1
t
dt =
_
ab
a
a
at
dt =
_
b
1
1
u
du =
_
b
1
1
t
dt.
(b) Usingpart (a),
G(ab) =
_
ab
1
1
t
dt =
_
a
1
1
t
dt +
_
ab
a
1
t
dt =
_
a
1
1
t
dt +
_
b
1
1
t
dt = G(a) +G(b).
(c) First,
G(1) =
_
1
1
1
t
dt = 0.
Next,
G(a
1
) = G
_
1
a
_
=
_
1/a
1
1
t
dt =
_
1
a
1
t
dt bypart (a) withb =
1
a
=
_
a
1
1
t
dt = G(a).
(d) Usingpart (a),
G(a
n
) =
_
a
n
1
1
t
dt =
_
a
1
1
t
dt +
_
a
2
a
1
t
dt + +
_
a
n
a
n1
1
t
dt
=
_
a
1
1
t
dt +
_
a
1
1
t
dt + +
_
a
1
1
t
dt = nG(a).
(e) G(a) = G((a
1/n
)
n
= nG(a
1/n
). Thus, G(a
1/n
) =
1
n
G(a).
(f) Let r = m/n wherem andn areintegers. Then
G(a
r
) = G(a
m/n
) = G((a
m
)
1/n
)
=
1
n
G(a
m
) bypart (e)
=
m
n
G(a) bypart d
= rG(a).
(g) BytheFundamental Theoremof Calculus, G(x) iscontinuouson(0, ) andG
(x) =
1
x
> 0for x > 0. Thus, G(x)
isincreasingandone-to-onefor x > 0.
(h) First notethat
G(2) =
_
2
1
1
t
dt >
1
2
> 0
because
1
t
>
1
2
for t (1, 2). Now, let a = 2
m
for m aninteger greater than1/G(2). Then
G(a) = G(2
m
) = mG(2) >
1
G(2)
G(2) = 1.
(i) First, let a bethevaluefrompart (h) for whichG(a) > 1(notethat a itself isgreater than1). Now,
lim
x
G(x) = lim
m
G(a
m
) = G(a) lim
m
m = .
For theother limit, let t = 1/x andnote
lim
x0+
G(x) = lim
t
G
_
1
t
_
= lim
t
G(t ) = .
(j) By part (c), G(1) = 0andby part (h) thereexists ana suchthat G(a) > 1. theIntermediateValueTheoremthen
guarantees thereexists anumber E such that 1 < E < a and G(E) = 1. Weknow that E is uniquebecauseG is
one-to-one.
S E C T I ON 7.4 Exponential Growth and Decay 393
(k) Usingpart (f) andthenpart (j),
G(E
r
) = rG(E) = r 1= r.
Deninge
x
UseExercise121toprovethefollowingstatements.
(a) G(x) hasaninversewithdomainR andrange{x : x > 0}. DenotetheinversebyF(x).
(b) F(x +y) = F(x)F(y) for all x, y. Hint: it sufcestoshowthat G(F(x)F(y)) = G(F(x +y)).
(c) F(r) = E
r
for all numbers. Inparticular, F(0) = 1.
(d) F
(x) = F(x). Hint: Usetheformulafor thederivativeof aninversefunction.

Thisshowsthat E = e andthat F(x) isthefunctione
x
asdenedinthetext.
123. Deningb
x
Let b > 0andlet f (x) = F(xG(b)) withF asinExercise122. UseExercise121(f) toprovethat
f (r) = b
r
for everyrational number r. Thisgivesusawayof deningb
x
for irrational x, namelyb
x
= f (x). Withthis
denition, b
x
isadifferentiablefunctionof x (becauseF isdifferentiable).
solution ByExercise121(f),
f (r) = F(rG(b)) = F(G(b
r
)) = b
r
,
for everyrational number r.
7.4 Exponential Growth and Decay
1. Two quantities increaseexponentially with growth constants k = 1.2 and k = 3.4, respectively. Which quantity
doublesmorerapidly?
solution Doublingtimeis inversely proportional to thegrowthconstant. Consequently, thequantity withk = 3.4
doublesmorerapidly.
2. A cell populationgrowsexponentiallybeginningwithonecell. Whichtakeslonger: increasingfromonetotwocells
or increasingfrom15millionto20millioncells?
solution It takeslonger for thepopulationtoincreasefromonecell totwocells, becausethisrequiresdoublingthe
population. Increasingfrom15millionto20millionislessthandoublingthepopulation.
3. Referringtohispopular bookA Brief History of Time, therenownedphysicistStephenHawkingsaid, Someonetold
methat eachequationI includedinthebookwouldhalveitssales. Findadifferential equationsatisedbythefunction
S(n), thenumber of copiessoldif thebookhasn equations.
solution LetS(0) denotethesaleswithnoequationsinthebook. TranslatingHawkingsobservationintoanequation
yields
S(n) =
S(0)
2
n
.
Differentiatingwithrespect ton thenyields
dS
dn
= S(0)
d
dn
2
n
= ln2S(0)2
n
= ln2S(n).
4. Carbondatingisbasedontheassumptionthat theratioR of C
14
toC
12
intheatmospherehasbeenconstant over
thepast 50,000years. If R wereactuallysmaller inthepast thanit istoday, wouldtheageestimatesproducedbycarbon
datingbetooancient or toorecent?
solution If R wereactuallysmaller inthepast thanit istoday, thenwewouldbeoverestimatingtheamount of decay
andthereforeoverestimatingtheage. Our estimateswouldbetooancient.
Exercises
1. A certainpopulationP of bacteriaobeystheexponential growthlawP(t ) = 2000e
1.3t
(t inhours).
(a) Howmanybacteriaarepresent initially?
(b) At what timewill therebe10,000bacteria?
solution
(a) P(0) = 2000e
0
= 2000bacteriainitially.
(b) Wesolve2000e
1.3t
= 10, 000for t . Thus, e
1.3t
= 5or
t =
1
1.3
ln5 1.24hours.
A quantityP obeystheexponential growthlawP(t ) = e
5t
(t inyears).
(a) At what timet isP = 10?
(b) What isthedoublingtimefor P?
3. Writef (t ) = 5(7)
t
intheformf (t ) = P
0
e
kt
for someP
0
andk.
solution Because7= e
ln7
, it followsthat
f (t ) = 5(7)
t
= 5(e
ln7
)
t
= 5e
t ln7
.
Thus, P
0
= 5andk = ln7.
Writef (t ) = 9e
1.4t
intheformf (t ) = P
0
b
t
for someP
0
andb.
5. A certainRNA moleculereplicatesevery3minutes. Findthedifferential equationfor thenumber N(t ) of molecules
present at timet (inminutes). Howmanymoleculeswill bepresent after onehour if thereisonemoleculeat t = 0?
solution The doubling time is
ln2
k
so k =
ln2
doublingtime
. Thus, the differential equation is N
(t ) = kN(t ) =
ln2
3
N(t ). Withonemoleculeinitially,
N(t ) = e
(ln2/3)t
= 2
t /3
.
Thus, after onehour, thereare
N(60) = 2
60/3
= 1,048,576
moleculespresent.
A quantity P obeystheexponential growthlawP(t ) = Ce
kt
(t inyears). Findtheformulafor P(t ), assuming
that thedoublingtimeis7yearsandP(0) = 100.
7. Findall solutionstothedifferential equationy
= 5y. Whichsolutionsatisestheinitial conditiony(0) = 3.4?

solution y
= 5y, so y(t ) = Ce
5t
for someconstant C. Theinitial conditiony(0) = 3.4determines C = 3.4.
Therefore, y(t ) = 3.4e
5t
.
Findthesolutiontoy
=

2y satisfyingy(0) = 20.
9. Findthesolutiontoy
= 3y satisfyingy(2) = 1000.
solution y
= 3y, soy(t ) = Ce
3t
for someconstant C. Theinitial conditiony(2) = 1000determines C =
1000
e
6
.
Therefore, y(t ) =
1000
e
6
e
3t
= 1000e
3(t 2)
.
Findthefunctiony = f (t ) thatsatisesthedifferential equationy
= 0.7y andtheinitial conditiony(0) = 10.

11. Thedecayconstant of cobalt-60is0.13year
1
. Finditshalf-life.
solution Half-life=
ln2
0.13
5.33years.
Thehalf-liferadium-226is1622years. Finditsdecayconstant.
13. Oneof theworldssmallest oweringplants, Wolfa globosa (Figure11), hasadoublingtimeof approximately30
hours. Findthegrowthconstant k anddeterminetheinitial populationif thepopulationgrewto1000after 48hours.
FIGURE 11 ThetinyplantsareWolfa, withplant bodiessmaller thantheheadof apin.
solution Bytheformulafor thedoublingtime, 30=
ln2
k
. Therefore,
k =
ln2
30
0.023hours
1
.
Theplant populationafter t hoursisP(t ) = P
0
e
0.023t
. If P(48) = 1000, then
P
0
e
(0.023)48
= 1000 P
0
= 1000e
(0.023)48
332
A 10-kgquantityof aradioactiveisotopedecaysto3kgafter 17years. Findthedecayconstant of theisotope.
15. Thepopulationof acity isP(t ) = 2 e
0.06t
(inmillions), wheret ismeasuredinyears. Calculatethetimeit takes
for thepopulationtodouble, totriple, andtoincreaseseven-fold.
solution Sincek = 0.06, thedoublingtimeis
ln2
k
11.55years.
Thetriplingtimeiscalculatedinthesamewayasthedoublingtime. Solvefor intheequation
P(t +) = 3P(t )
2 e
0.06(t +)
= 3(2e
0.06t
)
2 e
0.06t
e
0.06
= 3(2e
0.06t
)
e
0.06
= 3
0.06 = ln3,
or = ln3/0.06 18.31years. Workinginasimilarfashion, wendthatthetimerequiredforthepopulationtoincrease
seven-foldis
ln7
k
=
ln7
0.06
32.43years.
What is thedifferential equationsatisedby P(t ), thenumber of infectedcomputer hosts inExample4? Over
whichtimeinterval wouldP(t ) increaseonehundred-fold?
17. Thedecay constant for acertaindrugisk = 0.35day
1
. Calculatethetimeit takesfor thequantity present inthe
bloodstreamtodecreasebyhalf, byone-third, andbyone-tenth.
solution Thetimerequiredfor thequantitypresent inthebloodstreamtodecreasebyhalf is
ln2
k
=
ln2
0.35
1.98days.
Todecaybyone-third(meaningthat two-thirdsremainsinthebloodstream), thetimeis
ln3/2
k
=
ln3/2
0.35
1.16days.
Finally, todecaybyone-tenth(meaningthat nine-tenthsremainsinthebloodstream), thetimeis
ln10/9
k
=
ln10/9
0.35
0.30days.
LightIntensity Theintensity of light passingthroughanabsorbingmediumdecreasesexponentially withthe
distancetraveled. Supposethedecayconstant for acertainplasticblockisk = 4m
1
. Howthickmust theblockbe
toreducetheintensitybyafactor of one-third?
19. Assumingthatpopulationgrowthisapproximatelyexponential, whichof thefollowingtwosetsof dataismostlikely
torepresent thepopulation(inmillions) of acityover a5-year period?
Year 2000 2001 2002 2003 2004
Set I 3.14 3.36 3.60 3.85 4.11
Set II 3.14 3.24 3.54 4.04 4.74
solution If thepopulationgrowthisapproximatelyexponential, thentheratiobetweensuccessiveyears dataneeds
tobeapproximatelythesame.
Year 2000 2001 2002 2003 2004
DataI 3.14 3.36 3.60 3.85 4.11
Ratios 1.07006 1.07143 1.06944 1.06753
DataII 3.14 3.24 3.54 4.04 4.74
Ratios 1.03185 1.09259 1.14124 1.17327
Asyoucansee, theratioof successiveyearsinthedatafromDataI isverycloseto1.07. Therefore, wewouldexpect
exponential growthof about P(t ) (3.14)(1.07
t
).
The atmospheric pressureP(h) (in kilopascals) at a height h meters above sea level satises a differential
equationP
= kP for somepositiveconstant k.
(a) Barometricmeasurementsshowthat P(0) = 101.3andP(30, 900) = 1.013. What isthedecayconstant k?
(b) Determinetheatmosphericpressureat h = 500.
21. DegreesinPhysics Onestudysuggeststhatfrom1955to1970,thenumberof bachelorsdegreesinphysicsawarded
per year byU.S. universitiesgrewexponentially, withgrowthconstant k = 0.1.
(a) If exponential growthcontinues, howlongwill ittakeforthenumberof degreesawardedperyeartoincrease14-fold?
(b) If 2500degreeswereawardedin1955, inwhichyear were10,000degreesawarded?
solution
(a) Thetimerequiredfor thenumber of degreestoincrease14-foldis
ln14
k
=
ln14
0.1
26.39years.
(b) Thedoublingtimeis(ln2)/0.1 0.693/0.1= 6.93years. Sincedegreesareusuallyawardedonceayear, weround
off thedoublingtimeto7years. Thenumber quadruplesafter 14years, so10, 000degreeswouldbeawardedin1969.
TheBeerLambertLawisusedinspectroscopytodeterminethemolar absorptivity or theconcentrationc of
acompounddissolvedinasolutionat lowconcentrations(Figure12). Thelawstatesthat theintensityI of light as
it passesthroughthesolutionsatisesln(I/I
0
) = cx, whereI
0
istheinitial intensityandx isthedistancetraveled
bythelight. Showthat I satisesadifferential equationdI/dx = kI for someconstant k.
23. A sampleof sheepskinparchment discoveredbyarchaeologistshadaC
14
-to-C
12
ratioequal to40%of that foundin
theatmosphere. Approximatelyhowoldistheparchment?
solution Theratioof C
14
toC
12
isRe
0.000121t
= 0.4R so0.000121t = ln(0.4) or t = 7572.65 7600years.
Chauvet Caves In 1994, three French speleologists (geologists specializing in caves) discovered a cave in
southernFrancecontainingprehistoric cavepaintings. A C
14
analysis carriedout by archeologist HeleneValladas
showedthepaintingstobebetween29,700and32,400yearsold, mucholder thananypreviouslyknownhumanart.
Giventhat theC
14
-to-C
12
ratioof theatmosphereisR = 10
12
, what rangeof C
14
-to-C
12
ratiosdidValladasnd
inthecharcoal specimens?
25. A paleontologist discoversremainsof animalsthat appear tohavediedat theonset of theHoloceneiceage, between
10,000and12,000yearsago. What rangeof C
14
-to-C
12
ratiowouldthescientist expect tondintheanimal remains?
solution Thescientist wouldexpect tondC
14
-C
12
ratiosrangingfrom
10
12
e
0.000121(12,000)
2.3410
13
to
10
12
e
0.000121(10,000)
2.9810
13
.
Inversionof Sugar Whencanesugar isdissolvedinwater, it convertstoinvert sugar over aperiodof several
hours. Thepercentagef (t ) of unconvertedcanesugar at timet (inhours) satisesf
= 0.2f . What percentageof

canesugar remainsafter 5hours?After 10hours?
27. ContinuingwithExercise26, supposethat 50gramsof sugar aredissolvedinacontainer of water. After howmany
hourswill 20gramsof invert sugar bepresent?
solution If thereare20gramsof invert sugar present, thenthereare30gramsof unconvertedsugar. Thismeansthat
f (t ) = 60. Solving
100e
0.2t
= 60
for t yields
t =
1
0.2
ln0.6 2.55hours.
Twobacteriacoloniesarecultivatedinalaboratory. Therstcolonyhasadoublingtimeof 2hoursandthesecond
adoublingtimeof 3hours. Initially, therst colonycontains1000bacteriaandthesecondcolony3000bacteria. At
what timet will thesizesof thecoloniesbeequal?
29. MooresLaw In1965, GordonMoorepredictedthat thenumber N of transistorsonamicrochipwouldincrease
exponentially.
(a) Doesthetableof databelowconrmMoorespredictionfor theperiodfrom1971to2000?If so, estimatethegrowth
constant k.
(b) Plot thedatainthetable.
(c) Let N(t ) bethenumber of transistorst yearsafter 1971. FindanapproximateformulaN(t ) Ce
kt
, wheret isthe
number of yearsafter 1971.
(d) EstimatethedoublingtimeinMooresLawfor theperiodfrom1971to2000.
(e) Howmanytransistorswill achipcontainin2015if MooresLawcontinuestohold?
(f) CanMoorehaveexpectedhispredictiontoholdindenitely?
Processor Year No.Transistors
4004 1971 2250
8008 1972 2500
8080 1974 5000
8086 1978 29,000
286 1982 120,000
386processor 1985 275,000
486DX processor 1989 1,180,000
Pentiumprocessor 1993 3,100,000
PentiumII processor 1997 7,500,000
PentiumIII processor 1999 24,000,000
Pentium4processor 2000 42,000,000
Xeonprocessor 2008 1,900,000,000
solution
(a) Yes, thegraphlookslikeanexponential graphespecially towardsthelatter years. Weestimatethegrowthconstant
by setting1971asour startingpoint, soP
0
= 2250. Therefore, P(t ) = 2250e
kt
. In2008, t = 37. Therefore, P(37) =
2250e
37k
= 1,900,000,000, so k =
ln844,444.444
37
0.369. Note: A better estimatecanbefoundby calculatingk for
eachtimeperiodandthenaveragingthek values.
(b)
y
x
110
7
210
7
310
7
410
7
2000 1995 1990 1985 1980
(c) N(t ) = 2250e
0.369t
(d) Thedoublingtimeisln2/0.369 1.88years.
(e) In2015, t = 44years. Therefore, N(44) = 2250e
0.369(44)
2.5310
10
.
(f) No, youcant makeamicrochipsmaller thananatom.
Assumethat inacertaincountry, therateat whichjobsarecreatedisproportional tothenumber of peoplewho
alreadyhavejobs. If thereare15millionjobsatt = 0and15.1millionjobs3monthslater, howmanyjobswill there
beafter 2years?
31. Theonly functions withaconstant doublingtimearetheexponential functions P
0
e
kt
withk > 0. Showthat the
doublingtimeof linear functionf (t ) = at + b at timet
0
ist
0
+ b/a (whichincreaseswitht
0
). Computethedoubling
timesof f (t ) = 3t +12at t
0
= 10andt
0
= 20.
solution Let f (t ) = at + b andsupposef (t
0
) = P
0
. Thetimeat whichthevalueof f will havedoubledis the
solutionof theequation
2P
0
= 2(at
0
+b) = at +b or t = 2t
0
+b/a.
Thus, thetimeit takestodoubleis
t t
0
= 2t
0
+b/a t
0
= t
0
+b/a.
For thefunctionf (t ) = 3t +12, a = 3, b = 12andb/a = 4. Witht
0
= 10, thedoublingtimeisthen14; witht
0
= 20,
thedoublingtimeis24.
Verifythat thehalf-lifeof aquantitythat decaysexponentiallywithdecayconstant k isequal to(ln2)/k.
33. DrugDosingInterval Let y(t ) bethedrug concentration (in mg/kg) in apatients body at timet . The
initial concentrationis y(0) = L. Additional doses that increasetheconcentrationby anamount d areadministeredat
regular timeintervalsof lengthT . Inbetweendoses, y(t ) decaysexponentiallythat is, y
= ky. Findthevalueof T
(intermsof k andd) for whichthetheconcentrationvariesbetweenL andL d asinFigure13.
L
L d
t
y (mcg/ml)
T 2T 3T
Exponential decay
Doseadministered
FIGURE 13 Drugconcentrationwithperiodicdoses.
solution Becausey
= ky andy(0) = L, it followsthat y(t ) = Le

kt
. Wewant y(T ) = L d, thus
Le
kT
= L d or T =
1
k
ln
_
1
d
L
_
.
Exercises 34 and 35: The Gompertzdifferential equation
dy
dt
= ky ln
_
y
M
_
2
(where M and k are constants) was introduced in 1825 by the English mathematician Benjamin Gompertz and is still
used today to model aging and mortality.
Showthat y = Me
ae
kt
satisesEq. (2) for anyconstant a.
35. Tomodel mortality inapopulationof 200laboratory rats, ascientist assumesthat thenumber P(t ) of ratsaliveat
timet (inmonths) satisesEq. (2) withM = 204andk = 0.15month
1
(Figure14). FindP(t ) [notethat P(0) = 200]
anddeterminethepopulationafter 20months.
40 30 20 10
Rat population P(t)
t (mo)
100
200
FIGURE 14
solution ThesolutiontotheGompertz equationwithM = 204andk = 0.15isof theform:
P(t ) = 204e
ae
0.15t
Applyingtheinitial conditionallowsustosolvefor a:
200= 204e
a
200
204
= e
a
ln
_
200
204
_
= a
sothat a 0.02. After t = 20months,
P(20) = 204e
0.02e
0.15(20)
= 136.51,
sothereare136rats.
IsotopesforDating Whichof thefollowingwouldbemostsuitablefordatingextremelyoldrocks: carbon-14
(half-life5570years), lead-210(half-life22.26years), or potassium-49(half-life1.3billionyears)? Explainwhy.
37. Let P = P(t ) beaquantity that obeys anexponential growthlawwithgrowthconstant k. Showthat P increases
m-foldafter aninterval of (lnm)/k years.
solution For m-foldgrowth, P(t ) = mP
0
for somet . SolvingmP
0
= P
0
e
kt
for t , wendt =
lnm
k
.
AverageTimeof Decay Physicists usetheradioactivedecay lawR = R
0
e
kt
to computetheaverageor
mean time M until anatomdecays. Let F(t ) = R/R
0
= e
kt
bethefractionof atomsthat havesurvivedtotimet
without decaying.
(a) Findtheinversefunctiont (F).
(b) Bydenitionof t (F), afraction1/N of atomsdecaysinthetimeinterval
_
t
_
j
N
_
, t
_
j 1
N
__
Usethis to justify theapproximation M
1
N
N
j=1
t
_
j
N
_
. Then argue, by passing to thelimit as N , that
M =
_
1
0
t (F) dF. Strictlyspeaking, thisisanimproper integral becauset (0) isinnite(it takesaninniteamount
of timefor all atomstodecay). Therefore, wedeneM asalimit
M = lim
c0
_
1
c
t (F) dF
(c) Verifytheformula
_
lnx dx = x lnx x bydifferentiationanduseit toshowthat for c > 0,
M = lim
c0
_
1
k
+
1
k
(c lnc c)
_
(d) Verifynumericallythat lim
c0
(c lnc) = 0(wewill provethisusingLHpitalsRuleinSection7.7). Usethisto
showthat M = 1/k.
(e) What isthemeantimetodecayfor radon(withahalf-lifeof 3.825days)?
7.5 Compound Interest and Present Value
1. Whichispreferable: aninterest rateof 12%compoundedquarterly, or aninterest rateof 11%compoundedcontinu-
ously?
solution Toanswer thisquestion, weneedtodeterminetheyearly multiplier associatedwitheachinterest rate. The
multiplier associatedwithaninterest rateof 12%compoundedquarterlyis
_
1+
0.12
4
_
4
1.1255,
whilethemultiplier associatedwithaninterest rateof 11%compoundedcontinuouslyis
e
0.11
1.11627.
Thus, thecompoundedquarterlyrateispreferable.
2. Findtheyearlymultiplier if r = 9%andinterest iscompounded(a) continuouslyand(b) quarterly.
solution Withr = 9%, theyearlymultiplier for continuouslycompoundedinterest is
e
0.09
1.09417,
andtheyearlymultiplier for compoundedquarterlyinterest is
_
1+
0.09
4
_
4
1.09308.
3. ThePV of N dollarsreceivedat timeT is(choosethecorrect answer):
(a) Thevalueat timeT of N dollarsinvestedtoday
(b) Theamount youwouldhavetoinvest todayinorder toreceiveN dollarsat timeT
solution Thecorrect responseis(b): thePV of N dollarsreceivedat timeT istheamount youwouldhavetoinvest
todayinorder toreceiveN dollarsat timeT .
4. Inoneyear, youwill bepaid$1. Will thePV increaseor decreaseif theinterest rategoesup?
solution If theinterest rategoesup, thepresent valueof $1ayear fromnowwill decrease.
5. Xavier expectstoreceiveacheck for $1000oneyear fromtoday. Explainusingtheconcept of PV, whether hewill
behappyor sadtolearnthat theinterest ratehasjust increasedfrom6%to7%.
solution If theinterest rategoes up, thepresent valueof $1,000oneyear fromtoday decreases. Therefore, Xavier
will besadif theinterest ratehasjust increasedfrom6to7%.
Exercises
1. Computethebalanceafter10yearsif $2000isdepositedinanaccountpaying9%interestandinterestiscompounded
(a) quarterly, (b) monthly, and(c) continuously.
solution
(a) P(10) = 2000(1+0.09/4)
4(10)
= $4870.38
(b) P(10) = 2000(1+0.09/12)
12(10)
= $4902.71
(c) P(10) = 2000e
0.09(10)
= $4919.21
Suppose$500 is deposited into an account paying interest at arateof 7%, continuously compounded. Find a
formulafor thevalueof theaccount at timet . What isthevalueof theaccount after 3years?
3. A bankpaysinterest at arateof 5%. What istheyearlymultiplier if interest iscompounded
(a) threetimesayear? (b) continuously?
solution
(a) P(t ) = P
0
_
1+
0.05
3
_
3t
, sotheyearlymultiplier is
_
1+
0.05
3
_
3
1.0508.
(b) P(t ) = P
0
e
0.05t
, sotheyearlymultiplier ise
0.05
1.0513.
S E C T I ON 7.5 Compound Interest and Present Value 399
Howlongwill ittakefor$4000todoubleinvalueif itisdepositedinanaccountbearing7%interest, continuously
compounded?
5. Howmuchmust oneinvest todayinorder toreceive$20,000after 5yearsif interest iscompoundedcontinuouslyat
therater = 9%?
solution Solving20,000= P
0
e
0.09(5)
for P
0
yields
P
0
=
20,000
e
0.45
$12,752.56.
Aninvestmentincreasesinvalueatacontinuouslycompoundedrateof 9%. Howlargemusttheinitial investment
beinorder tobuildupavalueof $50,000over a7-year period?
7. ComputethePV of $5000receivedin3yearsif theinterest rateis(a) 6%and(b) 11%. What isthePV inthesetwo
casesif thesumisinsteadreceivedin5years?
solution In3years:
(a) PV = 5000e
0.06(3)
= $4176.35
(b) PV = 5000e
0.11(3)
= $3594.62
In5years:
(a) PV = 5000e
0.06(5)
= $3704.09
(b) PV = 5000e
0.11(5)
= $2884.75
Isit better toreceive$1000todayor $1300in4years? Consider r = 0.08andr = 0.03.
9. Findtheinterest rater if thePV of $8000tobereceivedin1year is$7300.
solution Solving7300= 8000e
r(1)
for r yields
r = ln
_
7300
8000
_
= 0.0916,
or 9.16%.
A companycanearnadditional protsof $500,000/year for 5yearsbyinvesting$2milliontoupgradeitsfactory.
Istheinvestmentworthwhileif theinterestrateis6%?(Assumethesavingsarereceivedasalumpsumattheendof
eachyear.)
11. A newcomputer systemcosting$25,000will reducelabor costsby$7000/year for 5years.
(a) Isit agoodinvestment if r = 8%?
(b) Howmuchmoneywill thecompanyactuallysave?
solution
(a) Thepresent valueof thereducedlabor costsis
7000(e
0.08
+e
0.16
+e
0.24
+e
0.32
+e
0.4
) = $27,708.50.
Thisismorethanthe$25,000cost of thecomputer system, sothecomputer systemshouldbepurchased.
(b) Thepresent valueof thesavingsis
$27,708.50$25,000= $2708.50.
Afterwinning$25millioninthestatelottery, J essicalearnsthatshewill receiveveyearlypaymentsof $5million
beginningimmediately.
(a) What isthePV of J essicasprizeif r = 6%?
(b) Howmuchmorewouldtheprizebeworthif theentireamount werepaidtoday?
13. Use Eq. (2) to compute the PV of an income streampaying out R(t ) = $5000/year continuously for 10 years,
assumingr = 0.05.
solution PV =
_
10
0
5000e
0.05t
dt = 100,000e
0.05t
10
0
= $39,346.93.
FindthePV of aninvestment that paysout continuouslyat arateof $800/year for 5years, assumingr = 0.08.
15. FindthePV of anincomestreamthat paysout continuouslyat arateR(t ) = $5000e
0.1t
/year for 7years, assuming
r = 0.05.
solution PV =
_
7
0
5000e
0.1t
e
0.05t
dt =
_
7
0
5000e
0.05t
dt = 100,000e
0.05t
7
0
= $41,906.75.
A commercial property generates income at the rate R(t ). Suppose that R(0) = $70,000/year and that R(t )
increases at a continuously compounded rate of 5%. Find the PV of the income generated in the rst 4 years if
r = 6%.
17. Showthat aninvestment that paysout R dollarsper year continuouslyfor T yearshasaPV of R(1e
rT
)/r.
solution Thepresent valueof aninvestment that paysout R dollars/year continuouslyfor T yearsis
PV =
_
T
0
Re
rt
dt.
Let u = rt, du = r dt . Then
PV =
1
r
_
rT
0
Re
u
du =
R
r
e
u
rT
0
=
R
r
(e
rT
1) =
R
r
(1e
rT
).
Explain this statement: If T is very large, then the PV of the income streamdescribed in Exercise 17 is
approximatelyR/r.
19. Supposethat r = 0.06. Usetheresult of Exercise18to estimatethepayout rateR neededto produceanincome
streamwhosePV is$20,000, assumingthat thestreamcontinuesfor alargenumber of years.
solution FromExercise18, PV =
R
r
so20000=
R
0.06
or R = $1200.
Verifybydifferentiation:
_
t e
rt
dt =
e
rt
(1+rt )
r
2
+C
UseEq. (4) tocomputethePV of aninvestment that paysout incomecontinuouslyat arateR(t ) = (5000+1000t )
21. Use Eq. (4) to compute the PV of an investment that pays out income continuously at a rate R(t ) = (5000+
1000t )e
0.02t
dollarsper year for 10years, assumingr = 0.08.
solution
PV =
_
10
0
(5000+1000t )(e
0.02t
)e
0.08t
dt =
_
10
0
5000e
0.06t
dt +
_
10
0
1000t e
0.06t
dt
=
5000
0.06
(e
0.06(10)
1) 1000
_
e
0.06(10)
(1+0.06(10))
(0.06)
2
_
+1000
1
(0.06)
2
= 37,599.03243,916.28+277,777.78 $71,460.53.
BankersRuleof70 If youearnaninterestrateof R percent, continuouslycompounded, yourmoneydoubles
afterapproximately70/R years. Forexample, atR = 5%, yourmoneydoublesafter70/5or14years. Usetheconcept
of doublingtimetojustifytheBankersRule. (Note: Sometimes, therule72/R isused. It islessaccuratebut easier
toapplybecause72isdivisiblebymorenumbersthan70.)
In Exercises 2326, calculate the limit.
23. lim
n
_
1+
1
n
_
6n
solution
lim
n
_
1+
1
n
_
6n
= lim
n
__
1+
1
n
_
n
_
6
=
_
lim
n
_
1+
1
n
_
n
_
6
= e
6
.
lim
n
_
1+
3
n
_
n
25. lim
n
_
1+
3
n
_
2n
solution Let t = n/3. Thenn = 3t and
lim
n
_
1+
3
n
_
2n
= lim
t
_
1+
1
t
_
6t
= lim
t
_
_
1+
1
t
_
t
_
6
=
_
lim
t
_
1+
1
t
_
t
_
6
= e
6
.
lim
n
_
1+
1
4n
_
12n
27. Modifytheproof of therelatione = lim
n
_
1+
1
n
_
n
giveninthetext toprovee
x
= lim
n
_
1+
x
n
_
n
. Hint: Express
ln(1+xn
1
) asanintegral andestimateaboveandbelowbyrectangles.
solution Start byexpressing
ln
_
1+
x
n
_
=
_
1+x/n
1
dt
t
.
Followingtheproof inthetext, wenotethat
x
n +x
ln
_
1+
x
n
_

x
n
providedx > 0, while
x
n
ln
_
1+
x
n
_

x
n +x
whenx < 0. Multiplyingbothsetsof inequalitiesbyn andpassingtothelimitasn , thesqueezetheoremguarantees
that
lim
n
_
ln
_
1+
x
n
__
n
= x.
Finally,
lim
n
_
1+
x
n
_
n
= e
x
.
Provethat, for n > 0,
_
1+
1
n
_
n
e
_
1+
1
n
_
n+1
Hint: TakelogarithmsanduseEq. (3).
29. A bank pays interest at therater, compoundedM times yearly. Theeffectiveinterest rater
e
is therateat which
interest, if compoundedannually, wouldhavetobepaidtoproducethesameyearlyreturn.
(a) Findr
e
if r = 9%compoundedmonthly.
(b) Showthat r
e
= (1+r/M)
M
1andthat r
e
= e
r
1if interest iscompoundedcontinuously.
(c) Findr
e
if r = 11%compoundedcontinuously.
(d) Findtherater that, compoundedweekly, wouldyieldaneffectiverateof 20%.
S E C T I ON 7.6 Models Involving y
= k ( y b) 401
solution
(a) Compoundedmonthly, P(t ) = P
0
(1+r/12)
12t
. Bythedenitionof r
e
,
P
0
(1+0.09/12)
12t
= P
0
(1+r
e
)
t
so
(1+0.09/12)
12t
= (1+r
e
)
t
or r
e
= (1+0.09/12)
12
1= 0.0938,
or 9.38%
(b) Ingeneral,
P
0
(1+r/M)
Mt
= P
0
(1+r
e
)
t
,
so(1+r/M)
Mt
= (1+r
e
)
t
orr
e
= (1+r/M)
M
1. If interestiscompoundedcontinuously, thenP
0
e
rt
= P
0
(1+r
e
)
t
soe
rt
= (1+r
e
)
t
or r
e
= e
r
1.
(c) Usingpart (b), r
e
= e
0.11
1 0.1163or 11.63%.
(d) Solving
0.20=
_
1+
r
52
_
52
1
for r yieldsr = 52(1.2
1/52
1) = 0.1826or 18.26%.
7.6 Models Involving y
= k ( y b)
1. Writedownasolutiontoy
= 4(y 5) that tendstoast .

solution Thegeneral solution is y(t ) = 5+ Ce
4t
for any constant C; thus thesolution tends to as t
whenever C < 0. Onespecicexampleisy(t ) = 5e
4t
.
2. Doesy
= 4(y 5) haveasolutionthat tendstoast ?

solution Thegeneral solutionis y(t ) = 5+ Ce
4t
for any constant C. As t , y(t ) 5. Thus, thereis no
solutionof y
= 4(y 5) that tendstoast .

3. Trueor false? If k > 0, thenall solutionsof y
= k(y b) approachthesamelimit ast .

solution True. Thegeneral solution of y
= k(y b) is y(t ) = b + Ce
kt
for any constant C. If k > 0, then
y(t ) b ast .
4. Asanobject cools, itsrateof coolingslows. ExplainhowthisfollowsfromNewtonsLawof Cooling.
solution NewtonsLawof Coolingstatesthat y
= k(y T
0
) wherey(t ) isthetemperatureandT
0
istheambient
temperature. Thusasy(t ) getscloser toT
0
, y
(t ), therateof cooling, getssmaller andtherateof coolingslows.

Exercises
1. Findthegeneral solutionof
y
= 2(y 10)
Thenndthetwosolutionssatisfyingy(0) = 25andy(0) = 5, andsketchtheir graphs.
solution The general solution of y
= 2(y 10) is y(t ) = 10+ Ce

2t
for any constant C. If y(0) = 25, then
10+ C = 25, or C = 15; therefore, y(t ) = 10+ 15e
2t
. Ontheother hand, if y(0) = 5, then10+ C = 5, or C = 5;
therefore, y(t ) = 105e
2t
. Graphsof thesetwofunctionsaregivenbelow.
y
x
200
400
600
800
1.5
y(0) = 25
1 0.5
x
y
50
100
150
200
250
y(0) = 5
1.5 1 0.5
Verifydirectlythat y = 12+Ce
3t
satises
y
= 3(y 12) for all C

Thenndthetwosolutionssatisfyingy(0) = 20andy(0) = 0, andsketchtheir graphs.
3. Solvey
= 4y +24subject toy(0) = 5.
solution Rewrite
y
= 4y +24 as
1
4y +24
dy = 1dt
Integratinggives
1
4
ln|4y +24| = t +C
ln|4y +24| = 4t +C
4y +24= e
4t +C
y = Ae
4t
6
whereA = e
C
/4isanyconstant. Sincey(0) = 5wehave5= A6sothatA = 11, andthesolutionisy = 11e
4t
6.
Solvey
+6y = 12subject toy(2) = 10.

In Exercises 512, use Newtons Law of Cooling.
5. A hot anvil withcoolingconstant k = 0.02s
1
is submergedinalargepool of water whosetemperatureis 10
C.
Let y(t ) betheanvilstemperaturet secondslater.
(a) What isthedifferential equationsatisedbyy(t )?
(b) Findaformulafor y(t ), assumingtheobjectsinitial temperatureis100
C.
(c) Howlongdoesit taketheobject tocool downto20
?
solution
(a) ByNewtonsLawof Cooling, thedifferential equationis
y
= 0.02(y 10)
(b) Separatingvariablesgives
1
y 10
dy = 0.02dt
Integratetoget
ln|y 10| = 0.02t +C
y 10= e
0.02t +C
y = 10+Ae
0.02t
whereA = e
C
isaconstant. Sincetheinitial temperatureis100
C, wehavey(0) = 100 = 10+ A sothat A = 90,

andy = 10+90e
0.02t
.
(c) Wemust ndthevalueof t suchthat y(t ) = 20, soweneedtosolve20= 10+90e
0.02t
. Thus
10= 90e
0.02t
1
9
= e
0.02t
ln9= 0.02t t = 50ln9 109.86s
Franksautomobileenginerunsat100
C. Onadaywhentheoutsidetemperatureis21
C, heturnsoff theignition
andnotesthat veminuteslater, theenginehascooledto70
C.
(a) Determinetheenginescoolingconstant k.
(b) What istheformulafor y(t )?
(c) Whenwill theenginecool to40
C?
7. At 10:30am, detectives discover adeadbody inaroomandmeasureits temperatureat 26
C. Onehour later, the

bodystemperaturehaddroppedto24.8
C. Determinethetimeof death(whenthebodytemperaturewasanormal 37
C),
assumingthat thetemperatureintheroomwasheldconstant at 20
C.
solution Let t = 0bethetimewhenthepersondied, andlet t
0
denote10:30am. Thedifferential equationsatised
bythebodytemperature, y(t ), is
y
= k(y 20)
byNewtonsLawof Cooling. Separatingvariablesgives
1
y 20
dy = k dt . Integratetoget
ln|y 20| = kt +C
y 20= e
kt +C
y = 20+Ae
kt
= k ( y b) 403
whereA = e
C
isaconstant. Sincenormal body temperatureis37
C, wehavey(0) = 37= 20+ A sothat A = 17.

Todeterminek, notethat
26= 20+17e
kt
0
and 24.8= 20+17e
k(t
0
+1)
kt
0
= ln
6
17
kt
0
+k = ln
4.8
17
Subtractingtheseequationsgives
k = ln
6
17
ln
4.8
17
= ln
6
4.8
0.223
Wethushave
y = 20+17e
0.223t
astheequationfor thebodytemperatureat timet . Sincey(t
0
) = 26, wehave
26= 20+17e
0.223t
0
e
0.223t
0
=
6
17
t
0
=
1
0.223
ln
6
17
4.667h
sothat thetimeof deathwasapproximately4hoursand40minutesago.
A cupof coffeewithcoolingconstant k = 0.09min
1
isplacedinaroomat temperature20
C.
(a) Howfast isthecoffeecooling(indegreesper minute) whenitstemperatureisT = 80
C?
(b) UsetheLinearApproximationtoestimatethechangeintemperatureover thenext 6swhenT = 80
C.
(c) If thecoffeeisservedat 90
C, howlongwill it taketoreachanoptimal drinkingtemperatureof 65
C?
9. A cold metal bar at 30
C is submerged in apool maintained at atemperatureof 40
C. Half aminutelater, the

temperatureof thebar is20
C. Howlongwill it takefor thebar toattainatemperatureof 30
C?
solution WithT
0
= 40
C, thetemperatureof thebar isgivenby F(t ) = 40+ Ce

kt
for someconstantsC andk.
Fromtheinitial condition, F(0) = 40+C = 30, soC = 70. After 30seconds, F(30) = 4070e
30k
= 20, so
k =
1
30
ln
_
20
70
_
0.0418seconds
1
.
Toattainatemperatureof 30
C wemust solve4070e
0.0418t
= 30for t . Thisyields
t =
ln
_
10
70
_
0.0418
46.55seconds.
Whenahot object is placedinawater bathwhosetemperatureis 25
C, it cools from100to50
C in150s. In
another bath, thesamecoolingoccursin120s. Findthetemperatureof thesecondbath.
11. Objects A and B areplaced in awarmbath at temperatureT
0
= 40
C. Object A has initial temperature

20
C andcoolingconstant k = 0.004s
1
. Object B hasinitial temperature0
C andcoolingconstant k = 0.002s
1
.
Plotthetemperaturesof A andB for 0 t 1000. After howmanysecondswill theobjectshavethesametemperature?
solution WithT
0
= 40
C, thetemperatureof A andB aregivenby

A(t ) = 40+C
A
e
0.004t
B(t ) = 40+C
B
e
0.002t
SinceA(0) = 20andB(0) = 0, wehave
A(t ) = 4060e
0.004t
B(t ) = 4040e
0.002t
Thetwoobjectswill havethesametemperaturewhenever A(t ) = B(t ), sowemust solve
4060e
0.004t
= 4040e
0.002t
3e
0.004t
= 2e
0.002t
Takelogstoget
0.004t +ln3= 0.002t +ln2 t =
ln3ln2
0.002
202.7s
or about 3minutes22seconds.
20
10
0
100 200 300
40 40e
0.002t
40 60e
0.004t
400
10
20
y
t
InNewtons Lawof Cooling, theconstant = 1/k is calledthecharacteristic time. Showthat is thetime
requiredfor thetemperaturedifference(y T
0
) todecreasebythefactor e
1
0.37. For example, if y(0) = 100
C
andT
0
= 0
C, thentheobject coolsto100/e 37
C intime, to100/e
2
13.5
C intime2, andsoon.
In Exercises 1316, use Eq. (3) as a model for free-fall with air resistance.
13. A 60-kg skydiver jumps out of an airplane. What is her terminal velocity, in meters per second, assuming that
k = 10kg/sfor free-fall (noparachute)?
solution Thefree-fall terminal velocityis
gm
k
=
9.8(60)
10
= 58.8m/s.
Findtheterminal velocity of askydiver of weight w = 192lbif k = 1.2lb-s/ft. Howlongdoes it takehimto
reachhalf of histerminal velocityif hisinitial velocityiszero?Massandweight arerelatedbyw = mg, andEq. (3)
becomesv
= (kg/w)(v +w/k) withg = 32ft/s

2
.
15. A80-kgskydiverjumpsoutof anairplane(withzeroinitial velocity).Assumethatk = 12kg/swithaclosedparachute
andk = 70kg/swithanopenparachute. What istheskydiversvelocityat t = 25sif theparachuteopensafter 20sof
freefall?
solution Werstcomputetheskydiversvelocityafter 20sof freefall, thenusethatastheinitial velocitytocalculate
her velocityafter anadditional 5sof restrainedfall. Wehavem = 80andg = 9.8; for freefall, k = 12, so
k
m
=
12
80
= 0.15,
mg
k
=
80 9.8
12
65.33
Thegeneral solutionisthusv(t ) = 65.33+Ce
0.15t
. Sincev(0) = 0, wehaveC = 65.33, sothat
v(t ) = 65.33(1e
0.15t
)
After 20sof freefall, thediversvelocityisthus
v(20) = 65.33(1e
0.1520
) 62.08m/s
Oncetheparachuteopens, k = 70, so
k
m
=
70
80
= 0.875,
mg
k
=
80 9.8
70
= 11.2
sothat thegeneral solutionfor therestrainedfall model is v
r
(t ) = 11.2+ Ce
0.875t
. Herev
r
(0) = 62.08, sothat
C = 11.2 62.08 = 50.88andv
r
(t ) = 11.20 50.88e
0.875t
. After 5additional seconds, thedivers velocity is
therefore
v
r
(5) = 11.2050.88e
0.8755
11.84m/s
Doesaheavier or alighter skydiver reachterminal velocityfaster?
17. A continuousannuitywithwithdrawal rateN = $5000/year andinterest rater = 5%isfundedbyaninitial deposit
of P
0
= $50,000.
(a) What isthebalanceintheannuityafter 10years?
(b) Whenwill theannuityrunout of funds?
solution
(a) FromEquation7, thevalueof theannuityisgivenby
P(t ) =
5000
0.05
+Ce
0.05t
= 100,000+Ce
0.05t
for someconstantC. SinceP(0) = 50,000, wehaveC = 50,000andP(t ) = 100,00050,000e
0.05t
.After tenyears,
then, thebalanceintheannuityis
P(10) = 100,00050,000e
0.0510
= 100,00050,000e
0.5
$17,563.94
(b) Theannuitywill runout of fundswhenP(t ) = 0:
0= 100,00050,000e
0.05t
e
0.05t
= 2 t =
ln2
0.05
13.86
Theannuitywill runout of fundsafter approximately13years10months.
Showthat acontinuous annuity withwithdrawal rateN = $5000/year andinterest rater = 8%, fundedby an
initial deposit of P
0
= $75,000, never runsout of money.
19. Findtheminimuminitial deposit P
0
that will allowanannuitytopayout $6000/year indenitelyif it earnsinterest
at arateof 5%.
solution Let P(t ) denotethebalanceof theannuityat timet measuredinyears. Then
P(t ) =
N
r
+Ce
rt
=
6000
0.05
+Ce
0.05t
= 120,000+Ce
0.05t
for some constant C. To fund the annuity indenitely, we must have C 0. If the initial deposit is P
0
, then P
0
=
120,000+C andC = P
0
120,000. Thus, tofundtheannuityindenitely, wemust haveP
0
$120,000.
= k ( y b) 405
Findtheminimuminitial deposit P
0
necessarytofundanannuityfor 20yearsif withdrawalsaremadeat arate
of $10,000/year andinterest isearnedat arateof 7%.
21. Aninitial deposit of 100,000eurosareplacedinanannuity withaFrenchbank. What istheminimuminterest rate
theannuitymust earntoallowwithdrawalsat arateof 8000euros/year tocontinueindenitely?
solution Let P(t ) denotethebalanceof theannuityat timet measuredinyears. Then
P(t ) =
N
r
+Ce
rt
=
8000
r
+Ce
rt
for some constant C. To fund the annuity indenitely, we need C 0. If the initial deposit is 100,000 euros, then
100,000=
8000
r
+C andC = 100,000
8000
r
. Thus, tofundtheannuityindenitely, weneed100,000
8000
r
0, or
r 0.08. Thebankmust payat least 8%.
Showthat acontinuous annuity never runs out of money if theinitial balanceis greater thanor equal to N/r,
whereN isthewithdrawal rateandr theinterest rate.
23. Samborrows$10,000fromabankat aninterest rateof 9%andpaysbacktheloancontinuouslyat arateof
N dollarsper year. Let P(t ) denotetheamount still owedat timet .
(a) ExplainwhyP(t ) satisesthedifferential equation
y
= 0.09y N
(b) Howlongwill it takeSamtopaybacktheloanif N = $1200?
(c) Will theloanever bepaidbackif N = $800?
solution
(a)
Rateof Changeof Loan= (Amount still owed)(Interest rate) (Paybackrate)
= P(t ) r N = r
_
P
N
r
_
.
Therefore, if y = P(t ),
y
= r
_
y
N
r
_
= ry N
(b) Fromthedifferential equationderivedinpart (a), weknowthat P(t ) =
N
r
+ Ce
rt
= 13,333.33+ Ce
0.09t
. Since
$10,000 was initially borrowed, P(0) = 13,333.33+ C = 10,000, and C = 3333.33. Theloan is paid off when
P(t ) = 13,333.333333.33e
0.09t
= 0. Thisyields
t =
1
0.09
ln
_
13,333.33
3333.33
_
15.4years.
(c) If the annual rate of payment is $800, then P(t ) = 800/0.09+ Ce
0.09t
= 8888.89+ Ce
0.09t
. With P(0) =
8888.89+ C = 10,000, it follows that C = 1111.11. SinceC > 0ande
0.09t
as t , P(t ) , andthe
loanwill never bepaidback.
April borrows$18,000at aninterest rateof 5%topurchaseanewautomobile. At what rate(indollarsper year)
mustshepaybacktheloan, if theloanmustbepaidoff in5years?Hint: Setupthedifferential equationasinExercise
23).
25. Let N(t ) bethefraction of thepopulation who haveheard agiven pieceof news t hours after its initial release.
Accordingtoonemodel, therateN
(t ) at whichthenewsspreadsisequal tok timesthefractionof thepopulationthat

hasnot yet heardthenews, for someconstant k > 0.
(a) Determinethedifferential equationsatisedbyN(t ).
(b) Findthesolutionof thisdifferential equationwiththeinitial conditionN(0) = 0intermsof k.
(c) Supposethat half of thepopulationis awareof anearthquake8hours after it occurs. Usethemodel tocalculatek
andestimatethepercentagethat will knowabout theearthquake12hoursafter it occurs.
solution
(a) N
(t ) = k(1N(t )) = k(N(t ) 1).

(b) Thegeneral solutionof thedifferential equationfrompart(a) isN(t ) = 1+Ce
kt
. Theinitial conditiondetermines
thevalueof C: N(0) = 1+C = 0soC = 1. Thus, N(t ) = 1e
kt
.
(c) Knowingthat N(8) = 1e
8k
=
1
2
, wendthat
k =
1
8
ln
_
1
2
_
0.0866hours
1
.
Withthevalueof k determined, weestimatethat
N(12) = 1e
0.0866(12)
0.6463= 64.63%
of thepopulationwill knowabout theearthquakeafter 12hours.
CurrentinaCircuit WhenthecircuitinFigure6(whichconsistsof abatteryof V volts, aresistor of R ohms,
andaninductor of L henries) isconnected, thecurrent I (t ) owinginthecircuit satises
L
dI
dt
+RI = V
27. Showthatthecoolingconstantof anobjectcanbedeterminedfromtwotemperaturereadingsy(t
1
) andy(t
2
) attimes
t
1
= t
2
bytheformula
k =
1
t
1
t
2
ln
_
y(t
2
) T
0
y(t
1
) T
0
_
solution Weknowthaty(t
1
) = T
0
+Ce
kt
1
andy(t
2
) = T
0
+Ce
kt
2
.Thus, y(t
1
) T
0
= Ce
kt
1
andy(t
2
) T
0
=
Ce
kt
2
. Dividingthelatter equationbytheformer yields
e
kt
2
+kt
1
=
y(t
2
) T
0
y(t
1
) T
0
,
sothat
k(t
1
t
2
) = ln
_
y(t
2
) T
0
y(t
1
) T
0
_
and k =
1
t
1
t
2
ln
_
y(t
2
) T
0
y(t
1
) T
0
_
.
ShowthatbyNewtonsLawof Cooling, thetimerequiredtocool anobjectfromtemperatureA totemperatureB
is
t =
1
k
ln
_
A T
0
B T
0
_
whereT
0
istheambient temperature.
29. Air Resistance A projectileof massm = 1travelsstraight upfromgroundlevel withinitial velocityv
0
. Suppose
that thevelocityv satisesv
= g kv.
(a) Findaformulafor v(t ).
(b) Showthat theprojectilesheight h(t ) isgivenby
h(t ) = C(1e
kt
)
g
k
t
whereC = k
2
(g +kv
0
).
(c) Showthat theprojectilereachesitsmaximumheight at timet
max
= k
1
ln(1+kv
0
/g).
(d) Intheabsenceof air resistance, themaximumheight isreachedat timet = v
0
/g. Inviewof this, explainwhy we
shouldexpect that
lim
k0
ln(1+
kv
0
g
)
k
=
v
0
g
8
(e) VerifyEq. (8). Hint: UseTheorem1inSection7.5toshowthat lim
k0
_
1+
kv
0
g
_
1/k
= e
v
0
/g
.
solution
(a) Sincev
= g kv = k
_
v
g
k
_
itfollowsthatv(t ) =
g
k
+Be
kt
forsomeconstantB.Theinitial condition
v(0) = v
0
determinesB: v
0
=
g
k
+B, soB = v
0
+
g
k
. Thus,
v(t ) =
g
k
+
_
v
0
+
g
k
_
e
kt
.
(b) v(t ) = h
(t ) so
h(t ) =
_
_
g
k
+
_
v
0
+
g
k
_
e
kt
_
dt =
g
k
t
1
k
_
v
0
+
g
k
_
e
kt
+D.
Theinitial conditionh(0) = 0determines
D =
1
k
_
v
0
+
g
k
_
=
1
k
2
(v
0
k +g).
Let C =
1
k
2
(v
0
k +g). Then
h(t ) = C(1e
kt
)
g
k
t.
(c) Theprojectilereachesitsmaximumheight whenv(t ) = 0. Thisoccurswhen
g
k
+
_
v
0
+
g
k
_
e
kt
= 0,
or
t =
1
k
ln
_
g
kv
0
+g
_
=
1
k
ln
_
1+
kv
0
g
_
.
S E C T I ON 7.7 LHpitals Rule 407
(d) Recall that k is theproportionality constant for theforcedueto air resistance. Thus, as k 0, theeffect of air
resistancedisappears. Weshouldthereforeexpectthat, ask 0, thetimeatwhichthemaximumheightisachievedfrom
part (c) shouldapproachv
0
/g. Inother words, weshouldexpect
lim
k0
1
k
ln
_
1+
kv
0
g
_
=
v
0
g
.
(e) Recall that
e
x
= lim
n
_
1+
x
n
_
n
.
If wesubstitutex = v
0
/g andk = 1/n, wend
e
v
0
/g
= lim
k0
_
1+
v
0
k
g
_
1/k
.
Then
lim
k0
1
k
ln
_
1+
kv
0
g
_
= lim
k0
ln
_
1+
v
0
k
g
_
1/k
= ln
_
lim
k0
_
1+
v
0
k
g
_
1/k
_
= ln(e
v
0
/g
) =
v
0
g
.
7.7 LH opitals Rule
1. What iswrongwithapplyingLHpitalsRuleto lim
x0
x
2
2x
3x 2
?
solution Asx 0,
x
2
2x
3x 2
isnot of theform
0
0
or

, soLHpitalsRulecannot beused.
2. DoesLHpitalsRuleapplyto lim
xa
f (x)g(x) if f (x) andg(x) bothapproachasx a?
solution No. LHpitalsRuleonlyappliestolimitsof theform
0
0
or

.
Exercises
In Exercises 110, use LHpitals Rule to evaluate the limit, or state that LHpitals Rule does not apply.
1. lim
x3
2x
2
5x 3
x 4
solution Becausethequotient isnot indeterminateat x = 3,
2x
2
5x 3
x 4
x=3
=
18153
34
=
0
1
,
LHpitalsRuledoesnot apply.
lim
x5
x
2
25
54x x
2
3. lim
x4
x
3
64
x
2
+16
x
3
64
x
2
+16
x=4
=
6464
16+16
=
0
32
,
lim
x1
x
4
+2x +1
x
5
2x 1
5. lim
x9
x
1/2
+x 6
x
3/2
27
x
1/2
+x 6
x
3/2
27
x=9
=
3+96
2727
=
6
0
,
lim
x3
x +12
x
3
7x 6
7. lim
x0
sin4x
x
2
+3x +1
sin4x
x
2
+3x +1
x=0
=
0
0+0+1
=
0
1
,
lim
x0
x
3
sinx x
9. lim
x0
cos2x 1
sin5x
solution Thefunctionscos2x 1andsin5x aredifferentiable, but thequotient isindeterminateat x = 0,
cos2x 1
sin5x
x=0
=
11
0
=
0
0
,
soLHpitalsRuleapplies. Wend
lim
x0
cos2x 1
sin5x
= lim
x0
2sin2x
5cos5x
=
0
5
= 0.
lim
x0
cosx sin
2
x
sinx
In Exercises 1116, show that LHpitals Rule is applicable to the limit as x and evaluate.
11. lim
x
9x +4
32x
solution Asx , thequotient
9x +4
32x
isof theform
, soLHpitalsRuleapplies. Wend
lim
x
9x +4
32x
= lim
x
9
2
=
9
2
.
lim
x
x sin
1
x
13. lim
x
lnx
x
1/2
lnx
x
1/2
isof theform
, soLHpitalsRuleapplies. Wend
lim
x
lnx
x
1/2
= lim
x
1
x
1
2
x
1/2
= lim
x
1
2x
1/2
= 0.
lim
x
x
e
x
15. lim
x
ln(x
4
+1)
x
ln(x
4
+1)
x
isof theform
, soLHpitalsRuleapplies. Here, weuseLHpitals

Ruletwicetond
lim
x
ln(x
4
+1)
x
= lim
x
4x
3
x
4
+1
1
= lim
x
12x
2
4x
3
= lim
x
3
x
= 0.
lim
x
x
2
e
x
In Exercises 1750, evaluate the limit.
17. lim
x1
8+x 3x
1/3
x
2
3x +2
solution lim
x1
8+x 3x
1/3
x
2
3x +2
= lim
x1
1
2
(8+x)
1/2
x
2/3
2x 3
=
1
6
1
1
=
5
6
.
lim
x4
_
1
x 2
4
x 4
_
19. lim
x
3x 2
15x
solution lim
x
3x 2
15x
= lim
x
3
5
=
3
5
.
lim
x
x
2/3
+3x
x
5/3
x
21. lim
x
7x
2
+4x
93x
2
solution lim
x
7x
2
+4x
93x
2
= lim
x
14x +4
6x
= lim
x
14
6
=
7
3
.
lim
x
3x
3
+4x
2
4x
3
7
23. lim
x1
(1+3x)
1/2
2
(1+7x)
1/3
2
solution ApplyLHpitalsRuleonce:
lim
x1
(1+3x)
1/2
2
(1+7x)
1/3
2
= lim
x1
3
2
(1+3x)
1/2
7
3
(1+7x)
2/3
=
(
3
2
)
1
2
(
7
3
)(
1
4
)
=
9
7
lim
x8
x
5/3
2x 16
x
1/3
2
25. lim
x0
sin2x
sin7x
solution lim
x0
sin2x
sin7x
= lim
x0
2cos2x
7cos7x
=
2
7
.
lim
x/2
tan4x
tan5x
27. lim
x0
tanx
x
solution lim
x0
tanx
x
= lim
x0
sec
2
x
1
= 1.
lim
x0
_
cotx
1
x
_
29. lim
x0
sinx x cosx
x sinx
solution
lim
x0
sinx x cosx
x sinx
= lim
x0
x sinx
1cosx
= lim
x0
sinx +x cosx
sinx
= lim
x0
cosx +cosx x sinx
cosx
= 2.
lim
x/2
_
x

2
_
tanx
31. lim
x0
cos(x +

2
)
sinx
solution lim
x0
cos(x +

2
)
sinx
= lim
x0
sin(x +

2
)
cosx
= 1.
lim
x0
x
2
1cosx
33. lim
x/2
cosx
sin(2x)
solution lim
x/2
cosx
sin(2x)
= lim
x/2
sinx
2cos(2x)
=
1
2
.
lim
x0
_
1
x
2
csc
2
x
_ 35. lim
x/2
(secx tanx)
solution
lim
x
2
( secx tanx) = lim
x
2
_
1
cosx
sinx
cosx
_
= lim
x
2
_
1sinx
cosx
_
= lim
x
2
_
cosx
sinx
_
= 0.
lim
x2
e
x
2
e
4
x 2
37. lim
x1
tan
_
x
2
_
lnx
solution lim
x1
tan
_
x
2
_
lnx = lim
x1
lnx
cot(
x
2
)
= lim
x1
1
x
2
csc
2
(
x
2
)
= lim
x1
2
x
sin
2
_
2
x
_
=
2
.
lim
x1
x(lnx 1) +1
(x 1) lnx
39. lim
x0
e
x
1
sinx
solution lim
x0
e
x
1
sinx
= lim
x0
e
x
cosx
= 1.
lim
x1
e
x
e
lnx
41. lim
x0
e
2x
1x
x
2
solution lim
x0
e
2x
1x
x
2
= lim
x0
2e
2x
1
2x
whichdoesnot exist.
lim
x
e
2x
1x
x
2
43. lim
t 0+
(sint )(lnt )
solution
lim
t 0+
(sint )(lnt ) = lim
t 0+
lnt
csct
= lim
t 0+
1
t
csct cott
= lim
t 0+
sin
2
t
t cost
= lim
t 0+
2sint cost
cost t sint
= 0.
lim
x
e
x
(x
3
x
2
+9)
45. lim
x0
a
x
1
x
(a > 0)
solution lim
x0
a
x
1
x
= lim
x0
lna a
x
1
= lna.
lim
x
x
1/x
2
47. lim
x1
(1+lnx)
1/(x1)
solution lim
x1
ln(1+lnx)
1/(x1)
= lim
x1
ln(1+lnx)
x 1
= lim
x1
1
x(1+lnx)
= 1. Hence,
lim
x1
(1+lnx)
1/(x1)
= lim
x1
e
(1+lnx)
1/(x1)
= e.
lim
x0+
x
sinx
49. lim
x0
(cosx)
3/x
2
solution
lim
x0
ln(cosx)
3/x
2
= lim
x0
3lncosx
x
2
= lim
x0
3tanx
2x
= lim
x0
3sec
2
x
2
=
3
2
.
Hence, lim
x0
(cosx)
3/x
2
= e
3/2
.
lim
x
_
x
x +1
_
x 51. Evaluate lim
x/2
cosmx
cosnx
, wherem, n = 0areintegers.
solution Supposem andn areeven. Thenthereexist integersk andl suchthat m = 2k andn = 2l and
lim
x/2
cosmx
cosnx
=
cosk
cosl
= (1)
kl
.
Now, supposem isevenandn isodd. Then
lim
x/2
cosmx
cosnx
doesnot exist (fromonesidethelimit tendstoward, whilefromtheother sidethelimit tendstoward+). Third,
supposem isoddandn iseven. Then
lim
x/2
cosmx
cosnx
= 0.
Finally, supposem andn areodd. Thisistheonlycasewhenthelimit isindeterminate. Thenthereexist integersk andl
suchthat m = 2k +1, n = 2l +1and, byLHpitalsRule,
lim
x/2
cosmx
cosnx
= lim
x/2
msinmx
n sinnx
= (1)
kl
m
n
.
Tosummarize,
lim
x/2
cosmx
cosnx
=
(1)
(mn)/2
, m, n even
doesnot exist, m even, n odd
0 m odd, n even
(1)
(mn)/2m
n
, m, n odd
Evaluate lim
x1
x
m
1
x
n
1
for anynumbersm, n = 0.
53. Provethefollowinglimit formulafor e:
e = lim
x0
(1+x)
1/x
Thenndavalueof x suchthat |(1+x)
1/x
e| 0.001.
solution UsingLHpitalsRule,
lim
x0
ln(1+x)
x
= lim
x0
1
1+x
1
= 1.
Thus,
lim
x0
ln
_
(1+x)
1/x
_
= lim
x0
1
x
ln(1+x) = lim
x0
ln(1+x)
x
= 1,
and lim
x0
(1+x)
1/x
= e
1
= e. For x = 0.0005,
(1+x)
1/x
e
= |(1.0005)
2000
e| 6.7910
4
< 0.001.
CanLHpitalsRulebeappliedto lim
x0+
x
sin(1/x)
? Doesagraphical or numerical investigationsuggest that
thelimit exists?
55. Let f (x) = x
1/x
for x > 0.
(a) Calculate lim
x0+
f (x) and lim
x
f (x).
(b) Findthemaximumvalueof f (x), anddeterminetheintervalsonwhichf (x) isincreasingor decreasing.
solution
(a) Let f (x) = x
1/x
. Notethat lim
x0+
x
1/x
isnot indeterminate. Asx 0+, thebaseof thefunctiontendstoward
0andtheexponenttendstoward+. Bothof thesefactorsforcex
1/x
toward0. Thus, lim
x0+
f (x) = 0. Ontheother
hand, lim
x
f (x) isindeterminate. Wecalculatethislimit asfollows:
lim
x
lnf (x) = lim
x
lnx
x
= lim
x
1
x
= 0,
solim
x
f (x) = e
0
= 1.
(b) Again, let f (x) = x
1/x
, sothat lnf (x) =
1
x
lnx. Tondthederivativef
, weapplythederivativetobothsides:
d
dx
lnf (x) =
d
dx
_
1
x
lnx
_
1
f (x)
f
(x) =
lnx
x
2
+
1
x
2
f
(x) = f (x)
_
lnx
x
2
+
1
x
2
_
=
x
1/x
x
2
(1lnx)
Thus, f is increasingfor 0 < x < e, is decreasingfor x > e andhas amaximumat x = e. Themaximumvalueis
f (e) = e
1/e
1.444668.
(a) Usetheresults of Exercise55to provethat x
1/x
= c has auniquesolutionif 0 < c 1or c = e
1/e
, two
solutionsif 1< c < e
1/e
, andnosolutionsif c > e
1/e
.
(b) Plot thegraphof f (x) = x
1/x
andverifythat it conrmstheconclusionsof (a).
57. Determinewhether f g or g f (or neither) for thefunctionsf (x) = log
10
x andg(x) = lnx.
solution Because
lim
x
f (x)
g(x)
= lim
x
log
10
x
lnx
= lim
x
lnx
ln10
lnx
=
1
ln10
,
neither f g or g f issatised.
Showthat (lnx)
2

x and(lnx)
4
x
1/10
.
59. J ustasexponential functionsaredistinguishedbytheirrapidrateof increase, thelogarithmfunctionsgrowparticularly
slowly. Showthat lnx x
a
for all a > 0.
solution UsingLHpitalsRule:
lim
x
lnx
x
a
= lim
x
x
1
ax
a1
= lim
x
1
a
x
a
= 0;
hence, lnx (x
a
).
Showthat (lnx)
N
x
a
for all N andall a > 0.
61. Determinewhether
x e
lnx
or e
lnx

x. Hint: Usethesubstitutionu = lnx insteadof LHpitalsRule.
solution Let u = lnx, thenx = e
u
, andasx , u . So
lim
x
e
lnx
x
= lim
u
e
u
e
u/2
= lim
u
e
u
u
2.
Weneedtoexamine lim
u
(
u
u
2
). Since
lim
u
u/2
u
= lim
u
1
2
1
2
u
= lim
u
u = ,
u = o(u/2) and lim

u
_
u
u
2
_
= . Thus
lim
u
e
u
u
2 = e
= 0 so lim
x
e
lnx
x
= 0
ande
lnx

x.
Showthat lim
x
x
n
e
x
= 0for all wholenumbersn > 0.
63. AssumptionsMatter Let f (x) = x(2+sinx) andg(x) = x
2
+1.
(a) Showdirectlythat lim
x
f (x)/g(x) = 0.
(b) Showthat lim
x
f (x) = lim
x
g(x) = , but lim
x
f
(x)/g
(x) doesnot exist.

Do(a) and(b) contradict LHpitalsRule? Explain.
solution
(a) 1 2+sinx 3, so
x
x
2
+1

x(2+sinx)
x
2
+1

3x
x
2
+1
.
Since,
lim
x
x
x
2
+1
= lim
x
3x
x
2
+1
= 0,
lim
x
x(2+sinx)
x
2
+1
= 0.
(b) lim
x
f (x) = lim
x
x(2+sinx) lim
x
x = and lim
x
g(x) = lim
x
(x
2
+1) = , but
lim
x
f
(x)
g
(x)
= lim
x
x(cosx) +(2+sinx)
2x
doesnot exist sincecosx oscillates. Thisdoesnot violateLHpitalsRulesincethetheoremclearlystates
lim
x
f (x)
g(x)
= lim
x
f
(x)
g
(x)
providedthelimit ontheright exists.
Let H(b) = lim
x
ln(1+b
x
)
x
for b > 0.
(a) Showthat H(b) = lnb if b 1
(b) DetermineH(b) for 0 0.
(b) Verifyyour resultgraphicallybyplottingy = (1+b
x
)
1/x
together withthehorizontal liney = G(b) for the
valuesb = 0.25, 0.5, 2, 3.
solution
(a) UsingExercise64, weseethat G(b) = e
H(b)
. Thus, G(b) = 1if 0 b 1andG(b) = b if b > 1.
(b)
1
5 10 15
2
3
4
y
x
b = 0.25
1
5 10 15
2
3
4
y
x
b = 0.5
1
5 10 15
2
3
4
y
x
b = 2.0
1
5 10 15
2
3
4
5
6
y
x
b = 3.0
Showthat lim
t
t
k
e
t
2
= 0for all k. Hint: Comparewith lim
t
t
k
e
t
= 0.
In Exercises 6769, let
f (x) =
_
e
1/x
2
for x = 0
0 for x = 0
These exercises show that f (x) has an unusual property: All of its derivatives at x = 0exist and are equal to zero.
67. Showthat lim
x0
f (x)
x
k
= 0for all k. Hint: Let t = x
1
andapplytheresult of Exercise66.
solution lim
x0
f (x)
x
k
= lim
x0
1
x
k
e
1/x
2
. Let t = 1/x. Asx 0, t . Thus,
lim
x0
1
x
k
e
1/x
2
= lim
t
t
k
e
t
2
= 0
byExercise66.
Showthat f
(0) existsandisequal tozero. Also, verifythat f
(0) existsandisequal tozero.

69. Showthat for k 1andx = 0,
f
(k)
(x) =
P(x)e
1/x
2
x
r
for somepolynomial P(x) andsomeexponent r 1. Usetheresult of Exercise67to showthat f
(k)
(0) exists andis
equal tozerofor all k 1.
solution For x = 0, f
(x) = e
1/x
2
_
2
x
3
_
. HereP(x) = 2andr = 3. Assumef
(k)
(x) =
P(x)e
1/x
2
x
r
. Then
f
(k+1)
(x) = e
1/x
2
_
x
3
P
(x) +(2rx
2
)P(x)
x
r+3
_
whichisof theformdesired.
Moreover, fromExercise68, f
(0) = 0. Supposef
(k)
(0) = 0. Then
f
(k+1)
(0) = lim
x0
f
(k)
(x) f
(k)
(0)
x 0
= lim
x0
P(x)e
1/x
2
x
r+1
= P(0) lim
x0
f (x)
x
r+1
= 0.
(a)Verifyfor r = 0:
_
T
0
t e
rt
dt =
e
rT
(rT 1) +1
r
2
Hint: For xedr, let F(T ) bethevalueof theintegral ontheleft. ByFTCII, F
(t ) = t e
rt
andF(0) = 0. Showthat
thesameistrueof thefunctionontheright.
(b)UseLHpitalsRuletoshowthat for xedT , thelimit asr 0of theright-handsideof Eq. (1) isequal tothe
valueof theintegral for r = 0.
71. Theformula
_
x
1
t
n
dt =
x
n+1
1
n +1
isvalidfor n = 1. UseLHpitalsRuletoprovethat
lim
n1
x
n+1
1
n +1
= lnx
Usethistoshowthat
lim
n1
_
x
1
t
n
dt =
_
x
1
t
1
dt
Thus, thedeniteintegral of x
1
isalimit of thedeniteintegralsof x
n
asn approaches1.
solution
lim
n1
_
x
1
t
n
dt = lim
n1
t
n+1
n +1
x
1
= lim
n1
_
x
n+1
n +1
1
n+1
n +1
_
= lim
n1
x
n+1
1
n +1
= lim
n1
(x
n+1
) lnx = lnx =
_
x
1
t
1
dt
Notethat whenusingLHpitalsRuleinthesecondline, weneedtodifferentiatewithrespect ton.
Showthat LHpitalsRuleappliesto lim
x
x
_
x
2
+1
but that it doesnot help. Thenevaluatethelimit directly.
73. TheSecondDerivativeTest for critical pointsfailsif f
(c) = 0. ThisexercisedevelopsaHigher DerivativeTest

basedonthesignof therst nonzeroderivative. Supposethat
f
(c) = f
(c) = = f
(n1)
(c) = 0, but f
(n)
(c) = 0
(a) Show, byapplyingLHpitalsRulen times, that
lim
xc
f (x) f (c)
(x c)
n
=
1
n!
f
(n)
(c)
wheren! = n(n 1)(n 2) (2)(1).
(b) Use(a) toshowthatif n iseven, thenf (c) isalocal minimumif f
(n)
(c) > 0andisalocal maximumif f
(n)
(c) < 0.
Hint: If n iseven, then(x c)
n
> 0for x = a, sof (x) f (c) must bepositivefor x near c if f
(n)
(c) > 0.
(c) Use(a) toshowthat if n isodd, thenf (c) isneither alocal minimumnor alocal maximum.
solution
(a) Repeatedapplicationof LHpitalsruleyields
lim
xc
f (x) f (c)
(x c)
n
= lim
xc
f
(x)
n(x c)
n1
= lim
xc
f
(x)
n(n 1)(x c)
n2
= lim
xc
f
(x)
n(n 1)(n 2)(x c)
n3
=
=
1
n!
f
(n)
(c)
(b) Supposen iseven. Then(x c)
n
> 0for all x = c. If f
(n)
(c) > 0, it followsthat f (x) f (c) must bepositive
for x near c. Inother words, f (x) > f (c) for x near c andf (c) isalocal minimum. Ontheother hand, if f
(n)
(c) < 0,
it followsthat f (x) f (c) must benegativefor x near c. Inother words, f (x) < f (c) for x near c andf (c) isalocal
maximum.
(c) If n isodd, then(x c)
n
> 0for x > c but (x c)
n
< 0for x < c. If f
(n)
(c) > 0, it followsthat f (x) f (c)
must bepositivefor x near c andx > c but isnegativefor x near c andx < c. Inother words, f (x) > f (c) for x near
c andx > c but f (x) < f (c) for x near c andx < c. Thus, f (c) isneither alocal minimumnor alocal maximum. We
obtainasimilar result if f
(n)
(c) < 0.
Whenaspringwithnatural frequency /2 is drivenwithasinusoidal forcesin(t ) with = , it oscillates
accordingto
y(t ) =
1
2
_
sin(t ) sin(t )
_
Let y
0
(t ) = lim
y(t ).
(a) UseLHpitalsRuletodeterminey
0
(t ).
(b) Showthat y
0
(t ) ceasestobeperiodicandthat itsamplitude|y
0
(t )| tendstoast (thesystemissaidto
beinresonance; eventually, thespringisstretchedbeyonditslimits).
(c) Plot y(t ) for = 1and = 0.8, 0.9, 0.99, and0.999. Dothegraphsconrmyour conclusionin(b)?
75. Weexpendedalot of effort toevaluate lim
x0
sinx
x
inChapter 2. Showthat wecouldhaveevaluatedit easily
usingLHpitalsRule. Thenexplainwhythismethodwouldinvolvecircular reasoning.
solution lim
x0
sinx
x
= lim
x0
cosx
1
= 1. To use LHpitals Rule to evaluate lim
x0
sinx
x
, we must know that the
derivativeof sinx iscosx, but todeterminethederivativeof sinx, wemust beabletoevaluate lim
x0
sinx
x
.
By afact fromalgebra, if f , g arepolynomials suchthat f (a) = g(a) = 0, thentherearepolynomials f
1
, g
1
suchthat
f (x) = (x a)f
1
(x), g(x) = (x a)g
1
(x)
UsethistoverifyLHpitalsRuledirectlyfor lim
xa
f (x)/g(x).
77. PatienceRequired UseLHpitalsRuletoevaluateandcheckyour answersnumerically:
(a) lim
x0+
_
sinx
x
_
1/x
2
(b) lim
x0
_
1
sin
2
x
1
x
2
_
solution
(a) Westart byevaluating
lim
x0+
ln
_
sinx
x
_
1/x
2
= lim
x0+
ln(sinx) lnx
x
2
.
RepeatedlyusingLHpitalsRule, wend
lim
x0+
ln
_
sinx
x
_
1/x
2
= lim
x0+
cotx x
1
2x
= lim
x0+
x cosx sinx
2x
2
sinx
= lim
x0+
x sinx
2x
2
cosx +4x sinx
= lim
x0+
x cosx sinx
8x cosx +4sinx 2x
2
sinx
= lim
x0+
2cosx +x sinx
12cosx 2x
2
cosx 12x sinx
=
2
12
=
1
6
.
Therefore, lim
x0+
_
sinx
x
_
1/x
2
= e
1/6
. Numericallywend:
x 1 0.1 0.01
_
sinx
x
_
1/x
2
0.841471 0.846435 0.846481
Notethat e
1/6
0.846481724.
S E C T I ON 7.8 Inverse Trigonometric Functions 415
(b) RepeatedlyusingLHpitalsRuleandsimplifying, wend
lim
x0
_
1
sin
2
x
1
x
2
_
= lim
x0
x
2
sin
2
x
x
2
sin
2
x
= lim
x0
2x 2sinx cosx
x
2
(2sinx cosx) +2x sin
2
x
= lim
x0
2x 2sin2x
x
2
sin2x +2x sin
2
x
= lim
x0
22cos2x
2x
2
cos2x +2x sin2x +4x sinx cosx +2sin
2
x
= lim
x0
22cos2x
2x
2
cos2x +4x sin2x +2sin
2
x
= lim
x0
4sin2x
4x
2
sin2x +4x cos2x +8x cos2x +4sin2x +4sinx cosx
= lim
x0
4sin2x
(64x
2
) sin2x +12x cos2x
= lim
x0
8cos2x
(128x
2
) cos2x 8x sin2x +12cos2x 24x sin2x
=
1
3
.
Numericallywend:
x 1 0.1 0.01
1
sin
2
x
1
x
2
0.412283 0.334001 0.333340
Inthefollowingcases, check that x = c is acritical point anduseExercise73to determinewhether f (c) is a
local minimumor alocal maximum.
(a) f (x) = x
5
6x
4
+14x
3
16x
2
+9x +12 (c = 1)
(b) f (x) = x
6
x
3
(c = 0)
7.8 Inverse Trigonometric Functions
1. Whichof thefollowingquantitiesisundened?
(a) sin
1
_
1
2
_
(b) cos
1
(2)
(c) csc
1
_
1
2
_
(d) csc
1
(2)
solution (b)and(c)areundened. sin
1
_
1
2
_
=
6
andcsc
1
(2) =

6
.
2. Giveanexampleof anangle suchthat cos
1
(cos) = . Doesthiscontradict thedenitionof inversefunction?
solution Anyangle < 0or > will work. No, thisdoesnot contradict thedenitionof inversefunction.
3. What isthegeometricinterpretationof theidentity
sin
1
x +cos
1
x =

2
What doesthisidentitytell usabout thederivativesof sin
1
x andcos
1
x?
solution Geometrically, theidentitytellsusthatangleswhosesineandcosinearex arecomplementary. Because/2
isaconstant, it followsthat thederivativesof sin
1
x andcos
1
x sumtozero.
4. Findb suchthat
_
b
0
dx
1+x
2
=

3
.
solution Ingeneral,
_
b
0
dx
1+x
2
= tan
1
x
b
0
= tan
1
b tan
1
0= tan
1
b.
For thevalueof thedeniteintegral toequal

3
, wemust have
tan
1
b =

3
or b = tan
3
=
3.
5. Whichrelationbetweenx andu yields
_
16+x
2
= 4
_
1+u
2
?
solution Totransform
_
16+x
2
into4
_
1+u
2
, makethesubstitutionx = 4u.
Exercises
In Exercises 16, evaluate without using a calculator.
1. cos
1
1
solution cos
1
1= 0.
sin
1
1
2
3. cot
1
1
solution cot
1
1=

4
.
sec
1
2
3
5. tan
1
3
solution tan
1
3= tan
1
_
3/2
1/2
_
=

3
.
sin
1
(1)
In Exercises 716, compute without using a calculator.
7. sin
1
_
sin
3
_
solution sin
1
(sin
3
) =

3
.
sin
1
_
sin
4
3
_
9. cos
1
_
cos
3
2
_
solution cos
1
(cos
3
2
) = cos
1
(0) =

2
. Theanswer isnot
3
2
because
3
2
isnotintherangeof theinversecosine
function.
sin
1
_
sin
_
5
6
__
11. tan
1
_
tan
3
4
_
solution tan
1
(tan
3
4
) = tan
1
(1) =
4
. Theanswer is not
3
4
because
3
4
is not intherangeof theinverse
tangent function.
tan
1
(tan)
13. sec
1
(sec3)
solution sec
1
(sec3) = sec
1
(1) = . Theanswer isnot3 because3 isnotintherangeof theinversesecant
function.
sec
1
_
sec
3
2
_
15. csc
1
_
csc()
_
solution Noinversesincecsc() =
1
sin()
=
1
0
.
cot
1
_
cot
_
4
__
In Exercises 1720, simplify by referring to the appropriate triangle or trigonometric identity.
17. tan(cos
1
x)
solution Let = cos
1
x. Thencos = x andwegeneratethetriangleshownbelow. Fromthetriangle,
tan(cos
1
x) = tan =
_
1x
2
x
.
1
x
1 x
2
cos(tan
1
x)
19. cot(sec
1
x)
solution Let = sec
1
x. Thensec = x andwegeneratethetriangleshownbelow. Fromthetriangle,
cot(sec
1
x) = cot =
1
_
x
2
1
.
x
1
x
2
1
cot(sin
1
x)
In Exercises 2128, refer to the appropriate triangle or trigonometric identity to compute the given value.
21. cos
_
sin
1 2
3
_
solution Let = sin
1 2
3
. Thensin =
2
3
andwegeneratethetriangleshownbelow. Fromthetriangle,
cos
_
sin
1
2
3
_
= cos =
5
3
.
3
2
5
tan
_
cos
1 2
3
_ 23. tan
_
sin
1
0.8
_
solution Let = sin
1
0.8. Thensin = 0.8=
4
5
andwegeneratethetriangleshownbelow. Fromthetriangle,
tan(sin
1
0.8) = tan =
4
3
.
5
4
3
cos
_
cot
1
1
_ 25. cot
_
csc
1
2
_
solution csc
1
2=

6
. Hence, cot(csc
1
2) = cot
6
=

3.
tan
_
sec
1
(2)
_ 27. cot
_
tan
1
20
_
solution Let = tan
1
20. Thentan = 20, socot(tan
1
20) = cot =
1
tan
=
1
20
.
sin
_
csc
1
20
_
In Exercises 2932, compute the derivative at the point indicated without using a calculator.
29. y = sin
1
x, x =
3
5
solution Let y = sin
1
x. Theny
=
1
1x
2
and
y
_
3
5
_
=
1
19/25
=
1
4/5
=
5
4
.
y = tan
1
x, x =
1
2
31. y = sec
1
x, x = 4
solution Let y = sec
1
x. Theny
=
1
|x|
x
2
1
and
y
(4) =
1
4
15
.
y = arccos(4x), x =
1
5
33. y = sin
1
(7x)
solution
d
dx
sin
1
(7x) =
1
_
1(7x)
2

d
dx
7x =
7
_
1(7x)
2
.
y = arctan
_
x
3
_
35. y = cos
1
(x
2
)
solution
d
dx
cos
1
(x
2
) =
1
_
1x
4

d
dx
x
2
=
2x
_
1x
4
.
y = sec
1
(t +1)
37. y = x tan
1
x
solution
d
dx
x tan
1
x = x
_
1
1+x
2
_
+tan
1
x.
y = e
cos
1
x
39. y = arcsin(e
x
)
solution
d
dx
sin
1
(e
x
) =
1
_
1e
2x
d
dx
e
x
=
e
x
_
1e
2x
.
y = csc
1
(x
1
)
41. y =
_
1t
2
+sin
1
t
solution
d
dt
__
1t
2
+sin
1
t
_
=
1
2
(1t
2
)
1/2
(2t ) +
1
_
1t
2
=
t
_
1t
2
+
1
_
1t
2
=
1t
_
1t
2
.
y = tan
1
_
1+t
1t
_
43. y = (tan
1
x)
3
solution
d
dx
_
(tan
1
x)
3
_
= 3(tan
1
x)
2
d
dx
tan
1
x =
3(tan
1
x)
2
x
2
+1
.
y =
cos
1
x
sin
1
x
45. y = cos
1
t
1
sec
1
t
solution
d
dx
(cos
1
t
1
sec
1
t )=
1
_
1(1/t )
2
_
1
t
2
_
1
|t |
_
t
2
1
=
1
_
t
4
t
2

1
|t |
_
t
2
1
=
1
|t |
_
t
2
1
1
|t |
_
t
2
1
= 0.
Alternately, let t = sec. Thent
1
= cos andcos
1
t
1
sec
1
t = = 0. Consequently,
d
dx
(cos
1
t
1
sec
1
t ) = 0.
y = cos
1
(x +sin
1
x)
47. y = arccos(lnx)
solution
d
dx
arccos(lnx) =
1
x
_
1(lnx)
2
.
y = ln(arcsinx) 49. UseFigure10toprovethat (cos
1
x)
=
1
_
1x
2
.
1
x
1 x
2
FIGURE 10 Right trianglewith = cos

1
x.
solution Let = cos
1
x. Thencos = x and
sin
d
dx
= 1 or
d
dx
=
1
sin
=
1
sin(cos
1
x)
.
FromFigure10, weseethat sin(cos
1
x) = sin =
_
1x
2
; hence,
d
dx
cos
1
x =
1
sin(cos
1
x)
=
1
_
1x
2
.
Showthat (tan
1
x)
= cos
2
(tan
1
x) andthenuseFigure11toprovethat (tan
1
x)
= (x
2
+1)
1
.
51. Let = sec
1
x. Showthat tan =
_
x
2
1if x 1andthat tan =
_
x
2
1if x 1. Hint: tan 0on
_
0,

2
_
andtan 0on
_
2
,
_
.
solution Ingeneral, 1+tan
2
= sec
2
, sotan =
_
sec
2
1. With = sec
1
x, it followsthat sec = x, so
tan =
_
x
2
1. Finally, if x 1then = sec
1
x [0, /2) sotan ispositive; ontheother hand, if x 1then
= sec
1
x (/2, 0] sotan isnegative.
UseExercise51toverifytheformula
(sec
1
x)
=
1
|x|
_
x
2
1
In Exercises 5356, evaluate the denite integral.
53.
_
tan8
tan1
dx
x
2
+1
solution
_
tan8
tan1
dx
1+x
2
= tan
1
x
tan8
tan1
= tan
1
(tan8) tan
1
(tan1) = 81= 7.
_
7
2
x dx
x
2
+1
55.
_
1/2
0
dx
_
1x
2
solution
_
1/2
0
dx
_
1x
2
= sin
1
x
1/2
0
= sin
1
1
2
sin
1
0=

6
.
_
2/
3
2
dx
|x|
_
x
2
1
57. Usethesubstitutionu = x/3toprove
_
dx
9+x
2
=
1
3
tan
1
x
3
+C
solution Let u = x/3. Then, x = 3u, dx = 3du, 9+x
2
= 9(1+u
2
), and
_
dx
9+x
2
=
_
3du
9(1+u
2
)
=
1
3
_
du
1+u
2
=
1
3
tan
1
u +C =
1
3
tan
1
x
3
+C.
Usethesubstitutionu = 2x toevaluate
_
dx
4x
2
+1
.
59.
_
3
0
dx
x
2
+3
solution Let x =

3u. Thendx =

3du and
_
3
0
dx
x
2
+3
=
1
3
_

3
0
du
u
2
+1
=
1
3
tan
1
u
3
0
=
1
3
(tan
1
3tan
1
0) =

3
3
.
_
4
0
dt
4t
2
+9
61.
_
dt
_
116t
2
solution Let u = 4t . Thendu = 4dt , and
_
dt
_
116t
2
=
_
du
4
_
1u
2
=
1
4
sin
1
u +C =
1
4
sin
1
(4t ) +C.
_
1/5
1/5
dx
_
425x
2
63.
_
dt
_
53t
2
solution Let t =

5/3u. Thendt =

5/3du and
_
dt
_
53t
2
=
_
5/3du
5
_
1u
2
=
1
3
_
du
_
1u
2
=
1
3
sin
1
u +C =
1
3
sin
1
_
3
5
t +C.
_
1/2
1/2
2
dx
x
_
16x
2
1
65.
_
dx
x
_
12x
2
3
solution Let u = 2x. Thendu = 2dx and
_
dx
x
_
12x
2
3
=
1
3
_
du
u
_
u
2
1
=
1
3
sec
1
u +C =
1
3
sec
1
(2x) +C.
_
x dx
x
4
+1
67.
_
dx
x
_
x
4
1
solution Let u = x
2
_
dx
x
_
x
4
1
=
_
du
2u
_
u
2
1
=
1
2
sec
1
u +C =
1
2
sec
1
x
2
+C.
_
0
1/2
(x +1) dx
_
1x
2
69.
_
ln(cos
1
x) dx
(cos
1
x)
_
1x
2
solution Let u = lncos
1
x. Thendu =
1
cos
1
x
1
_
1x
2
, and
_
ln(cos
1
x) dx
(cos
1
x)
_
1x
2
=
_
u du =
1
2
u
2
+C =
1
2
(lncos
1
x)
2
+C.
_
tan
1
x dx
1+x
2
71.
_

3
1
dx
(tan
1
x)(1+x
2
)
solution Let u = tan
1
x. Thendu =
dx
1+x
2
, and
_

3
1
dx
(tan
1
x)(1+x
2
)
=
_
/3
/4
1
u
du = ln|u|
/3
/4
= ln
3
ln
4
= ln
4
3
.
_
dx
_
5
2x
1
In Exercises 73110, evaluate the integral using the methods covered in the text so far.
73.
_
ye
y
2
dy
solution Usethesubstitutionu = y
2
, du = 2y dy. Then
_
ye
y
2
dy =
1
2
_
e
u
du =
1
2
e
u
+C =
1
2
e
y
2
+C.
_
dx
3x +5
75.
_
x dx
_
4x
2
+9
solution Let u = 4x
2
_
x
_
4x
2
+9
dx =
1
8
_
u
1/2
du =
1
4
u
1/2
+C =
1
4
_
4x
2
+9+C.
_
(x x
2
)
2
dx
77.
_
7
x
dx
solution Let u = x. Thendu = dx and
_
7
x
dx =
_
7
u
du =
7
u
ln7
+C =
7
x
ln7
+C.
_
e
912t
dt
79.
_
sec
2
tan
7
d
solution Let u = tan. Thendu = sec
2
d and
_
sec
2
tan
7
d =
_
u
7
du =
1
8
u
8
+C =
1
8
tan
8
+C.
_
cos(lnt ) dt
t
81.
_
t dt
_
7t
2
solution Let u = 7t
2
. Thendu = 2t dt and
_
t dt
_
7t
2
=
1
2
_
u
1/2
du = u
1/2
+C =
_
7t
2
+C.
_
2
x
e
4x
dx
83.
_
(3x +2) dx
x
2
+4
solution Write
_
(3x +2) dx
x
2
+4
=
_
3x dx
x
2
+4
+
_
2dx
x
2
+4
.
Intherst integral, let u = x
2
_
3x dx
x
2
+4
=
3
2
_
du
u
3
2
ln|u| +C
1
=
3
2
ln(x
2
+4) +C
1
.
For thesecondintegral, let x = 2u. Thendx = 2du and
_
2dx
x
2
+4
=
_
du
u
2
+1
= tan
1
u +C
2
= tan
1
(x/2) +C
2
.
Combiningthesetworesultsyields
_
(3x +2) dx
x
2
+4
=
3
2
ln(x
2
+4) +tan
1
(x/2) +C.
_
tan(4x +1) dx
85.
_
dx
_
116x
2
solution Let u = 4x. Thendu = 4dx and
_
dx
_
116x
2
=
1
4
_
du
_
1u
2
=
1
4
sin
1
u +C =
1
4
sin
1
(4x) +C.
_
e
t
_
e
t
+1dt
87.
_
(e
x
4x) dx
solution First, observethat
_
(e
x
4x) dx =
_
e
x
dx
_
4x dx =
_
e
x
dx 2x
2
.
Intheremainingintegral, usethesubstitutionu = x, du = dx. Then
_
e
x
dx =
_
e
u
du = e
u
+C = e
x
+C.
Finally,
_
(e
x
4x) dx = e
x
2x
2
+C.
_
(7e
10x
) dx
89.
_
e
2x
e
4x
e
x
dx
solution
_
_
e
2x
e
4x
e
x
_
dx =
_
(e
x
e
3x
) dx = e
x
e
3x
3
+C.
_
dx
x
_
25x
2
1
91.
_
(x +5) dx
_
4x
2
solution Write
_
(x +5) dx
_
4x
2
=
_
x dx
_
4x
2
+
_
5dx
_
4x
2
.
Intherst integral, let u = 4x
2
. Thendu = 2x dx and
_
x dx
_
4x
2
=
1
2
_
u
1/2
du = u
1/2
+C
1
=
_
4x
2
+C
1
.
Inthesecondintegral, let x = 2u. Thendx = 2du and
_
5dx
_
4x
2
= 5
_
du
_
1u
2
= 5sin
1
u +C
2
= 5sin
1
(x/2) +C
2
.
Combiningthesetworesultsyields
_
(x +5) dx
_
4x
2
=
_
4x
2
+5sin
1
(x/2) +C.
_
(t +1)
t +1dt
93.
_
e
x
cos(e
x
) dx
x
, du = e
x
dx. Then
_
e
x
cos(e
x
) dx =
_
cosu du = sinu +C = sin(e
x
) +C.
_
e
x
e
x
+1
dx
95.
_
dx
_
916x
2
solution First rewrite
_
dx
_
916x
2
=
1
3
_
dx
_
1
_
4
3
x
_
2
.
Now, let u =
4
3
x. Thendu =
4
3
dx and
_
dx
_
916x
2
=
1
4
_
du
_
1u
2
=
1
4
sin
1
u +C =
1
4
sin
1
_
4x
3
_
+C.
_
dx
(4x 1) ln(8x 2)
97.
_
e
x
(e
2x
+1)
3
dx
x
, du = e
x
dx. Then
_
e
x
(e
2x
+1)
3
dx =
_
_
u
2
+1
_
3
du =
_
_
u
6
+3u
4
+3u
2
+1
_
du
=
1
7
u
7
+
3
5
u
5
+u
3
+u +C =
1
7
(e
x
)
7
+
3
5
(e
x
)
5
+(e
x
)
3
+e
x
+C
=
e
7x
7
+
3e
5x
5
+e
3x
+e
x
+C.
_
dx
x(lnx)
5
99.
_
x
2
dx
x
3
+2
solution Let u = x
3
+2. Thendu = 3x
2
dx, and
_
x
2
dx
x
3
+2
=
1
3
_
du
u
=
1
3
ln|x
3
+2| +C.
_
(3x 1) dx
92x +3x
2
101.
_
cotx dx
solution Werewrite
_
cotx dx as
_
cosx
sinx
dx. Let u = sinx. Thendu = cosx dx, and
_
cosx
sinx
dx =
_
du
u
= ln| sinx| +C.
_
cosx
2sinx +3
dx
103.
_
4lnx +5
x
dx
solution Let u = 4lnx +5. Thendu = (4/x)dx, and
_
4lnx +5
x
dx =
1
4
_
u du =
1
8
u
2
+C =
1
8
(4lnx +5)
2
+C.
_
(sec tan)5
sec
d
105.
_
x3
x
2
dx
solution Let u = x
2
_
x3
x
2
dx =
1
2
_
3
u
du =
1
2
3
u
ln3
+C =
3
x
2
2ln3
+C.
_
ln(lnx)
x lnx
dx
107.
_
cotx ln(sinx) dx
solution Let u = ln(sinx). Then
du =
1
sinx
cosx dx = cotx dx,
and
_
cotx ln(sinx) dx =
_
u du =
u
2
2
+C =
(ln(sinx))
2
2
+C.
_
t dt
_
1t
4
109.
_
t
2
t 3dt
solution Let u = t 3. Thent = u +3, du = dt and
_
t
2
t 3dt =
_
(u +3)
2
u du
=
_
(u
2
+6u +9)
u du =
_
(u
5/2
+6u
3/2
+9u
1/2
) du
=
2
7
u
7/2
+
12
5
u
5/2
+6u
3/2
+C
=
2
7
(t 3)
7/2
+
12
5
(t 3)
5/2
+6(t 3)
3/2
+C.
_
cosx5
2sinx
dx
111. UseFigure12toprove
_
x
0
_
1t
2
dt =
1
2
x
_
1x
2
+
1
2
sin
1
x
x
x
y
1
FIGURE 12
_
x
0
_
1t
2
dt representstheareaof theregionunder theupper half of theunit circle
from0to x. Theregionconsists of asector of thecircleandaright triangle. Thesector has acentral angleof

2
,
wherecos = x. Hence, thesector hasanareaof
1
2
(1)
2
_
2
cos
1
x
_
=
1
2
sin
1
x.
Theright trianglehasabaseof lengthx, aheight of
_
1x
2
, andhenceanareaof
1
2
x
_
1x
2
. Thus,
_
x
0
_
1t
2
dt =
1
2
x
_
1x
2
+
1
2
sin
1
x.
Usethesubstitutionu = tanx toevaluate
_
dx
1+sin
2
x
Hint: Showthat
dx
1+sin
2
x
=
du
1+2u
2
113. Prove:
_
sin
1
t dt =
_
1t
2
+t sin
1
t
solution Let G(t ) =
_
1t
2
+t sin
1
t . Then
G
(t ) =
d
dt
_
1t
2
+
d
dt
_
t sin
1
t
_
=
t
_
1t
2
+
_
t
d
dt
sin
1
t +sin
1
t
_
=
t
_
1t
2
+
_
t
_
1t
2
+sin
1
t
_
= sin
1
t.
Thisprovestheformula
_
sin
1
t dt =
_
1t
2
+t sin
1
t .
A cylindrical tankof radiusR andlengthL lyinghorizontallyasinFigure13islledwithoil toheight h.
(a) Showthat thevolumeV(h) of oil inthetankasafunctionof height h is
V(h) = L
_
R
2
cos
1
_
1
h
R
_
(R h)
_
2hR h
2
_
(b) Showthat
dV
dh
= 2L
_
h(2R h).
(c) Supposethat R = 2mandL = 12m, andthat thetankislledat aconstant rateof 1.5m
3
/min. Howfast isthe
height h increasingwhenh = 3m?
115. (a) ExplainwhytheshadedregioninFigure14hasarea
_
lna
0
e
y
dy.
(b) Provetheformula
_
a
1
lnx dx = a lna
_
lna
0
e
y
dy.
(c) Concludethat
_
a
1
lnx dx = a lna a +1.
(d) Usetheresult of (a) tondanantiderivativeof lnx.
solution
(a) Interpretingthegraphwithy astheindependent variable, weseethat thefunctionisx = e
y
. Integratinginy then
givestheareaof theshadedregionas
_
lna
0
e
y
dy
(b) Wecanobtaintheareaunder thegraphof y = lnx fromx = 1to x = a by computingtheareaof therectangle
extendingfromx = 0tox = a horizontally andfromy = 0toy = lna vertically andthensubtractingtheareaof the
shadedregion. Thisyields
_
a
1
lnx dx = a lna
_
lna
0
e
y
dy.
(c) Bydirect calculation
_
lna
0
e
y
dy = e
y
lna
0
= a 1.
Thus,
_
a
1
lnx dx = a lna (a 1) = a lna a +1.
(d) Basedontheseresultsit appearsthat
_
lnx dx = x lnx x +C.
x
y
a
ln a
y = lnx
1
FIGURE 14
7.9 Hyperbolic Functions
1. Whichhyperbolicfunctionstakeononlypositivevalues?
solution coshx andsechx takeononlypositivevalues.
2. Whichhyperbolicfunctionsareincreasingontheir domains?
solution sinhx andtanhx areincreasingontheir domains.
3. Describethreepropertiesof hyperbolicfunctionsthat havetrigonometricanalogs.
solution Hyperbolicfunctionshavethefollowinganalogswithtrigonometricfunctions: parity, identitiesandderiva-
tiveformulas.
4. What arey
(100)
andy
(101)
for y = coshx?
solution Let y = coshx. Theny
= sinhx, y
= coshx, andthispatternrepeatsindenitely. Thus, y

(100)
= coshx
andy
(101)
= sinhx.
Exercises
1. Useacalculator tocomputesinhx andcoshx for x = 3, 0, 5.
solution
x 3 0 5
sinhx =
e
x
e
x
2
e
3
e
3
2
= 10.0179
e
0
e
0
2
= 0
e
5
e
5
2
= 74.203
coshx =
e
x
+e
x
2
e
3
+e
3
2
= 10.0677
e
0
+e
0
2
= 1
e
5
+e
5
2
= 74.210
S E C T I ON 7.9 Hyperbolic Functions 425
Computesinh(ln5) andtanh(3ln5) without usingacalculator.
3. For whichvaluesof x arey = sinhx andy = coshx increasinganddecreasing?
solution Thegraphof y = sinhx isshownbelowontheleft. Fromthisgraph, weseethat sinhx isincreasingfor all
x. Ontheother hand, fromthegraphof y = coshx shownbelowontheright, weseethat coshx isdecreasingfor x < 0
andisincreasingfor x > 0.
2 2
20
10
10
20
y
x
y = sinh x
2 2
5
10
15
20
25
y = cosh x
y
x
Showthat y = tanhx isanoddfunction.
5. Refer to the graphs to explain why the equation sinhx = t has a unique solution for every t and why
coshx = t hastwosolutionsfor everyt > 1.
solution Fromitsgraphweseethatsinhx isaone-to-onefunctionwith lim
x
sinhx = and lim
x
sinhx = .
Thus, for every real number t , theequationsinhx = t hasauniquesolution. Ontheother hand, fromitsgraph, wesee
that coshx isnot one-to-one. Rather, it isanevenfunctionwithaminimumvalueof cosh0= 1. Thus, for everyt > 1,
theequationcoshx = t hastwosolutions: onepositive, theother negative.
Computecoshx andtanhx, assumingthat sinhx = 0.8.
7. Provetheadditionformulafor coshx.
solution
cosh(x +y) =
e
x+y
+e
(x+y)
2
=
2e
x+y
+2e
(x+y)
4
=
e
x+y
+e
x+y
+e
xy
+e
(x+y)
4
+
e
x+y
e
x+y
e
xy
+e
(x+y)
4
=
_
e
x
+e
x
2
__
e
y
+e
y
2
_
+
_
e
x
e
x
2
__
e
y
e
y
2
_
= coshx coshy +sinhx sinhy.
Usetheadditionformulastoprove
sinh(2x) = 2coshx sinhx
cosh(2x) = cosh
2
x +sinh
2
x
9. y = sinh(9x)
solution
d
dx
sinh(9x) = 9cosh(9x).
y = sinh(x
2
)
11. y = cosh
2
(93t )
solution
d
dt
cosh
2
(93t ) = 2cosh(93t ) (3sinh(93t )) = 6cosh(93t ) sinh(93t ).
y = tanh(t
2
+1)
13. y =

coshx +1
solution
d
dx
coshx +1=
1
2
(coshx +1)
1/2
sinhx.
y = sinhx tanhx 15. y =
cotht
1+tanht
solution
d
dt
cotht
1+tanht
=
csch
2
t (1+tanht ) cotht (sech
2
t )
(1+tanht )
2
=
csch
2
t +2cscht secht
(1+tanht )
2
.
y = (ln(coshx))
5
17. y = sinh(lnx)
solution
d
dx
sinh(lnx) =
cosh(lnx)
x
.
y = e
cothx
19. y = tanh(e
x
)
solution
d
dx
tanh(e
x
) = e
x
sech
2
(e
x
).
y = sinh(cosh
3
x)
21. y = sech(
x)
solution
d
dx
sech(
x) =
1
2
x
1/2
sech
x tanh
x.
y = ln(cothx)
23. y = sechx cothx
solution
d
dx
sechx cothx =
d
dx
cschx = cschx cothx.
y = x
sinhx
25. y = cosh
1
(3x)
solution
d
dx
cosh
1
(3x) =
3
_
9x
2
1
.
y = tanh
1
(e
x
+x
2
)
27. y = (sinh
1
(x
2
))
3
solution
d
dx
(sinh
1
(x
2
))
3
= 3(sinh
1
(x
2
))
2
2x
_
x
4
+1
.
y = (csch
1
3x)
4
29. y = e
cosh
1
x
solution
d
dx
e
cosh
1
x
= e
cosh
1
x
_
1
_
x
2
1
_
.
y = sinh
1
(
_
x
2
+1)
31. y = tanh
1
(lnt )
solution
d
dt
tanh
1
(lnt ) =
1
t (1(lnt )
2
)
.
y = ln(tanh
1
x)
33. Showthat for anyconstantsM, k, anda, thefunction
y(t ) =
1
2
M
_
1+tanh
_
k(t a)
2
__
satisesthelogisticequation:
y
y
= k
_
1
y
M
_
.
solution Let
y(t ) =
1
2
M
_
1+tanh
_
k(t a)
2
__
.
Then
1
y(t )
M
=
1
2
_
1tanh
_
k(t a)
2
__
,
and
ky(t )
_
1
y(t )
M
_
=
1
4
Mk
_
1tanh
2
_
k(t a)
2
__
=
1
4
Mk sech
2
_
k(t a)
2
_
.
Finally,
y
(t ) =
1
4
Mk sech
2
_
k(t a)
2
_
= ky(t )
_
1
y(t )
M
_
.
Showthat V(x) = 2ln(tanh(x/2)) satises thePoisson-Boltzmannequation V
(x) = sinh(V(x)), which is

usedtodescribeelectrostaticforcesincertainmolecules.
35.
_
cosh(3x) dx
solution
_
cosh(3x) dx =
1
3
sinh3x +C.
_
sinh(x +1) dx
37.
_
x sinh(x
2
+1) dx
solution
_
x sinh(x
2
+1) dx =
1
2
cosh(x
2
+1) +C.
_
sinh
2
x coshx dx
39.
_
sech
2
(12x) dx
solution
_
sech
2
(12x) dx =
1
2
tanh(12x) +C.
_
tanh(3x) sech(3x) dx
41.
_
tanhx sech
2
x dx
solution Let u = tanhx. Thendu = sech
2
x dx nd
_
tanhx sech
2
x dx =
_
u du =
1
2
u
2
+C =
tanh
2
x
2
+C.
_
coshx
3sinhx +4
dx
43.
_
tanhx dx
solution
_
tanhx dx = lncoshx +C.
_
coshx
sinhx
dx
45.
_
e
x
sinhx dx
solution Sincesinhx =
e
x
e
x
2
wecancombinethetwofunctionstoget
_
e
x
sinhx dx =
1
2
_
e
x
(e
x
e
x
) dx =
1
2
_
_
1e
2x
_
dx =
1
2
x +
1
4
e
2x
+C.
_
coshx
sinh
2
x
dx
In Exercises 4752, prove the formula.
47.
d
dx
tanhx = sech
2
x
solution
d
dx
tanhx =
d
dx
sinhx
coshx
=
cosh
2
x sinh
2
x
cosh
2
x
=
1
cosh
2
x
= sech
2
x.
d
dx
sechx = sechx tanhx
49. cosh(sinh
1
t ) =
_
t
2
+1
solution Let w = sinh
1
t , so that sinhw = t and the stated problembecomes evaluating coshw given that
sinhw = t . Fromtheidentity cosh
2
w sinh
2
w = 1wendcosh
2
w = sinh
2
w + 1, or coshw =
_
sinh
2
w +1.
Becausehyperboliccosineisalwayspositive, weknowtochoosethepositivesquareroot. Finally, returningtothevariable
t wehave
cosh(sinh
1
t ) =
_
t
2
+1.
sinh(cosh
1
t ) =
_
t
2
1 for t 1
51.
d
dt
sinh
1
t =
1
_
t
2
+1
solution Let x = sinh
1
t . Thent = sinhx and
1= coshx
dx
dt
or
dx
dt
=
1
coshx
.
Thus,
d
dt
sinh
1
t =
1
cosh(sinh
1
t )
=
1
_
t
2
+1
byExercise49.
d
dt
cosh
1
t =
1
_
t
2
1
for t > 1
In Exercises 5360, calculate the integral in terms of inverse hyperbolic functions.
53.
_
4
2
dx
_
x
2
1
solution
_
4
2
dx
_
x
2
1
= cosh
1
x
4
2
= cosh
1
4cosh
1
2.
_
dx
_
x
2
4
55.
_
dx
_
9+x
2
solution
_
dt
_
9+x
2
=
_
dx
3
_
1+(x/3)
2
= sinh
1
x
3
+C.
_
dx
_
1+9x
2
57.
_
1/2
1/3
dx
1x
2
solution
_
1/2
1/3
dx
1x
2
= tanh
1
x
1/2
1/3
= tanh
1
1
2
tanh
1
1
3
.
_
1
0
dx
_
1+x
2
59.
_
10
2
dx
4x
2
1
solution
_
10
2
dx
4x
2
1
=
1
2
coth
1
(2x)
10
2
=
1
2
(coth
1
4coth
1
20).
_
1
3
dx
x
_
x
2
+16
61. Provethat sinh
1
t = ln(t +
_
t
2
+1). Hint: Let t = sinhx. Provethat coshx =
_
t
2
+1andusetherelation
sinhx +coshx = e
x
solution Let t = sinhx. Then
coshx =
_
1+sinh
2
x =
_
1+t
2
.
Moreover, because
sinhx +coshx =
e
x
e
x
2
+
e
x
+e
x
2
= e
x
,
it followsthat
sinh
1
t = x = ln(sinhx +coshx) = ln(t +
_
t
2
+1).
Provethat cosh
1
t = ln(t +
_
t
2
1) for t > 1.
63. Provethat tanh
1
t =
1
2
ln
_
1+t
1t
_
for |t | < 1.
solution Let A = tanh
1
t . Then
t = tanhA =
sinhA
coshA
=
e
A
e
A
e
A
+e
A
.
Solvingfor A yields
A =
1
2
ln
t +1
1t
;
hence,
tanh
1
t =
1
2
ln
t +1
1t
.
Usethesubstitutionu = sinhx toprove
_
sechx dx = tan
1
(sinhx) +C
65. An(imaginary) trainmoves alongatrack at velocity v. Bionicawalks downtheaisleof thetrainwithvelocity u
inthedirectionof thetrains motion. Computethevelocity w of Bionicarelativeto thegroundusingthelaws of both
GalileoandEinsteininthefollowingcases.
(a) v = 500m/sandu = 10m/s. Isyour calculator accurateenoughtodetect thedifferencebetweenthetwolaws?
(b) v = 10
7
m/sandu = 10
6
m/s.
solution Recall that thespeedof light isc 310
8
m/s.
(a) ByGalileoslaw, w = 500+10= 510m/s. UsingEinsteinslawandacalculator,
tanh
1
w
c
= tanh
1
500
c
+tanh
1
10
c
= 1.710
6
;
sow = c tanh(1.710
6
) 510m/s. No, thecalculator wasnot accurateenoughtodetect thedifferencebetweenthe
twolaws.
(b) ByGalileoslaw, u +v = 10
7
+10
6
= 1.110
7
m/s. ByEinsteinslaw,
tanh
1
w
c
= tanh
1
10
7
310
8
+tanh
1
10
6
310
8
0.036679,
sow c tanh(0.036679) 1.0998810
7
m/s.
Show that thelinearization of thefunction y = tanh
1
x at x = 0 is tanh
1
x x. Usethis to explain the
followingassertion: EinsteinsLawof VelocityAddition[Eq. (2)] reducestoGalileosLawif thevelocitiesaresmall
relativetothespeedof light.
67. (a) Usetheadditionformulasfor sinhx andcoshx toprove
tanh(u +v) =
tanhu +tanhv
1+tanhu tanhv
(b) Use(a) toshowthat EinsteinsLawof VelocityAddition[Eq. (2)] isequivalent to
w =
u +v
1+
uv
c
2
solution
(a)
tanh(u +v) =
sinh(u +v)
cosh(u +v)
=
sinhu coshv +coshu sinhv
coshu coshv +sinhu sinhv
=
sinhu coshv +coshu sinhv
coshu coshv +sinhu sinhv
1/(coshu coshv)
1/(coshu coshv)
=
tanhu +tanhv
1+tanhu tanhv
(b) Einsteinslawstates: tanh
1
(w/c) = tanh
1
(u/c) +tanh
1
(v/c). Thus
w
c
= tanh
_
tanh
1
(u/c) +tanh
1
(v/c)
_
=
tanh(tanh
1
(v/c)) +tanh(tanh
1
(u/c))
1+tanh(tanh
1
(v/c)) tanh(tanh
1
(u/c))
=
v
c
+
u
c
1+
v
c
u
c
=
(1/c)(u +v)
1+
uv
c
2
.
Hence,
w =
u +v
1+
uv
c
2
.
Provethat
_
a
a
coshx sinhx dx = 0for all a.
69. (a) Showthat y = tanht satisesthedifferential equationdy/dt = 1y
2
withinitial conditiony(0) = 0.
(b) Showthat for arbitraryconstantsA, B, thefunction
y = Atanh(Bt )
satises
dy
dt
= AB
B
A
y
2
, y(0) = 0
(c) Let v(t ) bethevelocityof afallingobject of massm. For largevelocities, air resistanceisproportional tothesquare
of velocityv(t )
2
. If wechoosecoordinatessothat v(t ) > 0for afallingobject, thenbyNewtonsLawof Motion, there
isaconstant k > 0suchthat
dv
dt
= g
k
m
v
2
Solvefor v(t ) byapplyingtheresult of (b) withA =

gm/k andB =

gk/m.
(d) Calculatetheterminal velocity lim
t
v(t ).
(e) Findk if m = 150lbandtheterminal velocityis100mph.
solution
(a) First, notethatif wedividetheidentitycosh
2
t sinh
2
t = 1bycosh
2
t , weobtaintheidentity1tanh
2
t = sech
2
t .
Now, let y = tanht . Then
dy
dt
= sech
2
t = 1tanh
2
t = 1y
2
.
Furthermore, y(0) = tanh0= 0.
(b) Let y = Atanh(Bt ). Then
dy
dt
= AB sech
2
(Bt ) = AB(1tanh
2
(Bt )) = AB
_
1
y
2
A
2
_
= AB
By
2
A
.
Furthermore, y(0) = Atanh(0) = 0.
(c) Matchingthedifferential equation
dv
dt
= g
k
m
v
2
withthetemplate
dv
dt
= AB
B
A
v
2
frompart (b) yields
AB = g and
B
A
=
k
m
.
Solvingfor A andB gives
A =
_
mg
k
and B =
_
kg
m
.
Thus
v(t ) = Atanh(Bt ) =
_
mg
k
tanh
_
_
kg
m
t
_
.
(d) lim
t
v(t ) =
_
mg
k
lim
t
tanh
_
_
kg
m
t
_
=
_
mg
k
(e) Substitutem = 150lbandg = 32ft/sec
2
= 78545.5miles/hr
2
intotheequationfor theterminal velocityobtained
inpart (d) andthensolvefor k. Thisgives
k =
150(78545.5)
100
2
= 1178.18lb/mile.
In Exercises 7072, a exible chain of length L is suspended between two poles of equal height separated by a distance
2M (Figure 9). By Newtons laws, the chain describes a curve (called a catenary) with equation y = a cosh(x/a) +C.
The constant C is arbitrary and a is the number such that L = 2a sinh(M/a). The sags is the vertical distance from the
highest to the lowest point on the chain.
y = a cosh(x/a)
2 M
s
x
y
FIGURE 9 Chainhangingbetweentwopolesdescribesthecurvey = a cosh(x/a).
Suppose that L = 120 and M = 50. Experiment with your calculator to nd an approximate value of a
satisfyingL = 2a sinh(M/a) (for greater accuracy, useNewtonsmethodor acomputer algebrasystem).
71. Let M beaxedconstant. Showthat thesagisgivenbys = a cosh(M/a) a.
(a) Calculate
ds
da
.
(b) Calculateda/dL byimplicit differentiationusingtherelationL = 2a sinh(M/a).
(c) Use(a) and(b) andtheChainRuletoshowthat
ds
dL
=
ds
da
da
dL
=
cosh(M/a) (M/a) sinh(M/a) 1
2sinh(M/a) (2M/a) cosh(M/a)
6
solution Thesaginthecurveis
s = y(M) y(0) = a cosh
_
M
a
_
+C (a cosh0+C) = a cosh
_
M
a
_
a.
(a)
ds
da
= cosh
_
M
a
_
M
a
sinh
_
M
a
_
1
(b) If wedifferentiatetherelationL = 2a sinh
_
M
a
_
withrespect toa, wend
0= 2
da
dL
sinh
_
M
a
_
2M
a
da
dL
cosh
_
M
a
_
.
Solvingfor da/dL yields
da
dL
=
_
2sinh
_
M
a
_
2M
a
cosh
_
M
a
__
1
.
(c) BytheChainRule,
ds
dL
=
ds
da
da
dL
.
Theformulafor ds/dL followsuponsubstitutingtheresultsfromparts(a) and(b).
Assumethat M = 50andL = 160. Inthiscase, aCAScanbeusedtoshowthat a 28.46.
(a) UseEq. (6) andtheLinear Approximationto estimatetheincreaseinsagif L is increasedfromL = 160to
L = 161andfromL = 160toL = 165.
(b) If youhaveaCAS, computes(161) s(160) ands(165) s(160) directlyandcomparewithyour estimates
i ( )
73. Provethat every functionf (x) is thesumof anevenfunctionf
+
(x) andanoddfunctionf
(x). [Hint:f
(x) =
1
2
(f (x) f (x)).] Expressf (x) = 5e
x
+8e
x
intermsof coshx andsinhx.
solution Let f
+
(x) =
f (x)+f (x)
2
andf
(x) =
f (x)f (x)
2
. Thenf
+
+f
=
2f (x)
2
= f (x). Moreover,
f
+
(x) =
f (x) +f ((x))
2
=
f (x) +f (x)
2
= f
+
(x),
sof
+
(x) isanevenfunction, while
f
(x) =
f (x) f ((x))
2
=
f (x) f (x)
2
=
(f (x) f (x))
2
= f
(x),
sof
(x) isanoddfunction.
For f (x) = 5e
x
+8e
x
, wehave
f
+
(x) =
5e
x
+8e
x
+5e
x
+8e
x
2
= 8coshx +5coshx = 13coshx
and
f
(x) =
5e
x
+8e
x
5e
x
8e
x
2
= 5sinhx 8sinhx = 3sinhx.
Therefore, f (x) = f
+
(x) +f
(x) = 13coshx 3sinhx.

Usethemethodof thepreviousproblemtoexpress
f (x) = 7e
3x
+4e
3x
intermsof sinh(3x) andcosh(3x).
75. IntheExcursion, wediscussedtherelations
cosh(it ) = cost and sinh(it ) = i sint
Usetheserelationstoshowthattheidentitycos
2
t +sin
2
t = 1resultsfromtheidentitycosh
2
x sinh
2
x = 1bysetting
x = it .
solution Substitutingx = it intocosh
2
x sinh
2
x = 1yieldscosh
2
(it ) sinh
2
(it ) = 1. Sincecosh
2
(it ) = cos
2
t
andsinh
2
(it ) = (i sint )
2
= sin
2
t , it followsthat cos
2
t +sin
2
t = 1.
1. Matcheachquantity(a)(d) with(i), (ii), or (iii) if possible, or statethat nomatchexists.
(a) 2
a
3
b
(b)
2
a
3
b
(c) (2
a
)
b
(d) 2
ab
3
ba
(i) 2
ab
(ii) 6
a+b
(iii)
_
2
3
_
ab
solution
(a) Nomatch.
(b) Nomatch.
(c) (i): (2
a
)
b
= 2
ab
.
(d) (iii): 2
ab
3
ba
= 2
ab
_
1
3
_
ab
=
_
2
3
_
ab
.
Matcheachquantity(a)(d) with(i), (ii), or (iii) if possible, or statethat nomatchexists.
(a) ln
_
a
b
_
(b)
lna
lnb
(c) e
lnalnb
(d) (lna)(lnb)
(i) lna +lnb (ii) lna lnb (iii)
a
b
3. Whichof thefollowingisequal to
d
dx
2
x
?
(a) 2
x
(b) (ln2)2
x
(c) x2
x1
(d)
1
ln2
2
x
solution Thederivativeof f (x) = 2
x
is
d
dx
2
x
= 2
x
ln2.
Hence, thecorrect answer is(b).
Findtheinverseof f (x) =
_
x
3
8anddetermineitsdomainandrange.
5. Findtheinverseof f (x) =
x 2
x 1
anddetermineitsdomainandrange.
solution Tondtheinverseof f (x) =
x2
x1
, wesolvey =
x2
x1
for x asfollows:
x 2= y(x 1) = yx y
x yx = 2y
x =
2y
1y
.
Therefore,
f
1
(x) =
2x
1x
=
x 2
x 1
.
Thedomainof f
1
istherangeof f , namely{x : x = 1}; therangeof f
1
isthedomainof f , namely{y : y = 1}.
Findadomainonwhichh(t ) = (t 3)
2
isone-to-oneanddeterminetheinverseonthisdomain.
7. Showthat g(x) =
x
x 1
isequal toitsinverseonthedomain{x : x = 1}.
solution Toshowthat g(x) =
x
x1
isequal toitsinverse, weneedtoshowthat for x = 1,
g (g(x)) = x.
First, wenoticethat for x = 1, g(x) = 1. Therefore,
g (g(x)) = g
_
x
x 1
_
=
x
x1
x
x1
1
=
x
x (x 1)
=
x
1
= x.
Describethegraphical interpretationof therelationg
(x) = 1/f
(g(x)), wheref (x) andg(x) areinversesof

eachother.
9. Supposethat g(x) istheinverseof f (x). Matchthefunctions(a)(d) withtheir inverses(i)(iv).
(a) f (x) +1
(b) f (x +1)
(c) 4f (x)
(d) f (4x)
(i) g(x)/4
(ii) g(x/4)
(iii) g(x 1)
(iv) g(x) 1
solution
(a) (iii): f (x) +1andg(x 1) areinversefunctions:
f (g(x 1)) +1= (x 1) +1= x;
g(f (x) +11) = g(f (x)) = x.
(b) (iv): f (x +1) andg(x) 1areinversefunctions:
f (g(x) 1+1) = f (g(x)) = x;
g(f (x +1)) 1= (x +1) 1= x.
(c) (ii): 4f (x) andg(x/4) areinversefunctions:
4f (g(x/4)) = 4(x/4) = x;
g(4f (x)/4) = g(f (x)) = x.
(d) (i): f (4x) andg(x)/4areinversefunctions:
f (4 g(x)/4) = f (g(x)) = x;
1
4
g(f (4x)) =
1
4
(4x) = x.
Findg
(8) whereg(x) istheinverseof adifferentiablefunctionf (x) suchthat f (1) = 8andf
(1) = 12.
11. Supposethat f (g(x)) = e
x
2
, whereg(1) = 2andg
(1) = 4. Findf
(2).
solution Wedifferentiatebothsidesof theequationf (g(x)) = e
x
2
toobtain,
f
(g(x)) g
(x) = 2xe
x
2
.
Settingx = 1yields
f
(g(1)) g
(1) = 2e.
Sinceg(1) = 2andg
(1) = 4, wend
f
(2) 4= 2e,
or
f
(2) =
e
2
.
Showthat if f (x) isafunctionsatisfyingf
(x) = f (x)
2
, thenitsinverseg(x) satisesg
(x) = x
2
.
13. f (x) = 9e
4x
solution
d
dx
9e
4x
= 36e
4x
.
f (x) = ln(4x
2
+1)
15. f (x) =
e
x
x
solution
d
dx
_
e
x
x
_
=
xe
x
e
x
x
2
=
e
x
(x +1)
x
2
.
f (x) = ln(x +e
x
)
17. G(s) = (ln(s))
2
solution
d
ds
(lns)
2
=
2lns
s
.
G(s) = ln(s
2
)
19. g(t ) = e
4t t
2
solution
d
dt
e
4t t
2
= (42t )e
4t t
2
.
g (t ) = t
2
e
1/t
21. f () = ln(sin)
solution
d
d
ln(sin) =
cos
sin
= cot.
f () = sin(ln)
23. f (x) = ln(e
x
4x)
solution
d
dx
ln(e
x
4x) =
e
x
4
e
x
4x
.
h(z) = sec(z +lnz)
25. f (x) = e
x+lnx
solution
d
dx
e
x+lnx
=
_
1+
1
x
_
e
x+lnx
.
f (x) = e
sin
2
x
27. h(y) = 2
1y
solution
d
dy
2
1y
= 2
1y
ln2.
h(y) =
1+e
y
1e
y
29. f (x) = 7
2x
solution
d
dx
7
2x
= 2ln7 7
2x
.
g(x) = tan
1
(lnx)
31. G(s) = cos
1
(s
1
)
solution
d
ds
cos
1
(s
1
) =
1
_
1
_
1
s
_
2
_
1
s
2
_
=
1
_
s
4
s
2
.
G(s) = tan
1
(
s)
33. f (x) = ln(csc
1
x)
solution
d
dx
ln(csc
1
x) =
1
|x|
_
x
2
1csc
1
x
.
f (x) = e
sec
1
x
35. R(s) = s
lns
solution Rewrite
R(s) =
_
e
lns
_
lns
= e
(lns)
2
.
Then
dR
ds
= e
(lns)
2
2lns
1
s
=
2lns
s
s
lns
.
Alternately, R(s) = s
lns
impliesthat lnR = ln
_
s
lns
_
= (lns)
2
. Thus,
1
R
dR
ds
= 2lns
1
s
or
dR
ds
=
2lns
s
s
lns
.
f (x) = (cos
2
x)
cosx
37. G(t ) = (sin
2
t )
t
solution Rewrite
G(t ) =
_
e
lnsin
2
t
_
t
= e
2t lnsint
.
Then
dG
dt
= e
2t lnsint
_
2t
cost
sint
+2lnsint
_
= 2(sin
2
t )
t
(t cott +lnsint ).
Alternately, G(t ) = (sin
2
t )
t
impliesthat lnG = t lnsin
2
t = 2t lnsint . Thus,
1
G
dG
dt
= 2t
cost
sint
+2lnsint,
and
dG
dt
= 2(sin
2
t )
t
(t cott +lnsint ).
h(t ) = t
(t
t
)
39. g(t ) = sinh(t
2
)
solution
d
dt
sinh(t
2
) = 2t cosh(t
2
).
h(y) = y tanh(4y)
41. g(x) = tanh
1
(e
x
)
solution
d
dx
tanh
1
(e
x
) =
1
1(e
x
)
2
e
x
=
e
x
1e
2x
.
g(t ) =
_
t
2
1sinh
1
t
43. Thetangent linetothegraphof y = f (x) at x = 4hasequationy = 2x + 12. Findtheequationof thetangent
linetoy = g(x) at x = 4, whereg(x) istheinverseof f (x).
solution Becausethetangent linetothegraphof y = f (x) at x = 4hasequationy = 2x +12, it followsthat
f (4) = 2(4) +12= 4 and f
(4) = 2.
Thus, g(4) = 4and
g
(4) =
1
f
(g(4))
=
1
f
(4)
=
1
2
,
whereg(x) istheinverseof f (x). Finally, theequationof thetangent linetoy = g(x) at x = 4is
y =
1
2
(x 4) +4=
1
2
x +6.
In Exercises 4446, let f (x) = xe
x
.
Plot f (x) andusethezoomfeaturetondtwosolutionsof f (x) = 0.3.
45. Showthatf (x) hasaninverseon[1, ). Letg(x) bethisinverse. Findthedomainandrangeof g(x) andcompute
g
(2e
2
).
solution Let f (x) = xe
x
. Thenf
(x) = e
x
(1 x). On[1, ), f
(x) < 0, sof (x) isdecreasingandtherefore

one-to-one.Itfollowsthatf (x) hasaninverseon[1, ).Letg(x) denotethisinverse.Becausef (1) = e
1
andf (x) 0
asx , thedomainof g(x) is(0, e
1
], andtherangeis[1, ).
Todetermineg
(2e
2
), weusetheformulag
(x) = 1/f
(g(x)). Becausef (2) = 2e

2
, it followsthat g(2e
2
) = 2.
Then,
g
(2e
2
) =
1
f
(g(2e
2
))
=
1
f
(2)
=
1
e
2
= e
2
.
Showthat f (x) = c hastwosolutionsif 0< c < e
1
.
47. Findthelocal extremaof f (x) = e
2x
4e
x
.
2x
4e
x
. Thenf
(x) = 2e
2x
4e
x
= 0whenx = ln2. Next, weusetheSecondDerivative
Test. Withf
(x) = 4e
2x
4e
x
, it followsthat
f
(ln2) = 4e
2ln2
4e
ln2
= 168= 8> 0.
Hence, x = ln2correspondstoalocal minimum. Since
f (ln2) = e
2ln2
4e
ln2
= 48= 4,
weconcludethat thepoint (ln2, 4) isalocal minimum.
Findthepointsof inectionof f (x) = ln(x
2
+1) anddeterminewhether theconcavitychangesfromuptodown
or viceversa.
In Exercises 4952, nd the local extrema and points of inection, and sketch the graph over the interval specied. Use
LHpitals Rule to determine the limits as x 0+ or x if necessary.
49. y = x lnx, x > 0
solution Let y = x lnx. Then
y
= lnx +x
_
1
x
_
= 1+lnx,
andy
=
1
x
. Solvingy
= 0yieldsthecritical pointx = e
1
. Sincey
(e
1
) = e > 0, thefunctionhasalocal minimum
at x = e
1
. y
ispositivefor x > 0, hencethefunctionisconcaveupfor x > 0andtherearenopointsof inection. As

x 0+ andasx , wend
lim
x0+
x lnx = lim
x0+
lnx
x
1
= lim
x0+
x
1
x
2
= lim
x0+
(x) = 0;
lim
x
x lnx = .
Thegraphisshownbelow:
2
1 2 3 4
4
6
y
x
y = xe
x
2
/2
51. y = x(lnx)
2
, x > 0
solution Let y = x(logx)
2
. Then
y
= (logx)
2
+x
2logx
x ln10
= (logx)
_
2
ln10
+logx
_
,
and
y
=
2logx
x ln10
+
2
ln10
1
x ln10
=
2
x ln10
_
logx +
1
ln10
_
.
Solvingy
= 0yieldsthecritical pointsx = 1andx = e

2
. Because
y
(1) =
2
(ln10)
2
> 0 and y
(e
2
) =
2(loge)
2
e
2
< 0,
weconcludethat thefunctionhasalocal minimumat x = 1andalocal maximumat x = e
2
. Weseethat y
> 0for
x > e
1
andy
< 0for 0< x < e

1
. Therefore, thereisapoint of inectionat x = e
1
. Asx 0+ andasx ,
wend
lim
x0+
x(logx)
2
= lim
x0+
(logx)
2
1/x
= lim
x0+
2logx
1
ln10
1
x
1/x
2
=
2
ln10
lim
x0+
logx
1/x
=
2
ln10
lim
x0+
1
ln10
1
x
1/x
2
=
2
(ln10)
2
lim
x0+
x = 0; and
lim
x
x(logx)
2
= .
Thegraphisshownbelow:
0.5
0.2
0.4
0.6
0.8
1.0
1.0 1.5 2.0 2.5 3.0
y
x
y = tan
1
_
x
2
4
_
In Exercises 5358, use logarithmic differentiation to nd the derivative.
53. y =
(x +1)
3
(4x 2)
2
solution Let y =
(x +1)
3
(4x 2)
2
. Then
lny = ln
_
(x +1)
3
(4x 2)
2
_
= ln(x +1)
3
ln(4x 2)
2
= 3ln(x +1) 2ln(4x 2).
Bylogarithmicdifferentiation,
y
y
=
3
x +1
2
4x 2
4=
3
x +1
4
2x 1
,
so
y
=
(x +1)
3
(4x 2)
2
_
3
x +1
4
2x 1
_
.
y =
(x +1)(x +2)
2
(x +3)(x +4)
55. y = e
(x1)
2
e
(x3)
2
solution Let y = e
(x1)
2
e
(x3)
2
. Then
lny = ln
_
e
(x1)
2
e
(x3)
2_
= ln
_
e
(x1)
2
+(x3)
2_
= (x 1)
2
+(x 3)
2
.
y
y
= 2(x 1) +2(x 3) = 4x 8,
so
y
= 4e
(x1)
2
e
(x3)
2
(x 2).
y =
e
x
sin
1
x
lnx
57. y =
e
3x
(x 2)
2
(x +1)
2
solution Let y =
e
3x
(x 2)
2
(x +1)
2
. Then
lny = ln
_
e
3x
(x 2)
2
(x +1)
2
_
= lne
3x
+ln(x 2)
2
ln(x +1)
2
= 3x +2ln(x 2) 2ln(x +1).
y
y
= 3+
2
x 2
2
x +1
,
so
y =
e
3x
(x 2)
2
(x +1)
2
_
3+
2
x 2
2
x +1
_
.
y = x
x
(x
lnx
)
59. ImageProcessingTheintensity of apixel inadigital imageismeasuredby anumber u between0and1. Often,
imagescanbeenhancedbyrescalingintensities(Figure1), wherepixelsof intensityu aredisplayedwithintensityg(u)
for asuitablefunctiong(u). Onecommonchoiceisthesigmoidal correction, denedfor constantsa, b by
g(u) =
f (u) f (0)
f (1) f (0)
where f (u) =
_
1+e
b(au)
_
1
Figure2shows that g(u) reduces theintensity of low-intensity pixels (whereg(u) < u) andincreases theintensity of
high-intensitypixels.
(a) Verifythat f
(u) > 0andusethistoshowthat g(u) increasesfrom0to1for 0 u 1.

(b) Wheredoesg(u) haveapoint of inection?
Original Sigmoidal correction
FIGURE 1
0.2 0.4 0.6 0.8 1.0
0.2
0.4
0.6
0.8
1.0
u
y
y = g(u)
y = u
FIGURE 2 Sigmoidal correctionwith
a = 0.47, b = 12.
solution
(a) Withf (u) = (1+e
b(au)
)
1
, it followsthat
f
(u) = (1+e
b(au)
)
2
be
b(au)
=
be
b(au)
(1+e
b(au)
)
2
> 0
for all u. Next, observethat
g(0) =
f (0) f (0)
f (1) f (0)
= 0, g(1) =
f (1) f (0)
f (1) f (0)
= 1,
and
g
(u) =
1
f (1) f (0)
f
(u) > 0
for all u. Thus, g(u) increasesfrom0to1for 0 u 1.
(b) Workingfrompart (a), wend
f
(u) =
b
2
e
b(au)
(2e
b(au)
1)
(1+e
b(au)
)
3
.
Because
g
(u) =
1
f (1) f (0)
f
(u),
it followsthat g(u) hasapoint of inectionwhen
2e
b(au)
1= 0 or u = a +
1
b
ln2.
Let N(t ) bethesizeof atumor (inunits of 10
6
cells) at timet (indays). Accordingto theGompertzModel,
dN/dt = N(a b lnN) wherea, b arepositiveconstants. Showthat themaximumvalueof N is e
a
b
andthat the
tumor increasesmost rapidlywhenN = e
a
b
1
.
In Exercises 6166, use the given substitution to evaluate the integral.
61.
_
(lnx)
2
dx
x
, u = lnx
solution Let u = lnx. Thendu =
dx
x
, and
_
(lnx)
2
dx
x
=
_
u
2
du =
u
3
3
+C =
(lnx)
3
3
+C.
_
dx
4x
2
+9
, u =
2x
3
63.
_
dx
_
e
2x
1
, u = e
x
solution Werst rewritetheintegrandintermsof e
x
. That is,
_
1
_
e
2x
1
dx =
_
1
_
e
2x
_
1e
2x
_
dx =
_
1
e
x
_
1e
2x
dx =
_
e
x
dx
_
1e
2x
Now, let u = e
x
. Thendu = e
x
dx, and
_
1
_
e
2x
1
dx =
_
du
_
1u
2
= sin
1
u +C = sin
1
(e
x
) +C.
_
cos
1
t dt
_
1t
2
, u = cos
1
t
65.
_
dt
t (1+(lnt )
2
)
, u = lnt
solution Let u = lnt . Then, du =
1
t
dt and
_
dt
t (1+(lnt )
2
)
=
_
du
1+u
2
= tan
1
u +C = tan
1
(lnt ) +C.
_
dt
cosh
2
t +sinh
2
t
, u = tanht
67.
_
e
92x
dx
solution Let u = 92x. Thendu = 2dx, and
_
e
92x
dx =
1
2
_
e
u
du =
1
2
e
u
+C =
1
2
e
92x
+C.
_
x
2
e
x
3
dx
69.
_
e
2x
sin(e
2x
) dx
solution Let u = e
2x
. Thendu = 2e
2x
dx, and
_
e
2x
sin
_
e
2x
_
dx =
1
2
_
sinu du =
cosu
2
+C =
1
2
cos
_
e
2x
_
+C.
_
cos(lnx) dx
x
71.
_
3
1
e
4x3
dx
solution
_
3
1
e
4x3
dx =
1
4
e
4x3
3
1
=
1
4
(e
9
e).
_
dx
x
lnx
73.
_
e
1
lnx dx
x
solution Let u = lnx. Thendu =
dx
x
andthenewlimitsof integrationareu = ln1= 0andu = lne = 1. Thus,
_
e
1
lnx dx
x
=
_
1
0
u du =
1
2
u
2
1
0
=
1
2
.
_
ln3
0
e
xe
x
dx
75.
_
2/3
1/3
dx
_
1x
2
solution
_
2/3
1/3
dx
_
1x
2
= sin
1
x
2/3
1/3
= sin
1
2
3
sin
1
1
3
.
_
12
4
dx
x
_
x
2
1
77.
_
1
0
cosh(2t ) dt
solution Let u = 2t . Thent =
u
2
anddt =
du
2
. Thenewlimitsof integrationareu = 0andu = 2. Thus,
_
1
0
cosh(2t ) dt =
1
2
_
2
0
coshu du =
1
2
sinhu
2
0
=
1
2
(sinh2sinh0) =
1
2
sinh2.
_
2
0
dt
4t +12
79.
_
3
0
x dx
x
2
+9
solution Let u = x
2
+9. Thendu = 2x dx, andthenewlimitsof integrationareu = 9andu = 18. Thus,
_
3
0
x dx
x
2
+9
=
1
2
_
18
9
du
u
=
1
2
lnu
18
9
=
1
2
(ln18ln9) =
1
2
ln
18
9
=
1
2
ln2.
_
3
0
dx
x
2
+9
81.
_
x dx
_
1x
4
solution Let u = x
2
_
1x
4
=
_
1u
2
. Thus,
_
x dx
_
1x
4
=
1
2
_
du
_
1u
2
=
1
2
sin
1
u +C =
1
2
sin
1
(x
2
) +C.
_
e
x
10
x
dx
83.
_
e
x
dx
(e
x
+2)
3
solution Let u = e
x
+2. Thendu = e
x
dx and
_
e
x
dx
(e
x
+2)
3
=
_
u
3
du =
1
2u
2
+C =
1
2(e
x
+2)
2
+C.
_
sin cose
cos
2
+1
d
85.
_
/6
0
tan2 d
solution
_
/6
0
tan2 d =
1
2
ln| sec2|
/6
0
=
1
2
ln2.
_
2/3
/3
cot
_
1
2
_
d
87.
_
sin
1
x dx
_
1x
2
solution Let u = sin
1
x. Thendu =
1
1x
2
dx and
_
sin
1
x dx
_
1x
2
=
_
u du =
1
2
u
2
+C =
1
2
(sin
1
x)
2
+C.
_
tanh5x dx
89.
_
sinh
3
x coshx dx
solution Let u = sinhx. Thendu = coshx dx and
_
sinh
3
x coshx dx =
_
u
3
du =
u
4
4
+C =
sinh
4
x
4
+C.
_
1
0
dx
25x
2
91.
_
4
0
dx
2x
2
+1
solution Let u =

2x. Thendu =

2dx, andthenewlimitsof integrationareu = 0andu = 4
2. Thus,
_
4
0
dx
2x
2
+1
=
_
4
2
0
1
2
du
u
2
+1
=
1
2
_
4
2
0
du
u
2
+1
=
1
2
tan
1
u
2
0
=
1
2
_
tan
1
(4
2) tan
1
0
_
=
1
2
tan
1
(4
2).
_
6
2
dx
x
_
x
2
+12
93. TheisotopeThorium-234hasahalf-lifeof 24.5days.
(a) Findthedifferential equationsatisedbytheamount y(t ) of Thorium-234inasampleat timet .
(b) At t = 0, asamplecontains2kgof Thorium-234. Howmuchremainsafter 1year?
solution
(a) Bytheequationfor half-life,
24.5=
ln2
k
, so k =
ln2
24.5
0.028days
1
.
Therefore, thedifferential equationfor y(t ) is
y
= 0.028y.
(b) If thereare2kgof Thorium-234att = 0,theny(t ) = 2e
0.028t
.Afteroneyear(365days),theamountof Thorium-234
is
y(365) = 2e
0.028(365)
= 7.2910
5
kg= 0.0729grams.
TheOldest Snack Food InBat Cave, NewMexico, archaeologists foundancient humanremains, including
cobsof poppingcorn, that hadaC
14
toC
12
ratioequal toaround48%of that foundinlivingmatter. Estimatethe
ageof thecorncobs.
95. TheC
14
to C
12
ratio of asampleis proportional to thedisintegrationrate(number of betaparticles emittedper
minute) thatismeasureddirectlywithaGeiger counter. Thedisintegrationrateof carboninalivingorganismis15.3beta
particles/minper gram. Findtheageof asamplethat emits9.5betaparticles/minper gram.
solution Let t betheageof thesampleinyears. Becausethedisintegrationratefor thesamplehas droppedfrom
15.3 beta particles/min per gramto 9.5 beta particles/min per gramand the C
14
to C
12
ratio is proportional to the
disintegrationrate, it followsthat
e
0.000121t
=
9.5
15.3
,
so
t =
1
0.000121
ln
9.5
15.3
3938.5.
Weconcludethat thesampleisapproximately3938.5yearsold.
Aninvestmentpaysout$5000attheendof theyear for 3years. ComputethePV, assuminganinterestrateof 8%.
97. In arst-order chemical reaction, thequantity y(t ) of reactant at timet satises y
= ky, wherek > 0. The

dependenceof k ontemperatureT (inkelvins) isgivenby theArrheniusequationk = Ae
E
a
/(RT )
, whereE
a
isthe
activationenergy (J -mol
1
), R = 8.314J -mol
1
-K
1
, andA is aconstant. Assumethat A = 72 10
12
hour
1
and
E
a
= 1.1 10
5
. Calculatedk/dT for T = 500andusetheLinear Approximationto estimatethechangeink if T is
raisedfrom500to510K.
solution Let
k = Ae
E
a
/(RT )
.
Then
dk
dT
=
AE
a
RT
2
e
E
a
/(RT )
.
For A = 7210
12
, R = 8.314andE
a
= 1.110
5
wehave
dk
dT
=
7210
12
1.110
5
8.314
e
1.110
5
8.314T
T
2
=
9.5310
17
e
1.3210
4
T
T
2
.
Thederivativefor T = 500isthus
dk
dT
T =500
=
9.5310
17
e
1.3210
4
500
500
2
12.27hours
1
K
1
.
Usingthelinear approximationwend
k
dk
dT
T =500
(510500) = 12.27 10= 122.7hours
1
.
Findthesolutionstoy
= 4(y 12) satisfyingy(0) = 20andy(0) = 0, andsketchtheir graphs.

99. Findthesolutionstoy
= 2y +8satisfyingy(0) = 3andy(0) = 4, andsketchtheir graphs.

solution First, rewritethedifferential equationasy
= 2(y 4); fromhereweseethat thegeneral solutionis

y(t ) = 4+Ce
2t
,
for someconstant C. If y(0) = 3, then
3= 4+Ce
0
and C = 1.
Thus, y(t ) = 4e
2t
. If y(0) = 4, then
4= 4+Ce
0
and C = 0;
hence, y(t ) = 4. Thegraphsof thetwosolutionsareshownbelow.
0.5
2
2
4
y
y = 4
y = 4 e
2t
x
0.5 1.0 1.5
Showthat y = sin
1
x satisesthedifferential equationy
= secy withinitial conditiony(0) = 0.

101. What isthelimit lim
t
y(t ) if y(t ) isasolutionof:
(a)
dy
dt
= 4(y 12)? (b)
dy
dt
= 4(y 12)?
(c)
dy
dt
= 4y 12?
solution
(a) Thegeneral solutionof
dy
dt
= 4(y 12) is y(t ) = 12+ Ce
4t
, whereC is anarbitrary constant. Regardless of
thevalueof C,
lim
t
y(t ) = lim
t
(12+Ce
4t
) = 12.
(b) Thegeneral solution of
dy
dt
= 4(y 12) is y(t ) = 12+ Ce
4t
, whereC is an arbitrary constant. Here, thelimit
dependsonthevalueof C. Specically,
lim
t
y(t ) = lim
t
(12+Ce
4t
) =
, C > 0
12, C = 0
, C < 0
(c) Thegeneral solution of
dy
dt
= 4y 12= 4(y +3) is y(t ) = 3+ Ce
4t
, whereC is an arbitrary constant.
Regardlessof thevalueof C,
lim
t
y(t ) = lim
t
(3+Ce
4t
) = 3.
Let A andB beconstants. Provethat if A > 0, thenall solutions of
dy
dt
+Ay = B approachthesamelimit as
t .
103. Anequipmentupgradecosting$1millionwill saveacompany$320,000peryearfor4years.Isthisagoodinvestment
if theinterest rateisr = 5%?What isthelargest interest ratethat wouldmaketheinvestment worthwhile?Assumethat
thesavingsarereceivedasalumpsumat theendof eachyear.
solution Withaninterest rateof r = 5%, thepresent valueof thefour paymentsis
$320,000
_
e
0.05
+e
0.1
+e
0.15
+e
0.2
_
= $1,131,361.78.
Asthisisgreater thanthe$1millioncost of theupgrade, thisisagoodinvestment. Todeterminethelargest interest rate
that wouldmaketheinvestment worthwhile, wemust solvetheequation
320,000
_
e
r
+e
2r
+e
3r
+e
4r
_
= 1,000,000
for r. Usingacomputer algebrasystem, wendr = 10.13%.
FindthePV of anincomestreampayingout continuously at arateof 5000e
0.1t
dollars per year for 5years,
assuminganinterest rateof r = 4%.
In Exercises 105108, let P(t ) denote the balance at time t (years) of an annuity that earns 5%interest continuously
compounded and pays out $2000/year continuously.
105. Findthedifferential equationsatisedbyP(t ).
solution Sincemoney iswithdrawncontinuously at arateof $2000ayear andthegrowthduetointerest is0.05P,
therateof changeof thebalanceis
P
(t ) = 0.05P 2000.
Thus, thedifferential equationsatisedbyP(t ) is
P
(t ) = 0.05(P 40, 000).

DetermineP(2) if P(0) = $5000.
107. Whendoestheannuityrunout of moneyif P(0) = $2000?
solution Inthepreviousexercise, wefoundthat
P(t ) = 40,000+Ce
0.05t
.
If P(0) = 2000, then
2000= 40,000+Ce
0.050
= 40,000+C
or
C = 38,000.
Thus,
P(t ) = 40,00038,000e
0.05t
.
Theannuityrunsout of moneywhenP(t ) = 0; that is, when
40,00038,000e
0.05t
= 0.
Solvingfor t yields
t =
1
0.05
ln
_
40,000
38,000
_
1.03.
Themoneyrunsout after roughly1.03years.
What istheminimuminitial balancethat will allowtheannuitytomakepaymentsindenitely?
In Exercises 109120, verify that LHpitals Rule applies and evaluate the limit.
109. lim
x3
4x 12
x
2
5x +6
solution Thegivenexpressionisanindeterminateformof type
0
0
, thereforeLHpitalsRuleapplies. Wend
lim
x3
4x 12
x
2
5x +6
= lim
x3
4
2x 5
=
4
2 35
= 4.
lim
x2
x
3
+2x
2
x 2
x
4
+2x
3
4x 8
111. lim
x0+
x
1/2
lnx
solution First rewritethefunctionas
lnx
x
1/2
. Thelimit isnowanindeterminateformof type
, hencewemayapply
LHpitalsRule. Wend
lim
x0+
x
1/2
lnx = lim
x0+
lnx
x
1/2
= lim
x0+
x
1
1
2
x
3/2
= lim
x0+
2x
1/2
= 0.
lim
t
ln(e
t
+1)
t
113. lim
0
2sin sin2
sin cos
solution Thegivenexpressionisanindeterminateformof type
0
0
; hence, wemayapplyLHpitalsRule. Wend
lim
0
2sin sin2
sin cos
= lim
0
2cos 2cos2
cos (cos sin)
= lim
0
2cos 2cos2
sin
= lim
0
2sin +4sin2
sin + cos
= lim
0
2cos +8cos2
cos +cos sin
=
2+8
1+10
= 3.
lim
x0
4+x 2
8
1+x
x
2
115. lim
t
ln(t +2)
log
2
t
solution Thelimit isanindeterminateformof type

lim
t
ln(t +2)
log
2
t
= lim
t
1
t +2
1
t ln2
= lim
t
t ln2
t +2
= lim
t
ln2
1
= ln2.
lim
x0
_
e
x
e
x
1
1
x
_
117. lim
y0
sin
1
y y
y
3
0
0
lim
y0
sin
1
y y
y
3
= lim
y0
1
1y
2
1
3y
2
= lim
y0
y(1y
2
)
3/2
6y
= lim
y0
(1y
2
)
3/2
6
=
1
6
.
lim
x1
_
1x
2
cos
1
x
119. lim
x0
sinh(x
2
)
coshx 1
0
0
lim
x0
sinh(x
2
)
coshx 1
= lim
x0
2x cosh(x
2
)
sinhx
= lim
x0
2cosh(x
2
) +4x
2
sinh(x
2
)
coshx
=
2+0
1
= 2.
lim
x0
tanhx sinhx
sinx x
121. ExplainwhyLHpitalsRulegivesnoinformationabout lim
x
2x sinx
3x +cos2x
. Evaluatethelimitbyanother
method.
solution As x , both 2x sinx and 3x + cos2x tend toward innity, so LHpitals Rule applies to
lim
x
2x sinx
3x +cos2x
; however, theresulting limit, lim
x
2cosx
32sin2x
, does not exist dueto theoscillation of sinx and
cosx.
Toevaluatethelimit, wenote
lim
x
2x sinx
3x +cos2x
= lim
x
2
sinx
x
3+
cos2x
x
=
2
3
.
Let f (x) beadifferentiablefunctionwithinverseg(x) suchthat f (0) = 0andf
(0) = 0. Provethat
lim
x0
f (x)
g(x)
= f
(0)
2
123. Calculatethelimit
lim
n
_
1+
4
n
_
n
solution Let t = n/4. Thenn = 4t and
lim
n
_
1+
4
n
_
n
= lim
t
_
1+
1
t
_
4t
= lim
t
_
_
1+
1
t
_
t
_
4
=
_
lim
t
_
1+
1
t
_
t
_
4
= e
4
.
Calculatethelimit
lim
n
_
1+
4
n
_
3n
125. Inthisexercise, weprovethat for all x > 0,
x
x
2
2
ln(1+x) x 1
(a) Showthat ln(1+x) =
_
x
0
dt
1+t
for x > 0.
(b) Verifythat 1t
1
1+t
1for all t > 0.
(c) Use(b) toproveEq. (1).
(d) VerifyEq. (1) for x = 0.5, 0.1, and0.01.
solution
(a) Let x > 0. Then
_
x
0
dt
1+t
= ln(1+t )
x
0
= ln(1+x) ln1= ln(1+x).
(b) For t > 0, 1+ t > 1, so
1
1+t
< 1. Moreover, (1 t )(1+ t ) = 1 t
2
< 1. Because1+ t > 0, it follows that
1t <
1
1+t
. Hence,
1t
1
1+t
1.
(c) Integratingeachexpressionintheresult frompart (b) fromt = 0tot = x yields
x
x
2
2
ln(1+x) x.
(d) For x = 0.5, x = 0.1andx = 0.01, weobtainthestringof inequalities
0.375 0.405465 0.5
0.095 0.095310 0.1
0.009950.00995033 0.01,
respectively.
Let
F(x) = x
_
x
2
12
_
x
1
_
t
2
1dt
Provethat F(x) andcosh
1
x differ by aconstant by showingthat they havethesamederivative. Thenprovethey
areequal byevaluatingbothat x = 1.
In Exercises 127130, let gd(y) = tan
1
(sinhy) be the so-called gudermannian, which arises in cartography. In a map
of the earth constructed by Mercator projection, points located y radial units from the equator correspond to points on
the globe of latitude gd(y).
127. Provethat
d
dy
gd(y) = sechy.
solution Let gd(y) = tan
1
(sinhy). Then
d
dy
gd(y) =
1
1+sinh
2
y
coshy =
1
coshy
= sechy,
wherewehaveusedtheidentity1+sinh
2
y = cosh
2
y.
Let f (y) = 2tan
1
(e
y
) /2. Provethat gd(y) = f (y). Hint: Showthat gd
(y) = f
(y) andf (0) = gd(0).

129. Showthat t (y) = sinh
1
(tany) istheinverseof gd(y) for 0 y < /2.
solution Let x = gd(y) = tan
1
(sinhy). Solvingfor y yieldsy = sinh
1
(tanx). Therefore,
gd
1
(y) = sinh
1
(tany).
Verifythat t (y) inExercise129satisest
(y) = secy andndavalueof a suchthat

t (y) =
_
y
a
dt
cost
131. UseLHpitalsRuletoprovethat for all a > 0andb > 0,
lim
n
_
a
1/n
+b
1/n
2
_
n
=
ab
solution
lim
n
ln
_
a
1/n
+b
1/n
2
_
n
= lim
n
n ln
_
a
1/n
+b
1/n
2
_
= lim
n
ln
_
a
1/n
+b
1/n
2
_
1
n
= lim
n
1
a
1/n
+b
1/n
_
a
1/n
lna
n
2

b
1/n
lnb
n
2
_
1
n
2
= lim
n
1
a
1/n
+b
1/n
_
a
1/n
lna +b
1/n
lnb
_
=
1
2
(lna +lnb) = ln
ab.
Hence,
lim
n
_
a
1/n
+b
1/n
2
_
n
= e
ln
ab
=
ab.
Let
F(x) =
_
x
2
dt
lnt
and G(x) =
x
lnx
Verifythat LHpitalsRulemaybeappliedtothelimit L lim
F(x)
andevaluateL
133. Let f (x) = e
Ax
2
/2
, whereA > 0. Givenanyn numbersa
1
, a
2
, . . . , a
n
, set
(x) = f (x a
1
)f (x a
2
) f (x a
n
)
(a) Assumen = 2 and provethat (x) attains its maximumvalueat theaveragex =
1
2
(a
1
+ a
2
). Hint: Showthat
d/dx ln(f (x)) = Ax andcalculate
(x) usinglogarithmicdifferentiation.
(b) Showthat for anyn, (x) attainsitsmaximumvalueat x =
1
n
(a
1
+a
2
+ +a
n
). Thisfact isrelatedtotherole
of f (x) (whosegraphisabell-shapedcurve) instatistics.
solution
(a) For n = 2wehave,
(x) = f (x a
1
) f (x a
2
) = e
A
2
(xa
1
)
2
e
A
2
(xa
2
)
2
= e
A
2
_
(xa
1
)
2
+(xa
2
)
2
_
.
Sincee
A
2
y
isadecreasingfunctionof y, itattainsitsmaximumvaluewherey isminimum. Therefore, wemustndthe
minimumvalueof
y = (x a
1
)
2
+(x a
2
)
2
= 2x
2
2(a
1
+a
2
) x +a
2
1
+a
2
2
.
Now, y
= 4x 2(a
1
+a
2
) = 0when
x =
a
1
+a
2
2
.
Weconcludethat (x) attainsamaximumvalueat thispoint.
(b) Wehave
(x) = e
A
2
(xa
1
)
2
e
A
2
(xa
2
)
2
e
A
2
(xa
n
)
2
= e
A
2
_
(xa
1
)
2
++(xa
n
)
2
_
.
Sincethefunctione
A
2
y
isadecreasingfunctionof y, it attainsamaximumvaluewherey isminimum. Thereforewe
must minimizethefunction
y = (x a
1
)
2
+(x a
2
)
2
+ +(x a
n
)
2
.
Wendthecritical pointsbysolving:
y
= 2(x a
1
) +2(x a
2
) + +2(x a
n
) = 0
2nx = 2(a
1
+a
2
+ +a
n
)
x =
a
1
+ +a
n
n
.
Weverifythatthispointcorrespondstheminimumvalueof y byexaminingthesignof y
atthispoint: y
= 2n > 0. We
concludethat y attainsaminimumvalueat thepoint x =
a
1
++a
n
n
, hence(x) attainsamaximumvalueat thispoint.
8 TECHNIQUES OF
INTEGRATION
8.1 Integration by Parts
1. WhichderivativeruleisusedtoderivetheIntegrationbyPartsformula?
solution TheIntegrationbyPartsformulaisderivedfromtheProduct Rule.
2. For eachof thefollowingintegrals, statewhether substitutionor IntegrationbyPartsshouldbeused:
_
x cos(x
2
) dx,
_
x cosx dx,
_
x
2
e
x
dx,
_
xe
x
2
dx
solution
(a)
_
x cos(x
2
) dx: usethesubstitutionu = x
2
.
(b)
_
x cosx dx: useIntegrationbyParts.
(c)
_
x
2
e
x
dx; useIntegrationbyParts.
(d)
_
xe
x
2
dx; usethesubstitutionu = x
2
.
3. Whyisu = cosx, v
= x apoor choicefor evaluating

_
x cosx dx?
solution Transformingv
= x into v =
1
2
x
2
increases thepower of x andmakes thenewintegral harder thanthe
original.
Exercises
InExercises16, evaluatetheintegral usingtheIntegrationbyPartsformulawiththegivenchoiceof u andv
.
1.
_
x sinx dx; u = x, v
= sinx
solution Usingthegivenchoiceof u andv
resultsin
u = x v = cosx
u
= 1 v
= sinx
UsingIntegrationbyParts,
_
x sinx dx = x(cosx)
_
(1)(cosx) dx = x cosx +
_
cosx dx = x cosx +sinx +C.
_
xe
2x
dx; u = x, v
= e
2x
3.
_
(2x +9)e
x
dx; u = 2x +9, v
= e
x
solution Usingu = 2x +9andv
= e
x
givesus
u = 2x +9 v = e
x
u
= 2 v
= e
x
IntegrationbyPartsgivesus
_
(2x +9)e
x
dx = (2x +9)e
x
_
2e
x
dx = (2x +9)e
x
2e
x
+C = e
x
(2x +7) +C.
446
S E C T I ON 8.1 Integration by Parts 447
_
x cos4x dx; u = x, v
= cos4x
5.
_
x
3
lnx dx; u = lnx, v
= x
3
solution Usingu = lnx andv
= x
3
givesus
u = lnx v =
1
4
x
4
u
=
1
x
v
= x
3
_
x
3
lnx dx = (lnx)
_
1
4
x
4
_
_ _
1
x
__
1
4
x
4
_
dx
=
1
4
x
4
lnx
1
4
_
x
3
dx =
1
4
x
4
lnx
1
16
x
4
+C =
x
4
16
(4lnx 1) +C.
_
tan
1
x dx; u = tan
1
x, v
= 1
InExercises736, evaluateusingIntegrationbyParts.
7.
_
(4x 3)e
x
dx
solution Let u = 4x 3andv
= e
x
. Thenwehave
u = 4x 3 v = e
x
u
= 4 v
= e
x
UsingIntegrationbyParts, weget
_
(4x 3)e
x
dx = (4x 3)(e
x
)
_
(4)(e
x
) dx
= e
x
(4x 3) +4
_
e
x
dx = e
x
(4x 3) 4e
x
+C = e
x
(4x +1) +C.
_
(2x +1)e
x
dx
9.
_
x e
5x+2
dx
solution Let u = x andv
= e
5x+2
. Thenwehave
u = x v =
1
5
e
5x+2
u
= 1 v
= e
5x+2
_
xe
5x+2
dx = x
_
1
5
e
5x+2
_
_
(1)
_
1
5
e
5x+2
_
dx =
1
5
xe
5x+2
1
5
_
e
5x+2
dx
=
1
5
xe
5x+2
1
25
e
5x+2
+C =
_
x
5
1
25
_
e
5x+2
+C
_
x
2
e
x
dx
11.
_
x cos2x dx
= cos2x. Thenwehave
u = x v =
1
2
sin2x
u
= 1 v
= cos2x
_
x cos2x dx = x
_
1
2
sin2x
_
_
(1)
_
1
2
sin2x
_
dx
=
1
2
x sin2x
1
2
_
sin2x dx =
1
2
x sin2x +
1
4
cos2x +C.
448 C HA P T E R 8 TECHNIQUES OF INTEGRATION
_
x sin(3x) dx
13.
_
x
2
sinx dx
solution Let u = x
2
andv
= sinx. Thenwehave
u = x
2
v = cosx
u
= 2x v
= sinx
_
x
2
sinx dx = x
2
(cosx)
_
2x(cosx) dx = x
2
cosx +2
_
x cosx dx.
Wemust applyIntegrationbyPartsagaintoevaluate
_
x cosx dx. Takingu = x andv
= cosx, weget
_
x cosx dx = x sinx
_
sinx dx = x sinx +cosx +C.
Pluggingthisintotheoriginal equationgivesus
_
x
2
sinx dx = x
2
cosx +2(x sinx +cosx) +C = x
2
cosx +2x sinx +2cosx +C.
_
x
2
cos3x dx
15.
_
e
x
sinx dx
solution Let u = e
x
andv
= sinx. Thenwehave
u = e
x
v = cosx
u
= e
x
v
= sinx
_
e
x
sinx dx = e
x
cosx
_
(e
x
)(cosx) dx = e
x
cosx
_
e
x
cosx dx.
_
e
x
cosx dx. Usingu = e
x
andv
= cosx, weget
_
e
x
cosx dx = e
x
sinx
_
(e
x
)(sinx) dx = e
x
sinx +
_
e
x
sinx dx.
Pluggingthisintotheoriginal equation, weget
_
e
x
sinx dx = e
x
cosx
_
e
x
sinx +
_
e
x
sinx dx
_
.
Solvingthisequationfor
_
e
x
sinx dx givesus
_
e
x
sinx dx =
1
2
e
x
(sinx +cosx) +C.
_
e
x
sin2x dx
17.
_
e
5x
sinx dx
solution Let u = sinx andv
= e
5x
. Thenwehave
u = sinx v =
1
5
e
5x
u
= cosx v
= e
5x
_
e
5x
sinx dx =
1
5
e
5x
sinx
_
cosx
_
1
5
e
5x
_
dx =
1
5
e
5x
sinx +
1
5
_
e
5x
cosx dx
ApplyIntegrationbyPartsagaintothisintegral, withu = cosx andv
= e
5x
toget
_
e
5x
cosx dx =
1
5
e
5x
cosx
1
5
_
e
5x
sinx dx
_
e
5x
sinx dx =
1
5
e
5x
sinx +
1
5
_
1
5
e
5x
cosx
1
5
_
e
5x
sinx dx
_
=
1
5
e
5x
sinx
1
25
e
5x
cosx
1
25
_
e
5x
sinx dx
_
e
5x
sinx dx givesus
_
e
5x
sinx dx =
5
26
e
5x
sinx
1
26
e
5x
cosx +C =
1
26
e
5x
(5sinx +cosx) +C
_
e
3x
cos4x dx
19.
_
x lnx dx
solution Let u = lnx andv
= x. Thenwehave
u = lnx v =
1
2
x
2
u
=
1
x
v
= x
_
x lnx dx =
1
2
x
2
lnx
_ _
1
x
__
1
2
x
2
_
dx
=
1
2
x
2
lnx
1
2
_
x dx =
1
2
x
2
lnx
1
2
_
x
2
2
_
+C =
1
4
x
2
(2lnx 1) +C.
_
lnx
x
2
dx
21.
_
x
2
lnx dx
= x
2
. Thenwehave
u = lnx v =
1
3
x
3
u
=
1
x
v
= x
2
_
x
2
lnx dx =
1
3
x
3
lnx
_
1
x
_
1
3
x
3
_
dx =
1
3
x
3
lnx
1
3
_
x
2
dx
=
1
3
x
3
lnx
1
3
_
x
3
3
_
+C =
x
3
3
_
lnx
1
3
_
+C.
_
x
5
lnx dx
23.
_
(lnx)
2
dx
solution Let u = (lnx)
2
andv
= 1. Thenwehave
u = (lnx)
2
v = x
u
=
2
x
lnx v
= 1
_
(lnx)
2
dx = (lnx)
2
(x)
_ _
2
x
lnx
_
x dx = x(lnx)
2
2
_
lnx dx.
_
lnx dx. Usingu = lnx andv
= 1, wehave
_
lnx dx = x lnx
_
1
x
x dx = x lnx
_
dx = x lnx x +C.
_
(lnx)
2
dx = x(lnx)
2
2(x lnx x) +C = x
_
(lnx)
2
2lnx +2
_
+C.
_
x(lnx)
2
dx
25.
_
x sec
2
x dx
= sec
2
x. Thenwehave
u = x v = tanx
u
= 1 v
= sec
2
x
_
x sec
2
x dx = x tanx
_
(1) tanx dx = x tanx ln| secx| +C.
_
x tanx secx dx
27.
_
cos
1
x dx
solution Let u = cos
1
x andv
= 1. Thenwehave
u = cos
1
x v = x
u
=
1
_
1x
2
v
= 1
_
cos
1
x dx = x cos
1
x
_
x
_
1x
2
dx.
Wecanevaluate
_
x
_
1x
2
dx bymakingthesubstitutionw = 1x
2
. Thendw = 2x dx, andwehave
_
cos
1
x dx = x cos
1
x
1
2
_
2x dx
_
1x
2
= x cos
1
x
1
2
_
w
1/2
dw
= x cos
1
x
1
2
(2w
1/2
) +C = x cos
1
x
_
1x
2
+C.
_
sin
1
x dx
29.
_
sec
1
x dx
solution Weareforcedtochooseu = sec
1
x, v
= 1, sothat u
=
1
x
x
2
1
andv = x. UsingIntegrationbyparts,
weget:
_
sec
1
x dx = x sec
1
x
_
x dx
x
_
x
2
1
= x sec
1
x
_
dx
_
x
2
1
.
Viathesubstitution
_
x
2
1= tan (sothat x = sec anddx = sec tand), weget:
_
sec
1
x dx = x sec
1
x
_
sec tand
tan
= x sec
1
x
_
secd
= x sec
1
x ln| sec +tan| +C = x sec
1
x ln|x +
_
x
2
1| +C.
_
x5
x
dx
31.
_
3
x
cosx dx
solution Let u = cosx andv
= 3
x
. Thenwehave
u = cosx v =
3
x
ln3
u
= sinx v
= 3
x
_
3
x
cosx dx =
3
x
ln3
cosx +
1
ln3
_
3
x
sinx dx
ApplyIntegrationbyPartstotheremainingintegral, withu = sinx andv
= 3
x
; then
_
3
x
sinx dx =
3
x
ln3
sinx
1
ln3
_
3
x
cosx dx
Plugthisintotherst equationtoget
_
3
x
cosx dx =
3
x
ln3
cosx +
1
ln3
_
3
x
ln3
sinx
1
ln3
_
3
x
cosx dx
_
=
3
x
ln3
cosx +
3
x
(ln3)
2
sinx
1
(ln3)
2
_
3
x
cosx dx
Solvingfor
_
3
x
cosx dx gives
_
3
x
cosx dx =
3
x
ln3cosx
1+(ln3)
2
+
3
x
sinx
1+(ln3)
2
+C =
3
x
1+(ln3)
2
(ln3cosx +sinx) +C
_
x sinhx dx
33.
_
x
2
coshx dx
solution Let u = x
2
, v
= coshx. Then
u = x
2
v = sinhx
u
= 2x v
= coshx
IntegrationbyPartsgivesus(alongwithExercise32)
_
x
2
coshx dx = x
2
sinhx 2
_
x sinhx, dx = x
2
sinhx 2x coshx +2sinhx +C
_
cosx coshx dx
35.
_
tanh
1
4x dx
solution Usingu = tanh
1
4x andv
= 1givesus
u = tanh
1
4x v = x
u
=
4
116x
2
v
= 1
_
tanh
1
4x dx = x tanh
1
4x
_ _
4
116x
2
_
x dx.
For theintegral ontheright well usethesubstitutionw = 116x
2
, dw = 32x dx. Thenwehave
_
tanh
1
4x dx = x tanh
1
4x +
1
8
_
dw
w
= x tanh
1
4x +
1
8
ln|w| +C
= x tanh
1
4x +
1
8
ln|116x
2
| +C.
_
sinh
1
x dx
InExercises37and38, evaluateusingsubstitutionandthenIntegrationbyParts.
37.
_
e
x
dx Hint: Let u = x
1/2
solution Let w = x
1/2
. Thendw =
1
2
x
1/2
dx, or dx = 2x
1/2
dw = 2wdw. Now,
_
e
x
dx = 2
_
we
w
dw.
UsingIntegrationbyPartswithu = w andv
= e
w
, weget
2
_
we
w
dw = 2(we
w
e
w
) +C.
Substitutingback, wend
_
e
x
dx = 2e
x
(
x 1) +C.
_
x
3
e
x
2
dx
InExercises3948, evaluateusingIntegrationbyParts, substitution, or bothif necessary.
39.
_
x cos4x dx
= cos4x. Thenwehave
u = x v =
1
4
sin4x
u
= 1 v
= cos4x
_
x cos4x dx =
1
4
x sin4x
_
(1)
1
4
sin4x dx =
1
4
x sin4x
1
4
_
1
4
cos4x
_
+C
=
1
4
x sin4x +
1
16
cos4x +C.
_
ln(lnx) dx
x
41.
_
x dx
x +1
solution Let u = x +1. Thendu = dx, x = u 1, and
_
x dx
x +1
=
_
(u 1) du
u
=
_ _
u
u
_
du =
_
(u
1/2
u
1/2
) du
=
2
3
u
3/2
2u
1/2
+C =
2
3
(x +1)
3/2
2(x +1)
1/2
+C.
_
x
2
(x
3
+9)
15
dx
43.
_
cosx ln(sinx) dx
solution Let w = sinx. Thendw = cosx dx, and
_
cosx ln(sinx) dx =
_
lnwdw.
NowuseIntegrationbyPartswithu = lnw andv
= 1. Thenu
= 1/w andv = w, whichgivesus

_
cosx ln(sinx) dx =
_
lnwdw = wlnw w +C = sinx ln(sinx) sinx +C.
_
sin
x dx
45.
_

xe
x
dx
solution Let w =
x. Thendw =
1
2
x
dx and
_

xe
x
dx = 2
_
w
2
e
w
dw.
Now, useIntegrationbyPartswithu = w
2
andv
= e
w
. Thisgives
_

xe
x
dx = 2
_
w
2
e
w
dw = 2w
2
e
w
4
_
we
w
dw.
WeneedtouseIntegrationbyPartsagain, thistimewithu = w andv
= e
w
. Wend
_
we
w
dw = we
w
_
e
w
dw = we
w
e
w
+C;
nally,
_

xe
x
dx = 2w
2
e
w
4we
w
+4e
w
+C = 2xe
x
4
xe
x
+4e
x
+C.
_
tan
x dx
x
47.
_
ln(lnx) lnx dx
x
solution Let w = lnx. Thendw = dx/x, and
_
ln(lnx) lnx dx
x
=
_
wlnwdw.
NowuseIntegrationbyParts, withu = lnw andv
= w. Then,
u = lnw v =
1
2
w
2
u
= w
1
v
= w
and
_
ln(lnx) lnx dx
x
=
1
2
w
2
lnw
1
2
_
wdw =
1
2
w
2
lnw
1
2
_
w
2
2
_
+C
=
1
2
(lnx)
2
ln(lnx)
1
4
(lnx)
2
+C =
1
4
(lnx)
2
[2ln(lnx) 1] +C.
_
sin(lnx) dx
InExercises4954, computethedeniteintegral.
49.
_
3
0
xe
4x
dx
solution Let u = x, v
= e
4x
. Thenu
= 1andv =
1
4
e
4x
. UsingIntegrationbyParts,
_
3
0
xe
4x
dx =
_
1
4
xe
4x
_
3
0
1
4
_
3
0
e
4x
dx =
3
4
e
12
1
16
e
12
+
1
16
=
11
16
e
12
+
1
16
_
/4
0
x sin2x dx
51.
_
2
1
x lnx dx
= x. Thenu
=
1
x
andv =
1
2
x
2
. UsingIntegrationbyPartsgives
_
2
1
x lnx dx =
_
1
2
x
2
lnx
_
2
1
1
2
_
2
1
x dx = 2ln2
1
4
x
2
2
1
= 2ln2
3
4
_
e
1
lnx dx
x
2
53.
_

0
e
x
sinx dx
solution Let u = sinx andv
= e
x
; thenu
= cosx andv = e
x
. IntegrationbyPartsgives
_

0
e
x
sinx dx = e
x
sinx
0

_

0
e
x
cosx dx =
_

0
e
x
cosx dx
Applyintegrationbypartsagaintothisintegral, withu = cosx andv
= e
x
; thenu
= sinx andv = e
x
, soweget
_

0
e
x
sinx dx =
_
_
e
x
cosx
_
0
+
_

0
e
x
sinx dx
_
= e
+1
_

0
e
x
sinx dx
Solvingfor
_

0
e
x
sinx dx gives
_

0
e
x
sinx dx =
e
+1
2
_
1
0
tan
1
x dx
55. UseEq. (5) toevaluate
_
x
4
e
x
dx.
solution
_
x
4
e
x
dx = x
4
e
x
4
_
x
3
e
x
dx = x
4
e
x
4
_
x
3
e
x
3
_
x
2
e
x
dx
_
= x
4
e
x
4x
3
e
x
+12
_
x
2
e
x
dx = x
4
e
x
4x
3
e
x
+12
_
x
2
e
x
2
_
xe
x
dx
_
= x
4
e
x
4x
3
e
x
+12x
2
e
x
24
_
xe
x
dx = x
4
e
x
4x
3
e
x
+12x
2
e
x
24
_
xe
x
_
e
x
dx
_
= x
4
e
x
4x
3
e
x
+12x
2
e
x
24
_
xe
x
e
x
_
+C.
Thus,
_
x
4
e
x
dx = e
x
(x
4
4x
3
+12x
2
24x +24) +C.
UsesubstitutionandthenEq. (5) toevaluate
_
x
4
e
7x
dx.
57. Findareductionformulafor
_
x
n
e
x
dx similar toEq. (5).
solution Let u = x
n
andv
= e
x
. Then
u = x
n
v = e
x
u
= nx
n1
v
= e
x
_
x
n
e
x
dx = x
n
e
x
_
nx
n1
(e
x
) dx = x
n
e
x
+n
_
x
n1
e
x
dx.
Evaluate
_
x
n
lnx dx for n = 1. Whichmethodshouldbeusedtoevaluate
_
x
1
lnx dx?
InExercises5966, indicateagoodmethodfor evaluatingtheintegral (but donot evaluate). Your choicesarealgebraic
manipulation, substitution(specifyu anddu), andIntegrationbyParts(specifyu andv
). If itappearsthatthetechniques
youhavelearnedthusfar arenot sufcient, statethis.
59.
_

x lnx dx
solution UseIntegrationbyParts, withu = lnx andv
x.
_
x
2
x
2x
dx
61.
_
x
3
dx
_
4x
2
solution Usesubstitution, followedby algebraic manipulation: Let u = 4 x
2
. Thendu = 2x dx, x
2
= 4 u,
and
_
x
3
_
4x
2
dx =
1
2
_
(x
2
)(2x dx)
u
=
1
2
_
(4u)(du)
u
=
1
2
_ _
4
u
_
du.
_
dx
_
4x
2
63.
_
x +2
x
2
+4x +3
dx
solution Usesubstitution. Let u = x
2
+4x +3; thendu = 2x +4dx = 2(x +2) dx, and
_
x +2
x
2
+4x +3
dx =
1
2
_
1
u
du
_
dx
(x +2)(x
2
+4x +3)
65.
_
x sin(3x +4) dx
solution UseIntegrationbyParts, withu = x andv
= sin(3x +4).
_
x cos(9x
2
) dx
67. Evaluate
_
(sin
1
x)
2
dx. Hint: UseIntegrationbyPartsrst andthensubstitution.
solution First useintegrationbypartswithv
= 1toget
_
(sin
1
x)
2
dx = x(sin
1
x)
2
2
_
x sin
1
x dx
_
1x
2
.
Nowusesubstitutionontheintegral ontheright, withu = sin
1
x. Thendu = dx/
_
1x
2
andx = sinu, andweget
(usingIntegrationbyPartsagain)
_
x sin
1
x dx
_
1x
2
=
_
u sinu du = u cosu +sinu +C =
_
1x
2
sin
1
x +x +C.
wherecosu =
_
1sin
2
u =
_
1x
2
. Sothenal answer is
_
(sin
1
x)
2
dx = x(sin
1
x)
2
+2
_
1x
2
sin
1
x 2x +C.
Evaluate
_
(lnx)
2
dx
x
2
. Hint: Usesubstitutionrst andthenIntegrationbyParts.
69. Evaluate
_
x
7
cos(x
4
) dx.
solution First, let w = x
4
. Thendw = 4x
3
dx and
_
x
7
cos(x
4
) dx =
1
4
_
wcosx dw.
Now, useIntegrationbyPartswithu = w andv
= cosw. Then
_
x
7
cos(x
4
) dx =
1
4
_
wsinw
_
sinwdw
_
=
1
4
wsinw +
1
4
cosw +C =
1
4
x
4
sin(x
4
) +
1
4
cos(x
4
) +C.
Findf (x), assumingthat
_
f (x)e
x
dx = f (x)e
x
_
x
1
e
x
dx
71. Findthevolumeof thesolidobtainedbyrevolvingtheregionunder y = e
x
for 0 x 2about they-axis.
solution BytheMethodof Cylindrical Shells, thevolumeV of thesolidis
V =
_
b
a
(2r)hdx = 2
_
2
0
xe
x
dx.
UsingIntegrationbyPartswithu = x andv
= e
x
, wend
V = 2 (xe
x
e
x
)
2
0
= 2
_
(2e
2
e
2
) (01)
_
= 2(e
2
+1).
Findtheareaenclosedbyy = lnx andy = (lnx)
2
.
73. Recall that thepresent value(PV) of aninvestment that paysout incomecontinuously at arateR(t ) for T yearsis
_
T
0
R(t )e
rt
dt , wherer istheinterest rate. FindthePV if R(t ) = 5000+100t $/year, r = 0.05andT = 10years.
solution Thepresent valueisgivenby
PV =
_
T
0
R(t )e
rt
dt =
_
10
0
(5000+100t )e
rt
dt = 5000
_
10
0
e
rt
dt +100
_
10
0
t e
rt
dt.
UsingIntegrationbyPartsfor theintegral ontheright, withu = t andv
= e
rt
, wend
PV = 5000
_
1
r
e
rt
_
10
0
+100
_
_
t
r
e
rt
_
10
0

_
10
0
1
r
e
rt
dt
_
=
5000
r
e
rt
10
0

100
r
_
t e
rt
+
1
r
e
rt
_
10
0
=
5000
r
(e
10r
1)
100
r
__
10e
10r
+
1
r
e
10r
_
_
0+
1
r
__
= e
10r
_
5000
r
1000
r
100
r
2
_
+
5000
r
+
100
r
2
=
5000r +100e
10r
(6000r +100)
r
2
.
Derivethereductionformula
_
(lnx)
k
dx = x(lnx)
k
k
_
(lnx)
k1
dx
75. UseEq. (6) tocalculate
_
(lnx)
k
dx for k = 2, 3.
solution
_
(lnx)
2
dx = x(lnx)
2
2
_
lnx dx = x(lnx)
2
2(x lnx x) +C = x(lnx)
2
2x lnx +2x +C;
_
(lnx)
3
dx = x(lnx)
3
3
_
(lnx)
2
dx = x(lnx)
3
3
_
x(lnx)
2
2x lnx +2x
_
+C
= x(lnx)
3
3x(lnx)
2
+6x lnx 6x +C.
Derivethereductionformulas
_
x
n
cosx dx = x
n
sinx n
_
x
n1
sinx dx
_
x
n
sinx dx = x
n
cosx +n
_
x
n1
cosx dx
77. Provethat
_
xb
x
dx = b
x
_
x
lnb
1
ln
2
b
_
+C.
= b
x
. Thenu
= 1andv = b
x
/ lnb. UsingIntegrationbyParts, weget
_
x b
x
dx =
xb
x
lnb
1
lnb
_
b
x
dx =
xb
x
lnb
1
lnb
b
x
lnb
+C = b
x
_
x
lnb
1
(lnb)
2
_
+C.
DeneP
n
(x) by
_
x
n
e
x
dx = P
n
(x) e
x
+C
79. TheIntegrationbyPartsformulacanbewritten
_
u(x)v(x) dx = u(x)V(x)
_
u
(x)V(x) dx 7
whereV(x) satisesV
(x) = v(x).
(a) Showdirectly that theright-handsideof Eq. (7) does not changeif V(x) is replacedby V(x) + C, whereC is a
constant.
(b) Useu = tan
1
x andv = x inEq. (7) to calculate
_
x tan
1
x dx, but carry out thecalculationtwice: rst with
V(x) =
1
2
x
2
andthenwithV(x) =
1
2
x
2
+
1
2
. Whichchoiceof V(x) resultsinasimpler calculation?
solution
(a) ReplacingV(x) withV(x) +C intheexpressionu(x)V(x)
_
V(x)u
(x) dx, weget

u(x)(V(x) +C)
_
(V(x) +C)u
(x) dx = u(x)V(x) +u(x)C

_
V(x)u
(x) dx C
_
u
(x) dx
= u(x)V(x)
_
V(x)u
(x) dx +C
_
u(x)
_
u
(x) dx
_
= u(x)V(x)
_
V(x)u
(x) dx +C [u(x) u(x)]

= u(x)V(x)
_
V(x)u
(x) dx.
(b) If weevaluate
_
x tan
1
x dx withu = tan
1
x andv
= x, andif wedontaddaconstanttov, IntegrationbyParts

givesus
_
x tan
1
x dx =
x
2
2
tan
1
x
1
2
_
x
2
dx
x
2
+1
.
Theintegral ontheright requiresalgebraicmanipulationinorder toevaluate. But if wetakeV(x) =
1
2
x
2
+
1
2
insteadof
V(x) =
1
2
x
2
, then
_
x tan
1
x dx =
_
1
2
x
2
+
1
2
_
tan
1
x
1
2
_
x
2
+1
x
2
+1
dx =
1
2
(x
2
+1) tan
1
x
1
2
x +C
=
1
2
(x
2
tan
1
x x +tan
1
x) +C.
Proveintwowaysthat
_
a
0
f (x) dx = af (a)
_
a
0
xf
(x) dx
First use Integration by Parts. Then assume f (x) is increasing. Use the substitution u = f (x) to prove that
_
a
0
xf
(x) dx isequal totheareaof theshadedregioninFigure1andderiveEq. (8) asecondtime.

81. Assumethat f (0) = f (1) = 0andthat f
exists. Prove
_
1
0
f
(x)f (x) dx =
_
1
0
f
(x)
2
dx 9
Usethis to provethat if f (0) = f (1) = 0andf
(x) = f (x) for someconstant , then < 0. Canyouthink of a

functionsatisfyingtheseconditionsfor some?
solution Let u = f (x) andv
= f
(x). UsingIntegrationbyParts, weget

_
1
0
f
(x)f (x) dx = f (x)f
(x)
1
0
_
1
0
f
(x)
2
dx = f (1)f
(1) f (0)f
(0)
_
1
0
f
(x)
2
dx =
_
1
0
f
(x)
2
dx.
Nowassumethat f
(x) = f (x) for someconstant . Then

_
1
0
f
(x)f (x) dx =
_
1
0
[f (x)]
2
dx =
_
1
0
f
(x)
2
dx < 0.
Since
_
1
0
[f (x)]
2
dx > 0, wemust have < 0. An exampleof afunction satisfying theseproperties for some is
f (x) = sinx.
Set I (a, b) =
_
1
0
x
a
(1x)
b
dx, wherea, b arewholenumbers.
(a) Usesubstitutiontoshowthat I (a, b) = I (b, a).
(b) Showthat I (a, 0) = I (0, a) =
1
a +1
.
(c) Provethat for a 1andb 0,
I (a, b) =
a
b +1
I (a 1, b +1)
(d) Use(b) and(c) tocalculateI (1, 1) andI (3, 2).
( ) Sh h I ( b)
a! b!
83. Let I
n
=
_
x
n
cos(x
2
) dx andJ
n
=
_
x
n
sin(x
2
) dx.
(a) Findareductionformulathat expressesI
n
intermsof J
n2
. Hint: Writex
n
cos(x
2
) asx
n1
(x cos(x
2
)).
(b) Usetheresult of (a) toshowthat I
n
canbeevaluatedexplicitlyif n isodd.
(c) EvaluateI
3
.
S E C T I ON 8.2 Trigonometric Integrals 457
solution
(a) IntegrationbyPartswithu = x
n1
andv
= x cos(x
2
) dx yields
I
n
=
1
2
x
n1
sin(x
2
)
n 1
2
_
x
n2
sin(x
2
) dx =
1
2
x
n1
sin(x
2
)
n 1
2
J
n2
.
(b) If n isodd, thereductionprocesswill eventuallyleadtoeither
_
x cos(x
2
) dx or
_
x sin(x
2
) dx,
bothof whichcanbeevaluatedusingthesubstitutionu = x
2
.
(c) Startingwiththereductionformulafrompart (a), wend
I
3
=
1
2
x
2
sin(x
2
)
2
2
_
x sin(x
2
) dx =
1
2
x
2
sin(x
2
) +
1
2
cos(x
2
) +C.
8.2 Trigonometric Integrals
1. Describethetechniqueusedtoevaluate
_
sin
5
x dx.
solution Becausethesinefunctionisraisedtoanoddpower, rewritesin
5
x = sinx sin
4
x = sinx(1cos
2
x)
2
and
thensubstituteu = cosx.
2. Describeawayof evaluating
_
sin
6
x dx.
solution Repeatedlyusethereductionformulafor powersof sinx.
3. Arereductionformulasneededtoevaluate
_
sin
7
x cos
2
x dx?Whyor whynot?
solution No, areductionformulaisnot neededbecausethesinefunctionisraisedtoanoddpower.
4. Describeawayof evaluating
_
sin
6
x cos
2
x dx.
solution Becausebothtrigonometricfunctionsareraisedtoevenpowers, writecos
2
x = 1 sin
2
x andthenapply
thereductionformulafor powersof thesinefunction.
5. Whichintegral requiresmoreworktoevaluate?
_
sin
798
x cosx dx or
_
sin
4
x cos
4
x dx
Explainyour answer.
solution Therst integral canbeevaluatedusingthesubstitutionu = sinx, whereasthesecondintegral requiresthe
useof reductionformulas. Thesecondintegral thereforerequiresmoreworktoevaluate.
Exercises
InExercises16, usethemethodfor oddpowerstoevaluatetheintegral.
1.
_
cos
3
x dx
solution Usetheidentitycos
2
x = 1sin
2
x torewritetheintegrand:
_
cos
3
x dx =
_
_
1sin
2
x
_
cosx dx.
Nowusethesubstitutionu = sinx, du = cosx dx:
_
cos
3
x dx =
_
_
1u
2
_
du = u
1
3
u
3
+C = sinx
1
3
sin
3
x +C.
_
sin
5
x dx
3.
_
sin
3
cos
2
d
solution Writesin
3
= sin
2
sin = (1cos
2
) sin. Then
_
sin
3
cos
2
d =
_
_
1cos
2
_
cos
2
sin d.
Nowusethesubstitutionu = cos, du = sin d:
_
sin
3
cos
2
d =
_
_
1u
2
_
u
2
du =
_
_
u
2
u
4
_
du
=
1
3
u
3
+
1
5
u
5
+C =
1
3
cos
3
+
1
5
cos
5
+C.
_
sin
5
x cosx dx
5.
_
sin
3
t cos
3
t dt
solution Writesin
3
t = (1cos
2
t ) sint dt . Then
_
sin
3
t cos
3
t dt =
_
(1cos
2
t ) cos
3
t sint dt =
_
_
cos
3
t cos
5
t
_
sint dt.
Nowusethesubstitutionu = cost , du = sint dt :
_
sin
3
t cos
3
t dt =
_
_
u
3
u
5
_
du =
1
4
u
4
+
1
6
u
6
+C =
1
4
cos
4
t +
1
6
cos
6
t +C.
_
sin
2
x cos
5
x dx
7. Findtheareaof theshadedregioninFigure1.
x
y
y= cos
3
x
1
1
p 3p
2
p
2
FIGURE 1 Graphof y = cos
3
x.
solution First evaluate the indenite integral by writing cos
3
x = (1 sin
2
x) cosx, and using the substitution
u = sinx, du = cosx dx:
_
cos
3
x dx =
_
_
1sin
2
x
_
cosx dx =
_
_
1u
2
_
du = u
1
3
u
3
+C = sinx
1
3
sin
3
x +C.
Theareaisgivenby
A =
_
/2
0
cos
3
x dx
_
3/2
/2
cos
3
x dx =
_
sinx
1
3
sin
3
x
_
/2
0

_
sinx
1
3
sin
3
x
_
3/2
/2
=
__
sin
2

1
3
sin
3

2
_
0
_
__
sin
3
2

1
3
sin
3
3
2
_
_
sin
2

1
3
sin
3

2
__
= 1
1
3
(1)
3
(1) +
1
3
(1)
3
+1
1
3
(1)
3
= 2.
Usetheidentitysin
2
x = 1cos
2
x towrite
_
sin
2
x cos
2
x dx asasumof twointegrals, andthenevaluateusing
thereductionformula.
InExercises912, evaluatetheintegral usingmethodsemployedinExamples3and4.
9.
_
cos
4
y dy
solution Usingthereductionformulafor cos
m
y, weget
_
cos
4
y dy =
1
4
cos
3
y siny +
3
4
_
cos
2
y dy =
1
4
cos
3
y siny +
3
4
_
1
2
cosy siny +
1
2
_
dy
_
=
1
4
cos
3
y siny +
3
8
cosy siny +
3
8
y +C.
_
cos
2
sin
2
d
11.
_
sin
4
x cos
2
x dx
solution Usetheidentitycos
2
x = 1sin
2
x towrite:
_
sin
4
x cos
2
x dx =
_
sin
4
x
_
1sin
2
x
_
dx =
_
sin
4
x dx
_
sin
6
x dx.
Usingthereductionformulafor sin
m
x:
_
sin
4
x cos
2
x dx =
_
sin
4
x dx
_
1
6
sin
5
x cosx +
5
6
_
sin
4
x dx
_
=
1
6
sin
5
x cosx +
1
6
_
sin
4
x dx =
1
6
sin
5
x cosx +
1
6
_
1
4
sin
3
x cosx +
3
4
_
sin
2
x dx
_
=
1
6
sin
5
x cosx
1
24
sin
3
x cosx +
1
8
_
sin
2
x dx
=
1
6
sin
5
x cosx
1
24
sin
3
x cosx +
1
8
_
1
2
sinx cosx +
1
2
_
dx
_
=
1
6
sin
5
x cosx
1
24
sin
3
x cosx
1
16
sinx cosx +
1
16
x +C.
_
sin
2
x cos
6
x dx
InExercises13and14, evaluateusingEq. (13).
13.
_
sin
3
x cos
2
x dx
solution First rewritesin
3
x = sinx sin
2
x = sinx(1cos
2
x), sothat
_
sin
3
x cos
2
x dx =
_
sinx(1cos
2
x) cos
2
x dx =
_
sinx(cos
2
x cos
4
x) dx
Nowmakethesubstitutionu = cosx, du = sinx dx:
_
sinx(cos
2
x cos
4
x) dx =
_
u
2
u
4
du =
1
5
u
5
1
3
u
3
+C =
1
5
cos
5
x
1
3
cos
3
x +C
_
sin
2
x cos
4
x dx
InExercises1518, evaluatetheintegral usingthemethoddescribedonpage409andthereductionformulasonpage
423asnecessary.
15.
_
tan
3
x secx dx
solution Use the identity tan
2
x = sec
2
x 1 to rewrite tan
3
x secx = (sec
2
x 1) secx tanx. Then use the
substitutionu = secx, du = secx tanx dx:
_
tan
3
x secx dx =
_
(sec
2
x 1) secx tanx dx =
_
u
2
1du =
1
3
u
3
u +C =
1
3
sec
3
x secx +C
_
tan
2
x secx dx
17.
_
tan
2
x sec
4
x dx
solution First usetheidentitytan
2
x = sec
2
x 1:
_
tan
2
x sec
4
x dx =
_
(sec
2
x 1) sec
4
x dx =
_
sec
6
x sec
4
x dx =
_
sec
6
x dx
_
sec
4
x, dx
Weevaluatethesecondintegral usingthereductionformula:
_
sec
4
x dx =
1
3
tanx sec
2
x +
2
3
_
sec
2
x dx
=
1
3
tanx sec
2
x +
2
3
tanx
Then
_
sec
6
x dx =
1
5
tanx sec
4
x +
4
5
_
sec
4
x dx
=
1
5
tanx sec
4
x +
4
5
_
1
3
tanx sec
2
x +
2
3
tanx
_
=
1
5
tanx sec
4
x +
4
15
tanx sec
2
x +
8
15
tanx
sothat
_
tan
2
x sec
4
x dx =
_
sec
6
x dx
_
sec
4
x dx
=
1
5
tanx sec
4
x
1
15
tanx sec
2
x
2
15
tanx +C
_
tan
8
x sec
2
x dx
InExercises1922, evaluateusingmethodssimilar tothosethat applytointegral tan
m
x sec
n
.
19.
_
cot
3
x dx
solution Usingthereductionformulafor cot
m
x, weget
_
cot
3
x dx =
1
2
cot
2
x
_
cotx dx =
1
2
cot
2
x +ln| cscx| +C.
_
sec
3
x dx
21.
_
cot
5
x csc
2
x dx
solution Makethesubstitutionu = cotx, du = csc
2
x dx; then
_
cot
5
x csc
2
x dx =
_
u
5
du =
1
6
u
6
=
1
6
cot
6
x +C
_
cot
4
x cscx dx
InExercises2346, evaluatetheintegral.
23.
_
cos
5
x sinx dx
solution Usethesubstitutionu = cosx, du = sinx dx. Then
_
cos
5
x sinx dx =
_
u
5
du =
1
6
u
6
+C =
1
6
cos
6
x +C.
_
cos
3
(2x) sin(2x) dx
25.
_
cos
4
(3x +2) dx
solution First usethesubstitutionu = 3x +2, du = 3dx andthenapplythereductionformulafor cos
n
x:
_
cos
4
(3x +2) dx =
1
3
cos
4
u du =
1
3
_
1
4
cos
3
u sinu +
3
4
_
cos
2
u du
_
=
1
12
cos
3
u sinu +
1
4
_
u
2
+
sin2u
4
_
+C
=
1
12
cos
3
(3x +2) sin(3x +2) +
1
8
(3x +2) +
1
16
sin(6x +4) +C
_
cos
7
3x dx
27.
_
cos
3
() sin
4
() d
solution Usethesubstitutionu = , du = d, andtheidentitycos
2
u = 1sin
2
u towrite
_
cos
3
() sin
4
() d =
1
_
cos
3
u sin
4
u du =
1
_
_
1sin
2
u
_
sin
4
u cosu du.
Nowusethesubstitutionw = sinu, dw = cosu du:
_
cos
3
() sin
4
() d =
1
_
_
1w
2
_
w
4
dw =
1
_
_
w
4
w
6
_
dw =
1
5
w
5
1
7
w
7
+C
=
1
5
sin
5
()
1
7
sin
7
() +C.
_
cos
498
y sin
3
y dy
29.
_
sin
4
(3x) dx
solution Usethesubstitutionu = 3x, du = 3dx andthereductionformulafor sin
m
x:
_
sin
4
(3x) dx =
1
3
_
sin
4
u du =
1
12
sin
3
u cosu +
1
4
_
sin
2
u du
=
1
12
sin
3
u cosu +
1
4
_
1
2
sinu cosu +
1
2
_
du
_
=
1
12
sin
3
u cosu
1
8
sinu cosu +
1
8
u +C
=
1
12
sin
3
(3x) cos(3x)
1
8
sin(3x) cos(3x) +
3
8
x +C.
_
sin
2
x cos
6
x dx
31.
_
csc
2
(32x) dx
solution First makethesubstitutionu = 32x, du = 2dx, sothat
_
csc
2
(32x) dx =
1
2
_
(csc
2
u) du =
1
2
cotu +C =
1
2
cot(32x) +C
_
csc
3
x dx
33.
_
tanx sec
2
x dx
solution Usethesubstitutionu = tanx, du = sec
2
x dx. Then
_
tanx sec
2
x dx =
_
u du =
1
2
u
2
+C =
1
2
tan
2
x +C.
_
tan
3
sec
3
d
35.
_
tan
5
x sec
4
x dx
solution Usetheidentitytan
2
x = sec
2
x 1towrite
_
tan
5
x sec
4
x dx =
_
_
sec
2
x 1
_
2
sec
3
x(secx tanx dx).
Nowusethesubstitutionu = secx, du = secx tanx dx:
_
tan
5
x sec
4
x dx =
_
_
u
2
1
_
2
u
3
du =
_
_
u
7
2u
5
+u
3
_
du
=
1
8
u
8
1
3
u
6
+
1
4
u
4
+C =
1
8
sec
8
x
1
3
sec
6
x +
1
4
sec
4
x +C.
_
tan
4
x secx dx
37.
_
tan
6
x sec
4
x dx
solution Usetheidentitysec
2
x = tan
2
x +1towrite
_
tan
6
x sec
4
x dx =
_
tan
6
x
_
tan
2
x +1
_
sec
2
x dx.
Nowusethesubstitutionu = tanx, du = sec
2
x dx:
_
tan
6
x sec
4
x dx =
_
u
6
_
u
2
+1
_
du =
_
_
u
8
+u
6
_
du =
1
9
u
9
+
1
7
u
7
+C =
1
9
tan
9
x +
1
7
tan
7
x +C.
_
tan
2
x sec
3
x dx
39.
_
cot
5
x csc
5
x dx
solution First usetheidentitycot
2
x = csc
2
x 1torewritetheintegral:
_
cot
5
x csc
5
x dx =
_
(csc
2
x 1)
2
csc
4
x(cotx cscx) dx =
_
(csc
8
x 2csc
6
x +csc
4
x)(cotx cscx) dx
Nowusethesubstitutionu = cscx anddu = cotx cscx dx toget
_
cot
5
x csc
5
x dx =
_
u
8
2u
6
+u
4
du =
1
9
u
9
+
2
7
u
7
1
5
u
5
+C
=
1
9
csc
9
x +
2
7
csc
7
x
1
5
csc
5
x +C
_
cot
2
x csc
4
x dx
41.
_
sin2x cos2x dx
solution Usethesubstitutionu = sin2x, du = 2cos2x dx:
_
sin2x cos2x dx =
1
2
_
sin2x(2cos2x dx) =
1
2
_
u du =
1
4
u
2
+C =
1
4
sin
2
2x +C.
_
cos4x cos6x dx
43.
_
t cos
3
(t
2
) dt
solution Usethesubstitutionu = t
2
, du = 2t dt , followedbythereductionformulafor cos
m
x:
_
t cos
3
(t
2
) dt =
1
2
_
cos
3
u du =
1
6
cos
2
u sinu +
1
3
_
cosu du
=
1
6
cos
2
u sinu +
1
3
sinu +C =
1
6
cos
2
(t
2
) sin(t
2
) +
1
3
sin(t
2
) +C.
_
tan
3
(lnt )
t
dt
45.
_
cos
2
(sint ) cost dt
solution Usethesubstitutionu = sint , du = cost dt , followedbythereductionformulafor cos
m
x:
_
cos
2
(sint ) cost dt =
_
cos
2
u du =
1
2
cosu sinu +
1
2
_
du
=
1
2
cosu sinu +
1
2
u +C =
1
2
cos(sint ) sin(sint ) +
1
2
sint +C.
_
e
x
tan
2
(e
x
) dx
InExercises4760, evaluatethedeniteintegral.
47.
_
2
0
sin
2
x dx
solution Usetheformulafor
_
sin
2
x dx:
_
2
0
sin
2
x dx =
_
x
2
sin2x
4
_
2
0
=
_
2
2

sin4
4
_
_
0
2
sin0
4
_
= .
_
/2
0
cos
3
x dx
49.
_
/2
0
sin
5
x dx
solution Usetheidentitysin
2
x = 1cos
2
x followedbythesubstitutionu = cosx, du = sinx dx toget
_
/2
0
sin
5
x dx =
_
/2
0
(1cos
2
x)
2
sinx dx =
_
/2
0
(12cos
2
x +cos
4
x) sinx dx
=
_
0
1
(12u
2
+u
4
) du =
_
u
2
3
u
3
+
1
5
u
5
_
0
1
= 1
2
3
+
1
5
=
8
15
_
/2
0
sin
2
x cos
3
x dx
51.
_
/4
0
dx
cosx
solution Usethedenitionof secx tosimplifytheintegral:
_
/4
0
dx
cosx
=
_
/4
0
secx dx = ln| secx +tanx|
/4
0
= ln
2+1
ln|1+0| = ln
_
2+1
_
.
_
/2
/4
dx
sinx
53.
_
/3
0
tanx dx
_
tanx dx:
_
/3
0
tanx dx = ln| secx|
/3
0
= ln2ln1= ln2.
_
/4
0
tan
5
x dx
55.
_
/4
/4
sec
4
x dx
solution First usethereductionformulafor sec
m
x toevaluatetheindeniteintegral:
_
sec
4
x dx =
1
3
tanx sec
2
x +
2
3
_
sec
2
x dx =
1
3
tanx sec
2
x +
2
3
tanx +C.
Nowcomputethedeniteintegral:
_
/4
/4
sec
4
x dx =
_
1
3
tanx sec
2
x +
2
3
tanx
_
/4
/4
=
_
1
3
(1)
_
2
_
2
+
2
3
(1)
_
_
1
3
(1)
_
2
_
2
+
2
3
(1)
_
=
4
3
4
3
_
=
8
3
.
_
3/4
/4
cot
4
x csc
2
x dx
57.
_

0
sin3x cos4x dx
_
sinmx cosnx dx:
_

0
sin3x cos4x dx =
_
cos(34)x
2(34)
cos(3+4)x
2(3+4)
_
0
=
_
cos(x)
2

cos7x
14
_
0
=
_
1
2
cosx
1
14
cos7x
_
0
=
_
1
2
(1)
1
14
(1)
_
_
1
2
(1)
1
14
(1)
_
=
6
7
.
_

0
sinx sin3x dx
59.
_
/6
0
sin2x cos4x dx
solution Usingtheformulafor
_
sinmx cosnx dx, wehave
_
/6
0
sin2x cos4x dx =
_
1
4
cos(2x)
1
2 6
cos(6x)
_
/6
0
=
_
1
4
cos2x
1
12
cos6x
_
/6
0
=
_
1
4
1
2
1
12
(1)
_
_
1
4
1
12
_
=
1
24
Hereweveusedthefact that cosx isanevenfunction: cos(x) = cosx.
_
/4
0
sin7x cos2x dx
61. Usetheidentitiesfor sin2x andcos2x onpage407toverifythat thefollowingformulasareequivalent.
_
sin
4
x dx =
1
32
(12x 8sin2x +sin4x) +C
_
sin
4
x dx =
1
4
sin
3
x cosx
3
8
sinx cosx +
3
8
x +C
solution First, observe
sin4x = 2sin2x cos2x = 2sin2x(12sin
2
x)
= 2sin2x 4sin2x sin
2
x = 2sin2x 8sin
3
x cosx.
Then
1
32
(12x 8sin2x +sin4x) +C =
3
8
x
3
16
sin2x
1
4
sin
3
x cosx +C
=
3
8
x
3
8
sinx cosx
1
4
sin
3
x cosx +C.
Evaluate
_
sin
2
x cos
3
x dx usingthemethoddescribedinthetext andverifythat your result isequivalent tothe
followingresult producedbyacomputer algebrasystem.
_
sin
2
x cos
3
x dx =
1
30
(7+3cos2x) sin
3
x +C
63. Findthevolumeof thesolidobtainedbyrevolvingy = sinx for 0 x about thex-axis.
solution Usingthediskmethod, thevolumeisgivenby
V =
_

0
(sinx)
2
dx =
_

0
sin
2
x dx =
_
x
2
sin2x
4
_
0
=
__
2
0
_
(0)
_
=

2
2
.
UseIntegrationbyPartstoproveEqs. (1) and(2).
InExercises6568, usethefollowingalternativemethodfor evaluatingtheintegral J =
_
sin
m
x cos
n
x dx whenmand
n arebotheven. Usetheidentities
sin
2
x =
1
2
(1cos2x), cos
2
x =
1
2
(1+cos2x)
to write J =
1
4
_
(1cos2x)
m/2
(1+cos2x)
n/2
dx, and expand the right-hand side as a sumof integrals involving
smaller powersof sineandcosineinthevariable2x.
65.
_
sin
2
x cos
2
x dx
solution Usingtheidentitiessin
2
x =
1
2
(1cos2x) andcos
2
x =
1
2
(1+cos2x), wehave
J =
_
sin
2
x cos
2
x dx =
1
4
_
(1cos2x)(1+cos2x) dx
=
1
4
_
_
1cos
2
2x
_
dx =
1
4
_
dx
1
4
_
cos
2
2x dx.
Nowusethesubstitutionu = 2x, du = 2dx, andtheformulafor
_
cos
2
u du:
J =
1
4
x
1
8
_
cos
2
u du =
1
4
x
1
8
_
u
2
+
1
2
sinu cosu
_
+C
=
1
4
x
1
16
(2x)
1
16
sin2x cos2x +C =
1
8
x
1
16
sin2x cos2x +C.
_
cos
4
x dx
67.
_
sin
4
x cos
2
x dx
solution Usingtheidentitiessin
2
x =
1
2
(1cos2x) andcos
2
x =
1
2
(1+cos2x), wehave
J =
_
sin
4
x cos
2
x dx =
1
8
_
(1cos2x)
2
(1+cos2x) dx
=
1
8
_
_
12cos2x +cos
2
2x
_
(1+cos2x) dx
=
1
8
_
_
1cos2x cos
2
2x +cos
3
2x
_
dx.
Nowusethesubstitutionu = 2x, du = 2dx, together withthereductionformulafor cos
m
x:
J =
1
8
x
1
16
_
cosu du
1
16
_
cos
2
u du +
1
16
_
cos
3
u du
=
1
8
x
1
16
sinu
1
16
_
u
2
+
1
2
sinu cosu
_
+
1
16
_
1
3
cos
2
u sinu +
2
3
_
cosu du
_
=
1
8
x
1
16
sin2x
1
32
(2x)
1
32
sin2x cos2x +
1
48
cos
2
2x sin2x +
1
24
sin2x +C
=
1
16
x
1
48
sin2x
1
32
sin2x cos2x +
1
48
cos
2
2x sin2x +C.
_
sin
6
x dx
69. Provethereductionformula
_
tan
k
x dx =
tan
k1
x
k 1

_
tan
k2
x dx
Hint: tan
k
x = (sec
2
x 1) tan
k2
x.
solution Usetheidentitytan
2
x = sec
2
x 1towrite
_
tan
k
x dx =
_
tan
k2
x
_
sec
2
x 1
_
dx =
_
tan
k2
x sec
2
x dx
_
tan
k2
x dx.
Nowusethesubstitutionu = tanx, du = sec
2
x dx:
_
tan
k
x dx =
_
u
k2
du
_
tan
k2
x dx =
1
k 1
u
k1
_
tan
k2
x dx =
tan
k1
x
k 1

_
tan
k2
x dx.
Usethesubstitutionu = cscx cotx toevaluate
_
cscx dx (seeExample5).
71. Let I
m
=
_
/2
0
sin
m
x dx.
(a) Showthat I
0
=

2
andI
1
= 1.
(b) Provethat, for m 2,
I
m
=
m1
m
I
m2
(c) Use(a) and(b) tocomputeI
m
for m = 2, 3, 4, 5.
solution
(a) Wehave
I
0
=
_
/2
0
sin
0
x dx =
_
/2
0
1dx =

2
I
1
=
_
/2
0
sinx dx = cosx
/2
0
= 1
(b) Usingthereductionformulafor sin
m
x, weget for m 2
I
m
=
_
/2
0
sin
m
x dx =
1
m
sin
m1
x cosx
/2
0
+
m1
m
_
/2
0
sin
m2
x dx
=
1
m
sin
m1
_
2
_
cos
_
2
_
+
1
m
sin
m1
(0) cos(0) +
m1
m
I
m2
=
1
m
(1 0+0 1) +
m1
m
I
m2
=
m1
m
I
m2
(c)
I
2
=
1
2
I
0
=
1
2

2
=

4
I
3
=
2
3
I
1
=
2
3
I
4
=
3
4
I
2
=
3
4

4
=
3
16
I
5
=
4
5
I
3
=
8
15
Evaluate
_

0
sin
2
mx dx for m anarbitraryinteger.
73. Evaluate
_
sinx ln(sinx) dx. Hint: UseIntegrationbyPartsasarst step.
solution Start by usingintegrationby partswithu = ln(sinx) andv
= sinx, sothat u
= cotx andv = cosx.

Then
I =
_
sinx ln(sinx) dx = cosx ln(sinx) +
_
cotx cosx dx = cosx ln(sinx) +
_
cos
2
x
sinx
dx
= cosx ln(sinx) +
_
1sin
2
x
sinx
dx = cosx ln(sinx)
_
sinx dx +
_
cscx dx
= cosx ln(sinx) +cosx +
_
cscx dx
Usingthetable,
_
cscx dx = ln| cscx cotx| +C, sonally
I = cosx ln(sinx) +cosx +ln| cscx cotx| +C
Total Energy A100-WlightbulbhasresistanceR = 144(ohms) whenattachedtohouseholdcurrent, where
thevoltagevariesasV = V
0
sin(2f t ) (V
0
= 110V, f = 60Hz). Theenergy(injoules) expendedbythebulbover
aperiodof T secondsis
U =
_
T
0
P(t ) dt
whereP = V
2
/R (J /s) isthepower. ComputeU if thebulbremainsonfor 5hours.
75. Let m, n beintegers withm = n. UseEqs. (23)(25) to provetheso-called orthogonalityrelationsthat play a
basicroleinthetheoryof Fourier Series(Figure2):
_

0
sinmx sinnx dx = 0
_

0
cosmx cosnx dx = 0
_
2
0
sinmx cosnx dx = 0
y= sin 2x sin 4x
y
x
p
y= sin 3x cos 4x
y
p
x
2p
FIGURE 2 Theintegralsarezerobytheorthogonalityrelations.
solution If m, n areintegers, thenm n andm + n areintegers, andthereforesin(m n) = sin(m + n) = 0,
sincesink = 0if k isaninteger. Thuswehave
_

0
sinmx sinnx dx =
_
sin(mn)x
2(mn)
sin(m+n)x
2(m+n)
_
0
=
_
sin(mn)
2(mn)
sin(m+n)
2(m+n)
_
0= 0;
_

0
cosmx cosnx dx =
_
sin(mn)x
2(mn)
+
sin(m+n)x
2(m+n)
_
0
=
_
sin(mn)
2(mn)
+
sin(m+n)
2(m+n)
_
0= 0.
If k isaninteger, thencos2k = 1. Usingthisfact, wehave
_
2
0
sinmx cosnx dx =
_
cos(mn)x
2(mn)
cos(m+n)x
2(m+n)
_
2
0
=
_
cos(mn)2
2(mn)
cos(m+n)2
2(m+n)
_
1
2(mn)
1
2(m+n)
_
=
_
1
2(mn)
1
2(m+n)
_
1
2(mn)
1
2(m+n)
_
= 0.
Usethetrigonometricidentity
sinmx cosnx =
1
2
_
sin(mn)x +sin(m+n)x
_
toproveEq. (24) inthetableof integralsonpage423.
77. UseIntegrationbyPartstoprovethat (for m = 1)
_
sec
m
x dx =
tanx sec
m2
x
m1
+
m2
m1
_
sec
m2
x dx
solution UsingIntegrationbyPartswithu = sec
m2
x andv
= sec
2
x, wehavev = tanx and
u
= (m2) sec
m3
x(secx tanx) = (m2) tanx sec
m2
x.
Then,
_
sec
m
x dx = tanx sec
m2
x (m2)
_
tan
2
x sec
m2
x dx
= tanx sec
m2
x (m2)
_
_
sec
2
x 1
_
sec
m2
x dx
= tanx sec
m2
x (m2)
_
sec
m
x dx +(m2)
_
sec
m2
x dx.
_
sec
m
x dx, weget
(m1)
_
sec
m
x dx = tanx sec
m2
x +(m2)
_
sec
m2
x dx
_
sec
m
x dx =
tanx sec
m2
x
m1
+
m2
m1
_
sec
m2
x dx.
Set I
m
=
_
/2
0
sin
m
x dx. UseExercise71toprovethat
I
2m
=
2m1
2m
2m3
2m2

1
2

2
I
2m+1
=
2m
2m+1
2m2
2m1

2
3
Concludethat
2
=
2 2
1 3
4 4
3 5

2m 2m
(2m1)(2m+1)
I
2m
I
2m+1
79. Thisisacontinuationof Exercise78.
(a) Provethat I
2m+1
I
2m
I
2m1
. Hint: sin
2m+1
x sin
2m
x sin
2m1
x for 0 x

2
.
(b) Showthat
I
2m1
I
2m+1
= 1+
1
2m
.
(c) Showthat 1
I
2m
I
2m+1
1+
1
2m
.
(d) Provethat lim
m
I
2m
I
2m+1
= 1.
(e) Finally, deducetheinniteproduct for

2
discoveredbyEnglishmathematicianJ ohnWallis(16161703):
2
= lim
m
2
1
2
3
4
3
4
5

2m 2m
(2m1)(2m+1)
solution
(a) For 0 x

2
, 0 sinx 1. Multiplyingthislast inequalitybysinx, weobtain
0 sin
2
x sinx.
Continuingtomultiplythisinequalitybysinx, weobtain, moregenerally,
sin
2m+1
x sin
2m
x sin
2m1
x.
Integratingthesefunctionsover [0,

2
], weget
_
/2
0
sin
2m+1
x dx
_
/2
0
sin
2m
x dx
_
/2
0
sin
2m1
x dx,
whichisthesameas
I
2m+1
I
2m
I
2m1
.
(b) UsingtherelationI
m
= ((m1)/m)I
m2
, wehave
I
2m1
I
2m+1
=
I
2m1
_
2m
2m+1
_
I
2m1
=
2m+1
2m
=
2m
2m
+
1
2m
= 1+
1
2m
.
S E C T I ON 8.3 Trigonometric Substitution 467
(c) First start withtheinequalityof part (a):
I
2m+1
I
2m
I
2m1
.
DividethroughbyI
2m+1
:
1
I
2m
I
2m+1

I
2m1
I
2m+1
.
Usetheresult frompart (b):
1
I
2m
I
2m+1
1+
1
2m
.
(d) Takingthelimit of thisinequality, andapplyingtheSqueezeTheorem, wehave
lim
m
1 lim
m
I
2m
I
2m+1
lim
m
_
1+
1
2m
_
.
Because
lim
m
1= 1 and lim
m
_
1+
1
2m
_
= 1,
weobtain
1 lim
m
I
2m
I
2m+1
1.
Therefore
lim
m
I
2m
I
2m+1
= 1.
(e) Takethelimit of bothsidesof theequationobtainedinExercise78(d):
lim
m
2
= lim
m
2 2
1 3
4 4
3 5

2m 2m
(2m1)(2m+1)
I
2m
I
2m+1
2
=
_
lim
m
2 2
1 3
4 4
3 5

2m 2m
(2m1)(2m+1)
__
lim
m
I
2m
I
2m+1
_
.
Finally, usingtheresult from(d), wehave
2
= lim
m
2 2
1 3
4 4
3 5

2m 2m
(2m1)(2m+1)
.
8.3 Trigonometric Substitution
1. Statethetrigonometricsubstitutionappropriatetothegivenintegral:
(a)
_
_
9x
2
dx (b)
_
x
2
(x
2
16)
3/2
dx
(c)
_
x
2
(x
2
+16)
3/2
dx (d)
_
(x
2
5)
2
dx
solution
(a) x = 3sin (b) x = 4sec (c) x = 4tan (d) x =
5sec
2. Istrigonometricsubstitutionneededtoevaluate
_
x
_
9x
2
dx?
solution No. Thereisafactor of x intheintegrandoutsidetheradical andthederivativeof 9x
2
is2x, sowemay
usethesubstitutionu = 9x
2
, du = 2x dx toevaluatethisintegral.
3. Expresssin2 intermsof x = sin.
solution First notethat if sin = x, thencos =
_
1sin
2
=
_
1x
2
. Thus,
sin2 = 2sin cos = 2x
_
1x
2
.
4. Drawatrianglethat wouldbeusedtogether withthesubstitutionx = 3sec.
solution
x
2
9
3
x
Exercises
InExercises14, evaluatetheintegral byfollowingthestepsgiven.
1. I =
_
dx
_
9x
2
(a) Showthat thesubstitutionx = 3sin transformsI into
_
d, andevaluateI intermsof .
(b) EvaluateI intermsof x.
solution
(a) Let x = 3sin. Thendx = 3cos d, and
_
9x
2
=
_
99sin
2
= 3
_
1sin
2
= 3
_
cos
2
= 3cos.
Thus,
I =
_
dx
_
9x
2
=
_
3cos d
3cos
=
_
d = +C.
(b) If x = 3sin, then = sin
1
(
x
3
). Thus,
I = +C = sin
1
_
x
3
_
+C.
I =
_
dx
x
2
_
x
2
2
(a) Showthat thesubstitutionx =
2sec transformstheintegral I into

1
2
_
cosd, andevaluateI intermsof
.
(b) Usearight triangletoshowthat withtheabovesubstitution, sin =
_
x
2
2/x.
(c) EvaluateI intermsof x.
3. I =
_
dx
_
4x
2
+9
(a) Showthat thesubstitutionx =
3
2
tan transformsI into
1
2
_
sec d.
(b) EvaluateI intermsof (refer tothetableof integralsonpage423inSection8.2if necessary).
(c) ExpressI intermsof x.
solution
(a) If x =
3
2
tan, thendx =
3
2
sec
2
d, and
_
4x
2
+9=
_
4
_
3
2
tan
_
2
+9=
_
9tan
2
+9= 3
_
sec
2
= 3sec
Thus,
I =
_
dx
_
4x
2
+9
=
_
3
2
sec
2
d
3sec
=
1
2
_
sec d
(b)
I =
1
2
_
sec d =
1
2
ln| sec +tan| +C
(c) Sincex =
3
2
tan, weconstruct aright trianglewithtan =
2x
3
:
4x
2
+9
2x
3
Fromthistriangle, weseethat sec =
1
3
_
4x
2
+9, andtherefore
I =
1
2
ln| sec +tan| +C =
1
2
ln
1
3
_
4x
2
+9+
2x
3
+C
=
1
2
ln
_
4x
2
+9+2x
3
+C =
1
2
ln|
_
4x
2
+9+2x|
1
2
ln3+C =
1
2
ln|
_
4x
2
+9+2x| +C
I =
_
dx
(x
2
+4)
2
(a) Showthat thesubstitutionx = 2tan transformstheintegral I into
1
8
_
cos
2
d.
(b) Usetheformula
_
cos
2
d =
1
2
+
1
2
sin cos toevaluateI intermsof .
(c) Showthat sin =
x
_
x
2
+4
andcos =
2
_
x
2
+4
.
(d) ExpressI intermsof x.
InExercises510, usetheindicatedsubstitutiontoevaluatetheintegral.
5.
_
_
165x
2
dx, x =
4
5
sin
solution Let x =
4
5
sin. Thendx =
4
5
cos d, and
I =
_
_
165x
2
dx =
_
_
165
_
4
5
sin
_
2
5
cos d =
4
5
_ _
1616sin
2
cos d
=
4
5
4
_
cos cos d =
16
5
_
cos
2
d
=
16
5
_
1
2
+
1
2
sin cos
_
+C =
8
5
( +sin cos) +C
Sincex =
4
5
sin, weconstruct aright trianglewithsin =
x
5
4
:
16 5x
2
x5
4
Fromthistriangleweseethat cos =
1
4
_
165x
2
, sowehave
I =
8
5
( +sin cos) +C
=
8
5
_
sin
1
_
x
5
4
_
+
x
5
4

1
4
_
165x
2
_
+C
=
8
5
sin
1
_
x
5
4
_
+
1
2
x
_
165x
2
+C
_
1/2
0
x
2
_
1x
2
dx, x = sin
7.
_
dx
x
_
x
2
9
, x = 3sec
solution Let x = 3sec. Thendx = 3sec tan d, and
_
x
2
9=
_
9sec
2
9= 3
_
sec
2
1= 3
_
tan
2
= 3tan.
Thus,
_
dx
x
_
x
2
9
=
_
(3sec tan d)
(3sec)(3tan)
=
1
3
_
d =
1
3
+C.
Sincex = 3sec, = sec
1
(
x
3
), and
_
dx
x
_
x
2
9
=
1
3
sec
1
_
x
3
_
+C.
_
1
1/2
dx
x
2
_
x
2
+4
, x = 2tan
9.
_
dx
(x
2
4)
3/2
, x = 2sec
solution Let x = 2sec. Thendx = 2sec tan d, and
x
2
4= 4sec
2
4= 4(sec
2
1) = 4tan
2
.
Thisgives
I =
_
dx
(x
2
4)
3/2
=
_
2sec tan d
(4tan
2
)
3/2
=
_
2sec tan d
8tan
3
=
1
4
_
sec d
tan
2
=
1
4
_
cos
sin
2
d.
Nowusesubstitutionwithu = sin anddu = cos d. Then
I =
1
4
_
u
2
du =
1
4
u
1
+C =
1
4sin
+C.
Sincex = 2sec, weconstruct aright trianglewithsec =
x
2
:
q
2
x
x
2
4
Fromthistriangleweseethat sin =
_
x
2
4/x, sotherefore
I =
1
4(
_
x
2
4/x)
+C =
x
4
_
x
2
4
+C.
_
1
0
dx
(4+9x
2
)
2
, x =
2
3
tan
11. Evaluate
_
x dx
_
x
2
4
intwoways: usingthedirect substitutionu = x
2
4andbytrigonometricsubstitution.
solution Let u = x
2
4. Thendu = 2x dx, and
I
1
=
_
x dx
_
x
2
4
=
1
2
_
du
u
=
1
2
_
2u
1/2
_
+C =
u +C =
_
x
2
4+C.
Tousetrigonometricsubstitution, let x = 2sec. Thendx = 2sec tan d, x
2
4= 4sec
2
4= 4tan
2
, and
I
1
=
_
x dx
_
x
2
4
=
_
2sec(2sec tan d)
2tan
= 2
_
sec
2
d = 2tan +C.
x
2
:
q
2
x
x
2
4
Fromthistriangleweseethat
I
1
= 2
__
x
2
4
2
_
+C =
_
x
2
4+C.
Isthesubstitutionu = x
2
4effectiveforevaluatingtheintegral
_
x
2
dx
_
x
2
4
?If not,evaluateusingtrigonometric
substitution.
13. Evaluateusingthesubstitutionu = 1x
2
or trigonometricsubstitution.
(a)
_
x
_
1x
2
dx (b)
_
x
2
_
1x
2
dx
(c)
_
x
3
_
1x
2
dx (d)
_
x
4
_
1x
2
dx
solution
(a) Let u = 1x
2
. Thendu = 2x dx, andwehave
_
x
_
1x
2
dx =
1
2
_
2x dx
_
1x
2
=
1
2
_
du
u
1/2
.
(b) Let x = sin. Thendx = cos d, 1x
2
= cos
2
, andso
_
x
2
_
1x
2
dx =
_
sin
2
(cos) cos d =
_
sin
2
cos
2
d.
(c) Usethesubstitutionu = 1x
2
. Thendu = 2x dx, x
2
= 1u, andso
_
x
3
_
1x
2
dx =
1
2
_
x
2
_
1x
2
(2x dx) =
1
2
_
(1u)u
1/2
du.
(d) Let x = sin. Thendx = cos d, 1x
2
= cos
2
, andso
_
x
4
_
1x
2
dx =
_
sin
4
cos
cos d =
_
sin
4
d.
Evaluate:
(a)
_
dt
(t
2
+1)
3/2
(b)
_
t dt
(t
2
+1)
3/2
InExercises1532, evaluateusingtrigonometricsubstitution. Refer tothetableof trigonometricintegralsasnecessary.
15.
_
x
2
dx
_
9x
2
solution Let x = 3sin. Thendx = 3cos d,
9x
2
= 99sin
2
= 9(1sin
2
) = 9cos
2
,
and
I =
_
x
2
dx
_
9x
2
=
_
9sin
2
(3cos d)
3cos
= 9
_
sin
2
d = 9
_
1
2

1
2
sin cos
_
+C.
Sincex = 3sin, weconstruct aright trianglewithsin =
x
3
:
q
x
3
9 x
2
Fromthisweseethat cos =
_
9x
2
/3, andso
I =
9
2
sin
1
_
x
3
_
9
2
_
x
3
_
__
9x
2
3
_
+C =
9
2
sin
1
_
x
3
_
1
2
x
_
9x
2
+C.
_
dt
(16t
2
)
3/2
17.
_
dx
x
_
x
2
+16
solution Usethesubstitutionx = 4tan, sothat dx = 4sec
2
d. Then
x
_
x
2
+16= 4tan
_
(4tan)
2
+16= 4tan
_
16(tan
2
+1) = 16tan sec
sothat
I =
_
dx
x
_
x
2
+16
=
_
4sec
2
16tan sec
d =
1
4
_
sec
tan
d =
1
4
_
csc d =
1
4
ln| cscx +cotx| +C
Sincex = 4tan, weconstruct aright trianglewithtan =
x
4
:
16 +x
2
4
x
Fromthis, weseethat cscx =
x
2
+16
x
andcotx =
4
x
, sothat
I =
1
4
ln| cscx +cotx| +C =
1
4
ln
_
x
2
+16
x
+
4
x
+C =
1
4
ln
4+
_
x
2
+16
x
+C
_
_
12+4t
2
dt
19.
_
dx
_
x
2
9
solution Let x = 3sec. Thendx = 3sec tan d,
x
2
9= 9sec
2
9= 9(sec
2
1) = 9tan
2
,
and
I =
_
dx
_
x
2
9
=
_
3sec tan d
3tan
=
_
sec d = ln| sec +tan| +C.
x
3
:
q
3
x
x
2
9
Fromthisweseethat tan =
_
x
2
9/3, andso
I = ln
x
3
+
_
x
2
9
3
+C
1
= ln
x +
_
x
2
9
+ln
_
1
3
_
+C
1
= ln
x +
_
x
2
9
+C,
whereC = ln
_
1
3
_
+C
1
.
_
dt
t
2
_
t
2
25
21.
_
dy
y
2
_
5y
2
solution Let y =
5sin. Thendy =
5cos d,
5y
2
= 55sin
2
= 5(1sin
2
) = 5cos
2
,
and
I =
_
dy
y
2
_
5y
2
=
_

5cos d
(5sin
2
)(
5cos)
=
1
5
_
d
sin
2
=
1
5
_
csc
2
d =
1
5
(cot) +C.
Sincey =
5sin, weconstruct aright trianglewithsin =

y
5
:
q
y
5 y
2
5
Fromthisweseethat cot =
_
5y
2
/y, whichgivesus
I =
1
5
_
_
5y
2
y
_
+C =
_
5y
2
5y
+C.
_
x
3
_
9x
2
dx
23.
_
dx
_
25x
2
+2
solution Let x =
2
5
tan. Thendx =
2
5
sec
2
d, 25x
2
+2= 2tan
2
+2= 2sec
2
, and
I =
_
dx
_
25x
2
+2
=
_

2
5
sec
2
d
2sec
=
1
5
_
sec d =
1
5
ln| sec +tan| +C.
Sincex =
2
5
tan, weconstruct aright trianglewithtan =
5x
2
:
25x
2
+ 2
5x
2
Fromthisweseethat sec =
1
2
_
25x
2
+2, sothat
I =
1
5
ln| sec +tan| +C =
1
5
ln
_
25x
2
+2
2
+
5x
+C
=
1
5
ln
5x +
_
25x
2
+2
+C =
1
5
ln
5x +
_
25x
2
+2

1
5
ln
2+C
=
1
5
ln
5x +
_
25x
2
+2
+C
_
dt
(9t
2
+4)
2
25.
_
dz
z
3
_
z
2
4
solution Let z = 2sec. Thendz = 2sec tan d,
z
2
4= 4sec
2
4= 4(sec
2
1) = 4tan
2
,
and
I =
_
dz
z
3
_
z
2
4
=
_
2sec tan d
(8sec
3
)(2tan)
=
1
8
_
d
sec
2
=
1
8
_
cos
2
d
=
1
8
_
1
2
+
1
2
sin cos
_
+C =
1
16
+
1
16
sin cos +C.
Asexplainedinthetext, thiscomputationisvalidif wechoose in[0, /2) if z 2andin[, 3/2) if z 2. If z 2,
weconstruct aright trianglewithsec =
z
2
:
q
2
z
z
2
4
Fromthisweseethat sin =
_
z
2
4/z andcos = 2/z. Then
I =
1
16
sec
1
_
z
2
_
+
1
16
__
z
2
4
z
_
_
2
z
_
+C =
1
16
sec
1
_
z
2
_
+
_
z
2
4
8z
2
+C.
However, if z 2, thensec
1
_
z
2
_
liesin(
2
, ] accordingtothedenitionof sec
1
x usedinthetext. But since is
theanglein[,
3
2
) satisfyingsec = z/2, wendthat = 2 sec
1
_
z
2
_
. Similarly,
sin =
_
z
2
4
z
and cos =
2
z
Sofor z 2,
I =
1
16
sec
1
_
z
2
_
+
_
z
2
4
8z
2
+C
Notethat although = 2 sec
1
_
z
2
_
, the2 isnot neededintheexpressionfor I becauseit maybeabsorbedinthe
constant C.
_
dy
_
y
2
9
27.
_
x
2
dx
(6x
2
49)
1/2
solution Let x =
7
6
sec; thendx =
7
6
sec tan d, and
6x
2
49= 6
_
7
6
sec
_
2
49= 49(sec
2
1) = 49tan
2

sothat
I =
_
x
2
dx
(6x
2
49)
1/2
=
_
49
6
sec
2
(
7
6
sec tan)
7tan
d
=
49
6
6
_
sec
3
d =
49
6
6
_
1
2
tan sec +
1
2
_
sec d
_
=
49
12
6
(tan sec +ln| sec +tan|) +C
Sincex =
7
6
sec, weconstruct aright trianglewithsec =
x
6
7
:
6x
2
49
x6
7
1
7
_
6x
2
49, sothat
I =
49
12
6
_
x
6
_
6x
2
49
49
+ln
6+
_
6x
2
49
7
_
+C
=
49
12
6
_
x
6
_
6x
2
49
49
+ln
6+
_
6x
2
49
ln7
_
+C
=
1
12
6
_
x
6
_
6x
2
49+49ln
6+
_
6x
2
49
_
+C
_
dx
(x
2
4)
2
29.
_
dt
(t
2
+9)
2
solution Let t = 3tan. Thendt = 3sec
2
d,
t
2
+9= 9tan
2
+9= 9(tan
2
+1) = 9sec
2
,
and
I =
_
dt
(t
2
+9)
2
=
_
3sec
2
d
81sec
4
=
1
27
_
cos
2
d =
1
27
_
1
2
+
1
2
sin cos
_
+C.
Sincet = 3tan, weconstruct aright trianglewithtan =
t
3
:
t
2
+ 9
t
3
Fromthisweseethat sin = t /
_
t
2
+9andcos = 3/
_
t
2
+9. Thus
I =
1
54
tan
1
_
t
3
_
+
1
54
_
t
_
t
2
+9
__
3
_
t
2
+9
_
+C =
1
54
tan
1
_
t
3
_
+
t
18(t
2
+9)
+C.
_
dx
(x
2
+1)
3
31.
_
x
2
dx
(x
2
1)
3/2
solution Let x = sec. Thendx = sec tan d, andx
2
1= sec
2
1= tan
2
. Thus
I =
_
x
2
(x
2
1)
3/2
dx =
_
sec
2
(tan
2
)
3/2
sec tan d
=
_
sec
2
sec tan
tan
3
d =
_
sec
3
tan
2
d
=
_
sec
2
tan
2
sec d =
_
csc
2
sec d =
_
(1+cot
2
) sec d
=
_
sec +cot csc d = ln| sec +tan| csc +C
Sincex = sec, weconstruct thefollowingright triangle:
x
2
1
x
1
_
x
2
1andthat csc =
x
x
2
1
, sothat
I = ln
x +
_
x
2
1

x
_
x
2
1
+C
_
x
2
dx
(x
2
+1)
3/2
33. Provefor a > 0:
_
dx
x
2
+a
=
1
a
tan
1
x
a
+C
solution Let x =
a u. Then, x
2
= au
2
, dx =
a du, and
_
dx
x
2
+a
=
1
a
_
du
u
2
+1
=
1
a
tan
1
u +C =
1
a
tan
1
_
x
a
_
+C.
Provefor a > 0:
_
dx
(x
2
+a)
2
=
1
2a
_
x
x
2
+a
+
1
a
tan
1
x
a
_
+C
35. Let I =
_
dx
_
x
2
4x +8
.
(a) Completethesquaretoshowthat x
2
4x +8= (x 2)
2
+4.
(b) Usethesubstitutionu = x 2toshowthat I =
_
du
_
u
2
+2
2
. Evaluatetheu-integral.
(c) Showthat I = ln
_
(x 2)
2
+4+x 2
+C.
solution
(a) Completingthesquare, weget
x
2
4x +8= x
2
4x +4+4= (x 2)
2
+4.
(b) Let u = x 2. Thendu = dx, and
I =
_
dx
_
x
2
4x +8
=
_
dx
_
(x 2)
2
+4
=
_
du
_
u
2
+4
.
Nowlet u = 2tan. Thendu = 2sec
2
d,
u
2
+4= 4tan
2
+4= 4(tan
2
+1) = 4sec
2
,
and
I =
_
2sec
2
d
2sec
=
_
Sinceu = 2tan, weconstruct aright trianglewithtan =
u
2
:
q
u
2
+ 4
u
2
_
u
2
+4/2. Thus
I = ln
_
u
2
+4
2
+
u
2
+C
1
= ln
_
u
2
+4+u
+
_
ln
1
2
+C
1
_
= ln
_
u
2
+4+u
+C.
(c) Substitutebackfor x intheresult of part (b):
I = ln
_
(x 2)
2
+4+x 2
+C.
Evaluate
_
dx
_
12x x
2
. First completethesquaretowrite12x x
2
= 36(x 6)
2
.
InExercises3742, evaluatetheintegral bycompletingthesquareandusingtrigonometricsubstitution.
37.
_
dx
_
x
2
+4x +13
solution First completethesquare:
x
2
+4x +13= x
2
+4x +4+9= (x +2)
2
+9.
Let u = x +2. Thendu = dx, and
I =
_
dx
_
x
2
+4x +13
=
_
dx
_
(x +2)
2
+9
=
_
du
_
u
2
+9
.
Nowlet u = 3tan. Thendu = 3sec
2
d,
u
2
+9= 9tan
2
+9= 9(tan
2
+1) = 9sec
2
,
and
I =
_
3sec
2
d
3sec
=
_
Sinceu = 3tan, weconstruct thefollowingright triangle:
q
u
2
+ 9
u
3
_
u
2
+9/3. Thus
I = ln
_
u
2
+9
3
+
u
3
+C
1
= ln
_
u
2
+9+u
+
_
ln
1
3
+C
1
_
= ln
_
(x +2)
2
+9+x +2
+C = ln
_
x
2
+4x +13+x +2
+C.
_
dx
_
2+x x
2
39.
_
dx
_
x +6x
2
6x
2
+x =
_
6x
2
+x +
1
24
_
1
24
=
_
6x +
1
2
6
_
2
1
24
Let u =
6x +
1
2
6
sothat du =
6dx. Then
I =
_
1
_
x +6x
2
dx =
_
1
_
_
6x +
1
2
6
_
2
1
24
dx =
1
6
_
1
_
u
2
1
24
du
Nowlet u =
1
2
6
sec. Thendu =
1
2
6
sec tan, and
u
2
1
24
=
1
24
(sec
2
1) =
1
24
tan
2
sothat
I =
1
6
_
1
1
2
6
tan
1
2
6
sec tan d =
1
6
_
sec d =
1
6
ln| sec +tan| +C
Sinceu =
1
2
6
sec, weconstruct thefollowingright triangle:
24u
2
1
2u6
1
fromwhichweseethat tan =
_
24u
2
1andsec = 2u
6. Thus
I =
1
6
ln
2u
6+
_
24u
2
1
+C =
1
6
ln
6
_
6x +
1
2
6
_
+
_
24
_
6x
2
+x +
1
24
_
1
+C
=
1
6
ln
12x +1+
_
144x
2
+24x
+C
_
_
x
2
4x +7dx
41.
_
_
x
2
4x +3dx
x
2
4x +3= x
2
4x +41= (x 2)
2
1.
Let u = x 2. Thendu = dx, and
I =
_
_
x
2
4x +3dx =
_ _
(x 2)
2
1dx =
_
_
u
2
1du.
Nowlet u = sec. Thendu = sec tan d, u
2
1= sec
2
1= tan
2
, and
I =
_
_
tan
2
(sec tan d) =
_
tan
2
sec d =
_
_
sec
2
1
_
sec d
=
_
sec
3
d
_
sec d =
_
tan sec
2
+
1
2
_
sec d
_
_
sec d
=
1
2
tan sec
1
2
_
sec d =
1
2
tan sec
1
2
ln| sec +tan| +C.
Sinceu = sec, weconstruct thefollowingright triangle:
q
1
u
u
2
1
_
u
2
1. Thus
I =
1
2
u
_
u
2
1
1
2
ln
u +
_
u
2
1
+C =
1
2
(x 2)
_
(x 2)
2
1
1
2
ln
x 2+
_
(x 2)
2
1
+C
=
1
2
(x 2)
_
x
2
4x +3
1
2
ln
x 2+
_
x
2
4x +3
+C.
_
dx
(x
2
+6x +6)
2
InExercises4352, indicateagoodmethodfor evaluatingtheintegral (butdonotevaluate). Your choicesare: substitution
(specifyu anddu), IntegrationbyParts(specifyu andv
), atrigonometricmethod, or trigonometricsubstitution(specify).
If it appearsthat thesetechniquesarenot sufcient, statethis.
43.
_
x dx
_
126x x
2
solution Completethesquaresothethedenominator is
_
15(x +3)
2
andthenusetrigonometricsubstitutionwith
x +3= sin.
_
_
4x
2
1dx
45.
_
sin
3
x cos
3
x dx
solution Useoneof thefollowingtrigonometric methods: rewritesin
3
x = (1 cos
2
x) sinx andlet u = cosx, or
rewritecos
3
x = (1sin
2
x) cosx andlet u = sinx.
_
x sec
2
x dx
47.
_
dx
_
9x
2
solution Either usethesubstitutionx = 3u andthenrecognizetheformulafor theinversesine:
_
du
_
1u
2
= sin
1
u +C,
or usetrigonometricsubstitution, withx = 3sin.
_
_
1x
3
dx
49.
_
sin
3/2
x dx
solution Not solvablebyanymethodyet considered.
_
x
2
x +1dx
51.
_
dx
(x +1)(x +2)
3
solution Thetechniques wehavecovered thus far arenot sufcient to treat this integral. This integral requires a
techniqueknownaspartial fractions.
_
dx
(x +12)
4
InExercises5356, evaluateusingIntegrationbyPartsasarst step.
53.
_
sec
1
x dx
solution Let u = sec
1
x andv
= 1. Thenv = x, u
= 1/x
_
x
2
1, and
I =
_
sec
1
x dx = x sec
1
x
_
x
x
_
x
2
1
dx = x sec
1
x
_
dx
_
x
2
1
.
Toevaluatetheintegral ontheright, let x = sec. Thendx = sec tan d, x
2
1= sec
2
1= tan
2
, and
_
dx
_
x
2
1
=
_
sec tan d
tan
=
_
sec d = ln| sec +tan| +C = ln
x +
_
x
2
1
+C.
Thus, thenal answer is
I = x sec
1
x ln
x +
_
x
2
1
+C.
_
sin
1
x
x
2
dx
55.
_
ln(x
2
+1) dx
solution Start byusingintegrationbyparts, withu = ln(x
2
+1) andv
= 1; thenu
=
2x
x
2
+1
andv = x, sothat
I =
_
ln(x
2
+1) dx = x ln(x
2
+1) 2
_
x
2
x
2
+1
dx = x ln(x
2
+1) 2
_ _
1
1
x
2
+1
_
dx
= x ln(x
2
+1) 2x +2
_
1
x
2
+1
dx
Todeal withtheremainingintegral, usethesubstitutionx = tan, sothat dx = sec
2
d and
_
1
x
2
+1
dx =
_
sec
2
tan
2
+1
d =
_
sec
2
sec
2
d =
_
1d = = tan
1
x +C
sothat nally
I = x ln(x
2
+1) 2x +2tan
1
x +C
_
x
2
ln(x
2
+1) dx
57. Findtheaverageheight of apoint onthesemicircley =
_
1x
2
for 1 x 1.
solution Theaverageheight isgivenbytheformula
y
ave
=
1
1(1)
_
1
1
_
1x
2
dx =
1
2
_
1
1
_
1x
2
dx
Let x = sin. Thendx = cos d, 1x
2
= cos
2
, and
_
_
1x
2
dx =
_
(cos)(cos d) =
_
cos
2
d =
1
2
+
1
2
sin cos +C.
Sincex = sin, weconstruct thefollowingright triangle:
q
1
1 x
2
x
Fromthisweseethat cos =
_
1x
2
. Therefore,
y
ave
=
1
2
_
1
2
sin
1
x +
1
2
x
_
1x
2
_
1
1
=
1
2
__
1
2
+0
_
1
2
+0
__
=

4
.
Findthevolumeof thesolidobtainedbyrevolvingthegraphof y = x
_
1x
2
over [0, 1] about they-axis.
59. Findthevolumeof thesolidobtainedbyrevolvingtheregionbetweenthegraphof y
2
x
2
= 1andtheliney = 2
about theliney = 2.
solution First solvetheequationy
2
x
2
= 1for y:
y =
_
x
2
+1.
Theregioninquestionisboundedinpart bythetophalf of thishyperbola, whichistheequation
y =
_
x
2
+1.
Thelimitsof integrationareobtainedbyndingthepointsof intersectionof thisequationwithy = 2:
2=
_
x
2
+1 x =
3.
Theradiusof eachdiskisgivenby2
_
x
2
+1; thevolumeisthereforegivenby
V =
_

3
3
r
2
dx = 2
_

3
0
_
2
_
x
2
+1
_
2
dx = 2
_

3
0
_
44
_
x
2
+1+(x
2
+1)
_
dx
= 8
_

3
0
dx 8
_

3
0
_
x
2
+1dx +2
_

3
0
(x
2
+1) dx.
Toevaluatetheintegral
_
_
x
2
+1dx, let x = tan. Thendx = sec
2
d, x
2
+1= sec
2
, and
_
_
x
2
+1dx =
_
sec
3
d =
1
2
tan sec +
1
2
_
sec d
=
1
2
tan sec +
1
2
ln| sec +tan| +C =
1
2
x
_
x
2
+1+
1
2
ln
_
x
2
+1+x
+C.
Nowwecancomputethevolume:
V =
_
8x 8
_
1
2
x
_
x
2
+1+
1
2
ln
_
x
2
+1+x
_
+
2
3
x
3
+2x
_
3
0
=
_
10x +
2
3
x
3
4x
_
x
2
+14 ln
_
x
2
+1+x
3
0
=
_
10
3+2
38
34 ln
2+
_
(0) = 4
_
3ln
2+
_
.
Findthevolumeof revolutionfor theregioninExercise59, but revolvearoundy = 3. 61. Compute
_
dx
x
2
1
intwowaysandverifythattheanswersagree: rstviatrigonometricsubstitutionandthenusing
theidentity
1
x
2
1
=
1
2
_
1
x 1
1
x +1
_
solution Usingtrigonometric substitution, let x = sec. Thendx = sec tand, x
2
1 = sec
2
1 = tan
2
,
and
I =
_
dx
x
2
1
=
_
sec tan d
tan
2
=
_
sec
tan
d =
_
d
sin
=
_
csc d = ln| csc cot| +C.
Sincex = sec, weconstruct thefollowingright triangle:
q
1
x
x
2
1
Fromthisweseethat csc = x/
_
x
2
1andcot = 1/
_
x
2
1. Thisgivesus
I = ln
x
_
x
2
1
1
_
x
2
1
+C = ln
x 1
_
x
2
1
+C.
Usingthegivenidentity, weget
I =
_
dx
x
2
1
=
1
2
_ _
1
x 1
1
x +1
_
dx =
1
2
_
dx
x 1
1
2
_
dx
x +1
=
1
2
ln|x 1|
1
2
ln|x +1| +C.
Toconrmthat theseanswersagree, notethat
1
2
ln|x 1|
1
2
ln|x +1| =
1
2
ln
x 1
x +1
= ln
_
x 1
x +1
= ln
x 1
x +1
x 1
x 1
= ln
x 1
_
x
2
1
.
Youwanttodividean18-inchpizzaequallyamongthreefriendsusingvertical slicesatx asinFigure6. Find
anequationsatisedbyx andndtheapproximatevalueof x usingacomputer algebrasystem.
63. A chargedwirecreatesanelectriceldat apoint P locatedat adistanceD fromthewire(Figure7). Thecomponent
E
of theeldperpendicular tothewire(inN/C) is
E
=
_
x
2
x
1
kD
(x
2
+D
2
)
3/2
dx
where isthechargedensity(coulombsper meter), k = 8.9910
9
Nm
2
/C
2
(Coulombconstant), andx
1
, x
2
areasin
thegure. Supposethat = 610
4
C/m, andD = 3m. FindE
if (a) x
1
= 0andx
2
= 30m, and(b) x
1
= 15m
andx
2
= 15m.
x
1
x
2
P
D
y
x
FIGURE 7
solution Let x = Dtan. Thendx = Dsec
2
d,
x
2
+D
2
= D
2
tan
2
+D
2
= D
2
(tan
2
+1) = D
2
sec
2
,
and
E
=
_
x
2
x
1
kD
(x
2
+D
2
)
3/2
dx = kD
_
x
2
x
1
Dsec
2
d
(D
2
sec
2
)
3/2
=
kD
2
D
3
_
x
2
x
1
sec
2
d
sec
3
=
k
D
_
x
2
x
1
cos d =
k
D
sin
x
2
x
1
Sincex = Dtan, weconstruct aright trianglewithtan = x/D:
q
x
2
+ D
2
x
D
Fromthisweseethat sin = x/
_
x
2
+D
2
. Then
E
=
k
D
_
x
_
x
2
+D
2
_
x
2
x
1
(a) Plugginginthevaluesfor theconstantsk, , D, andevaluatingtheantiderivativefor x
1
= 0andx
2
= 30, weget
E
=
(8.9910
9
)(610
4
)
3
_
30
_
30
2
+3
2
0
_
1.78910
6
V
m
(b) If x
1
= 15mandx
2
= 15m, weget
E
=
(8.9910
9
)(610
4
)
3
_
15
_
15
2
+3
2
15
_
(15)
2
+3
2
_
3.52610
6
V
m
S E C T I ON 8.4 Integrals Involving Hyperbolic and Inverse Hyperbolic Functions 481
Let J
n
=
_
dx
(x
2
+1)
n
. UseIntegrationbyPartstoprove
J
n+1
=
_
1
1
2n
_
J
n
+
_
1
2n
_
x
(x
2
+1)
n
ThenusethisrecursionrelationtocalculateJ
2
andJ
3
.
65. Provetheformula
_
_
1x
2
dx =
1
2
sin
1
x +
1
2
x
_
1x
2
+C
usinggeometrybyinterpretingtheintegral astheareaof part of theunit circle.
solution Theintegral
_
a
0
_
1x
2
dx istheareaboundedbytheunitcircle, thex-axis, they-axis, andthelinex = a.
Thisareacanbedividedintotworegionsasfollows:
1
I
II
a 0
x
y
1
q
Region I is atrianglewith basea andheight
_
1a
2
. Region II is asector of theunit circlewith central angle =
2
cos
1
a = sin
1
a. Thus,
_
a
0
_
1x
2
dx =
1
2
a
_
1a
2
+
1
2
sin
1
a =
_
1
2
x
_
1x
2
+
1
2
sin
1
x
_
a
0
.
8.4 Integrals Involving Hyperbolic and Inverse Hyperbolic Functions
1. Whichhyperbolicsubstitutioncanbeusedtoevaluatethefollowingintegrals?
(a)
_
dx
_
x
2
+1
(b)
_
dx
_
x
2
+9
(c)
_
dx
_
9x
2
+1
solution Theappropriatehyperbolicsubstitutionsare
(a) x = sinht
(b) x = 3sinht
(c) 3x = sinht
2. Whichtwoof thehyperbolicintegrationformulasdiffer fromtheir trigonometriccounterpartsbyaminussign?
solution Theintegrationformulasfor sinhx andtanhx differ fromtheir trigonometriccounterpartsbyaminussign.
3. Whichantiderivativeof y = (1x
2
)
1
shouldweusetoevaluatetheintegral
_
5
3
(1x
2
)
1
dx?
solution Becausetheintegrationinterval liesoutside1< x < 1, theappropriateantiderivativeof y = (1x
2
)
1
is
1
2
ln
1+x
1x
.
Exercises
InExercises116, calculatetheintegral.
1.
_
cosh(3x) dx
solution
_
cosh(3x) dx =
1
3
sinh3x +C.
_
sinh(x +1) dx
3.
_
x sinh(x
2
+1) dx
solution
_
x sinh(x
2
+1) dx =
1
2
cosh(x
2
+1) +C.
_
sinh
2
x coshx dx
5.
_
sech
2
(12x) dx
solution
_
sech
2
(12x) dx =
1
2
tanh(12x) +C.
_
tanh(3x) sech(3x) dx
7.
_
tanhx sech
2
x dx
solution Let u = tanhx. Thendu = sech
2
x dx nd
_
tanhx sech
2
x dx =
_
u du =
1
2
u
2
+C =
tanh
2
x
2
+C.
_
coshx
3sinhx +4
dx
9.
_
tanhx dx
solution
_
tanhx dx = lncoshx +C.
_
x csch(x
2
) coth(x
2
) dx
11.
_
coshx
sinhx
dx
solution
_
coshx
sinhx
dx = ln| sinhx| +C.
_
coshx
sinh
2
x
dx
13.
_
sinh
2
(4x 9) dx
solution
_
sinh
2
(4x 9) dx =
1
2
_
(cosh(8x 18) 1) dx =
1
16
sinh(8x 18)
1
2
x +C.
_
sinh
3
x cosh
6
x dx
15.
_
sinh
2
x cosh
2
x dx
solution
_
sinh
2
x cosh
2
x dx =
1
4
_
sinh
2
2x dx =
1
8
_
(cosh4x 1) dx =
1
32
sinh4x
1
8
x +C.
_
tanh
3
x dx
InExercises1730, calculatetheintegral intermsof theinversehyperbolicfunctions.
17.
_
dx
_
x
2
1
solution
_
dx
_
x
2
1
= cosh
1
x +C.
_
dx
_
9x
2
4
19.
_
dx
_
16+25x
2
solution
_
dx
_
16+25x
2
=
1
5
sinh
1
_
5x
4
_
+C.
_
dx
_
1+3x
2
21.
_
_
x
2
1dx
solution Let x = cosht . Thendx = sinht dt and
_
_
x
2
1dx =
_
sinh
2
t dt =
1
2
_
(cosh2t 1) dt =
1
4
sinh2t
1
2
t +C
=
1
2
sinht cosht
1
2
t +C =
1
2
x
_
x
2
1
1
2
cosh
1
x +C.
_
x
2
dx
_
x
2
+1
23.
_
1/2
1/2
dx
1x
2
solution
_
1/2
1/2
dx
1x
2
= tanh
1
x
1/2
1/2
= tanh
1
_
1
2
_
tanh
1
_
1
2
_
= 2tanh
1
_
1
2
_
.
S E C T I ON 8.4 Integrals Involving Hyperbolic and Inverse Hyperbolic Functions 483
_
5
4
dx
1x
2
25.
_
1
0
dx
_
1+x
2
solution
_
1
0
dx
_
1+x
2
= sinh
1
1
0
= sinh
1
(1) sinh
1
(0) = sinh
1
1.
_
10
2
dx
4x
2
1
27.
_
1
3
dx
x
_
x
2
+16
solution
_
1
3
dx
x
_
x
2
+16
=
1
4
csch
1
_
x
4
_
1
3
=
1
4
_
csch
1
_
1
4
_
csch
1
_
3
4
__
.
_
0.8
0.2
dx
x
_
1x
2
29.
_
_
x
2
1dx
x
2
solution Let x = cosht . Thendx = sinht dt and
_
_
x
2
1dx
x
2
=
_
sinh
2
t
cosh
2
t
dt =
_
tanh
2
t dt =
_
(1sech
2
t ) dt
= t tanht +C = cosh
1
x
_
x
2
1
x
+C.
_
9
1
dx
x
_
x
4
+1
31. Verifytheformulas
sinh
1
x = ln|x +
_
x
2
+1|
cosh
1
x = ln|x +
_
x
2
1| (for x 1)
solution Let x = sinht . Then
cosht =
_
1+sinh
2
t =
_
1+x
2
.
Moreover, because
sinht +cosht =
e
t
e
t
2
+
e
t
+e
t
2
= e
t
,
it followsthat
sinh
1
x = t = ln(sinht +cosht ) = ln(x +
_
x
2
+1).
Now, Let x = cosht . Then
sinht =
_
cosh
2
t 1=
_
x
2
1.
and
cosh
1
x = t = ln(sinht +cosht ) = ln(x +
_
x
2
1).
Becausecosht 1for all t , thislast expressionisonlyvalidfor x = cosht 1.
Verifythat tanh
1
x =
1
2
ln
1+x
1x
for |x| < 1.

33. Evaluate
_
_
x
2
+16dx usingtrigonometric substitution. ThenuseExercise31toverify that your answer agrees
withtheanswer inExample3.
solution Let x = 4tan. Thendx = 4sec
2
d and
_
_
x
2
+16dx = 16
_
sec
3
d = 8tan sec +8
_
sec d = 8tan sec +8ln|sec +tan| +C
= 8
x
4
_
x
2
+16
4
+8ln
_
x
2
+16
4
+
x
4
+C
=
1
2
x
_
x
2
+16+8ln
x
4
+
_
_
x
4
_
2
+1
+C.
UsingExercise31,
ln
x
4
+
_
_
x
4
_
2
+1
= sinh
1
_
x
4
_
,
sowecanwritetheantiderivativeas
1
2
x
_
x
2
+16+8sinh
1
_
x
4
_
+C,
whichagreeswiththeanswer inExample3.
Evaluate
_
_
x
2
9dx intwo ways: usingtrigonometric substitutionandusinghyperbolic substitution. Then
useExercise31toverifythat thetwoanswersagree.
35. Provethereductionformulafor n 2:
_
cosh
n
x dx =
1
n
cosh
n1
x sinhx +
n 1
n
_
cosh
n2
x dx 2
solution UsingIntegrationbyPartswithu = cosh
n1
x andv
= coshx, wehave
_
cosh
n
x dx = cosh
n1
x sinhx (n 1)
_
cosh
n2
x sinh
2
x dx
= cosh
n1
x sinhx (n 1)
_
cosh
n
x dx +(n 1)
_
cosh
n2
x dx.
Adding(n 1)
_
cosh
n
x dx tobothsidesthenyields
n
_
cosh
n
x dx = cosh
n1
x sinhx +(n 1)
_
cosh
n2
x dx.
Finally,
_
cosh
n
x dx =
1
n
cosh
n1
x sinhx +
n 1
n
_
cosh
n2
x dx.
UseEq. (2) toevaluate
_
cosh
4
x dx.
37.
_
tanh
1
x dx
x
2
1
solution Let u = tanh
1
x. Thendu =
1
1x
2
dx =
1
x
2
1
dx and
_
tanh
1
x
x
2
1
dx =
_
u du =
1
2
u
2
+C =
1
2
_
tanh
1
x
_
2
+C.
_
sinh
1
x dx
39.
_
tanh
1
x dx
solution UsingIntegrationbyPartswithu = tanh
1
x andv
= 1,
_
tanh
1
x dx = x tanh
1
x
_
x
1x
2
dx = x tanh
1
x +
1
2
ln|1x
2
| +C.
_
x tanh
1
x dx
41. Showthat if u = tanh(x/2), then
coshx =
1+u
2
1u
2
, sinhx =
2u
1u
2
, dx =
2du
1u
2
Hint: For therst relation, usetheidentities
sinh
2
_
x
2
_
=
1
2
(coshx 1), cosh
2
_
x
2
_
=
1
2
(coshx +1)
solution Let u = tanh(x/2). Then
u =
sinh(x/2)
cosh(x/2)
=
_
coshx 1
coshx +1
.
Solvingfor coshx yields
coshx =
1+u
2
1u
2
.
S E C T I ON 8.5 The Method of Partial Fractions 485
Next,
sinhx =
_
cosh
2
x 1=
_
(1+u
2
)
2
(1u
2
)
2
(1u
2
)
2
=
2u
1u
2
.
Finally, if u = tanh(x/2), thenx = 2tanh
1
u and
dx =
2du
1u
2
.
Exercises42and43: evaluateusingthesubstitutionof Exercise41.
_
sechx dx
43.
_
dx
1+coshx
solution Let u = tanh(x/2). Then, byExercise41,
1+coshx = 1+
1+u
2
1u
2
=
2
1u
2
and dx =
2du
1u
2
,
so
_
dx
1+coshx
=
_
du = u +C = tanh
x
2
+C.
Supposethat y = f (x) satisesy
= y. Prove:
(a) f (x)
2
(f
(x))
2
isconstant.
(b) If f (0) = f
(0) = 0, thenf (x) isthezerofunction.

(c) f (x) = f (0) coshx +f
(0) sinhx.
Hint: Refer toTheorem1inSection5.8.
Exercises4548refer tothefunctiongd(y) = tan
1
(sinhy), calledthegudermannian. Inamapof theearthconstructed
byMercator projection, pointslocatedy radial unitsfromtheequator correspondtopointsontheglobeof latitudegd(y).
45. Provethat
d
dy
gd(y) = sechy.
solution Let gd(y) = tan
1
(sinhy). Then
d
dy
gd(y) =
1
1+sinh
2
y
coshy =
1
coshy
= sechy,
wherewehaveusedtheidentity1+sinh
2
y = cosh
2
y.
Let f (y) = 2tan
1
(e
y
) /2. Provethat gd(y) = f (y). Hint: Showthat gd
(y) = f
(y) andf (0) = g(0).

47. Let t (y) = sinh
1
(tany). Showthat t (y) istheinverseof gd(y) for 0 y < /2.
solution Let x = gd(y) = tan
1
(sinhy). Solvingfor y yieldsy = sinh
1
(tanx). Therefore,
gd
1
(y) = sinh
1
(tany).
Verifythat t (y) inExercise47satisest
(y) = secy, andndavalueof a suchthat

t (y) =
_
y
a
dt
cost
49. Therelationscosh(it ) = cost andsinh(it ) = i sint werediscussedintheExcursion. Usetheserelationstoshow
that theidentitycos
2
t +sin
2
t = 1resultsfromsettingx = it intheidentitycosh
2
x sinh
2
x = 1.
solution Let x = it . Then
cosh
2
x = (cosh(it ))
2
= cos
2
t
and
sinh
2
x = (sinh(it ))
2
= i
2
sin
2
t = sin
2
t.
Thus,
1= cosh
2
(it ) sinh
2
(it ) = cos
2
t (sin
2
t ) = cos
2
t +sin
2
t,
asdesired.
8.5 The Method of Partial Fractions
1. Supposethat
_
f (x) dx = lnx +
x +1+C. Canf (x) bearational function? Explain.

solution No, f (x) cannotbearational functionbecausetheintegral of arational functioncannotcontainatermwith
anon-integer exponent suchas
x +1.
2. Whichof thefollowingareproper rational functions?
(a)
x
x 3
(b)
4
9x
(c)
x
2
+12
(x +2)(x +1)(x 3)
(d)
4x
3
7x
(x 3)(2x +5)(9x)
solution
(a) No, this is not a proper rational function because the degree of the numerator is not less than the degree of the
denominator.
(b) Yes, thisisaproper rational function.
(c) Yes, thisisaproper rational function.
(d) No, this is not a proper rational function because the degree of the numerator is not less than the degree of the
denominator.
3. Whichof thefollowingquadraticpolynomialsareirreducible?Tocheck, completethesquareif necessary.
(a) x
2
+5 (b) x
2
5
(c) x
2
+4x +6 (d) x
2
+4x +2
solution
(a) Squareisalreadycompleted; irreducible.
(b) Squareisalreadycompleted; factorsas(x
5)(x +
5).
(c) x
2
+4x +6= (x +2)
2
+2; irreducible.
(d) x
2
+4x +2= (x +2)
2
2; factorsas(x +2
2)(x +2+
2).
4. LetP(x)/Q(x) beaproper rational functionwhereQ(x) factorsasaproductof distinctlinear factors(x a
i
). Then
_
P(x) dx
Q(x)
(choosethecorrect answer):
(a) isasumof logarithmictermsA
i
ln(x a
i
) for someconstantsA
i
.
(b) maycontainaterminvolvingthearctangent.
solution Thecorrect answer is(a): theintegral isasumof logarithmictermsA
i
ln(x a
i
) for someconstantsA
i
.
Exercises
1. Matchtherational functions(a)(d) withthecorrespondingpartial fractiondecompositions(i)(iv).
(a)
x
2
+4x +12
(x +2)(x
2
+4)
(b)
2x
2
+8x +24
(x +2)
2
(x
2
+4)
(c)
x
2
4x +8
(x 1)
2
(x 2)
2
(d)
x
4
4x +8
(x +2)(x
2
+4)
(i) x 2+
4
x +2
4x 4
x
2
+4
(ii)
8
x 2
+
4
(x 2)
2
+
8
x 1
+
5
(x 1)
2
(iii)
1
x +2
+
2
(x +2)
2
+
x +2
x
2
+4
(iv)
1
x +2
+
4
x
2
+4
solution
(a)
x
2
+4x +12
(x +2)(x
2
+4)
=
1
x +2
+
4
x
2
+4
.
(b)
2x
2
+8x +24
(x +2)
2
(x
2
+4)
=
1
x +2
+
2
(x +2)
2
+
x +2
x
2
+4
.
(c)
x
2
4x +8
(x 1)
2
(x 2)
2
=
8
x 2
+
4
(x 2)
2
+
8
x 1
+
5
(x 1)
2
.
(d)
x
4
4x +8
(x +2)(x
2
+4)
= x 2+
4
x +2
4x 4
x
2
+4
.
DeterminetheconstantsA, B:
2x 3
(x 3)(x 4)
=
A
x 3
+
B
x 4
3. Clear denominatorsinthefollowingpartial fractiondecompositionanddeterminetheconstant B (substituteavalue
of x or usethemethodof undeterminedcoefcients).
3x
2
+11x +12
(x +1)(x +3)
2
=
1
x +1
B
x +3
3
(x +3)
2
solution Clearingdenominatorsgives
3x
2
+11x +12= (x +3)
2
B(x +1)(x +3) 3(x +1).
Settingx = 0thenyields
12= 9B(1)(3) 3(1) or B = 2.
Tousethemethodof undeterminedcoefcients, expandtheright-handsideandgather liketerms:
3x
2
+11x +12= (1B)x
2
+(34B)x +(63B).
Equatingx
2
-coefcientsonbothsides, wend
3= 1B or B = 2.
Findtheconstantsinthepartial fractiondecomposition
2x +4
(x 2)(x
2
+4)
=
A
x 2
+
Bx +C
x
2
+4
InExercises58, evaluateusinglongdivisionrsttowritef (x) asthesumof apolynomial andaproper rational function.
5.
_
x dx
3x 4
solution Longdivisiongivesus
x
3x 4
=
1
3
+
4/3
3x 4
Thereforetheintegral is
_
x
3x 4
dx =
_
1
3
4
9x 12
dx =
1
3
x
4
9
ln|9x 12| +C
_
(x
2
+2) dx
x +3
7.
_
(x
3
+2x
2
+1) dx
x +2
solution Longdivisiongivesus
x
3
+2x
2
+1
x +2
= x
2
+
1
x +2
Thereforetheintegral is
_
x
3
+2x
2
+1
x +2
dx =
_
x
2
+
1
x +2
dx =
1
3
x
3
+ln|x +2| +C
_
(x
3
+1) dx
x
2
+1
9.
_
dx
(x 2)(x 4)
solution Thepartial fractiondecompositionhastheform:
1
(x 2)(x 4)
=
A
x 2
+
B
x 4
.
Clearingdenominatorsgivesus
1= A(x 4) +B(x 2).
1= A(24) +0 or A =
1
2
,
whilesettingx = 4yields
1= 0+B(42) or B =
1
2
.
Theresult is:
1
(x 2)(x 4)
=

1
2
x 2
+
1
2
x 4
.
Thus,
_
dx
(x 2)(x 4)
=
1
2
_
dx
x 2
+
1
2
_
dx
x 4
=
1
2
ln|x 2| +
1
2
ln|x 4| +C.
_
(x +3) dx
x +4
11.
_
dx
x(2x +1)
1
x(2x +1)
=
A
x
+
B
2x +1
.
1= A(2x +1) +Bx.
1= A(1) +0 or A = 1,
whilesettingx =
1
2
yields
1= 0+B
_
1
2
_
or B = 2.
Theresult is:
1
x(2x +1)
=
1
x
+
2
2x +1
.
Thus,
_
dx
x(2x +1)
=
_
dx
x
_
2dx
2x +1
= ln|x| ln|2x +1| +C.
For theintegral ontheright, wehaveusedthesubstitutionu = 2x +1, du = 2dx.
_
(2x 1) dx
x
2
5x +6
13.
_
x
2
dx
x
2
+9
solution
_
x
2
x
2
+9
dx =
_
1
9
x
2
+9
dx = x 3tan
1
_
x
3
_
+C
_
dx
(x 2)(x 3)(x +2)
15.
_
(x
2
+3x 44) dx
(x +3)(x +5)(3x 2)
x
2
+3x 44
(x +3)(x +5)(3x 2)
=
A
x +3
+
B
x +5
+
C
3x 2
.
x
2
+3x 44= A(x +5)(3x 2) +B(x +3)(3x 2) +C(x +3)(x +5).
9944= A(2)(11) +0+0 or A = 2,
251544= 0+B(2)(17) +0 or B = 1,
andsettingx =
2
3
yields
4
9
+244= 0+0+C
_
11
3
__
17
3
_
or C = 2.
Theresult is:
x
2
+3x 44
(x +3)(x +5)(3x 2)
=
2
x +3
+
1
x +5
+
2
3x 2
.
Thus,
_
(x
2
+3x 44) dx
(x +3)(x +5)(3x 2)
= 2
_
dx
x +3
_
dx
x +5
2
_
dx
3x 2
= 2ln|x +3| ln|x +5|
2
3
ln|3x 2| +C.
Toevaluatethelast integral, wehavemadethesubstitutionu = 3x 2, du = 3dx.
_
3dx
(x +1)(x
2
+x)
17.
_
(x
2
+11x) dx
(x 1)(x +1)
2
x
2
+11x
(x 1)(x +1)
2
=
A
x 1
+
B
x +1
+
C
(x +1)
2
.
x
2
+11x = A(x +1)
2
+B(x 1)(x +1) +C(x 1).
12= A(4) +0+0 or A = 3,
10= 0+0+C(2) or C = 5.
Plugginginthesevaluesresultsin
x
2
+11x = 3(x +1)
2
+B(x 1)(x +1) +5(x 1).
Theconstant B canbedeterminedbyplugginginfor x anyvalueother than1or 1. If wepluginx = 0, weget
0= 3+B(1)(1) +5(1) or B = 2.
Theresult is
x
2
+11x
(x 1)(x +1)
2
=
3
x 1
+
2
x +1
+
5
(x +1)
2
.
Thus,
_
(x
2
+11x) dx
(x 1)(x +1)
2
= 3
_
dx
x 1
2
_
dx
x +1
+5
_
dx
(x +1)
2
= 3ln|x 1| 2ln|x +1|
5
x +1
+C.
_
(4x
2
21x) dx
(x 3)
2
(2x +3)
19.
_
dx
(x 1)
2
(x 2)
2
1
(x 1)
2
(x 2)
2
=
A
x 1
+
B
(x 1)
2
+
C
x 2
+
D
(x 2)
2
.
1= A(x 1)(x 2)
2
+B(x 2)
2
+C(x 2)(x 1)
2
+D(x 1)
2
.
1= B(1) or B = 1,
1= D(1) or D = 1.
Plugginginthesevaluesgivesus
1= A(x 1)(x 2)
2
+(x 2)
2
+C(x 2)(x 1)
2
+(x 1)
2
.
Settingx = 0nowyields
1= A(1)(4) +4+C(2)(1) +1 or 4= 4A 2C,
1= A(2)(1) +1+C(1)(4) +4 or 4= 2A +4C.
Solvingthissystemof twoequationsintwounknownsgivesA = 2andC = 2. Theresult is
1
(x 1)
2
(x 2)
2
=
2
x 1
+
1
(x 1)
2
+
2
x 2
+
1
(x 2)
2
.
Thus,
_
dx
(x 1)
2
(x 2)
2
= 2
_
dx
x 1
+
_
dx
(x 1)
2
2
_
dx
x 2
+
_
dx
(x 2)
2
= 2ln|x 1|
1
x 1
2ln|x 2|
1
x 2
+C.
_
(x
2
8x) dx
(x +1)(x +4)
3
21.
_
8dx
x(x +2)
3
solution Thepartial fractiondecompositionis
8
x(x +2)
3
=
A
x
+
B
x +2
+
C
(x +2)
2
+
D
(x +2)
3
Clearingfractionsgives
8= A(x +2)
3
+Bx(x +2)
2
+Cx(x +2) +Dx
Settingx = 0gives8= 8A soA = 1; settingx = 2gives8= 2D sothat D = 4; theresult is
8= (x +2)
3
+Bx(x +2)
2
+Cx(x +2) 4x
Thecoefcient of x
3
ontheright-handsidemust bezero, sinceit iszeroontheleft. Wecomputeit tobe1+B, sothat
B = 1. Finally, welookat thecoefcient of x
2
ontheright-handside; it must bezeroaswell. Wecomputeit tobe
3 24+C = C +2
sothat C = 2andthepartial fractiondecompositionis
8
x(x +2)
3
=
1
x
1
x +2
2
(x +2)
2
4
(x +2)
3
and
_
8
x(x +2)
3
dx =
_
1
x
dx
1
x +2
dx 2
_
(x +2)
2
dx 4
_
(x +2)
3
dx
= ln|x| ln|x +2| +2(x +2)
1
+2(x +2)
2
+C = ln
x
x +2
+
2
x +2
+
2
(x +2)
2
+C
_
x
2
dx
x
2
+3
23.
_
dx
2x
2
3
solution Thepartial fractiondecompositionhastheform
1
2x
2
3
=
1
(
2x
3)(
2x +
3)
=
A
2x
3
+
B
2x +
3
.
Clearingdenominators, weget
1= A
_
2x +
3
_
+B
_
2x
3
_
.
Settingx =
3/
2thenyields
1= A
_
3+
3
_
+0 or A =
1
2
3
,
whilesettingx =
3/
2yields
1= 0+B
_
3
_
or B =
1
2
3
.
Theresult is
1
2x
2
3
=
1/2
2x
1/2
2x +
3
.
Thus,
_
dx
2x
2
3
=
1
2
3
_
dx
2x
1
2
3
_
dx
2x +
3
.
For therst integral, let u =
2x
3, du =
2dx, andfor thesecond, let w =
2x +
3, dw =
2dx. Thenwe
have
_
dx
2x
2
3
=
1
2
3(
2)
_
du
u
1
2
3(
2)
_
dw
w
=
1
2
6
ln
2x

1
2
6
ln
2x +
+C.
_
dx
(x 4)
2
(x 1)
25.
_
4x
2
20
(2x +5)
3
dx
solution Thepartial fractiondecompositionis
4x
2
20
(2x +5)
3
=
A
2x +5
+
B
(2x +5)
2
+
C
(2x +5)
3
Clearingfractionsgives
4x
2
20= A(2x +5)
2
+B(2x +5) +C
Settingx = 5/2gives5= C sothat C = 5. Thecoefcient of x
2
ontheleft-handsideis4, andontheright-handside
is4A, sothat A = 1andwehave
4x
2
20= (2x +5)
2
+B(2x +5) +5
Consideringtheconstant termsnowgives20= 25+5B +5sothat B = 10. Thus
_
4x
2
20
(2x +5)
3
=
_
1
2x +5
dx 10
_
1
(2x +5)
2
dx +5
_
1
(2x +5)
3
dx
=
1
2
ln|2x +5| +
5
2x +5
5
4(2x +5)
2
+C
_
3x +6
x
2
(x 1)(x 3)
dx
27.
_
dx
x(x 1)
3
1
x(x 1)
3
=
A
x
+
B
x 1
+
C
(x 1)
2
+
D
(x 1)
3
.
1= A(x 1)
3
+Bx(x 1)
2
+Cx(x 1) +Dx.
1= A(1) +0+0+0 or A = 1,
1= 0+0+0+D(1) or D = 1.
PlugginginA = 1andD = 1givesus
1= (x 1)
3
+Bx(x 1)
2
+Cx(x 1) +x.
Now, settingx = 2yields
1= 1+2B +2C +2 or 2B +2C = 0,
andsettingx = 3yields
1= 8+12B +6C +3 or 2B +C = 1.
Solvingthesetwoequationsintwounknowns, wendB = 1andC = 1. Theresult is
1
x(x 1)
3
=
1
x
+
1
x 1
+
1
(x 1)
2
+
1
(x 1)
3
.
Thus,
_
dx
x(x 1)
3
=
_
dx
x
+
_
dx
x 1
_
dx
(x 1)
2
+
_
dx
(x 1)
3
= ln|x| +ln|x 1| +
1
x 1
1
2(x 1)
2
+C.
_
(3x
2
2) dx
x 4
29.
_
(x
2
x +1) dx
x
2
+x
solution First uselongdivisiontowrite
x
2
x +1
x
2
+x
= 1+
2x +1
x
2
+x
= 1+
2x +1
x(x +1)
.
Thepartial fractiondecompositionof thetermontheright hastheform:
2x +1
x(x +1)
=
A
x
+
B
x +1
.
2x +1= A(x +1) +Bx.
1= A(1) +0 or A = 1,
3= 0+B(1) or B = 3.
Theresult is
2x +1
x(x +1)
=
1
x
+
3
x +1
.
Thus,
_
(x
2
x +1) dx
x
2
+x
=
_
dx +
_
dx
x
3
_
dx
x +1
= x +ln|x| 3ln|x +1| +C.
_
dx
x(x
2
+1)
31.
_
(3x
2
4x +5) dx
(x 1)(x
2
+1)
3x
2
4x +5
(x 1)(x
2
+1)
=
A
x 1
+
Bx +C
x
2
+1
.
3x
2
4x +5= A(x
2
+1) +(Bx +C)(x 1).
34+5= A(2) +0 or A = 2.
Thisgivesus
3x
2
4x +5= 2(x
2
+1) +(Bx +C)(x 1) = (B +2)x
2
+(C B)x +(2C).
Equatingx
2
-coefcients, wend
3= B +2 or B = 1;
whileequatingconstant coefcientsyields
5= 2C or C = 3.
Theresult is
3x
2
4x +5
(x 1)(x
2
+1)
=
2
x 1
+
x 3
x
2
+1
.
Thus,
_
(3x
2
4x +5) dx
(x 1)(x
2
+1)
= 2
_
dx
x 1
+
_
(x 3) dx
x
2
+1
= 2
_
dx
x 1
+
_
x dx
x
2
+1
3
_
dx
x
2
+1
.
For thesecondintegral, usethesubstitutionu = x
2
+1, du = 2x dx. Thenal answer is
_
(3x
2
4x +5) dx
(x 1)(x
2
+1)
= 2ln|x 1| +
1
2
ln|x
2
+1| 3tan
1
x +C.
_
x
2
(x +1)(x
2
+1)
dx
33.
_
dx
x(x
2
+25)
1
x(x
2
+25)
=
A
x
+
Bx +C
x
2
+25
.
1= A(x
2
+25) +(Bx +C)x.
1= A(25) +0 or A =
1
25
.
Thisgivesus
1=
1
25
x
2
+1+Bx
2
+Cx =
_
B +
1
25
_
x
2
+Cx +1.
Equatingx
2
-coefcients, wend
0= B +
1
25
or B =
1
25
,
whileequatingx-coefcientsyieldsC = 0. Theresult is
1
x(x
2
+25)
=
1
25
x
+

1
25
x
x
2
+25
.
Thus,
_
dx
x(x
2
+25)
=
1
25
_
dx
x
1
25
_
x dx
x
2
+25
.
For theintegral ontheright, useu = x
2
+25, du = 2x dx. Thenwehave
_
dx
x(x
2
+25)
=
1
25
ln|x|
1
50
ln|x
2
+25| +C.
_
dx
x
2
(x
2
+25)
35.
_
(6x
2
+2) dx
x
2
+2x 3
solution Longdivisiongives
6x
2
+2
x
2
+2x 3
= 6
12x 20
x
2
+2x 3
= 6
12x 20
(x +3)(x 1)
Thepartial fractiondecompositionof thesecondtermis
12x 20
(x +3)(x 1)
=
A
x +3
+
B
x 1
Clear fractionstoget
12x 20= A(x 1) +B(x +3)
Set x = 1toget 8= 4B sothat B = 2. Set x = 3toget 56= 4A sothat A = 14, andwehave
_
6x
2
+2
x
2
+2x 3
=
_
6
14
x +3
+
2
x 1
dx =
_
6dx 14
_
1
x +3
dx +2
_
1
x 1
dx
= 6x 14ln|x +3| +2ln|x 1| +C
_
6x
2
+7x 6
(x
2
4)(x +2)
dx
37.
_
10dx
(x 1)
2
(x
2
+9)
10
(x 1)
2
(x
2
+9)
=
A
x 1
+
B
(x 1)
2
+
Cx +D
x
2
+9
.
10= A(x 1)(x
2
+9) +B(x
2
+9) +(Cx +D)(x 1)
2
.
10= 0+B(10) +0 or B = 1.
Expandingtheright-handside, wehave
10= (A +C)x
3
+(1A 2C +D)x
2
+(9A +C 2D)x +(99A +D).
Equatingcoefcientsof likepowersof x thenyields
A +C = 0
1A 2C +D = 0
9A +C 2D = 0
99A +D = 10
Fromtherst equation, wehaveC = A, andfromthefourthequationwehaveD = 1+ 9A. Substitutingtheseinto
thesecondequation, weget
1A 2(A) +(1+9A) = 0 or A =
1
5
.
Finally, C =
1
5
andD =
4
5
. Theresult is
10
(x 1)
2
(x
2
+9)
=

1
5
x 1
+
1
(x 1)
2
+
1
5
x
4
5
x
2
+9
.
Thus,
_
10dx
(x 1)
2
(x
2
+9)
=
1
5
_
dx
x 1
+
_
dx
(x 1)
2
+
1
5
_
x dx
x
2
+9
4
5
_
dx
x
2
+9
=
1
5
ln|x 1|
1
x 1
+
1
10
ln|x
2
+9|
4
15
tan
1
_
x
3
_
+C.
_
10dx
(x +1)(x
2
+9)
2
39.
_
dx
x(x
2
+8)
2
1
x(x
2
+8)
2
=
A
x
+
Bx +C
x
2
+8
+
Dx +E
(x
2
+8)
2
.
1= A(x
2
+8)
2
+(Bx +C)x(x
2
+8) +(Dx +E)x.
Expandingtheright-handsidegivesus
1= (A +B)x
4
+Cx
3
+(16A +8B +D)x
2
+(8C +E)x +64A.
Equatingcoefcientsof likepowersof x yields
A +B = 0
C = 0
16A +8B +D = 0
8C +E = 0
64A = 1
Thesolutiontothissystemof equationsis
A =
1
64
, B =
1
64
, C = 0, D =
1
8
, E = 0.
Therefore
1
x(x
2
+8)
2
=
1
64
x
+

1
64
x
x
2
+8
+

1
8
x
(x
2
+8)
2
,
and
_
dx
x(x
2
+8)
2
=
1
64
_
dx
x
1
64
_
x dx
x
2
+8
1
8
_
x dx
(x
2
+8)
2
.
For thesecondandthirdintegrals, usethesubstitutionu = x
2
+8, du = 2x dx. Thenwehave
_
dx
x(x
2
+8)
2
=
1
64
ln|x|
1
128
ln|x
2
+8| +
1
16(x
2
+8)
+C.
_
100x dx
(x 3)(x
2
+1)
2
41.
_
dx
(x +2)(x
2
+4x +10)
1
(x +2)(x
2
+4x +10)
=
A
x +2
+
Bx +C
x
2
+4x +10
.
1= A(x
2
+4x +10) +(Bx +C)(x +2).
1= A(6) +0 or A =
1
6
.
Expandingtheright-handsidegivesus
1=
_
1
6
+B
_
x
2
+
_
2
3
+2B +C
_
x +
_
5
3
+2C
_
.
Equatingx
2
-coefcientsyields
0=
1
6
+B or B =
1
6
,
whileequatingconstant coefcientsyields
1=
5
3
+2C or C =
1
3
.
Theresult is
1
(x +2)(x
2
+4x +10)
=
1
6
x +2
+

1
6
x
1
3
x
2
+4x +10
.
Thus,
_
dx
(x +2)(x
2
+4x +10)
=
1
6
_
dx
x +2
1
6
_
(x +2) dx
x
2
+4x +10
.
For thesecondintegral, let u = x
2
+4x +10. Thendu = (2x +4) dx, and
_
dx
(x +2)(x
2
+4x +10)
=
1
6
ln|x +2|
1
12
_
(2x +4) dx
x
2
+4x +10
=
1
6
ln|x +2|
1
12
ln|x
2
+4x +10| +C.
_
9dx
(x +1)(x
2
2x +6)
43.
_
25dx
x(x
2
+2x +5)
2
25
x(x
2
+2x +5)
2
=
A
x
+
Bx +C
x
2
+2x +5
+
Dx +E
(x
2
+2x +5)
2
.
Clearingdenominatorsyields:
25= A(x
2
+2x +5)
2
+x(Bx +C)(x
2
+2x +5) +x(Dx +E)
= (Ax
4
+4Ax
3
+14Ax
2
+20Ax +25A) +(Bx
4
+Cx
3
+2Bx
3
+2Cx
2
+5Bx
2
+5Cx) +Dx
2
+Ex.
Equatingconstant termsyields
25A = 25 or A = 1,
whileequatingx
4
-coefcientsyields
A +B = 0 or B = A = 1.
Equatingx
3
-coefcientsyields
4A +C +2B = 0 or C = 2,
andequatingx
2
-coefcientsyields
14A +2C +5B +D = 0 or D = 5.
Finally, equatingx-coefcientsyields
20A +5C +E = 0 or E = 10.
Thus,
_
25dx
x(x
2
+2x +5)
2
=
_ _
1
x
x +2
x
2
+2x +5
5
x +2
(x
2
+2x +5)
2
_
dx
= ln|x|
_
x +2
x
2
+2x +5
dx 5
_
x +2
(x
2
+2x +5)
2
dx.
Thetwointegralsontheright bothrequirethesubstitutionu = x + 1, sothat x
2
+ 2x + 5 = (x + 1)
2
+ 4 = u
2
+ 4
anddu = dx. Thismeans:
_
25dx
x(x
2
+2x +5)
2
= ln|x|
_
u +1
u
2
+4
du 5
_
u +1
(u
2
+4)
2
du
= ln|x|
_
u
u
2
+4
du
_
1
u
2
+4
du 5
_
u
(u
2
+4)
2
du 5
_
1
(u
2
+4)
2
du.
For therst andthirdintegrals, wemakethesubstitutionw = u
2
+4, dw = 2u du. Thenwehave
_
25dx
x(x
2
+2x +5)
2
= ln|x|
1
2
ln|u
2
+4|
1
2
tan
1
_
u
2
_
+
5
2(u
2
+4)
5
_
du
(u
2
+4)
2
= ln|x|
1
2
ln|x
2
+2x +5|
1
2
tan
1
_
x +1
2
_
+
5
2(x
2
+2x +5)
5
_
du
(u
2
+4)
2
.
For theremainingintegral, weusethetrigonometricsubstitution2tanw = u, sothat u
2
+4= 4tan
2
w +4= 4sec
2
w
anddu = 2sec
2
wdw. Thismeans
_
1
(u
2
+4)
2
du =
1
8
_
1
sec
4
w
sec
2
wdw =
1
8
_
cos
2
wdw
=
1
8
_
1
4
sin2w +
w
2
_
+C =
_
1
16
sinwcosw +
w
16
_
+C
=
1
16
u
_
u
2
+4
2
_
u
2
+4
+
1
16
tan
1
_
u
2
_
+C =
1
8
u
u
2
+4
+
1
16
tan
1
_
u
2
_
+C
=
1
8
x +1
x
2
+2x +5
+
1
16
tan
1
_
x +1
2
_
.
Hence, theintegral is
_
25dx
x(x
2
+2x +5)
2
= ln|x|
1
2
ln|x
2
+2x +5|
1
2
tan
1
_
x +1
2
_
+
5
2(x
2
+2x +5)
5
8
x +1
x
2
+2x +5
5
16
tan
1
_
x +1
2
_
= ln|x| +
155x
8(x
2
+2x +5)
13
16
tan
1
_
x +1
2
_
1
2
ln|x
2
+2x +5| +C.
_
(x
2
+3) dx
(x
2
+2x +3)
2
InExercises4548, evaluatebyusingrst substitutionandthenpartial fractionsif necessary.
45.
_
x dx
x
4
+1
2
sothat du = 2x dx, and
_
x
x
4
+1
dx =
1
2
_
1
u
2
+1
du =
1
2
tan
1
u =
1
2
tan
1
(x
2
)
_
x dx
(x +2)
4
47.
_
e
x
dx
e
2x
e
x
x
. Thendu = e
x
dx = u dx sothat dx =
1
u
du. Then
_
e
x
dx
e
2x
e
x
=
_
u
1
u
du
u
2
u
=
_
1
u(u 1)
du
Usingpartial fractions, wehave
1
u(u 1)
=
A
u
+
B
u 1
=
(A +B)u A
u(u 1)
Uponequatingcoefcientsinthenumerators, wehaveA +B = 0, A = 1sothat B = 1. Then
_
e
x
dx
e
2x
e
x
=
_
1
u
du +
_
1
u 1
du = ln|u 1| ln|u| +C = ln|e
x
1| lne
x
+C
_
sec
2
d
tan
2
1
49. Evaluate
_
x dx
x 1
. Hint: Usethesubstitutionu =
x (sometimescalledarationalizingsubstitution).
solution Let u =
x. Thendu = (1/2
x) dx = (1/2u) dx. Thus

_
x dx
x 1
=
_
u(2u du)
u
2
1
= 2
_
u
2
du
u
2
1
= 2
_
(u
2
1+1) du
u
2
1
= 2
_
_
u
2
1
u
2
1
+
1
u
2
1
_
du = 2
_
du +
_
2du
u
2
1
= 2u +
_
2du
u
2
1
.
Thepartial fractiondecompositionof theremainingintegral hastheform:
2
u
2
1
=
2
(u 1)(u +1)
=
A
u 1
+
B
u +1
.
2= A(u +1) +B(u 1).
Settingu = 1yields2= A(2) +0or A = 1, whilesettingu = 1yields2= 0+B(2) or B = 1. Theresult is
2
u
2
1
=
1
u 1
+
1
u +1
.
Thus,
_
2du
u
2
1
=
_
du
u 1
_
du
u +1
= ln|u 1| ln|u +1| +C.
Thenal answer is
_
x dx
x 1
= 2u +ln|u 1| ln|u +1| +C = 2
x +ln|
x 1| ln|
x +1| +C.
Evaluate
_
dx
x
1/2
x
1/3
.
51. Evaluate
_
dx
x
2
1
in two ways: using partial fractions and using trigonometric substitution. Verify that thetwo
answersagree.
1
x
2
1
=
1
(x 1)(x +1)
=
A
x 1
+
B
x +1
.
1= A(x +1) +B(x 1).
Settingx = 1, weget 1= A(2) or A =
1
2
; whilesettingx = 1, weget 1= B(2) or B =
1
2
. Theresult is
1
x
2
1
=
1
2
x 1
+

1
2
x +1
.
Thus,
_
dx
x
2
1
=
1
2
_
dx
x 1
1
2
_
dx
x +1
=
1
2
ln|x 1|
1
2
ln|x +1| +C.
Usingtrigonometricsubstitution, let x = sec. Thendx = tan sec d, andx
2
1= sec
2
1= tan
2
. Thus
_
dx
x
2
1
=
_
tan sec d
tan
2
=
_
sec d
tan
=
_
cos d
sin cos
=
_
csc d = ln| csc cot| +C.
Nowweconstruct aright trianglewithsec = x:
q
1
x
x
2
1
Fromthisweseethat csc = x/
_
x
2
1andcot = 1/
_
x
2
1. Thus
_
dx
x
2
1
= ln
x
_
x
2
1
1
_
x
2
1
+C = ln
x 1
_
x
2
1
+C.
Tocheckthat thesetwoanswersagree, wewrite
1
2
ln|x 1|
1
2
ln|x +1| =
1
2
x 1
x +1
= ln
_
x 1
x +1
= ln
x 1
x +1
x 1
x 1
= ln
x 1
_
x
2
1
.
Graph theequation (x 40)y
2
= 10x(x 30) and nd thevolumeof thesolid obtained by revolving the
regionbetweenthegraphandthex-axisfor 0 x 30aroundthex-axis.
InExercises 5366, evaluatetheintegral usingtheappropriatemethodor combinationof methods coveredthus far in
thetext.
53.
_
dx
x
2
_
4x
2
solution Usethetrigonometricsubstitutionx = 2sin. Thendx = 2cos d,
4x
2
= 44sin
2
= 4(1sin
2
) = 4cos
2
,
and
_
dx
x
2
_
4x
2
=
_
2cos d
(4sin
2
)(2cos)
=
1
4
_
csc
2
d =
1
4
cot +C.
Nowconstruct aright trianglewithsin = x/2:
q
x
2
4 x
2
Fromthisweseethat cot =
_
4x
2
/x. Thus
_
dx
x
2
_
4x
2
=
1
4
__
4x
2
x
_
+C =
_
4x
2
4x
+C.
_
dx
x(x 1)
2
55.
_
cos
2
4x dx
solution Usethesubstitutionu = 4x, du = 4dx. Thenwehave
_
cos
2
(4x) dx =
1
4
_
cos
2
(4x)4dx =
1
4
_
cos
2
u du =
1
4
_
1
2
u +
1
2
sinu cosu
_
+C
=
1
8
u +
1
8
sinu cosu +C =
1
2
x +
1
8
sin4x cos4x +C.
_
x sec
2
x dx
57.
_
dx
(x
2
+9)
2
solution Usethetrigonometricsubstitutionx = 3tan. Thendx = 3sec
2
d,
x
2
+9= 9tan
2
+9= 9(tan
2
+1) = 9sec
2
,
and
_
dx
(x
2
+9)
2
=
_
3sec
2
d
(9sec
2
)
2
=
3
81
_
sec
2
d
sec
4
=
1
27
_
cos
2
d =
1
27
_
1
2
+
1
2
sin cos
_
+C.
Nowconstruct aright trianglewithtan = x/3:
q
x
2
+ 9
x
3
Fromthisweseethat sin = x/
_
x
2
+9andcos = 3/
_
x
2
+9. Thus
_
dx
_
x
2
+9
2
=
1
54
tan
1
_
x
3
_
+
1
54
_
x
_
x
2
+9
__
3
_
x
2
+9
_
+C =
1
54
tan
1
_
x
3
_
+
x
18(x
2
+9)
+C.
_
sec
1
d
59.
_
tan
5
x secx dx
solution Usethetrigonometricidentitytan
2
x = sec
2
x 1towrite
_
tan
5
x secx dx =
_
_
sec
2
x 1
_
2
tanx secx dx.
Nowusethesubstitutionu = secx, du = secx tanx dx:
_
tan
5
x secx dx =
_
(u
2
1)
2
du =
_
_
u
4
2u
2
+1
_
du
=
1
5
u
5
2
3
u
3
+u +C =
1
5
sec
5
x
2
3
sec
3
x +secx +C.
_
(3x
2
1) dx
x(x
2
1)
61.
_
ln(x
4
1) dx
solution Applyintegrationbypartswithu = ln(x
4
1), v
= 1; thenu
=
4x
3
x
4
1
andv = x, soafter simplication,
_
ln(x
4
1) dx = x ln(x
4
1) 4
_
x
4
x
4
1
dx = x ln(x
4
1) 4
_
1+
1
x
4
1
dx
= x ln(x
4
1) 4
_
1dx 4
_
1
x
4
1
dx
= x ln(x
4
1) 4x 4
_
1
2
_
1
x
2
1
1
x
2
+1
_
dx
= x ln(x
4
1) 4x 2
_
1
x
2
1
dx +2
_
1
x
2
+1
dx
= x ln(x
4
1) 4x +2tanh
1
x +2tan
1
x +C
_
x dx
(x
2
1)
3/2
63.
_
x
2
dx
(x
2
1)
3/2
solution Usethetrigonometricsubstitutionx = sec. Thendx = sec tan d,
x
2
1= sec
2
1= tan
2
,
and
_
x
2
dx
(x
2
1)
3/2
=
_
(sec
2
) sec tan d
(tan
2
)
3/2
=
_
sec
3
d
tan
2
=
_
(tan
2
+1) sec d
tan
2
=
_
tan
2
sec d
tan
2
+
_
sec d
tan
2
=
_
sec d +
_
csc cot d
= ln| sec +tan| csc +C.
Nowconstruct aright trianglewithsec = x:
q
1
x
x
2
1
_
x
2
1andcsc = x/
_
x
2
1. Sothenal answer is
_
x
2
dx
(x
2
1)
3/2
= ln
x +
_
x
2
1

x
_
x
2
1
+C.
_
(x +1) dx
(x
2
+4x +8)
2
65.
_
x dx
x
3
+1
3/2
, du =
3
2
x
1/2
dx. Thenx
3
= (x
3/2
)
2
= u
2
, sowehave
_
x dx
x
3
+1
=
2
3
_
du
u
2
+1
=
2
3
tan
1
u +C =
2
3
tan
1
(x
3/2
) +C.
_
x
1/2
dx
x
1/3
+1
67. Showthat thesubstitution = 2tan
1
t (Figure2) yieldstheformulas
cos =
1t
2
1+t
2
, sin =
2t
1+t
2
, d =
2dt
1+t
2
10
Thissubstitutiontransformstheintegral of anyrational functionof cos andsin intoanintegral of arational function
of t (whichcanthenbeevaluatedusingpartial fractions). Useit toevaluate
_
d
cos +
3
4
sin
.
1
q/2
t
1+ t
2
FIGURE 2
solution If = 2tan
1
t , thend = 2dt /(1+t
2
). Wealsohavethatcos(
2
) = 1/
_
1+t
2
andsin(
2
) = t /
_
1+t
2
.
Tondcos, weusethedoubleangleidentitycos = 12sin
2
(
2
). Thisgivesus
cos = 12
_
t
_
1+t
2
_
2
= 1
2t
2
1+t
2
=
1+t
2
2t
2
1+t
2
=
1t
2
1+t
2
.
Tondsin, weusethedoubleangleidentitysin = 2sin(
2
) cos(
2
). Thisgivesus
sin = 2
_
t
_
1+t
2
__
1
_
1+t
2
_
=
2t
1+t
2
.
Withtheseformulas, wehave
_
d
cos +(3/4) sin
=
_ 2dt
1+t
2
_
1t
2
1+t
2
_
+
3
4
_
2t
1+t
2
_ =
_
8dt
4(1t
2
) +3(2t )
=
_
8dt
4+6t 4t
2
=
_
4dt
2+3t 2t
2
.
Thepartial fractiondecompositionhastheform
4
2+3t 2t
2
=
A
2t
+
B
1+2t
.
4= A(1+2t ) +B(2t ).
Settingt = 2thenyields
4= A(5) +0 or A =
4
5
,
whilesettingt =
1
2
yields
4= 0+B
_
5
2
_
or B =
8
5
.
Theresult is
4
2+3t 2t
2
=
4
5
2t
+
8
5
1+2t
.
Thus,
_
4
2+3t 2t
2
dt =
4
5
_
dt
2t
+
8
5
_
dt
1+2t
=
4
5
ln|2t | +
4
5
ln|1+2t | +C.
Theoriginal substitutionwas = 2tan
1
t , whichmeansthat t = tan(
2
). Thenal answer isthen
_
d
cos +
3
4
sin
=
4
5
ln
2tan
_
2
_
+
4
5
ln
1+2tan
_
2
_
+C.
Usethesubstitutionof Exercise67toevaluate
_
d
cos +sin
.
69. Provethegeneral formula
_
dx
(x a)(x b)
=
1
a b
ln
x a
x b
+C
wherea, b areconstantssuchthat a = b.
1
(x a)(x b)
=
A
x a
+
B
x b
.
1= A(x b) +B(x a).
Settingx = a thenyields
1= A(a b) +0 or A =
1
a b
,
whilesettingx = b yields
1= 0+B(b a) or B =
1
b a
.
Theresult is
1
(x a)(x b)
=
1
ab
x a
+
1
ba
x b
.
Thus,
_
dx
(x a)(x b)
=
1
a b
_
dx
x a
+
1
b a
_
dx
x b
=
1
a b
ln|x a| +
1
b a
ln|x b| +C
=
1
a b
ln|x a|
1
a b
ln|x b| +C =
1
a b
ln
x a
x b
+C.
Themethodof partial fractionsshowsthat
_
dx
x
2
1
=
1
2
ln
x 1
1
2
ln
x +1
+C
The computer algebra systemMathematica evaluates this integral as tanh
1
x, where tanh
1
x is the inverse
hyperbolictangent function. Canyoureconcilethetwoanswers?
71. Supposethat Q(x) = (x a)(x b), wherea = b, andlet P(x)/Q(x) beaproper rational functionsothat
P(x)
Q(x)
=
A
(x a)
+
B
(x b)
(a) Showthat A =
P(a)
Q
(a)
andB =
P(b)
Q
(b)
.
(b) Usethisresult tondthepartial fractiondecompositionfor P(x) = 3x 2andQ(x) = x
2
4x 12.
solution
(a) Clearingdenominatorsgivesus
P(x) = A(x b) +B(x a).
Settingx = a thenyields
P(a) = A(a b) +0 or A =
P(a)
a b
,
whilesettingx = b yields
P(b) = 0+B(b a) or B =
P(b)
b a
.
Nowusetheproduct ruletodifferentiateQ(x):
Q
(x) = (x a)(1) +(1)(x b) = x a +x b = 2x a b;

therefore,
Q
(a) = 2a a b = a b
Q
(b) = 2b a b = b a
S E C T I ON 8.6 Improper Integrals 503
Substitutingtheseintotheaboveresults, wend
A =
P(a)
Q
(a)
and B =
P(b)
Q
(b)
.
(b) Thepartial fractiondecompositionhastheform:
P(x)
Q(x)
=
3x 2
x
2
4x 12
=
3x 2
(x 6)(x +2)
=
A
x 6
+
B
x +2
;
A =
P(6)
Q
(6)
=
3(6) 2
2(6) 4
=
16
8
= 2;
B =
P(2)
Q
(2)
=
3(2) 2
2(2) 4
=
8
8
= 1.
Theresult is
3x 2
x
2
4x 12
=
2
x 6
+
1
x +2
.
Supposethat Q(x) = (x a
1
)(x a
2
) (x a
n
), wheretheroots a
j
areall distinct. Let P(x)/Q(x) bea
proper rational functionsothat
P(x)
Q(x)
=
A
1
(x a
1
)
+
A
2
(x a
2
)
+ +
A
n
(x a
n
)
(a) Showthat A
j
=
P(a
j
)
Q
(a
j
)
for j = 1, . . . , n.
(b) Usethis result to ndthepartial fractiondecompositionfor P(x) = 2x
2
1, Q(x) = x
3
4x
2
+ x + 6 =
(x +1)(x 2)(x 3).
8.6 Improper Integrals
1. Statewhether theintegral convergesor diverges:
(a)
_

1
x
3
dx (b)
_
1
0
x
3
dx
(c)
_

1
x
2/3
dx (d)
_
1
0
x
2/3
dx
solution
(a) Theintegral isimproper becauseoneof thelimitsof integrationisinnite. Becausethepower of x intheintegrand
islessthan1, thisintegral converges.
(b) Theintegral isimproper becausetheintegrandisundenedat x = 0. Becausethepower of x intheintegrandisless
than1, thisintegral diverges.
(c) Theintegral isimproper becauseoneof thelimitsof integrationisinnite. Becausethepower of x intheintegrand
isgreater than1, thisintegral diverges.
(d) Theintegral is improper becausetheintegrandis undenedat x = 0. Becausethepower of x intheintegrandis
greater than1, thisintegral converges.
2. Is
_
/2
0
cotx dx animproper integral? Explain.
solution Becausetheintegrandcotx isundenedat x = 0, thisisanimproper integral.
3. Findavalueof b > 0that makes
_
b
0
1
x
2
4
dx animproper integral.
solution Anyvalueof b satisfying|b| 2will makethisanimproper integral.
4. Whichcomparisonwouldshowthat
_

0
dx
x +e
x
converges?
solution Notethat, for x > 0,
1
x +e
x
<
1
e
x
= e
x
.
Moreover
_

0
e
x
dx
converges. Therefore,
_

0
1
x +e
x
dx
convergesbythecomparisontest.
5. Explainwhy it is not possibleto drawany conclusions about theconvergenceof
_

1
e
x
x
dx by comparingwith
theintegral
_

1
dx
x
.
solution For 1 x < ,
e
x
x
<
1
x
,
but
_

1
dx
x
diverges. Knowingthat anintegral issmaller thanadivergent integral doesnot allowustodrawany conclusionsusing
thecomparisontest.
Exercises
1. Whichof thefollowingintegralsisimproper? Explainyour answer, but donot evaluatetheintegral.
(a)
_
2
0
dx
x
1/3
(b)
_

1
dx
x
0.2
(c)
_

1
e
x
dx
(d)
_
1
0
e
x
dx (e)
_
/2
0
secx dx (f)
_

0
sinx dx
(g)
_
1
0
sinx dx (h)
_
1
0
dx
_
3x
2
(i)
_

1
lnx dx
(j)
_
3
0
lnx dx
solution
(a) Improper. Thefunctionx
1/3
isinniteat 0.
(b) Improper. Inniteinterval of integration.
(c) Improper. Inniteinterval of integration.
(d) Proper. Thefunctione
x
iscontinuousontheniteinterval [0, 1].
(e) Improper. Thefunctionsecx isinniteat

2
.
(f) Improper. Inniteinterval of integration.
(g) Proper. Thefunctionsinx iscontinuousontheniteinterval [0, 1].
(h) Proper. Thefunction1/
_
3x
2
iscontinuousontheniteinterval [0, 1].
(i) Improper. Inniteinterval of integration.
(j) Improper. Thefunctionlnx isinniteat 0.
Let f (x) = x
4/3
.
(a) Evaluate
_
R
1
f (x) dx.
(b) Evaluate
_

1
f (x) dx bycomputingthelimit
lim
R
_
R
1
f (x) dx
3. Provethat
_

1
x
2/3
dx divergesbyshowingthat
lim
R
_
R
1
x
2/3
dx =
solution First computetheproper integral:
_
R
1
x
2/3
dx = 3x
1/3
R
1
= 3R
1/3
3= 3
_
R
1/3
1
_
.
Thenshowdivergence:
_

1
x
2/3
dx = lim
R
_
R
1
x
2/3
dx = lim
R
3
_
R
1/3
1
_
= .
Determinewhether
_
3
0
dx
(3x)
3/2
convergesbycomputing
lim
R3
_
R
0
dx
(3x)
3/2
InExercises540, determinewhether theimproper integral convergesand, if so, evaluateit.
5.
_

1
dx
x
19/20
solution First evaluatetheintegral over theniteinterval [1, R] for R > 1:
_
R
1
dx
x
19/20
= 20x
1/20
R
1
= 20R
1/20
20.
Nowcomputethelimit asR :
_

1
dx
x
19/20
= lim
R
_
R
1
dx
x
19/20
= lim
R
_
20R
1/20
20
_
= .
Theintegral doesnot converge.
_

1
dx
x
20/19
7.
_
4
e
0.0001t
dt
solution First evaluatetheintegral over theniteinterval [R, 4] for R < 4:
_
4
R
e
(0.0001)t
dt =
e
(0.0001)t
0.0001
4
R
= 10,000
_
e
0.0004
e
(0.0001)R
_
.
_
4
e
(0.0001)t
dt = lim
R
_
4
R
e
(0.0001)t
dt = lim
R
10,000
_
e
0.0004
e
(0.0001)R
_
= 10,000
_
e
0.0004
0
_
= 10,000e
0.0004
.
_

20
dt
t
9.
_
5
0
dx
x
20/19
solution Thefunctionx
20/19
is inniteat theendpoint 0, so well rst evaluatetheintegral ontheniteinterval
[R, 5] for 0< R < 5:
_
5
R
dx
x
20/19
= 19x
1/19
5
R
= 19
_
5
1/19
R
1/19
_
= 19
_
1
R
1/19
1
5
1/19
_
.
Nowcomputethelimit asR 0
+
:
_
5
0
dx
x
20/19
= lim
R0
+
_
5
R
dx
x
20/19
= lim
R0
+
19
_
1
R
1/19
1
5
1/19
_
= ;
thus, theintegral doesnot converge.
_
5
0
dx
x
19/20
11.
_
4
0
dx
4x
solution Thefunction 1/
4x is inniteat x = 4, so well rst evaluatetheintegral on theinterval [0, R] for

0< R < 4:
_
R
0
dx
4x
= 2
4x
R
0
= 2
4R (2)
4= 42
4R.
:
_
4
0
dx
4x
= lim
R4
_
R
0
dx
4x
= lim
R4
_
42
4R
_
= 40= 4.
_
6
5
dx
(x 5)
3/2
13.
_

2
x
3
dx
solution First evaluatetheintegral ontheniteinterval [2, R] for 2< R:
_
R
2
x
3
dx =
x
2
2
R
2
=
1
2R
2
1
2(2
2
)
=
1
8
1
2R
2
.
_

2
x
3
dx = lim
R
_
R
2
x
3
dx = lim
R
_
1
8
1
2R
2
_
=
1
8
.
_

0
dx
(x +1)
3
15.
_

3
dx
(x +4)
3/2
solution First evaluatetheintegral ontheniteinterval [3, R] for R > 3:
_
R
3
dx
(x +4)
3/2
= 2(x +4)
1/2
R
3
=
2
R +4
1
= 2
2
R +4
.
_

3
dx
(x +4)
3/2
= lim
R
_
R
3
dx
(x +4)
3/2
= lim
R
_
2
2
R +4
_
= 20= 2.
_

2
e
2x
dx
17.
_
1
0
dx
x
0.2
solution Thefunctionx
0.2
isinniteatx = 0, sowell rstevaluatetheintegral ontheinterval [R, 1] for0< R < 1:
_
1
R
dx
x
0.2
=
x
0.8
0.8
1
R
= 1.25
_
1R
0.8
_
.
+
:
_
1
0
dx
x
0.2
= lim
R0
+
_
1
R
dx
x
0.2
= lim
R0
+
1.25
_
1R
0.8
_
= 1.25(10) = 1.25.
_

2
x
1/3
dx
19.
_

4
e
3x
dx
_
R
4
e
3x
dx =
e
3x
3
R
4
=
1
3
_
e
3R
e
12
_
=
1
3
_
e
12
e
3R
_
.
_

4
e
3x
dx = lim
R
_
R
4
e
3x
dx = lim
R
1
3
_
e
12
e
3R
_
=
1
3
_
e
12
0
_
=
1
3e
12
.
_

4
e
3x
dx
21.
_
0
e
3x
dx
solution First evaluatetheintegral ontheniteinterval [R, 0] for R < 0:
_
0
R
e
3x
dx =
e
3x
3
0
R
=
1
3
e
3R
3
.
_
0
e
3x
dx = lim
R
_
0
R
e
3x
dx = lim
R
_
1
3
e
3R
3
_
=
1
3
0=
1
3
.
_
2
1
dx
(x 1)
2
23.
_
3
1
dx
3x
solution Thefunctionf (x) = 1/
3x isinniteat x = 3, sowerst evaluatetheintegral ontheinterval [1, R]

for 1< R < 3:
_
R
1
dx
3x
= 2
3x
R
1
= 2
3R +2
2.
:
_
3
1
dx
3x
= lim
R3
_
R
1
dx
3x
= 0+2
2= 2
2.
_
4
2
dx
(x +2)
1/3
25.
_

0
dx
1+x
_
R
0
dx
1+x
= ln|1+x|
R
0
= ln|1+R| ln1= ln|1+R|.
_

0
dx
1+x
= lim
R
_
R
0
dx
1+x
= lim
R
ln|1+R| = ;
thus, theintegral doesnot converge.
_
0
xe
x
2
dx
27.
_

0
x dx
(1+x
2
)
2
solution First evaluatetheindeniteintegral, usingthesubstitutionu = x
2
, du = 2x dx; then
_
x dx
(1+x
2
)
2
=
1
2
_
1
(1+u)
2
du =
1
2(u +1)
+C =
1
2(x
2
+1)
+C
Thus, for R > 0,
_
R
0
x dx
(x
2
+1)
2
=
_
1
2(x
2
+1)
_
R
0
=
1
2(R
2
+1)
+
1
2
andthusinthelimit
_

0
x dx
(x
2
+1)
2
= lim
R
_
R
0
x dx
(x
2
+1)
2
=
1
2
lim
R
1
2(R
2
+1)
=
1
2
_
6
3
x dx
x 3
29.
_

0
e
x
cosx dx
solution Firstevaluatetheindeniteintegral usingIntegrationbyParts, withu = e
x
, v
= cosx. Thenu
= e
x
,
v = sinx, and
_
e
x
cosx dx = e
x
sinx
_
sinx(e
x
) dx = e
x
sinx +
_
e
x
sinx dx.
NowuseIntegrationbyPartsagain, withu = e
x
, v
= sinx. Thenu
= e
x
, v = cosx, and
_
e
x
cosx dx = e
x
sinx +
_
e
x
cosx
_
e
x
cosx dx
_
.
_
e
x
cosx dx, wend
_
e
x
cosx dx =
1
2
e
x
(sinx cosx) +C.
Thus,
_
R
0
e
x
cosx dx =
1
2
e
x
(sinx cosx)
R
0
=
sinR cosR
2e
R

sin0cos0
2
=
sinR cosR
2e
R
+
1
2
,
and
_

0
e
x
cosx dx = lim
R
_
sinR cosR
2e
R
+
1
2
_
= 0+
1
2
=
1
2
.
_

1
xe
2x
dx
31.
_
3
0
dx
_
9x
2
solution Thefunction(9x
2
)
1/2
isinniteat x = 3, sowell rst evaluatetheintegral ontheinterval [0, R] for
0< R < 3:
_
R
0
dx
_
9x
2
= sin
1
x
3
R
0
= sin
1
R
3
sin
1
0= sin
1
R
3
.
Thus,
_
3
0
dx
_
9x
2
= lim
R3
sin
1
R
3
= sin
1
1=

2
.
_
1
0
e
x
dx
x
33.
_

1
e
x
dx
x
solution Let u =
x, du =
1
2
x
1/2
dx. Then
_
e
x
dx
x
= 2
_
e
x
_
dx
2
x
_
= 2
_
e
u
du = 2e
u
+C = 2e
x
+C,
and
_

1
e
x
dx
x
= lim
R
_
R
1
e
x
dx
x
= lim
R
2e
R
1
= lim
R
_
2e
R
2e
_
= .
_
/2
0
sec d
35.
_

0
sinx dx
_
R
0
sinx dx = cosx
R
0
= cosR +cos0= 1cosR.
Thus,
_
R
0
sinx dx = lim
R
(1cosR) = 1 lim
R
cosR.
Thislimitdoesnotexist, sincethevalueof cosR oscillatesbetween1and1asR approachesinnity. Hencetheintegral
doesnot converge.
_
/2
0
tanx dx
37.
_
1
0
lnx dx
solution Thefunction lnx is inniteat x = 0, so well rst evaluatetheintegral on [R, 1] for 0 < R < 1. Use
IntegrationbyPartswithu = lnx andv
= 1. Thenu
= 1/x, v = x, andwehave
_
1
R
lnx dx = x lnx
1
R
_
1
R
dx = (x lnx x)
1
R
= (ln11) (RlnR R) = R 1RlnR.
Thus,
_
1
0
lnx dx = lim
R0
+
(R 1RlnR) = 1 lim
R0
+
RlnR.
Tocomputethelimit, rewritethefunctionasaquotient andapplyLHpitalsRule:
_
1
0
lnx dx = 1 lim
R0
+
lnR
1
R
= 1 lim
R0
+
1
R
1
R
2
= 1 lim
R0
+
(R) = 10= 1.
_
2
1
dx
x lnx
39.
_
1
0
lnx
x
2
dx
solution UseIntegrationbyParts, withu = lnx andv
= x
2
. Thenu
= 1/x, v = x
1
, and
_
lnx
x
2
dx =
1
x
lnx +
_
dx
x
2
=
1
x
lnx
1
x
+C.
Thefunctionisinniteat x = 0, sowell rst evaluatetheintegral on[R, 1] for 0< R < 1:
_
1
a
lnx
x
2
dx =
_
1
x
lnx
1
x
_
1
R
=
_
1
1
ln1
1
1
_
1
R
lnR
1
R
_
=
1
R
lnR +
1
R
1.
Thus,
_
1
0
lnx
x
2
dx = lim
R0
+
1
R
lnR +
1
R
1= 1+ lim
R0
+
lnR +1
R
= .
_

1
lnx
x
2
dx
41. Let I =
_

4
dx
(x 2)(x 3)
.
(a) Showthat for R > 4,
_
R
4
dx
(x 2)(x 3)
= ln
R 3
R 2
ln
1
2
(b) Thenshowthat I = ln2.
solution
(a) Thepartial fractiondecompositiontakestheform
1
(x 2)(x 3)
=
A
x 2
+
B
x 3
.
1= A(x 3) +B(x 2).
Settingx = 2thenyieldsA = 1, whilesettingx = 3yieldsB = 1. Thus,
_
dx
(x 2)(x 3)
=
_
dx
x 3
_
dx
x 2
= ln|x 3| ln|x 2| +C = ln
x 3
x 2
+C,
and, for R > 4,
_
R
4
dx
(x 2)(x 3)
= ln
x 3
x 2
R
4
= ln
R 3
R 2
ln
1
2
.
(b) Usingtheresult frompart (a),
I = lim
R
_
ln
R 3
R 2
ln
1
2
_
= ln1ln
1
2
= ln2.
Evaluatetheintegral I =
_

1
dx
x(2x +5)
.
43. EvaluateI =
_
1
0
dx
x(2x +5)
or statethat it diverges.
solution Thepartial fractiondecompositiontakestheform
1
x(2x +5)
=
A
x
+
B
2x +5
.
1= A(2x +5) +Bx.
Settingx = 0thenyieldsA =
1
5
, whilesettingx =
5
2
yieldsB =
2
5
. Thus,
_
dx
x(2x +5)
=
1
5
_
dx
x
2
5
_
dx
2x +5
=
1
5
ln|x|
1
5
ln|2x +5| +C =
1
5
ln
x
2x +5
+C,
and, for 0< R < 1,
_
1
R
dx
x(2x +5)
=
1
5
ln
x
2x +5
1
R
=
1
5
ln
1
7
1
5
ln
R
2R +5
.
Thus,
I = lim
R0+
_
1
5
ln
1
7
1
5
ln
R
2R +5
_
= .
EvaluateI =
_

2
dx
(x +3)(x +1)
2
or statethat it diverges.
InExercises4548, determinewhether thedoublyinniteimproper integral convergesand, if so, evaluateit. Usedeni-
tion(2).
45.
_

x dx
1+x
2
solution Usingthesubstitutionu = x
2
+1, du = 2x dx, weobtain
_
x dx
1+x
2
=
1
2
ln(x
2
+1) +C.
Thus,
_

0
x dx
1+x
2
= lim
R
_
R
0
x dx
1+x
2
= lim
R
1
2
ln(R
2
+1) = ;
_
0
x dx
1+x
2
= lim
R
_
0
R
x dx
1+x
2
= lim
R
1
2
ln(R
2
+1) = ;
It followsthat
_

x dx
1+x
2
diverges.
_

e
|x|
dx
47.
_

xe
x
2
dx
solution First notethat
_
xe
x
2
dx =
1
2
e
x
2
+C.
Thus,
_

0
xe
x
2
dx = lim
R
_
R
0
xe
x
2
dx = lim
R
_
1
2
1
2
e
R
2
_
=
1
2
;
_
0
xe
x
2
dx = lim
R
_
0
R
xe
x
2
dx = lim
R
_
1
2
+
1
2
e
R
2
_
=
1
2
;
and
_

xe
x
2
dx =
1
2
1
2
= 0.
_

dx
(x
2
+1)
3/2
49. DeneJ =
_
1
1
dx
x
1/3
asthesumof thetwoimproper integrals
_
0
1
dx
x
1/3
+
_
1
0
dx
x
1/3
. Showthat J convergesand
that J = 0.
solution Notethatsincex
1/3
isanoddfunction, onemightexpectthisintegral over asymmetricinterval tobezero.
Toprovethis, westart byevaluatingtheindeniteintegral:
_
dx
x
1/3
=
3
2
x
2/3
+C
Then
_
0
1
dx
x
1/3
= lim
R0
_
R
1
dx
x
1/3
= lim
R0
3
2
x
2/3
R
1
= lim
R0
3
2
R
2/3
3
2
=
3
2
_
1
0
dx
x
1/3
= lim
R0
+
_
1
R
dx
x
1/3
= lim
R0
+
3
2
x
2/3
1
R
=
3
2
lim
R0
+
3
2
R
2/3
=
3
2
sothat
J =
_
1
1
dx
x
1/3
=
_
0
1
dx
x
1/3
+
_
1
0
dx
x
1/3
=
3
2
+
3
2
= 0
Determinewhether J =
_
1
1
dx
x
2
(denedasinExercise49) converges.
51. For whichvaluesof a does
_

0
e
ax
dx converge?
_
R
0
e
ax
dx =
1
a
e
ax
R
0
=
1
a
_
e
aR
1
_
.
Thus,
_

0
e
ax
dx = lim
R
1
a
_
e
aR
1
_
.
If a > 0, thene
aR
asR . If a < 0, thene
aR
0asR , and
_

0
e
ax
dx = lim
R
1
a
_
e
aR
1
_
=
1
a
.
Theintegral convergesfor a < 0.
Showthat
_
1
0
dx
x
p
convergesif p < 1anddivergesif p 1.
53. Sketchtheregionunder thegraphof f (x) =
1
1+x
2
for < x < , andshowthat itsareais.
solution Thegraphisshownbelow.
1
0.4
0.2
0.8
0.6
y
x
2 4 2 4
Since(1+x
2
)
1
isanevenfunction, wecanrst computetheareaunder thegraphfor x > 0:
_
R
0
dx
1+x
2
= tan
1
x
R
0
= tan
1
R tan
1
0= tan
1
R.
Thus,
_

0
dx
1+x
2
= lim
R
tan
1
R =

2
.
Bysymmetry, wehave
_

dx
1+x
2
=
_
0
dx
1+x
2
+
_

0
dx
1+x
2
=

2
+

2
= .
Showthat
1
_
x
4
+1
1
x
2
for all x, andusethistoprovethat
_

1
dx
_
x
4
+1
converges.
55. Showthat
_

1
dx
x
3
+4
convergesbycomparingwith
_

1
x
3
dx.
solution Theintegral
_

1
x
3
dx convergesbecause3> 1. Sincex
3
+4 x
3
, it followsthat
1
x
3
+4
1
x
3
.
Therefore, bythecomparisontest,
_

1
dx
x
3
+4
converges.
Showthat
_

2
dx
x
3
4
convergesbycomparingwith
_

2
2x
3
dx.
57. Showthat 0 e
x
2
e
x
for x 1 (Figure10). UsetheComparisonTest to showthat
_
0
e
x
2
dx
converges. Hint: It sufces(why?) tomakethecomparisonfor x 1because
_

0
e
x
2
dx =
_
1
0
e
x
2
dx +
_

1
e
x
2
dx
y
y= e
|x|
y= e
x
2
x
1
1
2 4 2 3 4 1 3
FIGURE 10 Comparisonof y = e
|x|
andy = e
x
2
.
solution For x 1, x
2
x, sox
2
x ande
x
2
e
x
. Now
_

1
e
x
dx converges, so
_

1
e
x
2
dx converges
bythecomparisontest. Finally, becausee
x
2
iscontinuouson[0, 1],
_

0
e
x
2
dx converges.
Weconcludethat our integral convergesbywritingit asasum:
_

0
e
x
2
dx =
_
1
0
e
x
2
dx +
_

1
e
x
2
dx
Provethat
_

e
x
2
dx convergesbycomparingwith
_

e
|x|
dx (Figure10).
59. Showthat
_

1
1sinx
x
2
dx converges.
1sinx
x
2
. Sincef (x)
2
x
2
and
_

1
2x
2
dx = 2, it followsthat
_

1
1sinx
x
2
dx converges
bythecomparisontest.
Let a > 0. UseLHpitalsRuletoprovethat lim
x
x
a
lnx
= . Then:
(a) Showthat x
a
> 2lnx for all x sufcientlylarge.
(b) Showthat e
x
a
< x
2
for all x sufcientlylarge.
(c) Showthat
_

1
e
x
a
dx converges.
InExercises6174, usetheComparisonTest todeterminewhether or not theintegral converges.
61.
_

1
1
_
x
5
+2
dx
solution Since
_
x
5
+2
x
5
= x
5/2
, it followsthat
1
_
x
5
+2
1
x
5/2
.
Theintegral
_

1
dx/x
5/2
convergesbecause
5
2
> 1. Therefore, bythecomparisontest:
_

1
dx
_
x
5
+2
alsoconverges.
_

1
dx
(x
3
+2x +4)
1/2
63.
_

3
dx
x 1
solution Since
x 1, wehave(for x > 1)
1
x 1
.
Theintegral
_

1
dx/
x =
_

1
dx/x
1/2
divergesbecause
1
2
< 1. Sincethefunctionx
1/2
iscontinuous(andtherefore
nite) on[1, 3], wealsoknowthat
_

3
dx/x
1/2
diverges. Therefore, bythecomparisontest,
_

3
dx
x 1
alsodiverges.
_
5
0
dx
x
1/3
+x
3
65.
_

1
e
(x+x
1
)
dx
solution For all x 1,
1
x
> 0sox +
1
x
x. Then
_
x +x
1
_
x and e
(x+x
1
)
e
x
.
Theintegral
_

1
e
x
dx convergesbydirect computation:
_

1
e
x
dx = lim
R
_
R
1
e
x
dx = lim
R
e
x
R
1
= lim
R
e
R
+e
1
= 0+e
1
= e
1
.
Therefore, bythecomparisontest,
_

1
e
(x+x
1
)
alsoconverges.
_
1
0
| sinx|
x
dx
67.
_
1
0
e
x
x
2
dx
solution For 0< x < 1, e
x
> 1, andtherefore
1
x
2
<
e
x
x
2
.
Theintegral
_
1
0
dx/x
2
divergessince2> 1. Therefore, bythecomparisontest,
_
1
0
e
x
x
2
alsodiverges.
_

1
1
x
4
+e
x
dx
69.
_
1
0
1
x
4
+
x
dx
solution For 0< x < 1, x
4
+
x, and
1
x
4
+
x
.
Theintegral
_
1
0
(1/
x) dx converges, sincep =
1
2
< 1. Therefore, bythecomparisontest,
_
1
0
dx
x
4
+
x
alsoconverges.
_

1
lnx
sinhx
dx
71.
_

1
dx
_
x
1/3
+x
3
solution For x 0,
_
x
1/3
+x
3
x
3
= x
3/2
, sothat
1
_
x
1/3
+x
3

1
x
3/2
Theintegral
_

1
x
3/2
dx convergessincep = 3/2> 1. Therefore, bythecomparisontest,
_
1
_
x
1/3
+x
3
dx alsoconverges.
_
1
0
dx
(8x
2
+x
4
)
1/3
73.
_

1
dx
(x +x
2
)
1/3
Hint for Exercise73: Showthat for x 1,
1
(x +x
2
)
1/3

1
2
1/3
x
2/3
solution For x > 1, x < x
2
sothat x +x
2
< 2x
2
; then
_

1
1
(x +x
2
)
1/3
dx
_

1
1
(2x
2
)
1/3
dx =
1
2
1/3
_

1
1
x
2/3
dx
But
_

1
1
x
2/3
dx divergessincep = 2/3< 1. Therefore, bythecomparisontest,
_

1
1
(x +x
2
)
1/3
dx divergesaswell.
_
1
0
dx
xe
x
+x
2
Hint for Exercise74: Showthat for 0 x 1,
1
xe
x
+x
2

1
(e +1)x
75. DeneJ =
_

0
dx
x
1/2
(x +1)
asthesumof thetwoimproper integrals
_
1
0
dx
x
1/2
(x +1)
+
_

1
dx
x
1/2
(x +1)
UsetheComparisonTest toshowthat J converges.
solution For therst integral, notethat for 0 x 1, wehave1 1+ x, sothat x
1/2
(x + 1) x
1/2
. It follows
that
_
1
0
1
x
1/2
(x +1)
dx
_
1
0
1
x
1/2
dx
whichconvergessincep = 1/2 < 1. Thustherst integral convergesby thecomparisontest. For thesecondintegral,
for 1 x, wehavex
1/2
(x +1) = x
3/2
+x
1/2
x
3/2
, sothat
_

1
1
x
1/2
(x +1)
dx =
_

1
1
x
3/2
+x
1/2
dx
_

1
1
x
3/2
dx
whichconvergessincep = 3/2> 1. Thusthesecondintegral convergesaswell bythecomparisontest, andthereforeJ,
whichisthesumof thetwo, converges.
Determinewhether J =
_

0
dx
x
3/2
(x +1)
(denedasinExercise75) converges.
77. Aninvestmentpaysadividendof $250/year continuouslyforever. If theinterestrateis7%, whatisthepresentvalue
of theentireincomestreamgeneratedbytheinvestment?
solution Thepresent valueof theincomestreamafter T yearsis
_
T
0
250e
0.07t
dt =
250e
0.07t
0.07
T
0
=
250
0.07
_
e
0.07T
1
_
=
250
0.07
_
1e
0.07T
_
.
Thereforethepresent valueof theentireincomestreamis
_

0
250e
0.07t
= lim
T
_
T
0
250e
0.07t
= lim
T
250
0.07
_
1e
0.07T
_
=
250
0.07
(10) =
250
0.07
= $3571.43.
Aninvestment isexpectedtoearnprotsat arateof 10,000e
0.01t
dollarsper year forever. Findthepresent value
of theincomestreamif theinterest rateis4%.
79. Computethepresent valueof aninvestment that generatesincomeat arateof 5000t e
0.01t
dollarsper year forever,
assuminganinterest rateof 6%.
solution Thepresent valueof theincomestreamafter T yearsis
_
T
0
_
5000t e
0.01t
_
e
0.06t
dt = 5000
_
T
0
t e
0.05t
dt
Compute the indenite integral using Integration by Parts, with u = t and v
= e
0.05t
. Then u
= 1, v =
(1/0.05)e
0.05t
, and
_
t e
0.05t
dt =
t
0.05
e
0.05t
+
1
0.05
_
e
0.05t
dt = 20t e
0.05t
+
20
0.05
e
0.05t
+C
= e
0.05t
(20t 400) +C.
Thus,
5000
_
T
0
t e
0.05t
dt = 5000e
0.05t
(20t 400)
T
0
= 5000e
0.05T
(20T 400) 5000(400)
= 2,000,0005000e
0.05T
(20T +400).
UseLHpitalsRuletocomputethelimit:
lim
T
_
2,000,000
5000(20T +400)
e
0.05T
_
= 2,000,000 lim
T
5000(20)
0.05e
0.05T
= 2,000,0000= $2,000,000.
Findthevolumeof thesolidobtainedby rotatingtheregionbelowthegraphof y = e
x
about thex-axis for
0 x < .
81. ThesolidS obtainedby rotatingtheregionbelowthegraphof y = x
1
about thex-axisfor 1 x < iscalled
GabrielsHorn(Figure11).
(a) UsetheDisk Method(Section6.3) tocomputethevolumeof S. Notethat thevolumeisniteeventhoughS isan
inniteregion.
(b) It canbeshownthat thesurfaceareaof S is
A = 2
_

1
x
1
_
1+x
4
dx
Showthat A isinnite. If S wereacontainer, youcouldll itsinterior withaniteamount of paint, but youcouldnot
paint itssurfacewithaniteamount of paint.
y= x
1
y
x
FIGURE 11
solution
(a) Thevolumeisgivenby
V =
_

1

_
1
x
_
2
dx.
First computethevolumeover aniteinterval:
_
R
1

_
1
x
_
2
dx =
_
R
1
x
2
dx =
x
1
1
R
1
=
_
1
R
1
1
_
=
_
1
1
R
_
.
Thus,
V = lim
R
_

1
x
2
dx = lim
R
_
1
1
R
_
= .
(b) For x > 1, wehave
1
x
_
1+
1
x
4
=
1
x
_
x
4
+1
x
4
=
_
x
4
+1
x
3

x
4
x
3
=
x
2
x
3
=
1
x
.
Theintegral
_

1
1
x
dx diverges, sincep = 1 1. Therefore, bythecomparisontest,
_

1
1
x
_
1+
1
x
4
dx alsodiverges.
Finally,
A = 2
_

1
1
x
_
1+
1
x
4
dx
diverges.
Computethevolumeof thesolidobtainedbyrotatingtheregionbelowthegraphof y = e
|x|/2
aboutthex-axis
for < x < .
83. Whenacapacitor of capacitanceC ischargedbyasourceof voltageV, thepower expendedat timet is
P(t ) =
V
2
R
(e
t /RC
e
2t /RC
)
whereR istheresistanceinthecircuit. Thetotal energystoredinthecapacitor is
W =
_

0
P(t ) dt
Showthat W =
1
2
CV
2
.
solution Thetotal energycontainedafter thecapacitor isfullychargedis
W =
V
2
R
_

0
_
e
t /RC
e
2t /RC
_
dt.
Theenergyafter aniteamount of time(t = T ) is
V
2
R
_
T
0
_
e
t /RC
e
2t /RC
_
dt =
V
2
R
_
RCe
t /RC
+
RC
2
e
2t /RC
_
T
0
= V
2
C
__
e
T/RC
+
1
2
e
2T/RC
_
_
1+
1
2
__
= CV
2
_
1
2
e
T/RC
+
1
2
e
2T/RC
_
.
Thus,
W = lim
T
CV
2
_
1
2
e
T/RC
+
1
2
e
2T/RC
_
= CV
2
_
1
2
0+0
_
=
1
2
CV
2
.
For whichintegersp does
_
1/2
0
dx
x(lnx)
p
converge?
85. Conservationof Energycanbeusedtoshowthatwhenamassmoscillatesattheendof aspringwithspringconstant
k, theperiodof oscillationis
T = 4
m
_

2E/k
0
dx
_
2E kx
2
whereE isthetotal energyof themass. Showthat thisisanimproper integral withvalueT = 2
m/k.
solution Theintegrandisinniteat theupper limit of integration, x =
2E/k, sotheintegral isimproper. Now, let

T (R) = 4
m
_
R
0
dx
_
2E kx
2
= 4
m
1
2E
_
R
0
dx
_
1(
k
2E
)x
2
= 4
_
m
2E
_
2E
k
sin
1
__
k
2E
R
_
= 4
_
m/k sin
1
__
k
2E
R
_
.
Therefore
T = lim
R
2E/k
T (R) = 4
_
m
k
sin
1
(1) = 2
_
m
k
.
In Exercises 8689, theLaplacetransformof a function f (x) is thefunction Lf (s) of thevariables dened by the
improper integral (if it converges):
Lf (s) =
_

0
f (x)e
sx
dx
Laplacetransformsarewidelyusedinphysicsandengineering.
Showthat if f (x) = C, whereC isaconstant, thenLf (s) = C/s for s > 0.
87. Showthat if f (x) = sinx, thenLf (s) =

s
2
+
2
.
solution If f (x) = sinx, thentheLaplacetransformof f (x) is
Lf (s) =
_

0
e
sx
sinx dx
First evaluatetheindeniteintegral usingIntegrationby Parts, withu = sinx andv
= e
sx
. Thenu
= cosx,
v =
1
s
e
sx
, and
_
e
sx
sinx dx =
1
s
e
sx
sinx +

s
_
e
sx
cosx dx.
UseIntegrationbyPartsagain, withu = cosx, v
= e
sx
. Thenu
= sinx, v =
1
s
e
sx
, and
_
e
sx
cosx dx =
1
s
e
sx
cosx

s
_
e
sx
sinx dx.
Substitutingthisintotherst equationandsolvingfor
_
e
sx
sinx dx, weget
_
e
sx
sinx dx =
1
s
e
sx
sinx

s
2
e
sx
cosx

2
s
2
_
e
sx
sinx dx
_
e
sx
sinx dx =
e
sx
_
1
s
sinx +

s
2
cosx
_
_
1+

2
s
2
_ =
e
sx
(s sinx + cosx)
s
2
+
2
Thus,
_
R
0
e
sx
sinx dx =
1
s
2
+
2
_
s sinR + cosR
e
sR

0+
1
_
=
1
s
2
+
2
_

s sinR + cosR
e
sR
_
.
Finallywetakethelimit, notingthefact that, for all valuesof R, |s sinR + cosR| s +||
Lf (s) = lim
R
1
s
2
+
2
_

s sinR + cosR
e
sR
_
=
1
s
2
+
2
( 0) =

s
2
+
2
.
ComputeLf (s), wheref (x) = e
x
ands > .
89. ComputeLf (s), wheref (x) = cosx ands > 0.
solution If f (x) = cosx, thentheLaplacetransformof f (x) is
Lf (x) =
_

0
e
sx
cosx dx
First evaluatetheindeniteintegral usingIntegrationby Parts, withu = cosx andv
e
sx
. Thenu
= sinx,
v =
1
s
e
sx
, and
_
e
sx
cosx dx =
1
s
e
sx
cosx

s
_
e
sx
sinx dx.
UseIntegrationbyPartsagain, withu = sinx dx andv
= e
sx
. Thenu
= cosx, v =
1
s
e
sx
, and
_
e
sx
sinx dx =
1
s
e
sx
sinx +

s
_
e
sx
cosx dx.
Substitutingthisintotherst equationandsolvingfor
_
e
sx
cosx dx, weget
_
e
sx
cosx dx =
1
s
e
sx
cosx

s
_
1
s
e
sx
sinx +

s
_
e
sx
cos dx
_
=
1
s
e
sx
cosx +

s
2
e
sx
sinx

2
s
2
_
e
sx
cosx dx
_
e
sx
cosx dx =
e
sx
_
s
2
sinx
1
s
cosx
_
1+

2
s
2
=
e
sx
( sinx s cosx)
s
2
+
2
Thus,
_
R
0
e
sx
cosx dx =
1
s
2
+
2
_
sinR s cosR
e
sR

0s
1
_
.
Finallywetakethelimit, notingthefact that, for all valuesof R, | sinR s cosR| || +s
Lf (s) = lim
R
1
s
2
+
2
_
s +
sinR s cosR
e
sR
_
=
1
s
2
+
2
(s +0) =
s
s
2
+
2
.
Whenaradioactivesubstancedecays, thefractionof atomspresent at timet isf (t ) = e
kt
, wherek > 0is
thedecay constant. It canbeshownthat theaveragelifeof anatom(until it decays) is A =
_
0
tf
(t ) dt . Use
Integrationby Parts to showthat A =
_
0
f (t ) dt andcomputeA. What is theaveragedecay timeof radon-222,
whosehalf-lifeis3.825days?
91. Let J
n
=
_

0
x
n
e
x
dx, wheren 1isaninteger and > 0. Provethat
J
n
=
n
J
n1
andJ
0
= 1/. UsethistocomputeJ
4
. Showthat J
n
= n!/
n+1
.
solution UsingIntegrationbyParts, withu = x
n
andv
= e
x
, weget u
= nx
n1
, v =
1
e
x
, and
_
x
n
e
x
dx =
1
x
n
e
x
+
n
_
x
n1
e
x
dx.
Thus,
J
n
=
_

0
x
n
e
x
dx = lim
R
_
x
n
e
x
_
R
0
+
n
_

0
x
n1
e
x
dx = lim
R
R
n
e
R
+0+
n
J
n1
.
UseLHpitalsRulerepeatedlytocomputethelimit:
lim
R
R
n
e
R
= lim
R
nR
n1
2
e
R
= lim
R
n(n 1)R
n2
3
e
R
= = lim
R
n(n 1)(n 2) (3)(2)(1)
n+1
e
R
= 0.
Finally,
J
n
= 0+
n
J
n1
=
n
J
n1
.
J
0
canbecomputeddirectly:
J
0
=
_

0
e
x
dx = lim
R
_
R
0
e
x
dx = lim
R
e
x
R
0
= lim
R
_
e
R
1
_
=
1
(01) =
1
.
Withthisstartingpoint, wecanworkuptoJ
4
:
J
1
=
1
J
0
=
1
_
1
_
=
1
2
;
J
2
=
2
J
1
=
2
_
1
2
_
=
2
3
=
2!
2+1
;
J
3
=
3
J
2
=
3
_
2
3
_
=
6
4
=
3!
3+1
;
J
4
=
4
J
3
=
4
_
6
4
_
=
24
5
=
4!
4+1
.
Wecanuseinductiontoprovetheformulafor J
n
. If
J
n1
=
(n 1)!
n
,
thenwehave
J
n
=
n
J
n1
=
n
(n 1)!
n
=
n!
n+1
.
Let a > 0andn > 1. Denef (x) =
x
n
e
ax
1
for x = 0andf (0) = 0.
(a) UseLHpitalsRuletoshowthat f (x) iscontinuousat x = 0.
(b) Showthat
_
0
f (x) dx converges. Hint: Showthat f (x) 2x
n
e
ax
if x islargeenough. ThenusetheCom-
parisonTest andExercise91.
93. AccordingtoPlancksRadiationLaw, theamount of electromagnetic energy withfrequency between
and + that isradiatedbyaso-calledblackbodyat temperatureT isproportional toF() , where
F() =
_
8h
c
3
_

3
e
h/kT
1
wherec, h, k arephysical constants. UseExercise92toshowthat thetotal radiatedenergy
E =
_

0
F() d
is nite. To derive his law, Planck introduced the quantumhypothesis in 1900, which marked the birth of quantum
mechanics.
solution Thetotal radiatedenergyE isgivenby
E =
_

0
F() d =
8h
c
3
_

0
3
e
h/kT
1
d.
Let = h/kT . Then
E =
8h
c
3
_

0
3
e
1
d.
Because > 0and8h/c
3
isaconstant, weknowE isnitebyExercise92.
Let I =
_
1
0
x
p
lnx dx.
(a) Showthat I divergesfor p = 1.
(b) Showthat if p = 1, then
_
x
p
lnx dx =
x
p+1
p +1
_
lnx
1
p +1
_
+C
(c) UseLHpitalsRuletoshowthat I convergesif p > 1anddivergesif p < 1.
95. Let
F(x) =
_
x
2
dt
lnt
and G(x) =
x
lnx
Verifythat LHpitalsRuleappliestothelimit L = lim
x
F(x)
G(x)
andevaluateL.
solution Becauselnt < t for t > 2, wehave
1
lnt
>
1
t
for t > 2, andso
F(x) =
_
x
2
dt
lnt
>
_
x
2
dt
t
= lnx ln2
Thus, F(x) asx . Moreover, byLHpitalsRule
lim
x
G(x) = lim
x
1
1/x
= lim
x
x = .
Thus, lim
x
F(x)
G(x)
isof theform/, andLHpitalsRuleapplies. Finally,
L = lim
x
F(x)
G(x)
= lim
x
1
lnx
lnx1
(lnx)
2
= lim
x
lnx
lnx 1
= lim
x
1
1(1/ lnx)
= 1.
InExercises9698, animproper integral I =
_
a
f (x) dx iscalledabsolutelyconvergent if
_
a
|f (x)| dx converges.
It canbeshownthat if I isabsolutelyconvergent, thenit isconvergent.
Showthat
_

1
sinx
x
2
dx isabsolutelyconvergent.
97. Showthat
_

1
e
x
2
cosx dx isabsolutelyconvergent.
solution Bytheresultof Exercise57,weknowthat
_

0
e
x
2
dx isconvergent.Then
_

1
e
x
2
dx isalsoconvergent.
Because| cosx| 1for all x, wehave
e
x
2
cosx
= | cosx|
e
x
2
e
x
2
= e
x
2
.
Therefore, bythecomparisontest, wehave
_

1
e
x
2
cosx
dx alsoconverges.
Since
_

1
e
x
2
cosx dx convergesabsolutely, it itself converges.
Let f (x) = sinx/x andI =
_
0
f (x) dx. Wedenef (0) = 1. Thenf (x) iscontinuousandI isnot improper
at x = 0.
(a) Showthat
_
R
1
sinx
x
dx =
cosx
x
R
1

_
R
1
cosx
x
2
dx
(b) Showthat
_
1
(cosx/x
2
) dx converges. Concludethat thelimit as R of theintegral in(a) exists andis
nite.
(c) Showthat I converges.
ItisknownthatI =

2
. However, I isnot absolutelyconvergent. Theconvergencedependsoncancellation, asshown
inFigure12.
99. Thegammafunction, whichplaysanimportant roleinadvancedapplications, isdenedfor n 1by
(n) =
_

0
t
n1
e
t
dt
(a) Showthat theintegral dening (n) converges for n 1 (it actually converges for all n > 0). Hint: Showthat
t
n1
e
t
< t
2
for t sufcientlylarge.
(b) Showthat (n +1) = n(n) usingIntegrationbyParts.
(c) Showthat (n + 1) = n! if n 1is aninteger. Hint: Use(a) repeatedly. Thus, (n) provides away of dening
n-factorial whenn isnot aninteger.
solution
(a) Byrepeateduseof LHpitalsRule, wecancomputethefollowinglimit:
lim
t
e
t
t
n+1
= lim
t
e
t
(n +1)t
n
= = lim
t
e
t
(n +1)!
= .
Thisimpliesthat, for t sufcientlylarge, wehave
e
t
t
n+1
;
therefore
e
t
t
n1

t
n+1
t
n1
= t
2
or t
n1
e
t
t
2
.
Theintegral
_

1
t
2
dt convergesbecausep = 2> 1. Therefore, bythecomparisontest,
_

M
t
n1
e
t
dt alsoconverges,
whereM isthevalueabovewhichtheabovecomparisonshold. Finally, becausethefunctiont
n1
e
t
iscontinuousfor
all t , weknowthat
(n) =
_

0
t
n1
e
t
dt converges for all n 1.
(b) UsingIntegrationbyParts, withu = t
n
andv
e
t
, wehaveu
= nt
n1
, v = e
t
, and
(n +1) =
_

0
t
n
e
t
dt = t
n
e
t
0
+n
_

0
t
n1
e
t
dt
= lim
R
_
R
n
e
R
0
_
+n(n) = 0+n(n) = n(n).
Here, wevecomputedthelimit asinpart (a) withrepeateduseof LHpitalsRule.
(c) Bytheresult of part (b), wehave
(n +1) = n(n) = n(n 1)(n 1) = n(n 1)(n 2)(n 2) = = n! (1).
If n = 1, then
(1) =
_

0
e
t
dt = lim
R
e
t
R
0
= lim
R
_
1e
R
_
= 1.
Thus
(n +1) = n! (1) = n!
Usetheresultsof Exercise99toshowthat theLaplacetransform(seeExercises8689above) of x
n
is
n!
s
n+1
.
8.7 Probability and Integration
1. Thefunctionp(x) = cosx satises
_

/2
p(x) dx = 1. Isp(x) aprobabilitydensityfunctionon[/2, ]?
solution Sincep(x) = cosx < 0for somepointsin(/2, ), p(x) isnot aprobabilitydensityfunction.
2. EstimateP(2 X 2.1) assumingthat theprobabilitydensityfunctionof X satisesp(2) = 0.2.
solution P(2 X 2.1) p(2) (2.12) = 0.02.
3. Whichexponential probabilitydensityhasmean =
1
4
?
solution
1
1/4
e
x/(1/4)
= 4e
4x
.
S E C T I ON 8.7 Probability and Integration 521
Exercises
InExercises16, ndaconstantC suchthatp(x) isaprobabilitydensityfunctiononthegiveninterval, andcomputethe
probabilityindicated.
1. p(x) =
C
(x +1)
3
on[0, ); P(0 X 1)
solution Computetheindeniteintegral usingthesubstitutionu = x +1, du = dx:
_
p(x) dx =
_
C
(x +1)
3
dx =
1
2
C(x +1)
2
+K
For p(x) tobeaprobabilitydensityfunction, wemust have
1=
_

0
p(x) dx =
1
2
C lim
R
(x +1)
2
R
0
=
1
2
C
1
2
C lim
R
(R +1)
2
=
1
2
C
sothat C = 2, andp(x) =
2
(x+1)
3
. Thenusingtheindeniteintegral above,
P(0 X 1) =
_
1
0
2
(x +1)
3
=
1
2
2 (x +1)
2
1
0
=
1
4
+1=
3
4
p(x) = Cx(4x) on[0, 4]; P(3 X 4) 3. p(x) =
C
_
1x
2
on(1, 1); P
_
1
2
X
1
2
_
solution Computetheindeniteintegral:
_
p(x) dx = C
_
1
_
1x
2
dx = C sin
1
x +K
validfor 1< x < 1. For p(x) tobeaprobabilitydensityfunction, wemust have
1=
_
1
1
p(x) dx =
_
0
1
p(x) dx +
_
1
0
p(x) dx = C
_
lim
R1
+
sin
1
x
0
R
+ lim
R1
sin
1
x
R
0
_
= C
_
sin
1
(0) lim
R1
+
sin
1
(R) + lim
R1
sin
1
R sin
1
(0)
_
= C
_
sin
1
(1) +sin
1
(1)
_
= C
sothat C =
1
andp(x) =
1
1x
2
P
_
1
2
X
1
2
_
=
_
1/2
1/2
p(x) dx =
1
sin
1
x
1/2
1/2
=
1
6

6
_
=
1
3
p(x) =
Ce
x
1+e
2x
on(, ); P(X 4)
5. p(x) = C
_
1x
2
on(1, 1); P
_
1
2
X 1
_
solution Computetheindeniteintegral usingthesubstitutionx = sinu, sothat dx = cosu du:
_
p(x) dx = C
_
_
1x
2
dx = C
_ _
1sin
2
u cosu du = C
_
cos
2
u du
= C
_
1
2
u +
1
2
cosu sinu
_
+K
Sincex = sinu, weconstruct thefollowingright triangle:
1 x
2
x
1
andweseethat cosu =
_
1x
2
, sothat
_
p(x) dx =
1
2
C
_
sin
1
x +x
_
1x
2
_
+K
For p(x) tobeaprobabilitydensityfunction, wemust have
1=
_
1
1
p(x) dx =
1
2
C
_
sin
1
x +x
_
1x
2
_
1
1
=
1
2
C(sin
1
1sin
1
(1)) =

2
C
sothat C =
2
andp(x) =
2
_
1x
2
P
_
1
2
X 1
_
=
_
1
1/2
2
_
1x
2
dx =
1
_
sin
1
x +x
_
1x
2
_
1
1/2
=
1
_
sin
1
1+0sin
1
_
1
2
_
1
2
_
1
1
4
_
=
1
2

6
+
3
4
_
=
2
3
+
3
4
0.804
p(x) = Ce
x
e
e
x
on(, ); P(4 X 4) This function, calledtheGumbel density, is usedto
model extremeeventssuchasoodsandearthquakes.
7. Verifythat p(x) = 3x
4
isaprobabilitydensityfunctionon[1, ) andcalculateitsmeanvalue.
solution Wehave
_

1
3x
4
dx = lim
R
_
x
3
_
R
1
= lim
R
_
1
R
3
_
+1= 1
sothat p(x) isaprobabilitydensityfunctionon[1, ). Itsmeanvalueis
_

1
x 3x
4
dx =
_

1
3x
3
dx =
3
2
x
2
R
1
= lim
R
_
3
2R
2
_
+
3
2
=
3
2
Showthat thedensityfunctionp(x) =
2
(x
2
+1)
on[0, ) hasinnitemean.
9. Verifythat p(t ) =
1
50
e
t /50
satisesthecondition
_
0
p(t ) dt = 1.
solution Usethesubstitutionu =
t
50
, sothat du =
1
50
dt . Then
_

0
p(t ) dt =
_

0
1
50
e
t /50
dt =
_

0
e
u
du = lim
R
(e
u
)
R
0
= lim
R
1e
R
= 1
Verifythatfor all r > 0, theexponential densityfunctionp(t ) =
1
r
e
t /r
satisesthecondition
_
0
p(t ) dt = 1.
11. ThelifeX(inhours)of abatteryinconstantuseisarandomvariablewithexponential density.Whatistheprobability
that thebatterywill last morethan12hoursif theaveragelifeis8hours?
solution If theaveragelifeis8hours, thenthemeanof theexponential distributionis8, sothat thedistributionis
p(x) =
1
8
e
x/8
Theprobabilitythatthebatterywill lastmorethan12hoursisgivenby(usingthesubstitutionu = x/8,sothatdu = 1/8dx
andx = 12correspondstou = 3/2)
P(X 12) =
_

12
p(x) dx =
_

12
1
8
e
x/8
dx =
_

3/2
e
u
du = lim
R
(e
u
)
R
3/2
= e
3/2
lim
R
e
R
= e
3/2
0.223
Thetimebetweenincomingphonecallsat acall center isarandomvariablewithexponential density. Thereisa
50%probabilityof waiting20secondsor morebetweencalls. What istheaveragetimebetweencalls?
13. Thedistancer betweentheelectronandthenucleus inahydrogenatom(inits lowest energy state) is arandom
variablewithprobabilitydensityp(r) = 4a
3
0
r
2
e
2r/a
0
for r 0, wherea
0
istheBohr radius(Figure7). Calculatethe
probabilityP that theelectroniswithinoneBohr radiusof thenucleus. Thevalueof a
0
isapproximately5.2910
11
m, but thisvalueisnot neededtocomputeP.
a
0
2a
0
3a
0
4a
0
p(r)
r
0.4
FIGURE 7 Probabilitydensityfunctionp(r) = 4a
3
0
r
2
e
2r/a
0
.
S E C T I ON 8.7 Probability and Integration 523
solution TheprobabilityP istheareaof theshadedregioninFigure7. Tocalculatep, usethesubstitutionu = 2r/a
0
:
P =
_
a
0
0
p(r) dr =
4
a
3
0
_
a
0
0
r
2
e
2r/a
0
dr =
_
4
a
3
0
__
a
3
0
8
_
_
2
0
u
2
e
u
du
Theconstant infront simpliesto
1
2
andtheformulainthemargingivesus
P =
1
2
_
2
0
u
2
e
u
du =
1
2
_
(u
2
+2u +2)e
u
_
2
0
=
1
2
_
210e
2
_
0.32
Thus, theelectronwithinadistancea
0
of thenucleuswithprobability0.32.
Showthat thedistancer betweentheelectronandthenucleusinExercise13hasmean = 3a
0
/2.
InExercises1521, F(z) denotesthecumulativenormal distributionfunction. Refer toacalculator, computer algebra
system, or onlineresourcetoobtainvaluesof F(z).
15. Expresstheareaof regionA inFigure8intermsof F(z) andcomputeitsvalue.
55 100 120 165
x
y
A
B
FIGURE 8 Normal densityfunctionwith = 120and = 30.
solution Theareaof regionA isP(55 X 100). ByTheorem1, wehave
P(55 X 100) = F
_
100120
30
_
F
_
55120
30
_
= F
_
2
3
_
F
_
13
6
_
0.237
Showthat theareaof regionB inFigure8isequal to1F(1.5) andcomputeitsvalue. Verifynumericallythat
thisareaisalsoequal toF(1.5) andexplainwhygraphically.
17. AssumeX hasastandardnormal distribution( = 0, = 1). Find:
(a) P(X 1.2) (b) P(X 0.4)
solution
(a) P(X 1.2) = F(1.2) 0.8849
(b) P(X 0.4) = 1P(X 0.4) = 1F(0.4) 10.3446 0.6554
Evaluatenumerically:
1
3
2
_

14.5
e
(z10)
2
/18
dz.
19. Useagraphtoshowthat F(z) = 1 F(z) for all z. Thenshowthat if p(x) isanormal densityfunction
withmean andstandarddeviation, thenfor all r 0,
P( r X +r) = 2F(r) 1
solution Consider thegraphof thestandardnormal densityfunctioninFigure5. Thisgraphissymmetricaroundthe
y-axis, sothat theareaunder thecurvefromz to, whichis1 F(z), isequal totheareaunder thecurvefrom
toz, whichisF(z). Thus1 F(z) = F(z). Now, if p(x) isanormal density functionwithmean andstandard
deviation, thenfor r 0(sothat therange r X +r isnonempty),
P( r X +r) = F
_
+r
_
F
_
r
_
= F(r) F(r) = F(r) (1F(r)) = 2F(r) 1
TheaverageSeptember rainfall inErie, Pennsylvania, isarandomvariableX withmean = 102mm. Assume
that theamount of rainfall isnormallydistributedwithstandarddeviation = 48.
(a) ExpressP(128 X 150) intermsof F(z) andcomputeitsvaluenumerically.
(b) LetP betheprobabilitythatSeptemberrainfall will beatleast120mm. ExpressP asanintegral of anappropriate
densityfunctionandcomputeitsvaluenumerically.
21. A bottlingcompanyproducesbottlesof fruitjuicethatarelled, onaverage, with32ouncesof juice. Duetorandom
uctuationsinthemachinery, theactual volumeof juiceisnormally distributedwithastandarddeviationof 0.4ounce.
LetP betheprobabilityof abottlehavinglessthan31ounces. ExpressP asanintegral of anappropriatedensityfunction
andcomputeitsvaluenumerically.
solution Theassociatedcumulativedistributionfunctionis
f (z) =
1
0.4
2
_
z
e
(x32)
2
/(20.4
2
)
dx
Tocomputethevaluenumerically, weusethestandardnormal cumulativedistributionfunctionF(z):
P(X 31) = F
_
3132
0.4
_
= F
_
5
2
_
=
1
2
_
5/2
e
x
2
/2
dx 0.0062
AccordingtoMaxwellsDistributionLaw, inagasof molecular massm, thespeedv of amoleculeinagasat
temperatureT (kelvins) isarandomvariablewithdensity
p(v) = 4
_
m
2kT
_
3/2
v
2
e
mv
2
/(2kT )
(v 0)
wherek is Boltzmanns constant. Showthat theaveragemolecular speed is equal to (8kT /m)
1/2
. Theaverage
speedof oxygenmoleculesat roomtemperatureisaround450m/s.
InExercises2326, calculate and, where isthestandarddeviation, denedby
2
=
_

(x )
2
p(x) dx
Thesmaller thevalueof , themoretightlyclusteredarethevaluesof therandomvariableX about themean.
23. p(x) =
5
2x
7/2
on[1, )
solution Themeanis
_

1
xp(x) dx =
_

1
5
2
x
5/2
dx =
5
3
x
3/2
1
=
5
3
and
2
=
_

1
(x )
2
p(x) dx =
_

1
(x
2
2x +
2
)
5
2
x
7/2
dx
=
5
2
_

1
x
3/2
2x
5/2
+
2
x
7/2
dx =
5
2
_
2x
1/2
+
4
3
x
3/2
2
5
2
x
5/2
_
1
=
5
2
_
2
4
3
+
2
5
2
_
=
5
2
_
2
4
3
5
3
+
2
5
25
9
_
=
20
9
p(x) =
1
_
1x
2
on(1, 1)
25. p(x) =
1
3
e
x/3
on[0, )
solution Thisisanexponential densityfunctionwithmean = 3. Thestandarddeviationis
2
=
1
3
_

0
(x 3)
2
e
x/3
dx =
1
3
_

0
_
x
2
e
x/3
6xe
x/3
+9e
x/3
_
dx
=
1
3
_

0
x
2
e
x/3
dx 2
_

0
xe
x/3
dx +3
_

0
e
x/3
dx
Wetacklethethirdintegral rst:
_

0
e
x/3
dx = 3e
x/3
0
= 3
For thesecondintegral, useintegrationbypartswithu = x, v
= e
x/3
sothat u
= 1andv = 3e
x/3
. Then
_

0
xe
x/3
dx = 3xe
x/3
0
+3
_

0
e
x/3
dx = 0+3 3= 9
Finally, therstintegral issolvedusingintegrationbypartswithu = x
2
, v
= e
x/3
sothatu
= 2x andv = 3e
x/3
;
then
_

0
x
2
e
x/3
dx = 3x
2
e
x/3
0
+6
_

0
xe
x/3
dx = 0+6 9= 54
and, nally,
2
=
1
3
_

0
x
2
e
x/3
dx 2
_

0
xe
x/3
dx +3
_

0
e
x/3
dx
=
1
3
542 9+3 3= 9
p(x) =
1
r
e
x/r
on[0, ), wherer > 0
S E C T I ON 8.8 Numerical Integration 525
27. Thetimeto decay of anatominaradioactivesubstanceis arandomvariableX. Thelawof radioactive
decay states that if N atoms arepresent at timet = 0, thenNf (t ) atoms will bepresent at timet , wheref (t ) = e
kt
(k > 0isthedecayconstant). Explainthefollowingstatements:
(a) Thefractionof atomsthat decayinasmall timeinterval [t, t +t ] isapproximatelyf
(t )t .
(b) Theprobabilitydensityfunctionof X isf
(t ).
(c) Theaveragetimetodecayis1/k.
solution
(a) Thenumber of atomsthat decayintheinterval [t, t + t ] isjust f (t ) f (t + t ); thestatement simplysaysthat
f (t ) f (t +t ) f
(t )t , whichisthesameassayingthat
f
(t )
f (t ) f (t +t )
t
=
f (t +t ) f (t )
t
whichistruebythedenitionof thederivative. Intuitively, sincef
(t ) istheinstantaneousrateof decay, wewouldexpect

that over ashort interval, thenumber of atomsdecayingisproportional tobothf
(t ) andthesizeof theinterval.
(b) The probability density function is dened by the property in (a): the probability that X lies in a small interval
[t, t +t ] isapproximatelyp(t )t , sothat p(t ) = f
(t ).
(c) Theaveragetimetodecayisthemeanof thedistribution, whichis
=
_

0
t (f
(t )) dt =
_

0
tf
(t ) dt
Wecomputethisintegral usingintegrationbyparts, withu = t , v
= f
(t ). Thenu
= 1, v = f (t ), and
=
_

0
tf
(t ) dt = tf (t )
0
+
_

0
f (t ) dt.
Sincef (t ) = e
kt
, wehave
tf (t )
0
= lim
R
t e
kt
R
0
= lim
R
Re
Rt
+0= lim
R
R
e
Rt
= lim
R
1
Re
Rt
= 0.
HereweusedLHpitalsRuletocomputethelimit. Thus
=
_

0
f (t ) dt =
_

0
e
kt
dt.
Now,
_
R
0
e
kt
dt =
1
k
e
kt
R
0
=
1
k
_
e
kR
1
_
=
1
k
_
1e
kR
_
,
so
= lim
R
1
k
_
1e
kR
_
=
1
k
(10) =
1
k
.
Becausek hasunitsof (time)
1
, doesinfact havetheappropriateunitsof time.
Thehalf-lifeof radon-222, is3.825days. UseExercise27tocompute:
(a) Theaveragetimetodecayof aradon-222atom.
(b) Theprobabilitythat agivenatomwill decayinthenext 24hours.
8.8 Numerical Integration
1. What areT
1
andT
2
for afunctionon[0, 2] suchthat f (0) = 3, f (1) = 4, andf (2) = 3?
solution Usingthegivenfunctionvalues
T
1
=
1
2
(2)(3+3) = 6 and T
2
=
1
2
(1)(3+8+3) = 7.
2. For whichgraphinFigure16will T
N
overestimatetheintegral?What about M
N
?
x
y
y= f(x)
x
y
y= g(x)
FIGURE 16
solution T
N
overestimatesthevalueof theintegral whentheintegrandisconcaveup; thus, T
N
will overestimatethe
integral of y = g(x). Ontheother hand, M
N
overestimatesthevalueof theintegral whentheintegrandisconcavedown;
thus, M
N
will overestimatetheintegral of y = f (x).
3. Howlargeistheerror whentheTrapezoidal Ruleisappliedtoalinear function? Explaingraphically.
solution TheTrapezoidal Ruleintegrateslinear functionsexactly, sotheerror will bezero.
4. What isthemaximumpossibleerror if T
4
isusedtoapproximate
_
3
0
f (x) dx
where|f
(x)| 2for all x.

solution Themaximumpossibleerror inT
4
is
max|f
(x)|
(b a)
3
12n
2

2(30)
3
12(4)
2
=
9
32
.
5. What arethetwographical interpretationsof theMidpoint Rule?
solution Thetwographical interpretationsof theMidpoint Rulearethesumof theareasof themidpoint rectangles
andthesumof theareasof thetangential trapezoids.
Exercises
InExercises112, calculateT
N
andM
N
for thevalueof N indicated.
1.
_
2
0
x
2
dx, N = 4
2
. Wedivide[0, 2] into4subintervalsof width
x =
20
4
=
1
2
withendpoints0, 0.5, 1, 1.5, 2, andmidpoints0.25, 0.75, 1.25, 1.75. Withthisdata, weget
T
4
=
1
2
1
2
_
0
2
+2(0.5)
2
+2(1)
2
+2(1.5)
2
+2
2
_
= 2.75; and
M
4
=
1
2
_
0.25
2
+0.75
2
+1.25
2
+1.75
2
_
= 2.625.
_
4
0
x dx, N = 4
3.
_
4
1
x
3
dx, N = 6
3
x =
41
6
=
1
2
withendpoints1, 1.5, 2, 2.5, 3, 3.5, 4, andmidpoints1.25, 1.75, 2.25, 2.75, 3.25, 3.75. Withthisdata, weget
T
6
=
1
2
_
1
2
_
_
1
3
+2(1.5)
3
+2(2)
3
+2(2.5)
3
+2(3)
3
+2(3.5)
3
+4
3
_
= 64.6875; and
M
6
=
1
2
_
1.25
3
+1.75
3
+2.25
3
+2.75
3
+3.25
3
+3.75
3
_
= 63.28125.
_
2
1
_
x
4
+1dx, N = 5
5.
_
4
1
dx
x
, N = 6
solution Let f (x) = 1/x. Wedivide[1, 4] into6subintervalsof width
x =
41
6
=
1
2
withendpoints1, 1.5, 2, 2.5, 3, 3.5, 4, andmidpoints1.25, 1.75, 2.25, 2.75, 3.25, 3.75. Withthisdata, weget
T
6
=
1
2
_
1
2
__
1
1
+
2
1.5
+
2
2
+
2
2.5
+
2
3
+
2
3.5
+
1
4
_
1.40536; and
M
6
=
1
2
_
1
1.25
+
1
1.75
+
1
2.25
+
1
2.75
+
1
3.25
+
1
3.75
_
1.37693.
_
1
2
dx
x
, N = 5
7.
_
/2
0
sinx dx, N = 6
sinx. Wedivide[0, /2] into6subintervalsof width

x =
2
0
6
=

12
withendpoints
0,
12
,
2
12
, . . . ,
6
12
=

2
,
andmidpoints
24
,
3
24
, . . . ,
11
24
.
Withthisdata, weget
T
6
=
1
2
_

12
_ _
_
sin(0) +2
_
sin(/12) + +
_
sin(6/12)
_
1.17029; and
M
6
=

12
_
_
sin(/24) +
_
sin(3/24) + +
_
sin(11/24)
_
1.20630.
_
/4
0
secx dx, N = 6
9.
_
2
1
lnx dx, N = 5
solution Let f (x) = lnx. Wedivide[1, 2] into5subintervalsof width
x =
21
5
=
1
5
= 0.2
withendpoints1, 1.2, 1.4, 1.6, 1.8, 2, andmidpoints1.1, 1.3, 1.5, 1.7, 1.9. Withthisdata, weget
T
5
=
1
2
_
1
5
_
_
ln1+2ln1.2+2ln1.4+2ln1.6+2ln1.8+ln2
_
0.384632; and
M
5
=
1
5
_
ln1.1+ln1.3+ln1.5+ln1.7+ln1.9
_
0.387124.
_
3
2
dx
lnx
, N = 5
11.
_
1
0
e
x
2
dx, N = 5
x
2
x =
10
5
=
1
5
= 0.2
withendpoints
0,
1
5
,
2
5
,
3
5
,
4
5
, 1
andmidpoints
1
10
,
3
10
,
5
10
,
7
10
,
9
10
.
Withthisdata, weget
T
5
=
1
2
_
1
5
_
_
e
0
2
+2e
(1/5)
2
+2e
(2/5)
2
+2e
(3/5)
2
+2e
(4/5)
2
+e
1
2
_
0.74437; and
M
5
=
1
5
_
e
(1/10)
2
+e
(3/10)
2
+e
(5/10)
2
+e
(7/10)
2
+e
(9/10)
2
_
0.74805.
_
1
2
e
x
2
dx, N = 6
InExercises1322, calculateS
N
givenbySimpsonsRulefor thevalueof N indicated.
13.
_
4
0
x dx, N = 4
x. Wedivide[0, 4] into4subintervalsof width

x =
40
4
= 1
withendpoints0, 1, 2, 3, 4. Withthisdata, weget
S
4
=
1
3
(1)
_
0+4
1+2
2+4
3+
4
_
5.25221.
_
5
3
(9x
2
) dx, N = 4
15.
_
3
0
dx
x
4
+1
, N = 6
solution Let f (x) = 1/(x
4
+1). Wedivide[0, 3] into6subintervalsof length
x =
30
6
=
1
2
= 0.5
withendpoints0, 0.5, 1, 1.5, 2, 2.5, 3. Withthisdata, weget
S
6
=
1
3
_
1
2
__
1
0
4
+1
+
4
0.5
4
+1
+
2
1
4
+1
+
4
1.5
4
+1
+
2
2
4
+1
+
4
2.5
4
+1
+
1
3
4
+1
_
1.10903.
_
1
0
cos(x
2
) dx, N = 6
17.
_
1
0
e
x
2
dx, N = 4
x
2
. Wedivide[0, 1] into4subintervalsof length
x =
10
4
=
1
4
withendpoints0,
1
4
,
2
4
,
3
4
,
4
4
= 1. Withthisdata, weget
S
4
=
1
3
_
1
4
_
_
e
0
2
+4e
(1/4)
2
+2e
(2/4)
2
+4e
(3/4)
2
+e
(1)
2
_
0.746855.
_
2
1
e
x
dx, N = 6
19.
_
4
1
lnx dx, N = 8
solution Let f (x) = lnx. Wedivide[1, 4] into8subintervalsof length
x =
41
8
=
3
8
= 0.375
withendpoints1, 1.375, 1.75, 2.125, 2.5, 2.875, 3.25, 3.625, 4. Withthisdata, weget
S
8
=
1
3
_
3
8
_
_
ln1+4ln(1.375) +2ln(1.75) + +4ln(3.625) +ln4
_
2.54499.
_
4
2
_
x
4
+1dx, N = 8
21.
_
/4
0
tan d, N = 10
solution Let f () = tan. Wedivide[0,

4
] into10subintervalsof width
=
4
0
10
=

40
withendpoints0,

40
,
2
40
,
3
40
, . . . ,
10
40
=

4
. Withthisdata, weget
S
10
=
1
3
_

40
_
_
tan(0) +4tan
_

40
_
+2tan
_
2
40
_
+ +4tan
_
9
40
_
+tan
_
10
40
__
0.346576.
_
2
0
(x
2
+1)
1/3
dx, N = 10
InExercises 2326, calculatetheapproximationto thevolumeof thesolidobtainedbyrotatingthegrapharoundthe
givenaxis.
23. y = cosx;
_
0,

2
_
; x-axis; M
8
V =
_
/2
0
r
2
dx =
_
/2
0
(cosx)
2
dx
whichcanbeestimatedas
_
/2
0
(cosx)
2
dx [M
8
].
Let f (x) = cos
2
x. Wedivide[0, /2] into8subintervalsof length
x =
2
0
8
=

16
withmidpoints
32
,
3
32
,
5
32
, . . . ,
15
32
.
Withthisdata, weget
V [M
8
] =
_
x(y
1
+y
2
+ +y
8
)
_
=

2
16
_
cos
2
_

32
_
+cos
2
_
3
32
_
+ +cos
2
_
15
32
__
2.46740.
y = cosx;
_
0,

2
_
; y-axis; S
8
25. y = e
x
2
; [0, 1]; x-axis; T
8
V =
_
1
0
r
2
dx =
_
1
0
_
e
x
2_
2
dx =
_
1
0
e
2x
2
dx.
Wecanusetheapproximation
V =
_
1
0
e
2x
2
dx [T
8
],
wheref (x) = e
2x
2
. Divide[0, 1] into8subintervalsof length
x =
10
8
=
1
8
,
withendpoints
0,
1
8
,
2
8
, . . . , 1.
Withthisdata, weget
V [T
8
] =
_
1
2
1
8
_
e
2(0
2
)
+2e
2(1/8)
2
+ +2e
2(7/8)
2
+e
2(1)
2
_
_
1.87691.
y = e
x
2
; [0, 1]; y-axis; S
8
27. Anairplanesvelocity isrecordedat 5-minintervalsduringa1-hour periodwiththefollowingresults, inmilesper
hour:
550, 575, 600, 580, 610, 640, 625,
595, 590, 620, 640, 640, 630
UseSimpsonsRuletoestimatethedistancetraveledduringthehour.
solution Thedistancetraveledisequal totheintegral
_
1
0
v(t ) dt , wheret isinhours. Since5minutesis1/12of an
hour, wehavet = 1/12. SimpsonsRulegivesus
S
12
=
1
3
1
12
_
550+4 575+2 600+4 580+2 610+ +4 640+630
_
608.611.
Thedistancetraveledduringthehour isapproximately608.6miles.
UseSimpsonsRuletodeterminetheaveragetemperatureinamuseumover a3-hour period, if thetemperatures
(indegreesCelsius), recordedat 15-minintervals, are
21, 21.3, 21.5, 21.8, 21.6, 21.2, 20.8,
20.6, 20.9, 21.2, 21.1, 21.3, 21.2
29. Tsunami Arrival Times Scientistsestimatethearrival timesof tsunamis(seismicoceanwaves) basedon
thepoint of originP andoceandepths. Thespeeds of atsunami inmilesper hour isapproximatelys =
15d, whered
istheoceandepthinfeet.
(a) Let f (x) betheoceandepthx milesfromP (inthedirectionof thecoast). ArgueusingRiemannsumsthat thetime
T requiredfor thetsunami totravel M milestowardthecoast is
T =
_
M
0
dx
15f (x)
(b) UseSimpsonsRuletoestimateT if M = 1000andtheoceandepths(infeet), measuredat100-mileintervalsstarting
fromP, are
13,000, 11,500, 10,500, 9000, 8500,
7000, 6000, 4400, 3800, 3200, 2000
solution
(a) At agivendistancefromshore, say, x
i
, thespeedof thetsunami inmphiss =
_
15f (x
i
). If weassumethespeeds
isconstant over asmall interval x, thenthetimetocover that interval at that speedis
t
i
=
distance
speed
=
x
_
15f (x
i
)
.
Nowdividetheinterval [0, M] intoN subintervalsof lengthx. Thetotal timeT isgivenby
T =
N
i=1
t
i
=
N
i=1
x
_
15f (x
i
)
.
Takingthelimit asN , weget
T =
_
M
0
dx
15f (x)
.
(b) Wehavex = 100. SimpsonsRulegivesus
S
10
=
1
3
100
_
1
15(13,000)
+
4
15(11,500)
+ +
1
15(2000)
_
3.347.
It will takethetsunami about 3hoursand21minutestoreachshore.
UseS
8
toestimate
_
/2
0
sinx
x
dx, takingthevalueof
sinx
x
at x = 0tobe1.
31. CalculateT
6
for theintegral I =
_
2
0
x
3
dx.
(a) IsT
6
toolargeor toosmall? Explaingraphically.
(b) Showthat K
2
= |f
(2)| maybeusedintheerror boundandndaboundfor theerror.

(c) EvaluateI andcheckthat theactual error islessthantheboundcomputedin(b).
3
. Divide[0, 2] into6subintervalsof lengthx =
20
6
=
1
3
withendpoints0,
1
3
,
2
3
, . . . , 2.
Withthisdata, weget
T
6
=
1
2
1
3
_
0
3
+2
_
1
3
_
3
+2
_
2
3
_
3
+2
_
3
3
_
3
+2
_
4
3
_
3
+2
_
5
3
_
3
+(1)2
3
_
4.11111.
(a) Sincex
3
isconcaveupon[0, 2], T
6
istoolarge.
(b) Wehavef
(x) = 3x
2
andf
(x) = 6x. Since|f
(x)| = |6x| isincreasingon[0, 2], itsmaximumvalueoccursat

x = 2andwemaytakeK
2
= |f
(2)| = 12. Then

Error(T
6
)
K
2
(b a)
3
12N
2
=
12(20)
3
12(6)
2
=
2
9
0.22222.
(c) Theexact valueis
_
2
0
x
3
dx =
1
4
x
4
2
0
=
1
4
(160) = 4.
Wecanusethistocomputetheactual error:
Error(T
6
) = |T
6
4| |4.111114| 0.11111.
Since0.11111< 0.22222, theactual error isindeedlessthanthemaximumpossibleerror.
CalculateM
4
for theintegral I =
_
1
0
x sin(x
2
) dx.
(a) Useaplot of f
(x) toshowthat K
2
= 3.2maybeusedintheerror boundandndaboundfor theerror.
(b) EvaluateI numericallyandcheckthat theactual error islessthantheboundcomputedin(a).
InExercises3336, statewhether T
N
or M
N
underestimatesor overestimatestheintegral andndaboundfor theerror
(but donot calculateT
N
or M
N
).
33.
_
4
1
1
x
dx, T
10
1
x
. Then f
(x) =
1
x
2
and f
(x) =
2
x
3
> 0 on [1, 4], so f (x) is concave up, and T
10
overestimatestheintegral. Since|f
(x)| = |
2
x
3
| hasitsmaximumvalueon[1, 4] at x = 1, wecantakeK
2
=
2
1
3
= 2,
and
Error(T
10
)
K
2
(41)
3
12N
2
=
2(3)
3
12(10)
2
= 0.045.
_
2
0
e
x/4
dx, T
20
35.
_
4
1
lnx dx, M
10
solution Let f (x) = lnx. Thenf
(x) = 1/x and

f
(x) =
1
x
2
< 0
on[1, 4], so f (x) is concavedown, andM
10
overestimates theintegral. Since|f
(x)| = | 1/x
2
| has its maximum
valueon[1, 4] at x = 1, wecantakeK
2
= | 1/1
2
| = 1, and
Error(M
10
)
K
2
(41)
3
24N
2
=
(1)(3)
3
24(10)
2
= 0.01125.
_
/4
0
cosx, M
20
InExercises3740, usetheerror boundtondavalueof N for whichError(T
N
) 10
6
. If youhaveacomputer
algebrasystem, calculatethecorrespondingapproximationandconrmthat theerror satisestherequiredbound.
37.
_
1
0
x
4
dx
4
. Thenf
(x) = 4x
3
and|f
(x)| = |12x
2
|, whichhasitsmaximumvalueon[0, 1] at x = 1,
sowecantakeK
2
= |12(1)
2
| = 12. Thenwehave
Error(T
N
)
K
2
(10)
3
12N
2
=
12
12N
2
=
1
N
2
.
Toensurethat theerror isat most 10
6
, wemust chooseN suchthat
1
N
2

1
10
6
.
ThisgivesN
2
10
6
or N 10
3
. Thuslet N = 1000. Theexact valueof theintegral is
_
1
0
x
4
dx =
x
5
5
1
0
=
1
5
= 0.2.
UsingaCAS, wendthat
T
1000
0.2000003333.
Theactual error isapproximately|0.20000033330.2| 3.33310
7
, andisindeedlessthan10
6
.
_
3
0
(5x
4
x
5
) dx
39.
_
5
2
1
x
dx
solution Let f (x) = 1/x. Thenf
(x) = 1/x
2
and|f
(x)| = |2/x
3
|, whichhas its maximumvalueon[2, 5] at
x = 2, sowecantakeK
2
= |2/2
3
| = 1/4. Thenwehave
Error(T
N
)
K
2
(52)
3
12N
2
=
(1/4)3
3
12N
2
=
9
16N
2
.
Toensurethat theerror isat most 10
6
9
16N
2

1
10
6
.
Thisgivesus
N
2
9 10
6
16
N
_
9 10
6
16
= 750.
Thuslet N = 750. Theexact valueof theintegral is
_
5
2
1
x
dx = ln5ln2 0.9162907314.
UsingaCAS, wendthat
T
750
0.9162910119.
Theerror isapproximately
|0.91629073140.9162910119| 2.80510
7
andisindeedlessthan10
6
.
_
3
0
e
x
dx
41. Computetheerror boundfor theapproximationsT
10
andM
10
to
_
3
0
(x
3
+1)
1/2
dx, usingFigure17todetermine
avalueof K
2
. Thenndavalueof N suchthat theerror inM
N
isat most 10
6
.
1 2 3
1
1
x
y
FIGURE 17 Graphof f
(x), wheref (x) = (x

3
+1)
1/2
.
solution Clearly, intherange0 x 3, wehave|f
(x)| 1, sowemaychooseK
2
= 1. Then
Error(T
10
)
K
2
(30)
3
12N
2
=
27
12 10
2
=
27
1200
= 0.0225
Error(M
10
)
K
2
(30)
3
24N
2
=
27
24 10
2
=
27
2400
= 0.01125
Inorder for theerror inM
N
tobeat most 10
6
, wemust have
Error(M
N
)
K
2
(30)
3
24N
2
=
9
8N
2
10
6
sothat 8N
2
910
6
andN
2
1,125,000. Thuswemust chooseN
1,125,000 1060.7, sothat N = 1061.

(a) ComputeS
6
for theintegral I =
_
1
0
e
2x
dx.
(b) Showthat K
4
= 16maybeusedintheerror boundandcomputetheerror bound.
(c) EvaluateI andcheckthat theactual error islessthantheboundfor theerror computedin(b).
43. CalculateS
8
for
_
5
1
lnx dx andcalculatetheerror bound. Thenndavalueof N suchthatS
N
hasanerror of atmost
10
6
.
solution Let f (x) = lnx. Wedivide[1, 5] intoeight subintervalsof lengthx = (5 1)/8= 0.5, withendpoints
1, 1.5, 2, . . . , 5. Withthisdata, weget
S
8
=
1
3
1
2
_
ln1+4ln1.5+2ln2+ +4ln4.5+ln5
_
4.046655.
Tondthemaximumpossibleerror, werst takederivatives:
f
(x) =
1
x
, f
(x) =
1
x
2
, f
(3)
(x) =
2
x
3
, f
(4)
(x) =
6
x
4
.
Since|f
(4)
(x)| = | 6x
4
| = 6x
4
, assumes its maximumvalueon[1, 5] at x = 1, wecanset K
4
= 6(1)
4
= 6.
Thenwehave
Error(S
8
)
K
4
(51)
5
180N
4
=
6 4
5
180 8
4
0.0083333.
Toensurethat S
N
haserror at most 10
6
, wemust ndN suchthat
6 4
5
180N
4

1
10
6
.
Thisgivesus
N
4
6 4
5
10
6
180
N
_
6 4
5
10
6
180
_
1/4
76.435.
Thuslet N = 78(remember that N must beevenwhenusingSimpsonsRule).
Findaboundfor theerror intheapproximationS
10
to
_
3
0
e
x
2
dx (useFigure18todetermineavalueof K
4
).
Thenndavalueof N suchthat S
N
hasanerror of at most 10
6
.
45. Useacomputer algebrasystemtocomputeandgraphf
(4)
(x) for f (x) =
_
1+x
4
andndaboundfor the
error intheapproximationS
40
to
_
5
0
f (x) dx.
solution Fromthegraphof f
(4)
(x) shownbelow, weseethat |f
(4)
(x)| 15on[0, 5]. Thereforeweset K
4
= 15.
Nowwehave
Error(S
40
)
15(50)
5
180(40)
4
=
5
49152
1.01710
4
.
5 4 3 2 1
15
10
5
15
10
5
x
y
Useacomputer algebrasystemtocomputeandgraphf
(4)
(x) for f (x) = tanx secx andndaboundfor
theerror intheapproximationS
40
to
_
/4
0
f (x) dx.
InExercises4750, usetheerror boundtondavalueof N for whichError(S
N
) 10
9
.
47.
_
6
1
x
4/3
dx
4/3
. Westart bytakingderivatives:
f
(x) =
4
3
x
1/3
f
(x) =
4
9
x
2/3
f
(x) =
8
27
x
5/3
f
(4)
(x) =
40
81
x
8/3
For x 1, f
(4)
(x) isadecreasingfunctionof x, soit takesitsmaximumvalueon[1, 6] at x = 1. That maximumvalue
is
40
81
, whichisquitecloseto(but smaller than)
1
2
. For simplicity, wetakeK
4
=
1
2
. Then
Error(S
N
)
K
4
(b a)
5
180N
4
=
(61)
5
2 180 N
4
=
5
5
360N
4
=
625
72N
4
10
9
Thus72N
4
62510
9
, sothat
N
_
62510
9
72
_
1/4
305.24
sowecantakeN = 306.
_
4
0
xe
x
dx
49.
_
1
0
e
x
2
dx
x
2
. TondK
4
, werst takederivatives:
f
(x) = 2xe
x
2
f
(x) = 4x
2
e
x
2
+2e
x
2
f
(3)
(x) = 8x
3
e
x
2
+12xe
x
2
f
(4)
(x) = 16x
4
e
x
2
+48x
2
e
x
2
+12e
x
2
.
Ontheinterval [0, 1], |f
(4)
(x)| assumesitsmaximumvalueat x = 1. Thereforeweset
K
4
= |f
(4)
(1)| = 16e +48e +12e = 76e.
Nowwehave
Error(S
N
)
K
4
(10)
5
180N
4
=
76e
180N
4
.
Toensurethat S
N
haserror at most 10
9
76e
180N
4

1
10
9
.
Thisgivesus
N
4
76e 10
9
180
N
_
76e 10
9
180
_
1/4
184.06.
Thuswelet N = 186(remember that N must beevenwhenusingSimpsonsRule).
_
4
1
sin(lnx) dx
51. Showthat
_
1
0
dx
1+x
2
=

4
[useEq. (3) inSection5.7].
(a) Useacomputer algebrasystemtographf
(4)
(x) for f (x) = (1+x
2
)
1
andnditsmaximumon[0, 1].
(b) Findavalueof N suchthatS
N
approximatestheintegral withanerror of atmost10
6
. Calculatethecorresponding
approximationandconrmthat youhavecomputed
4
toat least four places.
solution Recall fromSection3.9that
d
dx
tan
1
(x) =
1
1+x
2
.
Sothen
_
1
0
dx
1+x
2
= tan
1
x
1
0
= tan
1
(1) tan
1
(0) =

4
.
(a) Fromthegraphof f
(4)
(x) shownbelow, wecanseethatthemaximumvalueof |f
(4)
(x)| ontheinterval [0, 1] is24.
1 0.8 0.6 0.4 0.2
10
30
10
20
x
y
(b) Frompart (a), weset K
4
= 24. Thenwehave
Error(S
N
)
24(10)
5
180N
4
=
2
15N
4
.
Toensurethat S
N
haserror at most 10
6
2
15N
4

1
10
6
.
Thisgivesus
N
4
2 10
6
15
N
_
2 10
6
15
_
1/4
19.1.
Thuslet N = 20. TocomputeS
20
, let x = (10)/20= 0.05. Theendpointsof [0, 1] are0, 0.05, . . . , 1. Withthis
data, weget
S
20
=
1
3
_
1
20
__
1
1+0
2
+
4
1+(0.05)
2
+
2
1+(0.1)
2
+ +
1
1+1
2
_
0.785398163242.
Theactual error is
|0.785398163242/4| = |0.7853981632420.785398163397| = 1.5510
10
.
LetJ =
_

0
e
x
2
dx andJ
N
=
_
N
0
e
x
2
dx. Althoughe
x
2
hasnoelementaryantiderivative, itisknownthat
J =
/2. Let T
N
betheNthtrapezoidal approximationtoJ
N
. CalculateT
4
andshowthat T
4
approximatesJ to
threedecimal places.
53. Let f (x) = sin(x
2
) andI =
_
1
0
f (x) dx.
(a) Checkthatf
(x) = 2cos(x
2
) 4x
2
sin(x
2
).Thenshowthat|f
(x)| 6forx [0, 1].Hint: Notethat|2cos(x

2
)|
2and|4x
2
sin(x
2
)| 4for x [0, 1].
(b) Showthat Error(M
N
) isat most
1
4N
2
.
(c) FindanN suchthat |I M
N
| 10
3
.
solution
(a) Takingderivatives, weget
f
(x) = 2x cos(x
2
)
f
(x) = 2x(sin(x
2
) 2x) +2cos(x
2
) = 2cos(x
2
) 4x
2
sin(x
2
).
Ontheinterval [0, 1],
|f
(x)| = |2cos(x
2
) 4x
2
sin(x
2
)| |2cos(x
2
)| +|4x
2
sin(x
2
)| 2+4= 6.
(b) UsingK
2
= 6, weget
Error(M
N
)
K
2
(10)
3
24N
2
=
6
24N
2
=
1
4N
2
.
(c) Toensurethat M
N
haserror at most 10
3
1
4N
2

1
10
3
.
Thisgivesus
N
2
10
3
4
= 250 N
250 15.81.
Thuslet N = 16.
Theerror boundfor M
N
isproportional to1/N
2
, sotheerror bounddecreasesby
1
4
if N isincreasedto2N.
Computetheactual error inM
N
for
_
0
sinx dx for N = 4, 8, 16, 32, and64. Doestheactual error seemtodecrease
by
1
4
asN isdoubled?
55. Observethattheerror boundfor T
N
(whichhas12inthedenominator) istwiceaslargeastheerror
boundfor M
N
(whichhas24inthedenominator). Computetheactual error inT
N
for
_
0
sinx dx for N = 4, 8, 16, 32,
and64andcomparewiththecalculationsof Exercise54. Doestheactual error inT
N
seemtoberoughlytwiceaslarge
astheerror inM
N
inthiscase?
solution Theexact valueof theintegral is
_

0
sinx dx = cosx
0
= (1) (1) = 2.
TocomputeT
4
, wehavex = ( 0)/4= /4, andendpoints0, /4, 2/4, 3/4, . Withthisdata, weget
T
4
=
1
2

4
_
sin(0) +2sin
_
4
_
+2sin
_
2
4
_
+2sin
_
3
4
_
+sin()
_
1.896119.
Thevaluesfor T
8
, T
16
, T
32
, andT
64
arecomputedsimilarly:
T
8
=
1
2

8
_
sin(0) +2sin
_
8
_
+2sin
_
2
8
_
+ +2sin
_
7
8
_
+sin()
_
1.974232;
T
16
=
1
2

16
_
sin(0) +2sin
_

16
_
+2sin
_
2
16
_
+ +2sin
_
15
16
_
+sin()
_
1.993570;
T
32
=
1
2

32
_
sin(0) +2sin
_

32
_
+2sin
_
2
32
_
+ +2sin
_
31
32
_
+sin()
_
1.998393;
T
64
=
1
2

64
_
sin(0) +2sin
_

64
_
+2sin
_
2
64
_
+ +2sin
_
63
64
_
+sin()
_
1.999598.
Nowwecancomputetheactual errorsfor eachN:
Error(T
4
) = |21.896119| = 0.103881
Error(T
8
) = |21.974232| = 0.025768
Error(T
16
) = |21.993570| = 0.006430
Error(T
32
) = |21.998393| = 0.001607
Error(T
64
) = |21.999598| = 0.000402
Comparingtheseresultswiththecalculationsof Exercise54, weseethat theactual error inT
N
isinfact about twiceas
largeastheerror inM
N
.
Explainwhy theerror boundfor S
N
decreasesby
1
16
if N isincreasedto2N. Computetheactual error in
S
N
for
_
0
sinx dx for N = 4, 8, 16, 32, and64. Doestheactual error seemtodecreaseby
1
16
asN isdoubled?
57. Verifythat S
2
yieldstheexact valueof
_
1
0
(x x
3
) dx.
solution Let f (x) = x x
3
. Clearlyf
(4)
(x) = 0, sowemaytakeK
4
= 0intheerror boundestimatefor S
2
. Then
Error(S
2
)
K
4
(10)
5
180 2
4
= 0
1
2880
= 0
sothat S
2
yieldstheexact valueof theintegral.
Verifythat S
2
yieldstheexact valueof
_
b
a
(x x
3
) dx for all a < b.
59. Showthat if f (x) = rx +s isalinear function(r, s constants), thenT
N
=
_
b
a
f (x) dx for all N andall endpoints
a, b.
solution First, notethat
_
b
a
(rx +s) dx =
r(b
2
a
2
)
2
+s(b a).
Now,
T
N
(rx +s) =
b a
2N
f (a) +2
N1
i=1
f (x
i
) +f (b)
=
r(b a)
2N
a +2
N1
i=1
a +2
b a
N
N1
i=1
i +b
+s
b a
2N
(2N)
=
r(b a)
2N
_
(2N 1)a +2
b a
N
(N 1)N
2
+b
_
+s(b a) =
r(b
2
a
2
)
2
+s(b a).
Showthat if f (x) = px
2
+qx +r isaquadraticpolynomial, thenS
2
=
_
b
a
f (x) dx. Inother words, showthat
_
b
a
f (x) dx =
b a
6
_
y
0
+4y
1
+y
2
_
wherey
0
= f (a), y
1
= f
_
a +b
2
_
, andy
2
= f (b). Hint: Showthisrst for f (x) = 1, x, x
2
anduselinearity.
61. For N even, divide[a, b] intoN subintervalsof widthx =
b a
N
. Set x
j
= a +j x, y
j
= f (x
j
), and
S
2j
2
=
b a
3N
_
y
2j
+4y
2j+1
+y
2j+2
_
(a) ShowthatS
N
isthesumof theapproximationsontheintervals[x
2j
, x
2j+2
]thatis, S
N
= S
0
2
+S
2
2
+ +S
N2
2
.
(b) ByExercise60, S
2j
2
=
_
x
2j+2
x
2j
f (x) dx if f (x) isaquadraticpolynomial. Use(a) toshowthatS
N
isexactfor all N
if f (x) isaquadraticpolynomial.
solution
(a) Thisresult followsbecausetheeven-numberedinterior endpointsoverlap:
(N2)/2
i=0
S
2j
2
=
b a
6
[(y
0
+4y
1
+y
2
) +(y
2
+4y
3
+y
4
) + ]
=
b a
6
_
y
0
+4y
1
+2y
2
+4y
3
+2y
4
+ +4y
N1
+y
N
_
= S
N
.
(b) If f (x) isaquadraticpolynomial, thenbypart (a) wehave
S
N
= S
0
2
+S
2
2
+ +S
N2
2
=
_
x
2
x
0
f (x) dx +
_
x
4
x
2
f (x) dx + +
_
x
N
x
N2
f (x) dx =
_
b
a
f (x) dx.
Showthat S
2
also gives theexact valuefor
_
b
a
x
3
dx andconclude, as inExercise61, that S
N
is exact for all
cubicpolynomials. Showbycounterexamplethat S
2
isnot exact for integralsof x
4
.
63. Usetheerror boundfor S
N
toobtainanother proof that SimpsonsRuleisexact for all cubicpolynomials.
solution Let f (x) = ax
3
+bx
2
+cx +d, witha = 0, beanycubicpolynomial. Then, f
(4)
(x) = 0, sowecantake
K
4
= 0. Thisyields
Error(S
N
)
0
180N
4
= 0.
Inother words, S
N
isexact for all cubicpolynomialsfor all N.
Sometimes,SimpsonsRulePerformsPoorly CalculateM
10
andS
10
fortheintegral
_
1
0
_
1x
2
dx, whose
valueweknowtobe
4
(one-quarter of theareaof theunit circle).
(a) Weusuallyexpect S
N
tobemoreaccuratethanM
N
. Whichof M
10
andS
10
ismoreaccurateinthiscase?
(b) Howdoyouexplaintheresult of part (a)? Hint: Theerror boundsarenot validbecause|f
(x)| and|f
(4)
(x)|
d 1 b |f
(4)
( )| i i f
1. Matchtheintegrals (a)(e) withtheir antiderivatives (i)(v) onthebasis of thegeneral form(do not evaluatethe
integrals).
(a)
_
x dx
x
2
4
(b)
_
(2x +9) dx
x
2
+4
(c)
_
sin
3
x cos
2
x dx (d)
_
dx
x
_
16x
2
1
(e)
_
16dx
x(x 4)
2
(i) sec
1
4x +C
(ii) log|x| log|x 4|
4
x 4
+C
(iii)
1
30
(3cos
5
x 3cos
3
x sin
2
x 7cos
3
x) +C
(iv)
9
2
tan
1
x
2
+ln(x
2
+4) +C (v)
_
x
2
4+C
solution
(a)
_
x dx
_
x
2
4
Sincex isaconstant multipleof thederivativeof x
2
4, thesubstitutionmethodimpliesthat theintegral isaconstant
multipleof
_
du
u
whereu = x
2
4, that isaconstant multipleof
u =
_
x
2
4. It correspondstothefunctionin(v).
(b)
_
(2x +9) dx
x
2
+4
Thepart
_
2x
x
2
+4
dx corresponds to ln(x
2
+ 4) in (iv) and thepart
_
9
x
2
+4
dx corresponds to
9
2
tan
1 x
2
. Hencethe
integral correspondstothefunctionin(iv).
(c)
_
sin
3
xcos
2
x dx
Thereductionformulafor
_
sin
m
x cos
n
x dx showsthatthisintegral isequal toasumof constantmultiplesof products
intheformcos
i
x sin
j
x asin(iii).
(d)
_
dx
x
_
16x
2
1
Since
_
dx
|x|
x
2
1
= sec
1
x +C, weexpect theintegral
_
dx
x
16x
2
1
tobeequal tothefunctionin(i).
(e)
_
16dx
x(x 4)
2
Thepartial fractiondecompositionof theintegrandhastheform:
A
x
+
B
x 4
+
C
(x 4)
2
Theterm
A
x
contributesthefunctionAln|x| totheintegral, theterm
B
x4
contributesB ln|x 4| andtheterm
C
(x4)
2
contributes
C
x4
. Therefore, weexpect theintegral tobeequal tothefunctionin(ii).
Evaluate
_
x dx
x +2
intwoways: usingsubstitutionandusingtheMethodof Partial Fractions.
InExercises312, evaluateusingthesuggestedmethod.
3.
_
cos
3
sin
8
d [writecos
3
ascos(1sin
2
)]
solution Weusetheidentitycos
2
= 1sin
2
torewritetheintegral:
_
cos
3
sin
8
d =
_
cos
2
sin
8
cos d =
_
_
1sin
2
_
sin
8
cos d.
Now, weusethesubstitutionu = sin, du = cos d:
_
cos
3
sin
8
d =
_
_
1u
2
_
u
8
du =
_
_
u
8
u
10
_
du =
u
9
9

u
11
11
+C =
sin
9
9

sin
11
11
+C.
_
xe
12x
dx (IntegrationbyParts)
5.
_
sec
3
tan
4
d (trigonometricidentity, reductionformula)
solution Weusetheidentity1+tan
2
= sec
2
towritetan
4
=
_
sec
2
1
_
2
andtorewritetheintegral as
_
sec
3
tan
4
d
_
sec
3
_
1sec
2
_
2
d =
_
sec
3
_
12sec
2
+sec
4
_
d
=
_
sec
7
d 2
_
sec
5
d +
_
sec
3
d.
Nowweusethereductionformula
_
sec
m
d =
tansec
m2
m1
+
m2
m1
_
sec
m2
d.
Wehave
_
sec
5
d =
tansec
3
4
+
3
4
_
sec
3
d +C,
and
_
sec
7
d =
tansec
5
6
+
5
6
_
sec
5
d =
tansec
5
6
+
5
6
_
tansec
3
4
+
3
4
_
sec
3
d
_
+C
=
tansec
5
6
+
5
24
tansec
3
+
5
8
_
sec
3
d +C.
Therefore,
_
sec
3
tan
4
d =
_
tansec
5
6
+
5
24
tansec
3
+
5
8
_
sec
3
d
_
2
_
tansec
3
4
+
3
4
_
sec
3
d
_
+
_
sec
3
d
=
tansec
5
6

7tansec
3
24
+
1
8
_
sec
3
d.
Weagainusethereductionformulatocompute
_
sec
3
d =
tan sec
2
+
1
2
_
sec d =
tan sec
2
+
1
2
ln| sec +tan| +C.
Finally,
_
sec
3
tan
4
d =
tansec
5
6

7tansec
3
24
+
tan sec
16
+
1
16
ln| sec +tan| +C.
_
4x +4
(x 5)(x +3)
dx (partial fractions)
7.
_
dx
x(x
2
1)
3/2
dx (trigonometricsubstitution)
solution Substitutex = sec, dx = sec tan d. Then,
_
x
2
1
_
3/2
=
_
sec
2
1
_
3/2
=
_
tan
2
_
3/2
= tan
3
,
and
_
dx
x
_
x
2
1
_
3/2
=
_
sec tan d
sectan
3
=
_
d
tan
2
=
_
cot
2
d.
Usingareductionformulawendthat:
_
cot
2
d = cot +C
so
_
dx
x
_
x
2
1
_
3/2
= cot +C.
Wenowmust returntotheoriginal variablex. Weusetherelationx = sec andtheguretoobtain:
_
dx
x
_
x
2
1
_
3/2
=
1
_
x
2
1
sec
1
x +C.
q
1
x
x
2
1
_
(1+x
2
)
3/2
dx (trigonometricsubstitution)
9.
_
dx
x
3/2
+x
1/2
(substitution)
solution Let t = x
1/2
. Thendt =
1
2
x
1/2
dx or dx = 2x
1/2
dt = 2t dt . Therefore,
_
dx
x
3/2
+x
1/2
=
_
2t dt
t
3
+t
=
_
2dt
t
2
+1
= 2tan
1
t +C = 2tan
1
x +C.
_
dx
x +x
1
(rewriteintegrand)
11.
_
x
2
tan
1
x dx (IntegrationbyParts)
solution WeuseIntegrationbyPartswithu = tan
1
x andv
= x
2
. Thenu
=
1
1+x
2
, v = x
1
and
_
x
2
tan
1
x dx =
tan
1
x
x
+
_
dx
x
_
1+x
2
_ .
For theremainingintegral, thepartial fractiondecompositiontakestheform
1
x(1+x
2
)
=
A
x
+
Bx +C
1+x
2
.
1= A(1+x
2
) +(Bx +C)x.
Settingx = 0thenyieldsA = 1. Next, equatingthex
2
-coefcientsgives
0= A +B so B = 1,
whileequatingx-coefcientsgivesC = 0. Hence,
1
x
_
1+x
2
_ =
1
x
x
1+x
2
,
and
_
dx
x(1+x
2
)
=
_
1
x
dx
_
x dx
1+x
2
= ln|x|
1
2
ln
_
1+x
2
_
+C.
Therefore,
_
x
2
tan
1
x dx =
tan
1
x
x
+ln|x|
1
2
ln
_
1+x
2
_
+C.
_
dx
x
2
+4x 5
(completethesquare, substitution, partial fractions)
InExercises1364, evaluateusingtheappropriatemethodor combinationof methods.
13.
_
1
0
x
2
e
4x
dx
solution Weevaluatetheindeniteintegral usingIntegrationby Parts withu = x
2
andv
= e
4x
. Thenu
= 2x,
v =
1
4
e
4x
and
_
x
2
e
4x
dx =
x
2
4
e
4x
1
2
_
xe
4x
dx.
Wecomputetheresultingintegral usingIntegrationby Parts again, this timewithu = x andv
= e
4x
. Thenu
= 1,
v =
1
4
e
4x
and
_
xe
4x
dx = x
1
4
e
4x
_
1
4
e
4x
dx =
x
4
e
4x
1
16
e
4x
+C.
Therefore,
_
x
2
e
4x
dx =
x
2
4
e
4x
1
2
_
x
4
e
4x
1
16
e
4x
_
+C =
e
4x
32
_
8x
2
4x +1
_
+C.
Finally,
_
1
0
x
2
e
4x
dx =
_
e
4x
32
_
8x
2
4x +1
_
_
1
0
=
e
4
32
(84+1)
1
32
(1) =
5e
4
1
32
_
x
2
_
9x
2
dx
15.
_
cos
9
6 sin
3
6 d
solution Weusetheidentitysin
2
6 = 1cos
2
6 torewritetheintegral:
_
cos
9
6sin
3
6 d =
_
cos
9
6sin
2
6 sin6 d =
_
cos
9
6
_
1cos
2
6
_
sin6 d.
Now, weusethesubstitutionu = cos6, du = 6sin6 d:
_
cos
9
6sin
3
6 d =
_
u
9
_
1u
2
_
_
du
6
_
=
1
6
_
_
u
9
u
11
_
du
=
1
6
_
u
10
10

u
12
12
_
+C =
cos
12
6
72

cos
10
6
60
+C.
_
sec
2
tan
4
d
17.
_
(6x +4) dx
x
2
1
solution Thepartial fractiondecompositiontakestheform
6x +4
(x 1)(x +1)
=
A
x 1
+
B
x +1
.
Clearingthedenominatorsgivesus
6x +4= A(x +1) +B(x 1).
Settingx = 1thenyieldsA = 5, whilesettingx = 1yieldsB = 1. Hence,
_
(6x +4)dx
x
2
1
=
_
5
x 1
dx +
_
1
x +1
dx = 5ln|x 1| +ln|x +1| +C.
_
9
4
dt
(t
2
1)
2
19.
_
d
cos
4
solution Weusetheidentity1+tan
2
= sec
2
_
d
cos
4
=
_
sec
4
d =
_
_
1+tan
2
_
sec
2
d.
Now, wesubstituteu = tan. Then, du = sec
2
d and
_
d
cos
4
=
_
_
1+u
2
_
du = u +
u
3
3
+C =
tan
3
3
+tan +C.
_
sin2 sin
2
d
21.
_
1
0
ln(42x) dx
solution Notethat ln(4 2x) = ln(2(2 x)) = ln2+ ln(2 x). Useintegrationby parts tointegrateln(2 x),
withu = ln(2x), v
= 1, sothat u
=
1
2x
andv = x. Then
I =
_
1
0
ln(42x) dx =
_
1
0
ln2dx +
_
1
0
ln(2x) dx = ln2+(x ln(2x))
1
0
+
_
1
0
x
2x
dx
Nowuselongdivisionontheremainingintegral, andthesubstitutionu = 2x:
I = ln2+(x ln(2x))
1
0
+
_
1
0
_
1+
2
2x
_
dx
= ln2+1ln1
_
1
0
1dx +2
_
1
0
1
2x
dx = ln212
_
1
2
1
u
du
= ln212lnu
1
2
= ln21+2ln2= 3ln21
_
(ln(x +1))
2
dx
23.
_
sin
5
d
solution Weusethetrigonometricidentitysin
2
= 1cos
2
_
sin
5
d =
_
sin
4
sin d =
_
_
1cos
2
_
2
sin d.
Now, wesubstituteu = cos. Thendu = sin d and
_
sin
5
d =
_
_
1u
2
_
2
(du) =
_
_
12u
2
+u
4
_
du
=
_
u
2
3
u
3
+
u
5
5
_
+C =
cos
5
5
+
2cos
3
3
cos +C.
_
cos
4
(9x 2) dx
25.
_
/4
0
sin3x cos5x dx
solution First computetheindeniteintegral, usingthetrigonometricidentity:
sin cos =
1
2
(sin( +) +sin( )) .
For = 3x and = 5x weget:
sin3x cos5x =
1
2
(sin8x +sin(2x)) =
1
2
(sin8x sin2x).
Hence,
_
sin3x cos5x dx =
1
2
_
sin8x dx
1
2
_
sin2x dx =
1
16
cos8x +
1
4
cos2x +C.
Then
_
/4
0
sin3x cos5x dx =
_
1
4
cos2x
1
16
cos8x
_
/4
0
=
1
4
cos
2

1
16
cos2
1
4
cos0+
1
16
cos0=
1
4
_
sin2x sec
2
x dx
27.
_

tanx sec
2
x dx
solution Wesubstituteu = tanx. Thendu = sec
2
x dx andweobtain:
_

tanx sec
2
x dx =
_

u du =
2
3
u
3/2
+C =
2
3
(tanx)
3/2
+C.
_
(secx +tanx)
2
dx
29.
_
sin
5
cos
3
d
solution Weusetheidentitycos
2
= 1sin
2
_
sin
5
cos
3
d =
_
sin
5
cos
2
cos d =
_
sin
5
_
1sin
2
_
cos d.
Now, weusethesubstitutionu = sin, du = cos d:
_
sin
5
cos
3
d =
_
u
5
_
1u
2
_
du =
_
_
u
5
u
7
_
du =
u
6
6

u
8
8
+C =
sin
6
6

sin
8
8
+C.
_
cot
3
x cscx dx
31.
_
cot
2
x csc
2
x dx
solution Usethesubstitutionu = cotx, du = csc
2
x dx:
_
cot
2
x csc
2
x dx =
_
cot
2
x
_
csc
2
x dx
_
=
_
u
2
du =
1
3
u
3
+C =
1
3
cot
3
x +C.
_

/2
cot
2

2
d
33.
_
/2
/4
cot
2
x csc
3
x dx
solution Tocomputetheindeniteintegral, usetheidentitycot
2
x = csc
2
x 1towrite
_
cot
2
x csc
3
x dx =
_
_
csc
2
x 1
_
csc
3
x dx =
_
csc
5
x dx
_
csc
3
x dx.
Nowusethereductionformulafor csc
m
x:
_
cot
2
x csc
3
x dx =
_
1
4
cotx csc
3
x +
3
4
_
csc
3
x dx
_
_
csc
3
x dx
=
1
4
cotx csc
3
x
1
4
_
csc
3
x dx
=
1
4
cotx csc
3
x
1
4
_
1
2
cotx cscx +
1
2
_
cscx dx
_
=
1
4
cotx csc
3
x +
1
8
cotx cscx
1
8
ln| cscx cotx| +C.
Then
_
/2
/4
cos
2
x csc
3
x dx =
_
1
4
cotx csc
3
x +
1
8
cotx cscx
1
8
ln| cscx cotx|
_
/2
/4
=
1
4
cot
2
csc
3

2
+
1
8
cot
2
csc
2

1
8
ln
csc
2
cot
+
1
4
cot
4
csc
3

4

1
8
cot
4
csc
4
+
1
8
ln
csc
4
cot
= 0+0
1
8
ln|10| +
1
4
1 (
2)
3
1
8
1
2+
1
8
ln
21
2
2

2
8
+
1
8
ln(
21) =
3
8
2+
1
8
ln(
21)
_
6
4
dt
(t 3)(t +4)
35.
_
dt
(t 3)
2
(t +4)
1
(t 3)
2
(t +4)
=
A
t +4
+
B
t 3
+
C
(t 3)
2
.
1= A(t 3)
2
+B(t 3)(t +4) +C(t +4).
Settingt = 3thenyieldsC =
1
7
, whilesettingt = 4yieldsA =
1
49
. Lastly, settingt = 0yields
1= 9A 12B +4C or B =
1
49
.
Hence,
_
dt
(t 3)
2
(t +4)
=
1
49
_
dt
t +4
1
49
_
dt
t 3
+
1
7
_
dt
(t 3)
2
=
1
49
ln|t +4|
1
49
ln|t 3| +
1
7
1
t 3
+C =
1
49
ln
t +4
t 3
1
7
1
t 3
+C.
_
_
x
2
+9dx
37.
_
dx
x
_
x
2
4
solution Substitutex = 2sec, dx = 2sec tan d. Then
_
x
2
4=
_
4sec
2
4=
_
4
_
sec
2
1
_
=
_
4tan
2
= 2tan,
and
_
dx
x
_
x
2
4
=
_
2sec tan d
2sec 2tan
=
1
2
_
d =
1
2
+C.
Now, returntotheoriginal variablex. Sincex = 2sec, wehavesec =
x
2
or = sec
1x
2
. Thus,
_
dx
x
_
x
2
4
=
1
2
sec
1
x
2
+C.
_
27
8
dx
x +x
2/3
39.
_
dx
x
3/2
+ax
1/2
solution Let u = x
1/2
or x = u
2
. Thendx = 2u du and
_
dx
x
3/2
+ax
1/2
=
_
2u du
u
3
+au
= 2
_
du
u
2
+a
.
If a > 0, then
_
dx
x
3/2
+ax
1/2
= 2
_
du
u
2
+a
=
2
a
tan
1
_
u
a
_
+C =
2
a
tan
1
_
x
a
+C.
If a = 0, then
_
dx
x
3/2
=
2
x
+C.
Finally, if a < 0, then
_
du
u
2
+a
=
_
du
u
2
_
a
_
2
,
andthepartial fractiondecompositiontakestheform
1
u
2
_
a
_
2
=
A
u
a
+
B
u +
a
.
1= A(u +
a) +B(u
a).
Settingu =
a thenyieldsA =
1
2
a
, whilesettingu =
a yieldsB =
1
2
a
. Hence,
_
dx
x
3/2
+ax
1/2
= 2
_
du
u
2
+a
=
1
a
_
du
u
a
_
du
u +
a
=
1
a
ln|u
a|
1
a
ln
u +
+C
=
1
a
ln
a
u +
+C =
1
a
ln
x +
+C.
Insummary,
_
dx
x
3/2
+ax
1/2
=
a
tan
1
_
x
a
+C a > 0
1
a
ln
x+
+C a < 0
x
+C a = 0
_
dx
(x b)
2
+4
41.
_
(x
2
x) dx
(x +2)
3
x
2
x
(x +2)
3
=
A
x +2
+
B
(x +2)
2
+
C
(x +2)
3
.
x
2
x = A(x +2)
2
+B(x +2) +C.
Setting x = 2 then yields C = 6. Equating x
2
-coefcients gives us A = 1, and equating x-coefcients yields
4A +B = 1, or B = 5. Thus,
_
x
2
x
(x +2)
3
dx =
_
dx
x +2
+
_
5dx
(x +2)
2
+
_
6dx
(x +2)
3
= ln|x +2| +
5
x +2
3
(x +2)
2
+C.
_
(7x
2
+x) dx
(x 2)(2x +1)(x +1)
43.
_
16dx
(x 2)
2
(x
2
+4)
16
(x 2)
2
_
x
2
+4
_ =
A
x 2
+
B
(x 2)
2
+
Cx +D
x
2
+4
.
16= A(x 2)
_
x
2
+4
_
+B
_
x
2
+4
_
+(Cx +D)(x 2)
2
.
Settingx = 2thenyieldsB = 2. WithB = 2,
16= A
_
x
3
2x
2
+4x 8
_
+2
_
x
2
+4
_
+Cx
3
+(D 4C)x
2
+(4C 4D)x +4D
16= (A +C)x
3
+(2A +2+D 4C) x
2
+(4A +4C 4D)x +(8A +8+4D)
Equatingcoefcientsof likepowersof x nowgivesusthesystemof equations
A +C = 0
2A 4C +D +2= 0
4A +4C 4D = 0
8A +4D +8= 1
whosesolutionis
A = 1, C = 1, D = 0.
Thus,
_
dx
(x 2)
2
_
x
2
+4
_ =
_
dx
x 2
+2
_
dx
(x 2)
2
+
_
x
x
2
+4
dx
= ln|x 2| 2
1
x 2
+
1
2
ln
_
x
2
+4
_
+C.
_
dx
(x
2
+25)
2
45.
_
dx
x
2
+8x +25
solution Completethesquaretorewritethedenominator as
x
2
+8x +25= (x +4)
2
+9.
Now, let u = x +4, du = dx. Then,
_
dx
x
2
+8x +25
=
_
du
u
2
+9
=
1
3
tan
1
u
3
+C =
1
3
tan
1
_
x +4
3
_
+C.
_
dx
x
2
+8x +4
47.
_
(x
2
x) dx
(x +2)
3
x
2
x
(x +2)
3
=
A
x +2
+
B
(x +2)
2
+
C
(x +2)
3
.
x
2
x = A(x +2)
2
+B(x +2) +C.
Setting x = 2 then yields C = 6. Equating x
2
-coefcients gives us A = 1, and equating x-coefcients yields
4A +B = 1, or B = 5. Thus,
_
x
2
x
(x +2)
3
dx =
_
dx
x +2
+
_
5dx
(x +2)
2
+
_
6dx
(x +2)
3
= ln|x +2| +
5
x +2
3
(x +2)
2
+C.
_
1
0
t
2
_
1t
2
dt
49.
_
dx
x
4
_
x
2
+4
solution Substitutex = 2tan, dx = 2sec
2
d. Then
_
x
2
+4=
_
4tan
2
+4=
_
4
_
tan
2
+1
_
= 2
_
sec
2
= 2sec,
and
_
dx
x
4
_
x
2
+4
=
_
2sec
2
d
16tan
4
2sec
=
_
sec d
16tan
4
.
Wehave
sec
tan
4
=
cos
3
sin
4
.
Hence,
_
dx
x
4
_
x
2
+4
=
1
16
_
cos
3
d
sin
4
=
1
16
_
cos
2
cos d
sin
4
=
1
16
_
_
1sin
2
_
cos d
sin
4
.
Nowsubstituteu = sin anddu = cos d toobtain
_
dx
x
4
_
x
2
+4
=
1
16
_
1u
2
u
4
du =
1
16
_
_
u
4
u
2
_
du =
1
48u
3
+
1
16
1
u
+C
=
1
48
1
sin
3
+
1
16
1
sin
+C =
1
48
csc
3
+
1
16
csc +C.
Finally, returntotheoriginal totheoriginal variablex usingtherelationx = 2tan andthegurebelow.
_
dx
x
4
_
x
2
+4
=
1
48
__
x
2
+4
x
_
3
+
1
16
_
x
2
+4
x
+C =
_
x
2
+4
_
3/2
48x
3
+
_
x
2
+4
16x
+C.
q
x
2
+ 4
x
2
_
dx
(x
2
+5)
3/2
51.
_
(x +1)e
43x
dx
solution We compute the integral using Integration by Parts with u = x + 1 and v
= e
43x
. Then u
= 1,
v =
1
3
e
43x
and
_
(x +1)e
43x
dx =
1
3
(x +1)e
43x
+
1
3
_
e
43x
dx =
1
3
(x +1)e
43x
+
1
3
1
3
_
e
43x
+C
=
1
9
e
43x
(3x +4) +C.
_
x
2
tan
1
x dx
53.
_
x
3
cos(x
2
) dx
solution Substitutet = x
2
, dt = 2x dx. Then
_
x
3
cos
_
x
2
_
dx =
1
2
_
t cost dt.
Wecomputetheresultingintegral usingIntegrationbyPartswithu = t andv
= cost . Thenu
= 1, v = sint and
_
t cost dt = t sint
_
sint dt = t sint +cost +C.
Thus,
_
x
3
cos
_
x
2
_
dx =
1
2
x
2
sinx
2
+
1
2
cosx
2
+C.
_
x
2
(lnx)
2
dx
55.
_
x tanh
1
x dx
solution WeuseIntegrationbyPartswithu = tanh
1
x andv
= x. Thenu
=
1
1x
2
, v =
x
2
2
and
_
x tanh
1
x dx =
x
2
2
tanh
1
x
1
2
_
x
2
1x
2
dx.
Now
x
2
1x
2
=
x
2
1+1
1x
2
= 1+
1
1x
2
,
andthepartial fractiondecompositionfor theremainingfractiontakestheform
1
1x
2
=
A
1x
+
B
1+x
.
1= A(1+x) +B(1x).
Settingx = 1thenyieldsA =
1
2
, whilesettingx = 1yieldsB =
1
2
. Thus,
_
x
2
1x
2
=
_
dx +
1
2
_
1
1x
dx +
1
2
_
1
1+x
dx
= x
1
2
ln|1x| +
1
2
ln|1+x| +C = x +
1
2
ln
1+x
1x
+C.
Therefore,
_
x tanh
1
x dx =
x
2
2
tanh
1
x
1
2
_
x +
1
2
ln
1+x
1x
_
+C =
x
2
2
tanh
1
x +
x
2
1
4
ln
1+x
1x
+C.
_
tan
1
t dt
1+t
2
57.
_
ln(x
2
+9) dx
solution Wecomputetheintegral usingIntegrationby Parts withu = ln
_
x
2
+9
_
andv
= 1. Thenu
=
2x
x
2
+9
,
v = x, and
_
ln
_
x
2
+9
_
dx = x ln
_
x
2
+9
_
_
2x
2
x
2
+9
dx.
Tocomputethisintegral wewrite:
x
2
x
2
+9
=
_
x
2
+9
_
9
x
2
+9
= 1
9
x
2
+9
;
hence,
_
x
2
x
2
+9
dx =
_
1dx 9
_
dx
x
2
+9
= x 3tan
1
x
3
+C.
Therefore,
_
ln
_
x
2
+9
_
dx = x ln
_
x
2
+9
_
2x +6tan
1
_
x
3
_
+C.
_
(sinx)(coshx) dx
59.
_
1
0
cosh2t dt
solution
_
1
0
cosh2t dt =
1
2
sinh2t
1
0
=
1
2
sinh2.
_
sinh
3
x coshx dx
61.
_
coth
2
(14t ) dt
solution
_
coth
2
(14t ) dt =
_
_
1+csch
2
(14t )
_
dt = t +
1
4
coth(14t ) +C.
_
0.3
0.3
dx
1x
2
63.
_
3
3/2
0
dx
_
9x
2
solution
_
3
3/2
0
dx
_
9x
2
= sin
1
x
3
3/2
0
= sin
1
3
2
=

3
.
_
_
x
2
+1dx
x
2
65. Usethesubstitutionu = tanht toevaluate
_
dt
cosh
2
t +sinh
2
t
.
solution Let u = tanht . Thendu = sech
2
t dt and
_
dt
cosh
2
t +sinh
2
t
=
_
sech
2
t
1+tanh
2
t
dt =
_
du
1+u
2
= tan
1
u +C = tan
1
(tanhx) +C.
Findthevolumeobtainedbyrotatingtheregionenclosedbyy = lnx andy = (lnx)
2
about they-axis.
67. Let I
n
=
_
x
n
dx
x
2
+1
.
(a) Provethat I
n
=
x
n1
n 1
I
n2
.
(b) Use(a) tocalculateI
n
for 0 n 5.
(c) Showthat, ingeneral,
I
2n+1
=
x
2n
2n
x
2n2
2n 2
+ +(1)
n1
x
2
2
+(1)
n
1
2
ln(x
2
+1) +C
I
2n
=
x
2n1
2n 1
x
2n3
2n 3
+ +(1)
n1
x +(1)
n
tan
1
x +C
solution
(a) I
n
=
_
x
n
x
2
+1
dx =
_
x
n2
(x
2
+11)
x
2
+1
dx =
_
x
n2
dx
_
x
n2
x
2
+1
dx =
x
n1
n 1
I
n2
.
(b) First computeI
0
andI
1
directly:
I
0
=
_
x
0
dx
x
2
+1
=
_
dx
x
2
+1
= tan
1
x +C and I
1
=
_
x dx
x
2
+1
=
1
2
ln
_
x
2
+1
_
+C.
Wenowusetheequalityobtainedinpart (a) tocomputeI
2
, I
3
, I
4
andI
5
:
I
2
=
x
21
21
I
22
= x I
0
= x tan
1
x +C;
I
3
=
x
31
31
I
32
=
x
2
2
I
1
=
x
2
2

1
2
ln
_
x
2
+1
_
+C;
I
4
=
x
41
41
I
42
=
x
3
3
I
2
=
x
3
3

_
x tan
1
x
_
+C =
x
3
3
x +tan
1
x +C;
I
5
=
x
51
51
I
52
=
x
4
4
I
3
=
x
4
4

_
x
2
2

1
2
ln
_
x
2
+1
_
_
+C =
x
4
4

x
2
2
+
1
2
ln
_
x
2
+1
_
+C.
(c) Weprovethetwoidentitiesusingmathematical induction. Werst provethat for n 1:
I
2n+1
=
x
2n
2n
x
2n2
2n 2
+ +(1)
n
1
2
ln
_
x
2
+1
_
+C.
Weverifytheequalityfor n = 1. Settingn = 1, wend
I
3
=
x
2
2
+(1)
1
1
2
ln
_
x
2
+1
_
+C =
x
2
2

1
2
ln
_
x
2
+1
_
+C,
whichagreeswiththevalueobtainedinpart (b). Wenowassumethat for n = k:
I
2k+1
=
x
2k
2k
x
2k2
2k 2
+ +(1)
k
1
2
ln
_
x
2
+1
_
+C.
Weusethisassumptiontoprovetheequalityfor n = k +1. Bypart (a) andtheinductionhypothesis
I
2k+3
=
x
2k+2
2k +2
I
2k+1
=
x
2k+2
2k +2
x
2k
2k
+
x
2k2
2k 2
(1)
k
1
2
ln
_
x
2
+1
_
+C
=
x
2k+2
2k +2
x
2k
2k
+ +(1)
k+1
1
2
ln
_
x
2
+1
_
+C
asrequired. Wenowprovethesecondidentityfor n 1:
I
2n
=
x
2n1
2n 1
x
2n3
2n 3
+ +(1)
n
tan
1
x +C.
Weverifythisequalityfor n = 1:
I
2
= x tan
1
x +C,
whichagreeswiththevalueobtainedinpart (b). Wenowassumethat for n = k
I
2k
=
x
2k1
2k 1
x
2k3
2k 3
+ +(1)
k
tan
1
x +C.
Weusethisassumptiontoprovetheequalityfor n = k +1. Bypart (a) andtheinductionhypothesis
I
2k+2
=
x
2k+1
2k +1
I
2k
=
x
2k+1
2k +1
x
2k1
2k 1
+
x
2k3
2k 3
(1)
k
tan
1
x +C
=
x
2k+1
2k +1
x
2k1
2k 1
+ +(1)
k+1
tan
1
x +C
asrequired.
Let J
n
=
_
x
n
e
x
2
/2
dx.
(a) Showthat J
1
= e
x
2
/2
.
(b) Provethat J
n
= x
n1
e
x
2
/2
+(n 1)J
n2
.
(c) Use(a) and(b) tocomputeJ
3
andJ
5
.
69. Computep(X 1), whereX isacontinuousrandomvariablewithprobabilitydensityp(x) =
1
(x
2
+1)
.
solution
P(X 1) =
_
1
p(x) dx =
1
_
1
1
x
2
+1
dx =
1
tan
1
x
=
1
4

2
_
=
3
4
Showthat p(x) =
1
4
e
x/2
+
1
6
e
x/3
isaprobabilitydensityon[0, ) andnditsmean.
71. Findaconstant C suchthat p(x) = Cx
3
e
x
2
isaprobabilitydensityandcomputep(0 X 1).
solution Werst ndtheindeniteintegral of p(x) usingintegrationby parts, withu = x
2
, v
= xe
x
2
, so that
u
= 2x andv =
1
2
e
x
2
:
_
Cx
3
e
x
2
dx = C
_
1
2
x
2
e
x
2
+
_
xe
x
2
dx
_
= C
_
1
2
x
2
e
x
2
1
2
e
x
2
_
=
C
2
e
x
2
(x
2
+1)
Todeterminetheconstant C, thevalueof theintegral ontheinterval [0, ) must be1:
1=
_

0
Cx
3
e
x
2
dx =
C
2
e
x/2
(x
2
+1)
0
=
C
2
_
lim
R
x
2
+1
e
x/2
1
_
=
C
2
sothat C = 2. Then
P(0 X 1) =
_
1
0
2x
3
e
x
2
dx = e
x
2
(x
2
+1)
1
0
= 12e
1
0.13212
Theinterval betweenpatientarrivalsinanemergencyroomisarandomvariablewithexponential densityfunction
p(t ) = 0.125e
0.125t
(t inminutes). What istheaveragetimebetweenpatient arrivals? What istheprobability of
twopatientsarrivingwithin3minutesof eachother?
73. Calculatethefollowingprobabilities, assumingthat X isnormallydistributedwithmean = 40and = 5.
(a) p(X 45) (b) p(0 X 40)
solution Let F bethestandardnormal cumulativedistributionfunction. ThenbyTheorem1inSection7.7,
(a)
p(X 45) = 1p(X 45) = 1F
_
4540
5
_
= 1F(1) 10.8413 0.1587
(b)
p(0 X 40) = p(X 40) p(X 0) = F
_
4040
5
_
F
_
040
5
_
= F(0) F(8) =
1
2
F(8)
1
2
0=
1
2
Note that p(X 40) is exactly
1
2
since 40 is the mean. Also, since 8 is so far to the left in the standard normal
distribution, theprobabilityof itsoccurrenceisquitesmall (approximately810
11
).
Accordingtokinetictheory, themoleculesof ordinarymatter areinconstant randommotion. TheenergyE of a
moleculeisarandomvariablewithdensity functionp(E) =
1
kT
e
E/(kT )
, whereT isthetemperature(inkelvins)
andk isBoltzmannsconstant. ComputethemeankineticenergyE intermsof k andT .
InExercises7584, determinewhether theimproper integral convergesand, if so, evaluateit.
75.
_

0
dx
(x +2)
2
solution
_

0
dx
(x +2)
2
= lim
R
_
R
0
dx
(x +2)
2
= lim
R
1
x +2
R
0
= lim
R
_
1
R +2
+
1
0+2
_
= lim
R
_
1
R +2
+
1
2
_
= 0+
1
2
=
1
2
.
_

4
dx
x
2/3
77.
_
4
0
dx
x
2/3
solution
_
4
0
dx
x
2/3
= lim
R0+
_
4
R
dx
x
2/3
= lim
R0+
3x
1/3
4
R
= lim
R0+
_
3 4
1/3
3 R
1/3
_
= 3
3
4.
_

9
dx
x
12/5
79.
_
0
dx
x
2
+1
solution
_
0
dx
x
2
+1
= lim
R
_
0
R
dx
x
2
+1
= lim
R
tan
1
x
0
R
= lim
R
_
tan
1
0tan
1
R
_
= lim
R
_
tan
1
R
_
=
_
2
_
=

2
.
_
9
e
4x
dx
81.
_
/2
0
cot d
solution
_
/2
0
cot d = lim
R0+
_
/2
R
cot d = lim
R0+
ln| sin|
/2
R
= lim
R0+
_
ln
_
sin
2
_
ln(sinR)
_
= lim
R0+
(ln1ln(sinR)) = lim
R0+
ln
_
1
sinR
_
= .
Weconcludethat theimproper integral diverges.
_

1
dx
(x +2)(2x +3)
83.
_

0
(5+x)
1/3
dx
solution
_

0
(5+x)
1/3
dx = lim
R
_
R
0
(5+x)
1/3
dx = lim
R
3
2
(5+x)
2/3
R
0
= lim
R
_
3
2
(5+R)
2/3
3
2
5
2/3
_
= .
Weconcludethat theimproper integral diverges.
_
5
2
(5x)
1/3
dx
InExercises8590, usetheComparisonTest todeterminewhether theimproper integral convergesor diverges.
85.
_

8
dx
x
2
4
solution For x 8,
1
2
x
2
4, sothat
1
2
x
2
4
1
2
x
2
x
2
4
and
1
x
2
4
2
x
2
.
Now,
_

1
dx
x
2
converges, so
_

8
2
x
2
dx alsoconverges. Therefore, bythecomparisontest,
_

8
dx
x
2
4
converges.
_

8
(sin
2
x)e
x
dx
87.
_

3
dx
x
4
+cos
2
x
solution For x 1, wehave
1
x
4
+cos
2
x
1
x
4
.
Since
_

1
dx
x
4
converges, the Comparison Test guarantees that
_

1
dx
x
4
+cos
2
x
also converges. The integral
_
3
1
dx
x
4
+cos
2
x
has anitevalue(noticethat x
4
+ cos
2
x = 0) henceweconcludethat theintegral
_

3
dx
x
4
+cos
2
x
alsoconverges.
_

1
dx
x
1/3
+x
2/3
89.
_
1
0
dx
x
1/3
+x
2/3
solution For 0 x 1,
x
1/3
+x
2/3
x
1/3
so
1
x
1/3
+x
2/3

1
x
1/3
.
Now,
_
1
0
x
1/3
dx converges.Therefore, bytheComparisonTest, theimproperintegral
_
1
0
dx
x
1/3
+x
2/3
alsoconverges.
_

0
e
x
3
dx
91. Calculatethevolumeof theinnitesolidobtainedby rotatingtheregionunder y = (x
2
+ 1)
2
for 0 x <
about they-axis.
solution UsingtheShell Method, thevolumeof theinnitesolidobtainedbyrotatingtheregionunder thegraphof
y =
_
x
2
+1
_
2
over theinterval [0, ) about they-axisis
V = 2
_

0
x
_
x
2
+1
_
2
dx.
Now,
_

0
x
_
x
2
+1
_
2
dx = lim
R
_
R
0
x dx
_
x
2
+1
_
2
Wesubstitutet = x
2
+1, dt = 2x dx. Thenewlimitsof integrationaret = 1andt = R
2
+1. Thus,
_
R
0
x dx
_
x
2
+1
_
2
=
_
R
2
+1
1
1
2
dt
t
2
=
1
2t
R
2
+1
1
=
1
2
_
1
1
R
2
+1
_
.
Takingthelimit asR yields:
_

0
x dx
_
x
2
+1
_
2
= lim
R
1
2
_
1
1
R
2
+1
_
=
1
2
(10) =
1
2
.
Therefore,
V = 2
1
2
= .
LetR betheregionunder thegraphof y = (x +1)
1
for 0 x < . Whichof thefollowingquantitiesisnite?
(a) Theareaof R
(b) Thevolumeof thesolidobtainedbyrotatingR about thex-axis
(c) Thevolumeof thesolidobtainedbyrotatingR about they-axis
93. Showthat
_
0
x
n
e
x
2
dx convergesfor all n > 0. Hint: Firstobservethatx
n
e
x
2
< x
n
e
x
for x > 1. Thenshow
that x
n
e
x
< x
2
for x sufcientlylarge.
2
> x; hencee
x
2
> e
x
, and0< e
x
2
< e
x
. Therefore, for x > 1thefollowinginequality
holds:
x
n+2
e
x
2
< x
n+2
e
x
.
Now, usingLHpitalsRulen +2times, wend
lim
x
x
n+2
e
x
= lim
x
x
n+2
e
x
= lim
x
(n +2)x
n+1
e
x
= lim
x
(n +2)(n +1)x
n
e
x
= = lim
x
(n +2)!
e
x
= 0.
Therefore,
lim
x
x
n+2
e
x
2
= 0
bytheSqueezeTheorem, andthereexistsanumber R > 1suchthat, for all x > R:
x
n+2
e
x
2
< 1 or x
n
e
x
2
< x
2
.
Finally, write
_

0
x
n
e
x
2
dx =
_
R
0
x
n
e
x
2
dx +
_

R
x
n
e
x
2
dx.
Therst integral ontheright-handsidehasnitevaluesincetheintegrandisacontinuousfunction. Thesecondintegral
convergessinceontheinterval of integration, x
n
e
x
2
< x
2
andweknowthat
_

R
x
2
dx =
_

R
dx
x
2
converges. We
concludethat theintegral
_

0
x
n
e
x
2
dx converges.
ComputetheLaplacetransformLf (s) of thefunctionf (x) = x for s > 0. SeeExercises8689inSection8.6
for thedenitionof Lf (s).
95. ComputetheLaplacetransformLf (s) of thefunctionf (x) = x
2
e
x
for s > .
solution TheLaplacetransformisthefollowingintegral:
L
_
x
2
e
x
_
(s) =
_

0
x
2
e
x
e
sx
dx =
_

0
x
2
e
(s)x
dx = lim
R
_
R
0
x
2
e
(s)x
dx.
Wecomputethedeniteintegral usingIntegrationbyPartswithu = x
2
, v
= e
(s)x
. Thenu
= 2x, v =
1
s
e
(s)x
and
_
R
0
x
2
e
(s)x
dx =
1
s
x
2
e
(s)x
R
x=0
_
R
0
2x
1
s
e
(s)x
dx
=
1
s
R
2
e
(s)R
2
s
_
R
0
xe
(s)x
dx.
Wecomputetheresultingintegral usingIntegrationbyPartsagain, thistimewithu = x andv
= e
(s)x
. Thenu
= 1,
v =
1
s
e
(s)x
and
_
R
0
xe
(s)x
dx = x
1
s
e
(s)x
R
x=0
1
s
_
R
0
e
(s)x
dx =
_
x
s
e
(s)x
1
( s)
2
e
(s)x
_
R
x=0
=
R
s
e
(s)R
1
( s)
2
_
e
(s)R
e
0
_
=
1
( s)
2
1
( s)
2
e
(s)R
+
R
s
e
(s)R
.
Thus,
_
R
0
x
2
e
(s)x
dx =
1
s
R
2
e
(s)R
2
s
_
1
( s)
2
1
( s)
2
e
(s)R
+
R
s
e
(s)R
_
=
1
s
R
2
e
(s)R
2
( s)
3
+
2
( s)
3
e
(s)R
2R
( s)
2
e
(s)R
,
and
L
_
x
2
e
x
_
(s) =
2
(s )
3
1
s
lim
R
R
2
e
(s)R
2
(s )
3
lim
R
e
(s)R
2
(s )
2
lim
R
Re
(s)R
.
Now, sinces > , lim
R
e
(s)R
= 0. WeuseLHpitalsRuletocomputetheother twolimits:
lim
R
Re
(s)R
= lim
R
R
e
(s)R
= lim
R
1
(s )e
(s)R
= 0;
lim
R
R
2
e
(s)R
= lim
R
R
2
e
(s)R
= lim
R
2R
(s )e
(s)R
= lim
R
2
(s )
2
e
(s)R
= 0.
Finally,
L
_
x
2
e
x
_
(s) =
2
(s )
3
000=
2
(s )
3
.
Estimate
_
5
2
f (x) dx bycomputingT
2
, M
3
, T
6
, andS
6
for afunctionf (x) takingonthevaluesinthefollowing
table:
x 2 2.5 3 3.5 4 4.5 5
f (x)
1
2
2 1 0
3
2
4 2
97. Statewhether theapproximationM
N
or T
N
islarger or smaller thantheintegral.
(a)
_

0
sinx dx (b)
_
2
sinx dx
(c)
_
8
1
dx
x
2
(d)
_
5
2
lnx dx
solution
(a) Becausef (x) = sinx isconcavedownontheinterval [0, ],
T
N

_

0
sinx dx M
N
;
that is, T
N
issmaller andM
N
islarger thantheintegral.
(b) Ontheinterval [, 2], thefunctionf (x) = sinx isconcaveup, therefore
M
N

_
2
sinx dx T
N
;
that is, M
N
issmaller andT
N
(c) Thefunctionf (x) =
1
x
2
isconcaveupontheinterval [1, 8]; therefore,
M
N

_
8
1
dx
x
2
T
N
;
that is, M
N
issmaller andT
N
(d) Theintegrandy = lnx isconcavedownontheinterval [2, 5]; hence,
T
N

_
5
2
lnx dx M
N
;
that is, T
N
issmaller andM
N
Therainfall rate(in inches per hour) was measured hourly during a10-hour thunderstormwith thefollowing
results:
0, 0.41, 0.49, 0.32, 0.3, 0.23,
0.09, 0.08, 0.05, 0.11, 0.12
UseSimpsonsRuletoestimatethetotal rainfall duringthe10-hour period.
InExercises99104, computethegivenapproximationtotheintegral.
99.
_
1
0
e
x
2
dx, M
5
solution Dividetheinterval [0, 1] into5subintervalsof lengthx =
10
5
=
1
5
, withmidpointsc
1
=
1
10
, c
2
=
3
10
,
c
3
=
1
2
, c
4
=
7
10
, andc
5
=
9
10
. Then
M
5
= x
_
f
_
1
10
_
+f
_
3
10
_
+f
_
1
2
_
+f
_
7
10
_
+f
_
9
10
__
=
1
5
_
e
(1/10)
2
+e
(3/10)
2
+e
(1/2)
2
+e
(7/10)
2
+e
(9/10)
2
_
= 0.748053.
_
4
2
_
6t
3
+1dt , T
3
101.
_
/2
/4
sin d, M
4
solution Dividetheinterval
_
4
,

2
_
into4subintervalsof lengthx =
4
4
=

16
withmidpoints
9
32
,
11
32
,
13
32
,
and
15
32
. Then
M
4
= x
_
f
_
9
32
_
+f
_
11
32
_
+f
_
13
32
_
+f
_
15
32
__
=

16
__
sin
9
32
+
_
sin
11
32
+
_
sin
13
32
+
_
sin
15
32
_
= 0.744978.
_
4
1
dx
x
3
+1
, T
6
103.
_
1
0
e
x
2
dx, S
4
solution Dividetheinterval [0, 1] into4subintervalsof lengthx =
1
4
withendpoints0,
1
4
,
1
2
,
3
4
, 1. Then
S
6
=
1
3
x
_
f (0) +4f
_
1
4
_
+2f
_
1
2
_
+4f
_
3
4
_
+f (1)
_
=
1
3
1
4
_
e
0
2
+4e
(1/4)
2
+2e
(1/2)
2
+4e
(3/4)
2
+e
1
2
_
= 0.746855.
_
9
5
cos(x
2
) dx, S
8
105. ThefollowingtablegivestheareaA(h) of ahorizontal crosssectionof apondatdepthh. UsetheTrapezoidal Rule
toestimatethevolumeV of thepond(Figure1).
h (ft) A(h) (acres) h (ft) A(h) (acres)
0 2.8 10 0.8
2 2.4 12 0.6
4 1.8 14 0.2
6 1.5 16 0.1
8 1.2 18 0
Areaof horizontal
cross section isA(h)
h
FIGURE 1
solution Thevolumeof thepondisthefollowingintegral:
V =
_
18
0
A(h)dh
Weapproximatetheintegral usingthetrapezoidal approximationT
9
. Theinterval of depth[0, 18] isdividedto9subin-
tervalsof lengthx = 2withendpoints0, 2, 4, 6, 8, 10, 12, 14, 16, 18. Thus,
V T
9
=
1
2
2(2.8+2 2.4+2 1.8+2 1.5+2 1.2+2 0.8+2 0.6+2 0.2+2 0.1+0)
= 20acre ft = 871,200ft
3
,
wherewehaveusedthefact that 1acre= 43,560ft
2
.
Supposethat thesecondderivativeof thefunctionA(h) inExercise105satises |A
(h)| 1.5. Usetheerror

boundtondthemaximumpossibleerror inyour estimateof thevolumeV of thepond.
107. Findaboundfor theerror
M
16
_
3
1
x
3
dx
.
solution TheError Boundfor theMidpoint Rulestatesthat
M
N

_
b
a
f (x) dx
K
2
(b a)
3
24N
2
,
whereK
2
isanumber suchthat
(x)
K
2
for all x [1, 3]. Hereb a = 31= 2andN = 16. Therefore,
M
16
_
3
1
x
3
dx
K
2
2
3
24 16
2
=
K
2
768
.
TondK
2
, wedifferentiatef (x) = x
3
twice:
f
(x) = 3x
2
and f
(x) = 6x.
Ontheinterval [1, 3] wehave
(x)
= 6x 6 3= 18; hence, wemaytakeK

2
= 18. Thus,
M
16
_
3
1
x
3
dx
18
768
=
3
128
= 0.0234375.
Let f (x) = sin(x
3
). Findaboundfor theerror
T
24
_
/2
0
f (x) dx
Hint: FindaboundK
2
for |f
(x)| byplottingf
(x) withagraphingutility.
109. Findavalueof N suchthat
M
N

_
/4
0
tanx dx
10
4
solution To usetheError Boundwemust ndthesecondderivativeof f (x) = tanx. Wedifferentiatef twiceto
obtain:
f
(x) = sec
2
x
f
(x) = 2secx tanx =

2sinx
cos
2
x
For 0 x

4
, wehavesinx sin
4
=
1
2
andcosx
1
2
or cos
2
x
1
2
. Therefore, for 0 x

4
wehave:
f
(x) =
2sinx
cos
2
x
2
1
2
1
2
= 2
2.
UsingtheError Boundwithb =

4
, a = 0andK
2
= 2
2wehave:
M
N

_
/4
0
tanx dx
2
_
4
0
_
3
24N
2
=

3
2
768N
2
.
Wemust chooseavalueof N suchthat:
2
768N
2
10
4
N
2
10
4
2
3
768
N 23.9
Thesmallest integer that isneededtoobtaintherequiredprecisionisN = 24.
Findavalueof N suchthat S
N
approximates
_
5
2
x
1/4
dx withanerror of at most 10
2
(but donot calculate
S
N
).
9 FURTHER APPLICATIONS
OF THE INTEGRAL AND
TAYLOR POLYNOMIALS
9.1 Arc Length and Surface Area
1. Whichintegral representsthelengthof thecurvey = cosx between0and?
_

0
_
1+cos
2
x dx,
_

0
_
1+sin
2
x dx
solution Let y = cosx. Theny
= sinx, and1+ (y
)
2
= 1+ sin
2
x. Thus, thelengthof thecurvey = cosx
between0and is
_

0
_
1+sin
2
x dx.
2. Usetheformulafor arc lengthtoshowthat for any constant C, thegraphs y = f (x) andy = f (x) + C havethe
samelengthover everyinterval [a, b]. Explaingeometrically.
solution Thegraphof y = f (x) +C isavertical translationof thegraphof y = f (x); hence, thetwographsshould
havethesamearclength. Wecanexplicitlyestablishthisasfollows:
lengthof y = f (x) +C =
_
b
a
_
1+
_
d
dx
(f (x) +C)
_
2
dx =
_
b
a
_
1+[f
(x)]
2
dx = lengthof y = f (x).
3. Usetheformulafor arclengthtoshowthat thelengthof agraphover [1, 4] cannot belessthan3.
solution Notethat f
(x)
2
0, so that
_
1+[f
(x)]
2

1 = 1. Then thearc length of thegraph of f (x) on
[1, 4] is
_
4
1
_
1+[f
(x)]
2
dx
_
4
1
1dx = 3
Exercises
1. Expressthearclengthof thecurvey = x
4
betweenx = 2andx = 6asanintegral (but donot evaluate).
solution Let y = x
4
. Theny
= 4x
3
and
s =
_
6
2
_
1+(4x
3
)
2
dx =
_
6
2
_
1+16x
6
dx.
Expressthearclengthof thecurvey = tanx for 0 x

4
asanintegral (but donot evaluate).
3. Findthearclengthof y =
1
12
x
3
+x
1
for 1 x 2. Hint: Showthat 1+(y
)
2
=
_
1
4
x
2
+x
2
_
2
.
solution Let y =
1
12
x
3
+x
1
. Theny
=
x
2
4
x
2
, and
(y
)
2
+1=
_
x
2
4
x
2
_
2
+1=
x
4
16
1
2
+x
4
+1=
x
4
16
+
1
2
+x
4
=
_
x
2
4
+x
2
_
2
.
555
556 C HA P T E R 9 FURTHER APPLICATIONS OF THE INTEGRAL AND TAYLOR POLYNOMIALS
Thus,
s =
_
2
1
_
1+(y
)
2
dx =
_
2
1
_
_
x
2
4
+
1
x
2
_
2
dx =
_
2
1
x
2
4
+
1
x
2
dx
=
_
2
1
_
x
2
4
+
1
x
2
_
dx since
x
2
4
+
1
x
2
> 0
=
_
x
3
12
1
x
_
2
1
=
13
12
.
Findthearclengthof y =
_
x
2
_
4
+
1
2x
2
over [1, 4]. Hint: Showthat 1+(y
)
2
isaperfect square.
In Exercises 510, calculate the arc length over the given interval.
5. y = 3x +1, [0, 3]
solution Let y = 3x +1. Theny
= 3, ands =
_
3
0
1+9dx = 3
10.
y = 93x, [1, 3]
7. y = x
3/2
, [1, 2]
solution Let y = x
3/2
. Theny
=
3
2
x
1/2
, and
s =
_
2
1
_
1+
9
4
x dx =
8
27
_
1+
9
4
x
_
3/2
2
1
=
8
27
_
_
11
2
_
3/2
_
13
4
_
3/2
_
=
1
27
_
22
2213
13
_
.
y =
1
3
x
3/2
x
1/2
, [2, 8]
9. y =
1
4
x
2
1
2
lnx, [1, 2e]
solution Let y =
1
4
x
2
1
2
lnx. Then
y
=
x
2
1
2x
,
and
1+(y
)
2
= 1+
_
x
2
1
2x
_
2
=
x
2
4
+
1
2
+
1
4x
2
=
_
x
2
+
1
2x
_
2
.
Hence,
s =
_
2e
1
_
1+(y
)
2
dx =
_
2e
1
_
_
x
2
+
1
2x
_
2
dx =
_
2e
1
x
2
+
1
2x
dx
=
_
2e
1
_
x
2
+
1
2x
_
dx since
x
2
+
1
2x
> 0on[1, 2e]
=
_
x
2
4
+
1
2
lnx
_
2e
1
= e
2
+
ln2
2
+
1
4
.
y = ln(cosx),
_
0,

4
_
In Exercises 1114, approximate the arc length of the curve over the interval using the Trapezoidal Rule T
N
, the Midpoint
Rule M
N
, or Simpsons Rule S
N
as indicated.
11. y =
1
4
x
4
, [1, 2], T
5
solution Let y =
1
4
x
4
. Then
1+(y
)
2
= 1+(x
3
)
2
= 1+x
6
.
Therefore, thearclengthover [1, 2] is
_
2
1
_
1+x
6
dx.
Now, let f (x) =
_
1+x
6
. Withn = 5,
x =
21
5
=
1
5
and {x
i
}
5
i=0
=
_
1,
6
5
,
7
5
,
8
5
,
9
5
, 2
_
.
S E C T I ON 9.1 Arc Length and Surface Area 557
UsingtheTrapezoidal Rule,
_
2
1
_
1+x
6
dx
x
2
f (x
0
) +2
4
i=1
f (x
i
) +f (x
5
)
= 3.957736.
Thearclengthisapproximately3.957736units.
y = sinx,
_
0,

2
_
, M
8
13. y = x
1
, [1, 2], S
8
solution Let y = x
1
. Theny
= x
2
and
1+(y
)
2
= 1+
1
x
4
.
Therefore, thearclengthover [1, 2] is
_
2
1
_
1+
1
x
4
dx.
Now, let f (x) =
_
1+
1
x
4
. Withn = 8,
x =
21
8
=
1
8
and {x
i
}
8
i=0
=
_
1,
9
8
,
5
4
,
11
8
,
3
2
,
13
8
,
7
4
,
15
8
, 2
_
.
UsingSimpsonsRule,
_
2
1
_
1+
1
x
4
dx
x
3
f (x
0
) +4
4
i=1
f (x
2i1
) +2
3
i=1
f (x
2i
) +f (x
8
)
= 1.132123.
Thearclengthisapproximately1.132123units.
y = e
x
2
, [0, 2], S
8
15. Calculatethelengthof theastroidx
2/3
+y
2/3
= 1(Figure11).
y
1
1
1
1
x
FIGURE 11 Graphof x
2/3
+y
2/3
= 1.
solution Wewill calculatethearclengthof theportionof theasteroidintherstquadrantandthenmultiplyby4. By
implicit differentiation
2
3
x
1/3
+
2
3
y
1/3
y
= 0,
so
y
=
x
1/3
y
1/3
=
y
1/3
x
1/3
.
Thus
1+(y
)
2
= 1+
y
2/3
x
2/3
=
x
2/3
+y
2/3
x
2/3
=
1
x
2/3
,
and
s =
_
1
0
1
x
1/3
dx =
3
2
.
Thetotal arclengthistherefore4
3
2
= 6.
Showthat thearclengthof theasteroidx
2/3
+y
2/3
= a
2/3
(for a > 0) isproportional toa.
17. Let a, r > 0. Showthat thearclengthof thecurvex
r
+y
r
= a
r
for 0 x a isproportional toa.
solution Usingimplicit differentiation, wendy
= (x/y)
r1
and
1+(y
)
2
= 1+(x/y)
2r2
=
x
2r1
+y
2r2
y
2r2
=
x
2r2
+(a
r
x
r
)
22/r
(a
r
x
r
)
22/r
.
Thearclengthisthen
s =
_
a
0
_
x
2r2
+(a
r
x
r
)
22/r
(a
r
x
r
)
22/r
dx.
Usingthesubstitutionx = au, weobtain
s = a
_
1
0
_
u
2r2
+(1u
r
)
22/r
(1u
r
)
22/r
du,
wheretheintegral isindependent of a.
Findthearclengthof thecurveshowninFigure12.
19. Findthevalueof a suchthat thearclengthof thecatenary y = coshx for a x a equals10.
= sinhx and
1+(y
)
2
= 1+sinh
2
x = cosh
2
x.
Thus,
s =
_
a
a
coshx dx = sinh(a) sinh(a) = 2sinha.
Settingthisexpressionequal to10andsolvingfor a yieldsa = sinh
1
(5) = ln(5+
26).
Calculatethearclengthof thegraphof f (x) = mx +r over [a, b] intwoways: usingthePythagoreantheorem
(Figure13) andusingthearclengthintegral.
21. Showthat thecircumferenceof theunit circleisequal to
2
_
1
1
dx
_
1x
2
(animproper integral)
Evaluate, thusverifyingthat thecircumferenceis2.
solution Notethecircumferenceof theunit circleis twicethearc lengthof theupper half of thecurvedenedby
x
2
+y
2
= 1. Thus, let y =
_
1x
2
. Then
y
=
x
_
1x
2
and 1+(y
)
2
= 1+
x
2
1x
2
=
1
1x
2
.
Finally, thecircumferenceof theunit circleis
2
_
1
1
dx
_
1x
2
= 2sin
1
x
1
1
= () = 2.
Generalizetheresult of Exercise21toshowthat thecircumferenceof thecircleof radiusr is2r.
23. Calculatethearclengthof y = x
2
over [0, a]. Hint: Usetrigonometricsubstitution. Evaluatefor a = 1.
solution Let y = x
2
. Theny
= 2x and
s =
_
a
0
_
1+4x
2
dx.
Usingthesubstitution2x = tan, 2dx = sec
2
d, wend
s =
1
2
_
x=a
x=0
sec
3
d.
Next, usingareductionformulafor theintegral of sec
3
, weseethat
s =
_
1
4
sec tan +
1
4
ln| sec +tan|
_
x=a
x=0
=
_
1
2
x
_
1+4x
2
+
1
4
ln|
_
1+4x
2
+2x|
_
a
0
=
a
2
_
1+4a
2
+
1
4
ln|
_
1+4a
2
+2a|
Thus, whena = 1,
s =
1
2
5+
1
4
ln(
5+2) 1.478943.
Express thearc lengthof g(x) =

x over [0, 1] as adeniteintegral. Thenusethesubstitutionu =

x to
showthat thisarc lengthisequal tothearc lengthof x
2
over [0, 1] (but donot evaluatetheintegrals). Explainthis
result graphically.
25. Findthearclengthof y = e
x
over [0, a]. Hint: Trythesubstitutionu =
_
1+e
2x
followedbypartial fractions.
solution Let y = e
x
. Then1+(y
)
2
= 1+e
2x
, andthearclengthover [0, a] is
_
a
0
_
1+e
2x
dx.
Now, let u =
_
1+e
2x
. Then
du =
1
2
2e
2x
_
1+e
2x
dx =
u
2
1
u
dx
andthearclengthis
_
a
0
_
1+e
2x
dx =
_
x=a
x=0
u
u
u
2
1
du =
_
x=a
x=0
u
2
u
2
1
du =
_
x=a
x=0
_
1+
1
u
2
1
_
du
=
_
x=a
x=0
_
1+
1
2
1
u 1
1
2
1
u +1
_
du =
_
u +
1
2
ln(u 1)
1
2
ln(u +1)
_
x=a
x=0
=
_
_
1+e
2x
+
1
2
ln
__
1+e
2x
1
_
1+e
2x
+1
__
a
0
=
_
1+e
2a
+
1
2
ln
_
1+e
2a
1
_
1+e
2a
+1
2+
1
2
ln
1+
21
=
_
1+e
2a
+
1
2
ln
_
1+e
2a
1
_
1+e
2a
+1
2+ln(1+
2).
Showthat thearclengthof y = ln(f (x)) for a x b is
_
b
a
_
f (x)
2
+f
(x)
2
f (x)
dx
27. UseEq. (4) tocomputethearclengthof y = ln(sinx) for

4
x

2
.
solution Withf (x) = sinx, Eq. (4) yields
s =
_
/2
/4
_
sin
2
x +cos
2
x
sinx
dx =
_
/2
/4
cscx dx = ln(cscx cotx)
/2
/4
= ln1ln(
21) = ln
1
21
= ln(
2+1).
UseEq. (4) tocomputethearclengthof y = ln
_
e
x
+1
e
x
1
_
over [1, 3].
29. Showthat if 0 f
(x) 1for all x, thenthearc lengthof y = f (x) over [a, b] isat most
2(b a). Showthat

for f (x) = x, thearclengthequals
2(b a).
solution If 0 f
(x) 1for all x, then

s =
_
b
a
_
1+f
(x)
2
dx
_
b
a
1+1dx =
2(b a).
If f (x) = x, thenf
(x) = 1and
s =
_
b
a
1+1dx =
2(b a).
UsetheComparisonTheorem(Section5.2) toprovethatthearclengthof y = x
4/3
over [1, 2] isnotlessthan
5
3
.
31. Approximatethearc lengthof one-quarter of theunit circle(whichweknowis

2
) by computingthelengthof the
polygonal approximationwithN = 4segments(Figure14).
y
1 0.75 0.5 0.25
x
FIGURE 14 One-quarter of theunit circle
solution Withy =
_
1x
2
, thevepointsalongthecurveare
P
0
(0, 1), P
1
(1/4,
15/4), P
2
(1/2,
3/2), P
3
(3/4,
7/4), P
4
(1, 0)
Then
P
0
P
1
=
_
1
16
+
_
4
15
4
_
2
0.252009
P
1
P
2
=
_
1
16
+
_
2
15
4
_
2
0.270091
P
2
P
3
=
_
1
16
+
_
2
7
4
_
2
0.323042
P
3
P
4
=
_
1
16
+
7
16
0.707108
andthetotal approximatedistanceis1.552250whereas/2 1.570796.
A merchant intendstoproducespecialty carpetsintheshapeof theregioninFigure15, boundedby theaxes
andgraphof y = 1x
n
(unitsinyards). Assumethat material costs$50/yd
2
andthat it costs50L dollarstocut the
carpet, whereL isthelengthof thecurvedsideof thecarpet. Thecarpetcanbesoldfor 150A dollars, whereA isthe
carpetsarea. Usingnumerical integrationwithacomputer algebrasystem, ndthewholenumber n for whichthe
merchantsprotsaremaximal.
In Exercises 3340, compute the surface area of revolution about the x-axis over the interval.
33. y = x, [0, 4]
solution 1+(y
)
2
= 2sothat
SA = 2
_
4
0
x
2dx = 2
2
1
2
x
2
4
0
= 16
2
y = 4x +3, [0, 1]
35. y = x
3
, [0, 2]
solution 1+(y
)
2
= 1+9x
4
, sothat
SA = 2
_
2
0
x
3
_
1+9x
4
dx =
2
36
_
2
0
36x
3
_
1+9x
4
dx =

18
(1+9x
4
)
3/2
2
0
=

18
_
145
3/2
1
_
y = x
2
, [0, 4]
37. y = (4x
2/3
)
3/2
, [0, 8]
solution Let y = (4x
2/3
)
3/2
. Then
y
= x
1/3
(4x
2/3
)
1/2
,
and
1+(y
)
2
= 1+
4x
2/3
x
2/3
=
4
x
2/3
.
Therefore,
SA = 2
_
8
0
(4x
2/3
)
3/2
_
2
x
1/3
_
dx.
Usingthesubstitutionu = 4x
2/3
, du =
2
3
x
1/3
dx, wend
SA = 2
_
0
4
u
3/2
(3) du = 6
_
4
0
u
3/2
du =
12
5
u
5/2
4
0
=
384
5
.
y = e
x
, [0, 1]
39. y =
1
4
x
2
1
2
lnx, [1, e]
solution Wehavey
=
x
2

1
2x
, and
1+(y
)
2
= 1+
_
x
2
1
2x
_
2
= 1+
x
2
4

1
2
+
1
4x
2
=
x
2
4
+
1
2
+
1
4x
2
=
_
x
2
+
1
2x
_
2
.
Thus,
SA = 2
_
e
1
_
x
2
4

lnx
2
_
_
x
2
+
1
2x
_
dx = 2
_
e
1
x
3
8
+
x
8
x lnx
4

lnx
4x
dx
= 2
_
x
4
32
+
x
2
16
x
2
lnx
8
+
x
2
16
(lnx)
2
8
_
e
1
= 2
_
e
4
32
+
e
2
16
e
2
8
+
e
2
16
1
8
_
1
32
+
1
16
+0+
1
16
0
_
_
= 2
_
e
4
32
1
8
1
32
1
16
1
16
_
=

16
(e
4
9)
y = sinx, [0, ]
In Exercises 4144, use a computer algebra system to nd the approximate surface area of the solid generated
by rotating the curve about the x-axis.
41. y = x
1
, [1, 3]
solution
SA = 2
_
3
1
1
x
_
1+
_
1
x
2
_
2
dx = 2
_
3
1
1
x
_
1+
1
x
4
dx 7.603062807
usingMaple.
y = x
4
, [0, 1]
43. y = e
x
2
/2
, [0, 2]
solution
SA = 2
_
2
0
e
x
2
/2
_
1+(xe
x
2
/2
)
2
dx = 2
_
2
0
e
x
2
/2
_
1+x
2
e
x
2
dx 8.222695606
usingMaple.
y = tanx,
_
0,

4
_ 45. Findtheareaof thesurfaceobtainedbyrotatingy = coshx over [ln2, ln2] aroundthex-axis.
= sinhx, and
_
1+(y
)
2
=
_
1+sinh
2
x =
_
cosh
2
x = coshx.
Therefore,
SA = 2
_
ln2
ln2
cosh
2
x dx =
_
ln2
ln2
(1+cosh2x) dx =
_
x +
1
2
sinh2x
_
ln2
ln2
=
_
ln2+
1
2
sinh(2ln2) +ln2
1
2
sinh(2ln2)
_
= 2 ln2+ sinh(2ln2).
Wecansimplifythisanswer byrecognizingthat
sinh(2ln2) =
e
2ln2
e
2ln2
2
=
4
1
4
2
=
15
8
.
Thus,
SA = 2 ln2+
15
8
.
Showthat thesurfaceareaof aspherical capof height h andradiusR (Figure16) hassurfacearea2Rh.
47. Findthesurfaceareaof thetorusobtainedbyrotatingthecirclex
2
+(y b)
2
= a
2
about thex-axis(Figure17).
y
x
(0, b + a)
(0, b)
FIGURE 17 Torusobtainedbyrotatingacircleabout thex-axis.
solution y = b +
_
a
2
x
2
givesthetophalf of thecircleandy = b
_
a
2
x
2
givesthebottomhalf. Notethat
ineachcase,
1+(y
)
2
= 1+
x
2
a
2
x
2
=
a
2
a
2
x
2
.
Rotatingthetwohalvesof thecirclearoundthex-axisthenyields
SA = 2
_
a
a
(b +
_
a
2
x
2
)
a
_
a
2
x
2
dx +2
_
a
a
(b
_
a
2
x
2
)
a
_
a
2
x
2
dx
= 2
_
a
a
2b
a
_
a
2
x
2
dx = 4ba
_
a
a
1
_
a
2
x
2
dx
= 4ba sin
1
_
x
a
_
a
a
= 4ba
_
2

_
2
__
= 4
2
ba.
Showthat thesurfaceareaof aright circular coneof radius r andheight h is r
_
r
2
+h
2
. Hint: Rotatealine
y = mx about thex-axisfor 0 x h, wherem isdeterminedsuitablybytheradiusr.
49. Findthesurfaceareaof theellipsoidobtainedbyrotatingtheellipse
_
x
a
_
2
+
_
y
b
_
2
= 1about thex-axis.
solution Takingadvantageof symmetry, wecanndthesurfaceareaof theellipsoidby doublingthesurfacearea
obtainedby rotatingtheportionof theellipseintherst quadrant about thex-axis. Theequationfor theportionof the
ellipseintherst quadrant is
y =
b
a
_
a
2
x
2
.
Thus,
1+(y
)
2
= 1+
b
2
x
2
a
2
(a
2
x
2
)
=
a
4
+(b
2
a
2
)x
2
a
2
(a
2
x
2
)
,
and
SA = 4
_
a
0
b
a
_
a
2
x
2
_
a
4
+(b
2
a
2
)x
2
a
_
a
2
x
2
dx = 4b
_
a
0
_
1+
_
b
2
a
2
a
4
_
x
2
dx.
Wenowconsider twocases. If b
2
> a
2
, thenwemakethesubstitution
_
b
2
a
2
a
2
x = tan, dx =
a
2
_
b
2
a
2
sec
2
d,
andndthat
SA = 4b
a
2
_
b
2
a
2
_
x=a
x=0
sec
3
d = 2b
a
2
_
b
2
a
2
(sec tan +ln| sec +tan|)
x=a
x=0
=
2bx
_
1+
_
b
2
a
2
a
4
_
x
2
+2b
a
2
_
b
2
a
2
ln
_
1+
_
b
2
a
2
a
4
_
x
2
+
_
b
2
a
2
a
2
x
a
0
= 2b
2
+2b
a
2
_
b
2
a
2
ln
_
b
a
+
_
b
2
a
2
a
_
.
Ontheother hand, if a
2
> b
2
, thenwemakethesubstitution
_
a
2
b
2
a
2
x = sin, dx =
a
2
_
a
2
b
2
cos d,
andndthat
SA = 4b
a
2
_
a
2
b
2
_
x=a
x=0
cos
2
d = 2b
a
2
_
a
2
b
2
( +sin cos)
x=a
x=0
=
2bx
_
1
_
a
2
b
2
a
4
_
x
2
+2b
a
2
_
a
2
b
2
sin
1
__
a
2
b
2
a
2
x
_
a
0
= 2b
2
+2b
a
2
_
a
2
b
2
sin
1
__
a
2
b
2
a
_
.
Observethat inbothcases, asa approachesb, thevalueof thesurfaceareaof theellipsoidapproaches4b
2
, thesurface
areaof asphereof radiusb.
Showthat if thearclengthof f (x) over [0, a] isproportional toa, thenf (x) must bealinear function.
51. Let L bethearclengthof theupper half of theellipsewithequation
y =
b
a
_
a
2
x
2
(Figure18) andlet =
_
1(b
2
/a
2
). Usesubstitutiontoshowthat
L = a
_
/2
/2
_
1
2
sin
2
d
Useacomputer algebrasystemtoapproximateL for a = 2, b = 1.
x
y
2 2
1
FIGURE 18 Graphof theellipsey =
1
2
_
4x
2
.
solution Let y =
b
a
_
a
2
x
2
. Then
1+(y
)
2
=
b
2
x
2
+a
2
(a
2
x
2
)
a
2
(a
2
x
2
)
and
s =
_
a
a
_
b
2
x
2
+a
2
(a
2
x
2
)
a
2
(a
2
x
2
)
dx.
Withthesubstitutionx = a sint , dx = a cost dt , a
2
x
2
= a
2
cos
2
t and
s = a
_
/2
/2
cost
_
a
2
b
2
sin
2
t +a
2
a
2
cos
2
t
a
2
(a
2
cos
2
t )
dt = a
_
/2
/2
_
b
2
sin
2
t
a
2
+cos
2
t dt
Because
=
_
1
b
2
a
2
,
2
= 1
b
2
a
2
wethenhave
1
2
sin
2
t = 1
_
1
b
2
a
2
_
sin
2
t = 1sin
2
t +
b
2
a
2
sin
2
t = cos
2
t +
b
2
a
2
sin
2
t
whichisthesameastheexpressionunder thesquareroot above. Substituting, weget
s = a
_
/2
/2
_
1
2
sin
2
t dt
Whena = 2andb = 1,
2
=
3
4
. Usingacomputer algebrasystemtoapproximatethevalueof thedeniteintegral, we
nds 4.84422.
Prove that the portion of a sphere of radius R seen by an observer located at a distance d above the North
PolehasareaA = 2dR
2
/(d +R). Hint: AccordingtoExercise46, thecaphassurfaceareais2Rh. Showthat
h = dR/(d +R) byapplyingthePythagoreanTheoremtothethreeright trianglesinFigure19.
53. Supposethat theobserver inExercise52moves off to innitythat is, d . What do youexpect the
limitingvalueof theobservedareatobe? Checkyour guessbycalculatingthelimit usingtheformulafor theareainthe
previousexercise.
solution Wewouldassumetheobservedsurfaceareawouldapproach2R
2
whichisthesurfaceareaof ahemisphere
of radiusR. Toverifythis, observe:
lim
d
SA = lim
d
2R
2
d
R +d
= lim
d
2R
2
1
= 2R
2
.
LetM bethetotal massof ametal rodintheshapeof thecurvey = f (x) over [a, b] whosemassdensity(x)
variesasafunctionof x. UseRiemannsumstojustifytheformula
M =
_
b
a
(x)
_
1+f
(x)
2
dx
55. Let f (x) beanincreasingfunctionon[a, b] andlet g(x) beitsinverse. Argueonthebasisof arclengththat
thefollowingequalityholds:
_
b
a
_
1+f
(x)
2
dx =
_
f (b)
f (a)
_
1+g
(y)
2
dy 5
Thenusethesubstitutionu = f (x) toproveEq. (5).
solution Sincethegraphsof f (x) andg(x) aresymmetricwithrespecttotheliney = x, thearclengthof thecurves
will beequal ontherespectivedomains. Sincethedomainof g is therangeof f , onf (a) to f (b), g(x) will havethe
samearclengthasf (x) ona tob. If g(x) = f
1
(x) andu = f (x), thenx = g(u) anddu = f
(x) dx. But

g
(u) =
1
f
(g(u))
=
1
f
(x)
f
(x) =
1
g
(u)
Nowsubstitutingu = f (x),
s =
_
b
a
_
1+f
(x)
2
dx =
_
f (b)
f (a)
_
1+
_
1
g
(u)
_
2
g
(u) du =
_
f (b)
f (a)
_
g
(u)
2
+1du
9.2 Fluid Pressure and Force
1. Howispressuredened?
solution Pressureisdenedasforceper unit area.
2. Fluidpressureisproportional todepth. What isthefactor of proportionality?
solution Thefactor of proportionalityistheweight densityof theuid, w = g, where isthemassdensityof the
uid.
3. Whenuidforceactsonthesideof asubmergedobject, inwhichdirectiondoesit act?
solution Fluidforceactsinthedirectionperpendicular tothesideof thesubmergedobject.
4. Whyisuidpressureonasurfacecalculatedusingthinhorizontal stripsrather thanthinvertical strips?
solution Pressuredependsonlyondepthanddoesnot changehorizontallyat agivendepth.
5. If athinplateissubmergedhorizontally, thentheuidforceononesideof theplateisequal topressuretimesarea.
Isthistrueif theplateissubmergedvertically?
solution Whenaplateis submergedvertically, thepressureis not constant alongtheplate, sotheuidforceis not
equal tothepressuretimesthearea.
Exercises
1. A boxof height6mandsquarebaseof side3missubmergedinapool of water. Thetopof theboxis2mbelowthe
surfaceof thewater.
(a) Calculatetheuidforceonthetopandbottomof thebox.
(b) WriteaRiemannsumthat approximates theuidforceonasideof thebox by dividingthesideinto N horizontal
stripsof thicknessy = 6/N.
(c) Towhichintegral doestheRiemannsumconverge?
(d) Computetheuidforceonasideof thebox.
solution
(a) At adepthof 2m, thepressureonthetopof theboxisgh = 10
3
9.8 2= 19,600Pa. Thetophasarea9m
2
, and
thepressureisconstant, sotheforceonthetopof theboxis19,600 9= 176,400N.Atadepthof 8m, thepressureonthe
bottomof theboxisgh = 10
3
9.8 8= 78,400Pa, sotheforceonthebottomof theboxis78,400 9= 705,600N.
(b) Lety
j
denotethedepthof thej
th
strip, forj = 1, 2, 3, . . . , N; thepressureatthisdepthis10
3
9.8 y
j
= 9800y
j
Pa.
Thestriphas thickness y mandlength3m, sohas area3y m
2
. Thus theforceonthestripis 29,400y
j
y N. Sum
over all thestripstoconcludethat theforceononesideof theboxisapproximately
F
N
j=1
29,400y
j
y.
S E C T I ON 9.2 Fluid Pressure and Force 565
(c) AsN , theRiemannsuminpart (b) convergestothedeniteintegral 29,400
_
8
2
y dy.
(d) Usingtheresult frompart (c), theuidforceononesideof theboxis
29,400
_
8
2
y dy = 14,700y
2
8
2
= 882,000N
A plateintheshapeof anisosceles trianglewithbase1mandheight 2mis submergedvertically inatank of
water sothat itsvertextouchesthesurfaceof thewater (Figure7).
(a) Showthat thewidthof thetriangleat depthy isf (y) =
1
2
y.
(b) Considerathinstripof thicknessy atdepthy. Explainwhytheuidforceonasideof thisstripisapproximately
equal tog
1
2
y
2
y.
(c) Writeanapproximationforthetotal uidforceF onasideof theplateasaRiemannsumandindicatetheintegral
towhichit converges.
(d) CalculateF.
3. Repeat Exercise2, but assumethat thetopof thetriangleislocated3mbelowthesurfaceof thewater.
solution
(a) Examinethegurebelow. Bysimilar triangles,
y 3
2
=
f (y)
1
sof (y) =
y 3
2
.
f (y)
y
3
(b) Thepressureatadepthof y feetisgy lb/ Pa, andtheareaof thestripisapproximatelyf (y) y =
1
2
(y 3)y m
2
.
Therefore, theuidforceonthisstripisapproximately
gy
_
1
2
(y 3)y
_
=
1
2
gy(y 3)y N.
(c) F
N
j=1
g
y
2
j
3y
j
2
y. AsN , theRiemannsumconvergestothedeniteintegral
g
2
_
5
3
(y
2
3y) dy.
(d) Usingtheresult of part (c),
F =
g
2
_
5
3
(y
2
3y) dy =
g
2
_
y
3
3

3y
2
2
_
5
3
=
9800
2
__
125
3

75
2
_
_
9
27
2
__
=
127,400
3
N.
TheplateR inFigure8, boundedby theparabolay = x
2
andy = 1, issubmergedvertically inwater (distance
inmeters).
(a) Showthatthewidthof R atheighty isf (y) = 2
y andtheuidforceonasideof ahorizontal stripof thickness

y at height y isapproximately(g)2y
1/2
(1y)y.
(b) WriteaRiemannsumthat approximatestheuidforceF onasideof R anduseit toexplainwhy
F = g
_
1
0
2y
1/2
(1y) dy
(c) CalculateF.
5. Let F betheuidforceonasideof asemicircular plateof radiusr meters, submergedverticallyinwater sothat its
diameter islevel withthewaterssurface(Figure9).
(a) Showthat thewidthof theplateat depthy is2
_
r
2
y
2
.
(b) CalculateF asafunctionof r usingEq. (2).
y
r
r
2 r
2
y
2
x
FIGURE 9
solution
(a) Placetheoriginat thecenter of thesemicircleandpoint thepositivey-axisdownward. Theequationfor theedgeof
thesemicircular plateisthenx
2
+y
2
= r
2
. At adepthof y, theplateextendsfromthepoint (
_
r
2
y
2
, y) ontheleft
tothepoint (
_
r
2
y
2
, y) ontheright. Thewidthof theplateat depthy isthen
_
r
2
y
2
_
r
2
y
2
_
= 2
_
r
2
y
2
.
(b) Withw = 9800N/m
3
,
F = 2w
_
r
0
y
_
r
2
y
2
dy =
19,600
3
(r
2
y
2
)
3/2
r
0
=
19,600r
3
3
N.
Calculatetheforceononesideof acircular platewithradius2m, submergedverticallyinatankof water sothat
thetopof thecircleistangent tothewater surface.
7. A semicircular plateof radiusr meters, orientedasinFigure9, issubmergedinwater sothat itsdiameter islocated
at adepthof m meters. Calculatetheuidforceononesideof theplateintermsof m andr.
solution Placetheoriginatthecenterof thesemicircularplatewiththepositivey-axispointingdownward. Thewater
surfaceisthenat y = m. Moreover, at locationy, thewidthof theplateis2
_
r
2
y
2
andthedepthisy +m. Thus,
F = 2g
_
r
0
(y +m)
_
r
2
y
2
dy.
Now,
_
r
0
y
_
r
2
y
2
dy =
1
3
(r
2
y
2
)
3/2
r
0
=
1
3
r
3
.
Geometrically,
_
r
0
_
r
2
y
2
dy
representstheareaof onequarter of acircleof radiusr, andthushasthevalue
r
2
4
. Bringingtheseresultstogether, we
ndthat
F = 2g
_
1
3
r
3
+

4
r
2
_
=
19,600
3
r
3
+4900mr
2
N.
A plateextendingfromdepthy = 2mto y = 5mis submergedinauidof density = 850kg/m
3
. The
horizontal widthof theplateat depthy isf (y) = 2(1+y
2
)
1
. Calculatetheuidforceononesideof theplate.
9. Figure10showsthewall of adamonawaterreservoir.UsetheTrapezoidal Ruleandthewidthanddepthmeasurements
intheguretoestimatetheuidforceonthewall.
Depth (ft)
20
0
600
900
1,100
1,400
1,650
1,800 (ft)
40
60
80
100
FIGURE 10
solution Let f (y) denotethewidthof thedamwall at depthy feet. Thentheforceonthedamwall is
F = w
_
100
0
yf (y) dy.
UsingtheTrapezoidal Ruleandthewidthanddepthmeasurementsinthegure,
F w
20
2
[0 f (0) +2 20 f (20) +2 40 f (40) +2 60 f (60) +2 80 f (80) +100 f (100)]
= 10w(0+66,000+112,000+132,000+144,000+60,000) = 321,250,000lb.
Calculatetheuidforceonasideof theplateinFigure11(A), submergedinwater.
11. Calculatetheuidforceonasideof theplateinFigure11(B), submergedinauidof massdensity = 800kg/m
3
.
solution Becausetheuidhasamassdensityof = 800kg/m
3
,
w = (800)(9.8) = 7840N/m
3
.
For depthsupto2meters, thewidthof theplateat depthy isy; for depthsfrom2metersto6meters, thewidthof the
plateisaconstant 2meters. Thus,
F = w
_
2
0
y(y) dy +w
_
6
2
2y dy = w
y
3
3
2
0
+ wy
2
6
2
=
8w
3
+32w =
104w
3
=
815,360
3
N.
S E C T I ON 9.2 Fluid Pressure and Force 567
Findtheuidforceonthesideof theplateinFigure12, submergedinauidof density = 1200kg/m
3
. The
topof theplaceislevel withtheuidsurface. Theedgesof theplatearethecurvesy = x
1/3
andy = x
1/3
.
13. Let R betheplateintheshapeof theregionunder y = sinx for 0 x

2
inFigure13(A). Findtheuidforceon
asideof R if itisrotatedcounterclockwiseby90
andsubmergedinauidof density1100kg/m
3
withitstopedgelevel
withthesurfaceof theuidasin(B).
1
(A) (B)
Fluid level
y
y = sin x
x
R
Fluid level
R
p
2
FIGURE 13
solution Placetheoriginatthebottomcorner of theplatewiththepositivey-axispointingupward. Theuidsurface
isthenat height y =

2
, andthehorizontal stripof theplateat height y isat adepthof

2
y andhasawidthof siny.
Now, usingintegrationbypartswend
F = g
_
/2
0
_
2
y
_
siny dy = g
_
2
y
_
cosy siny
_
/2
0
= g
_
2
1
_
= 1100 9.8
_
2
1
_
6153.184N.
Inthenotationof Exercise13, calculatetheuidforceonasideof theplateR if it isorientedasinFigure13(A).
YoumayneedtouseIntegrationbyPartsandtrigonometricsubstitution.
15. Calculatetheuidforceononesideof aplateintheshapeof regionA shownFigure14. Thewater surfaceis at
y = 1, andtheuidhasdensity = 900kg/m
3
.
y = ln x
1
y
1 e
x
A
B
FIGURE 14
solution Becausetheuidsurfaceisatheighty = 1, thehorizontal stripatheighty isatadepthof 1y. Moreover,
thisstriphasawidthof e e
y
. Thus,
F = g
_
1
0
(1y)(e e
y
) dy = eg
_
1
0
(1y) dy g
_
1
0
(1y)e
y
dy.
Now,
_
1
0
(1y) dy =
_
y
1
2
y
2
_
1
0
=
1
2
,
andusingintegrationbyparts
_
1
0
(1y)e
y
dy =
_
(1y)e
y
+e
y
_
1
0
= e 2.
Combiningtheseresults, wendthat
F = g
_
1
2
e (e 2)
_
= g
_
2
1
2
e
_
= 900 9.8
_
2
1
2
e
_
5652.37N.
Calculatetheuidforceononesideof theinniteplateB inFigure14, assumingtheuidhasdensity = 900
kg/m
3
.
17. Figure15(A) showsarampinclinedat 30
leadingintoaswimmingpool. Calculatetheuidforceontheramp.
4
6
Water surface
(A)
30
3
10
y
f (y)
Vertical
changey
(B)
Water surface
60
FIGURE 15
solution A horizontal stripat depthy haslength6andwidth
y
sin30
= 2y.
Thus,
F = 2g
_
4
0
6y dy = 96g.
If distancesareinfeet, theng = w = 62.5lb/ft
3
andF = 6000lb; if distancesareinmeters, theng = 9800N/m
3
andF = 940,800N.
Calculatetheuidforceononesideof theplate(anisoscelestriangle) showninFigure15(B).
19. ThemassiveThreeGorgesDamonChinasYangtzeRiver hasheight 185m(Figure16). Calculatetheforceonthe
dam, assumingthat thedamis atrapezoidof base2000mandupper edge3000m, inclinedat anangleof 55
to the
horizontal (Figure17).
FIGURE 16 ThreeGorgesDamon
theYangtzeRiver
2000 m
3000 m
185 m
55
FIGURE 17
solution Let y = 0beat thebottomof thedam, sothat thetopof thedamisat y = 185. Thenthewidthof thedam
at height y is2000+
1000y
185
. Thedamisinclinedat anangleof 55
tothehorizontal, sotheheight of ahorizontal stripis

y
sin55
1.221y
sothat theareaof suchastripis
1.221
_
2000+
1000y
185
_
y
Then
F = g
_
185
0
1.221y
_
2000+
1000y
185
_
dy = g
_
185
0
2442y +6.6y
2
dy = g(1221y
2
+2.2y
3
)
185
0
= 55,718,300g = 55,718,300 9800= 5.46039340010
11
N.
A squareplateof side3missubmergedinwater at aninclineof 30
withthehorizontal. Calculatetheuidforce
ononesideof theplateif thetopedgeof theplateliesat adepthof 6m.
21. ThetroughinFigure18islledwithcornsyrup, whoseweight densityis90lb/ft
3
. Calculatetheforceonthefront
sideof thetrough.
a
d
h
b
FIGURE 18
solution Placetheoriginalongthetopedgeof thetroughwiththepositivey-axispointingdownward. Thewidthof
thefront sideof thetroughvarieslinearly fromb wheny = 0toa wheny = h; thus, thewidthof thefront sideof the
troughat depthy feet isgivenby
b +
a b
h
y.
S E C T I ON 9.3 Center of Mass 569
Now,
F = w
_
h
0
y
_
b +
a b
h
y
_
dy = w
_
1
2
by
2
+
a b
3h
y
3
_
h
0
= w
_
b
6
+
a
3
_
h
2
= (15b +30a)h
2
lb.
Calculatetheuidpressureononeof theslantedsidesof thetroughinFigure18whenit islledwithcornsyrup
asinExercise21.
23. Theendof thetroughinFigure19isanequilateral triangleof side3. Assumethat thetroughislledwithwater to
height H. Calculatetheuidforceoneachsideof thetroughasafunctionof H andthelengthl of thetrough.
H
l
3
FIGURE 19
solution Place the origin at the lower vertex of the trough and orient the positive y-axis pointing upward. First,
consider thefacesat thefront andback endsof thetrough. A horizontal stripat height y hasalengthof
2y
3
andisat a
depthof H y. Thus,
F = w
_
H
0
(H y)
2y
3
dy = w
_
H
3
y
2
2
3
3
y
3
_
H
0
=
3
9
wH
3
.
For theslantedsides, wenotethat eachsidemakesanangleof 60
withthehorizontal. If welet denotethelengthof

thetrough, then
F =
2w
3
_
H
0
(H y) dy =
3
3
wH
2
.
A rectangular plateof side is submergedvertically inauidof density w, withits topedgeat depthh. Show
that if thedepthisincreasedbyanamount h, thentheforceonasideof theplateincreasesbywAh, whereA is
theareaof theplate.
25. Provethattheforceonthesideof arectangular plateof areaA submergedverticallyinauidisequal top
0
A, where
p
0
istheuidpressureat thecenter point of therectangle.
solution Let denotethelengthof thevertical sideof therectangle, x denotethelengthof thehorizontal sideof the
rectangle, andsupposethetopedgeof therectangleisat depthy = m. Thepressureat thecenter of therectangleisthen
p
0
= w
_
m+

2
_
,
andtheforceonthesideof therectangular plateis
F =
_
+m
m
wxy dy =
wx
2
_
( +m)
2
m
2
_
=
wx
2
( +2m) = Aw
_
2
+m
_
= Ap
0
.
If thedensityof auidvarieswithdepth, thenthepressureat depthy isafunctionp(y) (whichneednot equal
wy asinthecaseof constantdensity). UseRiemannsumstoarguethatthetotal forceF ontheatsideof asubmerged
object submergedverticallyisF =
_
b
a
f (y)p(y) dy, wheref (y) isthewidthof thesideat depthy.
9.3 Center of Mass
1. What arethex- andy-momentsof alaminawhosecenter of massislocatedat theorigin?
solution Becausethecenter of massislocatedat theorigin, it followsthat M
x
= M
y
= 0.
2. A thinplatehasmass3. What isthex-moment of theplateif itscenter of masshascoordinates(2, 7)?
solution Thex-momentof theplateistheproductof themassof theplateandthey-coordinateof thecenter of mass.
Thus, M
x
= 3(7) = 21.
3. Thecenter of massof alaminaof total mass5hascoordinates(2, 1). What arethelaminasx- andy-moments?
solution Thex-moment of theplateis theproduct of themass of theplateand they-coordinateof thecenter of
mass, whereas they-moment is theproduct of themass of theplateandthex-coordinateof thecenter of mass. Thus,
M
x
= 5(1) = 5, andM
y
= 5(2) = 10.
4. ExplainhowtheSymmetryPrincipleisusedtoconcludethatthecentroidof arectangleisthecenter of therectangle.
solution Becausearectangleis symmetric withrespect toboththevertical lineandthehorizontal linethroughthe
center of therectangle, theSymmetryPrincipleguaranteesthat thecentroidof therectanglemust liealongbothof these
lines. Theonlypoint incommontobothlinesof symmetryisthecenter of therectangle, sothecentroidof therectangle
must bethecenter of therectangle.
Exercises
1. Four particlesarelocatedat points(1, 1), (1, 2), (4, 0), (3, 1).
(a) FindthemomentsM
x
andM
y
andthecenter of massof thesystem, assumingthat theparticleshaveequal massm.
(b) Findthecenter of massof thesystem, assumingtheparticleshavemasses3, 2, 5, and7, respectively.
solution
(a) Becauseeachparticlehasmassm,
M
x
= m(1) +m(2) +m(0) +m(1) = 4m;
M
y
= m(1) +m(1) +m(4) +m(3) = 9m;
andthetotal massof thesystemis4m. Thus, thecoordinatesof thecenter of massare
_
M
y
M
,
M
x
M
_
=
_
9m
4m
,
4m
4m
_
=
_
9
4
, 1
_
.
(b) Withtheindicatedmassesof theparticles,
M
x
= 3(1) +2(2) +5(0) +7(1) = 14;
M
y
= 3(1) +2(1) +5(4) +7(3) = 46;
andthetotal massof thesystemis17. Thus, thecoordinatesof thecenter of massare
_
M
y
M
,
M
x
M
_
=
_
46
17
,
14
17
_
.
Findthecenter of massfor thesystemof particlesof masses4, 2, 5, 1locatedat (1, 2), (3, 2), (2, 1), (4, 0).
3. Point massesof equal sizeareplacedat theverticesof thetrianglewithcoordinates(a, 0), (b, 0), and(0, c). Show
that thecenter of massof thesystemof masseshascoordinates
_
1
3
(a +b),
1
3
c
_
.
solution Let eachparticlehavemassm. Thetotal massof thesystemisthen3m. andthemomentsare
M
x
= 0(m) +0(m) +c(m) = cm; and
M
y
= a(m) +b(m) +0(m) = (a +b)m.
Thus, thecoordinatesof thecenter of massare
_
M
y
M
,
M
x
M
_
=
_
(a +b)m
3m
,
cm
3m
_
=
_
a +b
3
,
c
3
_
.
Point massesof massm
1
, m
2
, andm
3
areplacedat thepoints(1, 0), (3, 0), and(0, 4).
(a) Supposethat m
1
= 6. Findm
2
suchthat thecenter of massliesonthey-axis.
(b) Supposethat m
1
= 6andm
2
= 4. Findthevalueof m
3
suchthat y
CM
= 2.
5. SketchthelaminaS of constantdensity = 3g/cm
2
occupyingtheregionbeneaththegraphof y = x
2
for0 x 3.
(a) UseEqs. (1) and(2) tocomputeM
x
andM
y
.
(b) Findtheareaandthecenter of massof S.
solution A sketchof thelaminaisshownbelow
y
x
8
4
6
2
0 2 3 1 1.5 2.5 0.5
(a) UsingEq. (2),
M
x
= 3
_
9
0
y(3
y) dy =
_
9y
2
2

6
5
y
5/2
_
9
0
=
729
10
.
UsingEq. (1),
M
y
= 3
_
3
0
x(x
2
) dx =
3x
4
4
3
0
=
243
4
.
(b) Theareaof thelaminais
A =
_
3
0
x
2
dx =
x
3
3
3
0
= 9cm
2
.
Withaconstant density of = 3g/cm
2
, themassof thelaminaisM = 27grams, andthecoordinatesof thecenter of
massare
_
M
y
M
,
M
x
M
_
=
_
243/4
27
,
729/10
27
_
=
_
9
4
,
27
10
_
.
UseEqs. (1) and(3) to ndthemoments andcenter of mass of thelaminaS of constant density = 2g/cm
2
occupyingtheregionbetweeny = x
2
andy = 9x over [0, 3]. SketchS, indicatingthelocationof thecenter of mass.
7. Findthemomentsandcenter of massof thelaminaof uniformdensity occupyingtheregionunderneathy = x
3
for 0 x 2.
solution Withuniformdensity,
M
x
=
1
2
_
2
0
(x
3
)
2
dx =
64
7
and M
y
=
_
2
0
x(x
3
) dx =
32
5
.
Themassof thelaminais
M =
_
2
0
x
3
dx = 4,
sothecoordinatesof thecenter of massare
_
M
y
M
,
M
x
M
_
=
_
8
5
,
16
7
_
.
CalculateM
x
(assuming = 1) for theregionunderneaththegraphof y = 1 x
2
for 0 x 1intwoways,
rst usingEq. (2) andthenusingEq. (3).
9. Let T bethetriangular laminainFigure17.
(a) Showthat thehorizontal cut at height y haslength4
2
3
y anduseEq. (2) tocomputeM
x
(with = 1).
(b) UsetheSymmetryPrincipletoshowthat M
y
= 0andndthecenter of mass.
y
2 2
6
x
FIGURE 17 Isoscelestriangle.
solution
(a) Theequationof thelinefrom(2, 0) to(0, 6) isy = 3x +6, so
x = 2
1
3
y.
Thelengthof thehorizontal cut at height y isthen
2
_
2
1
3
y
_
= 4
2
3
y,
and
M
x
=
_
6
0
y
_
4
2
3
y
_
dy = 24.
(b) Becausethetriangular laminais symmetric withrespect to they-axis, x
cm
= 0, whichimplies that M
y
= 0. The
total massof thelaminais
M = 2
_
2
0
(3x +6) dx = 12,
soy
cm
= 24/12. Finally, thecoordinatesof thecenter of massare(0, 2).
In Exercises 1017, nd the centroid of the region lying underneath the graph of the function over the given interval.
f (x) = 62x, [0, 3]
11. f (x) =

x, [1, 4]
solution Themomentsof theregionare
M
x
=
1
2
_
4
1
x dx =
15
4
and M
y
=
_
4
1
x
x dx =
62
5
.
Theareaof theregionis
A =
_
4
1
x dx =
14
3
,
sothecoordinatesof thecentroidare
_
M
y
A
,
M
x
A
_
=
_
93
35
,
45
56
_
.
f (x) = x
3
, [0, 1]
13. f (x) = 9x
2
, [0, 3]
M
x
=
1
2
_
3
0
(9x
2
)
2
dx =
324
5
and M
y
=
_
3
0
x(9x
2
) dx =
81
4
.
A =
_
3
0
(9x
2
) dx = 18,
_
M
y
A
,
M
x
A
_
=
_
9
8
,
18
5
_
.
f (x) = (1+x
2
)
1/2
, [0, 3]
15. f (x) = e
x
, [0, 4]
M
x
=
1
2
_
4
0
e
2x
dx =
1
4
_
1e
8
_
and M
y
=
_
4
0
xe
x
dx = e
x
(x +1)
4
0
= 15e
4
.
A =
_
4
0
e
x
dx = 1e
4
,
_
M
y
A
,
M
x
A
_
=
_
15e
4
1e
4
,
1e
8
4(1e
4
)
_
.
f (x) = lnx, [1, 2]
17. f (x) = sinx, [0, ]
M
x
=
1
2
_

0
sin
2
x dx =
1
4
(x sinx cosx)
0
=

4
; and
M
y
=
_

0
x sinx dx = (x cosx +sinx)
0
= .
A =
_

0
sinx dx = 2,
_
M
y
A
,
M
x
A
_
=
_
2
,
8
_
.
Calculatethemoments andcenter of mass of thelaminaoccupyingtheregionbetweenthecurves y = x and
y = x
2
for 0 x 1.
19. Sketchtheregionbetweeny = x + 4andy = 2 x for 0 x 2. Usingsymmetry, explainwhy thecentroidof
theregionliesontheliney = 3. Verifythisbycomputingthemomentsandthecentroid.
solution A sketchof theregionisshownbelow.
0.5
1
2
3
4
5
y
x
1.0 1.5 2.0
Theregionisclearlysymmetricabouttheliney = 3, soweexpectthecentroidof theregiontoliealongthisline. Wend
M
x
=
1
2
_
2
0
_
(x +4)
2
(2x)
2
_
dx = 24;
M
y
=
_
2
0
x ((x +4) (2x)) dx =
28
3
; and
A =
_
2
0
((x +4) (2x)) dx = 8.
Thus, thecoordinatesof thecentroidare
_
7
6
, 3
_
.
In Exercises 2025, nd the centroid of the region lying between the graphs of the functions over the given interval.
y = x, y =

x, [0, 1]
21. y = x
2
, y =

x, [0, 1]
M
x
=
1
2
_
1
0
(x x
4
) dx =
3
20
and M
y
=
_
1
0
x(
x x
2
) dx =
3
20
.
A =
_
1
0
(
x x
2
) dx =
1
3
,
_
9
20
,
9
20
_
.
Note: Thismakessense, sincethefunctionsareinversesof eachother. Thismakestheregionsymmetricwithrespect to
theliney = x. Thus, bythesymmetryprinciple, thecenter of massmust lieonthat line.
y = x
1
, y = 2x, [1, 2]
23. y = e
x
, y = 1, [0, 1]
M
x
=
1
2
_
1
0
(e
2x
1) dx =
e
2
3
4
and M
y
=
_
1
0
x(e
x
1) dx =
_
xe
x
e
x
1
2
x
2
_
1
0
=
1
2
.
A =
_
1
0
(e
x
1) dx = e 2,
_
1
2(e 2)
,
e
2
3
4(e 2)
_
.
y = lnx, y = x 1, [1, 3]
25. y = sinx, y = cosx, [0, /4]
M
x
=
1
2
_
/4
0
(cos
2
x sin
2
x) dx =
1
2
_
/4
0
cos2x dx =
1
4
; and
M
y
=
_
/4
0
x(cosx sinx) dx = [(x 1) sinx +(x +1) cosx]
/4
0
=

2
4
1.
A =
_
/4
0
(cosx sinx) dx =
21,
_

24
4(
21)
,
1
4(
21)
_
.
Sketchtheregionenclosedbyy = x +1, andy = (x 1)
2
, andnditscentroid.
27. Sketchtheregionenclosedbyy = 0, y = (x +1)
3
, andy = (1x)
3
, andnditscentroid.
solution A sketchof theregionisshownbelow.
1
1
1
y
x
Themomentsof theregionare
M
x
=
1
2
_
_
0
1
(x +1)
6
dx +
_
1
0
(1x)
6
dx
_
=
1
7
; and
M
y
= 0bytheSymmetryPrinciple.
A =
_
0
1
(x +1)
3
dx +
_
1
0
(1x)
3
dx =
1
2
,
_
0,
2
7
_
.
In Exercises 2832, nd the centroid of the region.
Tophalf of theellipse
_
x
2
_
2
+
_
y
4
_
2
= 1
29. Tophalf of theellipse
_
x
a
_
2
+
_
y
b
_
2
= 1for arbitrarya, b > 0
solution Theequationof thetophalf of theellipseis
y =
_
b
2
b
2
x
2
a
2
Thus,
M
x
=
1
2
_
a
a
_
b
2
b
2
x
2
a
2
2
dx =
2ab
2
3
.
By theSymmetry Principle, M
y
= 0. Theareaof theregionisone-half theareaof anellipsewithaxesof lengtha and
b; i.e.,
1
2
ab. Finally, thecoordinatesof thecentroidare
_
0,
4b
3
_
.
Semicircleof radiusr withcenter at theorigin
31. Quarter of theunit circlelyingintherst quadrant
solution By theSymmetry Principle, thecenter of massmust lieontheliney = x intherst quadrant. Therefore,
weneedonlycalculateoneof themomentsof theregion. Withy =
_
1x
2
, wend
M
y
=
_
1
0
x
_
1x
2
dx =
1
3
.
Theareaof theregionisone-quarter of theareaof aunit circle; i.e.,
1
4
. Thus, thecoordinatesof thecentroidare
_
4
3
,
4
3
_
.
Regionbetweeny = x(a x) andthex-axisfor a > 0.
33. Findthecentroidof theshadedregionof thesemicircleof radiusr inFigure18. What is thecentroidwhenr = 1
andh =
1
2
? Hint: Usegeometry rather thanintegrationtoshowthat thearea of theregionisr
2
sin
1
(
_
1h
2
/r
2
)
h
_
r
2
h
2
).
y
x
h r
FIGURE 18
solution Fromthesymmetry of theregion, it is obvious that thecentroid lies along they-axis. To determinethe
y-coordinateof thecentroid, wemust calculatethemoment about thex-axisandtheareaof theregion. Now, thelength
of thehorizontal cut of thesemicircleat height y is
_
r
2
y
2
_
r
2
y
2
_
= 2
_
r
2
y
2
.
Therefore, taking = 1, wend
M
x
= 2
_
r
h
y
_
r
2
y
2
dy =
2
3
(r
2
h
2
)
3/2
.
Observethat theregioniscomprisedof asector of thecirclewiththetrianglebetweenthetworadii removed. Theangle
of thesector is 2, where = sin
1
_
1h
2
/r
2
, so theareaof thesector is
1
2
r
2
(2) = r
2
sin
1
_
1h
2
/r
2
. The
trianglehasbase2
_
r
2
h
2
andheight h, sotheareaish
_
r
2
h
2
. Therefore,
Y
CM
=
M
x
A
=
2
3
(r
2
h
2
)
3/2
r
2
sin
1
_
1h
2
/r
2
h
_
r
2
h
2
.
Whenr = 1andh = 1/2, wend
Y
CM
=
2
3
(3/4)
3/2
sin
1

3
2

3
4
=
3
3
4 3
3
.
Sketchtheregionbetweeny = x
n
andy = x
m
for 0 x 1, wherem > n 0andndtheCOMof theregion.
Findapair (n, m) suchthat theCOM liesoutsidetheregion.
In Exercises 3537, use the additivity of moments to nd the COM of the region.
35. Isoscelestriangleof height 2ontopof arectangleof base4andheight 3(Figure19)
y
2 2
2
3
x
FIGURE 19
solution Theregionis symmetric withrespect to they-axis, so M
y
= 0by theSymmetry Principle. Themoment
about thex-axisfor therectangleis
M
rect
x
=
1
2
_
2
2
3
2
dx = 18,
whereasthemoment about thex-axisfor thetriangleis
M
triangle
x
=
_
5
3
y(102y) dy =
44
3
.
Thetotal moment about thex-axisisthen
M
x
= M
rect
x
+M
triangle
x
= 18+
44
3
=
98
3
.
Becausetheareaof theregionis12+4= 16, thecoordinatesof thecenter of massare
_
0,
49
24
_
.
Anicecreamconeconsistingof asemicircleontopof anequilateral triangleof side6(Figure20)
37. Three-quartersof theunit circle(removethepart inthefourthquadrant)
solution BytheSymmetryPrinciple, thecenter of massmust lieontheliney = x. Let region1bethesemicircle
abovethex-axisandregion2bethequarter circleinthethirdquadrant. Becauseregion1issymmetric withrespect to
they-axis, M
1
y
= 0bytheSymmetryPrinciple. Furthermore
M
2
y
=
_
0
1
x
_
1x
2
dx =
1
3
.
Thus, M
y
= M
1
y
+M
2
y
= 0+(
1
3
) =
1
3
. Theareaof theregionis3/4, sothecoordinatesof thecentroidare
_
4
9
,
4
9
_
.
Let S bethelaminaof massdensity = 1obtainedbyremovingacircleof radiusr fromthecircleof radius2r
showninFigure21. Let M
S
x
andM
S
y
denotethemomentsof S. Similarly, let M
big
y
andM
small
y
bethey-momentsof
thelarger andsmaller circles.
(a) UsetheSymmetryPrincipletoshowthat M
S
x
= 0.
(b) Showthat M
S
y
= M
big
y
M
small
y
usingtheadditivityof moments.
(c) FindM
big
y
andM
small
y
usingthefact that theCOM of acircleisitscenter. ThencomputeM
S
y
using(b).
(d) DeterminetheCOM of S.
39. FindtheCOM of thelaminasinFigure22obtainedbyremovingsquaresof side2fromasquareof side8.
8
2 2
8
FIGURE 22
solution Startwiththesquareontheleft. Placethesquaresothatthebottomleftcorner isat(0, 0). BytheSymmetry
Principle, thecenter of massmust lieonthelinesy = x andy = 8x. Theonlypoint incommontothesetwolinesis
(4, 4), sothecenter of massis(4, 4).
Nowconsider thesquareontheright. Placethesquareas above. By thesymmetry principle, x
cm
= 4. Now, let s1
denotethesquareintheupper left, s2denotethesquareintheupper right, andB denotetheentiresquare. Then
M
s1
x
=
1
2
_
2
0
(8
2
6
2
) dx = 28;
M
s2
x
=
1
2
_
8
6
(8
2
6
2
) dx = 28; and
M
B
x
=
1
2
_
8
0
8
2
dx = 256.
Bytheadditivityof moments, M
x
= 256 28 28= 200. Finally, theareaof theregionisA = 64 4 4= 56, so
thecoordinatesof thecenter of massare
_
4,
200
56
_
=
_
4,
25
7
_
.
A medianof atriangleisasegment joiningavertextothemidpoint of theoppositeside. Showthat thecentroid
of atriangleliesoneachof itsmedians, at adistancetwo-thirdsdownfromthevertex. Thenusethisfact toprove
that thethreemediansintersect at asinglepoint. Hint: Simplify thecalculationby assumingthat onevertex liesat
theoriginandanother onthex-axis.
41. LetP betheCOMof asystemof twoweightswithmassesm
1
andm
2
separatedbyadistanced. ProveArchimedes
Lawof the(weightless) Lever: P isthepoint onalinebetweenthetwoweightssuchthat m
1
L
1
= m
2
L
2
, whereL
j
is
thedistancefrommassj toP.
solution Placethelever alongthex-axiswithmassm
1
at theorigin. ThenM
y
= m
2
d andthex-coordinateof the
center of mass, P, is
m
2
d
m
1
+m
2
.
Thus,
L
1
=
m
2
d
m
1
+m
2
, L
2
= d
m
2
d
m
1
+m
2
=
m
1
d
m
1
+m
2
,
and
L
1
m
1
= m
1
m
2
d
m
1
+m
2
= m
2
m
1
d
m
1
+m
2
= L
2
m
2
.
FindtheCOM of asystemof twoweights of masses m
1
andm
2
connectedby alever of lengthd whosemass
density isuniform. Hint: Themoment of thesystemisthesumof themomentsof theweightsandthelever.
43. SymmetryPrinciple LetRbetheregionunderthegraphof f (x) overtheinterval [a, a], wheref (x)
0. Assumethat Rissymmetricwithrespect tothey-axis.
(a) Explainwhyf (x) iseventhat is, whyf (x) = f (x).
(b) Showthat xf (x) isanodd function.
(c) Use(b) toprovethat M
y
= 0.
(d) Provethat theCOM of Rliesonthey-axis(asimilar argument appliestosymmetrywithrespect tothex-axis).
S E C T I ON 9.4 Taylor Polynomials 577
solution
(a) Bythedenitionof symmetrywithrespect tothey-axis, f (x) = f (x), sof iseven.
(b) Let g(x) = xf (x) wheref iseven. Then
g(x) = xf (x) = xf (x) = g(x),
andthusg isodd.
(c) M
y
=
_
a
a
xf (x) dx = 0sincexf (x) isanoddfunction.
(d) Bypart (c), x
cm
=
M
y
M
=
0
M
= 0sothecenter of massliesalongthey-axis.
Provedirectly that Eqs. (2) and(3) areequivalent inthefollowingsituation. Let f (x) beapositivedecreasing
functionon[0, b] suchthat f (b) = 0. Set d = f (0) andg(y) = f
1
(y). Showthat
1
2
_
b
0
f (x)
2
dx =
_
d
0
yg(y) dy
Hint: First apply thesubstitutiony = f (x) totheintegral ontheleft andobservethat dx = g
(y) dy. Thenapply

IntegrationbyParts.
45. LetR bealaminaof uniformdensitysubmergedinauidof densityw(Figure23). Provethefollowinglaw: Theuid
forceononesideof R isequal totheareaof R timestheuidpressureonthecentroid. Hint: Let g(y) bethehorizontal
widthof R at depthy. Expressboththeuidpressure[Eq. (2) inSection8.2] andy-coordinateof thecentroidinterms
of g(y).
y
y
CM
y (depth)
Fluid level
Centroid
g(y)
FIGURE 23
solution Let denotetheuniformdensityof thesubmergedlamina. Then
M
x
=
_
b
a
yg(y) dy,
andthemassof thelaminais
M =
_
b
a
g(y) dy = A,
whereA istheareaof thelamina. Thus, they-coordinateof thecentroidis
y
cm
=

_
b
a
yg(y) dy
A
=
_
b
a
yg(y) dy
A
.
Now, theuidforceonthelaminais
F = w
_
b
a
yg(y) dy = w
_
b
a
yg(y) dy
A
A = wy
cm
A.
Inother words, theuidforceonthelaminaisequal totheuidpressureat thecentroidof thelaminatimestheareaof
thelamina.
9.4 Taylor Polynomials
1. What isT
3
(x) centeredat a = 3for afunctionf (x) suchthat f (3) = 9, f
(3) = 8, f
(3) = 4, andf
(3) = 12?
solution Ingeneral, witha = 3,
T
3
(x) = f (3) +f
(3)(x 3) +
f
(3)
2
(x 3)
2
+
f
(3)
6
(x 3)
3
.
Usingtheinformationprovided, wend
T
3
(x) = 9+8(x 3) +2(x 3)
2
+2(x 3)
3
.
2. ThedashedgraphsinFigure9areTaylorpolynomialsforafunctionf (x).Whichof thetwoisaMaclaurinpolynomial?
x x
2 3 1
2
3 1
-1 -1
y = f (x) y = f (x)
y y
(A) (B)
FIGURE 9
solution A Maclaurin polynomial always gives the value of f (0) exactly. This is true for theTaylor polynomial
sketchedin(B); thus, thisistheMaclaurinpolynomial.
3. For whichvalueof x doestheMaclaurinpolynomial T
n
(x) satisfyT
n
(x) = f (x), nomatter what f (x) is?
solution A Maclaurinpolynomial alwaysgivesthevalueof f (0) exactly.
4. LetT
n
(x) betheMaclaurinpolynomial of afunctionf (x) satisfying|f
(4)
(x)| 1for all x. Whichof thefollowing
statementsfollowfromtheerror bound?
(a) |T
4
(2) f (2)|
2
3
(b) |T
3
(2) f (2)|
2
3
(c) |T
3
(2) f (2)|
1
3
solution For afunctionf (x) satisfying|f
(4)
(x)| 1for all x,
|T
3
(2) f (2)|
1
24
|f
(4)
(x)|2
4
16
24
<
2
3
.
Thus, (b)isthecorrect answer.
Exercises
In Exercises 114, calculate the Taylor polynomials T
2
(x) and T
3
(x) centered at x = a for the given function and value
of a.
1. f (x) = sinx, a = 0
solution First, wecalculateandevaluatetheneededderivatives:
f (x) = sinx f (a) = 0
f
(x) = cosx f
(a) = 1
f
(x) = sinx f
(a) = 0
f
(x) = cosx f
(a) = 1
Now,
T
2
(x) = f (a) +f
(a)(x a) +
f
(a)
2
(x a)
2
= 0+1(x 0) +
0
2
(x 0)
2
= x; and
T
3
(x) = f (a) +f
(a)(x a) +
f
(a)
2
(x a)
2
+
f
(a)
6
(x a)
3
= 0+1(x 0) +
0
2
(x 0)
2
+
1
6
(x 0)
3
= x
1
6
x
3
.
f (x) = sinx, a =

2
3. f (x) =
1
1+x
, a = 2
f (x) =
1
1+x
f (a) =
1
3
f
(x) =
1
(1+x)
2
f
(a) =
1
9
f
(x) =
2
(1+x)
3
f
(a) =
2
27
f
(x) =
6
(1+x)
4
f
(a) =
2
27
Now,
T
2
(x) = f (a) +f
(a)(x a) +
f
(a)
2!
(x a)
2
=
1
3
1
9
(x 2) +
2/27
2!
(x 2)
2
=
1
3
1
9
(x 2) +
1
27
(x 2)
2
T
3
(x) = f (a) +f
(a)(x a) +
f
(a)
2!
(x a)
2
+
f
(a)
3!
(x a)
3
=
1
3
1
9
(x 2) +
2/27
2!
(x 2)
2
2/27
3!
(x 2)
3
=
1
3
1
9
(x 2) +
1
27
(x 2)
2
1
81
(x 2)
3
f (x) =
1
1+x
2
, a = 1
5. f (x) = x
4
2x, a = 3
solution First calculateandevaluatetheneededderivatives:
f (x) = x
4
2x f (a) = 75
f
(x) = 4x
3
2 f
(a) = 106
f
(x) = 12x
2
f
(a) = 108
f
(x) = 24x f
(a) = 72
Now,
T
2
(x) = f (a) +f
(a)(x a) +
f
(a)
2
(x a)
2
= 75+106(x 3) +
108
2
(x 3)
2
= 75+106(x 3) +54(x 3)
2
T
3
(x) = f (a) +f
(a)(x a) +
f
(a)
2
(x a)
2
+
f
(a)
3!
(x a)
3
= 75+106(x 3) +
108
2
(x 3)
2
+
72
3!
(x 3)
3
= 75+106(x 3) +54(x 3)
2
+12(x 3)
3
f (x) =
x
2
+1
x +1
, a = 2
7. f (x) = tanx, a = 0
f (x) = tanx f (a) = 0
f
(x) = sec
2
x f
(a) = 1
f
(x) = 2sec
2
x tanx f
(a) = 0
f
(x) = 2sec
4
x +4sec
2
x tan
2
x f
(a) = 2
Now,
T
2
(x) = f (a) +f
(a)(x a) +
f
(a)
2
(x a)
2
= 0+1(x 0) +
0
2
(x 0)
2
= x; and
T
3
(x) = f (a) +f
(a)(x a) +
f
(a)
2
(x a)
2
+
f
(a)
6
(x a)
3
= 0+1(x 0) +
0
2
(x 0)
2
+
2
6
(x 0)
3
= x +
1
3
x
3
.
f (x) = tanx, a =

4
9. f (x) = e
x
+e
2x
, a = 0
f (x) = e
x
+e
2x
f (a) = 2
f
(x) = e
x
2e
2x
f
(a) = 3
f
(x) = e
x
+4e
2x
f
(a) = 5
f
(x) = e
x
8e
2x
f
(a) = 9
Now,
T
2
(x) = f (a) +f
(a)(x a) +
f
(a)
2
(x a)
2
= 2+(3)(x 0) +
5
2
(x 0)
2
= 23x +
5
2
x
2
; and
T
3
(x) = f (a) +f
(a)(x a) +
f
(a)
2
(x a)
2
+
f
(a)
6
(x a)
3
= 2+(3)(x 0) +
5
2
(x 0)
2
+
9
6
(x 0)
3
= 23x +
5
2
x
2
3
2
x
3
.
f (x) = e
2x
, a = ln2
11. f (x) = x
2
e
x
, a = 1
f (x) = x
2
e
x
f (a) = 1/e
f
(x) = (2x x
2
)e
x
f
(a) = 1/e
f
(x) = (x
2
4x +2)e
x
f
(a) = 1/e
f
(x) = (x
2
+6x 6)e
x
f
(a) = 1/e
Now,
T
2
(x) = f (a) +f
(a)(x a) +
f
(a)
2
(x a)
2
=
1
e
+
1
e
(x 1) +
1/e
2
(x 1)
2
=
1
e
+
1
e
(x 1)
1
2e
(x 1)
2
; and
T
3
(x) = f (a) +f
(a)(x a) +
f
(a)
2
(x a)
2
+
f
(a)
6
(x a)
3
=
1
e
+
1
e
(x 1) +
1/e
2
(x 1)
2
+
_
1/e
6
_
(x 1)
3
=
1
e
+
1
e
(x 1)
1
2e
(x 1)
2
1
6e
(x 1)
3
.
f (x) = cosh2x, a = 0 13. f (x) =
lnx
x
, a = 1
solution First calculateandevaluatetheneededderivatives:
f (x) =
lnx
x
f (a) = 0
f
(x) =
1lnx
x
2
f (a) = 1
f
(x) =
3+2lnx
x
3
f (a) = 3
f
(x) =
116lnx
x
4
f (a) = 11
sothat
T
2
(x) = f (a) +f
(a)(x a) +
f
(a)
2!
(x a)
2
= 0+1(x 1) +
3
2!
(x 1)
2
= (x 1)
3
2
(x 1)
2
T
3
(x) = f (a) +f
(a)(x a) +
f
(a)
2!
(x a)
2
+
f
(a)
3!
(x a)
3
= 0+1(x 1) +
3
2!
(x 1)
2
+
11
3!
(x 1)
3
= (x 1)
3
2
(x 1)
2
+
11
6
(x 1)
3
f (x) = ln(x +1), a = 0
15. Showthat thenthMaclaurinpolynomial for e
x
is
T
n
(x) = 1+
x
1!
+
x
2
2!
+ +
x
n
n!
solution Withf (x) = e
x
, it followsthat f
(n)
(x) = e
x
andf
(n)
(0) = 1for all n. Thus,
T
n
(x) = 1+1(x 0) +
1
2
(x 0)
2
+ +
1
n!
(x 0)
n
= 1+x +
x
2
2
+ +
x
n
n!
.
Showthat thenthTaylor polynomial for
1
x +1
at a = 1is
T
n
(x) =
1
2
(x 1)
4
+
(x 1)
2
8
+ +(1)
n
(x 1)
n
2
n+1
17. Showthat theMaclaurinpolynomialsfor sinx are
T
2n+1
(x) = T
2n+2
(x) = x
x
3
3!
+
x
5
5!
+(1)
n
x
2n+1
(2n +1)!
solution Let f (x) = sinx. Then
f (x) = sinx f (0) = 0
f
(x) = cosx f
(0) = 1
f
(x) = sinx f
(0) = 0
f
(x) = cosx f
(0) = 1
f
(4)
(x) = sinx f
(4)
(0) = 0
f
(5)
(x) = cosx f
(5)
(0) = 1
.
.
.
.
.
.
Consequently,
T
2n+1
(x) = x
x
3
3!
+
x
5
5!
+ +(1)
n
x
2n+1
(2n +1)!
and
T
2n+2
(x) = x
x
3
3!
+
x
5
5!
+ +(1)
n
x
2n+1
(2n +1)!
+0= T
2n+1
(x).
Showthat theMaclaurinpolynomialsfor ln(1+x) are
T
n
(x) = x
x
2
2
+
x
3
3
+ +(1)
n1
x
n
n
In Exercises 1924, nd T
n
(x) at x = a for all n.
19. f (x) =
1
1+x
, a = 0
solution Wehave
1
1+x
= (ln(1+x))
sothat fromExercise18, lettingg(x) = ln(1+x),

f
(n)
(x) = g
(n+1)
(x) = (1)
n
n!(x +1)
1n
and f
(n)
(0) = (1)
n
n!
Then
T
n
(x) = f (0) +f
(0)x +
f
(0)
2!
x
2
+ +
f
(n)
(0)
n!
x
n
= 1x +
2!
2!
x
2
3!
3!
x
3
+ +(1)
n
n!
n!
x
n
= 1x +x
2
x
3
+ +(1)
n
x
n
f (x) =
1
x 1
, a = 4
21. f (x) = e
x
, a = 1
x
. Thenf
(n)
(x) = e
x
andf
(n)
(1) = e for all n. Therefore,
T
n
(x) = e +e(x 1) +
e
2!
(x 1)
2
+ +
e
n!
(x 1)
n
.
f (x) = x
2
, a = 2
23. f (x) = cosx, a =

4
solution Let f (x) = cosx. Then
f (x) = cosx f (/4) =
1
2
f
(x) = sinx f
(/4) =
1
2
f
(x) = cosx f
(/4) =
1
2
f
(x) = sinx f
(/4) =
1
2
Thispatternof four valuesrepeatsindenitely. Thus,
f
(n)
(/4) =
(1)
(n+1)/2
1
2
, n odd
(1)
n/2
1
2
, n even
and
T
n
(x) =
1
2
_
x

4
_
1
2
2
_
x

4
_
2
+
1
6
2
_
x

4
_
3
.
Ingeneral, thecoefcient of (x /4)
n
is
1
(
2)n!
withthepatternof signs+, , , +, +, , , . . . .
f () = sin3, a = 0
In Exercises 2528, nd T
2
(x) and use a calculator to compute the error |f (x) T
2
(x)| for the given values of a and x.
25. y = e
x
, a = 0, x = 0.5
x
. Thenf
(x) = e
x
, f
(x) = e
x
, f (a) = 1, f
(a) = 1andf
(a) = 1. Therefore
T
2
(x) = 1+1(x 0) +
1
2
(x 0)
2
= 1+x +
1
2
x
2
,
and
T
2
(0.5) = 1+(0.5) +
1
2
(0.5)
2
= 0.625.
Usingacalculator, wend
f (0.5) =
1
e
= 0.606531,
so
|T
2
(0.5) f (0.5)| = 0.0185.
y = cosx, a = 0, x =

12
27. y = x
2/3
, a = 1, x = 1.2
2/3
. Thenf
(x) =
2
3
x
5/3
, f
(x) =
10
9
x
8/3
, f (1) = 1, f
(1) =
2
3
, andf
(1) =
10
9
.
Thus
T
2
(x) = 1
2
3
(x 1) +
10
2 9
(x 1)
2
= 1
2
3
(x 1) +
5
9
(x 1)
2
and
T
2
(1.2) = 1
2
3
(0.2) +
5
9
(0.2)
2
=
8
9
0.88889
Usingacalculator, f (1.2) = (1.2)
2/3
0.88555sothat
|T
2
(1.2) f (1.2)| 0.00334
y = e
sinx
, a =

2
, x = 1.5
29. ComputeT
3
(x) for f (x) =

x centeredat a = 1. Thenuseaplotof theerror |f (x) T
3
(x)| tondavalue
c > 1suchthat theerror ontheinterval [1, c] isat most 0.25.
solution Wehave
f (x) = x
1/2
f (1) = 1
f
(x) =
1
2
x
1/2
f
(1) =
1
2
f
(x) =
1
4
x
3/2
f
(1) =
1
4
f
(x) =
3
8
x
5/2
f
(1) =
3
8
Therefore
T
3
(x) = 1+
1
2
(x 1)
1
4 2!
(x 1)
2
+
3
8 3!
(x 1)
3
= 1+
1
2
(x 1)
1
8
(x 1)
2
+
1
16
(x 1)
3
A plot of |f (x) T
3
(x)| isbelow.
1.0 1.5 2.0 2.5 3.0
0.05
0.10
0.15
0.20
0.25
y
x
It appearsthat for x [1, 2.9] that theerror doesnot exceed0.25. Theerror at x = 3appearstojust exceed0.25.
Plot f (x) = 1/(1+x) together withtheTaylor polynomials T
n
(x) at a = 1for 1 n 4ontheinterval
[2, 8] (besuretolimit theupper plot range).
(a) Over whichinterval doesT
4
(x) appear toapproximatef (x) closely?
(b) What happensfor x < 1?
(c) Useyour computer algebrasystemtoproduceandplot T
30
together withf (x) on[2, 8]. Over whichinterval
doesT
30
appear togiveacloseapproximation?
31. Let T
3
(x) betheMaclaurinpolynomial of f (x) = e
x
. Usetheerror boundtondthemaximumpossiblevalueof
|f (1.1) T
3
(1.1)|. Showthat wecantakeK = e
1.1
.
solution Sincef (x) = e
x
, wehavef
(n)
(x) = e
x
for all n; sincee
x
isincreasing, themaximumvalueof e
x
onthe
interval [0, 1.1] isK = e
1.1
. Thenbytheerror bound,
e
1.1
T
3
(1.1)
K
(1.10)
4
4!
=
e
1.1
1.1
4
24
0.183
LetT
2
(x) betheTaylor polynomial of f (x) =

x ata = 4. Applytheerror boundtondthemaximumpossible
valueof theerror |f (3.9) T
2
(3.9)|.
In Exercises 3336, compute the Taylor polynomial indicated and use the error bound to nd the maximum possible size
of the error. Verify your result with a calculator.
33. f (x) = cosx, a = 0; |cos0.25T
5
(0.25)|
solution TheMaclaurinseriesfor cosx is
1
x
2
2!
+
x
4
4!

x
6
6!
+. . .
sothat
T
5
(x) = 1
x
2
2
+
x
4
24
T
5
(0.25) 0.9689127604
Inaddition, f
(6)
(x) = cosx sothat |f
(6)
(x)| 1andwemaytakeK = 1intheerror boundformula. Then
|cos0.25T
5
(0.25)| K
0.25
6
6!
=
1
2
12
6!
3.390842014 10
7
(Thetruevalueiscos0.25 0.9689124217andthedifferenceisinfact 3.387 10
7
.)
f (x) = x
11/2
, a = 1; |f (1.2) T
4
(1.2)|
35. f (x) = x
1/2
, a = 4; |f (4.3) T
3
(4.3)|
solution Wehave
f (x) = x
1/2
f (4) =
1
2
f
(x) =
1
2
x
3/2
f
(4) =
1
16
f
(x) =
3
4
x
5/2
f
(4) =
3
128
f
(x) =
15
8
x
7/2
f
(4) =
15
1024
f
(4)
(x) =
105
16
x
9/2
sothat
T
3
(x) =
1
2
1
16
(x 4) +
3
256
(x 4)
2
5
2048
(x 4)
3
Usingtheerror boundformula,
|f (4.3) T
3
(4.3)| K
|4.34|
4
4!
=
27K
80,000
whereK isanumber suchthat |f
(4)
(x)| K for x between4and4.3. Now, f
(4)
(x) isadecreasingfunctionfor x > 1,
soit takesitsmaximumvalueon[4, 4.3] at x = 4; there, itsvalueis
K =
105
16
4
9/2
=
105
8192
sothat
|f (4.3) T
3
(4.3)|
27
105
8192
80,000
=
27 105
8192 80,000
4.3258667 10
6
f (x) =

1+x, a = 8; |
9.02T
3
(8.02)|
37. CalculatetheMaclaurinpolynomial T
3
(x) for f (x) = tan
1
x. ComputeT
3
_
1
2
_
andusetheerror boundto nda
boundfor theerror
tan
1 1
2
T
3
_
1
2
_
. Refer to thegraphinFigure10to ndanacceptablevalueof K. Verify your

result bycomputing
tan
1 1
2
T
3
_
1
2
_
usingacalculator.
y
x
2 1 3
1
1
2
3
4
5
FIGURE 10 Graphof f
(4)
(x) =
24x(x
2
1)
(x
2
+1)
4
, wheref (x) = tan
1
x.
solution Let f (x) = tan
1
x. Then
f (x) = tan
1
x f (0) = 0
f
(x) =
1
1+x
2
f
(0) = 1
f
(x) =
2x
(1+x
2
)
2
f
(0) = 0
f
(x) =
(1+x
2
)
2
(2) (2x)(2)(1+x
2
)(2x)
(1+x
2
)
4
f
(0) = 2
and
T
3
(x) = 0+1(x 0) +
0
2
(x 0)
2
+
2
6
(x 0)
3
= x
x
3
3
.
Sincef
(4)
(x) 5for x 0, wemaytakeK = 5intheerror bound; then,
tan
1
_
1
2
_
T
3
_
1
2
_

5(1/2)
4
4!
=
5
384
.
Let f (x) = ln(x
3
x +1). ThethirdTaylor polynomial at a = 1is
T
3
(x) = 2(x 1) +(x 1)
2
7
3
(x 1)
3
Findthemaximumpossiblevalueof |f (1.1) T
3
(1.1)|, usingthegraphinFigure11to ndanacceptablevalue
of K. Verifyyour result bycomputing|f (1.1) T
3
(1.1)| usingacalculator.
39. LetT
2
(x) betheTaylor polynomial ata = 0.5for f (x) = cos(x
2
). Usetheerror boundtondthemaximum
possiblevalueof |f (0.6) T
2
(0.6)|. Plot f
(3)
(x) tondanacceptablevalueof K.
solution Wehave
f (x) = cos(x
2
) f (0.5) = cos(0.25) 0.9689
f
(x) = 2x sin(x
2
) f
(0.5) = sin(0.25) 0.2474039593

f
(x) = 4x
2
cos(x
2
) 2sin(x
2
) f
(0.5) = cos(0.25) 2sin(0.25) 1.463720340

f
(3)
(x) = 8x
3
sin(x
2
) 12x cos(x
2
)
sothat
T
2
(x) = 0.96890.2472039593(x 0.5) 0.73186017(x 0.5)
2
andT
2
(0.6) 0.9368534237. A graphof f
(3)
(x) for x near 0.5isbelow.
3
0.4 0.5 0.6 0.7
4
5
6
7
y
x
Clearlythemaximumvalueof |f
(3)
(x)| on[0.5, 0.6] isboundedby7(near x = 0.5), sowemaytakeK = 7; then
|f (0.6) T
2
(0.6)| K
|0.60.5|
3
3!
=
7
6000
0.0011666667
CalculatetheMaclaurinpolynomial T
2
(x) for f (x) = sechx andusetheerror boundto ndthemaximum
possiblevalueof
f
_
1
2
_
T
2
_
1
2
_
. Plot f
(x) tondanacceptablevalueof K.
In Exercises 4144, use the error bound to nd a value of n for which the given inequality is satised. Then verify your
result using a calculator.
41. | cos0.1T
n
(0.1)| 10
7
, a = 0
solution Usingtheerror boundwithK = 1(everyderivativeof f (x) = cosx issinx or cosx, so|f
(n)
(x)| 1
for all n), wehave
|T
n
(0.1) cos0.1|
(0.1)
n+1
(n +1)!
.
Withn = 3,
(0.1)
4
4!
4.1710
6
> 10
7
,
but withn = 4,
(0.1)
5
5!
8.3310
8
< 10
7
,
sowechoosen = 4. Now,
T
4
(x) = 1
1
2
x
2
+
1
24
x
4
,
so
T
4
(0.1) = 1
1
2
(0.1)
2
+
1
24
(0.1)
4
= 0.995004166.
Usingacalculator, cos0.1= 0.995004165, so
|T
4
(0.1) cos0.1| = 1.38710
8
< 10
7
.
| ln1.3T
n
(1.3)| 10
4
, a = 1
43. |
1.3T
n
(1.3)| 10
6
, a = 1
solution UsingtheError Bound, wehave
|
1.3T
n
(1.3)| K
|1.31|
n+1
(n +1)!
= K
|0.3|
n+1
(n +1)!
,
whereK isanumber suchthat |f
(n+1)
(x)| K for x between1and1.3. For f (x) =

x, |f
(n)
(x)| isdecreasingfor
x > 1, hencethemaximumvalueof |f
(n+1)
(x)| occursat x = 1. Wemaythereforetake
K = |f
(n+1)
(1)| =
1 3 5 (2n +1)
2
n+1
=
1 3 5 (2n +1)
2
n+1

2 4 6 (2n +2)
2 4 6 (2n +2)
=
(2n +2)!
(n +1)!2
2n+2
.
Then
|
1.3T
n
(1.3)|
(2n +2)!
(n +1)!2
2n+2

|0.3|
n+1
(n +1)!
=
(2n +2)!
[(n +1)!]
2
(0.075)
n+1
.
Withn = 9
(20)!
[(10)!]
2
(0.075)
10
= 1.04010
6
> 10
6
,
but withn = 10
(22)!
[(11)!]
2
(0.075)
11
= 2.97910
7
< 10
6
.
Hence, n = 10will guaranteethedesiredaccuracy. UsingtechnologytocomputeandevaluateT
10
(1.3) gives
T
10
(1.3) 1.140175414,
1.3 1.140175425
and
|
1.3T
10
(1.3)| 1.110
8
< 10
6
|e
0.1
T
n
(0.1)| 10
6
, a = 0 45. Let f (x) = e
x
andT
3
(x) = 1x +
x
2
2

x
3
6
. Usetheerror boundtoshowthat for all x 0,
|f (x) T
3
(x)|
x
4
24
If youhaveaGU, illustratethisinequalitybyplottingf (x) T
3
(x) andx
4
/24together over [0, 1].
solution Notethatf
(n)
(x) = e
x
, sothat|f
(n)
(x)| = f (x). Now, f (x) isadecreasingfunctionfor x 0, sothat
for any c > 0, |f
(n)
(x)| takesitsmaximumvalueat x = 0; thisvalueise
0
= 1. Thuswemay takeK = 1intheerror
boundequation. Thusfor anyx,
|f (x) T
3
(x)| K
|x 0|
4
4!
=
x
4
24
A plot of f (x) T
3
(x) and
x
4
24
isshownbelow.
2
1 2 3 4 5 6 7
4
6
8
10
1
24
y
x
4
x
e
x
T
3
(x)
Usetheerror boundwithn = 4toshowthat
sinx
_
x
x
3
6
_

|x|
5
120
(for all x)
47. Let T
n
(x) betheTaylor polynomial for f (x) = lnx at a = 1, andlet c > 1. Showthat
| lnc T
n
(c)|
|c 1|
n+1
n +1
Thenndavalueof n suchthat | ln1.5T
n
(1.5)| 10
2
.
solution Withf (x) = lnx, wehave
f
(x) = x
1
, f
(x) = x
2
, f
(x) = 2x
3
, f
(4)
(x) = 6x
4
,
and, ingeneral,
f
(k+1)
(x) = (1)
k
k! x
k1
.
Noticethat |f
(k+1)
(x)| = k!|x|
k1
isadecreasingfunctionfor x > 0. Therefore, themaximumvalueof |f
(k+1)
(x)|
on[1, c] is|f
(k+1)
(1)|. UsingtheError Bound, wehave
|lnc T
n
(c)| K
|c 1|
n+1
(n +1)!
,
whereK is anumber such that |f
(n+1)
(x)| K for x between 1 and c. Frompart (a), weknow that wemay take
K = |f
(n+1)
(1)| = n!. Then
|lnc T
n
(c)| n!
|c 1|
n+1
(n +1)!
=
|c 1|
n+1
n +1
.
Evaluatingat c = 1.5gives
|ln1.5T
n
(1.5)|
|1.51|
n+1
n +1
=
(0.5)
n+1
n +1
.
Withn = 3,
(0.5)
4
4
= 0.015625> 10
2
.
but withn = 4,
(0.5)
5
5
= 0.00625< 10
2
.
Hence, n = 4will guaranteethedesiredaccuracy.
Let n 1. Showthat if |x| issmall, then
(x +1)
1/n
1+
x
n
+
1n
2n
2
x
2
Usethisapproximationwithn = 6toestimate1.5
1/6
.
49. Verify that the third Maclaurin polynomial for f (x) = e
x
sinx is equal to the product of the third Maclaurin
polynomialsof e
x
andsinx (after discardingtermsof degreegreater than3intheproduct).
x
sinx. Then
f (x) = e
x
sinx f (0) = 0
f
(x) = e
x
(cosx +sinx) f
(0) = 1
f
(x) = 2e
x
cosx f
(0) = 2
f
(x) = 2e
x
(cosx sinx) f
(0) = 2
and
T
3
(x) = 0+(1)x +
2
2!
x
2
+
2
3!
x
3
= x +x
2
+
x
3
3
.
Now, thethirdMaclaurinpolynomial for e
x
is1+x +
x
2
2
+
x
3
6
, andthethirdMaclaurinpolynomial for sinx isx
x
3
6
.
Multiplyingthesetwopolynomials, andthendiscardingtermsof degreegreater than3, yields
e
x
sinx x +x
2
+
x
3
3
,
whichagreeswiththeMaclaurinpolynomial obtainedfromthedenition.
FindthefourthMaclaurinpolynomial for f (x) = sinx cosx by multiplyingthefourthMaclaurinpolynomials
for f (x) = sinx andf (x) = cosx.
51. FindtheMaclaurinpolynomialsT
n
(x) for f (x) = cos(x
2
). Youmay usethefact that T
n
(x) isequal tothesumof
thetermsuptodegreen obtainedbysubstitutingx
2
for x inthenthMaclaurinpolynomial of cosx.
solution TheMaclaurinpolynomialsfor cosx areof theform
T
2n
(x) = 1
x
2
2
+
x
4
4!
+ +(1)
n
x
2n
(2n)!
.
Accordingly, theMaclaurinpolynomialsfor cos(x
2
) areof theform
T
4n
(x) = 1
x
4
2
+
x
8
4!
+ +(1)
n
x
4n
(2n)!
.
Find the Maclaurin polynomials of 1/(1+x
2
) by substituting x
2
for x in the Maclaurin polynomials of
1/(1x).
53. Let f (x) = 3x
3
+ 2x
2
x 4. Calculate T
j
(x) for j = 1, 2, 3, 4, 5 at both a = 0 and a = 1. Show that
T
3
(x) = f (x) inbothcases.
3
+2x
2
x 4. Then
f (x) = 3x
3
+2x
2
x 4 f (0) = 4 f (1) = 0
f
(x) = 9x
2
+4x 1 f
(0) = 1 f
(1) = 12
f
(x) = 18x +4 f
(0) = 4 f
(1) = 22
f
(x) = 18 f
(0) = 18 f
(1) = 18
f
(4)
(x) = 0 f
(4)
(0) = 0 f
(4)
(1) = 0
f
(5)
(x) = 0 f
(5)
(0) = 0 f
(5)
(1) = 0
At a = 0,
T
1
(x) = 4x;
T
2
(x) = 4x +2x
2
;
T
3
(x) = 4x +2x
2
+3x
3
= f (x);
T
4
(x) = T
3
(x); and
T
5
(x) = T
3
(x).
At a = 1,
T
1
(x) = 12(x 1);
T
2
(x) = 12(x 1) +11(x 1)
2
;
T
3
(x) = 12(x 1) +11(x 1)
2
+3(x 1)
3
= 4x +2x
2
+3x
3
= f (x);
T
4
(x) = T
3
(x); and
T
5
(x) = T
3
(x).
Let T
n
(x) bethenth Taylor polynomial at x = a for apolynomial f (x) of degreen. Based on theresult of
Exercise53, guessthevalueof |f (x) T
n
(x)|. Provethat your guessiscorrect usingtheerror bound.
55. Let s(t ) bethedistanceof atruck to an intersection. At timet = 0, thetruck is 60meters fromtheintersection,
travelsawayfromit withavelocityof 24m/s, andbeginstoslowdownwithanaccelerationof a = 3m/s
2
. Determine
thesecondMaclaurinpolynomial of s(t ), anduseit toestimatethetrucksdistancefromtheintersectionafter 4s.
solution Placetheoriginattheintersection, sothats(0) = 60(thetruckistravelingawayfromtheintersection). The
secondMaclaurinpolynomial of s(t ) is
T
2
(t ) = s(0) +s
(0)t +
s
(0)
2
t
2
Theconditionsof theproblemtell usthat s(0) = 60, s
(0) = 24, ands
(0) = 3. Thus
T
2
(t ) = 60+24t
3
2
t
2
sothat after 4seconds,
T
2
(4) = 60+24 4
3
2
4
2
= 132m
Thetruckis132mpast theintersection.
A bank owns aportfolio of bonds whosevalueP(r) depends on theinterest rater (measured in percent; for
example, r = 5meansa5%interest rate). Thebanksquantitativeanalyst determinesthat
P(5) = 100,000,
dP
dr
r=5
= 40,000,
d
2
P
dr
2
r=5
= 50,000
57. A narrow, negativelychargedringof radiusR exertsaforceonapositivelychargedparticleP locatedat distancex
abovethecenter of theringof magnitude
F(x) =
kx
(x
2
+R
2
)
3/2
wherek > 0isaconstant (Figure12).
(a) Computethethird-degreeMaclaurinpolynomial for F(x).
(b) Showthat F (k/R
3
)x to secondorder. This shows that whenx is small, F(x) behaves likearestoringforce
similar totheforceexertedbyaspring.
(c) Showthat F(x) k/x
2
whenx islargebyshowingthat
lim
x
F(x)
k/x
2
= 1
Thus, F(x) behaveslikeaninversesquarelaw, andthechargedringlookslikeapoint chargefromfar away.
x
x
R
F(x)
Nearly linear
here
Nearly inversesquare
here
P
FIGURE 12
solution
(a) Start bycomputingandevaluatingthenecessaryderivatives:
F(x) =
kx
(x
2
+R
2
)
3/2
F(0) = 0
F
(x) =
k(2x
2
R
2
)
(x
2
+R
2
)
5/2
F
(0) =
k
R
3
F
(x) =
3kx(3R
2
2x
2
)
(x
2
+R
2
)
7/2
F
(0) = 0
F
(x) =
3k(8x
4
24x
2
R
2
+3R
4
)
(x
2
+R
2
)
9/2
F
(0) =
9k
R
5
sothat
T
3
(x) = F(0) +F
(0)x +
F
(0)
2!
x
2
+
F
(0)
3!
x
3
=
k
R
3
x +
3k
2R
5
x
3
(b) Todegree2, F(x) T
3
(x)
k
R
3
x aswemayignorethex
3
termof T
3
(x).
(c) Wehave
lim
x
F(x)
k/x
2
= lim
x
_
x
2
k
kx
(x
2
+R
2
)
3/2
_
= lim
x
x
3
(x
2
+R
2
)
3/2
= lim
x
1
x
3
(x
2
+R
2
)
3/2
= lim
x
1
(1+R
2
/x
2
)
3/2
= 1
Thusasx growslarge, F(x) lookslikeaninversesquarefunction.
A light waveof wavelength travels fromA toB by passingthroughanaperture(circular region) locatedina
planethat isperpendicular toAB (seeFigure13for thenotation). Let f (r) = d
+ h
; that is, f (r) isthedistance

AC +CB asafunctionof r.
(a) Showthat f (r) =
_
d
2
+r
2
+
_
h
2
+r
2
, andusetheMaclaurinpolynomial of order 2toshowthat
f (r) d +h +
1
2
_
1
d
+
1
h
_
r
2
(b) TheFresnel zones, used to determinetheoptical disturbanceat B, aretheconcentric bands bounded by the
circlesof radiusR
n
suchthat f (R
n
) = d +h +n/2. Showthat R
n

nL, whereL = (d
1
+h
1
)
1
.
(c) Estimatetheradii R
1
andR
100
for bluelight ( = 47510
7
cm) if d = h = 100cm.
59. Referring to Figure 14, let a be the length of the chord AC of angle of the unit circle. Derive the following
approximationfor theexcessof thearcover thechord.
a

3
24
Hint: Showthat a = 2sin(/2) andusethethirdMaclaurinpolynomial asanapproximation.
C
1
B
A
b
a
2
FIGURE 14 Unit circle.
solution Drawalinefromthecenter O of thecircletoB, andlabel thepoint of intersectionof thislinewithAC as
D. ThenCD =
a
2
, andtheangleCOB is
2
. SinceCO = 1, wehave
sin
2
=
a
2
sothat a = 2sin(/2). Thus a = 2sin(/2). Now, thethirdMaclaurinpolynomial for f () = sin(/2) canbe
computedasfollows: f (0) = 0, f
(x) =
1
2
cos(/2) sothat f
(0) =
1
2
. f
(x) =
1
4
sin(/2) andf
(0) = 0. Finally,
f
(x) =
1
8
cos(/2) andf
(0) =
1
8
. Thus
T
3
() = f (0) +f
(0) +
f
(0)
2!

2
+
f
(0)
3!

3
=
1
2

1
48
3
Finally,
a = 2sin
2
2T
3
() =
_

1
24
3
_
=

3
24
To estimatethelength of a circular arc of theunit circle, theseventeenth-century Dutch scientist Christian
Huygensusedtheapproximation (8b a)/3, wherea isthelengthof thechordAC of angle andb islength
of thechordAB of angle/2(Figure14).
(a) Prove that a = 2sin(/2) and b = 2sin(/4), and show that the Huygens approximation amounts to the
approximation

16
3
sin
2
3
sin
2
(b) ComputethefthMaclaurinpolynomial of thefunctionontheright.
(c) Usetheerror boundtoshowthat theerror intheHuygensapproximationislessthan0.00022||
5
.
61. Showthat thenthMaclaurinpolynomial of f (x) = arcsinx for n oddis
T
n
(x) = x +
1
2
x
3
3
+
1 3
2 4
x
5
5
+ +
1 3 5 (n 2)
2 4 6 (n 1)
x
n
n
1
x. Then
f (x) = sin
1
x f (0) = 0
f
(x) =
1
_
1x
2
f
(0) = 1
f
(x) =
1
2
(1x
2
)
3/2
(2x) f
(0) = 0
f
(x) =
2x
2
+1
(1x
2
)
5/2
f
(0) = 1
f
(4)
(x) =
3x(2x
2
+3)
(1x
2
)
7/2
f
(4)
(0) = 0
f
(5)
(x) =
24x
4
+72x
2
+9
(1x
2
)
9/2
f
(5)
(0) = 9
.
.
.
.
.
.
f
(7)
(0) = 225
and
T
7
(x) = x +
x
3
3!
+
9x
5
5!
+
225x
7
7!
= x +
1
2
x
3
3
+
1
2
3
4
x
5
5
+
1
2
3
4
5
6
x
7
7
.
Thus, wecaninfer that
T
n
(x) = x +
1
2
x
3
3
+
1
2
3
4
x
5
5
+
1
2
3
4
5
6
x
7
7
+ +
1
2
3
4

n 2
n 1
x
n
n
.
Let x 0andassumethat f
(n+1)
(t ) 0for 0 t x. UseTaylorsTheoremtoshowthat thenthMaclaurin
polynomial T
n
(x) satises
T
n
(x) f (x) for all x 0
63. UseExercise62toshowthat for x 0andall n,
e
x
1+x +
x
2
2!
+ +
x
n
n!
Sketchthegraphsof e
x
, T
1
(x), andT
2
(x) onthesamecoordinateaxes. Doesthisinequalityremaintruefor x < 0?
x
. Thenf
(n)
(x) = e
x
for all n. Becausee
x
> 0for all x, it follows fromExercise62that
f (x) T
n
(x) for all x 0andfor all n. For f (x) = e
x
,
T
n
(x) = 1+x +
x
2
2!
+ +
x
n
n!
,
thus,
e
x
1+x +
x
2
2!
+ +
x
n
n!
.
Fromthegurebelow, weseethat theinequalitydoesnot remaintruefor x < 0, asT
2
(x) e
x
for x < 0.
x
y
2 1 1 2
2
4
6
e
x
T
1
T
2
Thisexerciseisintendedtoreinforcetheproof of TaylorsTheorem.
(a) Showthat f (x) = T
0
(x) +
_
x
a
f
(u) du.
(b) UseIntegrationbyPartstoprovetheformula
_
x
a
(x u)f
(2)
(u) du = f
(a)(x a) +
_
x
a
f
(u) du
(c) Provethecasen = 2of TaylorsTheorem:
f (x) = T
1
(x) +
_
x
a
(x u)f
(2)
(u) du.
In Exercises 6569, we estimate integrals using Taylor polynomials. Exercise 66 is used to estimate the error.
65. Find the fourth Maclaurin polynomial T
4
(x) for f (x) = e
x
2
, and calculate I =
_
1/2
0
T
4
(x) dx as an estimate
_
1/2
0
e
x
2
dx. A CAS yields thevalueI 0.461281. How largeis theerror in your approximation? Hint: T
4
(x) is
obtainedbysubstitutingx
2
inthesecondMaclaurinpolynomial for e
x
.
solution Followingthehint, sincethesecondMaclaurinpolynomial for e
x
is
1+x +
x
2
2
wesubstitutex
2
for x toget thefourthMaclaurinpolynomial for e
x
2
:
T
4
(x) = 1x
2
+
x
4
2
Then
_
1/2
0
e
x
2
dx
_
1/2
0
T
4
(x) dx =
_
x
1
3
x
3
+
1
10
x
5
_
1/2
0
=
443
960
0.4614583333
UsingaCAS, wehave
_
1/2
0
e
x
2
dx 0.4612810064, sotheerror isabout 1.7710
4
.
ApproximatingIntegrals Let L > 0. Showthat if twofunctionsf (x) andg(x) satisfy|f (x) g(x)| < L for
all x [a, b], then
_
b
a
f (x) dx
_
b
a
g(x) dx
< L(b a)
67. Let T
4
(x) bethefourthMaclaurinpolynomial for cosx.
(a) Showthat | cosx T
4
(x)|
_
1
2
_
6
/6! for all x
_
0,
1
2
_
. Hint: T
4
(x) = T
5
(x).
(b) Evaluate
_
1/2
0
T
4
(x) dx as anapproximationto
_
1/2
0
cosx dx. UseExercise66to ndaboundfor thesizeof the
error.
solution
(a) Let f (x) = cosx. Then
T
4
(x) = 1
x
2
2
+
x
4
24
.
Moreover, witha = 0, T
4
(x) = T
5
(x) and
|cosx T
4
(x)| K
|x|
6
6!
,
whereK is anumber suchthat |f
(6)
(u)| K for u between0andx. Now|f
(6)
(u)| = | cosu| 1, so wemay take
K = 1. Finally, withtherestrictionx [0,
1
2
],
|cosx T
4
(x)|
(1/2)
6
6!
0.000022.
(b)
_
1/2
0
_
1
x
2
2
+
x
4
24
_
dx =
1841
3840
0.479427.
By(a) andExercise66, theerror associatedwiththisapproximationislessthanor equal to
(1/2)
6
6!
_
1
2
0
_
=
1
92,160
1.110
5
.
Notethat
_
1/2
0
cosx dx 0.4794255, sotheactual error isroughly1.510
6
.
Let Q(x) = 1x
2
/6. Usetheerror boundfor sinx toshowthat
sinx
x
Q(x)

|x|
4
5!
Thencalculate
_
1
0
Q(x) dx asanapproximationto
_
1
0
(sinx/x) dx andndaboundfor theerror.
69. (a) ComputethesixthMaclaurinpolynomial T
6
(x) for sin(x
2
) by substitutingx
2
inP(x) = x x
3
/6, thethird
Maclaurinpolynomial for sinx.
(b) Showthat | sin(x
2
) T
6
(x)|
|x|
10
5!
.
Hint: Substitutex
2
for x intheerror boundfor | sinx P(x)|, notingthat P(x) isalsothefourthMaclaurinpolynomial
for sinx.
(c) UseT
6
(x) toapproximate
_
1/2
0
sin(x
2
) dx andndaboundfor theerror.
solution Let s(x) = sinx andf (x) = sin(x
2
). Then
(a) ThethirdMaclaurinpolynomial for sinx is
S
3
(x) = x
x
3
6
so, substitutingx
2
for x, weseethat thesixthMaclaurinpolynomial for sin(x
2
) is
T
6
(x) = x
2
x
6
6
(b) Sinceall derivativesof s(x) areeither cosx or sinx, theyareboundedinmagnitudeby1, sowemaytakeK = 1
intheError Boundfor sinx. SincethethirdMaclaurinpolynomial S
3
(x) for sinx isalsothefourthMaclaurinpolynomial
S
4
(x), wehave
|sinx S
3
(x)| = |sinx S
4
(x)| K
|x|
5
5!
=
|x|
5
5!
Nowsubstitutex
2
for x intheaboveinequalityandnotefrompart (a) that S
3
(x
2
) = T
6
(x) toget
|sin(x
2
) S
3
(x
2
)| = |sin(x
2
) T
6
(x)|
|x
2
|
5
5!
=
|x|
10
5!
(c)
_
1/2
0
sin(x
2
) dx
_
1/2
0
T
6
(x) dx =
_
1
3
x
3
1
42
x
7
_
1/2
0
0.04148065476
Frompart (b), theerror isboundedby
x
10
5!
=
(1/2)
10
120
=
1
1024 120
8.13802083310
6
Thetruevalueof theintegral isapproximately0.04148102420, whichisconsistent withthecomputederror bound.
Provebyinductionthat for all k,
d
j
dx
j
_
(x a)
k
k!
_
=
k(k 1) (k j +1)(x a)
kj
k!
d
j
dx
j
_
(x a)
k
k!
_
x=a
=
_
1 for k = j
0 for k = j
Usethistoprovethat T
n
(x) agreeswithf (x) at x = a toorder n.
71. Let a beanynumber andlet
P(x) = a
n
x
n
+a
n1
x
n1
+ +a
1
x +a
0
beapolynomial of degreen or less.
(a) Showthat if P
(j)
(a) = 0for j = 0, 1, . . . , n, thenP(x) = 0, that is, a
j
= 0for all j. Hint: Useinduction, noting
that if thestatement istruefor degreen 1, thenP
(x) = 0.
(b) Provethat T
n
(x) istheonlypolynomial of degreen or lessthat agreeswithf (x) at x = a toorder n. Hint: If Q(x)
isanother suchpolynomial, apply(a) toP(x) = T
n
(x) Q(x).
solution
(a) Noterst that if n = 0, i.e. if P(x) = a
0
is aconstant, then thestatement holds: if P
(0)
(a) = P(a) = 0, then
a
0
= 0sothat P(x) = 0. Next, assumethestatement holdsfor all polynomialsof degreen 1or less, andlet P(x) be
apolynomial of degreeat most n withP
(j)
(a) = 0for j = 0, 1, . . . , n. If P(x) hasdegreelessthann, thenweknow
P(x) = 0byinduction, soassumethedegreeof P(x) isexactlyn. Then
P(x) = a
n
x
n
+a
n1
x
n1
+ +a
1
x +a
0
wherea
n
= 0; also,
P
(x) = na
n
x
n1
+(n 1)a
n1
x
n2
+ +a
1
Notethat P
(j+1)
(a) = (P
)
(j)
(a) for j = 0, 1, . . . , n 1. But then
0= P
(j+1)
(a) = (P
)
(j)
(a) for all j = 0, 1, . . . , n 1
SinceP
(x) hasdegreeat most n 1, it followsbyinductionthat P
(x) = 0. Thusa
n
= a
n1
= = a
1
= 0sothat
P(x) = a
0
. But P(a) = 0sothat a
0
= 0aswell andthusP(x) = 0.
(b) Suppose Q(x) is a polynomial of degree at most n that agrees with f (x) at x = a up to order n. Let P(x) =
T
n
(x) Q(x). Notethat P(x) isapolynomial of degreeat most n sincebothT
n
(x) andQ(x) are. SincebothT
n
(x) and
Q(x) agreewithf (x) at x = a toorder n, wehave
T
(j)
n
(a) = f
(j)
(a) = Q
(j)
(a), j = 0, 1, 2, . . . , n
Thus
P
(j)
(a) = T
(j)
n
(a) Q
(j)
(a) = 0 for j = 0, 1, 2, . . . , n
But thenbypart (a), P(x) = 0sothat T
n
(x) = Q(x).
In Exercises 14, calculate the arc length over the given interval.
1. y =
x
5
10
+
x
3
6
, [1, 2]
solution Let y =
x
5
10
+
x
3
6
. Then
1+(y
)
2
= 1+
_
x
4
2

x
4
2
_
2
= 1+
x
8
4

1
2
+
x
8
4
=
x
8
4
+
1
2
+
x
8
4
=
_
x
4
2
+
x
4
2
_
2
.
Because
1
2
(x
4
+x
4
) > 0on[1, 2], thearclengthis
s =
_
2
1
_
1+(y
)
2
dx =
_
2
1
_
x
4
2
+
x
4
2
_
dx =
_
x
5
10
x
3
6
_
2
1
=
779
240
.
y = e
x/2
+e
x/2
, [0, 2]
3. y = 4x 2, [2, 2]
solution Let y = 4x 2. Then
_
1+
_
y
_
2
=
_
1+4
2
=
17.
Hence,
s =
_
2
2
17dx = 4
17.
y = x
2/3
, [1, 8]
5. Show that the arc length of y = 2
x over [0, a] is equal to

a(a +1) + ln(
a +

a +1). Hint: Apply the
substitutionx = tan
2
tothearclengthintegral.
solution Let y = 2
x. Theny
=
1
x
, and
_
1+
_
y
_
2
=
_
1+
1
x
=
_
x +1
x
=
1
x +1.
Thus,
s =
_
a
0
1
1+x dx.
Wemakethesubstitutionx = tan
2
, dx = 2tan sec
2
d. Then
s =
_
x=a
x=0
1
tan
sec 2tan sec
2
d = 2
_
x=a
x=0
sec
3
d.
Weuseareductionformulatoobtain
s = 2
_
tan sec
2
+
1
2
ln| sec +tan|
_
x=a
x=0
= (
1+x +ln|
1+x +
x|)
a
0
=

a
1+a +ln|
1+a +
a| =
_
a(a +1) +ln
_
a +
a +1
_
.
Computethetrapezoidal approximationT
5
tothearclengths of y = tanx over
_
0,

4
_
.
In Exercises 710, calculate the surface area of the solid obtained by rotating the curve over the given interval about the
x-axis.
7. y = x +1, [0, 4]
solution Let y = x +1. Theny
= 1, and
y
_
1+y
2
= (x +1)
1+1=
2(x +1).
Thus,
SA = 2
_
4
0
2(x +1) dx = 2
2
_
x
2
2
+x
_
4
0
= 24
2.
y =
2
3
x
3/4
2
5
x
5/4
, [0, 1]
9. y =
2
3
x
3/2
1
2
x
1/2
, [1, 2]
solution Let y =
2
3
x
3/2
1
2
x
1/2
. Then
y
=

x
1
4
x
,
and
1+
_
y
_
2
= 1+
_
x
1
4
x
_
2
= 1+
_
x
1
2
+
1
16x
_
= x +
1
2
+
1
16x
=
_
x +
1
4
x
_
2
.
Because
x +
1
x
0, thesurfaceareais
2
_
b
a
y
_
1+
_
y
_
2
dx = 2
_
2
1
_
2
3
x
3/2
x
2
__
x +
1
4
x
_
dx
= 2
_
2
1
_
2
3
x
2
+
1
6
x
1
2
x
1
8
_
dx = 2
_
2x
3
9

x
2
6

1
8
x
_
2
1
=
67
36
.
y =
1
2
x
2
, [0, 2]
11. Computethetotal surfaceareaof thecoinobtainedbyrotatingtheregioninFigure1about thex-axis. Thetopand
bottompartsof theregionaresemicircleswitharadiusof 1mm.
1 mm
4 mm
x
y
FIGURE 1
solution Thegeneratinghalf circleof theedgeisy = 2+
_
1x
2
. Then,
y
=
2x
2
_
1x
2
=
x
_
1x
2
,
and
1+(y
)
2
= 1+
x
2
1x
2
=
1
1x
2
.
Thesurfaceareaof theedgeof thecoinis
2
_
1
1
y
_
1+
_
y
_
2
dx = 2
_
1
1
_
2+
_
1x
2
_
1
_
1x
2
dx
= 2
_
2
_
1
1
dx
_
1x
2
+
_
1
1
_
1x
2
_
1x
2
dx
_
= 2
_
2arcsinx|
1
1
+
_
1
1
dx
_
= 2(2 +2) = 4
2
+4.
Wenowaddthesurfaceareaof thetwosidesof thedisk, whicharecirclesof radius2. Hencethesurfaceareaof thecoin
is:
_
4
2
+4
_
+2 2
2
= 4
2
+12.
Calculatetheuidforceonthesideof aright triangleof height 3mandbase2msubmergedinwater vertically,
withitsupper vertexat thesurfaceof thewater.
13. Calculatetheuidforceonthesideof arighttriangleof height3mandbase2msubmergedinwater vertically, with
itsupper vertexlocatedat adepthof 4m.
solution Weneedtondanexpressionfor thehorizontal widthf (y) at depthy.
3
2
y
f (y)
y 4
4
Bysimilar triangleswehave:
f (y)
y 4
=
2
3
so f (y) =
2(y 4)
3
.
Hence, theforceonthesideof thetriangleis
F = g
_
7
4
yf (y) dy =
2g
3
_
7
4
_
y
2
4y
_
dy =
2g
3
_
y
3
3
2y
2
_
7
4
= 18g.
For water, = 10
3
; g = 9.8, soF = 18 9800= 176,400N.
A plateintheshapeof theshadedregioninFigure2issubmergedinwater. Calculatetheuidforceonasideof
theplateif thewater surfaceisy = 1.
15. Figure3showsanobjectwhosefaceisanequilateral trianglewith5-msides. Theobjectis2mthickandissubmerged
inwater withits vertex 3mbelowthewater surface. Calculatetheuidforceonbothatriangular faceandaslanted
rectangular edgeof theobject.
5
2
3
Water level
FIGURE 3
solution Start witheachtriangular faceof theobject. Placetheoriginat theupper vertex of thetriangle, withthe
positivey-axispointingdownward. Notethat becausetheequilateral trianglehassidesof length5feet, theheight of the
triangleis
5
3
2
feet. Moreover, thewidthof thetriangleat locationy is
2y
3
. Thus,
F =
2g
3
_
5
3/2
0
(y +3)y dy =
2g
3
_
1
3
y
3
+
3
2
y
2
_
3/2
0
=
g
4
(125+75
3) 624,514N.
Now, consider theslantedrectangular edgesof theobject. Eachedgeisaconstant 2feet wideandmakesanangleof 60
withthehorizontal. Therefore,
F =
g
sin60
_
5
3/2
0
2(y +3) dy =
2g
3
_
y
2
+6y
_
3/2
0
= g
_
25
3
2
+30
_
506,176N.
Theforceonthebottomfacecanbecomputedwithout calculus:
F =
_
3+
5
3
2
_
(2)(5)g 718,352N.
Theendof ahorizontal oil tank is anellipse(Figure4) withequation(x/4)
2
+ (y/3)
2
= 1(lengthinmeters).
Assumethat thetankislledwithoil of density900kg/m
3
.
(a) Calculatethetotal forceF ontheendof thetankwhenthetankisfull.
(b) Wouldyouexpect thetotal forceonthelower half of thetank tobegreater than, lessthan, or equal to
1
2
F?
Explain. Thencomputetheforceonthelower half exactlyandconrm(or refute) your expectation.
17. CalculatethemomentsandCOM of thelaminaoccupyingtheregionunder y = x(4x) for 0 x 4, assuming
adensityof = 1200kg/m
3
.
solution Becausethelaminaissymmetricwithrespecttothevertical linex = 2, bythesymmetryprinciple, weknow
that x
cm
= 2. Now,
M
x
=

2
_
4
0
f (x)
2
dx =
1200
2
_
4
0
x
2
(4x)
2
dx =
1200
2
_
16
3
x
3
2x
4
+
1
5
x
5
_
4
0
= 20,480.
Moreover, themassof thelaminais
M =
_
4
0
f (x) dx = 1200
_
4
0
x(4x) dx = 1200
_
2x
2
1
3
x
3
_
4
0
= 12,800.
Thus, thecoordinatesof thecenter of massare
_
2,
20,480
12,800
_
=
_
2,
8
5
_
.
Sketchtheregionbetweeny = 4(x +1)
1
andy = 1for 0 x 3, andnditscentroid.
19. Findthecentroidof theregionbetweenthesemicircley =
_
1x
2
andthetophalf of theellipsey =
1
2
_
1x
2
(Figure2).
solution Sincetheregion is symmetric with respect to they-axis, thecentroid lies on they-axis. To nd y
cm
we
calculate
M
x
=
1
2
_
1
1
__
1x
2
_
2
__
1x
2
2
_
2
dx
=
1
2
_
1
1
3
4
_
1x
2
_
dx =
3
8
_
x
1
3
x
3
_
1
1
=
1
2
.
Theareaof thelaminais

2

4
=

4
, sothecoordinatesof thecentroidare
_
0,
1/2
/4
_
=
_
0,
2
_
.
Findthecentroidof theshadedregioninFigure5boundedontheleft by x = 2y
2
2andontheright by a
semicircleof radius1. Hint: Usesymmetryandadditivityof moments.
In Exercises 2126, nd the Taylor polynomial at x = a for the given function.
21. f (x) = x
3
, T
3
(x), a = 1
solution Westart bycomputingtherst threederivativesof f (x) = x
3
:
f
(x) = 3x
2
f
(x) = 6x
f
(x) = 6
Evaluatingthefunctionanditsderivativesat x = 1, wend
f (1) = 1, f
(1) = 3, f
(1) = 6, f
(1) = 6.
Therefore,
T
3
(x) = f (1) +f
(1)(x 1) +
f
(1)
2!
(x 2)
2
+
f
(1)
3!
(x 1)
3
= 1+3(x 1) +
6
2!
(x 2)
2
+
6
3!
(x 1)
3
= 1+3(x 1) +3(x 2)
2
+(x 1)
3
.
f (x) = 3(x +2)
3
5(x +2), T
3
(x), a = 2
23. f (x) = x ln(x), T
4
(x), a = 1
solution Westart bycomputingtherst four derivativesof f (x) = x lnx:
f
(x) = lnx +x
1
x
= lnx +1
f
(x) =
1
x
f
(x) =
1
x
2
f
(4)
(x) =
2
x
3
f (1) = 0, f
(1) = 1, f
(1) = 1, f
(1) = 1, f
(4)
(1) = 2.
Therefore,
T
4
(x) = f (1) +f
(1)(x 1) +
f
(1)
2!
(x 1)
2
+
f
(1)
3!
(x 1)
3
+
f
(4)
(1)
4!
(x 1)
4
= 0+1(x 1) +
1
2!
(x 1)
2
1
3!
(x 1)
3
+
2
4!
(x 1)
4
= (x 1) +
1
2
(x 1)
2
1
6
(x 1)
3
+
1
12
(x 1)
4
.
f (x) = (3x +2)
1/3
, T
3
(x), a = 2
25. f (x) = xe
x
2
, T
4
(x), a = 0
solution Westart bycomputingtherst four derivativesof f (x) = xe
x
2
:
f
(x) = e
x
2
+x (2x)e
x
2
= (12x
2
)e
x
2
f
(x) = 4xe
x
2
+(12x
2
) (2x)e
x
2
= (4x
3
6x)e
x
2
f
(x) = (12x
2
6)e
x
2
+(4x
3
6x) (2x)e
x
2
= (8x
4
+24x
2
6)e
x
2
f
(4)
(x) = (32x
3
+48x)e
x
2
+(8x
4
+24x
2
6) (2x)e
x
2
= (16x
5
80x
3
+60x)e
x
2
f (0) = 0, f
(0) = 1, f
(0) = 0, f
(0) = 6, f
(4)
(0) = 0.
Therefore,
T
4
(x) = f (0) +f
(0)x +
f
(0)
2!
x
2
+
f
(0)
3!
x
3
+
f
(4)
(0)
4!
x
4
= 0+x +0 x
2
6
3!
x
3
+0 x
4
= x x
3
.
f (x) = ln(cosx), T
3
(x), a = 0
27. FindthenthMaclaurinpolynomial for f (x) = e
3x
.
solution Wedifferentiatethefunctionf (x) = e
3x
repeatedly, lookingfor apattern:
f
(x) = 3e
3x
= 3
1
e
3x
f
(x) = 3 3e
3x
= 3
2
e
3x
f
(x) = 3 3
2
e
3x
= 3
3
e
3x
Thus, for general n, f
(n)
(x) = 3
n
e
3x
andf
(n)
(0) = 3
n
. Substitutingintotheformulafor thenthTaylor polynomial, we
obtain:
T
n
(x) = 1+
3x
1!
+
3
2
x
2
2!
+
3
3
x
3
3!
+
3
4
x
4
4!
+ +
3
n
x
n
n!
= 1+3x +
1
2!
(3x)
2
+
1
3!
(3x)
3
+ +
1
n!
(3x)
n
.
UsethefthMaclaurinpolynomial of f (x) = e
x
toapproximate
e. Useacalculator todeterminetheerror.
29. UsethethirdTaylor polynomial of f (x) = tan
1
x at a = 1toapproximatef (1.1). Useacalculator todetermine
theerror.
solution Westart bycomputingtherst threederivativesof f (x) = tan
1
x:
f
(x) =
1
1+x
2
f
(x) =
2x
_
1+x
2
_
2
f
(x) =
2
_
1+x
2
_
2
+2x 2
_
1+x
2
_
2x
_
1+x
2
_
4
=
2
_
3x
2
1
_
_
1+x
2
_
3
f (1) =

4
, f
(1) =
1
2
, f
(1) =
1
2
, f
(1) =
1
2
.
Therefore,
T
3
(x) = f (1) +f
(1)(x 1) +
f
(1)
2!
(x 1)
2
+
f
(1)
3!
(x 1)
3
=

4
+
1
2
(x 1)
1
4
(x 1)
2
+
1
12
(x 1)
3
.
Settingx = 1.1yields:
T
3
(1.1) =

4
+
1
2
(0.1)
1
4
(0.1)
2
+
1
12
(0.1)
3
= 0.832981496.
Usingacalculator, wendtan
1
1.1= 0.832981266. Theerror intheTaylor polynomial approximationis
T
3
(1.1) tan
1
1.1
= |0.8329814960.832981266| = 2.30110
7
.
LetT
4
(x) betheTaylor polynomial for f (x) =

x ata = 16. Usetheerror boundtondthemaximumpossible
sizeof |f (17) T
4
(17)|.
31. Findn suchthat |e T
n
(1)| < 10
8
, whereT
n
(x) isthenthMaclaurinpolynomial for f (x) = e
x
.
solution UsingtheError Bound, wehave
|e T
n
(1)| K
|10|
n+1
(n +1)!
=
K
(n +1)!
whereK is anumber suchthat
f
(n+1)
(x)
= e
x
K for all 0 x 1. Sincee
x
is increasing, themaximumvalue
ontheinterval 0 x 1isattainedat theendpoint x = 1. Thus, for 0 u 1, e
u
e
1
< 2.8. Hencewemay take
K = 2.8toobtain:
|e T
n
(1)|
2.8
(n +1)!
Wenowchoosen suchthat
2.8
(n +1)!
< 10
8
(n +1)!
2.8
> 10
8
(n +1)! > 2.810
8
For n = 10, (n + 1)! = 3.9910
7
< 2.810
8
andfor n = 11, (n + 1)! = 4.7910
8
> 2.810
8
. Hence, tomake
theerror lessthan10
8
, n = 11issufcient; that is,
|e T
11
(1)| < 10
8
.
Let T
4
(x) betheTaylor polynomial for f (x) = x lnx at a = 1computedinExercise23. Usetheerror boundto
ndaboundfor |f (1.2) T
4
(1.2)|.
33. Verify that T
n
(x) = 1+ x + x
2
+ + x
n
is thenth Maclaurin polynomial of f (x) = 1/(1x). Showusing
substitutionthat thenthMaclaurinpolynomial for f (x) = 1/(1x/4) is
T
n
(x) = 1+
1
4
x +
1
4
2
x
2
+ +
1
4
n
x
n
What isthenthMaclaurinpolynomial for g(x) =
1
1+x
?
solution Let f (x) = (1 x)
1
. Then, f
(x) = (1 x)
2
, f
(x) = 2(1 x)
3
, f
(x) = 3!(1 x)
4
, and, in
general, f
(n)
(x) = n!(1x)
(n+1)
. Therefore, f
(n)
(0) = n! and
T
n
(x) = 1+
1!
1!
x +
2!
2!
x
2
+ +
n!
n!
x
n
= 1+x +x
2
+ +x
n
.
Uponsubstitutingx/4for x, wendthat thenthMaclaurinpolynomial for f (x) =
1
1x/4
is
T
n
(x) = 1+
1
4
x +
1
4
2
x
2
+ +
1
4
n
x
n
.
Substitutingx for x, thenthMaclaurinpolynomial for g(x) =
1
1+x
is
T
n
(x) = 1x +x
2
x
3
+ +(x)
n
.
Let f (x) =
5
4+3x x
2
andlet a
k
bethecoefcient of x
k
intheMaclaurinpolynomial T
n
(x) of for k n.
(a) Showthat f (x) =
_
1/4
1x/4
+
1
1+x
_
.
(b) UseExercise33toshowthat a
k
=
1
4
k+1
+(1)
k
.
(c) ComputeT
3
(x).
35. Let T
n
(x) bethenthMaclaurinpolynomial for thefunctionf (x) = sinx +sinhx.
(a) Showthat T
5
(x) = T
6
(x) = T
7
(x) = T
8
(x).
(b) Showthat |f
n
(x)| 1+coshx for all n. Hint: Notethat | sinhx| | coshx| for all x.
(c) Showthat |T
8
(x) f (x)|
2.6
9!
|x|
9
for 1 x 1.
solution
(a) Let f (x) = sinx +sinhx. Then
f
(x) = cosx +coshx

f
(x) = sinx +sinhx

f
(x) = cosx +coshx

f
(4)
(x) = sinx +sinhx.
Fromthispoint onward, thepatternof derivativesrepeatsindenitely. Thus
f (0) = f
(4)
(0) = f
(8)
(0) = sin0+sinh0= 0
f
(0) = f
(5)
(0) = cos0+cosh0= 2
f
(0) = f
(6)
(0) = sin0+sinh0= 0
f
(0) = f
(7)
(0) = cos0+cosh0= 0.
Consequently,
T
5
(x) = f
(0)x +
f
(5)
(0)
5!
x
5
= 2x +
1
60
x
5
,
and, becausef
(6)
(0) = f
(7)
(0) = f
(8)
(0) = 0, it followsthat
T
6
(x) = T
7
(x) = T
8
(x) = T
5
(x) = 2x +
1
60
x
5
.
(b) First notethat | sinx| 1and| cosx| 1for all x. Moreover,
| sinhx| =
e
x
e
x
2

e
x
+e
x
2
= coshx.
Now, recall frompart (a), that all derivativesof f (x) containtwoterms: therst issinx or cosx, whilethesecond
iseither sinhx or coshx. Inabsolutevalue, thetrigonometrictermisalwayslessthanor equal to1, whilethehyperbolic
termisalwayslessthanor equal tocoshx. Thus, for all n,
f
(n)
(x) 1+coshx.
(c) UsingtheError Bound, wehave
|T
8
(x) f (x)|
K|x 0|
9
9!
=
K|x|
9
9!
,
whereK isanumber suchthat
f
(9)
(u)
K for all u between0andx. Bypart (b), weknowthat

f
(9)
(u) 1+coshu.
Now, coshu is anevenfunctionthat is increasingon(0, ). Themaximumvaluefor u between0andx is therefore
coshx. Moreover, for 1 x 1, coshx cosh1 1.543< 1.6. Hence, wemaytakeK = 1+1.6= 2.6, and
|T
8
(x) f (x)|
2.6
9!
|x|
9
.
10 INTRODUCTION TO
DIFFERENTIAL EQUATIONS
10.1 Solving Differential Equations
1. Determinetheorder of thefollowingdifferential equations:
(a) x
5
y
= 1 (b) (y
)
3
+x = 1
(c) y
+x
4
y
= 2 (d) sin(y
) +x = y
solution
(a) Thehighest order derivativethat appearsinthisequationisarst derivative, sothisisarst order equation.
(b) Thehighest order derivativethat appearsinthisequationisarst derivative, sothisisarst order equation.
(c) Thehighest order derivativethat appearsinthisequationisathirdderivative, sothisisathirdorder equation.
(d) Thehighest order derivativethat appearsinthisequationisasecondderivative, sothisisasecondorder equation.
2. Isy
= sinx alinear differential equation?

solution Yes.
3. Giveanexampleof anonlinear differential equationof theformy
= f (y).
solution Onepossibilityisy
= y
2
.
4. Cananonlinear differential equationbeseparable? If so, giveanexample.
solution Yes. Anexampleisy
= y
2
.
5. Giveanexampleof alinear, nonseparabledifferential equation.
solution Oneexampleisy
+y = x.
Exercises
1. Whichof thefollowingdifferential equationsarerst-order?
(a) y
= x
2
(b) y
= y
2
(c) (y
)
3
+yy
= sinx (d) x
2
y
e
x
y = siny
(e) y
+3y
=
y
x
(f) yy
+x +y = 0
solution
(a) Thehighest order derivativethat appearsinthisequationisarst derivative, sothisisarst order equation.
(b) Thehighest order derivativethat appearsinthisequationisasecondderivative, sothisisnot arst order equation.
(c) Thehighest order derivativethat appearsinthisequationisarst derivative, sothisisarst order equation.
(d) Thehighest order derivativethat appearsinthisequationisarst derivative, sothisisarst order equation.
(e) Thehighest order derivativethat appearsinthisequationisasecondderivative, sothisisnot arst order equation.
(f) Thehighest order derivativethat appearsinthisequationisarst derivative, sothisisarst order equation.
Whichof theequationsinExercise1arelinear?
In Exercises 38, verify that the given function is a solution of the differential equation.
3. y
8x = 0, y = 4x
2
solution Let y = 4x
2
. Theny
= 8x and
y
8x = 8x 8x = 0.
yy
+4x = 0, y =
_
124x
2
5. y
+4xy = 0, y = 25e
2x
2
solution Let y = 25e
2x
2
. Theny
= 100xe
2x
2
, and
y
+4xy = 100xe
2x
2
+4x(25e
2x
2
) = 0.
601
602 C HA P T E R 10 INTRODUCTION TO DIFFERENTIAL EQUATIONS
(x
2
1)y
+xy = 0, y = 4(x
2
1)
1/2
7. y
2xy
+8y = 0, y = 4x
4
12x
2
+3
solution Let y = 4x
4
12x
2
+3. Theny
= 16x
3
24x, y
= 48x
2
24, and
y
2xy
+8y = (48x
2
24) 2x(16x
3
24x) +8(4x
4
12x
2
+3)
= 48x
2
2432x
4
+48x
2
+32x
4
96x
2
+24= 0.
y
2y
+5y = 0, y = e
x
sin2x
9. Whichof thefollowingequationsareseparable? Writethosethat areseparableintheformy
= f (x)g(y) (but do
not solve).
(a) xy
9y
2
= 0 (b)
_
4x
2
y
= e
3y
sinx
(c) y
= x
2
+y
2
(d) y
= 9y
2
solution
(a) xy
9y
2
= 0isseparable:
xy
9y
2
= 0
xy
= 9y
2
y
=
9
x
y
2
(b)
_
4x
2
y
= e
3y
sinx isseparable:
_
4x
2
y
= e
3y
sinx
y
= e
3y
sinx
_
4x
2
.
(c) y
= x
2
+y
2
isnot separable; y
isalreadyisolated, but isnot equal toaproduct f (x)g(y).

(d) y
= 9y
2
isseparable: y
= (1)(9y
2
).
Thefollowingdifferential equationsappear similar but haveverydifferent solutions.
dy
dx
= x,
dy
dx
= y
Solvebothsubject totheinitial conditiony(1) = 2.
11. Consider thedifferential equationy
3
y
9x
2
= 0.
(a) Writeit asy
3
dy = 9x
2
dx.
(b) Integratebothsidestoobtain
1
4
y
4
= 3x
3
+C.
(c) Verifythat y = (12x
3
+C)
1/4
isthegeneral solution.
(d) Findtheparticular solutionsatisfyingy(1) = 2.
solution Solvingy
3
y
9x
2
= 0for y
givesy
= 9x
2
y
3
.
(a) Separatingvariablesintheequationaboveyields
y
3
dy = 9x
2
dx
(b) Integratingbothsidesgives
y
4
4
= 3x
3
+C
(c) Simplifytheequationabovetoget y
4
= 12x
3
+C, or y = (12x
3
+C)
1/4
.
(d) Solve2= (12 1
3
+C)
1/4
toget 16= 12+C, or C = 4. Thustheparticular solutionisy = (12x
3
+4)
1/4
.
Verifythat x
2
y
+e
y
= 0isseparable.
(a) Writeit ase
y
dy = x
2
dx.
(b) Integratebothsidestoobtaine
y
= x
1
+C.
(c) Verifythat y = ln(x
1
+C) isthegeneral solution.
In Exercises 1328, use separation of variables to nd the general solution.
13. y
+4xy
2
= 0
solution Rewrite
y
+4xy
2
= 0 as
dy
dx
= 4xy
2
andthenas y
2
dy = 4x dx
Integratingbothsidesof thisequationgives
_
y
2
dy = 4
_
x dx
y
1
= 2x
2
+C
y
1
= 2x
2
+C
S E C T I ON 10.1 Solving Differential Equations 603
Solvingfor y gives
y =
1
2x
2
+C
whereC isanarbitraryconstant.
y
+x
2
y = 0
15.
dy
dt
20t
4
e
y
= 0
solution Rewrite
dy
dt
20t
4
e
y
= 0 as
dy
dt
= 20t
4
e
y
andthenas e
y
dy = 20t
4
dt
Integratingbothsidesof thisequationgives
_
e
y
dy =
_
20t
4
dt
e
y
= 4t
5
+C
Solvefor y toget y = ln(4t
5
+C), whereC isanarbitraryconstant.
t
3
y
+4y
2
= 0
17. 2y
+5y = 4
solution Rewrite
2y
+5y = 4 as y
= 2
5
2
y andthenas (45y)
1
dy =
1
2
dx
Integratingbothsidesandsolvingfor y gives
_
dy
45y
=
1
2
_
1dx
1
5
ln|45y| =
1
2
x +C
1
ln|45y| = C
2
5
2
x
45y = C
3
e
5x/2
5y = 4C
3
e
5x/2
y = Ce
5x/2
+
4
5
dy
dt
= 8
y
19.
_
1x
2
y
= xy
solution Rewrite
_
1x
2
dy
dx
= xy as
dy
y
=
x
_
1x
2
dx.
Integratingbothsidesof thisequationyields
_
dy
y
=
_
x
_
1x
2
dx
ln|y| =
_
1x
2
+C.
Solvingfor y, wend
|y| = e
1x
2
+C
= e
C
e
1x
2
y = e
C
e
1x
2
= Ae
1x
2
,
whereA isanarbitraryconstant.
y
= y
2
(1x
2
)
21. yy
= x
solution Rewrite
y
dy
dx
= x as y dy = x dx.
_
y dy =
_
x dx
1
2
y
2
=
1
2
x
2
+C.
Solvingfor y, wend
y
2
= x
2
+2C
y =
_
x
2
+A,
whereA = 2C isanarbitraryconstant.
(lny)y
ty = 0
23.
dx
dt
= (t +1)(x
2
+1)
solution Rewrite
dx
dt
= (t +1)(x
2
+1) as
1
x
2
+1
dx = (t +1) dt.
_
1
x
2
+1
dx =
_
(t +1) dt
tan
1
x =
1
2
t
2
+t +C.
Solvingfor x, wend
x = tan
_
1
2
t
2
+t +C
_
.
whereA = tanC isanarbitraryconstant.
(1+x
2
)y
= x
3
y
25. y
= x secy
solution Rewrite
dy
dx
= x secy as cosy dy = x dx.
_
cosy dy =
_
x dx
siny =
1
2
x
2
+C.
Solvingfor y, wend
y = sin
1
_
1
2
x
2
+C
_
,
dy
d
= tany
27.
dy
dt
= y tant
solution Rewrite
dy
dt
= y tant as
1
y
dy = tant dt.
_
1
y
dy =
_
tant dt
ln|y| = ln|sect | +C.
Solvingfor y, wend
|y| = e
ln|sect |+C
= e
C
|sect |
y = e
C
sect = Asect,
whereA = e
C
isanarbitraryconstant.
dx
dt
= t tanx
29. y
+2y = 0, y(ln5) = 3
solution First, wendthegeneral solutionof thedifferential equation. Rewrite
dy
dx
+2y = 0 as
1
y
dy = 2dx,
andthenintegratetoobtain
ln|y| = 2x +C.
Thus,
y = Ae
2x
,
whereA = e
C
isanarbitraryconstant. Theinitial conditiony(ln5) = 3allowsustodeterminethevalueof A.
3= Ae
2(ln5)
; 3= A
1
25
; so 75= A.
Finally,
y = 75e
2x
.
y
3y +12= 0, y(2) = 1
31. yy
= xe
y
2
, y(0) = 2
y
dy
dx
= xe
y
2
as ye
y
2
dy = x dx,
1
2
e
y
2
=
1
2
x
2
+C.
Thus,
y =
_
ln(x
2
+A),
whereA = 2C is anarbitrary constant. Theinitial conditiony(0) = 2allows us to determinethevalueof A. Since
y(0) < 0, wehavey =
_
ln(x
2
+A), and
2=
_
ln(A); 4= ln(A); so e
4
= A.
Finally,
y =
_
ln(x
2
+e
4
).
y
2
dy
dx
= x
3
, y(1) = 0
33. y
= (x 1)(y 2), y(2) = 4

dy
dx
= (x 1)(y 2) as
1
y 2
dy = (x 1) dx,
ln|y 2| =
1
2
x
2
x +C.
Thus,
y = Ae
(1/2)x
2
x
+2,
whereA = e
C
isanarbitraryconstant. Theinitial conditiony(2) = 4allowsustodeterminethevalueof A.
4= Ae
0
+2 so A = 2.
Finally,
y = 2e
(1/2)x
2
x
+2.
y
= (x 1)(y 2), y(2) = 2

35. y
= x(y
2
+1), y(0) = 0
solution First, ndthegeneral solutionof thedifferential equation. Rewrite
dy
dx
= x(y
2
+1) as
1
y
2
+1
dy = x dx
andintegratetoobtain
tan
1
y =
1
2
x
2
+C
sothat
y = tan
_
1
2
x
2
+C
_
whereC isanarbitrary constant. Theinitial conditiony(0) = 0allowsustodeterminethevalueof C: 0 = tan(C), so
C = 0. Finally,
y = tan
_
1
2
x
2
_
(1t )
dy
dt
y = 0, y(2) = 4
37.
dy
dt
= ye
t
, y(0) = 1
dy
dt
= ye
t
as
1
y
dy = e
t
dt,
ln|y| = e
t
+C.
Thus,
y = Ae
e
t
,
whereA = e
C
1= Ae
1
so A = e.
Finally,
y = (e)e
e
t
= e
1e
t
.
dy
dt
= t e
y
, y(1) = 0
39. t
2
dy
dt
t = 1+y +ty, y(1) = 0
t
2
dy
dt
= 1+t +y +ty = (1+t )(1+y)
as
1
1+y
dy =
1+t
t
2
dt,
ln|1+y| = t
1
+ln|t | +C.
Thus,
y = A
t
e
1/t
1,
whereA = e
C
0= A
_
1
e
_
1 so A = e.
Finally,
y =
et
e
1/t
1.
_
1x
2
y
= y
2
+1, y(0) = 0
41. y
= tany, y(ln2) =

2
dy
dx
= tany as
dy
tany
= dx,
ln|siny| = x +C.
Thus,
y = sin
1
(Ae
x
),
whereA = e
C
isanarbitraryconstant. Theinitial conditiony(ln2) =

2
allowsustodeterminethevalueof A.
2
= sin
1
(2A); 1= 2A so A =
1
2
.
Finally,
y = sin
1
_
1
2
e
x
_
.
y
= y
2
sinx, y() = 2
43. Findall valuesof a suchthat y = x
a
isasolutionof
y
12x
2
y = 0
solution Let y = x
a
. Then
y
= ax
a1
and y
= a(a 1)x
a2
.
Substitutingintothedifferential equation, wend
y
12x
2
y = a(a 1)x
a2
12x
a2
= x
a2
(a
2
a 12).
Thus, y
12x
2
y = 0if andonlyif
a
2
a 12= (a 4)(a +3) = 0.
Hence, y = x
a
isasolutionof thedifferential equationy
12x
2
y = 0provideda = 4or a = 3.
Findall valuesof a suchthat y = e
ax
isasolutionof
y
+4y
12y = 0
In Exercises 45 and 46, let y(t ) be a solution of (cosy +1)
dy
dt
= 2t such that y(2) = 0.
45. Showthat siny +y = t
2
+C. Wecannot solvefor y asafunctionof t , but, assumingthat y(2) = 0, ndthevalues
of t at whichy(t ) = .
solution Rewrite
(cosy +1)
dy
dt
= 2t as (cosy +1) dy = 2t dt
andintegratetoobtain
siny +y = t
2
+C
whereC isanarbitraryconstant. Sincey(2) = 0, wehavesin0+0= 4+C sothat C = 4andtheparticular solution
weseekissiny +y = t
2
4. Tondvaluesof t at whichy(t ) = , wemust solvesin + = t
2
4, or t
2
4= ;
thust =
+4.
Assumingthat y(6) = /3, ndanequationof thetangent linetothegraphof y(t ) at (6, /3).
In Exercises 4752, use Eq. (4) and Torricellis Law [Eq. (5)].
47. Water leaksthroughaholeof area0.002m
2
atthebottomof acylindrical tankthatislledwithwater andhasheight
3mandabaseof area10m
2
. Howlongdoesit take(a) for half of thewater toleakout and(b) for thetanktoempty?
solution Becausethetank has aconstant cross-sectional areaof 10m
2
andtheholehas anareaof 0.002m
2
, the
differential equationfor theheight of thewater inthetankis
dy
dt
=
0.002v
10
= 0.0002v.
ByTorricellisLaw,
v =
_
2gy =
_
19.6y,
usingg = 9.8m/s
2
. Thus,
dy
dt
= 0.0002
_
19.6y = 0.0002
19.6
y.
Separatingvariablesandthenintegratingyields
y
1/2
dy = 0.0002
19.6dt
2y
1/2
= 0.0002
19.6t +C
Solvingfor y, wend
y(t ) =
_
C 0.0001
19.6t
_
2
.
Sincethetankisoriginallyfull, wehavetheinitial conditiony(0) = 10, whence
10= C. Therefore,
y(t ) =
_
100.0001
19.6t
_
2
.
Whenhalf of thewater isout of thetank, y = 1.5, sowesolve:
1.5=
_
100.0001
19.6t
_
2
for t , nding
t =
1
0.0002
19.6
(2
10
6) 4376.44sec.
Whenall of thewater isout of thetank, y = 0, so
100.0001
19.6t = 0 and t =
10
0.0001
19.6
7142.86sec.
At t = 0, aconical tank of height 300cmandtopradius100cm[Figure7(A)] islledwithwater. Water leaks
throughaholeinthebottomof area3cm
2
. Let y(t ) bethewater level at timet .
(a) Showthat thetankscross-sectional areaat height y isA(y) =

9
y
2
.
(b) Findandsolvethedifferential equationsatisedbyy(t )
(c) Howlongdoesit takefor thetanktoempty?
49. Thetank inFigure7(B) isacylinder of radius4mandheight 15m. Assumethat thetank ishalf-lledwithwater
andthat water leaksthroughaholeinthebottomof areaB = 0.001m
2
. Determinethewater level y(t ) andthetimet
e
whenthetankisempty.
solution Whenthewater isatheighty over thebottom, thetopcrosssectionisarectanglewithlength15m, andwith
widthx satisfyingtheequation:
(x/2)
2
+(y 4)
2
= 16.
Thus, x = 2
_
8y y
2
, and
A(y) = 15x = 30
_
8y y
2
.
WithB = 0.001m
2
andv =
2gy =
19.6
y, it followsthat
dy
dt
=
0.001
19.6
y
30
_
8y y
2
=
0.001
19.6
30
8y
.
Separatingvariablesandintegratingthenyields:
_
8y dy =
0.001
19.6
30
dt =
0.0001
19.6
3
dt
2
3
(8y)
3/2
=
0.0001
19.6
3
t +C
Whent = 0, y = 4, soC =
2
3
4
3/2
=
16
3
, and
2
3
(8y)
3/2
=
0.0001
19.6
3
t
16
3
y(t ) = 8
_
0.0001
19.6
2
t +8
_
2/3
.
Thetankisemptywheny = 0. Thus, t
e
satisestheequation
8
_
0.0001
19.6
2
t +8
_
2/3
= 0.
It followsthat
t
e
=
2(8
3/2
8)
0.0001
19.6
66,079.9seconds.
A tank has the shape of the parabola y = x
2
, revolved around the y-axis. Water leaks froma hole of area
B = 0.0005m
2
at thebottomof thetank. Let y(t ) bethewater level at timet . Howlongdoesit takefor thetankto
emptyif it isinitiallylledtoheight y
0
= 1m.
51. A tankhastheshapeof theparabolay = ax
2
(wherea isaconstant) revolvedaroundthey-axis. Water drainsfrom
aholeof areaB m
2
at thebottomof thetank.
(a) Showthat thewater level at timet is
y(t ) =
_
y
3/2
0

3aB
2g
2
t
_
2/3
wherey
0
isthewater level at timet = 0.
(b) Showthat if thetotal volumeof water inthetank hasvolumeV at timet = 0, theny
0
=
2aV/. Hint: Compute

thevolumeof thetankasavolumeof rotation.
(c) Showthat thetankisemptyat time
t
e
=
_
2
3B
g
_
_
2V
3
a
_
1/4
Weseethat for xedinitial water volumeV, thetimet
e
isproportional toa
1/4
. A largevalueof a correspondstoatall
thintank. Suchatankdrainsmorequicklythanashort widetankof thesameinitial volume.
solution
(a) Whenthewater isat height y, thesurfaceof thewater isacircleof radius
y/a, sothat thecross-sectional areais

A(y) = y/a. Withv =
2gy =
2g
y, wehave
dy
dt
=
B
2g
y
A
=
aB
2g
y
y
=
aB
2g
y
1/2
Separatingvariablesandintegratinggives
y dy =
aB
2g
dt
2
3
y
3/2
=
aB
2g
t +C
1
y
3/2
=
3aB
2g
2
t +C
Sincey(0) = y
0
, wehaveC = y
3/2
0
; solvingfor y gives
y =
_
y
3/2
0

3aB
2g
2
t
_
2/3
(b) Thevolumeof thetankcanbecomputedasavolumeof rotation. Usingthediskmethodandapplyingittothefunction
x =
y/a, wehave
V =
_
y
0
0

_
y
a
2
dy =

a
_
y
0
0
y dy =

2a
y
2
y
0
0
=

2a
y
2
0
Solvingfor y
0
gives
y
0
=
_
2aV/
(c) Thetankisemptywheny = 0; thisoccurswhen
y
3/2
0

3aB
2g
2
t = 0
Frompart (b), wehave
y
3/2
0
=
_
2aV/
3/2
= ((2aV/)
1/2
)
3/2
= (2aV/)
3/4
sothat
t
e
=
2y
3/2
0
3aB
2g
=
2
4
8a
3
V
3
3
3/4
B
4
a
4
4
g
=
2
1/4
4
2V
3
a
1
3B
g
=
_
2
3B
g
_
_
2V
3
a
_
1/4
A cylindrical tanklledwithwater hasheight h andabaseof areaA. Water leaksthroughaholeinthebottom
of areaB.
(a) Showthat thetimerequiredfor thetanktoemptyisproportional toA
h/B.
(b) Showthat theemptyingtimeisproportional toVh
1/2
, whereV isthevolumeof thetank.
(c) Twotankshavethesamevolumeandaholeof thesamesize, but theyhavedifferent heightsandbases. Which
tankemptiesrst: thetaller or theshorter tank?
53. Figure8showsacircuitconsistingof aresistor of R ohms, acapacitor of C farads, andabatteryof voltageV. When
thecircuit is completed, theamount of chargeq(t ) (incoulombs) ontheplates of thecapacitor varies accordingtothe
differential equation(t inseconds)
R
dq
dt
+
1
C
q = V
whereR, C, andV areconstants.
(a) Solvefor q(t ), assumingthat q(0) = 0.
(b) Showthat lim
t
q(t ) = CV.
(c) Showthat thecapacitor chargestoapproximately63%of itsnal valueCV after atimeperiodof length = RC (
iscalledthetimeconstant of thecapacitor).
V C
R
FIGURE 8 AnRC circuit.
solution
(a) Uponrearrangingthetermsof thedifferential equation, wehave
dq
dt
=
q CV
RC
.
Separatingthevariablesandintegratingbothsides, weobtain
dq
q CV
=
dt
RC
_
dq
q CV
=
_
dt
RC
and
ln|q CV| =
t
RC
+k,
wherek isanarbitraryconstant. Solvingfor q(t ) yields
q(t ) = CV +Ke
1
RC
t
,
whereK = e
k
. Weusetheinitial conditionq(0) = 0tosolvefor K:
0= CV +K K = CV
sothat theparticular solutionis
q(t ) = CV(1e
1
RC
t
)
(b) Usingtheresult frompart (a), wecalculate
lim
t
q(t ) = lim
t
CV(1e
1
RC
t
) = CV(1 lim
t
1e
1
RC
t
) = CV.
(c) Wehave
q() = q(RC) = CV(1e
1
RC
RC
) = CV(1e
1
) 0.632CV.
Assumeinthecircuit of Figure8that R = 200, C = 0.02F, andV = 12V. Howmany secondsdoesit take
for thechargeonthecapacitor platestoreachhalf of itslimitingvalue?
55. Accordingtoonehypothesis, thegrowthratedV/dt of acellsvolumeV isproportional toitssurfacearea
A. SinceV hascubicunitssuchascm
3
andA hassquareunitssuchascm
2
, wemayassumeroughlythatA V
2/3
, and
hencedV/dt = kV
2/3
for someconstantk. If thishypothesisiscorrect, whichdependenceof volumeontimewouldwe
expect tosee(again, roughlyspeaking) inthelaboratory?
(a) Linear (b) Quadratic (c) Cubic
solution Rewrite
dV
dt
= kV
2/3
as V
2/3
dv = k dt,
andthenintegratebothsidestoobtain
3V
1/3
= kt +C
V = (kt /3+C)
3
.
Thus, weexpect toseeV increasingroughlylikethecubeof time.
Wemight alsoguessthat thevolumeV of ameltingsnowball decreasesat arateproportional toitssurfacearea.
Argueas inExercise55tondadifferential equationsatisedby V. Supposethesnowball has volume1000cm
3
andthat it loseshalf of itsvolumeafter 5min. Accordingtothismodel, whenwill thesnowball disappear?
57. Ingeneral, (fg)
isnot equal tof
, but let f (x) = e

3x
andndafunctiong(x) suchthat (fg)
= f
. Dothe
samefor f (x) = x.
solution If (fg)
= f
, wehave
f
(x)g(x) +g
(x)f (x) = f
(x)g
(x)
g
(x)(f (x) f
(x)) = g(x)f
(x)
g
(x)
g(x)
=
f
(x)
f
(x) f (x)
Now, let f (x) = e
3x
. Thenf
(x) = 3e
3x
and
g
(x)
g(x)
=
3e
3x
3e
3x
e
3x
=
3
2
.
Integratingandsolvingfor g(x), wend
dg
g
=
3
2
dx
ln|g| =
3
2
x +C
g(x) = Ae
(3/2)x
,
whereA = e
C
If f (x) = x, thenf
(x) = 1, and
g
(x)
g(x)
=
1
1x
.
Thus,
dg
g
=
1
1x
dx
ln|g| = ln|1x| +C
g(x) =
A
1x
,
whereA = e
C
AboystandingatpointB onadockholdsaropeof length attachedtoaboatatpointA[Figure9(A)].Astheboy
walksalongthedock, holdingtheropetaut, theboat movesalongacurvecalledatractrix(fromtheLatintractus,
meaningtopull). Thesegment fromapointP onthecurvetothex-axisalongthetangent linehasconstant length
. Let y = f (x) betheequationof thetractrix.
(a) Showthat y
2
+(y/y
)
2

2
andconcludey
y
Whymust wechoosethenegativesquareroot?
59. Showthat thedifferential equationsy
= 3y/x andy
= x/3y deneorthogonal familiesof curves; that is, the

graphsof solutionstotherstequationintersectthegraphsof thesolutionstothesecondequationinrightangles(Figure
10). Findthesecurvesexplicitly.
x
y
FIGURE 10 Twoorthogonal familiesof curves.
solution Let y
1
beasolutiontoy
=
3y
x
andlet y
2
beasolutiontoy
=
x
3y
. Supposethesetwocurvesintersect at
apoint (x
0
, y
0
). Thelinetangent tothecurvey
1
(x) at (x
0
, y
0
) hasaslopeof
3y
0
x
0
andthelinetangent tothecurvey
2
(x)
hasaslopeof
x
0
3y
0
. Theslopesarenegativereciprocalsof oneanother; hencethetangent linesareperpendicular.
Separationof variablesandintegrationappliedtoy
=
3y
x
gives
dy
y
= 3
dx
x
ln|y| = 3ln|x| +C
y = Ax
3
Ontheother hand, separationof variablesandintegrationappliedtoy
=
x
3y
gives
3y dy = x dx
3y
2
/2= x
2
/2+C
y =
_
C x
2
/3
Findthefamilyof curvessatisfyingy
= x/y andsketchseveral membersof thefamily. Thenndthedifferential

equationfortheorthogonal family(seeExercise59),nditsgeneral solution,andaddsomemembersof thisorthogonal
familytoyour plot.
61. A 50-kgmodel rocket lifts off by expellingfuel downwardat arateof k = 4.75kg/s for 10s. Thefuel leaves the
endof therocket withanexhaust velocityof b = 100m/s. Let m(t ) bethemassof therocket at timet . Fromthelaw
of conservationof momentum, wendthefollowingdifferential equationfor therockets velocity v(t ) (inmeters per
second):
m(t )v
(t ) = 9.8m(t ) +b
dm
dt
(a) Showthat m(t ) = 504.75t kg.
(b) Solvefor v(t ) andcomputetherocketsvelocityat rocket burnout (after 10s).
solution
(a) For 0 t 10, the rocket is expelling fuel at a constant rate of 4.75 kg/s, giving m
(t ) = 4.75. Hence,
m(t ) = 4.75t +C. Initially, therocket hasamassof 50kg, soC = 50. Therefore, m(t ) = 504.75t .
(b) Withm(t ) = 504.75t and
dm
dt
= 4.75, theequationfor v becomes
dv
dt
= 9.8+
b
dm
dt
504.75t
= 9.8+
(100)(4.75)
504.75t
andtherefore
v(t ) = 9.8t +100
_
4.75dt
504.75t
= 9.8t 100ln(504.75t ) +C
Becausev(0) = 0, wendC = 100ln50and
v(t ) = 9.8t 100ln(504.75t ) +100ln(50).
After 10secondsthevelocityis:
v(10) = 98100ln(2.5) +100ln(50) 201.573m/s.
Letv(t ) bethevelocityof anobjectof massminfreefall near theearthssurface. If weassumethatair resistance
isproportional tov
2
, thenv satisesthedifferential equationm
dv
dt
= g +kv
2
for someconstant k > 0.
(a) Set = (g/k)
1/2
andrewritethedifferential equationas
d k
63. If abucket of water spins about avertical axis withconstant angular velocity (inradians per second), thewater
climbsupthesideof thebucket until it reachesanequilibriumposition(Figure11). Twoforcesact onaparticlelocated
at adistancex fromthevertical axis: thegravitational forcemg actingdownwardandtheforceof thebucket onthe
particle(transmittedindirectly throughtheliquid) inthedirectionperpendicular to thesurfaceof thewater. Thesetwo
forcesmust combinetosupply acentripetal forcem
2
x, andthisoccursif thediagonal of therectangleinFigure11is
normal to thewaters surface(that is, perpendicular to thetangent line). Provethat if y = f (x) is theequationof the
curveobtainedby takingavertical cross sectionthroughtheaxis, then1/y
= g/(
2
x). Showthat y = f (x) is a
parabola.
mg
m
2
x
x
x
y
y = f (x)
FIGURE 11
solution At any point alongthesurfaceof thewater, theslopeof thetangent lineisgivenby thevalueof y
at that
point; hence, theslopeof thelineperpendicular tothesurfaceof thewater isgivenby1/y
. Theslopeof theresultant
forcegeneratedbythegravitational forceandthecentrifugal forceis
mg
m
2
x
=
g
2
x
.
Therefore, thecurveobtainedbytakingavertical cross-sectionof thewater surfaceisdeterminedbytheequation
1
y
=
g
2
x
or y
=

2
g
x.
Performingoneintegrationyields
y = f (x) =

2
2g
x
2
+C,
whereC isaconstant of integration. Thus, y = f (x) isaparabola.
InSection6.2, wecomputedthevolumeV of asolidastheintegral of cross-sectional area. Explainthisformula
intermsof differential equations. LetV(y) bethevolumeof thesoliduptoheighty, andletA(y) bethecross-sectional
areaat height y asinFigure12.
(a) Explainthefollowingapproximationfor small y:
V(y +y) V(y) A(y) y
(b) UseEq. (8) tojustifythedifferential equationdV/dy = A(y). Thenderivetheformula
V =
_
b
a
A(y) dy
65. A basictheoremstatesthat alinear differential equationof order n hasageneral solutionthat dependsonn arbitrary
constants. Thereare, however, nonlinear exceptions.
(a) Showthat (y
)
2
+y
2
= 0isarst-order equationwithonlyonesolutiony = 0.
(b) Showthat (y
)
2
+y
2
+1= 0isarst-order equationwithnosolutions.
solution
(a) (y
)
2
+y
2
0andequalszeroif andonlyif y
= 0andy = 0
(b) (y
)
2
+y
2
+1 1> 0for all y
andy, so(y
)
2
+y
2
+1= 0hasnosolution
Showthat y = Ce
rx
isasolutionof y
+ ay
+ by = 0if andonly if r isaroot of P(r) = r

2
+ ar + b. Then
verifydirectlythat y = C
1
e
3x
+C
2
e
x
isasolutionof y
2y
3y = 0for anyconstantsC
1
, C
2
.
67. A spherical tank of radiusR ishalf-lledwithwater. Supposethat water leaksthroughaholeinthebottomof area
B. Let y(t ) bethewater level at timet (seconds).
(a) Showthat
dy
dt
=

2gB
y
(2Ry y
2
)
.
(b) Showthat for someconstant C,
2
15B
2g
_
10Ry
3/2
3y
5/2
_
= C t
(c) Usetheinitial conditiony(0) = R tocomputeC, andshowthat C = t
e
, thetimeat whichthetankisempty.
(d) Showthat t
e
isproportional toR
5/2
andinverselyproportional toB.
solution
(a) At height y abovethebottomof thetank, thecrosssectionisacircleof radius
r =
_
R
2
(R y)
2
=
_
2Ry y
2
.
Thecross-sectional areafunctionisthenA(y) = (2Ry y
2
). Thedifferential equationfor theheight of thewater in
thetankisthen
dy
dt
=
2gB
y
(2Ry y
2
)
byTorricellislaw.
(b) Rewritethedifferential equationas
2gB
_
2Ry
1/2
y
3/2
_
dy = dt,
2
2gB
_
2
3
Ry
3/2
1
5
y
5/2
_
= C t,
whereC isanarbitraryconstant. Simplifyinggives
2
15B
2g
(10Ry
3/2
3y
5/2
) = C t (*)
(c) FromEquation (*) weseethat y = 0 when t = C. It follows that C = t
e
, thetimeat which thetank is empty.
Moreover, theinitial conditiony(0) = R allowsustodeterminethevalueof C:
2
15B
2g
(10R
5/2
3R
5/2
) =
14
15B
2g
R
5/2
= C
(d) Frompart (c),
t
e
=
14
15
2g
R
5/2
B
,
fromwhichit isclear that t
e
isproportional toR
5/2
andinverselyproportional toB.
10.2 Graphical and Numerical Methods
1. What istheslopeof thesegment intheslopeeldfor

y = ty +1at thepoint (2, 3)?
solution Theslopeof thesegment intheslopeeldfor

y = ty +1at thepoint (2, 3) is(2)(3) +1= 7.
2. What istheequationof theisoclineof slopec = 1for

y = y
2
t ?
solution Theisoclineof slopec = 1hasequationy
2
t = 1, or y =
1+t .
3. For whichof thefollowingdifferential equationsaretheslopesat pointsonavertical linet = C all equal?
(a)

y = lny (b)

y = lnt
solution Only for theequationinpart (b). Theslopeat apoint issimply thevalueof y at that point, sofor part (a),
theslopedependsony, whilefor part (b), theslopedependsonlyont .
4. Let y(t ) bethesolution to

y = F(t, y) with y(1) = 3. How many iterations of Eulers Method arerequired to
approximatey(3) if thetimestepish = 0.1?
solution Theinitial conditionisspeciedatt = 1andwewanttoobtainanapproximationtothevalueof thesolution
at t = 3. Withatimestepof h = 0.1,
31
0.1
= 20
iterationsof Eulersmethodarerequired.
S E C T I ON 10.2 Graphical and Numerical Methods 615
Exercises
1. Figure8showstheslopeeldfor

y = siny sint . Sketchthegraphsof thesolutionswithinitial conditionsy(0) = 1
andy(0) = 1. Showthat y(t ) = 0isasolutionandadditsgraphtotheplot.
0 3
3
3 1 2 2 1
t 0
2
1
1
2
3
y
FIGURE 8 Slopeeldfor

y = siny sint .
solution Thesketchesof thesolutionsappear below.
0 3
3
3 1 2 2 1
t 0
2
1
1
2
3
y
If y(t ) = 0, theny
= 0; moreover, sin0sint = 0. Thus, y(t ) = 0isasolutionof

y = siny sint .
Figure9showstheslopeeldfor

y = y
2
t
2
. Sketchtheintegral curvepassingthroughthepoint (0, 1), the
curvethrough(0, 0), andthecurvethrough(0, 2). Isy(t ) = 0asolution?
3. Showthatf (t ) =
1
2
_
t
1
2
_
isasolutionto
y = t 2y. Sketchthefour solutionswithy(0) = 0.5, 1ontheslope

eldinFigure10. Theslopeeldsuggeststhat everysolutionapproachesf (t ) ast . Conrmthisbyshowingthat
y = f (t ) +Ce
2t
isthegeneral solution.
t
1 0.5 2 1 0.5 1.5 0
1
0.5
0
0.5
1
y
y = (t )
1
2
1
2
FIGURE 10 Slopeeldfor

y = t 2y.
solution Let y = f (t ) =
1
2
(t
1
2
). Then
y =
1
2
and
y +2y =
1
2
+t
1
2
= t,
sof (t ) =
1
2
(t
1
2
) isasolutionto

y = t 2y. Theslopeeldwiththefour requiredsolutionsisshownbelow.
0 1 1 2
0
1
1
y
t
Now, let y = f (t ) +Ce
2t
=
1
2
(t
1
2
) +Ce
2t
. Then
y =
1
2
2Ce
2t
,
and
y +2y =
1
2
2Ce
2t
+
_
t
1
2
_
+2Ce
2t
= t.
Thus, y = f (t ) +Ce
2t
isthegeneral solutiontotheequation
y = t 2y.
Oneof theslopeeldsinFigures11(A) and(B) istheslopeeldfor

y = t
2
. Theother isfor

y = y
2
. Identify
whichiswhich. Ineachcase, sketchthesolutionswithinitial conditionsy(0) = 1, y(0) = 0, andy(0) = 1.
5. Consider thedifferential equation
y = t y.
(a) Sketchtheslopeeldof thedifferential equation
y = t y intherange1 t 3, 1 y 3.Asanaid, observe

that theisoclineof slopec isthelinet y = c, sothesegmentshaveslopec at pointsontheliney = t c.
(b) Showthat y = t 1+ Ce
t
is asolutionfor all C. Since lim
t
e
t
= 0, thesesolutions approachtheparticular
solutiony = t 1ast . Explainhowthisbehavior isreectedinyour slopeeld.
solution
(a) Hereisasketchof theslopeeld:
0 3
1
1 1 2
t 0
1
2
3
y
(b) Let y = t 1+Ce
t
. Then
y = 1C
t
, and
t y = t (t 1+Ce
t
) = 1Ce
t
.
Thus, y = t 1+ Ce
t
is asolutionof

y = t y. Ontheslopeeld, wecanseethat theisoclines of 1all liealong
theliney = t 1. Whenever y > t 1,

y = t y < 1, so thesolutioncurvewill convergedownwardtowards the
liney = t 1. Ontheother hand, if y < t 1,

y = t y > 1, so thesolutioncurvewill convergeupwardtowards
y = t 1. Ineither case, thesolutionisapproachingt 1.
Showthat theisoclinesof

y = 1/y arehorizontal lines. Sketchtheslopeeldfor 2 t 2, 2 y 2and
plot thesolutionswithinitial conditionsy(0) = 0andy(0) = 1.
7. Showthat theisoclinesof

y = t arevertical lines. Sketchtheslopeeldfor 2 t 2, 2 y 2andplot the
integral curvespassingthrough(0, 1) and(0, 1).
solution Theisoclineof slopec for thedifferential equation

y = t has equationt = c, whichis theequationof a
vertical line. Theslopeeldandtherequiredsolutioncurvesareshownbelow.
2
2
1
2 1 0 1
t
2
1
0
y
Sketchtheslopeeldof

y = ty for 2 t 2, 2 y 2. Basedonthesketch, determine lim
t
y(t ), where
y(t ) isasolutionwithy(0) > 0. What is lim
t
y(t ) if y(0) < 0?
9. Matcheachdifferential equationwithitsslopeeldinFigures12(A)(F).
(i)

y = 1
(ii)

y =
y
t
(iii)

y = t
2
y
(iv)

y = ty
2
(v)

y = t
2
+y
2
(vi)

y = t
0 3
3
3 1 2 2 1
t 0
2
1
1
2
3
y
FIGURE 12(A)
0 3
3
3 1 2 2 1
t 0
2
1
1
2
3
y
FIGURE 12(B)
0 3
3
3 1 2 2 1
t 0
2
1
1
2
3
y
FIGURE 12(C)
0 3
3
3 1 2 2 1
t 0
2
1
1
2
3
y
FIGURE 12(D)
0 3
3
3 1 2 2 1
t 0
2
1
1
2
3
y
FIGURE 12(E)
0 3
3
3 1 2 2 1
t 0
2
1
1
2
3
y
FIGURE 12(F)
solution
(i) Everysegment intheslopeeldfor

y = 1will haveslope1; thismatchesFigure12(C).
(ii) Thesegmentsintheslopeeldfor
y =
y
t
will havepositiveslopeintherstandthirdquadrantsandnegativeslopes
inthesecondandfourthquadrant; thismatchesFigure12(B).
(iii) Thesegmentsintheslopeeldfor
y = t
2
y will havepositiveslopeintheupper half of theplaneandnegativeslopes
inthelower half of theplane; thismatchesFigure12(F).
(iv) Thesegmentsintheslopeeldfor
y = ty
2
will havepositiveslopeontherightsideof theplaneandnegativeslopes
ontheleft sideof theplane; thismatchesFigure12(D).
(v) Every segment in the slope eld for

y = t
2
+ y
2
, except at the origin, will have positive slope; this matches
Figure12(A).
(vi) Theisoclinesfor

y = t arevertical lines; thismatchesFigure12(E).
Sketchthesolutionof

y = ty
2
satisfyingy(0) = 1intheappropriateslopeeldof Figure12(A)(F). Thenshow,
usingseparationof variables, thatif y(t ) isasolutionsuchthaty(0) > 0, theny(t ) tendstoinnityast
2/y(0).
11. (a) Sketchtheslopeeldof

y = t /y intheregion2 t 2, 2 y 2.
(b) Checkthat y =
_
t
2
+C isthegeneral solution.
(c) Sketchthesolutionsontheslopeeldwithinitial conditionsy(0) = 1andy(0) = 1.
solution
(a) Theslopeeldisshownbelow:
2
2
1
2 1 0 1
t
2
1
0
y
(b) Rewrite
dy
dt
=
t
y
as y dy = t dt,
1
2
y
2
=
1
2
t
2
+C.
Solvingfor y, wendthat thegeneral solutionis
y =
_
t
2
+C.
(c) Thesketchesof thetwosolutionsareshownbelow:
2
2
1
2 1 0 1
t
2
1
0
y
Sketchtheslopeeldof

y = t
2
y intheregion3 t 3, 3 y 3andsketchthesolutionssatisfying
y(1) = 0, y(1) = 1, andy(1) = 1.
13. Let F(t, y) = t
2
y andlet y(t ) bethesolutionof

y = F(t, y) satisfyingy(2) = 3. Let h = 0.1bethetimestepin
EulersMethod, andset y
0
= y(2) = 3.
(a) Calculatey
1
= y
0
+hF(2, 3).
(b) Calculatey
2
= y
1
+hF(2.1, y
1
).
(c) Calculatey
3
= y
2
+hF(2.2, y
2
) andcontinuecomputingy
4
, y
5
, andy
6
.
(d) Findapproximationstoy(2.2) andy(2.5).
solution
(a) Withy
0
= 3, t
0
= 2, h = 0.1, andF(t, y) = t
2
y, wend
y
1
= y
0
+hF(t
0
, y
0
) = 3+0.1(1) = 3.1.
(b) Withy
1
= 3.1, t
1
= 2.1, h = 0.1, andF(t, y) = t
2
y, wend
y
2
= y
1
+hF(t
1
, y
1
) = 3.1+0.1(4.413.1) = 3.231.
(c) Continuingasintheprevioustwoparts, wend
y
3
= y
2
+hF(t
2
, y
2
) = 3.3919;
y
4
= y
3
+hF(t
3
, y
3
) = 3.58171;
y
5
= y
4
+hF(t
4
, y
4
) = 3.799539;
y
6
= y
5
+hF(t
5
, y
5
) = 4.0445851.
(d) y(2.2) y
2
= 3.231, andy(2.5) y
5
= 3.799539.
Let y(t ) bethesolutionto

y = t e
y
satisfyingy(0) = 0.
(a) UseEulersMethodwithtimesteph = 0.1toapproximatey(0.1), y(0.2), . . . , y(0.5).
(b) Useseparationof variablestondy(t ) exactly.
(c) Computetheerrorsintheapproximationstoy(0.1) andy(0.5).
In Exercises 1520, use Eulers Method to approximate the given value of y(t ) with the time step h indicated.
15. y(0.5);

y = y +t , y(0) = 1, h = 0.1
solution Withy
0
= 1, t
0
= 0, h = 0.1, andF(t, y) = y +t , wecompute
n t
n
y
n
0 0 1
1 0.1 y
0
+hF(t
0
, y
0
) = 1.1
2 0.2 y
1
+hF(t
1
, y
1
) = 1.22
3 0.3 y
2
+hF(t
2
, y
2
) = 1.362
4 0.4 y
3
+hF(t
3
, y
3
) = 1.5282
5 0.5 y
4
+hF(t
4
, y
4
) = 1.72102
y(0.7);

y = 2y, y(0) = 3, h = 0.1
17. y(3.3);

y = t
2
y, y(3) = 1, h = 0.05
solution Withy
0
= 1, t
0
= 3, h = 0.05, andF(t, y) = t
2
y, wecompute
n t
n
y
n
0 3 1
1 3.05 y
0
+hF(t
0
, y
0
) = 1.4
2 3.1 y
1
+hF(t
1
, y
1
) = 1.795125
3 3.15 y
2
+hF(t
2
, y
2
) = 2.185869
4 3.2 y
3
+hF(t
3
, y
3
) = 2.572700
5 3.25 y
4
+hF(t
4
, y
4
) = 2.956065
6 3.3 y
5
+hF(t
5
, y
5
) = 3.336387
y(3);

y =
t +y, y(2.7) = 5, h = 0.05

19. y(2);

y = t siny, y(1) = 2, h = 0.2
solution Let F(t, y) = t siny. Witht
0
= 1, y
0
= 2andh = 0.2, wecompute
n t
n
y
n
0 1 2
1 1.2 y
0
+hF(t
0
, y
0
) = 2.181859
2 1.4 y
1
+hF(t
1
, y
1
) = 2.378429
3 1.6 y
2
+hF(t
2
, y
2
) = 2.571968
4 1.8 y
3
+hF(t
3
, y
3
) = 2.744549
5 2.0 y
4
+hF(t
4
, y
4
) = 2.883759
y(5.2);

y = t secy, y(4) = 2, h = 0.2
21. If f (t ) iscontinuouson[a, b], thenthesolutionto

y = f (t ) withinitial conditiony(a) = 0isy(t ) =
_
t
a
f (u) du.
Show that Eulers Method with timestep h = (b a)/N for N steps yields theNth left-endpoint approximation to
y(b) =
_
b
a
f (u) du.
solution For adifferential equationof theform
y = f (t ), theequationfor Eulersmethodreducesto
y
k
= y
k1
+hf (t
k1
).
Withastepsizeof h = (b a)/N, y(b) = y
N
. Startingfromy
0
= 0, wecompute
y
1
= y
0
+hf (t
0
) = hf (t
0
)
y
2
= y
1
+hf (t
1
) = h[f (t
0
) +f (t
1
)]
y
3
= y
2
+hf (t
2
) = h[f (t
0
) +f (t
1
) +f (t
2
)]
.
.
.
y
N
= y
N
1
+hf (t
N1
) = h
_
f (t
0
) +f (t
1
) +f (t
2
) +. . . +f (t
N1
)
_
= h
N1
k=0
f (t
k
)
Observethislast expressionisexactlytheNthleft-endpoint approximationtoy(b) =
_
b
a
f (u) du.
Exercises 2227: EulersMidpointMethodis a variation on Eulers Method that is signicantly more accurate in general.
For time step h and initial value y
0
= y(t
0
), the values y
k
are dened successively by
y
k
= y
k1
+hm
k1
where m
k1
= F
_
t
k1
+
h
2
, y
k1
+
h
2
F(t
k1
, y
k1
)
_
.
ApplybothEulersMethodandtheEuler Midpoint Methodwithh = 0.1toestimatey(1.5), wherey(t ) satises
y = y withy(0) = 1. Findy(t ) exactlyandcomputetheerrorsinthesetwoapproximations.

In Exercises 2326, use Eulers Midpoint Method with the time step indicated to approximate the given value of y(t ).
23. y(0.5);

y = y +t , y(0) = 1, h = 0.1
solution Witht
0
= 0, y
0
= 1, F(t, y) = y +t , andh = 0.1wecompute
n t
n
y
n
0 0 1
1 0.1 y
0
+hF(t
0
+h/2, y
0
+(h/2)F(t
0
, y
0
)) = 1.11
2 0.2 y
1
+hF(t
1
+h/2, y
1
+(h/2)F(t
1
, y
1
)) = 1.242050
3 0.3 y
2
+hF(t
2
+h/2, y
2
+(h/2)F(t
2
, y
2
)) = 1.398465
4 0.4 y
3
+hF(t
3
+h/2, y
3
+(h/2)F(t
3
, y
3
)) = 1.581804
5 0.5 y
4
+hF(t
4
+h/2, y
4
+(h/2)F(t
4
, y
4
)) = 1.794894
y(2);

y = t
2
y, y(1) = 3, h = 0.2
25. y(0.25);

y = cos(y +t ), y(0) = 1, h = 0.05
solution Witht
0
= 0, y
0
= 1, F(t, y) = cos(y +t ), andh = 0.05wecompute
n t
n
y
n
0 0 1
1 0.05 y
0
+hF(t
0
+h/2, y
0
+(h/2)F(t
0
, y
0
)) = 1.025375
2 0.10 y
1
+hF(t
1
+h/2, y
1
+(h/2)F(t
1
, y
1
)) = 1.047507
3 0.15 y
2
+hF(t
2
+h/2, y
2
+(h/2)F(t
2
, y
2
)) = 1.066425
4 0.20 y
3
+hF(t
3
+h/2, y
3
+(h/2)F(t
3
, y
3
)) = 1.082186
5 0.25 y
4
+hF(t
4
+h/2, y
4
+(h/2)F(t
4
, y
4
)) = 1.094871
y(2.3);

y = y +t
2
, y(2) = 1, h = 0.05
27. Assume that f (t ) is continuous on [a, b]. Show that Eulers Midpoint Method applied to

y = f (t ) with initial
conditiony(a) = 0andtimesteph = (b a)/N for N stepsyieldstheNthmidpoint approximationto
y(b) =
_
b
a
f (u) du
solution For adifferential equationof theform
y = f (t ), theequationsfor Eulersmidpoint methodreduceto

m
k1
= f
_
t
k1
+
h
2
_
and y
k
= y
k1
+hf
_
t
k1
+
h
2
_
.
Withastepsizeof h = (b a)/N, y(b) = y
N
. Startingfromy
0
= 0, wecompute
y
1
= y
0
+hf
_
t
0
+
h
2
_
= hf
_
t
0
+
h
2
_
y
2
= y
1
+hf
_
t
1
+
h
2
_
= h
_
f
_
t
0
+
h
2
_
+f
_
t
1
+
h
2
__
y
3
= y
2
+hf
_
t
2
+
h
2
_
= h
_
f
_
t
0
+
h
2
_
+f
_
t
1
+
h
2
_
+f
_
t
2
+
h
2
__
.
.
.
y
N
= y
N
1
+hf
_
t
N1
+
h
2
_
= h
_
f
_
t
0
+
h
2
_
+f
_
t
1
+
h
2
_
+f
_
t
2
+
h
2
_
+. . . +f
_
t
N1
+
h
2
__
= h
N1
k=0
f
_
t
k
+
h
2
_
Observethislast expressionisexactlytheNthmidpoint approximationtoy(b) =
_
b
a
f (u) du.
S E C T I ON 10.3 The Logistic Equation 621
10.3 The Logistic Equation
1. Whichof thefollowingdifferential equationsisalogisticdifferential equation?
(a)

y = 2y(1y
2
) (b)

y = 2y
_
1
y
3
_
(c)

y = 2y
_
1
t
4
_
(d)

y = 2y(13y)
solution Thedifferential equationsin(b)and(d)arelogisticequations. Theequationin(a)isnot alogisticequation
becauseof they
2
terminsidetheparenthesesontheright-handside; theequationin(c)isnotalogisticequationbecause
of thepresenceof theindependent variableontheright-handside.
2. Isthelogisticequationalinear differential equation?
solution No, thelogisticequationisnot linear.
y = ky
_
1
y
A
_
canberewritten

y = ky
k
A
y
2
andweseethat aterminvolvingy
2
occurs.
3. Isthelogisticequationseparable?
solution Yes, thelogisticequationisaseparabledifferential equation.
Exercises
1. Findthegeneral solutionof thelogisticequation
y = 3y
_
1
y
5
_
Thenndtheparticular solutionsatisfyingy(0) = 2.
solution

y = 3y(1y/5) isalogisticequationwithk = 3andA = 5; therefore, thegeneral solutionis
y =
5
1e
3t
/C
.
Theinitial conditiony(0) = 2allowsustodeterminethevalueof C:
2=
5
11/C
; 1
1
C
=
5
2
; so C =
2
3
.
Theparticular solutionisthen
y =
5
1+
3
2
e
3t
=
10
2+3e
3t
.
Findthesolutionof

y = 2y(3y), y(0) = 10.
3. Lety(t ) beasolutionof

y = 0.5y(10.5y) suchthaty(0) = 4. Determine lim
t
y(t ) withoutndingy(t ) explicitly.
solution Thisisalogisticequationwithk =
1
2
andA = 2, sothecarryingcapacityis2. Thustherequiredlimit is2.
Lety(t ) beasolutionof

y = 5y(1y/5). Statewhethery(t ) isincreasing, decreasing, orconstantinthefollowing
cases:
(a) y(0) = 2 (b) y(0) = 5 (c) y(0) = 8
5. A population of squirrels lives in aforest with acarrying capacity of 2000. Assumelogistic growth with growth
constant k = 0.6yr
1
.
(a) Findaformulafor thesquirrel populationP(t ), assuminganinitial populationof 500squirrels.
(b) Howlongwill it takefor thesquirrel populationtodouble?
solution
(a) Sincek = 0.6andthecarryingcapacity is A = 2000, thepopulationP(t ) of thesquirrels satises thedifferential
equation
P
(t ) = 0.6P(t )(1P(t )/2000),

withgeneral solution
P(t ) =
2000
1e
0.6t
/C
.
Theinitial conditionP(0) = 500allowsustodeterminethevalueof C:
500=
2000
11/C
; 1
1
C
= 4; so C =
1
3
.
Theformulafor thepopulationisthen
P(t ) =
2000
1+3e
0.6t
.
(b) Thesquirrel populationwill havedoubledat thetimet whereP(t ) = 1000. Thisgives
1000=
2000
1+3e
0.6t
; 1+3e
0.6t
= 2; so t =
5
3
ln3 1.83.
It thereforetakesapproximately1.83yearsfor thesquirrel populationtodouble.
ThepopulationP(t ) of mosquitolarvaegrowinginatreeholeincreasesaccordingtothelogistic equationwith
growthconstant k = 0.3day
1
andcarryingcapacityA = 500.
(a) Findaformulafor thelarvaepopulationP(t ), assuminganinitial populationof P
0
= 50larvae.
(b) After howmanydayswill thelarvaepopulationreach200?
7. SunsetLakeisstockedwith2000rainbowtrout, andafter1yearthepopulationhasgrownto4500.Assuminglogistic
growthwithacarryingcapacityof 20,000, ndthegrowthconstantk (specifytheunits)anddeterminewhenthepopulation
will increaseto10,000.
solution SinceA = 20,000, thetrout populationP(t ) satisesthelogisticequation
P
(t ) = kP(t )(1P(t )/20,000),

withgeneral solution
P(t ) =
20,000
1e
kt
/C
.
Theinitial conditionP(0) = 2000allowsustodeterminethevalueof C:
2000=
20,000
11/C
; 1
1
C
= 10; so C =
1
9
.
After oneyear, weknowthepopulationhasgrownto4500. Letsmeasuretimeinyears. Then
4500=
20,000
1+9e
k
1+9e
k
=
40
9
e
k
=
31
81
k = ln
81
31
0.9605years
1
.
Thepopulationwill increaseto10,000at timet whereP(t ) = 10,000. Thisgives
10,000=
20,000
1+9e
0.9605t
1+9e
0.9605t
= 2
e
0.9605t
=
1
9
t =
1
0.9605
ln9 2.29years.
Spreadof aRumor A rumor spreadsthroughasmall town. Let y(t ) bethefractionof thepopulationthat has
heardtherumor at timet andassumethat therateat whichtherumor spreadsisproportional totheproduct of the
fractiony of thepopulationthat hasheardtherumor andthefraction1y that hasnot yet heardtherumor.
(a) Writedownthedifferential equationsatisedbyy intermsof aproportionalityfactor k.
(b) Findk (inunitsof day
1
), assumingthat 10%of thepopulationknowstherumor at t = 0and40%knowsit at
t = 2days.
(c) Usingtheassumptionsof part (b), determinewhen75%of thepopulationwill knowtherumor.
9. A rumor spreadsthroughaschool with1000students. At 8am, 80studentshaveheardtherumor, andbynoon, half
theschool hasheardit. Usingthelogisticmodel of Exercise8, determinewhen90%of thestudentswill haveheardthe
rumor.
solution Let y(t ) betheproportionof studentsthat haveheardtherumor at atimet hoursafter 8am. Inthelogistic
model of Exercise8, wehaveacapacityof A = 1(100%of students) andanunknowngrowthfactor of k. Hence,
y(t ) =
1
1e
kt
/C
.
Theinitial conditiony(0) = 0.08allowsustodeterminethevalueof C:
2
25
=
1
11/C
; 1
1
C
=
25
2
; so C =
2
23
.
sothat
y(t ) =
2
2+23e
kt
.
Theconditiony(4) = 0.5nowallowsustodeterminethevalueof k:
1
2
=
2
2+23e
4k
; 2+23e
4k
= 4; so k =
1
4
ln
23
2
0.6106hours
1
.
90%of thestudentshaveheardtherumor wheny(t ) = 0.9. Thus
9
10
=
2
2+23e
0.6106t
2+23e
0.6106t
=
20
9
t =
1
0.6106
ln
207
2
7.6hours.
Thus, 90%of thestudentshaveheardtherumor after 7.6hours, or at 3:36pm.
A simpler model for thespreadof arumor assumes that therateat whichtherumor spreads is proportional
(withfactor k) tothefractionof thepopulationthat hasnot yet heardtherumor.
(a) Computethesolutionstothismodel andthemodel of Exercise8withthevaluesk = 0.9andy
0
= 0.1.
(b) Graphthetwosolutionsonthesameaxis.
(c) Whichmodel seemsmorerealistic?Why?
11. Let k = 1andA = 1inthelogisticequation.
(a) Findthesolutionssatisfyingy
1
(0) = 10andy
2
(0) = 1.
(b) Findthetimet wheny
1
(t ) = 5.
(c) Whendoesy
2
(t ) becomeinnite?
solution Thegeneral solutionof thelogisticequationwithk = 1andA = 1is
y(t ) =
1
1e
t
/C
.
(a) Giveny
1
(0) = 10, wendC =
10
9
, and
y
1
(t ) =
1
1
10
9
e
t
=
10
109e
t
.
Ontheother hand, giveny
2
(0) = 1, wendC =
1
2
, and
y
2
(t ) =
1
12e
t
.
(b) Frompart (a), wehave
y
1
(t ) =
10
109e
t
.
Thus, y
1
(t ) = 5when
5=
10
109e
t
; 109e
t
= 2; so t = ln
9
8
.
(c) Frompart (a), wehave
y
2
(t ) =
1
12e
t
.
Thus, y
2
(t ) becomesinnitewhen
12e
t
= 0 or t = ln2.
A tissueculturegrows until it has amaximumareaof M cm
2
. TheareaA(t ) of thecultureat timet may be
modeledbythedifferential equation
A = k
A
_
1
A
M
_
wherek isagrowthconstant.
(a) Showthat if weset A = u
2
, then
u =
1
2
k
_
1
u
2
M
_
Thenndthegeneral solutionusingseparationof variables.
13. Inthemodel of Exercise12, let A(t ) betheareaat timet (hours) of agrowingtissueculturewithinitial size
A(0) = 1cm
2
, assumingthat themaximumareaisM = 16cm
2
andthegrowthconstant isk = 0.1.
(a) Findaformulafor A(t ). Note: Theinitial conditionissatisedfor twovaluesof theconstantC. Choosethevalueof
C for whichA(t ) isincreasing.
(b) Determinetheareaof thecultureat t = 10hours.
(c) Graphthesolutionusingagraphingutility.
solution
(a) Fromthevaluesfor M andk wehave
A(t ) = 16
_
Ce
t /40
1
Ce
t /40
+1
_
2
andtheinitial conditionthengivesus
A(0) = 1= 16
_
Ce
0/40
1
Ce
0/40
+1
_
2
so, simplifying,
1= 16
_
C 1
C +1
_
2
C
2
+2C +1= 16C
2
32C +16 15C
2
34C +15= 0
andthusC =
5
3
or C =
3
5
. Thederivativeof A(t ) is
A
(t ) =
16Ce
t /40
(Ce
t /40
+1)
3
(Ce
t /40
1)
For C = 3/5, A
(t ) canbenegative, whilefor C = 5/3, it isalwayspositive. Solet C = 5/3.

(b) Frompart (a), wehave
A(t ) = 16
_
5
3
e
t /40
1
5
3
e
t /40
+1
_
2
andA(10) 2.11.
(c)
50
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
y
t
100 150 200
Showthatif atissueculturegrowsaccordingtoEq. (7), thenthegrowthratereachesamaximumwhenA = M/3.
15. In 1751, Benjamin Franklin predicted that the U.S. population P(t ) would increase with growth constant k =
0.028year
1
. Accordingto thecensus, theU.S. populationwas 5millionin1800and76millionin1900. Assuming
logistic growthwithk = 0.028, ndthepredictedcarryingcapacity for theU.S. population. Hint: UseEqs. (3) and(4)
toshowthat
P(t )
P(t ) A
=
P
0
P
0
A
e
kt
solution Assumingthepopulationgrowsaccordingtothelogisticequation,
P(t )
P(t ) A
= Ce
kt
.
But
C =
P
0
P
0
A
,
so
P(t )
P(t ) A
=
P
0
P
0
A
e
kt
.
Now, let t = 0correspondtotheyear 1800. Thentheyear 1900correspondstot = 100, andwithk = 0.028, wehave
76
76A
=
5
5A
e
(0.028)(100)
.
Solvingfor A, wend
A =
5(e
2.8
1)
5
76
e
2.8
1
943.07.
Thus, thepredictedcarryingcapacityfor theU.S. populationisapproximately943million.
ReverseLogisticEquation Consider thefollowinglogisticequation(withk, B > 0):
dP
dt
= kP
_
1
P
B
_
(a) Sketchtheslopeeldof thisequation.
(b) Thegeneral solutionisP(t ) = B/(1 e
kt
/C), whereC isanonzeroconstant. Showthat P(0) > B if C > 1
and0< P(0) B.
(d) Showthat P = 0isastableequilibriumandP = B anunstableequilibrium.
In Exercises 17 and 18, let y(t ) be a solution of the logistic equation
dy
dt
= ky
_
1
y
A
_
9
where A > 0and k > 0.
17. (a) DifferentiateEq. (9) withrespect tot andusetheChainRuletoshowthat
d
2
y
dt
2
= k
2
y
_
1
y
A
_
_
1
2y
A
_
(b) Showthat y(t ) isconcaveupif 0< y < A/2andconcavedownif A/2< y < A.
(c) Showthat if 0< y(0) < A/2, theny(t ) hasapoint of inectionat y = A/2(Figure6).
(d) Assumethat 0< y(0) < A/2. Findthetimet wheny(t ) reachestheinectionpoint.
A
A
2
y
y(0)
t
Inflection point
FIGURE 6 Aninectionpoint occursat y = A/2inthelogisticcurve.
solution
(a) Thederivativeof Eq. (9) withrespect tot is
y
= ky
2kyy
A
= ky
_
1
2y
A
_
= k
_
1
y
A
_
ky
_
1
2y
A
_
= k
2
y
_
1
y
A
_
_
1
2y
A
_
.
(b) If 0 < y < A/2, 1
y
A
and1
2y
A
arebothpositive, so y
> 0. Therefore, y is concaveup. If A/2 < y < A,

1
y
A
> 0, but 1
2y
A
< 0, soy
< 0, soy isconcavedown.

(c) If y
0
< A, y growsand lim
t
y(t ) = A. If 0 < y < A/2, y isconcaveupat rst. Oncey passesA/2, y becomes
concavedown, soy hasaninectionpoint at y = A/2.
(d) Thegeneral solutiontoEq. (9) is
y =
A
1e
kt
/C
;
thus, y = A/2when
A
2
=
A
1e
kt
/C
1e
kt
/C = 2
t =
1
k
ln(C)
Now, C = y
0
/(y
0
A), so
t =
1
k
ln
y
0
A y
0
=
1
k
ln
A y
0
y
0
.
Let y =
A
1e
kt
/C
bethegeneral nonequilibriumEq. (9). If y(t ) hasavertical asymptoteat t = t
b
, that is, if
lim
t t
b
y(t ) = , wesaythat thesolutionblowsup at t = t

b
.
(a) Showthat if 0< y(0) < A theny doesnot blowupat anytimet
b
10.4 First-Order Linear Equations
1. Whichof thefollowingarerst-order linear equations?
(a) y
+x
2
y = 1 (b) y
+xy
2
= 1
(c) x
5
y
+y = e
x
(d) x
5
y
+y = e
y
solution Theequations in (a) and (c) arerst-order linear differential equations. Theequation in (b) is not linear
becauseof they
2
factor inthesecondtermontheleft-handsideof theequation; theequationin(d)isnot linear because
of thee
y
termontheright-handsideof theequation.
2. If (x) isanintegratingfactor for y
+A(x)y = B(x), then
(x) isequal to(choosethecorrect answer):

(a) B(x) (b) (x)A(x)
(c) (x)A
(x) (d) (x)B(x)

solution Thecorrect answer is(b): (x)A(x).
Exercises
1. Consider y
+x
1
y = x
3
.
(a) Verifythat (x) = x isanintegratingfactor.
(b) Showthat whenmultipliedby(x), thedifferential equationcanbewritten(xy)
= x
4
.
(c) Concludethat xy isanantiderivativeof x
4
andusethisinformationtondthegeneral solution.
solution
(a) Theequationisof theform
y
+A(x)y = B(x)
for A(x) = x
1
andB(x) = x
3
. ByTheorem1, (x) isdenedby
(x) = e
_
A(x) dx
= e
lnx
= x.
(b) Whenmultipliedby(x), theequationbecomes:
xy
+y = x
4
.
Now, xy
+y = xy
+(x)
y = (xy)
, so
(xy)
= x
4
.
(c) Since(xy)
= x
4
, (xy) =
x
5
5
+C and
y =
x
4
5
+
C
x
(d) If y(1) = 0, wend
0=
1
5
+C so
1
5
= C.
Thesolution, therefore, is
y =
x
4
5

1
5x
.
Consider
dy
dt
+2y = e
3t
.
(a) Verifythat (t ) = e
2t
isanintegratingfactor.
(b) UseEq. (4) tondthegeneral solution.
(c) Findtheparticular solutionwithinitial conditiony(0) = 1.
3. Let (x) = e
x
2
. Verifytheidentity
((x)y)
= (x)(y
+2xy)
andexplainhowit isusedtondthegeneral solutionof
y
+2xy = x
solution Let (x) = e
x
2
. Then
((x)y)
= (e
x
2
y)
= 2xe
x
2
y +e
x
2
y
= e
x
2 _
2xy +y
_
= (x)
_
y
+2xy
_
.
S E C T I ON 10.4 First-Order Linear Equations 627
If wenowmultiplybothsidesof thedifferential equationy
+2xy = x by(x), weobtain

(x)(y
+2xy) = x(x) = xe
x
2
.
But (x)(y
+2xy) = ((x)y)
, sobyintegrationwend
(x)y =
_
xe
x
2
dx =
1
2
e
x
2
+C.
Finally,
y(x) =
1
2
+Ce
x
2
.
Findthesolutionof y
y = e
2x
, y(0) = 1.
In Exercises 518, nd the general solution of the rst-order linear differential equation.
5. xy
+y = x
solution Rewritetheequationas
y
+
1
x
y = 1,
whichisinstandardlinear formwithA(x) =
1
x
andB(x) = 1. ByTheorem1, theintegratingfactor is
(x) = e
_
A(x) dx
= e
lnx
= x.
Whenmultipliedbytheintegratingfactor, therewrittendifferential equationbecomes
xy
+y = x or (xy)
= x.
Integrationof bothsidesnowyields
xy =
1
2
x
2
+C.
Finally,
y(x) =
1
2
x +
C
x
.
xy
y = x
2
x
7. 3xy
y = x
1
y
1
3x
y =
1
3x
2
,
whichisinstandardformwithA(x) =
1
3
x
1
andB(x) =
1
3
x
2
. ByTheorem1, theintegratingfactor is
(x) = e
_
A(x) dx
= e
(1/3) lnx
= x
1/3
.
x
1/3
y
1
3
x
4/3
=
1
3
x
7/3
or (x
1/3
y)
=
1
3
x
7/3
.
x
1/3
y =
1
4
x
4/3
+C.
Finally,
y(x) =
1
4
x
1
+Cx
1/3
.
y
+xy = x
9. y
+3x
1
y = x +x
1
solution ThisequationisinstandardformwithA(x) = 3x
1
andB(x) = x +x
1
. ByTheorem1, theintegrating
factor is
(x) = e
_
3x
1
= e
3lnx
= x
3
.
Whenmultipliedbytheintegratingfactor, theoriginal differential equationbecomes
x
3
y
+3x
2
y = x
4
+x
2
or (x
3
y)
= x
4
+x
3
.
x
3
y =
1
5
x
5
+
1
3
x
3
+C.
Finally,
y(x) =
1
5
x
2
+
1
3
+Cx
3
.
y
+x
1
y = cos(x
2
)
11. xy
= y x
y
1
x
y = 1,
1
x
andB(x) = 1. ByTheorem1, theintegratingfactor is
(x) = e
_
(1/x) dx
= e
lnx
= x
1
.
1
x
y
1
x
2
y =
1
x
or
_
1
x
y
_
=
1
x
.
Integrationonbothsidesnowyields
1
x
y = lnx +C.
Finally,
y(x) = x lnx +Cx.
xy
= x
2
3y
x
13. y
+y = e
x
solution ThisequationisinstandardformwithA(x) = 1andB(x) = e
x
(x) = e
_
1dx
= e
x
.
e
x
y
+e
x
y = e
2x
or (e
x
y)
= e
2x
.
e
x
y =
1
2
e
2x
+C.
Finally,
y(x) =
1
2
e
x
+Ce
x
.
y
+(secx)y = cosx
15. y
+(tanx)y = cosx
solution ThisequationisinstandardformwithA(x) = tanx andB(x) = cosx. ByTheorem1, theintegratingfactor
is
(x) = e
_
tanx dx
= e
lnsecx
= secx.
secxy
+secx tanxy = 1 or (y secx)
= 1.
y secx = x +C.
Finally,
y(x) = x cosx +C cosx.
e
2x
y
= 1e
x
y
17. y
(lnx)y = x
x
solution ThisequationisinstandardformwithA(x) = lnx andB(x) = x
x
. ByTheorem1, theintegratingfactor
is
(x) = e
_
lnx dx
= e
xx lnx
=
e
x
x
x
.
x
x
e
x
y
(lnx)x
x
e
x
y = e
x
or (x
x
e
x
y)
= e
x
.
x
x
e
x
y = e
x
+C.
Finally,
y(x) = x
x
+Cx
x
e
x
.
y
+y = cosx
19. y
+3y = e
2x
, y(0) = 1
solution First, wendthegeneral solutionof thedifferential equation. Thislinear equationisinstandardformwith
A(x) = 3andB(x) = e
2x
(x) = e
3x
.
(e
3x
y)
= e
5x
.
(e
3x
y) =
1
5
e
5x
+C;
hence,
y(x) =
1
5
e
2x
+Ce
3x
.
1=
1
5
+C so C =
6
5
.
Thesolutiontotheinitial valueproblemistherefore
y(x) =
1
5
e
2x
6
5
e
3x
.
xy
+y = e
x
, y(1) = 3
21. y
+
1
x +1
y = x
2
, y(1) = 2
solution First, wendthegeneral solutionof thedifferential equation. Thislinear equationisinstandardformwith
A(x) =
1
x+1
andB(x) = x
2
(x) = e
_
1/(x+1) dx
= e
ln(x+1)
= x +1.
((x +1)y)
= x
1
+x
2
.
(x +1)y = lnx x
1
+C;
hence,
y(x) =
1
x +1
_
C +lnx
1
x
_
.
2=
1
2
(C 1) so C = 5.
y(x) =
1
x +1
_
5+lnx
1
x
_
.
y
+y = sinx, y(0) = 1
23. (sinx)y
= (cosx)y +1, y
_
4
_
= 0
solution First, wendthegeneral solutionof thedifferential equation. Rewritetheequationas
y
(cotx)y = cscx,
whichisinstandardformwithA(x) = cotx andB(x) = cscx. ByTheorem1, theintegratingfactor is
(x) = e
_
cotx dx
= e
lnsinx
= cscx.
(cscxy)
= csc
2
x.
(cscx)y = cotx +C;
hence,
y(x) = cosx +C sinx.
Theinitial conditiony(/4) = 0allowsustodeterminethevalueof C:
0=
2
2
+C
2
2
so C = 1.
y(x) = cosx +sinx.
y
+(sect )y = sect , y
_
4
_
= 1
25. y
+(tanhx)y = 1, y(0) = 3
solution First, wendthegeneral solutionof thedifferential equation. ThisequationisinstandardformwithA(x) =
tanhx andB(x) = 1. ByTheorem1, theintegratingfactor is
(x) = e
_
tanhx dx
= e
lncoshx
= coshx.
(coshxy)
= coshx.
(coshxy) = sinhx +C;
hence,
y(x) = tanhx +C sechx.
3= C.
y(x) = tanhx +3sechx.
y
+
x
1+x
2
y =
1
(1+x
2
)
3/2
, y(1) = 0
27. Findthegeneral solutionof y
+ny = e
mx
for all m, n. Note: Thecasem = n must betreatedseparately.
solution For anym, n, Theorem1givesustheformulafor (x):
(x) = e
_
n dx
= e
nx
.
(e
nx
y)
= e
(m+n)x
.
If m = n, integrationonbothsidesyields
e
nx
y =
1
m+n
e
(m+n)x
+C,
so
y(x) =
1
m+n
e
mx
+Ce
nx
.
However, if m = n, thenm+n = 0andtheequationreducesto
(e
nx
y)
= 1,
sointegrationyields
e
nx
y = x +C or y(x) = (x +C)e
nx
.
Findthegeneral solutionof y
+ny = cosx for all n.

In Exercises 2932, a 1000 L tank contains 500 L of water with a salt concentration of 10 g/L. Water with a salt
concentration of 50g/L ows into the tank at a rate of 80L/min. The uid mixes instantaneously and is pumped out at a
specied rate R
out
. Let y(t ) denote the quantity of salt in the tank at time t .
29. Assumethat R
out
= 40L/min.
(a) Set upandsolvethedifferential equationfor y(t ).
(b) What isthesalt concentrationwhenthetankoverows?
solution Becausewater owsintothetankattherateof 80L/minbutowsoutattherateof R
out
= 40L/min, there
isanet inowof 40L/min. Therefore, at anytimet , thereare500+40t litersof water inthetank.
(a) Thenet owof salt intothetankat timet is
dy
dt
= salt rateinsalt rateout =
_
80
L
min
_
_
50
g
L
_
_
40
L
min
__
y g
500+40t L
_
= 400040
y
500+40t
Rewritingthislinear equationinstandardform, wehave
dy
dt
+
4
50+4t
y = 4000,
soA(t ) =
4
50+4t
andB(t ) = 4000. ByTheorem1, theintegratingfactor is
(t ) = e
_
4(50+4t )
1
dt
= e
ln(50+4t )
= 50+4t.
((50+4t )y)
= 4000(50+4t ).
(50+4t )y = 200,000t +16,000t
2
+C;
hence,
y(t ) =
200,000t +8000t
2
+C
50+4t
.
10=
C
50
so C = 500.
y(t ) =
200,000t +8000t
2
+500
50+4t
=
250+4000t
2
+100,000t
25+2t
.
(b) Thetankoverowswhent = 25/2= 12.5. Theamount of salt inthetankat that timeis
y(12.5) = 37,505g,
sotheconcentrationof salt is
37,505g
1000L
= 37.505g/L.
Findthesalt concentrationwhenthetankoverows, assumingthat R
out
= 60L/min.
31. Findthelimitingsalt concentrationast assumingthat R
out
= 80L/min.
solution Thetotal volumeof water isnowconstant at 500liters, sothenet owof salt at timet is
dy
dt
_
80
L
min
_
_
50
g
L
_
_
80
L
min
_
_
y g
500L
_
= 4000
8
50
y
Rewritingthisequationinstandardformgives
dy
dt
+
8
50
y = 4000
sothat theintegratingfactor is
e
_
(8/50) dt
= e
0.16t
Multiplyingbothsidesbytheintegratingfactor gives
(e
0.16t
y)
= 4000e
0.16t
Integratebothsidestoget
e
0.16t
y = 25,000e
0.16t
+C sothat y = 25,000+Ce
0.16t
Ast , theexponential termtendstozero, sothat theamount of salt tendsto25,000g, or 50g/L. (Notethat thisis
preciselywhat wouldbeexpectednavely, sincethesalt concentrationowinginisalso50g/L).
Assumingthat R
out
= 120L/min. Findy(t ). Thencalculatethetankvolumeandthesalt concentrationat t = 10
minutes.
33. Water owsintoatankatthevariablerateof R
in
= 20/(1+t ) gal/minandoutattheconstantrateR
out
= 5gal/min.
Let V(t ) bethevolumeof water inthetankat timet .
(a) Set upadifferential equationfor V(t ) andsolveit withtheinitial conditionV(0) = 100.
(b) Findthemaximumvalueof V.
(c) Plot V(t ) andestimatethetimet whenthetankisempty.
solution
(a) Therateof changeof thevolumeof water inthetankisgivenby
dV
dt
= R
in
R
out
=
20
1+t
5.
Becausetheright-handsidedependsonlyontheindependent variablet , weintegratetoobtain
V(t ) = 20ln(1+t ) 5t +C.
Theinitial conditionV(0) = 100allowsustodeterminethevalueof C:
100= 20ln10+C so C = 100.
Therefore
V(t ) = 20ln(1+t ) 5t +100.
(b) Usingtheresult frompart (a),
dV
dt
=
20
1+t
5= 0
whent = 3. Because
dV
dt
> 0for t < 3and
dV
dt
< 0for t > 3, it followsthat
V(3) = 20ln415+100 112.726gal
isthemaximumvolume.
(c) V(t ) isplottedinthegurebelowat theleft. Ontheright, wezoominnear thelocationwherethecurvecrossesthe
t -axis. Fromthisgraph, weestimatethat thetankisemptyafter roughly34.25minutes.
10 20 30 40
20
40
60
80
100
120
32 34 36 38
A streamfeedsintoalakeat arateof 1000m
3
/day. Thestreamispollutedwithatoxinwhoseconcentrationis
5g/m
3
.Assumethatthelakehasvolume10
6
m
3
andthatwater owsoutof thelakeatthesamerateof 1000m
3
/day.
(a) Set upadifferential equationfor theconcentration c(t ) of toxininthelakeandsolvefor c(t ), assumingthat
c(0) = 0. Hint: Findthedifferential equationfor thequantityof toxiny(t ), andobservethat c(t ) = y(t )/10
6
.
(b) What isthelimitingconcentrationfor larget ?
In Exercises 3538, consider a series circuit (Figure 4) consisting of a resistor of R ohms, an inductor of L henries, and
a variable voltage source of V(t ) volts (time t in seconds). The current through the circuit I (t ) (in amperes) satises the
differential equation
dI
dt
+
R
L
I =
1
L
V(t ) 10
L V(t)
R
FIGURE 4 RL circuit.
35. SolveEq. (10) with initial condition I (0) = 0, assuming that R = 100 , L = 5 H, and V(t ) is constant with
V(t ) = 10volts.
solution If R = 100, V(t ) = 10, andL = 5, thedifferential equationbecomes
dI
dt
+20I = 2,
whichisalinear equationinstandardformwithA(t ) = 20andB(t ) = 2. Theintegratingfactor is(t ) = e
20t
, andwhen
multipliedbytheintegratingfactor, thedifferential equationbecomes
(e
20t
I)
= 2e
20t
.
e
20t
I =
1
10
e
20t
+C;
hence,
I (t ) =
1
10
+Ce
20t
.
Theinitial conditionI (0) = 0allowsustodeterminethevalueof C:
0=
1
10
+C so C =
1
10
.
Finally,
I (t ) =
1
10
_
1e
20t
_
.
Assumethat R = 110, L = 10H, andV(t ) = e
t
volts.
(a) SolveEq. (10) withinitial conditionI (0) = 0.
(b) Calculatet
m
andI (t
m
), wheret
m
isthetimeat whichI (t ) hasamaximumvalue.
(c) Useacomputer algebrasystemtosketchthegraphof thesolutionfor 0 t 3.
37. Assumethat V(t ) = V isconstant andI (0) = 0.
(a) Solvefor I (t ).
(b) Showthat lim
t
I (t ) = V/R andthat I (t ) reachesapproximately63%of itslimitingvalueafter L/R seconds.
(c) Howlongdoesit takefor I (t ) toreach90%of itslimitingvalueif R = 500, L = 4H, andV = 20volts?
solution
(a) Theequation
dI
dt
+
R
L
I =
1
L
V
isalinear equationinstandardformwithA(t ) =
R
L
andB(t ) =
1
L
V(t ). ByTheorem1, theintegratingfactor is
(t ) = e
_
(R/L) dt
= e
(R/L) t
.
(e
(R/L) t
I)
= e
(R/L) t
V
L
.
(e
(R/L) t
I) =
V
R
e
(R/L) t
+C;
hence,
I (t ) =
V
R
+Ce
(R/L) t
.
Theinitial conditionI (0) = 0allowsustodeterminethevalueof C:
0=
V
R
+C so C =
V
R
.
Thereforethecurrent isgivenby
I (t ) =
V
R
_
1e
(R/L) t
_
.
(b) Ast , e
(R/L) t
0, soI (t )
V
R
. Moreover, whent = (L/R) seconds, wehave
I
_
L
R
_
=
V
R
_
1e
(R/L) (L/R)
_
=
V
R
_
1e
1
_
0.632
V
R
.
(c) Usingtheresultsfrompart (a) andpart (b), I (t ) reaches90%of itslimitingvaluewhen
9
10
= 1e
(R/L) t
,
or when
t =
L
R
ln10.
WithL = 4andR = 500, thistakesapproximately0.0184seconds.
Solvefor I (t ), assumingthat R = 500, L = 4H, andV = 20cos(80t ) volts.
39. Tank 1inFigure5islledwithV
1
litersof water containingbluedyeat aninitial concentrationof c
0
g/L.
Water owsintothetankat arateof R L/min, ismixedinstantaneouslywiththedyesolution, andowsout throughthe
bottomat thesamerateR. Let c
1
(t ) bethedyeconcentrationinthetankat timet .
(a) Explainwhyc
1
satisesthedifferential equation
dc
1
dt
=
R
V
1
c
1
.
(b) Solvefor c
1
(t ) withV
1
= 300L, R = 50, andc
0
= 10g/L.
R (L/min)
Tank 2
R (L/min)
R(L/min)
Tank 1
FIGURE 5
solution
(a) Let g
1
(t ) bethenumber of gramsof dyeinthetankat timet . Theng
1
(t ) = V
1
c
1
(t ) andg
1
(t ) = V
1
c
1
(t ). Now,
g
1
(t ) = gramsof dyeingramsof dyeout = 0
g(t )
V
1
g/L R L/min=
R
V
1
g(t )
Substitutinggives
V
1
c
1
(t ) =
R
V
1
c
1
(t )V
1
andsimplifyingyields c
1
(t ) =
R
V
1
c
1
(t )
(b) Instandardform, theequationis
c
1
(t ) +
R
V
1
c
1
(t ) = 0
sothat A(t ) =
R
V
1
andB(t ) = 0. Theintegratingfactor ise
(R/V
1
)t
; multiplyingthroughgives
(e
(R/V
1
)t
c
1
(t ))
= 0 so, integrating, e
(R/V
1
)t
c
1
(t ) = C
andthusc
1
(t ) = Ce
(R/V
1
)t
.WithR = 50andV
1
= 300wehavec
1
(t ) = Ce
t /6
; theinitial conditionc
1
(0) = c
0
= 10
givesC = 10. Finally,
c
1
(t ) = 10e
t /6
Continuingwiththepreviousexercise, let Tank 2beanother tank lledwithV
2
gal of water. Assumethat the
dyesolutionfromTank 1emptiesintoTank 2asinFigure5, mixesinstantaneously, andleavesTank 2at thesame
rateR. Let c
2
(t ) bethedyeconcentrationinTank2at timet .
(a) Explainwhyc
2
satisesthedifferential equation
dc
2
dt
=
R
V
2
(c
1
c
2
)
(b) UsethesolutiontoExercise39tosolvefor c
2
(t ) if V
1
= 300, V
2
= 200, R = 50, andc
0
= 10.
(c) FindthemaximumconcentrationinTank2.
(d) Plot thesolution.
41. Let a, b, r beconstants. Showthat
y = Ce
kt
+a +bk
_
k sinrt r cosrt
k
2
+r
2
_
isageneral solutionof
dy
dt
= k
_
y a b sinrt
_
solution Thisisalinear differential equation; instandardform, it is
dy
dt
+ky = k(a +b sinrt )
Theintegratingfactor isthene
kt
; multiplyingthroughgives
(e
kt
y)
= kae
kt
+kbe
kt
sinrt (*)
Therst termontheright-handsidehasintegral ae
kt
. Tointegratethesecondterm, useintegrationby partstwice; this
result inanequationof theform
_
kbe
kt
sinrt = F(t ) +A
_
kbe
kt
sinrt
for somefunctionF(t ) andconstant A. Solvingfor theintegral gives
_
kbe
kt
sinrt = kbe
kt
k sinrt r cosrt
k
2
+r
2
sothat integratingequation(*) gives
e
kt
y = ae
kt
+kbe
kt
k sinrt r cosrt
k
2
+r
2
+C
Dividethroughbye
kt
toget
y = a +bk
_
k sinrt r cosrt
k
2
+r
2
_
+Ce
kt
Assumethat theoutsidetemperaturevariesas
T (t ) = 15+5sin(t /12)
wheret = 0is12noon. A houseisheatedto25
C at t = 0andafter that, itstemperaturey(t ) variesaccordingto

NewtonsLawof Cooling(Figure6):
dy
dt
= 0.1
_
y(t ) T (t )
_
UseExercise41tosolvefor y(t ).
43. Let (x) beanintegratingfactor for y
+ A(x)y = B(x). Thedifferential equationy
+ A(x)y = 0is calledthe

associatedhomogeneousequation.
(a) Showthat 1/(x) isasolutionof theassociatedhomogeneousequation.
(b) Showthat if y = f (x) isaparticular solutionof y
+A(x)y = B(x), thenf (x) +C/(x) isalsoasolutionfor any

constant C.
solution
(a) Remember that
(x) = A(x)(x). Now, let y(x) = ((x))

1
. Then
y
+A(x)y =
1
((x))
2
(x) +
A(x)
(x)
=
1
((x))
2
A(x)(x) +
A(x)
(x)
= 0.
(b) Suppose f (x) satises f
(x) + A(x)f (x) = B(x). Now, let y(x) = f (x) + C/(x), where C is an arbitrary
constant. Then
y
+A(x)y = f
(x)
C
((x))
2
(x) +A(x)f (x) +

CA(x)
(x)
=
_
f
(x) +A(x)f (x)

_
+
C
(x)
_
A(x)

(x)
(x)
_
= B(x) +0= B(x).
UsetheFundamental Theoremof CalculusandtheProduct Ruletoverifydirectlythat for anyx
0
, thefunction
f (x) = (x)
1
_
x
x
0
(t )B(t ) dt
isasolutionof theinitial valueproblem
y
+A(x)y = B(x), y(x

0
) = 0
where(x) isanintegratingfactor [asolutiontoEq. (3)].
45. Transient Currents Supposethecircuit described by Eq. (10) is driven by asinusoidal voltagesourceV(t ) =
V sint (whereV and areconstant).
(a) Showthat
I (t ) =
V
R
2
+L
2
2
(Rsint Lcost ) +Ce
(R/L) t
(b) Let Z =
_
R
2
+L
2
2
. Choose so that Zcos = R and Zsin = L. Usetheaddition formulafor thesine
functiontoshowthat
I (t ) =
V
Z
sin(t ) +Ce
(R/L) t
This shows that the current in the circuit varies sinusoidally apart froma DC term(called the transient current in
electronics) that decreasesexponentially.
solution
(a) WithV(t ) = V sint , theequation
dI
dt
+
R
L
I =
1
L
V(t )
becomes
dI
dt
+
R
L
I =
V
L
sint.
Thisisalinear equationinstandardformwithA(t ) =
R
L
andB(t ) =
V
L
sint . ByTheorem1, theintegratingfactor is
(t ) =
_
e
_
A(t ) dt
= e
(R/L) t
.
Whenmultipliedbytheintegratingfactor, theequationbecomes
(e
(R/L) t
I)
=
V
L
e
(R/L) t
sint.
Integrationonbothsides(integrationbypartsisneededfor theintegral ontheright-handside) nowyields
(e
(R/L) t
I) =
V
R
2
+L
2
2
e
(R/L) t
(Rsint Lcost ) +C;
hence,
I (t ) =
V
R
2
+L
2
2
(Rsint Lcost ) +Ce
(R/L) t
.
(b) Let Z =
_
R
2
+L
2
2
, andchoose sothat Zcos = R andZsin = L. Then
V
R
2
+L
2
2
(Rsint Lcost ) =
V
Z
2
(Zcos sint Zsin cost )
=
V
Z
(cos sint sin cost ) =
V
Z
sin(t ).
Thus,
I (t ) =
V
Z
sin(t ) +Ce
(R/L) t
.
1. Whichof thefollowingdifferential equationsarelinear? Determinetheorder of eachequation.
(a) y
= y
5
3x
4
y (b) y
= x
5
3x
4
y
(c) y = y
3x
y (d) sinx y
= y 1
solution
(a) y
5
isanonlinear terminvolvingthedependent variable, sothisisnot alinear equation; thehighest order derivative
that appearsintheequationisarst derivative, sothisisarst-order equation.
(b) This is linear equation; the highest order derivative that appears in the equation is a rst derivative, so this is a
rst-order equation.
(c)

y isanonlinear terminvolvingthedependent variable, sothisisnot alinear equation; thehighest order derivative
that appearsintheequationisathirdderivative, sothisisathird-order equation.
(d) This is linear equation; thehighest order derivativethat appears intheequationis asecondderivative, so this is a
second-order equation.
Findavalueof c suchthat y = x 2+e
cx
isasolutionof 2y
+y = x.
In Exercises 36, solve using separation of variables.
3.
dy
dt
= t
2
y
3
y
3
dy = t
2
dt.
Uponintegratingbothsidesof thisequation, weobtain:
_
y
3
dy =
_
t
2
dt
y
4
4
=
t
3
3
+C.
Thus,
y =
_
4
3
t
3
+C
_
1/4
,
xyy
= 1x
2 5. x
dy
dx
y = 1
dy
1+y
=
dx
x
.
uponintegratingbothsidesof thisequation, weobtain
_
dy
1+y
=
_
dx
x
ln|1+y| = ln|x| +C.
Thus,
y = 1+Ax,
whereA = e
C
y
=
xy
2
x
2
+1
In Exercises 710, solve the initial value problem using separation of variables.
7. y
= cos
2
x, y(0) =

4
solution First, wendthegeneral solutionof thedifferential equation. Becausethevariablesarealreadyseparated,
weintegratebothsidestoobtain
y =
_
cos
2
x dx =
_ _
1
2
+
1
2
cos2x
_
dx =
x
2
+
sin2x
4
+C.
Theinitial conditiony(0) =

4
allowsustodetermineC =

4
. Thus, thesolutionis:
y(x) =
x
2
+
sin2x
4
+

4
.
y
= cos
2
y, y(0) =

4
9. y
= xy
2
, y(1) = 2
dy
dx
= xy
2
as
dy
y
2
= x dx.
Uponintegratingbothsidesof thisequation, wend
_
dy
y
2
=
_
x dx
1
y
=
1
2
x
2
+C.
Thus,
y =
1
1
2
x
2
+C
.
2=
1
1
2
1
2
+C
=
2
1+2C
1+2C = 1
C = 1
Hence, thesolutiontotheinitial valueproblemis
y =
1
1
2
x
2
1
=
2
x
2
2
.
xyy
= 1, y(3) = 2
11. Figure 1 shows the slope eld for

y = siny +ty. Sketch the graphs of the solutions with the initial conditions
y(0) = 1, y(0) = 0, andy(0) = 1.
0 1 2 2 1
2
1
0
1
2
t
y
FIGURE 1
solution
3
3
2
1
0
1
2
3
2 1 0 1 2 3
y
t
Whichof theequations(i)(iii) correspondstotheslopeeldinFigure2?
(i)

y = 1y
2
(ii)

y = 1+y
2
(iii)

y = y
2
13. Lety(t ) bethesolutiontothedifferential equationwithslopeeldasshowninFigure2, satisfyingy(0) = 0. Sketch
thegraphof y(t ). Thenuseyour answer toExercise12tosolvefor y(t ).
solution Asexplainedinthepreviousexercise, theslopeeldinFigure2correspondstotheequation

y = 1+ y
2
.
Thegraphof thesolutionsatisfyingy(0) = 0is:
3
3
2
1
0
1
2
3
2 1 1 0 2 3
y
t
Tosolvetheinitial valueproblem
y = 1+ y
2
, y(0) = 0, werst ndthegeneral solutionof thedifferential equation.
Separatingvariablesyields:
dy
1+y
2
= dt.
tan
1
y = t +C or y = tan(t +C).
Theinitial conditiongivesC = 0, sothesolutionisy = tanx.
Let y(t ) bethesolutionof 4
y = y
2
+ t satisfyingy(2) = 1. Carryout EulersMethodwithtimesteph = 0.05
for n = 6steps.
15. Lety(t ) bethesolutionof (x
3
+1)
y = y satisfyingy(0) = 1. Computeapproximationstoy(0.1), y(0.2), andy(0.3)

usingEulersMethodwithtimesteph = 0.1.
solution Rewritingtheequationas
y =
y
x
3
+1
wehaveF(x, y) =
y
x
3
+1
. UsingEulersMethodwithx
0
= 0, y
0
= 1
andh = 0.1, wecalculate
y(0.1) y
1
= y
0
+hF(x
0
, y
0
) = 1+0.1
1
0
3
+1
= 1.1
y(0.2) y
2
= y
1
+hF(x
1
, y
1
) = 1.209890
y(0.3) y
3
= y
2
+hF(x
2
, y
2
) = 1.329919
In Exercises 1619, solve using the method of integrating factors.
dy
dt
= y +t
2
, y(0) = 4
17.
dy
dx
=
y
x
+x, y(1) = 3
solution First, wendthegeneral solutionof thedifferential equation. Rewritetheequationas
y
1
x
y = x,
1
x
andB(x) = x. Theintegratingfactor is
(x) = e
_

1
x
dx
= e
lnx
=
1
x
.
_
1
x
y
_
= 1.
1
x
y = x +C;
hence,
y(x) = x
2
+Cx.
3= 1+C so C = 2.
Thesolutiontotheinitial valueproblemisthen
y = x
2
+2x.
dy
dt
= y 3t , y(1) = 2
19. y
+2y = 1+e
x
, y(0) = 4
solution TheequationisalreadyinstandardformwithA(x) = 2andB(x) = 1+e
x
. Theintegratingfactor is
(x) = e
_
2dx
= e
2x
.
(e
2x
y)
= e
2x
+e
x
.
e
2x
y =
1
2
e
2x
+e
x
+C;
hence,
y(x) =
1
2
+e
x
+Ce
2x
.
4=
1
2
+1+C so C =
11
2
.
y(x) =
1
2
+e
x
11
2
e
2x
.
In Exercises 2027, solve using the appropriate method.
x
2
y
= x
2
+1, y(1) = 10
21. y
+(tanx)y = cos
2
x, y() = 2
solution First, wendthegeneral solutionof thedifferential equation. As this is arst order linear equationwith
A(x) = tanx andB(x) = cos
2
x, wecomputetheintegratingfactor
(x) = e
_
A(x) dx
= e
_
tanx dx
= e
lncosx
=
1
cosx
.
_
1
cosx
y
_
= cosx.
1
cosx
y = sinx +C;
hence,
y(x) = sinx cosx +C cosx =
1
2
sin2x +C cosx.
Theinitial conditiony() = 2allowsustodeterminethevalueof C:
2= 0+C(1) so C = 2.
y =
1
2
sin2x 2cosx.
xy
= 2y +x 1, y
_
3
2
_
= 9
23. (y 1)y
= t , y(1) = 3
solution First, wendthegeneral solutionof thedifferential equation. Thisisaseparableequationthatwerewriteas
(y 1) dy = t dt.
_
(y 1) dy =
_
t dt
y
2
2
y =
1
2
t
2
+C
y
2
2y +1= t
2
+C
(y 1)
2
= t
2
+C
y(t ) =
_
t
2
+C +1
Tosatisfytheinitial conditiony(1) = 3wemust choosethenegativesquareroot; moreover,
3=
1+C +1 so C = 15.
y(t ) =
_
t
2
+15+1
_
y +1
_
y
= yt e
t
2
, y(0) = 1
25.
dw
dx
= k
1+w
2
x
, w(1) = 1
solution First, wendthegeneral solutionof thedifferential equation. Thisisaseparableequationthatwerewriteas
dw
1+w
2
=
k
x
dx.
_
dw
1+w
2
=
_
k
x
dx
tan
1
w = k lnx +C
w(x) = tan(k lnx +C).
Becausetheinitial conditionis speciedat x = 1, weareinterestedinthesolutionfor x > 0; wecanthereforeomit
theabsolutevaluewithinthenatural logarithmfunction. Theinitial conditionw(1) = 1allowsustodeterminethevalue
of C:
1= tan(k ln1+C) so C = tan
1
1=

4
.
w = tan
_
k lnx +

4
_
.
y
+
3y 1
t
= t +2
27. y
+
y
x
= sinx
solution Thisisarst order linear equationwithA(x) =
1
x
andB(x) = sinx. Theintegratingfactor is
(x) = e
_
A(x) dx
= e
lnx
= x.
(xy)
= x sinx.
Integrationonbothsides(integrationbypartsisneededfor theintegral ontheright-handside) nowyields
xy = x cosx +sinx +C;
hence,
y(x) = cosx +
sinx
x
+
C
x
.
Findthesolutionstoy
= 4(y 12) satisfyingy(0) = 20andy(0) = 0, andsketchtheir graphs.

29. Findthesolutionstoy
= 2y +8satisfyingy(0) = 3andy(0) = 4, andsketchtheir graphs.

solution First, rewritethedifferential equationasy
= 2(y 4); fromhereweseethat thegeneral solutionis

y(t ) = 4+Ce
2t
,
for someconstant C. If y(0) = 3, then
3= 4+Ce
0
and C = 1.
Thus, y(t ) = 4e
2t
. If y(0) = 4, then
4= 4+Ce
0
and C = 0;
hence, y(t ) = 4. Thegraphsof thetwosolutionsareshownbelow.
0.5
2
2
4
y
y = 4
y = 4 e
2t
x
0.5 1.0 1.5
Showthat y = sin
1
x satisesthedifferential equationy
= secy withinitial conditiony(0) = 0.

31. Findthesolutiony = f (x) of y
=
_
y
2
1satisfyingy(0) = 2.
solution Usingseparationof variables, wehave
dy
_
y
2
1
= 1dx
Integratinggives
ln
y +
_
y
2
1
= x +C
1
sothat
y +
_
y
2
1= Ce
x
Theinitial conditiony(0) = 2gives2+
_
2
2
1= C, sothat C = 2+
3, andtheanswer is
y +
_
y
2
1= (2+
3)e
x
or, solvingfor y, y =
(2+
3)e
2x
+2
3
2e
x
Statewhether thedifferential equation can besolved using separation of variables, themethod of integrating
factors, both, or neither.
(a) y
= y +x
2
(b) xy
= y +1
(c) y
= y
2
+x
2
(d) xy
= y
2
33. LetAandB beconstants. Provethatif A > 0, thenall solutionsof
dy
dt
+Ay = B approachthesamelimitast .
solution Thisisalinear rst-order equationinstandardformwithintegratingfactor
(t ) = e
_
Adt
= e
At
.
(e
At
y)
= Be
At
.
e
At
y =
B
A
e
At
+C;
hence,
y(t ) =
B
A
+Ce
At
.
BecauseA > 0,
lim
t
y(t ) = lim
t
_
B
A
+Ce
At
_
=
B
A
.
Weconcludethat if A > 0, all solutionsapproachthelimit
B
A
ast .
At timet = 0, atankof height 5mintheshapeof aninvertedpyramidwhosecrosssectionat thetopisasquare
of side2mislledwithwater. Water owsthroughaholeat thebottomof area0.002m
2
. UseTorricellisLawto
determinethetimerequiredfor thetanktoempty.
35. ThetroughinFigure3(dimensions incentimeters) is lledwithwater. At timet = 0(inseconds), water begins
leakingthroughaholeat thebottomof area4cm
2
. Let y(t ) bethewater height at timet . Findadifferential equationfor
y(t ) andsolveit todeterminewhenthewater level decreasesto60cm.
180
120
360
260
FIGURE 3
solution y(t ) obeysthedifferential equation:
dy
dt
=
Bv(y)
A(y)
,
wherev(y) denotesthevelocityof thewater owingthroughtheholewhenthetroughislledtoheight y, B denotesthe
areaof theholeandA(y) denotestheareaof thehorizontal crosssectionof thetroughat height y. Sincemeasurements
areall incentimeters, wewill workincentimeters. Wehave
g = 9.8m/s
2
= 980cm/s
2
ByTorricellisLaw, v(y) =
2 980
y = 14
10
y m/s. Theareaof theholeisB = 4cm

2
. Thehorizontal cross
sectionof thetroughat height y is arectangleof length360andwidthw(y). As w(y) varies linearly from180when
y = 0to260wheny = 120, it followsthat
w(y) = 180+
80y
120
= 180+
2
3
y
sothat theareaof thehorizontal cross-sectionat height y is
A(y) = 360w(y) = 64800+240y = 240(y +270)
Thedifferential equationfor y(t ) thenbecomes
dy
dt
=
Bv(y)
A(y)
=
4 14
10
y
240(y +270)
=
7
10
30

y
y +270
Thisequationisseparable, so
y +270
y
dy =
7
10
30
dt
(y
1/2
+270y
1/2
) dy =
7
10
30
dt
_
(y
1/2
+270y
1/2
) dy =
7
10
30
_
1dt
2
3
y
3/2
+540y
1/2
=
7
10
30
t +C
y
3/2
+810y
1/2
=
7
10
20
t +C
120
3/2
+810 120
1/2
= 0+C so C = 930
120= 1860
30
Thustheheight of thewater isgivenimplicitlybytheequation
y
3/2
+810y
1/2
=
7
10
20
t +1860
30
Wewant tondt suchthat y(t ) = 60:
60
3/2
+810 60
1/2
=
7
10
20
t +1860
30
1740
15=
7
10
20
t +1860
30
t =
120
7

10(31
3029
15) 3115.88s
Theheight of thewater inthetankis60cmafter approximately3116seconds, or 51minutes56seconds.
Findthesolutionof thelogisticequation
y = 0.4y(4y) satisfyingy(0) = 8.
37. Let y(t ) bethesolutionof

y = 0.3y(2y) withy(0) = 1. Determine lim
t
y(t ) without solvingfor y explicitly.
solution Wewritethegivenequationintheform
y = 0.6y
_
1
y
2
_
.
Thisisalogistic equationwithA = 2andk = 0.6. Becausetheinitial conditiony(0) = y
0
= 1satises0< y
0
< A,
thesolutionisincreasingandapproachesA ast . That is, lim
t
y(t ) = 2.
Supposethat y
= ky(1y/8) hasasolutionsatisfyingy(0) = 12andy(10) = 24. Findk.

39. A lakehasacarryingcapacityof 1000sh. Assumethat theshpopulationgrowslogisticallywithgrowthconstant
k = 0.2day
1
. Howmanydayswill it takefor thepopulationtoreach900shif theinitial populationis20sh?
solution Lety(t ) representtheshpopulation. Becausethepopulationgrowslogisticallywithk = 0.2andA = 1000,
y(t ) =
1000
1e
0.2t
/C
.
20=
1000
1
1
C
; 1
1
C
= 50; so C =
1
49
.
Hence,
y(t ) =
1000
1+49e
0.2t
.
Thepopulationwill reach900shwhen
1000
1+49e
0.2t
= 900.
Solvingfor t , wend
t = 5ln441 30.44days.
A rabbit population on an island increases exponentially with growth rate k = 0.12 months
1
. When the
populationreaches300rabbits(say, at timet = 0), wolvesbegineatingtherabbitsat arateof r rabbitsper month.
(a) Findadifferential equationsatisedbytherabbit populationP(t ).
(b) Howlargecanr bewithout therabbit populationbecomingextinct?
41. Showthat y = sin(tan
1
x + C) isthegeneral solutionof y
=
_
1y
2
/
_
1+x
2
_
. Thenusetheadditionformula
for thesinefunctiontoshowthat thegeneral solutionmaybewritten
y =
(cosC)x +sinC
_
1+x
2
solution Rewrite
dy
dx
=
_
1y
2
1+x
2
as
dy
_
1y
2
=
dx
1+x
2
.
_
dy
_
1y
2
=
_
dx
1+x
2
sin
1
y = tan
1
x +C
Thus,
y(x) = sin
_
tan
1
x +C
_
.
Toexpressthesolutionintherequiredform, weusetheadditionformula
sin( +) = sin cos +sin cos
Thisyields
y(x) = sin
_
tan
1
x
_
cosC +sinC cos
_
tan
1
x
_
.
Usingthegurebelow, weseethat
sin
_
tan
1
x
_
=
x
_
1+x
2
; and
cos
_
tan
1
x
_
=
1
_
1+x
2
.
Finally,
y =
x cosC
_
1+x
2
+
sinC
_
1+x
2
=
(cosC)x +sinC
_
1+x
2
.
x
1
1+ x
2
tan
1
x
Atankislledwith300litersof contaminatedwater containing3kgof toxin. Purewater ispumpedinatarateof
40L/min, mixesinstantaneously, andisthenpumpedout at thesamerate. Let y(t ) bethequantity of toxinpresent
inthetankat timet .
(a) Findadifferential equationsatisedbyy(t ).
(b) Solvefor y(t ).
(c) Findthetimeat whichthereis0.01kgof toxinpresent.
43. Att = 0, atankof volume300L islledwith100L of water containingsaltataconcentrationof 8g/L. Freshwater
ows inat arateof 40L/min, mixes instantaneously, andexits at thesamerate. Let c
1
(t ) bethesalt concentrationat
timet .
(a) Findadifferential equationsatisedby c
1
(t ) Hint: Findthedifferential equationfor thequantity of salt y(t ), and
observethat c
1
(t ) = y(t )/100.
(b) Findthesalt concentrationc
1
(t ) inthetankasafunctionof time.
solution
(a) Let y(t ) betheamount of salt inthetank at timet ; thenc
1
(t ) = y(t )/100. Therateof changeof theamount of salt
inthetankis
dy
dt
_
40
L
min
_
_
0
g
L
_
_
40
L
min
_
_
y
100
g
L
_
=
2
5
y
Now, c
1
(t ) = y
(t )/100andc(t ) = y(t )/100, sothat c

1
satisesthesamedifferential equation:
dc
1
dt
=
2
5
c
1
(b) Thisisalinear differential equation. Puttingit instandardformgives
dc
1
dt
+
2
5
c
1
= 0
Theintegratingfactor ise
2t /5
; multiplyingbothsidesbytheintegratingfactor gives
(e
2t /5
c
1
)
= 0
Integrateandmultiplythroughbye
2t /5
toget
c
1
(t ) = Ce
2t /5
Theinitial conditiontellsusthat y(0) = Ce
20/5
= C = 8, sothat nally,
c
1
(t ) = 8e
2t /5
Theoutowof thetank inExercise43isdirectedintoasecondtank containingV litersof freshwater whereit
mixesinstantaneously andexitsat thesamerateof 40L/min. Determinethesalt concentrationc
2
(t ) inthesecond
tankasafunctionof timeinthefollowingtwocases:
(a) V = 200 (b) V = 300
I h d t i th i t ti
11 INFINITE SERIES
11.1 Sequences
1. What isa
4
for thesequencea
n
= n
2
n?
solution Substitutingn = 4intheexpressionfor a
n
gives
a
4
= 4
2
4= 12.
2. Whichof thefollowingsequencesconvergetozero?
(a)
n
2
n
2
+1
(b) 2
n
(c)
_
1
2
_
n
solution
(a) Thissequencedoesnot convergetozero:
lim
n
n
2
n
2
+1
= lim
x
x
2
x
2
+1
= lim
x
1
1+
1
x
2
=
1
1+0
= 1.
(b) Thissequencedoesnot convergetozero: thisisageometricsequencewithr = 2> 1; hence, thesequencediverges
to.
(c) Recall that if |a
n
| convergesto0, thena
n
must alsoconvergetozero. Here,
1
2
_
n
=
_
1
2
_
n
,
whichisageometricsequencewith0< r < 1; hence, (
1
2
)
n
convergestozero. Itthereforefollowsthat(
1
2
)
n
converges
tozero.
3. Let a
n
bethenthdecimal approximationto
2. That is, a
1
= 1, a
2
= 1.4, a
3
= 1.41, etc. What is lim
n
a
n
?
solution lim
n
a
n
=
2.
4. Whichof thefollowingsequencesisdenedrecursively?
(a) a
n
=
4+n (b) b
n
=
_
4+b
n1
solution
(a) a
n
canbecomputeddirectly, sinceitdependsonn onlyandnotonprecedingterms. Thereforea
n
isdenedexplicitly
andnot recursively.
(b) b
n
iscomputedintermsof theprecedingtermb
n1
, hencethesequence{b
n
} isdenedrecursively.
5. Theorem5saysthat every convergent sequenceisbounded. Determineif thefollowingstatementsaretrueor false
andif false, giveacounterexample.
(a) If {a
n
} isbounded, thenit converges.
(b) If {a
n
} isnot bounded, thenit diverges.
(c) If {a
n
} diverges, thenit isnot bounded.
solution
(a) This statement is false. Thesequencea
n
= cosn is bounded since1 cosn 1for all n, but it does not
converge: sincea
n
= cosn = (1)
n
, thetermsassumethetwovalues1and1alternately, hencetheydonotapproach
onevalue.
(b) ByTheorem5, aconvergingsequencemust bebounded. Therefore, if asequenceisnot bounded, it certainly does
not converge.
(c) Thestatement isfalse. Thesequencea
n
= (1)
n
isbounded, but it doesnot approachonelimit.
646
S E C T I ON 11.1 Sequences 647
Exercises
1. Matcheachsequencewithitsgeneral term:
a
1
, a
2
, a
3
, a
4
, . . . General term
(a)
1
2
,
2
3
,
3
4
,
4
5
, . . . (i) cosn
(b) 1, 1, 1, 1, . . . (ii)
n!
2
n
(c) 1, 1, 1, 1, . . . (iii) (1)
n+1
(d)
1
2
,
2
4
,
6
8
,
24
16
. . . (iv)
n
n +1
solution
(a) Thenumerator of eachtermisthesameastheindexof theterm, andthedenominator isonemorethanthenumerator;
hencea
n
=
n
n+1
, n = 1, 2, 3, . . . .
(b) Thetermsof thissequencearealternatingbetween1and1sothat thepositivetermsareintheevenplaces. Since
cosn = 1for evenn andcosn = 1for oddn, wehavea
n
= cosn, n = 1, 2, . . . .
(c) Thetermsa
n
are1for oddn and1for evenn. Hence, a
n
= (1)
n+1
, n = 1, 2, . . .
(d) Thenumerator of eachtermisn!, andthedenominator is2
n
; hence, a
n
=
n!
2
n
, n = 1, 2, 3, . . . .
Let a
n
=
1
2n 1
for n = 1, 2, 3, . . . . Writeout therst threetermsof thefollowingsequences.
(a) b
n
= a
n+1
(b) c
n
= a
n+3
(c) d
n
= a
2
n
(d) e
n
= 2a
n
a
n+1
In Exercises 312, calculate the rst four terms of the sequence, starting with n = 1.
3. c
n
=
3
n
n!
solution Settingn = 1, 2, 3, 4intheformulafor c
n
gives
c
1
=
3
1
1!
=
3
1
= 3, c
2
=
3
2
2!
=
9
2
,
c
3
=
3
3
3!
=
27
6
=
9
2
, c
4
=
3
4
4!
=
81
24
=
27
8
.
b
n
=
(2n 1)!
n!
5. a
1
= 2, a
n+1
= 2a
2
n
3
solution For n = 1, 2, 3wehave:
a
2
= a
1+1
= 2a
2
1
3= 2 43= 5;
a
3
= a
2+1
= 2a
2
2
3= 2 253= 47;
a
4
= a
3+1
= 2a
2
3
3= 2 22093= 4415.
Therst four termsof {a
n
} are2, 5, 47, 4415.
b
1
= 1, b
n
= b
n1
+
1
b
n1
7. b
n
= 5+cosn
solution For n = 1, 2, 3, 4wehave
b
1
= 5+cos = 4;
b
2
= 5+cos2 = 6;
b
3
= 5+cos3 = 4;
b
4
= 5+cos4 = 6.
Therst four termsof {b
n
} are4, 6, 4, 6.
c
n
= (1)
2n+1 9. c
n
= 1+
1
2
+
1
3
+ +
1
n
solution
c
1
= 1;
c
2
= 1+
1
2
=
3
2
;
c
3
= 1+
1
2
+
1
3
=
3
2
+
1
3
=
11
6
;
c
4
= 1+
1
2
+
1
3
+
1
4
=
11
6
+
1
4
=
25
12
.
a
n
= n +(n +1) +(n +2) + +(2n)
648 C HA P T E R 11 INFINITE SERIES
11. b
1
= 2, b
2
= 3, b
n
= 2b
n1
+b
n2
solution Weneedtondb
3
andb
4
. Settingn = 3andn = 4andusingthegivenvaluesfor b
1
andb
2
weobtain:
b
3
= 2b
31
+b
32
= 2b
2
+b
1
= 2 3+2= 8;
b
4
= 2b
41
+b
42
= 2b
3
+b
2
= 2 8+3= 19.
Therst four termsof thesequence{b
n
} are2, 3, 8, 19.
c
n
= n-placedecimal approximationtoe
13. Findaformulafor thenthtermof eachsequence.
(a)
1
1
,
1
8
,
1
27
, . . . (b)
2
6
,
3
7
,
4
8
, . . .
solution
(a) Thedenominators arethethird powers of thepositiveintegers starting with n = 1. Also, thesign of theterms is
alternatingwiththesignof therst termbeingpositive. Thus,
a
1
=
1
1
3
=
(1)
1+1
1
3
; a
2
=
1
2
3
=
(1)
2+1
2
3
; a
3
=
1
3
3
=
(1)
3+1
3
3
.
Thisruleleadstothefollowingformulafor thenthterm:
a
n
=
(1)
n+1
n
3
.
(b) Assumingastartingindex of n = 1, weseethat eachnumerator isonemorethantheindex andthedenominator is
four morethanthenumerator. Thus, thegeneral terma
n
is
a
n
=
n +1
n +5
.
Supposethat lim
n
a
n
= 4and lim
n
b
n
= 7. Determine:
(a) lim
n
(a
n
+b
n
) (b) lim
n
a
3
n
(c) lim
n
cos(b
n
) (d) lim
n
(a
2
n
2a
n
b
n
)
In Exercises 1526, use Theorem 1 to determine the limit of the sequence or state that the sequence diverges.
15. a
n
= 12
solution Wehavea
n
= f (n) wheref (x) = 12; thus,
lim
n
a
n
= lim
x
f (x) = lim
x
12= 12.
a
n
= 20
4
n
2
17. b
n
=
5n 1
12n +9
solution Wehaveb
n
= f (n) wheref (x) =
5x 1
12x +9
; thus,
lim
n
5n 1
12n +9
= lim
x
5x 1
12x +9
=
5
12
.
a
n
=
4+n 3n
2
4n
2
+1
19. c
n
= 2
n
solution Wehavec
n
= f (n) wheref (x) = 2
x
; thus,
lim
n
_
2
n
_
= lim
x
2
x
= lim
x
1
2
x
= 0.
z
n
=
_
1
3
_
n
21. c
n
= 9
n
solution Wehavec
n
= f (n) wheref (x) = 9
x
; thus,
lim
n
9
n
= lim
x
9
x
=
Thus, thesequence9
n
diverges.
z
n
= 10
1/n
23. a
n
=
n
_
n
2
+1
solution Wehavea
n
= f (n) wheref (x) =
x
_
x
2
+1
; thus,
lim
n
n
_
n
2
+1
= lim
x
x
_
x
2
+1
= lim
x
x
x
x
2
+1
x
= lim
x
1
_
x
2
+1
x
2
= lim
x
1
_
1+
1
x
2
=
1
1+0
= 1.
a
n
=
n
_
n
3
+1
25. a
n
= ln
_
12n +2
9+4n
_
solution Wehavea
n
= f (n) wheref (x) = ln
_
12x +2
9+4x
_
; thus,
lim
n
ln
_
12n +2
9+4n
_
= lim
x
ln
_
12x +2
9+4x
_
= ln lim
x
_
12x +2
9+4x
_
= ln3
r
n
= lnn ln(n
2
+1)
In Exercises 2730, use Theorem 4 to determine the limit of the sequence.
27. a
n
=
_
4+
1
n
solution Wehave
lim
n
4+
1
n
= lim
x
4+
1
x
= 4
Since
x isacontinuousfunctionfor x > 0, Theorem4tellsusthat

lim
n
_
4+
1
n
=
_
lim
n
4+
1
n
=
4= 2
a
n
= e
4n/(3n+9)
29. a
n
= cos
1
_
n
3
2n
3
+1
_
solution Wehave
lim
n
n
3
2n
3
+1
=
1
2
Sincecos
1
(x) iscontinuousfor all x, Theorem4tellsusthat
lim
n
cos
1
_
n
3
2n
3
+1
_
= cos
1
_
lim
n
n
3
2n
3
+1
_
= cos
1
(1/2) =

3
a
n
= tan
1
(e
n
)
31. Let a
n
=
n
n +1
. Findanumber M suchthat:
(a) |a
n
1| 0.001for n M.
(b) |a
n
1| 0.00001for n M.
Thenusethelimit denitiontoprovethat lim
n
a
n
= 1.
solution
(a) Wehave
|a
n
1| =
n
n +1
1
n (n +1)
n +1
1
n +1
=
1
n +1
.
Therefore|a
n
1| 0.001provided
1
n+1
0.001, that is, n 999. It followsthat wecantakeM = 999.
(b) Bypart(a), |a
n
1| 0.00001provided
1
n+1
0.00001, thatis, n 99999. ItfollowsthatwecantakeM = 99999.
Wenowproveformallythat lim
n
a
n
= 1. Usingpart (a), weknowthat
|a
n
1| =
1
n +1
< ,
providedn >
1
1. Thus, Let > 0andtakeM =

1
1. Then, for n > M, wehave

|a
n
1| =
1
n +1
<
1
M +1
= .
Let b
n
=
_
1
3
_
n
.
(a) Findavalueof M suchthat |b
n
| 10
5
for n M.
(b) Usethelimit denitiontoprovethat lim
n
b
n
= 0.
33. Usethelimit denitiontoprovethat lim
n
n
2
= 0.
solution Weseethat
|n
2
0| =
1
n
2
=
1
n
2
<
provided
n >
1
.
Thus, let > 0andtakeM =
1
. Then, for n > M, wehave

|n
2
0| =
1
n
2
=
1
n
2
<
1
M
2
= .
Usethelimit denitiontoprovethat lim
n
n
n +n
1
= 1.
In Exercises 3562, use the appropriate limit laws and theorems to determine the limit of the sequence or show that it
diverges.
35. a
n
= 10+
_
1
9
_
n
solution BytheLimit Lawsfor Sequenceswehave:
lim
n
_
10+
_
1
9
_
n
_
= lim
n
10+ lim
n
_
1
9
_
n
= 10+ lim
n
_
1
9
_
n
.
Now,
_
1
9
_
n
1
9
_
n
_
1
9
_
n
.
Because
lim
n
_
1
9
_
n
= 0,
bytheLimit Lawsfor Sequences,
lim
n
_
1
9
_
n
= lim
n
_
1
9
_
n
= 0.
Thus, wehave
lim
n
_
1
9
_
n
= 0,
and
lim
n
_
10+
_
1
9
_
n
_
= 10+0= 10.
d
n
=
n +3
n
37. c
n
= 1.01
n
solution Sincec
n
= f (n) wheref (x) = 1.01
x
, wehave
lim
n
1.01
n
= lim
x
1.01
x
=
sothat thesequencediverges.
b
n
= e
1n
2
39. a
n
= 2
1/n
solution Because2
x
isacontinuousfunction,
lim
n
2
1/n
= lim
x
2
1/x
= 2
lim
x
(1/x)
= 2
0
= 1.
b
n
= n
1/n
41. c
n
=
9
n
n!
solution For n 9, write
c
n
=
9
n
n!
=
9
1
9
2

9
9
. ,, .
call thisC
9
10
9
11

9
n 1
9
n
. ,, .
Eachfactor islessthan1
Thenclearly
0
9
n
n!
C
9
n
sinceeachfactor after therst nineis< 1. Thesqueezetheoremtellsusthat
lim
n
0 lim
n
9
n
n!
lim
n
C
9
n
= C lim
n
9
n
= C 0= 0
sothat lim
n
c
n
= 0aswell.
a
n
=
8
2n
n!
43. a
n
=
3n
2
+n +2
2n
2
3
solution
lim
n
3n
2
+n +2
2n
2
3
= lim
x
3x
2
+x +2
2x
2
3
=
3
2
.
a
n
=
n +4
45. a
n
=
cosn
n
solution Since1 cosn 1thefollowingholds:
1
n
cosn
n
1
n
.
WenowapplytheSqueezeTheoremfor Sequencesandthelimits
lim
n
1
n
= lim
n
1
n
= 0
toconcludethat lim
n
cosn
n
= 0.
c
n
=
(1)
n
n
47. d
n
= ln5
n
lnn!
solution Notethat
d
n
= ln
5
n
n!
sothat
e
d
n
=
5
n
n!
so lim
n
e
d
n
= lim
n
5
n
n!
= 0
bythemethodof Exercise41. If d
n
converged, wecould, sincef (x) = e
x
iscontinuous, thenwrite
lim
n
e
d
n
= e
lim
n
d
n
= 0
whichisimpossible. Thus{d
n
} diverges.
d
n
= ln(n
2
+4) ln(n
2
1) 49. a
n
=
_
2+
4
n
2
_
1/3
solution Let a
n
=
_
2+
4
n
2
_
1/3
. Takingthenatural logarithmof bothsidesof thisexpressionyields
lna
n
= ln
_
2+
4
n
2
_
1/3
=
1
3
ln
_
2+
4
n
2
_
.
Thus,
lim
n
lna
n
= lim
n
1
3
ln
_
2+
4
n
2
_
1/3
=
1
3
lim
x
ln
_
2+
4
x
2
_
=
1
3
ln
_
lim
x
_
2+
4
x
2
__
=
1
3
ln(2+0) =
1
3
ln2= ln2
1/3
.
Becausef (x) = e
x
isacontinuousfunction, it followsthat
lim
n
a
n
= lim
n
e
lna
n
= e
lim
n
(lna
n
)
= e
ln2
1/3
= 2
1/3
.
b
n
= tan
1
_
1
2
n
_
51. c
n
= ln
_
2n +1
3n +4
_
solution Becausef (x) = lnx isacontinuousfunction, it followsthat
lim
n
c
n
= lim
x
ln
_
2x +1
3x +4
_
= ln
_
lim
x
2x +1
3x +4
_
= ln
2
3
.
c
n
=
n
n +n
1/n
53. y
n
=
e
n
2
n
solution
e
n
2
n
=
_
e
2
_
n
and
e
2
> 1. BytheLimit of GeometricSequences,weconcludethat lim
n
_
e
2
_
n
= . Thus,
thegivensequencediverges.
a
n
=
n
2
n
55. y
n
=
e
n
+(3)
n
5
n
solution
lim
n
e
n
+(3)
n
5
n
= lim
n
_
e
5
_
n
+ lim
n
_
3
5
_
n
assumingbothlimitsontheright-handsideexist. But bytheLimit of GeometricSequences, since
1<
3
5
< 0<
e
5
< 1
bothlimitsontheright-handsideare0, sothat y
n
convergesto0.
b
n
=
(1)
n
n
3
+2
n
3n
3
+4
n
57. a
n
= n sin
n
solution BytheTheoremonSequencesDenedbyaFunction, wehave
lim
n
n sin
n
= lim
x
x sin
x
.
Now,
lim
x
x sin
x
= lim
x
sin
x
1
x
= lim
x
_
cos
x
_
_
x
2
_
1
x
2
= lim
x
_
cos
x
_
= lim
x
cos
x
= cos0= 1= .
Thus,
lim
n
n sin
n
= .
b
n
=
n!
n
59. b
n
=
34
n
2+7 4
n
solution Dividethenumerator anddenominator by4
n
toobtain
a
n
=
34
n
2+7 4
n
=
3
4
n

4
n
4
n
2
4
n
+
74
n
4
n
=
3
4
n
1
2
4
n
+7
.
Thus,
lim
n
a
n
= lim
x
3
4
x
1
2
4
x
+7
=
lim
x
_
3
4
x
1
_
lim
x
_
2
4
x
+7
_ =
3lim
x
1
4
x
lim
x
1
2lim
x
1
4
x
lim
x
7
=
3 01
2 0+7
=
1
7
.
a
n
=
34
n
2+7 3
n
61. a
n
=
_
1+
1
n
_
n
solution Takingthenatural logarithmof bothsidesof thisexpressionyields
lna
n
= ln
_
1+
1
n
_
n
= n ln
_
1+
1
n
_
=
ln
_
1+
1
n
_
1
n
.
Thus,
lim
n
(lna
n
) = lim
x
ln
_
1+
1
x
_
1
x
= lim
x
d
dx
_
ln
_
1+
1
x
__
d
dx
_
1
x
_ = lim
x
1
1+
1
x
1
x
2
_
1
x
2
= lim
x
1
1+
1
x
=
1
1+0
= 1.
Becausef (x) = e
x
isacontinuousfunction, it followsthat
lim
n
a
n
= lim
n
e
lna
n
= e
lim
n
(lna
n
)
= e
1
= e.
a
n
=
_
1+
1
n
2
_
n In Exercises 6366, nd the limit of the sequence using LHpitals Rule.
63. a
n
=
(lnn)
2
n
solution
lim
n
(lnn)
2
n
= lim
x
(lnx)
2
x
= lim
x
d
dx
(lnx)
2
d
dx
x
= lim
x
2lnx
x
1
= lim
x
2lnx
x
= lim
x
d
dx
2lnx
d
dx
x
= lim
x
2
x
1
= lim
x
2
x
= 0
b
n
=
n ln
_
1+
1
n
_
65. c
n
= n
_
_
n
2
+1n
_
solution
lim
n
n
__
n
2
+1n
_
= lim
x
x
__
x
2
+1x
_
= lim
x
x
__
x
2
+1x
_ __
x
2
+1+x
_
_
x
2
+1+x
= lim
x
x
_
x
2
+1+x
= lim
x
d
dx
x
d
dx
_
x
2
+1+x
= lim
x
1
1+
x
x
2
+1
= lim
x
1
1+
_
x
2
x
2
+1
= lim
x
1
1+
_
1
1+(1/x
2
)
=
1
2
d
n
= n
2
_
3
_
n
3
+1n
_ In Exercises 6770, use the Squeeze Theorem to evaluate lim
n
a
n
by verifying the given inequality.
67. a
n
=
1
_
n
4
+n
8
,
1
2n
4
a
n

1
2n
2
solution For all n > 1wehaven
4
< n
8
, sothequotient
1
n
4
+n
8
issmaller than
1
n
4
+n
4
andlarger than
1
n
8
+n
8
.
That is,
a
n
<
1
_
n
4
+n
4
=
1
n
4
2
=
1
2n
2
; and
a
n
>
1
_
n
8
+n
8
=
1
2n
8
=
1
2n
4
.
Now, since lim
n
1
2n
4
= lim
n
1
2n
2
= 0, theSqueezeTheoremfor Sequencesimpliesthat lim
n
a
n
= 0.
c
n
=
1
_
n
2
+1
+
1
_
n
2
+2
+ +
1
_
n
2
+n
,
n
_
n
2
+n
c
n

n
_
n
2
+1
69. a
n
= (2
n
+3
n
)
1/n
, 3 a
n
(2 3
n
)
1/n
= 2
1/n
3
solution Clearly2
n
+3
n
3
n
for all n 1. Therefore:
(2
n
+3
n
)
1/n
(3
n
)
1/n
= 3.
Also2
n
+3
n
3
n
+3
n
= 2 3
n
, so
(2
n
+3
n
)
1/n
(2 3
n
)
1/n
= 2
1/n
3.
Thus,
3 (2
n
+3
n
)
1/n
2
1/n
3.
Because
lim
n
2
1/n
3= 3 lim
n
2
1/n
= 3 1= 3
andlim
n
3= 3, theSqueezeTheoremfor Sequencesguarantees
lim
n
(2
n
+3
n
)
1/n
= 3.
a
n
= (n +10
n
)
1/n
, 10 a
n
(2 10
n
)
1/n 71. Whichof thefollowingstatementsisequivalent totheassertion lim
n
a
n
= L? Explain.
(a) For every > 0, theinterval (L , L +) containsat least oneelement of thesequence{a
n
}.
(b) For every > 0, theinterval (L , L +) containsall but at most nitelymanyelementsof thesequence{a
n
}.
solution Statement (b) is equivalent to Denition1of thelimit, sincetheassertion|a
n
L| < for all n > M
meansthat L < a
n
< L + for all n > M; that is, theinterval (L , L + ) containsall theelementsa
n
except
(maybe) thenitenumber of elementsa
1
, a
2
, . . . , a
M
.
Statement (a) isnot equivalent totheassertion lim
n
a
n
= L. Weshowthis, byconsideringthefollowingsequence:
a
n
=
1
n
for oddn
1+
1
n
for evenn
Clearly for every > 0, theinterval (, ) = (L , L + ) for L = 0containsat least oneelement of {a
n
}, but the
sequencediverges(rather thanconvergestoL = 0). Sincethetermsintheoddplacesconvergeto0andthetermsinthe
evenplacesconvergeto1. Hence, a
n
doesnot approachonelimit.
Showthat a
n
=
1
2n +1
isdecreasing.
73. Showthat a
n
=
3n
2
n
2
+2
isincreasing. Findanupper bound.
3x
2
x
2
+2
. Then
f
(x) =
6x(x
2
+2) 3x
2
2x
(x
2
+2)
2
=
12x
(x
2
+2)
2
.
f
(x) > 0for x > 0, hencef isincreasingonthisinterval. It followsthat a

n
= f (n) isalsoincreasing. Wenowshow
that M = 3isanupper boundfor a
n
, bywriting:
a
n
=
3n
2
n
2
+2
3n
2
+6
n
2
+2
=
3(n
2
+2)
n
2
+2
= 3.
That is, a
n
3for all n.
Showthat a
n
=
3
n +1n isdecreasing.
75. Giveanexampleof adivergent sequence{a
n
} suchthat lim
n
|a
n
| converges.
solution Leta
n
= (1)
n
. Thesequence{a
n
} divergesbecausethetermsalternatebetween+1and1; however, the
sequence{|a
n
|} convergesbecauseit isaconstant sequence, all of whosetermsareequal to1.
Giveanexampleof divergent sequences{a
n
} and{b
n
} suchthat {a
n
+b
n
} converges.
77. Usingthelimit denition, provethat if {a
n
} convergesand{b
n
} diverges, then{a
n
+b
n
} diverges.
solution Wewill provethis result by contradiction. Supposelim
n
a
n
= L
1
andthat {a
n
+ b
n
} converges to a
limitL
2
. Now, let > 0. Because{a
n
} convergestoL
1
and{a
n
+b
n
} convergestoL
2
, itfollowsthatthereexistnumbers
M
1
andM
2
suchthat:
|a
n
L
1
| <
2
for all n > M
1
,
| (a
n
+b
n
) L
2
| <
2
for all n > M
2
.
Thus, for n > M = max{M
1
, M
2
},
|a
n
L
1
| <
2
and | (a
n
+b
n
) L
2
| <
2
.
Bythetriangleinequality,
|b
n
(L
2
L
1
)| = |a
n
+b
n
a
n
(L
2
L
1
)| = |(a
n
+L
1
) +(a
n
+b
n
L
2
)|
|L
1
a
n
| +|a
n
+b
n
L
2
|.
Thus, for n > M,
|b
n
(L
2
L
1
) | <
2
+

2
= ;
that is, {b
n
} convergestoL
2
L
1
, incontradictiontothegivendata. Thus, {a
n
+b
n
} must diverge.
Usethelimit denitiontoprovethat if {a
n
} isaconvergent sequenceof integerswithlimit L, thenthereexistsa
number M suchthat a
n
= L for all n M.
79. Theorem1statesthat if lim
x
f (x) = L, thenthesequencea
n
= f (n) convergesand lim
n
a
n
= L. Showthat the
converse isfalse. Inother words, ndafunctionf (x) suchthat a
n
= f (n) convergesbut lim
x
f (x) doesnot exist.
solution Let f (x) = sinx anda
n
= sinn. Thena
n
= f (n). Sincesinx is oscillatingbetween1and1the
limit lim
x
f (x) doesnotexist. However, thesequence{a
n
} istheconstantsequenceinwhicha
n
= sinn = 0for all n,
henceit convergestozero.
Usethelimit denitiontoprovethat thelimit doesnot changeif anitenumber of termsareaddedor removed
fromaconvergent sequence.
81. Let b
n
= a
n+1
. Usethelimit denitionto provethat if {a
n
} converges, then{b
n
} also converges and lim
n
a
n
=
lim
n
b
n
.
solution Suppose{a
n
} convergestoL. Let b
n
= a
n+1
, andlet > 0. Because{a
n
} convergestoL, thereexistsan
M
suchthat |a
n
L| < for n > M
. Now, let M = M
1. Then, whenever n > M, n + 1> M + 1= M
. Thus,
for n > M,
|b
n
L| = |a
n+1
L| < .
Hence, {b
n
} convergestoL.
Let {a
n
} beasequencesuchthat lim
n
|a
n
| existsandisnonzero. Showthat lim
n
a
n
existsif andonly if there
existsaninteger M suchthat thesignof a
n
doesnot changefor n > M.
83. ProceedasinExample12toshowthat thesequence
3,
_
3
3,
_
3
_
3
3, . . . isincreasingandboundedaboveby
M = 3. Thenprovethat thelimit existsandnditsvalue.
solution Thissequenceisdenedrecursivelybytheformula:
a
n+1
=
_
3a
n
, a
1
=
3.
Consider thefollowinginequalities:
a
2
=
3a
1
=
_
3
3>
3= a
1
a
2
> a
1
;
a
3
=
3a
2
>
3a
1
= a
2
a
3
> a
2
;
a
4
=
3a
3
>
3a
2
= a
3
a
4
> a
3
.
Ingeneral, if weassumethat a
k
> a
k1
, then
a
k+1
=
_
3a
k
>
_
3a
k1
= a
k
.
Hence, bymathematical induction, a
n+1
> a
n
for all n; that is, thesequence{a
n
} isincreasing.
Becausea
n+1
=
3a
n
, it followsthat a
n
0for all n. Now, a
1
=
3< 3. If a
k
3, then
a
k+1
=
_
3a
k

3 3= 3.
Thus, bymathematical induction, a
n
3for all n.
Since{a
n
} isincreasingandbounded, itfollowsbytheTheoremonBoundedMonotonicSequencesthatthissequence
isconverging. Denotethelimit byL = lim
n
a
n
. UsingExercise81, it followsthat
L = lim
n
a
n+1
= lim
n
_
3a
n
=
_
3 lim
n
a
n
=
3L.
Thus, L
2
= 3L, soL = 0or L = 3. Becausethesequenceisincreasing, wehavea
n
a
1
=
3for all n. Hence, the

limit alsosatisesL
3. Weconcludethat theappropriatesolutionisL = 3; that is, lim

n
a
n
= 3.
Let {a
n
} bethesequencedenedrecursivelyby
a
0
= 0, a
n+1
=
_
2+a
n
Thus, a
1
=
2, a
2
=
_
2+
2, a
3
=
_
2+
_
2+
2, . . . .
(a) Showthat if a
n
< 2, thena
n+1
< 2. Concludebyinductionthat a
n
< 2for all n.
(b) Showthat if a
n
< 2, thena
n
a
n+1
. Concludebyinductionthat {a
n
} isincreasing.
(c) Use(a) and(b) toconcludethat L = lim
n
a
n
exists. ThencomputeL byshowingthat L =
2+L.
85. Showthat lim
n
n
n! = . Hint: Verify that n! (n/2)

n/2
by observingthat half of thefactors of n! aregreater
thanor equal ton/2.
solution Weshowthat n!
_
n
2
_
n/2
. For n 4even, wehave:
n! = 1
n
2
. ,, .
n
2
factors
_
n
2
+1
_
n
. ,, .
n
2
factors
_
n
2
+1
_
n
. ,, .
n
2
factors
.
Sinceeachoneof the
n
2
factorsisgreater than
n
2
, wehave:
n!
_
n
2
+1
_
n
. ,, .
n
2
factors
n
2

n
2
. ,, .
n
2
factors
=
_
n
2
_
n/2
.
For n 3odd, wehave:
n! = 1
n 1
2
. ,, .
n1
2
factors
n +1
2
n
. ,, .
n+1
2
factors
n +1
2
n
. ,, .
n+1
2
factors
.
Sinceeachoneof the
n+1
2
factorsisgreater than
n
2
, wehave:
n!
n +1
2
n
. ,, .
n+1
2
factors
n
2

n
2
. ,, .
n+1
2
factors
=
_
n
2
_
(n+1)/2
=
_
n
2
_
n/2
_
n
2
_
n
2
_
n/2
.
Ineither casewehaven!
_
n
2
_
n/2
. Thus,
n
n!
_
n
2
.
Since lim
n
_
n
2
= , it followsthat lim
n
n
n! = . Thus, thesequencea
n
=
n
n! diverges.
Let b
n
=
n
n!
n
.
(a) Showthat lnb
n
=
1
n
n
k=1
ln
k
n
.
(b) Showthat lnb
n
convergesto
_
1
0
lnx dx, andconcludethat b
n
e
1
.
87. Givenpositivenumbersa
1
< b
1
, denetwosequencesrecursivelyby
a
n+1
=
_
a
n
b
n
, b
n+1
=
a
n
+b
n
2
(a) Showthat a
n
b
n
for all n (Figure13).
(b) Showthat {a
n
} isincreasingand{b
n
} isdecreasing.
(c) Showthat b
n+1
a
n+1
b
n
a
n
2
.
(d) Provethat both {a
n
} and {b
n
} convergeand havethesamelimit. This limit, denoted AGM(a
1
, b
1
), is called the
arithmetic-geometricmeanof a
1
andb
1
.
(e) EstimateAGM(1,
2) tothreedecimal places.
x
a
n
a
n+1
b
n+1
b
n
Geometric
mean
AGM(a
1
, b
1
)
Arithmetic
mean
FIGURE 13
solution
(a) Examinethefollowing:
b
n+1
a
n+1
=
a
n
+b
n
2

_
a
n
b
n
=
a
n
+b
n
2
a
n
b
n
2
=
_
a
n
_
2
2
a
n
b
n
+
_
b
n
_
2
2
=
_
a
n
b
n
_
2
2
0.
Weconcludethat b
n+1
a
n+1
for all n > 1. Bythegiveninformationb
1
> a
1
; hence, b
n
a
n
for all n.
(b) Bypart (a), b
n
a
n
for all n, so
a
n+1
=
_
a
n
b
n

a
n
a
n
=
_
a
2
n
= a
n
for all n. Hence, thesequence{a
n
} isincreasing. Moreover, sincea
n
b
n
for all n,
b
n+1
=
a
n
+b
n
2

b
n
+b
n
2
=
2b
n
2
= b
n
for all n; that is, thesequence{b
n
} isdecreasing.
(c) Since{a
n
} isincreasing, a
n+1
a
n
. Thus,
b
n+1
a
n+1
b
n+1
a
n
=
a
n
+b
n
2
a
n
=
a
n
+b
n
2a
n
2
=
b
n
a
n
2
.
Now, bypart(a), a
n
b
n
forall n. Bypart(b), {b
n
} isdecreasing. Henceb
n
b
1
forall n. Combiningthetwoinequalities
weconcludethata
n
b
1
for all n. Thatis, thesequence{a
n
} isincreasingandbounded(0 a
n
b
1
). BytheTheorem
onBoundedMonotonicSequencesweconcludethat {a
n
} converges. Similarly, since{a
n
} isincreasing, a
n
a
1
for all
n. Wecombinethis inequality withb
n
a
n
to concludethat b
n
a
1
for all n. Thus, {b
n
} is decreasingandbounded
(a
1
b
n
b
1
); hencethissequenceconverges.
Toshowthat {a
n
} and{b
n
} convergetothesamelimit, notethat
b
n
a
n

b
n1
a
n1
2

b
n2
a
n2
2
2

b
1
a
1
2
n1
.
Thus,
lim
n
(b
n
a
n
) = (b
1
a
1
) lim
n
1
2
n1
= 0.
(d) Wehave
a
n+1
=
_
a
n
b
n
, a
1
= 1; b
n+1
=
a
n
+b
n
2
, b
1
=
2
Computingthevaluesof a
n
andb
n
until therst threedecimal digitsareequal insuccessiveterms, weobtain:
a
2
=
_
a
1
b
1
=
_
1
2= 1.1892
b
2
=
a
1
+b
1
2
=
1+
2
2
= 1.2071
a
3
=
_
a
2
b
2
=
1.1892 1.2071= 1.1981

b
3
=
a
2
+b
2
2
=
1.1892 1.2071
2
= 1.1981
a
4
=
_
a
3
b
3
= 1.1981
b
4
=
a
3
+b
3
2
= 1.1981
Thus,
AGM
_
1,
2
_
1.198.
Let c
n
=
1
n
+
1
n +1
+
1
n +2
+ +
1
2n
.
(a) Calculatec
1
, c
2
, c
3
, c
4
.
(b) Useacomparisonof rectangleswiththeareaunder y = x
1
over theinterval [n, 2n] toprovethat
_
2n
n
dx
x
+
1
2n
c
n

_
2n
n
dx
x
+
1
n
(c) UsetheSqueezeTheoremtodetermine lim
n
c
n
.
89. Let a
n
= H
n
lnn, whereH
n
isthenthharmonicnumber
H
n
= 1+
1
2
+
1
3
+ +
1
n
(a) Showthat a
n
0for n 1. Hint: Showthat H
n

_
n+1
1
dx
x
.
(b) Showthat {a
n
} isdecreasingbyinterpretinga
n
a
n+1
asanarea.
(c) Provethat lim
n
a
n
exists.
This limit, denoted , is knownas Eulers Constant. It appears inmany areas of mathematics, includinganalysis and
number theory, andhasbeencalculatedtomorethan100milliondecimal places, but it isstill not knownwhether isan
irrational number. Therst 10digitsare 0.5772156649.
solution
(a) Sincethefunctiony =
1
x
isdecreasing, theleft endpoint approximationtotheintegral
_
n+1
1
dx
x
isgreater thanthis
integral; that is,
1 1+
1
2
1+
1
3
1+ +
1
n
1
_
n+1
1
dx
x
or
H
n

_
n+1
1
dx
x
.
1
1
y
x
2 3 n n + 1
1/n
1
2
1
3
Moreover, sincethefunctiony =
1
x
ispositivefor x > 0, wehave:
_
n+1
1
dx
x
_
n
1
dx
x
.
Thus,
H
n

_
n
1
dx
x
= lnx
n
1
= lnn ln1= lnn,
and
a
n
= H
n
lnn 0 for all n 1.
(b) Toshowthat {a
n
} isdecreasing, weconsider thedifferencea
n
a
n+1
:
a
n
a
n+1
= H
n
lnn
_
H
n+1
ln(n +1)
_
= H
n
H
n+1
+ln(n +1) lnn
= 1+
1
2
+ +
1
n
_
1+
1
2
+ +
1
n
+
1
n +1
_
+ln(n +1) lnn
=
1
n +1
+ln(n +1) lnn.
Now, ln(n +1) lnn =
_
n+1
n
dx
x
, whereas
1
n+1
istheright endpoint approximationtotheintegral
_
n+1
n
dx
x
. Recalling
y =
1
x
isdecreasing, it followsthat
_
n+1
n
dx
x
1
n +1
y
x
n n + 1
y =
1
x
1
n + 1
so
a
n
a
n+1
0.
(c) By parts (a) and(b), {a
n
} is decreasingand0is alower boundfor this sequence. Hence0 a
n
a
1
for all n. A
monotonicandboundedsequenceisconvergent, solim
n
a
n
exists.
11.2 Summing an Infinite Series
1. What roledopartial sumsplayindeningthesumof aninniteseries?
solution Thesumof aninniteseriesisdenedasthelimitof thesequenceof partial sums. If thelimitof thissequence
doesnot exist, theseriesissaidtodiverge.
S E C T I ON 11.2 Summing an Infinite Series 659
2. What isthesumof thefollowinginniteseries?
1
4
+
1
8
+
1
16
+
1
32
+
1
64
+
solution Thisisageometricserieswithc =
1
4
andr =
1
2
. Thesumof theseriesistherefore
1
4
1
1
2
=
1
4
1
2
=
1
2
.
3. Whathappensif youapplytheformulafor thesumof ageometricseriestothefollowingseries?Istheformulavalid?
1+3+3
2
+3
3
+3
4
+
solution This is ageometric series withc = 1andr = 3. Applyingtheformulafor thesumof ageometric series
thengives
n=0
3
n
=
1
13
=
1
2
.
Clearly, thisisnot valid: aserieswithall positivetermscannot haveanegativesum. Theformulaisnot validinthiscase
becauseageometricserieswithr = 3diverges.
4. Arvindassertsthat

n=1
1
n
2
= 0because
1
n
2
tendstozero. Isthisvalidreasoning?
solution Arvinds reasoning is not valid. Though the terms in the series do tend to zero, the general termin the
sequenceof partial sums,
S
n
= 1+
1
2
2
+
1
3
2
+ +
1
n
2
,
isclearlylarger than1. Thesumof theseriesthereforecannot bezero.
5. Colleenclaimsthat

n=1
1
n
convergesbecause
lim
n
1
n
= 0
Isthisvalidreasoning?
solution Colleensreasoningisnotvalid. Althoughthegeneral termof aconvergentseriesmusttendtozero, aseries
whosegeneral termtendstozeroneednot converge. Inthecaseof

n=1
1
n
, theseriesdivergeseventhoughitsgeneral
termtendstozero.
6. FindanN suchthat S
N
> 25for theseries

n=1
2.
solution TheNthpartial sumof theseriesis:
S
N
=
N
n=1
2= 2+ +2
. ,, .
N
= 2N.
7. Doesthereexist anN suchthat S
N
> 25for theseries

n=1
2
n
? Explain.
solution Theseries

n=1
2
n
isaconvergentgeometricserieswiththecommonratior =
1
2
. Thesumof theseriesis:
S =
1
2
1
1
2
= 1.
Noticethat thesequenceof partial sums{S
N
} isincreasingandconvergesto1; thereforeS
N
1for all N. Thus, there
doesnot exist anN suchthat S
N
> 25.
8. Giveanexampleof adivergent inniteserieswhosegeneral termtendstozero.
solution Consider theseries

n=1
1
n
9
10
. Thegeneral termtendstozero, since lim
n
1
n
9
10
= 0. However, theNthpartial
sumsatisesthefollowinginequality:
S
N
=
1
1
9
10
+
1
2
9
10
+ +
1
N
9
10

N
N
9
10
= N
1
9
10 = N
1
10.
Thatis, S
N
N
1
10 for all N. Since lim
N
N
1
10 = , thesequenceof partial sumsS
n
diverges; hence, theseries

n=1
1
n
9
10
diverges.
Exercises
1. Findaformulafor thegeneral terma
n
(not thepartial sum) of theinniteseries.
(a)
1
3
+
1
9
+
1
27
+
1
81
+ (b)
1
1
+
5
2
+
25
4
+
125
8
+
(c)
1
1
2
2
2 1
+
3
3
3 2 1
4
4
4 3 2 1
+
(d)
2
1
2
+1
+
1
2
2
+1
+
2
3
2
+1
+
1
4
2
+1
+
solution
(a) Thedenominatorsof thetermsarepowersof 3, startingwiththerst power. Hence, thegeneral termis:
a
n
=
1
3
n
.
(b) Thenumeratorsarepowersof 5, andthedenominatorsarethesamepowersof 2. Therst termisa
1
= 1so,
a
n
=
_
5
2
_
n1
.
(c) Thegeneral termof thisseriesis,
a
n
= (1)
n+1
n
n
n!
.
(d) Noticethat thenumeratorsof a
n
equal 2for oddvaluesof n and1for evenvaluesof n. Thus,
a
n
=
2
n
2
+1
oddn
1
n
2
+1
evenn
Theformulacanalsoberewrittenasfollows:
a
n
=
1+
(1)
n+1
+1
2
n
2
+1
.
Writeinsummationnotation:
(a) 1+
1
4
+
1
9
+
1
16
+ (b)
1
9
+
1
16
+
1
25
+
1
36
+
(c) 1
1
3
+
1
5
1
7
+
(d)
125
9
+
625
16
+
3125
25
+
15,625
36
+
In Exercises 36, compute the partial sums S
2
, S
4
, and S
6
.
3. 1+
1
2
2
+
1
3
2
+
1
4
2
+
solution
S
2
= 1+
1
2
2
=
5
4
;
S
4
= 1+
1
2
2
+
1
3
2
+
1
4
2
=
205
144
;
S
6
= 1+
1
2
2
+
1
3
2
+
1
4
2
+
1
5
2
+
1
6
2
=
5369
3600
.
k=1
(1)
k
k
1
5.
1
1 2
+
1
2 3
+
1
3 4
+
solution
S
2
=
1
1 2
+
1
2 3
=
1
2
+
1
6
=
4
6
=
2
3
;
S
4
= S
2
+a
3
+a
4
=
2
3
+
1
3 4
+
1
4 5
=
2
3
+
1
12
+
1
20
=
4
5
;
S
6
= S
4
+a
5
+a
6
=
4
5
+
1
5 6
+
1
6 7
=
4
5
+
1
30
+
1
42
=
6
7
.
j=1
1
j!
7. TheseriesS = 1+
_
1
5
_
+
_
1
5
_
2
+
_
1
5
_
3
+ convergesto
5
4
. CalculateS
N
for N = 1, 2, . . . until youndanS
N
that approximates
5
4
withanerror lessthan0.0001.
solution
S
1
= 1
S
2
= 1+
1
5
=
6
5
= 1.2
S
3
= 1+
1
5
+
1
25
=
31
25
= 1.24
S
3
= 1+
1
5
+
1
25
+
1
125
=
156
125
= 1.248
S
4
= 1+
1
5
+
1
25
+
1
125
+
1
625
=
781
625
= 1.2496
S
5
= 1+
1
5
+
1
25
+
1
125
+
1
625
+
1
3125
=
3906
3125
= 1.24992
Notethat
1.25S
5
= 1.251.24992= 0.00008< 0.0001
Theseries S =
1
0!

1
1!
+
1
2!

1
3!
+ is known to convergeto e
1
(recall that 0! = 1). CalculateS
N
for
N = 1, 2, . . . until youndanS
N
that approximatese
1
withanerror lessthan0.001.
In Exercises 9 and 10, use a computer algebra system to compute S
10
, S
100
, S
500
, and S
1000
for the series. Do these
values suggest convergence to the given value?
9.
3
4
=
1
2 3 4
1
4 5 6
+
1
6 7 8
1
8 9 10
+
solution Write
a
n
=
(1)
n+1
2n (2n +1) (2n +2)
Then
S
N
=
N
i=1
a
n
Computing, wend
3
4
0.0353981635
S
10
0.03535167962
S
100
0.03539810274
S
500
0.03539816290
S
1000
0.03539816334
It appearsthat S
N

3
4
.
4
90
= 1+
1
2
4
+
1
3
4
+
1
4
4
+
11. CalculateS
3
, S
4
, andS
5
andthenndthesumof thetelescopingseries
S =
n=1
_
1
n +1
1
n +2
_
solution
S
3
=
_
1
2
1
3
_
+
_
1
3
1
4
_
+
_
1
4
1
5
_
=
1
2
1
5
=
3
10
;
S
4
= S
3
+
_
1
5
1
6
_
=
1
2
1
6
=
1
3
;
S
5
= S
4
+
_
1
6
1
7
_
=
1
2
1
7
=
5
14
.
Thegeneral terminthesequenceof partial sumsis
S
N
=
_
1
2
1
3
_
+
_
1
3
1
4
_
+
_
1
4
1
5
_
+ +
_
1
N +1
1
N +2
_
=
1
2
1
N +2
;
thus,
S = lim
N
S
N
= lim
N
_
1
2
1
N +2
_
=
1
2
.
Thesumof thetelescopingseriesistherefore
1
2
.
Write

n=3
1
n(n 1)
asatelescopingseriesandnditssum.
13. CalculateS
3
, S
4
, andS
5
andthenndthesumS =
n=1
1
4n
2
1
usingtheidentity
1
4n
2
1
=
1
2
_
1
2n 1
1
2n +1
_
solution
S
3
=
1
2
_
1
1
1
3
_
+
1
2
_
1
3
1
5
_
+
1
2
_
1
5
1
7
_
=
1
2
_
1
1
7
_
=
3
7
;
S
4
= S
3
+
1
2
_
1
7
1
9
_
=
1
2
_
1
1
9
_
=
4
9
;
S
5
= S
4
+
1
2
_
1
9
1
11
_
=
1
2
_
1
1
11
_
=
5
11
.
S
N
=
1
2
_
1
1
1
3
_
+
1
2
_
1
3
1
5
_
+
1
2
_
1
5
1
7
_
+ +
1
2
_
1
2N 1
1
2N +1
_
=
1
2
_
1
1
2N +1
_
;
thus,
S = lim
N
S
N
= lim
N
1
2
_
1
1
2N +1
_
=
1
2
.
Usepartial fractionstorewrite

n=1
1
n(n +3)
asatelescopingseriesandnditssum.
15. Findthesumof
1
1 3
+
1
3 5
+
1
5 7
+ .
solution Wemaywritethissumas
n=1
1
(2n 1)(2n +1)
=
n=1
1
2
_
1
2n 1
1
2n +1
_
.
S
N
=
1
2
_
1
1
1
3
_
+
1
2
_
1
3
1
5
_
+
1
2
_
1
5
1
7
_
+ +
1
2
_
1
2N 1
1
2N +1
_
=
1
2
_
1
1
2N +1
_
;
thus,
lim
N
S
N
= lim
N
1
2
_
1
1
2N +1
_
=
1
2
,
and
n=1
1
(2n 1)(2n +1)
=
1
2
.
Findaformulafor thepartial sumS
N
of

n=1
(1)
n1
andshowthat theseriesdiverges.
In Exercises 1722, use Theorem 3 to prove that the following series diverge.
17.

n=1
n
10n +12
solution Thegeneral term,
n
10n +12
, haslimit
lim
n
n
10n +12
= lim
n
1
10+(12/n)
=
1
10
Sincethegeneral termdoesnot tendtozero, theseriesdiverges.
n=1
n
_
n
2
+1
19.
0
1
1
2
+
2
3
3
4
+
solution Thegeneral terma
n
= (1)
n1n1
n
doesnottendtozero. Infact, becauselim
n
n1
n
= 1, lim
n
a
n
doesnot exist. ByTheorem3, weconcludethat thegivenseriesdiverges.
n=1
(1)
n
n
2
21. cos
1
2
+cos
1
3
+cos
1
4
+
solution Thegeneral terma
n
= cos
1
n+1
tends to 1, not zero. By Theorem3, weconcludethat thegiven series
diverges.
n=0
_
_
4n
2
+1n
_
In Exercises 2336, use the formula for the sum of a geometric series to nd the sum or state that the series diverges.
23.
1
1
+
1
8
+
1
8
2
+
solution Thisisageometricserieswithc = 1andr =
1
8
, soitssumis
1
1
1
8
=
1
7/8
=
8
7
4
3
5
3
+
4
4
5
4
+
4
5
5
5
+
25.

n=3
_
3
11
_
n
solution Rewritethisseriesas
n=3
_
11
3
_
n
Thisisageometricserieswithr =
11
3
> 1, soit isdivergent.
n=2
7 (3)
n
5
n
27.

n=4
_
4
9
_
n
4
9
, startingat n = 4. Itssumisthus
cr
4
1r
=
c
r
4
r
5
=
1
4
4
9
4
+
4
5
9
5
=
9
5
9 4
4
+4
5
=
59,049
3328
n=0
_
e
_
n 29.

n=1
e
n
solution Rewritetheseriesas
n=1
_
1
e
_
n
torecognizeit asageometricserieswithc =
1
e
andr =
1
e
. Thus,
n=1
e
n
=
1
e
1
1
e
=
1
e 1
.
n=2
e
32n
31.

n=0
8+2
n
5
n
solution Rewritetheseriesas
n=0
8
5
n
+
n=0
2
n
5
n
=
n=0
8
_
1
5
_
n
+
n=0
_
2
5
_
n
,
whichisasumof twogeometricseries. Therst serieshasc = 8
_
1
5
_
0
= 8andr =
1
5
; thesecondhasc =
_
2
5
_
0
= 1
andr =
2
5
. Thus,
n=0
8
_
1
5
_
n
=
8
1
1
5
=
8
4
5
= 10,
n=0
_
2
5
_
n
=
1
1
2
5
=
1
3
5
=
5
3
,
and
n=0
8+2
n
5
n
= 10+
5
3
=
35
3
.
n=0
3(2)
n
5
n
8
n
33. 5
5
4
+
5
4
2
5
4
3
+
1
4
. Thus,
n=0
5
_
1
4
_
n
=
5
1
_
1
4
_ =
5
1+
1
4
=
5
5
4
= 4.
2
3
7
+
2
4
7
2
+
2
5
7
3
+
2
6
7
4
+
35.
7
8
49
64
+
343
512
2401
4096
+
solution Thisisageometricserieswithc =
7
8
andr =
7
8
. Thus,
n=0
7
8
7
8
_
n
=
7
8
1
_
7
8
_ =
7
8
15
8
=
7
15
.
25
9
+
5
3
+1+
3
5
+
9
25
+
27
125
+
37. Whichof thefollowingarenot geometricseries?
(a)

n=0
7
n
29
n
(b)

n=3
1
n
4
(c)

n=0
n
2
2
n
(d)

n=5
n
solution
(a)

n=0
7
n
29
n
=
n=0
_
7
29
_
n
: thisisageometricserieswithcommonratior =
7
29
.
(b) Theratiobetweentwosuccessivetermsis
a
n+1
a
n
=
1
(n+1)
4
1
n
4
=
n
4
(n +1)
4
=
_
n
n +1
_
4
.
Thisratioisnot constant sinceit dependsonn. Hence, theseries

n=3
1
n
4
isnot ageometricseries.
(c) Theratiobetweentwosuccessivetermsis
a
n+1
a
n
=
(n+1)
2
2
n+1
n
2
2
n
=
(n +1)
2
n
2

2
n
2
n+1
=
_
1+
1
n
_
2
1
2
.
Thisratioisnot constant sinceit dependsonn. Hence, theseries

n=0
n
2
2
n
isnot ageometricseries.
(d)

n=5
n
=
n=5
_
1
_
n
: thisisageometricserieswithcommonratior =
1
.
Usethemethodof Example8toshowthat

k=1
1
k
1/3
diverges.
39. Provethatif

n=1
a
n
convergesand

n=1
b
n
diverges, then

n=1
(a
n
+b
n
) diverges. Hint: If not, deriveacontradiction
bywriting
n=1
b
n
=
n=1
(a
n
+b
n
)
n=1
a
n
solution Supposetothecontrarythat
n=1
a
n
converges,
n=1
b
n
diverges, but
n=1
(a
n
+b
n
) converges. Then
bytheLinearityof InniteSeries, wehave
n=1
b
n
=
n=1
(a
n
+b
n
)
n=1
a
n
sothat
n=1
b
n
converges, acontradiction.
Provethedivergenceof

n=0
9
n
+2
n
5
n
.
41. Giveacounterexampletoshowthat eachof thefollowingstatementsisfalse.
(a) If thegeneral terma
n
tendstozero, then

n=1
a
n
= 0.
(b) TheNthpartial sumof theinniteseriesdenedby{a
n
} isa
N
.
(c) If a
n
tendstozero, then

n=1
a
n
converges.
(d) If a
n
tendstoL, then

n=1
a
n
= L.
solution
(a) Let a
n
= 2
n
. Thenlim
n
a
n
= 0, but a
n
is ageometric series withc = 2
0
= 1andr = 1/2, so its sumis
1
1(1/2)
= 2.
(b) Let a
n
= 1. Thenthen
th
partial sumisa
1
+a
2
+ +a
n
= n whilea
n
= 1.
(c) Let a
n
=
1
n
. Anexampleinthetext showsthat whilea
n
tendstozero, thesum
n=1
a
n
doesnot converge.
(d) Let a
n
= 1. Thenclearlya
n
tendstoL = 1, whiletheseries
n=1
a
n
obviouslydiverges.
Supposethat S =
n=1
a
n
isaninniteserieswithpartial sumS
N
= 5
2
N
2
.
(a) What arethevaluesof
10
n=1
a
n
and
16
n=5
a
n
?
(b) What isthevalueof a
3
?
(c) Findageneral formulafor a
n
.
(d) Findthesum

n=1
a
n
.
43. Computethetotal areaof the(innitelymany) trianglesinFigure4.
1
8
1
4
1
2
1
16
1
2
y
x
1
FIGURE 4
solution Theareaof atrianglewithbaseB andheight H isA =
1
2
BH. Becauseall of thetrianglesinFigure4have
height
1
2
, theareaof each triangleequals one-quarter of thebase. Now, for n 0, thenth trianglehas abasewhich
extendsfromx =
1
2
n+1
tox =
1
2
n
. Thus,
B =
1
2
n

1
2
n+1
=
1
2
n+1
and A =
1
4
B =
1
2
n+3
.
Thetotal areaof thetrianglesisthengivenbythegeometricseries
n=0
1
2
n+3
=
n=0
1
8
_
1
2
_
n
=
1
8
1
1
2
=
1
4
.
Thewinner of alotteryreceivesmdollarsattheendof eachyear for N years. Thepresentvalue(PV) of thisprize
intodaysdollarsisPV =
N
i=1
m(1+r)
i
, wherer istheinterest rate. CalculatePV if m = $50,000, r = 0.06, and
N = 20. What isPV if N = ?
45. Findthetotal lengthof theinnitezigzagpathinFigure5(eachzagoccursat anangleof

4
).
1
/4 /4
FIGURE 5
solution Becausetheangleat thelower left inFigure5hasmeasure
4
andeachzaginthepathoccursat anangleof
4
, every triangleinthegureisanisoscelesright triangle. Accordingly, thelengthof eachnewsegment inthepathis
1
2
timesthelengthof theprevioussegment. Sincetherst segment haslength1, thetotal lengthof thepathis
n=0
_
1
2
_
n
=
1
1
1
2
=
21
= 2+
2.
Evaluate

n=1
1
n(n +1)(n +2)
. Hint: FindconstantsA, B, andC suchthat
1
n(n +1)(n +2)
=
A
n
+
B
n +1
+
C
n +2
47. Showthat if a isapositiveinteger, then
n=1
1
n(n +a)
=
1
a
_
1+
1
2
+ +
1
a
_
solution Bypartial fractiondecomposition
1
n (n +a)
=
A
n
+
B
n +a
;
clearingthedenominatorsgives
1= A(n +a) +Bn.
Settingn = 0thenyieldsA =
1
a
, whilesettingn = a yieldsB =
1
a
. Thus,
1
n (n +a)
=
1
a
n
1
a
n +a
=
1
a
_
1
n
1
n +a
_
,
and
n=1
1
n(n +a)
=
n=1
1
a
_
1
n
1
n +a
_
.
For N > a, theNthpartial sumis
S
N
=
1
a
_
1+
1
2
+
1
3
+ +
1
a
_
1
a
_
1
N +1
+
1
N +2
+
1
N +3
+ +
1
N +a
_
.
Thus,
n=1
1
n(n +a)
= lim
N
S
N
=
1
a
_
1+
1
2
+
1
3
+ +
1
a
_
.
A ball droppedfromaheight of 10ft beginstobounce. Eachtimeit strikestheground, it returnstotwo-thirdsof
itspreviousheight. What isthetotal distancetraveledbytheball if it bouncesinnitelymanytimes?
49. Let {b
n
} beasequenceandlet a
n
= b
n
b
n1
. Showthat

n=1
a
n
convergesif andonlyif lim
n
b
n
exists.
solution Let a
n
= b
n
b
n1
. Thegeneral terminthesequenceof partial sumsfor theseries

n=1
a
n
isthen
S
N
= (b
1
b
0
) +(b
2
b
1
) +(b
3
b
2
) + +(b
N
b
N1
) = b
N
b
0
.
Now, if lim
N
b
N
exists, then so does lim
N
S
N
and

n=1
a
n
converges. On theother hand, if

n=1
a
n
converges, then
lim
N
S
N
exists, whichimpliesthat lim
N
b
N
alsoexists. Thus,

n=1
a
n
convergesif andonlyif lim
n
b
n
exists.
AssumptionsMatter Show, by givingcounterexamples, that theassertions of Theorem1arenot validif the
series

n=0
a
n
and

n=0
b
n
arenot convergent.
Exercises 5153 use the formula
1+r +r
2
+ +r
N1
=
1r
N
1r
7
51. ProfessorGeorgeAndrewsof PennsylvaniaStateUniversityobservedthatwecanuseEq.(7)tocalculatethederivative
of f (x) = x
N
(for N 0). Assumethat a = 0andlet x = ra. Showthat
f
(a) = lim
xa
x
N
a
N
x a
= a
N1
lim
r1
r
N
1
r 1
andevaluatethelimit.
solution Accordingtothedenitionof derivativeof f (x) at x = a
f
(a) = lim
xa
x
N
a
N
x a
.
Now, let x = ra. Thenx a if andonlyif r 1, and
f
(a) = lim
xa
x
N
a
N
x a
= lim
r1
(ra)
N
a
N
ra a
= lim
r1
a
N
_
r
N
1
_
a (r 1)
= a
N1
lim
r1
r
N
1
r 1
.
ByEq. (7) for ageometricsum,
1r
N
1r
=
r
N
1
r 1
= 1+r +r
2
+ +r
N1
,
so
lim
r1
r
N
1
r 1
= lim
r1
_
1+r +r
2
+ +r
N1
_
= 1+1+1
2
+ +1
N1
= N.
Therefore, f
(a) = a
N1
N = Na
N1
Pierre de Fermat used geometric series to compute the area under the graph of f (x) = x
N
over [0, A]. For
0< r < 1, let F(r) bethesumof theareasof theinnitelymanyright-endpoint rectangleswithendpointsAr
n
, as
inFigure6. Asr tendsto1, therectanglesbecomenarrower andF(r) tendstotheareaunder thegraph.
(a) Showthat F(r) = A
N+1
1r
1r
N+1
.
(b) UseEq. (7) toevaluate
_
A
0
x
N
dx = lim
r1
F(r).
53. VerifytheGregoryLeibniz formulaasfollows.
(a) Set r = x
2
inEq. (7) andrearrangetoshowthat
1
1+x
2
= 1x
2
+x
4
+(1)
N1
x
2N2
+
(1)
N
x
2N
1+x
2
(b) Show, byintegratingover [0, 1], that
4
= 1
1
3
+
1
5
1
7
+ +
(1)
N1
2N 1
+(1)
N
_
1
0
x
2N
dx
1+x
2
(c) UsetheComparisonTheoremfor integralstoprovethat
0
_
1
0
x
2N
dx
1+x
2

1
2N +1
Hint: Observethat theintegrandis x
2N
.
(d) Provethat
4
= 1
1
3
+
1
5
1
7
+
1
9

Hint: Use (b) and (c) to show that the partial sums S
N
of satisfy

S
N

4

1
2N+1
, and thereby conclude that
lim
N
S
N
=

4
.
solution
(a) Start withEq. (7), andsubstitutex
2
for r:
1+r +r
2
+ +r
N1
=
1r
N
1r
1x
2
+x
4
+ +(1)
N1
x
2N2
=
1(1)
N
x
2N
1(x
2
)
1x
2
+x
4
+ +(1)
N1
x
2N2
=
1
1+x
2
(1)
N
x
2N
1+x
2
1
1+x
2
= 1x
2
+x
4
+ +(1)
N1
x
2N2
+
(1)
N
x
2N
1+x
2
(b) Theintegralsof bothsidesmust beequal. Now,
_
1
0
1
1+x
2
dx = tan
1
x
1
0
= tan
1
1tan
1
0=

4
while
_
1
0
_
1x
2
+x
4
+ +(1)
N1
x
2N2
+
(1)
N
x
2N
1+x
2
_
dx
=
_
x
1
3
x
3
+
1
5
x
5
+ +(1)
N1
1
2N 1
x
2N1
_
+(1)
N
_
1
0
x
2N
dx
1+x
2
= 1
1
3
+
1
5
+ +(1)
N1
1
2N 1
+(1)
N
_
1
0
x
2N
dx
1+x
2
(c) Notethat for x [0, 1], wehave1+x
2
1, sothat
0
x
2N
1+x
2
x
2N
BytheComparisonTheoremfor integrals, wethenseethat
0
_
1
0
x
2N
dx
1+x
2

_
1
0
x
2N
dx =
1
2N +1
x
2N+1
1
0
=
1
2N +1
(d) Write
a
n
= (1)
n
1
2n 1
, n 1
andlet S
N
bethepartial sums. Then
S
N

4
(1)
N
_
1
0
x
2N
dx
1+x
2
=
_
1
0
x
2N
dx
1+x
2

1
2N +1
Thuslim
N
S
N
=

4
sothat
4
= 1
1
3
+
1
5
1
7
+
1
9
. . .
CantorsDisappearingTable(followingLarry Knopof HamiltonCollege) Takeatableof lengthL (Figure
7). Atstage1, removethesectionof lengthL/4centeredatthemidpoint. Twosectionsremain, eachwithlengthless
thanL/2. Atstage2, removesectionsof lengthL/4
2
fromeachof thesetwosections(thisstageremovesL/8of the
table). Nowfour sectionsremain, eachof lengthlessthanL/4. At stage3, removethefour central sectionsof length
L/4
3
, etc.
(a) Showthat at theNthstage, eachremainingsectionhaslengthlessthanL/2
N
andthat thetotal amount of table
removedis
L
_
1
4
+
1
8
+
1
16
+ +
1
2
N+1
_
(b) Showthat inthelimit asN , preciselyone-half of thetableremains.
Thisresult iscurious, becausetherearenononzerointervalsof tableleft (at eachstage, theremainingsectionshave
alengthlessthanL/2
N
). Sothetablehasdisappeared. However, wecanplaceanyobject longer thanL/4onthe
table. It will not fall throughbecauseit will not t throughanyof theremovedsections.
55. TheKochsnowake(describedin1904bySwedishmathematicianHelgevonKoch) isaninnitelyjaggedfractal
curve obtained as a limit of polygonal curves (it is continuous but has no tangent line at any point). Begin with an
equilateral triangle(stage0) andproducestage1byreplacingeachedgewithfour edgesof one-thirdthelength, arranged
asinFigure8. Continuetheprocess: At thenthstage, replaceeachedgewithfour edgesof one-thirdthelength.
(a) Showthat theperimeter P
n
of thepolygonat thenthstagesatises P
n
=
4
3
P
n1
. Provethat lim
n
P
n
= . The
snowakehasinnitelength.
(b) Let A
0
betheareaof theoriginal equilateral triangle. Showthat (3)4
n1
newtriangles areaddedat thenthstage,
eachwithareaA
0
/9
n
(for n 1). Showthat thetotal areaof theKochsnowakeis
8
5
A
0
.
Stage1 Stage3 Stage2
FIGURE 8
S E C T I ON 11.3 Convergence of Series with Positive Terms 669
solution
(a) Eachedgeof thepolygonatthe(n 1)ststageisreplacedbyfour edgesof one-thirdthelength; hencetheperimeter
of thepolygonatthenthstageis
4
3
timestheperimeter of thepolygonatthe(n 1)thstage. Thatis, P
n
=
4
3
P
n1
. Thus,
P
1
=
4
3
P
0
; P
2
=
4
3
P
1
=
_
4
3
_
2
P
0
, P
3
=
4
3
P
2
=
_
4
3
_
3
P
0
,
and, ingeneral, P
n
=
_
4
3
_
n
P
0
. Asn , it followsthat
lim
n
P
n
= P
0
lim
n
_
4
3
_
n
= .
(b) Wheneachedgeisreplacedbyfour edgesof one-thirdthelength, onenewtriangleiscreated. Atthe(n 1)ststage,
thereare3 4
n1
edgesinthesnowake, so3 4
n1
newtrianglesaregeneratedat thenthstage. Becausetheareaof an
equilateral triangleis proportional tothesquareof its sidelengthandthesidelengthfor eachnewtriangleis one-third
thesidelengthof trianglesfromthepreviousstage, it followsthat theareaof thetrianglesaddedat eachstageisreduced
byafactor of
1
9
fromtheareaof thetrianglesaddedat thepreviousstage. Thus, eachtriangleaddedat thenthstagehas
anareaof A
0
/9
n
. Thismeansthat thenthstagecontributes
3 4
n1
A
0
9
n
=
3
4
A
0
_
4
9
_
n
totheareaof thesnowake. Thetotal areaistherefore
A = A
0
+
3
4
A
0

n=1
_
4
9
_
n
= A
0
+
3
4
A
0
4
9
1
4
9
= A
0
+
3
4
A
0
4
5
=
8
5
A
0
.
11.3 Convergence of Series with Positive Terms
1. Let S =
n=1
a
n
. If thepartial sumsS
N
areincreasing, then(choosethecorrect conclusion):
(a) {a
n
} isanincreasingsequence.
(b) {a
n
} isapositivesequence.
solution Thecorrect responseis(b). Recall that S
N
= a
1
+a
2
+a
3
+ +a
N
; thus, S
N
S
N1
= a
N
. If S
N
is
increasing, thenS
N
S
N1
0. It thenfollowsthat a
N
0; that is, {a
n
} isapositivesequence.
2. What arethehypothesesof theIntegral Test?
solution Thehypothesesfor theIntegral Testare: A functionf (x) suchthata
n
= f (n) mustbepositive, decreasing,
andcontinuousfor x 1.
3. Whichtest wouldyouusetodeterminewhether

n=1
n
3.2
converges?
solution Becausen
3.2
=
1
n
3.2
, weseethat theindicatedseriesisap-serieswithp = 3.2> 1. Therefore, theseries
converges.
4. Whichtest wouldyouusetodeterminewhether

n=1
1
2
n
+
n
converges?
solution Because
1
2
n
+
n
<
1
2
n
=
_
1
2
_
n
,
and
n=1
_
1
2
_
n
is aconvergent geometric series, thecomparisontest wouldbeanappropriatechoiceto establishthat thegivenseries
converges.
5. Ralphhopestoinvestigatetheconvergenceof

n=1
e
n
n
bycomparingit with

n=1
1
n
. IsRalphontheright track?
solution No, Ralphisnot ontheright track. For n 1,
e
n
n
<
1
n
;
however,

n=1
1
n
is adivergent series. TheComparisonTest thereforedoes not allowus todrawaconclusionabout the
convergenceor divergenceof theseries

n=1
e
n
n
.
Exercises
In Exercises 114, use the Integral Test to determine whether the innite series is convergent.
1.

n=1
1
n
4
1
x
4
. Thisfunctioniscontinuous, positiveanddecreasingontheinterval x 1, sotheIntegral
Test applies. Moreover,
_

1
dx
x
4
= lim
R
_
R
1
x
4
dx =
1
3
lim
R
_
1
R
3
1
_
=
1
3
.
Theintegral converges; hence, theseries

n=1
1
n
4
alsoconverges.
n=1
1
n +3
3.

n=1
n
1/3
1
3 =
1
3
x
. Thisfunctioniscontinuous, positiveanddecreasingontheinterval x 1, sothe
Integral Test applies. Moreover,
_

1
x
1/3
dx = lim
R
_
R
1
x
1/3
dx =
3
2
lim
R
_
R
2/3
1
_
= .
Theintegral diverges; hence, theseries

n=1
n
1/3
alsodiverges.
n=5
1
n 4
5.

n=25
n
2
(n
3
+9)
5/2
x
2
_
x
3
+9
_
5/2
. Thisfunctionispositiveandcontinuousfor x 25. Moreover, because
f
(x) =
2x(x
3
+9)
5/2
x
2
5
2
(x
3
+9)
3/2
3x
2
(x
3
+9)
5
=
x(3611x
3
)
2(x
3
+9)
7/2
,
weseethat f
(x) < 0 for x 25, so f is decreasing on theinterval x 25. TheIntegral Test thereforeapplies. To
evaluatetheimproper integral, weusethesubstitutionu = x
3
+9, du = 3x
2
dx. Wethennd
_

25
x
2
(x
3
+9)
5/2
dx = lim
R
_
R
25
x
2
(x
3
+9)
5/2
dx =
1
3
lim
R
_
R
3
+9
15634
du
u
5/2
=
2
9
lim
R
_
1
(R
3
+9)
3/2
1
15634
3/2
_
=
2
9 15634
3/2
.

n=25
n
2
_
n
3
+9
_
5/2
alsoconverges.
n=1
n
(n
2
+1)
3/5
7.

n=1
1
n
2
+1
1
x
2
+1
. Thisfunctionispositive, decreasingandcontinuousontheinterval x 1, hencethe
Integral Test applies. Moreover,
_

1
dx
x
2
+1
= lim
R
_
R
1
dx
x
2
+1
= lim
R
_
tan
1
R

4
_
=

2

4
=

4
.

n=1
1
n
2
+1
alsoconverges.
n=4
1
n
2
1
9.

n=1
1
n(n +1)
1
x(x +1)
. This functionis positive, continuous anddecreasingontheinterval x 1, so the
Integral Test applies. Wecomputetheimproper integral usingpartial fractions:
_

1
dx
x(x +1)
= lim
R
_
R
1
_
1
x
1
x +1
_
dx = lim
R
ln
x
x +1
R
1
= lim
R
_
ln
R
R +1
ln
1
2
_
= ln1ln
1
2
= ln2.

n=1
1
n(n +1)
converges.
n=1
ne
n
2 11.

n=2
1
n(lnn)
2
1
x(lnx)
2
. Thisfunctionispositiveandcontinuousfor x 2. Moreover,
f
(x) =
1
x
2
(lnx)
4
_
1 (lnx)
2
+x 2(lnx)
1
x
_
=
1
x
2
(lnx)
4
_
(lnx)
2
+2lnx
_
.
Sincelnx > 0forx > 1, f
(x) isnegativeforx > 1; hence, f isdecreasingforx 2. Tocomputetheimproperintegral,

wemakethesubstitutionu = lnx, du =
1
x
dx. Weobtain:
_

2
1
x(lnx)
2
dx = lim
R
_
R
2
1
x(lnx)
2
dx = lim
R
_
lnR
ln2
du
u
2
= lim
R
_
1
lnR
1
ln2
_
=
1
ln2
.

n=2
1
n(lnn)
2
alsoconverges.
n=1
lnn
n
2
13.

n=1
1
2
lnn
solution Notethat
2
lnn
= (e
ln2
)
lnn
= (e
lnn
)
ln2
= n
ln2
.
Thus,
n=1
1
2
lnn
=
n=1
1
n
ln2
.
Now, let f (x) =
1
x
ln2
. Thisfunctionispositive, continuousanddecreasingontheinterval x 1; therefore, theIntegral
Test applies. Moreover,
_

1
dx
x
ln2
= lim
R
_
R
1
dx
x
ln2
=
1
1ln2
lim
R
(R
1ln2
1) = ,
because1ln2> 0. Theintegral diverges; hence, theseries

n=1
1
2
lnn
alsodiverges.
n=1
1
3
lnn
15. Showthat

n=1
1
n
3
+8n
convergesbyusingtheComparisonTest with

n=1
n
3
.
solution Wecomparetheserieswiththep-series

n=1
n
3
. For n 1,
1
n
3
+8n
1
n
3
.
Since

n=1
1
n
3
converges(itisap-serieswithp = 3> 1), theseries

n=1
1
n
3
+8n
alsoconvergesbytheComparisonTest.
Showthat

n=2
1
_
n
2
3
divergesbycomparingwith

n=2
n
1
.
17. Let S =
n=1
1
n +
n
. Verifythat for n 1,
1
n +
1
n
,
1
n +
n
Caneither inequalitybeusedtoshowthat S diverges? Showthat
1
n +
1
2n
andconcludethat S diverges.
solution For n 1, n +
n n andn +
n. Takingthereciprocal of eachof theseinequalitiesyields

1
n +
1
n
and
1
n +
n
.
Theseinequalities indicatethat theseries

n=1
1
n +
n
is smaller than both

n=1
1
n
and

n=1
1
n
; however,

n=1
1
n
and
n=1
1
n
bothdivergesoneither inequalityallowsustoshowthat S diverges.
Ontheother hand, for n 1, n
n, so2n n +
n and
1
n +
1
2n
.
Theseries

n=1
1
2n
= 2

n=1
1
n
diverges, sincetheharmonic series diverges. TheComparisonTest thenlets us conclude
that thelarger series

n=1
1
n +
n
alsodiverges.
Whichof thefollowinginequalitiescanbeusedtostudytheconvergenceof

n=2
1
n
2
+
n
? Explain.
1
n
2
+
n
,
1
n
2
+
1
n
2
In Exercises 1930, use the Comparison Test to determine whether the innite series is convergent.
19.

n=1
1
n2
n
solution Wecomparewiththegeometricseries

n=1
_
1
2
_
n
. For n 1,
1
n2
n

1
2
n
=
_
1
2
_
n
.
Since

n=1
_
1
2
_
n
converges(itisageometricserieswithr =
1
2
), weconcludebytheComparisonTestthat

n=1
1
n2
n
also
converges.
n=1
n
3
n
5
+4n +1
21.

n=1
1
n
1/3
+2
n
solution For n 1,
1
n
1/3
+2
n

1
2
n
Theseries
n=1
1
2
n
isageometricserieswithr =
1
2
, soit converges. BytheComparisontest, sodoes

n=1
1
n
1/3
+2
n
.
n=1
1
_
n
3
+2n 1
23.

m=1
4
m! +4
m
solution For m 1,
4
m! +4
m

4
4
m
=
_
1
4
_
m1
.
Theseries

m=1
_
1
4
_
m1
is ageometric series withr =
1
4
, so it converges. By theComparisonTest wecantherefore
concludethat theseries

m=1
4
m! +4
m
alsoconverges.
n=4
n
n 3
25.

k=1
sin
2
k
k
2
solution For k 1, 0 sin
2
k 1, so
0
sin
2
k
k
2

1
k
2
.
Theseries

k=1
1
k
2
isap-serieswithp = 2> 1, soit converges. BytheComparisonTest wecanthereforeconcludethat
theseries

k=1
sin
2
k
k
2
alsoconverges.
k=2
k
1/3
k
5/4
k
27.

n=1
2
3
n
+3
n
solution Since3
n
> 0for all n,
2
3
n
+3
n

2
3
n
= 2
_
1
3
_
n
.
Theseries

n=1
2
_
1
3
_
n
1
3
, soit converges. BytheComparisonTheoremwecantherefore

n=1
2
3
n
+3
n
alsoconverges.
k=1
2
k
2 29.

n=1
1
(n +1)!
solution Notethat for n 2,
(n +1)! = 1 2 3 n (n +1)
. ,, .
n factors
2
n
sothat
n=1
1
(n +1)!
= 1+
n=2
1
(n +1)!
1+
n=2
1
2
n
But
n=2
1
2
n
isageometricserieswithratior =
1
2
, soitconverges. Bythecomparisontest,

n=1
1
(n +1)!
convergesas
well.
n=1
n!
n
3
Exercise 3136: For all a > 0and b > 1, the inequalities
lnn n
a
, n
a
< b
n
are true for n sufciently large (this can be proved using LHopitals Rule). Use this, together with the Comparison
Theorem, to determine whether the series converges or diverges.
31.

n=1
lnn
n
3
solution For n sufcientlylarge(sayn = k, althoughinthiscasen = 1sufces), wehavelnn n, sothat
n=k
lnn
n
3

n=k
n
n
3
=
n=k
1
n
2
Thisisap-serieswithp = 2> 1, soit converges. Thus
n=k
lnn
n
3
alsoconverges; addingback inthenitenumber of
termsfor 1 n k doesnot affect thisresult.
m=2
1
lnm
33.

n=1
(lnn)
100
n
1.1
solution ChooseN sothat lnn n
0.0005
for n N. Thenalsofor n > N, (lnn)
100
(n
0.0005
)
100
= n
0.05
. Then
n=N
(lnn)
100
n
1.1

n=N
n
0.05
n
1.1
=
n=N
1
n
1.05
But

n=N
1
n
1.05
is ap-series withp = 1.05 > 1, so is convergent. It follows that
n=N
(lnn)
1
00
n
1.1
is also convergent;
addingbackinthenitenumber of termsfor n = 1, 2, . . . , N 1showsthat

n=1
(lnn)
100
n
1.1
convergesaswell.
n=1
1
(lnn)
10
35.

n=1
n
3
n
solution ChooseN suchthat n 2
n
for n N. Then
n=N
n
3
n

n=N
_
2
3
_
n
Thelatter sumisageometricserieswithr =
2
3
< 1, soitconverges. Thustheseriesontheleftconvergesaswell.Adding
backinthenitenumber of termsfor n < N showsthat

n=1
n
3
n
converges.
n=1
n
5
2
n
37. Showthat

n=1
sin
1
n
2
converges. Hint: Usetheinequalitysinx x for x 0.
solution For n 1,
0
1
n
2
1< ;
therefore, sin
1
n
2
> 0for n 1. Moreover, for n 1,
sin
1
n
2

1
n
2
.
Theseries

n=1
1
n
2
isap-serieswithp = 2> 1, soit converges. BytheComparisonTest wecanthereforeconcludethat
theseries

n=1
sin
1
n
2
alsoconverges.
Does

n=2
sin(1/n)
lnn
converge? Hint: ByTheorem1inSection2.6, sin(1/n) > (cos(1/n))/n. Thussin(1/n) >
1/(2n) for n > 2
_
becausecos(1/n) >
1
2
_
.
In Exercises 3948, use the Limit Comparison Test to prove convergence or divergence of the innite series.
39.

n=2
n
2
n
4
1
solution Leta
n
=
n
2
n
4
1
. For largen,
n
2
n
4
1
n
2
n
4
=
1
n
2
, soweapplytheLimitComparisonTestwithb
n
=
1
n
2
.
Wend
L = lim
n
a
n
b
n
= lim
n
n
2
n
4
1
1
n
2
= lim
n
n
4
n
4
1
= 1.
Theseries

n=1
1
n
2
isap-serieswithp = 2> 1, soit converges; hence,

n=2
1
n
2
alsoconverges. BecauseL exists, bythe
Limit ComparisonTest wecanconcludethat theseries

n=2
n
2
n
4
1
converges.
n=2
1
n
2
n
41.

n=2
n
_
n
3
+1
solution Let a
n
=
n
_
n
3
+1
. For largen,
n
_
n
3
+1
n
3
=
1
n
, so weapply theLimit Comparison test with
b
n
=
1
n
. Wend
L = lim
n
a
n
b
n
= lim
n
n
n
3
+1
1
n
= lim
n
n
3
_
n
3
+1
= 1.
Theseries

n=1
1
n
is ap-series withp =
1
2
< 1, soit diverges; hence,

n=2
1
n
alsodiverges. BecauseL > 0, by the
Limit ComparisonTest wecanconcludethat theseries

n=2
n
_
n
3
+1
diverges.
n=2
n
3
_
n
7
+2n
2
+1
43.

n=3
3n +5
n(n 1)(n 2)
solution Let a
n
=
3n +5
n(n 1)(n 2)
. For largen,
3n +5
n(n 1)(n 2)
3n
n
3
=
3
n
2
, soweapplytheLimit Comparison
Test withb
n
=
1
n
2
. Wend
L = lim
n
a
n
b
n
= lim
n
3n+5
n(n+1)(n+2)
1
n
2
= lim
n
3n
3
+5n
2
n(n +1)(n +2)
= 3.
Theseries

n=1
1
n
2
is ap-series withp = 2 > 1, so it converges; hence, theseries

n=3
1
n
2
also converges. BecauseL
exists, bytheLimit ComparisonTest wecanconcludethat theseries

n=3
3n +5
n(n 1)(n 2)
converges.
n=1
e
n
+n
e
2n
n
2
45.

n=1
1
n +lnn
solution Let
a
n
=
1
n +lnn
For largen,
n +lnn
n, soapplytheComparisonTest withb
n
=
1
n
. Wend
L = lim
n
a
n
b
n
= lim
n
1
n +lnn
n
1
= lim
n
1
1+
lnn
n
= 1
Theseries

n=1
1
n
isap-serieswithp =
1
2
< 1, soitdiverges. BecauseL exists, theLimitComparisonTesttellsusthe
theoriginal seriesalsodiverges.
n=1
ln(n +4)
n
5/2
47.

n=1
_
1cos
1
n
_
Hint: Comparewith

n=1
n
2
.
solution Let a
n
= 1cos
1
n
, andapplytheLimit ComparisonTest withb
n
=
1
n
2
. Wend
L = lim
n
a
n
b
n
= lim
n
1cos
1
n
1
n
2
= lim
x
1cos
1
x
1
x
2
= lim
x
1
x
2
sin
1
x
2
x
3
=
1
2
lim
x
sin
1
x
1
x
.
Asx , u =
1
x
0, so
L =
1
2
lim
x
sin
1
x
1
x
=
1
2
lim
u0
sinu
u
=
1
2
.
Theseries

n=1
1
n
2
isap-serieswithp = 2> 1, soit converges. BecauseL exists, bytheLimit ComparisonTest wecan

n=1
_
1cos
1
n
_
alsoconverges.
n=1
(12
1/n
) Hint: Comparewiththeharmonicseries.
In Exercises 4978, determine convergence or divergence using any method covered so far.
49.

n=4
1
n
2
9
solution ApplytheLimit ComparisonTest witha
n
=
1
n
2
9
andb
n
=
1
n
2
:
L = lim
n
a
n
b
n
= lim
n
1
n
2
9
1
n
2
= lim
n
n
2
n
2
9
= 1.
Sincethep-series

n=1
1
n
2
converges, theseries

n=4
1
n
2
alsoconverges. BecauseL exists, bytheLimit ComparisonTest
wecanconcludethat theseries

n=4
1
n
2
9
converges.
n=1
cos
2
n
n
2
51.

n=1
n
4n +9
n
=
n
4n +9
andb
n
=
1
n
:
L = lim
n
a
n
b
n
= lim
n
n
4n+9
1
n
= lim
n
n
4n +9
=
1
4
.
Theseries

n=1
1
n
isadivergentp-series. BecauseL > 0, bytheLimitComparisonTestwecanconcludethattheseries
n=1
n
4n +9
alsodiverges.
n=1
n cosn
n
3
53.

n=1
n
2
n
n
5
+n
solution First rewritea
n
=
n
2
n
n
5
+n
=
n (n 1)
n
_
n
4
+1
_ =
n 1
n
4
+1
andobserve
n 1
n
4
+1
<
n
n
4
=
1
n
3
for n 1. The series

n=1
1
n
3
is a convergent p-series, so by the Comparison Test we can conclude that the series
n=1
n
2
n
n
5
+n
alsoconverges.
n=1
1
n
2
+sinn
55.

n=5
(4/5)
n
solution
n=5
_
4
5
_
n
=
n=5
_
5
4
_
n
whichisageometricseriesstartingat n = 5withratior =
5
4
> 1. Thustheseriesdiverges.
n=1
1
3
n
2
57.

n=2
1
n
3/2
lnn
solution For n 3, lnn > 1, son
3/2
lnn > n
3/2
and
1
n
3/2
lnn
<
1
n
3/2
.
Theseries

n=1
1
n
3/2
is aconvergent p-series, so theseries

n=3
1
n
3/2
also converges. By theComparisonTest wecan
thereforeconcludethat theseries

n=3
1
n
3/2
lnn
converges. Hence, theseries

n=2
1
n
3/2
lnn
alsoconverges.
n=2
(lnn)
12
n
9/8
59.

k=1
4
1/k
solution
lim
k
a
k
= lim
k
4
1/k
= 4
0
= 1= 0;
therefore, theseries

k=1
4
1/k
divergesbytheDivergenceTest.
n=1
4
n
5
n
2n
61.

n=2
1
(lnn)
4
solution Bythecomment precedingExercise31, wecanchooseN sothat for n N, wehavelnn < n
1/8
, sothat
(lnn)
4
< n
1/2
. Then
n=N
1
(lnn)
4
>
n=N
1
n
1/2
whichisadivergent p-series. Thustheseriesontheleft divergesaswell, andaddingback inthenitenumber of terms
for n < N doesnot affect theresult. Thus

n=2
1
(lnn)
4
diverges.
n=1
2
n
3
n
n
63.

n=1
1
n lnn n
solution For n 2, n lnn n n lnn; therefore,
1
n lnn n
1
n lnn
.
Now, letf (x) =
1
x lnx
. Forx 2, thisfunctioniscontinuous, positiveanddecreasing, sotheIntegral Testapplies. Using
thesubstitutionu = lnx, du =
1
x
dx, wend
_

2
dx
x lnx
= lim
R
_
R
2
dx
x lnx
= lim
R
_
lnR
ln2
du
u
= lim
R
(ln(lnR) ln(ln2)) = .

n=2
1
n lnn
alsodiverges. BytheComparisonTest wecanthereforeconcludethat
theseries

n=2
1
n lnn n
diverges.
n=1
1
n(lnn)
2
n
65.

n=1
1
n
n
solution For n 2, n
n
2
n
; therefore,
1
n
n

1
2
n
=
_
1
2
_
n
.
Theseries

n=1
_
1
2
_
n
is aconvergent geometric series, so

n=2
_
1
2
_
n

n=2
1
n
n
converges. Hence, theseries

n=1
1
n
n
converges.
n=1
n
2
4n
3/2
n
3
67.

n=1
1+(1)
n
n
solution Let
a
n
=
1+(1)
n
n
Then
a
n
=
_
0 n odd
2
2k
=
1
k
n = 2k even
Therefore, {a
n
} consistsof 0sintheoddplacesandtheharmonicseriesintheevenplaces, so
i=1
a
n
isjust thesumof
theharmonicseries, whichdiverges. Thus
i=1
a
n
divergesaswell.
n=1
2+(1)
n
n
3/2
69.

n=1
sin
1
n
n
= sin
1
n
andb
n
=
1
n
:
L = lim
n
sin
1
n
1
n
= lim
u0
sinu
u
= 1,
whereu =
1
n
. Theharmonic series diverges. BecauseL > 0, by theLimit ComparisonTest wecanconcludethat the
series

n=1
sin
1
n
alsodiverges.
n=1
sin(1/n)
n
71.

n=1
2n +1
4
n
solution For n 3, 2n +1< 2
n
, so
2n +1
4
n
<
2
n
4
n
=
_
1
2
_
n
.
Theseries

n=1
_
1
2
_
n
is aconvergent geometric series, so

n=3
_
1
2
_
n

n=3
2n +1
4
n
converges. Finally, theseries

n=1
2n +1
4
n
converges.
n=3
1
e
n
73.

n=4
lnn
n
2
3n
solution Bythecomment precedingExercise31, wecanchooseN 4sothat for n N, lnn < n
1/2
. Then
n=N
lnn
n
2
3n
n=N
n
1/2
n
2
3n
=
n=N
1
n
3/2
3n
1/2
To evaluateconvergenceof thelatter series, let a
n
=
1
n
3/2
3n
1/2
andb
n
=
1
n
3/2
, andapply theLimit Comparison
Test:
L = lim
n
a
n
b
n
= lim
n
1
n
3/2
3n
1/2
n
3/2
= lim
n
1
13n
1
= 0
Thus
a
n
converges if

b
n
does. But
b
n
is aconvergent p-series. Thus
a
n
converges and, by thecomparison
test, sodoestheoriginal series. Addingbackinthenitenumber of termsfor n < N doesnot affect convergence.
n=1
1
3
lnn
75.

n=2
1
n
1/2
lnn
solution Bythecomment precedingExercise31, wecanchooseN 2sothat for n N, lnn < n
1/4
. Then
n=N
1
n
1/2
lnn
>
n=N
1
n
3/4
whichis adivergent p-series. Thus theoriginal series diverges as well - as usual, addingback inthenitenumber of
termsfor n < N doesnot affect convergence.
n=1
1
n
3/2
ln
4
n
77.

n=1
4n
2
+15n
3n
4
5n
2
17
solution ApplytheLimit ComparisonTest with
a
n
=
4n
2
+15n
3n
4
5n
2
17
, b
n
=
4n
2
3n
4
=
4
3n
2
Wehave
L = lim
n
a
n
b
n
= lim
n
4n
2
+15n
3n
4
5n
2
17
3n
2
4
= lim
n
12n
4
+45n
3
12n
4
20n
2
68
= lim
n
12+45/n
1220/n
2
68/n
4
= 1
Now,
n=1
b
n
isap-serieswithp = 2> 1, soconverges. SinceL = 1, weseethat

n=1
4n
2
+15n
3n
4
5n
2
17
convergesas
well.
n=1
n
4
n
+5
n
79. For whicha does

n=2
1
n(lnn)
a
converge?
solution First consider thecasea > 0but a = 1. Let f (x) =
1
x(lnx)
a
. This functionis continuous, positiveand
decreasingfor x 2, sotheIntegral Test applies. Now,
_

2
dx
x(lnx)
a
= lim
R
_
R
2
dx
x(lnx)
a
= lim
R
_
lnR
ln2
du
u
a
=
1
1a
lim
R
_
1
(lnR)
a1
1
(ln2)
a1
_
.
Because
lim
R
1
(lnR)
a1
=
_
, 0< a < 1
0, a > 1
weconcludetheintegral divergeswhen0< a < 1andconvergeswhena > 1. Therefore
n=2
1
n(lnn)
a
convergesfor a > 1anddivergesfor 0< a < 1.
Next, consider thecasea = 1. Theseriesbecomes

n=2
1
n lnn
. Let f (x) =
1
x lnx
. For x 2, thisfunctioniscontinuous,
positiveanddecreasing, sotheIntegral Test applies. Usingthesubstitutionu = lnx, du =
1
x
dx, wend
_

2
dx
x lnx
= lim
R
_
R
2
dx
x lnx
= lim
R
_
lnR
ln2
du
u
= lim
R
(ln(lnR) ln(ln2)) = .
Theintegral diverges; hence, theseriesalsodiverges.
Finally, consider thecasea < 0. Let b = a > 0sotheseriesbecomes

n=2
(lnn)
b
n
. Sincelnn > 1for all n 3, it
followsthat
(lnn)
b
> 1 so
(lnn)
b
n
>
1
n
.
Theseries

n=3
1
n
diverges, so by theComparisonTest wecan concludethat

n=3
(lnn)
b
n
also diverges. Consequently,
n=2
(lnn)
b
n
diverges. Thus,
n=2
1
n(lnn)
a
divergesfor a < 0.
Tosummarize:
n=2
1
n(lnn)
a
convergesif a > 1anddivergesif a 1.
For whicha does

n=2
1
n
a
lnn
converge?
ApproximatingI nniteSums In Exercises 8183, let a
n
= f (n), where f (x) is a continuous, decreasing function such
that f (x) 0and
_
1
f (x) dx converges.
81. Showthat
_

1
f (x) dx
n=1
a
n
a
1
+
_

1
f (x) dx 3
solution Fromtheproof of theIntegral Test, weknowthat
a
2
+a
3
+a
4
+ +a
N

_
N
1
f (x) dx
_

1
f (x) dx;
that is,
S
N
a
1
_

1
f (x) dx or S
N
a
1
+
_

1
f (x) dx.
Alsofromtheproof of theIntegral test, weknowthat
_
N
1
f (x) dx a
1
+a
2
+a
3
+ +a
N1
= S
N
a
N
S
N
.
Thus,
_
N
1
f (x) dx S
N
a
1
+
_

1
f (x) dx.
Takingthelimit asN yieldsEq. (3), asdesired.
UsingEq. (3), showthat
5
n=1
1
n
1.2
6
Thisseriesconvergesslowly. Useacomputer algebrasystemtoverifythat S
N
< 5for N 43,128andS
43,129
5.00000021.
83. Let S =
n=1
a
n
. ArguingasinExercise81, showthat
M
n=1
a
n
+
_

M+1
f (x) dx S
M+1
n=1
a
n
+
_

M+1
f (x) dx 4
Concludethat
0 S
n=1
a
n
+
_

M+1
f (x) dx
a
M+1
5
Thisprovidesamethodfor approximatingS withanerror of at most a
M+1
.
solution Followingtheproof of theIntegral TestandtheargumentinExercise81, butstartingwithn = M +1rather
thann = 1, weobtain
_

M+1
f (x) dx
n=M+1
a
n
a
M+1
+
_

M+1
f (x) dx.
Adding
M
n=1
a
n
toeachpart of thisinequalityyields
M
n=1
a
n
+
_

M+1
f (x) dx
n=1
a
n
= S
M+1
n=1
a
n
+
_

M+1
f (x) dx.
Subtracting
M
n=1
a
n
+
_

M+1
f (x) dx fromeachpart of thislast inequalitythengivesus
0 S
n=1
a
n
+
_

M+1
f (x) dx
a
M+1
.
UseEq. (4) withM = 43,129toprovethat
5.5915810
n=1
1
n
1.2
5.5915839
85. ApplyEq. (4) withM = 40,000toshowthat
1.644934066
n=1
1
n
2
1.644934068
Isthisconsistent withEulersresult, accordingtowhichthisinniteserieshassum
2
/6?
solution UsingEq. (4) withf (x) =
1
x
2
, a
n
=
1
n
2
andM = 40,000, wend
S
40,000
+
_

40,001
dx
x
2

n=1
1
n
2
S
40,001
+
_

40,001
dx
x
2
.
Now,
S
40,000
= 1.6449090672;
S
40,001
= S
40,000
+
1
40,001
= 1.6449090678;
and
_

40,001
dx
x
2
= lim
R
_
R
40,001
dx
x
2
= lim
R
_
1
R
1
40,001
_
=
1
40,001
= 0.0000249994.
Thus,
1.6449090672+0.0000249994
n=1
1
n
2
1.6449090678+0.0000249994,
or
1.6449340665
n=1
1
n
2
1.6449340672.
Since
2
6
1.6449340668, our approximationisconsistent withEulersresult.
UsingaCAS andEq. (5), determinethevalueof

n=1
n
6
towithinanerror lessthan10
4
. Check that your
result isconsistent withthat of Euler, whoprovedthat thesumisequal to
6
/945.
87. UsingaCASandEq. (5), determinethevalueof

n=1
n
5
towithinanerror lessthan10
4
.
solution UsingEq. (5) withf (x) = x
5
anda
n
= n
5
, wehave
0
n=1
n
5
M+1
n=1
n
5
+
_

M+1
x
5
dx
(M +1)
5
.
To guaranteeanerror less than10
4
, weneed(M + 1)
5
10
4
. This yields M 10
4/5
1 5.3, so wechoose
M = 6. Now,
7
n=1
n
5
= 1.0368498887,
and
_

7
x
5
dx = lim
R
_
R
7
x
5
dx =
1
4
lim
R
_
R
4
7
4
_
=
1
4 7
4
= 0.0001041233.
Thus,
n=1
n
5
n=1
n
5
+
_

7
x
5
dx = 1.0368498887+0.0001041233= 1.0369540120.
Howfar canastack of identical books(of massm andunit length) extendwithout tippingover? Thestack will
not tipover if the(n +1)st bookisplacedat thebottomof thestackwithitsright edgelocatedat thecenter of mass
of therstn books(Figure5). Letc
n
bethecenter of massof therst n books, measuredalongthex-axis, wherewe
takethepositivex-axistotheleft of theoriginasinFigure6. Recall that if anobject of massm
1
hascenter of mass
at x
1
andasecondobject of m
2
hascenter of massx
2
, thenthecenter of massof thesystemhasx-coordinate
m
1
x
1
+m
2
x
2
m
1
+m
2
(a) Showthat if the(n +1)st bookisplacedwithitsright edgeat c
n
, thenitscenter of massislocatedat c
n
+
1
2
.
(b) Considertherstnbooksasasingleobjectof massnmwithcenterof massatc
n
andthe(n +1)stbookasasecond
object of massm. Showthat if the(n +1)st bookisplacedwithitsright edgeat c
n
, thenc
n+1
= c
n
+
1
2(n +1)
.
(c) Provethat lim
n
c
n
= . Thus, by using enough books, thestack can beextended as far as desired without
tippingover.
89. Thefollowingargument provesthedivergenceof theharmonicseriesS =
n=1
1/n without usingtheIntegral Test.
Let
S
1
= 1+
1
3
+
1
5
+ , S
2
=
1
2
+
1
4
+
1
6
+
Showthat if S converges, then
(a) S
1
andS
2
alsoconvergeandS = S
1
+S
2
.
(b) S
1
> S
2
andS
2
=
1
2
S.
Observethat (b) contradicts(a), andconcludethat S diverges.
solution Assumethroughout that S converges; wewill deriveacontradiction. Write
a
n
=
1
n
, b
n
=
1
2n 1
, c
n
=
1
2n
for then
th
termsintheseriesS, S
1
, andS
2
. Since2n 1 n for n 1, wehaveb
n
< a
n
. SinceS =
a
n
converges,
sodoes S
1
=

b
n
by theComparisonTest. Also, c
n
=
1
2
a
n
, soagainby theComparisonTest, theconvergenceof S
impliestheconvergenceof S
2
=
c
n
. Now, denetwosequences
b
n
=
_
b
(n+1)/2
n odd
0 n even
c
n
=
_
0 n odd
c
n/2
n even
Thatis, b
n
andc
n
looklikeb
n
andc
n
, buthavezerosinsertedinthemissingplacescomparedtoa
n
. Thena
n
= b
n
+c
n
;
alsoS
1
=
b
n
=
n
andS
2
=
c
n
=
n
. Finally, sinceS, S
1
, andS
2
all converge, wehave
S =
n=1
a
n
=
n=1
(b
n
+c
n
) =
n=1
b
n
+
n=1
c
n
=
n=1
b
n
+
n=1
c
n
= S
1
+S
2
Now, b
n
> c
n
for every n, so that S
1
> S
2
. Also, weshowedabovethat c
n
=
1
2
a
n
, so that 2S
2
= S. Puttingall this
together gives
S = S
1
+S
2
> S
2
+S
2
= 2S
2
= S
sothat S > S, acontradiction. ThusS must diverge.
Let S =
n=2
a
n
, wherea
n
= (ln(lnn))
lnn
.
(a) Show, bytakinglogarithms, that a
n
= n
ln(ln(lnn))
.
(b) Showthat ln(ln(lnn)) 2if n > C, whereC = e
e
e
2
.
(c) Showthat S converges.
91. KummersAccelerationMethod SupposewewishtoapproximateS =
n=1
1/n
2
. Thereisasimilar telescoping
serieswhosevaluecanbecomputedexactly(Example1inSection11.2):
n=1
1
n(n +1)
= 1
S E C T I ON 11.4 Absolute and Conditional Convergence 683
(a) Verifythat
S =
n=1
1
n(n +1)
+
n=1
_
1
n
2
1
n(n +1)
_
Thusfor M large,
S 1+
M
n=1
1
n
2
(n +1)
6
(b) Explainwhat hasbeengained. WhyisEq. (6) abetter approximationtoS thanis
M
n=1
1/n
2
?
(c) Compute
1000
n=1
1
n
2
, 1+
100
n=1
1
n
2
(n +1)
Whichisabetter approximationtoS, whoseexact valueis
2
/6?
solution
(a) Becausetheseries

n=1
1
n
2
and

n=1
1
n(n +1)
bothconverge,
n=1
1
n(n +1)
+
n=1
_
1
n
2
1
n(n +1)
_
=
n=1
1
n(n +1)
+
n=1
1
n
2
n=1
1
n(n +1)
=
n=1
1
n
2
= S.
Now,
1
n
2
1
n(n +1)
=
n +1
n
2
(n +1)
n
n
2
(n +1)
=
1
n
2
(n +1)
,
so, for M large,
S 1+
M
n=1
1
n
2
(n +1)
.
(b) Theseries
n=1
1
n
2
(n+1)
convergesmorerapidlythan

n=1
1
n
2
sincethedegreeof n inthedenominator islarger.
(c) Usingacomputer algebrasystem, wend
1000
n=1
1
n
2
= 1.6439345667 and 1+
100
n=1
1
n
2
(n +1)
= 1.6448848903.
Thesecondsumismoreaccuratebecauseit iscloser totheexact solution
2
6
1.6449340668.
TheseriesS =
k=1
k
3
hasbeencomputedtomorethan100milliondigits. Therst 30digitsare
S = 1.202056903159594285399738161511
ApproximateS usingtheAccelerationMethodof Exercise91withM = 100andauxiliaryseries
R =
n=1
(n(n +1)(n +2))
1
.
AccordingtoExercise46inSection11.2, R isatelescopingserieswiththesumR =
1
4
.
11.4 Absolute and Conditional Convergence
1. Giveanexampleof aseriessuchthat
a
n
convergesbut
|a
n
| diverges.
solution Theseries
(1)
n
3
n
convergesbytheLeibnizTest, but thepositiveseries
1
3
n
isadivergent p-series.
2. Whichof thefollowingstatementsisequivalent toTheorem1?
(a) If

n=0
|a
n
| diverges, then

n=0
a
n
alsodiverges.
(b) If

n=0
a
n
diverges, then

n=0
|a
n
| alsodiverges.
(c) If

n=0
a
n
converges, then

n=0
|a
n
| alsoconverges.
solution Thecorrect answer is (b): If

n=0
a
n
diverges, then

n=0
|a
n
| also diverges. Takea
n
= (1)
n 1
n
to seethat
statements(a)and(c)arenot trueingeneral.
3. Lathikaarguesthat

n=1
(1)
n
n isanalternatingseriesandthereforeconverges. IsLathikaright?
solution No. Although

n=1
(1)
n
n isanalternatingseries, thetermsa
n
=
n donot formadecreasingsequence
thattendstozero. Infact, a
n
=
n isanincreasingsequencethattendsto, so

n=1
(1)
n
n divergesbytheDivergence
Test.
4. Supposethata
n
ispositive, decreasing, andtendsto0, andletS =
n=1
(1)
n1
a
n
. Whatcanwesayabout|S S
100
|
if a
101
= 10
3
? IsS larger or smaller thanS
100
?
solution Fromthetext,weknowthat|S S
100
| < a
101
= 10
3
.Also,theLeibniztesttellsusthatS
2N
< S < S
2N+1
for anyN 1, sothat S
100
< S.
Exercises
1. Showthat
n=0
(1)
n
2
n
convergesabsolutely.
solution Thepositiveseries

n=0
1
2
n
is ageometric series withr =
1
2
. Thus, thepositiveseries converges, andthe
givenseriesconvergesabsolutely.
Showthat thefollowingseriesconvergesconditionally:
n=1
(1)
n1
1
n
2/3
=
1
1
2/3
1
2
2/3
+
1
3
2/3
1
4
2/3
+
In Exercises 310, determine whether the series converges absolutely, conditionally, or not at all.
3.

n=1
(1)
n1
n
1/3
solution Thesequencea
n
=
1
n
1/3
ispositive, decreasing, andtendstozero; hence, theseries

n=1
(1)
n1
n
1/3
converges
by the Leibniz Test. However, the positive series

n=1
1
n
1/3
is a divergent p-series, so the original series converges
conditionally.
n=1
(1)
n
n
4
n
3
+1
5.

n=0
(1)
n1
(1.1)
n
solution Thepositiveseries

n=0
_
1
1.1
_
n
isaconvergent geometric series; thus, theoriginal seriesconvergesabso-
lutely.
n=1
sin(
n
4
)
n
2
7.

n=2
(1)
n
n lnn
solution Let a
n
=
1
n lnn
. Thena
n
formsadecreasingsequence(notethat n andlnn arebothincreasingfunctionsof
n) that tendstozero; hence, theseries

n=2
(1)
n
n lnn
convergesbytheLeibnizTest. However, thepositiveseries

n=2
1
n lnn
diverges, sotheoriginal seriesconvergesconditionally.
n=1
(1)
n
1+
1
n
9.

n=2
cosn
(lnn)
2
solution Sincecosn alternatesbetween+1and1,
n=2
cosn
(lnn)
2
=
n=2
(1)
n
(lnn)
2
Thisisanalternatingserieswhosegeneral termdecreasestozero, soit converges. Theassociatedpositiveseries,
n=2
1
(lnn)
2
isadivergent series, sotheoriginal seriesconvergesconditionally.
n=1
cosn
2
n
11. Let S =
n=1
(1)
n+1
1
n
3
.
(a) CalculateS
n
for 1 n 10.
(b) UseEq. (2) toshowthat 0.9 S 0.902.
solution
(a)
S
1
= 1 S
6
= S
5
1
6
3
= 0.899782407
S
2
= 1
1
2
3
=
7
8
= 0.875 S
7
= S
6
+
1
7
3
= 0.902697859
S
3
= S
2
+
1
3
3
= 0.912037037 S
8
= S
7
1
8
3
= 0.900744734
S
4
= S
3
1
4
3
= 0.896412037 S
9
= S
8
+
1
9
3
= 0.902116476
S
5
= S
4
+
1
5
3
= 0.904412037 S
10
= S
9
1
10
3
= 0.901116476
(b) ByEq. (2),
|S
10
S| a
11
=
1
11
3
,
so
S
10
1
11
3
S S
10
+
1
11
3
,
or
0.900365161 S 0.901867791.
UseEq. (2) toapproximate
n=1
(1)
n+1
n!
tofour decimal places.
13. Approximate

n=1
(1)
n+1
n
4
tothreedecimal places.
solution Let S =
n=1
(1)
n+1
n
4
, sothat a
n
=
1
n
4
. ByEq. (2),
|S
N
S| a
N+1
=
1
(N +1)
4
.
Toguaranteeaccuracytothreedecimal places, wemust chooseN sothat
1
(N +1)
4
< 510
4
or N >
4
20001 5.7.
Thesmallest valuethat satisestherequiredinequalityisthenN = 6. Thus,
S S
6
= 1
1
2
4
+
1
3
4
1
4
4
+
1
5
4
1
6
4
= 0.946767824.
Let
S =
n=1
(1)
n1
n
n
2
+1
In Exercises 15 and 16, nd a value of N such that S
N
approximates the series with an error of at most 10
5
. If you have
a CAS, compute this value of S
N
.
15.

n=1
(1)
n+1
n(n +2)(n +3)
solution Let S =
n=1
(1)
n+1
n (n +2) (n +3)
, sothat a
n
=
1
n (n +2) (n +3)
. ByEq. (2),
|S
N
S| a
N+1
=
1
(N +1)(N +3)(N +4)
.
Wemust chooseN sothat
1
(N +1)(N +3)(N +4)
10
5
or (N +1)(N +3)(N +4) 10
5
.
For N = 43, theproduct ontheleft handsideis 95,128, whilefor N = 44theproduct is 101,520; hence, thesmallest
valueof N whichsatisestherequiredinequalityisN = 44. Thus,
S S
44
=
44
n=1
(1)
n+1
n(n +2)(n +3)
= 0.0656746.
n=1
(1)
n+1
lnn
n!
In Exercises 1732, determine convergence or divergence by any method.
17.

n=0
7
n
solution Thisisa(positive) geometricserieswithr =
1
7
< 1, soit converges.
n=1
1
n
7.5
19.

n=1
1
5
n
3
n
solution UsetheLimit ComparisonTest with
1
5
n
:
L = lim
n
1/(5
n
3
n
)
1/5
n
= lim
n
5
n
5
n
3
n
= lim
n
1
1(3/5)
n
= 1
But
n=1
1
5
n
isaconvergent geometricseries. SinceL = 1, theLimit ComparisonTest tellsusthat theoriginal series
convergesaswell.
n=2
n
n
2
n
21.

n=1
1
3n
4
+12n
solution UsetheLimit ComparisonTest with
1
3n
4
:
L = lim
n
(1/(3n
4
+12n)
1/3n
4
= lim
n
3n
4
3n
4
+12n
= lim
n
1
1+4n
3
= 1
But
n=1
1
3n
4
=
1
3
n=1
1
n
4
isaconvergentp-series. SinceL = 1, theLimitComparisonTesttellsusthattheoriginal
seriesconvergesaswell.
n=1
(1)
n
_
n
2
+1
23.

n=1
1
_
n
2
+1
solution ApplytheLimit ComparisonTest andcomparetheserieswiththedivergent harmonicseries:
L = lim
n
1
n
2
+1
1
n
= lim
n
n
_
n
2
+1
= 1.
BecauseL > 0, weconcludethat theseries

n=1
1
_
n
2
+1
diverges.
n=0
(1)
n
n
_
n
2
+1
25.

n=1
3
n
+(2)
n
5
n
solution Theseries
n=1
3
n
5
n
=
n=1
_
3
5
_
n
isaconvergent geometricseries, asistheseries
n=1
(1)
n
2
n
5
n
=
n=1
_
2
5
_
n
.
Hence,
n=1
3
n
+(1)
n
2
n
5
n
=
n=1
_
3
5
_
n
+
n=1
_
2
5
_
n
alsoconverges.
n=1
(1)
n+1
(2n +1)!
27.

n=1
(1)
n
n
2
e
n
3
/3
solution Consider theassociatedpositiveseries

n=1
n
2
e
n
3
/3
. This series canbeseento convergeby theIntegral
Test:
_

1
x
2
e
x
3
/3
dx = lim
R
_
R
1
x
2
e
x
3
/3
dx = lim
R
e
x
3
/3
R
1
= e
1/3
+ lim
R
e
R
3
/3
= e
1/3
.
Theintegral converges, sotheoriginal seriesconvergesabsolutely.
n=1
ne
n
3
/3
29.

n=2
(1)
n
n
1/2
(lnn)
2
solution Thisisanalternatingserieswitha
n
=
1
n
1/2
(lnn)
2
. Becausea
n
isadecreasingsequencewhichconverges
tozero, theseries

n=2
(1)
n
n
1/2
(lnn)
2
convergesbytheLeibnizTest. (Notethat theseriesconvergesonlyconditionally, not
absolutely; theassociatedpositiveseriesiseventuallygreater than
1
n
3/4
, whichisadivergent p-series).
n=2
1
n(lnn)
1/4
31.

n=1
lnn
n
1.05
solution ChooseN sothat for n N wehavelnn n
0.01
. Then
n=N
lnn
n
1.05

n=N
n
0.01
n
1.05
=
n=N
1
n
1.04
Thisisaconvergent p-series, sobytheComparisonTest, theoriginal seriesconvergesaswell.
n=2
1
(lnn)
2
33. Showthat
S =
1
2
1
2
+
1
3
1
3
+
1
4
1
4
+
convergesbycomputingthepartial sums. Doesit convergeabsolutely?
solution Thesequenceof partial sumsis
S
1
=
1
2
S
2
= S
1
1
2
= 0
S
3
= S
2
+
1
3
=
1
3
S
4
= S
3
1
3
= 0
and, ingeneral,
S
N
=
1
N
, for oddN
0, for evenN
Thus, lim
N
S
N
= 0, andtheseriesconvergesto0. Thepositiveseriesis
1
2
+
1
2
+
1
3
+
1
3
+
1
4
+
1
4
+ = 2

n=2
1
n
;
whichdiverges. Therefore, theoriginal seriesconvergesconditionally, not absolutely.
TheLeibnizTest cannot beappliedto
1
2
1
3
+
1
2
2
1
3
2
+
1
2
3
1
3
3
+
Whynot? Showthat it convergesbyanother method.
35. AssumptionsMatter Showbycounterexamplethat theLeibnizTest doesnot remaintrueif thesequence
a
n
tendstozerobut isnot assumednonincreasing. Hint: Consider
R =
1
2
1
4
+
1
3
1
8
+
1
4
1
16
+ +
_
1
n
1
2
n
_
+
solution Let
R =
1
2
1
4
+
1
3
1
8
+
1
4
1
16
+ +
_
1
n +1
1
2
n+1
_
+
Thisisanalternatingserieswith
a
n
=
1
k +1
, n = 2k 1
1
2
k+1
, n = 2k
Notethat a
n
0asn , but thesequence{a
n
} isnot decreasing. Wewill nowestablishthat R diverges.
For sakeof contradiction, supposethat R converges. Thegeometricseries
n=1
1
2
n+1
converges, sothesumof R andthisgeometricseriesmust alsoconverge; however,
R +
n=1
1
2
n+1
=
n=2
1
n
,
whichdivergesbecausetheharmonicseriesdiverges. Thus, theseriesR must diverge.
Determinewhether thefollowingseriesconvergesconditionally:
1
1
3
+
1
2
1
5
+
1
3
1
7
+
1
4
1
9
+
1
5
1
11
+
37. Provethat if

a
n
converges absolutely, then
a
2
n
also converges. Thengiveanexamplewhere
a
n
is only
conditionallyconvergent and
a
2
n
diverges.
solution Supposetheseries
a
n
convergesabsolutely. Because
|a
n
| converges, weknowthat
lim
n
|a
n
| = 0.
Therefore, thereexistsapositiveinteger N suchthat |a
n
| < 1for all n N. It thenfollowsthat for n N,
0 a
2
n
= |a
n
|
2
= |a
n
| |a
n
| < |a
n
| 1= |a
n
|.
BytheComparisonTest wecanthenconcludethat
a
2
n
alsoconverges.
Consider theseries

n=1
(1)
n
n
. This series converges by theLeibniz Test, but thecorrespondingpositiveseries is a
divergent p-series; that is,

n=1
(1)
n
n
isconditionally convergent. Now,

n=1
a
2
n
isthedivergent harmonic series

n=1
1
n
.
Thus,
a
2
n
neednot convergeif
a
n
isonlyconditionallyconvergent.
Provethefollowingvariantof theLeibnizTest: If {a
n
} isapositive, decreasingsequencewith lim
n
a
n
= 0, then
theseries
a
1
+a
2
2a
3
+a
4
+a
5
2a
6
+
converges. Hint: Showthat S
3N
isincreasingandboundedby a
1
+ a
2
, andcontinueasintheproof of theLeibniz
Test.
39. UseExercise38toshowthat thefollowingseriesconverges:
S =
1
ln2
+
1
ln3
2
ln4
+
1
ln5
+
1
ln6
2
ln7
+
solution Thegivenserieshasthestructureof thegenericseriesfromExercise38witha
n
=
1
ln(n+1)
. Becausea
n
is
apositive, decreasingsequencewith lim
n
a
n
= 0, wecanconcludefromExercise38that thegivenseriesconverges.
Provetheconditional convergenceof
R = 1+
1
2
+
1
3
3
4
+
1
5
+
1
6
+
1
7
3
8
+
41. Showthat thefollowingseriesdiverges:
S = 1+
1
2
+
1
3
2
4
+
1
5
+
1
6
+
1
7
2
8
+
Hint: Usetheresult of Exercise40towriteS asthesumof aconvergent seriesandadivergent series.
solution Let
R = 1+
1
2
+
1
3
3
4
+
1
5
+
1
6
+
1
7
3
8
+
and
S = 1+
1
2
+
1
3
2
4
+
1
5
+
1
6
+
1
7
2
8
+
For sakeof contradiction, supposetheseriesS converges. FromExercise40, weknowthattheseriesR converges. Thus,
theseriesS R must converge; however,
S R =
1
4
+
1
8
+
1
12
+ =
1
4
k=1
1
k
,
whichdivergesbecausetheharmonicseriesdiverges. Thus, theseriesS must diverge.
Provethat
n=1
(1)
n+1
(lnn)
a
n
convergesfor all exponentsa. Hint: Showthat f (x) = (lnx)
a
/x isdecreasingfor x sufcientlylarge.
43. Wesaythat{b
n
} isarearrangementof {a
n
} if {b
n
} hasthesametermsas{a
n
} butoccurringinadifferentorder. Show
that if {b
n
} isarearrangement of {a
n
} andS =
n=1
a
n
convergesabsolutely, thenT =
n=1
b
n
alsoconvergesabsolutely.
(Thisresult doesnot holdif S isonlyconditionallyconvergent.) Hint: Provethat thepartial sums
N
n=1
|b
n
| arebounded.
It canbeshownfurther that S = T .
solution SupposetheseriesS =
n=1
a
n
convergesabsolutelyanddenotethecorrespondingpositiveseriesby
S
+
=
n=1
|a
n
|.
Further, letT
N
=
N
n=1
|b
n
| denotetheNthpartial sumof theseries

n=1
|b
n
|. Because{b
n
} isarearrangementof {a
n
}, we
knowthat
0 T
N

n=1
|a
n
| = S
+
;
that is, thesequence{T
N
} isbounded. Moreover,
T
N+1
=
N+1
n=1
|b
n
| = T
N
+|b
N+1
| T
N
;
thatis, {T
N
} isincreasing. Itfollowsthat{T
N
} converges, sotheseries

n=1
|b
n
| converges, whichmeanstheseries

n=1
b
n
AssumptionsMatter In 1829, Lejeune Dirichlet pointed out that the great French mathematicianAugustin
LouisCauchy madeamistakeinapublishedpaper by improperly assumingtheLimit ComparisonTest tobevalid
for nonpositiveseries. HereareDirichletstwoseries:
(1)
n

(1)
n
_
1+
(1)
n
_
11.5 The Ratio and Root Tests
1. IntheRatioTest, is equal to lim
n
a
n+1
a
n
or lim
n
a
n
a
n+1
?
solution IntheRatioTest isthelimit lim
n
a
n+1
a
n
.
2. IstheRatioTest conclusivefor

n=1
1
2
n
? Isit conclusivefor

n=1
1
n
?
solution Thegeneral termof

n=1
1
2
n
isa
n
=
1
2
n
; thus,
a
n+1
a
n
=
1
2
n+1
2
n
1
=
1
2
,
and
= lim
n
a
n+1
a
n
=
1
2
< 1.
Consequently, theRatioTest guaranteesthat theseries

n=1
1
2
n
converges.
Thegeneral termof

n=1
1
n
isa
n
=
1
n
; thus,
a
n+1
a
n
=
1
n +1
n
1
=
n
n +1
,
and
= lim
n
a
n+1
a
n
= lim
n
n
n +1
= 1.
TheRatioTest isthereforeinconclusivefor theseries

n=1
1
n
.
3. CantheRatioTest beusedtoshowconvergenceif theseriesisonlyconditionallyconvergent?
solution No. TheRatioTest canonlyestablishabsoluteconvergenceanddivergence, not conditional convergence.
Exercises
In Exercises 120, apply the Ratio Test to determine convergence or divergence, or state that the Ratio Test is inconclusive.
1.

n=1
1
5
n
solution Witha
n
=
1
5
n
,
a
n+1
a
n
=
1
5
n+1
5
n
1
=
1
5
and = lim
n
a
n+1
a
n
=
1
5
< 1.
Therefore, theseries

n=1
1
5
n
convergesbytheRatioTest.
n=1
(1)
n1
n
5
n
3.

n=1
1
n
n
solution Witha
n
=
1
n
n
,
a
n+1
a
n
=
1
(n +1)
n+1
n
n
1
=
1
n +1
_
n
n +1
_
n
=
1
n +1
_
1+
1
n
_
n
,
S E C T I ON 11.5 The Ratio and Root Tests 691
and
= lim
n
a
n+1
a
n
= 0
1
e
= 0< 1.

n=1
1
n
n
n=0
3n +2
5n
3
+1
5.

n=1
n
n
2
+1
solution Witha
n
=
n
n
2
+1
,
a
n+1
a
n
=
n +1
(n +1)
2
+1
n
2
+1
n
=
n +1
n
n
2
+1
n
2
+2n +2
,
and
= lim
n
a
n+1
a
n
= 1 1= 1.
Therefore, for theseries

n=1
n
n
2
+1
, theRatioTest isinconclusive.
WecanshowthatthisseriesdivergesbyusingtheLimitComparisonTestandcomparingwiththedivergentharmonic
series.
n=1
2
n
n
7.

n=1
2
n
n
100
solution Witha
n
=
2
n
n
100
,
a
n+1
a
n
=
2
n+1
(n +1)
100
n
100
2
n
= 2
_
n
n +1
_
100
and = lim
n
a
n+1
a
n
= 2 1
100
= 2> 1.

n=1
2
n
n
100
divergesbytheRatioTest.
n=1
n
3
3
n
2
9.

n=1
10
n
2
n
2
solution Witha
n
=
10
n
2
n
2
,
a
n+1
a
n
=
10
n+1
2
(n+1)
2

2
n
2
10
n
= 10
1
2
2n+1
and = lim
n
a
n+1
a
n
= 10 0= 0< 1.

n=1
10
n
2
n
2
n=1
e
n
n!
11.

n=1
e
n
n
n
solution Witha
n
=
e
n
n
n
,
a
n+1
a
n
=
e
n+1
(n +1)
n+1
n
n
e
n
=
e
n +1
_
n
n +1
_
n
=
e
n +1
_
1+
1
n
_
n
,
and
= lim
n
a
n+1
a
n
= 0
1
e
= 0< 1.

n=1
e
n
n
n
n=1
n
40
n!
13.

n=0
n!
6
n
solution Witha
n
=
n!
6
n
,
a
n+1
a
n
=
(n +1)!
6
n+1

6
n
n!
=
n +1
6
and = lim
n
a
n+1
a
n
= > 1.

n=0
n!
6
n
divergesbytheRatioTest.
n=1
n!
n
9
15.

n=2
1
n lnn
solution Witha
n
=
1
n lnn
,
a
n+1
a
n
=
1
(n +1) ln(n +1)
n lnn
1
=
n
n +1
lnn
ln(n +1)
,
and
= lim
n
a
n+1
a
n
= 1 lim
n
lnn
ln(n +1)
.
Now,
lim
n
lnn
ln(n +1)
= lim
x
lnx
ln(x +1)
= lim
x
1/(x +1)
1/x
= lim
x
x
x +1
= 1.
Thus, = 1, andtheRatioTest isinconclusivefor theseries

n=2
1
n lnn
.
UsingtheIntegral Test, wecanshowthat theseries

n=2
1
n lnn
diverges.
n=1
1
(2n)!
17.

n=1
n
2
(2n +1)!
solution Witha
n
=
n
2
(2n+1)!
,
a
n+1
a
n
=
(n +1)
2
(2n +3)!

(2n +1)!
n
2
=
_
n +1
n
_
2
1
(2n +3)(2n +2)
,
and
= lim
n
a
n+1
a
n
= 1
2
0= 0< 1.

n=1
n
2
(2n +1)!
n=1
(n!)
3
(3n)!
19.

n=2
1
2
n
+1
solution Witha
n
=
1
2
n
+1
,
a
n+1
a
n
=
1
2
n+1
+1
2
n
+1
1
=
1+2
n
2+2
n
and
= lim
n
a
n+1
a
n
=
1
2
< 1

n=2
1
2
n
+1
n=2
1
lnn
21. Showthat

n=1
n
k
3
n
convergesfor all exponentsk.
solution Witha
n
= n
k
3
n
,
a
n+1
a
n
=
(n +1)
k
3
(n+1)
n
k
3
n
=
1
3
_
1+
1
n
_
k
,
and, for all k,
= lim
n
a
n+1
a
n
=
1
3
1=
1
3
< 1.

n=1
n
k
3
n
convergesfor all exponentsk bytheRatioTest.
Showthat

n=1
n
2
x
n
convergesif |x| < 1.
23. Showthat

n=1
2
n
x
n
convergesif |x| <
1
2
.
solution Witha
n
= 2
n
x
n
,
a
n+1
a
n
=
2
n+1
|x|
n+1
2
n
|x|
n
= 2|x| and = lim
n
a
n+1
a
n
= 2|x|.
Therefore, < 1andtheseries

n=1
2
n
x
n
convergesbytheRatioTest provided|x| <
1
2
.
Showthat

n=1
r
n
n!
convergesfor all r.
25. Showthat

n=1
r
n
n
convergesif |r| < 1.
solution Witha
n
=
r
n
n
,
a
n+1
a
n
=
|r|
n+1
n +1

n
|r|
n
= |r|
n
n +1
and = lim
n
a
n+1
a
n
= 1 |r| = |r|.
Therefore, bytheRatioTest, theseries

n=1
r
n
n
convergesprovided|r| < 1.
Isthereanyvalueof k suchthat

n=1
2
n
n
k
converges?
27. Showthat

n=1
n!
n
n
converges. Hint: Use lim
n
_
1+
1
n
_
n
= e.
solution Witha
n
=
n!
n
n
,
a
n+1
a
n
=
(n +1)!
(n +1)
n+1
n
n
n!
=
_
n
n +1
_
n
=
_
1+
1
n
_
n
,
and
= lim
n
a
n+1
a
n
=
1
e
< 1.

n=1
n!
n
n
In Exercises 2833, assume that |a
n+1
/a
n
| converges to =
1
3
. What can you say about the convergence of the given
series?
n=1
na
n
29.

n=1
n
3
a
n
solution Let b
n
= n
3
a
n
. Then
= lim
n
b
n+1
b
n
= lim
n
_
n +1
n
_
3
a
n+1
a
n
= 1
3
1
3
=
1
3
< 1.

n=1
n
3
a
n
n=1
2
n
a
n
31.

n=1
3
n
a
n
solution Let b
n
= 3
n
a
n
. Then
= lim
n
b
n+1
b
n
= lim
n
3
n+1
3
n
a
n+1
a
n
= 3
1
3
= 1.
Therefore, theRatioTest isinconclusivefor theseries

n=1
3
n
a
n
.
n=1
4
n
a
n
33.

n=1
a
2
n
solution Let b
n
= a
2
n
. Then
= lim
n
b
n+1
b
n
= lim
n
a
n+1
a
n
2
=
_
1
3
_
2
=
1
9
< 1.

n=1
a
2
n
Assumethat
a
n+1
/a
n
convergesto = 4. Does
n=1
a
1
n
converge(assumethat a
n
= 0for all n)? 35. IstheRatioTest conclusivefor thep-series

n=1
1
n
p
?
solution Witha
n
=
1
n
p
,
a
n+1
a
n
=
1
(n +1)
p

n
p
1
=
_
n
n +1
_
p
and = lim
n
a
n+1
a
n
= 1
p
= 1.
Therefore, theRatioTest isinconclusivefor thep-series

n=1
1
n
p
.
In Exercises 3641, use the Root Test to determine convergence or divergence (or state that the test is inconclusive).
n=0
1
10
n
37.

n=1
1
n
n
solution Witha
n
=
1
n
n
,
n
a
n
=
n
_
1
n
n
=
1
n
and lim
n
n
a
n
= 0< 1.

n=1
1
n
n
convergesbytheRoot Test.
k=0
_
k
k +10
_
k
39.

k=0
_
k
3k +1
_
k
solution Witha
k
=
_
k
3k+1
_
k
,
k
a
k
=
k
_
_
k
3k +1
_
k
=
k
3k +1
and lim
k
k
a
k
=
1
3
< 1.

k=0
_
k
3k +1
_
k
n=1
_
1+
1
n
_
n
41.

n=4
_
1+
1
n
_
n
2
solution Witha
k
=
_
1+
1
n
_
n
2
,
n
a
n
=
n
_
_
1+
1
n
_
n
2
=
_
1+
1
n
_
n
and lim
n
n
a
n
= e
1
< 1.

n=4
_
1+
1
n
_
n
2
Provethat

n=1
2
n
2
n!
diverges. Hint: Use2
n
2
= (2
n
)
n
andn! n
n
.
In Exercises 4356, determine convergence or divergence using any method covered in the text so far.
43.

n=1
2
n
+4
n
7
n
solution Becausetheseries
n=1
2
n
7
n
=
n=1
_
2
7
_
n
and

n=1
4
n
7
n
=
n=1
_
4
7
_
n
arebothconvergent geometricseries, it followsthat
n=1
2
n
+4
n
7
n
=
n=1
_
2
7
_
n
+
n=1
_
4
7
_
n
alsoconverges.
n=1
n
3
n!
45.

n=1
n
3
5
n
solution Thepresenceof theexponential termsuggestsapplyingtheRatioTest. Witha
n
=
n
3
5
n
,
a
n+1
a
n
=
(n +1)
3
5
n+1

5
n
n
3
=
1
5
_
1+
1
n
_
3
and = lim
n
a
n+1
a
n
=
1
5
1
3
=
1
5
< 1.

n=1
n
3
5
n
n=2
1
n(lnn)
3
47.

n=2
1
_
n
3
n
2
solution Thisseriesissimilar toap-series; because
1
_
n
3
n
2

1
n
3
=
1
n
3/2
for largen, wewill applytheLimit ComparisonTest comparingwiththep-serieswithp =
3
2
. Now,
L = lim
n
1
n
3
n
2
1
n
3/2
= lim
n
_
n
3
n
3
n
2
= 1.
Thep-serieswithp =
3
2
convergesandL exists; therefore, theseries

n=2
1
_
n
3
n
2
alsoconverges.
n=1
n
2
+4n
3n
4
+9
49.

n=1
n
0.8
solution
n=1
n
0.8
=
n=1
1
n
0.8
sothat thisisadivergent p-series.
n=1
(0.8)
n
n
0.8
51.

n=1
4
2n+1
solution Observe
n=1
4
2n+1
=
n=1
4 (4
2
)
n
=
n=1
4
_
1
16
_
n
1
16
; therefore, thisseriesconverges.
n=1
(1)
n1
n
53.

n=1
sin
1
n
2
solution Here, wewill applytheLimit ComparisonTest, comparingwiththep-serieswithp = 2. Now,
L = lim
n
sin
1
n
2
1
n
2
= lim
u0
sinu
u
= 1,
whereu =
1
n
2
. Thep-serieswithp = 2convergesandL exists; therefore, theseries

n=1
sin
1
n
2
alsoconverges.
n=1
(1)
n
cos
1
n
55.

n=1
(2)
n
n
solution Because
lim
n
2
n
n
= lim
x
2
x
x
= lim
x
2
x
ln2
1
2
x
= lim
x
2
x+1
x ln2= = 0,
thegeneral termintheseries

n=1
(2)
n
n
doesnottendtowardzero; therefore, theseriesdivergesbytheDivergenceTest.
n=1
_
n
n +12
_
n
57. Proof of theRootTest Let S =
n=0
a
n
beapositiveseries, andassumethat L = lim
n
n
a
n
exists.
(a) Showthat S convergesif L < 1. Hint: ChooseR withL < R < 1andshowthat a
n
R
n
for n sufciently large.
Thencomparewiththegeometricseries
R
n
.
(b) Showthat S divergesif L > 1.
solution Suppose lim
n
n
a
n
= L exists.
(a) If L < 1, let =
1L
2
. Bythedenitionof alimit, thereisapositiveinteger N suchthat

n
a
n
L
for n N. Fromthis, weconcludethat
0
n
a
n
L +
for n N. Now, let R = L +. Then
R = L +
1L
2
=
L +1
2
<
1+1
2
= 1,
and
0
n
a
n
R or 0 a
n
R
n
for n N. Because0 R < 1, theseries

n=N
R
n
isaconvergent geometric series, sotheseries

n=N
a
n
convergesby
theComparisonTest. Therefore, theseries

n=0
a
n
alsoconverges.
(b) If L > 1, let =
L 1
2
. Bythedenitionof alimit, thereisapositiveinteger N suchthat

n
a
n
L
for n N. Fromthis, weconcludethat
L
n
a
n
for n N. Now, let R = L . Then
R = L
L 1
2
=
L +1
2
>
1+1
2
= 1,
S E C T I ON 11.6 Power Series 697
and
R
n
a
n
or R
n
a
n
for n N. BecauseR > 1, theseries

n=N
R
n
is adivergent geometric series, so theseries

n=N
a
n
diverges by the
ComparisonTest. Therefore, theseries

n=0
a
n
alsodiverges.
Showthat theRatioTest doesnot apply, but verifyconvergenceusingtheComparisonTest for theseries
1
2
+
1
3
2
+
1
2
3
+
1
3
4
+
1
2
5
+
59. Let S =
n=1
c
n
n!
n
n
, wherec isaconstant.
(a) Provethat S convergesabsolutelyif |c| < e anddivergesif |c| > e.
(b) It isknownthat lim
n
e
n
n!
n
n+1/2
=
2. Verifythisnumerically.
(c) UsetheLimit ComparisonTest toprovethat S divergesfor c = e.
solution
(a) Witha
n
=
c
n
n!
n
n
,
a
n+1
a
n
=
|c|
n+1
(n +1)!
(n +1)
n+1

n
n
|c|
n
n!
= |c|
_
n
n +1
_
n
= |c|
_
1+
1
n
_
n
,
and
= lim
n
a
n+1
a
n
= |c|e
1
.
Thus, by theRatio Test, theseries

n=1
c
n
n!
n
n
converges when|c|e
1
< 1, or when|c| < e. Theseries diverges when
|c| > e.
(b) Thetablebelowliststhevalueof
e
n
n!
n
n+1/2
forseveral increasingvaluesof n. Since
2 = 2.506628275, thenumerical
evidenceveriesthat
lim
n
e
n
n!
n
n+1/2
=
2.
n 100 1000 10000 100000
e
n
n!
n
n+1/2
2.508717995 2.506837169 2.506649163 2.506630363
(c) Withc = e, theseriesS becomes

n=1
e
n
n!
n
n
. Usingtheresult frompart (b),
L = lim
n
e
n
n!
n
n
n
= lim
n
e
n
n!
n
n+1/2
=
2.
Becausetheseries

n=1
n divergesbytheDivergenceTestandL > 0, weconcludethat

n=1
e
n
n!
n
n
divergesbytheLimit
ComparisonTest.
11.6 Power Series
1. Supposethat
a
n
x
n
convergesfor x = 5. Must it alsoconvergefor x = 4?What about x = 3?
solution The power series
a
n
x
n
is centered at x = 0. Because the series converges for x = 5, the radius of
convergencemust beat least 5 and theseries converges absolutely at least for theinterval |x| < 5. Both x = 4 and
x = 3areinsidethisinterval, sotheseriesconvergesfor x = 4andfor x = 3.
2. Supposethat
a
n
(x 6)
n
convergesfor x = 10. At whichof thepoints(a)(d) must it alsoconverge?
(a) x = 8 (b) x = 11 (c) x = 3 (d) x = 0
solution The given power series is centered at x = 6. Because the series converges for x = 10, the radius of
convergencemust beat least |10 6| = 4andtheseries converges absolutely at least for theinterval |x 6| < 4, or
2< x < 10.
(a) x = 8isinsidetheinterval 2< x < 10, sotheseriesconvergesfor x = 8.
(b) x = 11isnot insidetheinterval 2< x < 10, sotheseriesmayor maynot convergefor x = 11.
(c) x = 3isinsidetheinterval 2< x < 10, sotheseriesconvergesfor x = 2.
(d) x = 0isnot insidetheinterval 2< x < 10, sotheseriesmayor maynot convergefor x = 0.
3. What istheradiusof convergenceof F(3x) if F(x) isapower serieswithradiusof convergenceR = 12?
solution If thepower series F(x) has radius of convergenceR = 12, then thepower series F(3x) has radius of
convergenceR =
12
3
= 4.
4. Thepower seriesF(x) =
n=1
nx
n
hasradiusof convergenceR = 1. What isthepower seriesexpansionof F
(x)
andwhat isitsradiusof convergence?
solution We obtain the power series expansion for F
(x) by differentiating the power series expansion for F(x)

term-by-term. Thus,
F
(x) =
n=1
n
2
x
n1
.
Theradius of convergencefor this series is R = 1, thesameas theradius of convergencefor theseries expansionfor
F(x).
Exercises
1. UsetheRatioTest todeterminetheradiusof convergenceR of

n=0
x
n
2
n
. Doesit convergeat theendpointsx = R?
solution Witha
n
=
x
n
2
n
,
a
n+1
a
n
=
|x|
n+1
2
n+1

2
n
|x|
n
=
|x|
2
and = lim
n
a
n+1
a
n
=
|x|
2
.
By theRatioTest, theseries converges when =
|x|
2
< 1, or |x| < 2, anddiverges when =
|x|
2
> 1, or |x| > 2.
Theradiusof convergenceisthereforeR = 2. For x = 2, theleft endpoint, theseriesbecomes
n=0
(1)
n
, whichis
divergent. For x = 2, theright endpoint, theseriesbecomes
n=0
1, whichisalsodivergent. Thustheseriesdivergesat
bothendpoints.
UsetheRatioTesttoshowthat

n=1
x
n
n2
n
hasradiusof convergenceR = 2. Thendeterminewhetheritconverges
at theendpointsR = 2.
3. Showthatthepowerseries(a)(c)havethesameradiusof convergence.Thenshowthat(a)divergesatbothendpoints,
(b) convergesat oneendpoint but divergesat theother, and(c) convergesat bothendpoints.
(a)

n=1
x
n
3
n
(b)

n=1
x
n
n3
n
(c)

n=1
x
n
n
2
3
n
solution
(a) Witha
n
=
x
n
3
n
,
= lim
n
a
n+1
a
n
= lim
n
x
n+1
3
n+1

3
n
x
n
= lim
n
x
3
x
3
Then < 1if |x| < 3, sothat theradiusof convergenceisR = 3. For theendpoint x = 3, theseriesbecomes
n=1
3
n
3
n
=
n=1
1,
whichdivergesbytheDivergenceTest. For theendpoint x = 3, theseriesbecomes
n=1
(3)
n
3
n
=
n=1
(1)
n
,
whichalsodivergesbytheDivergenceTest.
(b) Witha
n
=
x
n
n3
n
,
= lim
n
a
n+1
a
n
= lim
n
x
n+1
(n +1)3
n+1
n3
n
x
n
= lim
n
x
3
_
n
n +1
_
x
3
.
Then < 1when|x| < 3, sothat theradiusof convergenceisR = 3. For theendpoint x = 3, theseriesbecomes
n=1
3
n
n3
n
=
n=1
1
n
,
whichisthedivergent harmonicseries. For theendpoint x = 3, theseriesbecomes
n=1
(3)
n
n3
n
=
n=1
(1)
n
n
,
whichconvergesbytheLeibnizTest.
(c) Witha
n
=
x
n
n
2
3
n
,
= lim
n
a
n+1
a
n
= lim
n
x
n+1
(n +1)
2
3
n+1
n
2
3
n
x
n
= lim
n
x
3
_
n
n +1
_
2
x
3
Then < 1when|x| < 3, sothat theradiusof convergenceisR = 3. For theendpoint x = 3, theseriesbecomes
n=1
3
n
n
2
3
n
=
n=1
1
n
2
,
whichisaconvergent p-series. For theendpoint x = 3, theseriesbecomes
n=1
(3)
n
n
2
3
n
=
n=1
(1)
n
n
2
,
whichconvergesbytheLeibnizTest.
Repeat Exercise3for thefollowingseries:
(a)

n=1
(x 5)
n
9
n
(b)

n=1
(x 5)
n
n9
n
(c)

n=1
(x 5)
n
n
2
9
n
5. Showthat

n=0
n
n
x
n
divergesfor all x = 0.
solution Witha
n
= n
n
x
n
, andassumingx = 0,
= lim
n
a
n+1
a
n
= lim
n
(n +1)
n+1
x
n+1
n
n
x
n
= lim
n
x
_
1+
1
n
_
n
(n +1)
=
< 1only if x = 0, sothat theradiusof convergenceisthereforeR = 0. Inother words, thepower seriesconverges
onlyfor x = 0.
For whichvaluesof x does

n=0
n!x
n
converge?
7. UsetheRatioTest toshowthat

n=0
x
2n
3
n
hasradiusof convergenceR =
3.
solution Witha
n
=
x
2n
3
n
,
= lim
n
a
n+1
a
n
= lim
n
x
2(n+1)
3
n+1

3
n
x
2n
= lim
n
x
2
3
x
2
3
Then < 1when|x

2
| < 3, or x =
3, sotheradiusof convergenceisR =
3.
Showthat

n=0
x
3n+1
64
n
hasradiusof convergenceR = 4.
In Exercises 934, nd the interval of convergence.
9.

n=0
nx
n
solution Witha
n
= nx
n
,
= lim
n
a
n+1
a
n
= lim
n
(n +1)x
n+1
nx
n
= lim
n
x
n +1
n
= |x|
Then < 1when|x| < 1, sothattheradiusof convergenceisR = 1, andtheseriesconvergesabsolutelyontheinterval
|x| < 1, or 1 < x < 1. For theendpoint x = 1, theseries becomes

n=0
n, whichdiverges by theDivergenceTest.
For theendpoint x = 1, theseriesbecomes

n=1
(1)
n
n, whichalsodivergesby theDivergenceTest. Thus, theseries
n=0
nx
n
convergesfor 1< x < 1anddivergeselsewhere.
n=1
2
n
n
x
n
11.

n=1
(1)
n
x
2n+1
2
n
n
solution Witha
n
= (1)
n
x
2n+1
2
n
n
,
= lim
n
x
2(n+1)+1
2
n+1
(n +1)
2
n
n
x
2n+1
= lim
n
x
2
2

n
n +1
x
2
2
Then < 1when|x| <
2, sotheradiusof convergenceisR =
2, andtheseriesconvergesabsolutelyontheinterval
2< x <
2. For theendpoint x =
2, theseriesbecomes

n=1
(1)
n
2
n
=
n=1
(1)
n+1
2
n
, whichconverges
bytheLeibniztest. For theendpointx =
2, theseriesbecomes

n=1
(1)
n
2
n
whichalsoconvergesbytheLeibniztest.
Thustheseries

n=1
(1)
n
x
2n+1
2
n
n
convergesfor
2 x
2anddivergeselsewhere.
n=0
(1)
n
n
4
n
x
2n
13.

n=4
x
n
n
5
solution Witha
n
=
x
n
n
5
,
= lim
n
a
n+1
a
n
= lim
n
x
n+1
(n +1)
5
n
5
x
n
= lim
n
x
_
n
n +1
_
5
= |x|
Then < 1when|x| < 1, so theradius of convergenceis R = 1, andtheseries converges absolutely ontheinterval
|x| < 1, or 1 < x < 1. For theendpoint x = 1, theseries becomes

n=1
1
n
5
, whichis aconvergent p-series. For the
endpointx = 1, theseriesbecomes

n=1
(1)
n
n
5
, whichconvergesbytheLeibnizTest.Thus, theseries

n=4
x
n
n
5
converges
for 1 x 1anddivergeselsewhere.
n=8
n
7
x
n
15.

n=0
x
n
(n!)
2
solution Witha
n
=
x
n
(n!)
2
,
= lim
n
a
n+1
a
n
= lim
n
x
n+1
((n +1)!)
2
(n!)
2
x
n
= lim
n
x
_
1
n +1
_
2
= 0
< 1for all x, sotheradiusof convergenceisR = , andtheseriesconvergesabsolutelyfor all x.
n=0
8
n
n!
x
n
17.

n=0
(2n)!
(n!)
3
x
n
solution Witha
n
=
(2n)!x
n
(n!)
3
, andassumingx = 0,
= lim
n
a
n+1
a
n
= lim
n
(2(n +1))!x
n+1
((n +1)!)
3

(n!)
3
(2n)!x
n
= lim
n
x
(2n +2)(2n +1)
(n +1)
3
= lim
n
x
4n
2
+6n +2
n
3
+3n
2
+3n +1
= lim
n
x
4n
1
+6n
1
+2n
3
1+3n
1
+3n
2
+n
3
= 0
Then < 1for all x, sotheradiusof convergenceisR = , andtheseriesconvergesabsolutelyfor all x.
n=0
4
n
(2n +1)!
x
2n1
19.

n=0
(1)
n
x
n
_
n
2
+1
solution Witha
n
=
(1)
n
x
n
n
2
+1
,
= lim
n
a
n+1
a
n
= lim
n
(1)
n+1
x
n+1
_
n
2
+2n +2
_
n
2
+1
(1)
n
x
n
= lim
n
x
_
n
2
+1
_
n
2
+2n +2
= lim
n
x
_
n
2
+1
n
2
+2n +2
= lim
n
x
_
1+1/n
2
1+2/n +2/n
2
= |x|
Then < 1when|x| < 1, so theradius of convergenceis R = 1, andtheseries converges absolutely ontheinterval
1 < x < 1. For theendpoint x = 1, theseriesbecomes

n=1
(1)
n
_
n
2
+1
, whichconvergesby theLeibniz Test. For the
endpoint x = 1, theseriesbecomes

n=1
1
_
n
2
+1
, whichdivergesby theLimit ComparisonTest comparingwiththe
divergent harmonicseries. Thus, theseries

n=0
(1)
n
x
n
_
n
2
+1
convergesfor 1< x 1anddivergeselsewhere.
n=0
x
n
n
4
+2
21.

n=15
x
2n+1
3n +1
solution Witha
n
=
x
2n+1
3n +1
,
= lim
n
a
n+1
a
n
= lim
n
x
2n+3
3n +4
3n +1
x
2n+1
= lim
n
x
2
3n +1
3n +4
= |x
2
|
Then < 1when|x
2
| < 1, sotheradiusof convergenceisR = 1, andtheseriesconvergesabsolutelyfor 1< x < 1.
For the endpoint x = 1, the series becomes

n=15
1
3n +1
, which diverges by the Limit Comparison Test comparing
withthedivergent harmonic series. For theendpoint x = 1, theseriesbecomes

n=15
1
3n +1
, whichalsodivergesby
theLimit ComparisonTest comparingwiththedivergent harmonic series. Thus, theseries

n=15
x
2n+1
3n +1
converges for
1< x < 1anddivergeselsewhere.
n=1
x
n
n 4lnn
23.

n=2
x
n
lnn
solution Witha
n
=
x
n
lnn
,
= lim
n
a
n+1
a
n
= lim
n
x
n+1
ln(n +1)
lnn
x
n
= lim
n
x
ln(n +1)
lnn
= lim
n
x
1/(n +1)
1/n
= lim
n
x
n
n +1
= |x|
usingLHpitalsrule. Then < 1when|x| < 1, sotheradiusof convergenceis1, andtheseriesconvergesabsolutely
ontheinterval |x| < 1, or 1 < x < 1. For theendpoint x = 1, theseries becomes

n=2
1
lnn
. Because
1
lnn
>
1
n
and
n=2
1
n
isthedivergent harmonicseries, theendpoint seriesdivergesbytheComparisonTest. For theendpoint x = 1,
theseriesbecomes

n=2
(1)
n
lnn
, whichconvergesbytheLeibnizTest. Thus, theseries

n=2
x
n
lnn
convergesfor1 x < 1
anddivergeselsewhere.
n=2
x
3n+2
lnn
25.

n=1
n(x 3)
n
solution Witha
n
= n(x 3)
n
,
= lim
n
a
n+1
a
n
= lim
n
(n +1)(x 3)
n+1
n(x 3)
n
= lim
n
(x 3)
n +1
n
= |x 3|
Then < 1when|x 3| < 1, so theradius of convergenceis 1, andtheseries converges absolutely ontheinterval
|x 3| < 1, or 2 < x < 4. For theendpoint x = 4, theseriesbecomes

n=1
n, whichdivergesby theDivergenceTest.
For theendpoint x = 2, theseries becomes

n=1
(1)
n
n, whichalso diverges by theDivergenceTest. Thus, theseries
n=1
n(x 3)
n
n=1
(5)
n
(x 3)
n
n
2
27.

n=1
(1)
n
n
5
(x 7)
n
solution Witha
n
= (1)
n
n
5
(x 7)
n
,
= lim
n
a
n+1
a
n
= lim
n
(1)
n+1
(n +1)
5
(x 7)
n+1
(1)
n
n
5
(x 7)
n
= lim
n
(x 7)
(n +1)
5
n
5
= lim
n
(x 7)
n
5
+. . .
n
5
= |x 7|
Then < 1when|x 7| < 1, so theradius of convergenceis 1, andtheseries converges absolutely ontheinterval
|x 7| < 1, or 6< x < 8. For theendpoint x = 6, theseriesbecomes

n=1
(1)
2n
n
5
=
n=1
n
5
, whichdivergesbythe
DivergenceTest. For theendpoint x = 8, theseriesbecomes

n=1
(1)
n
n
5
, whichalsodivergesbytheDivergenceTest.
Thus, theseries

n=1
(1)
n
n
5
(x 7)
n
n=0
27
n
(x 1)
3n+2
29.

n=1
2
n
3n
(x +3)
n
solution Witha
n
=
2
n
(x+3)
n
3n
,
= lim
n
a
n+1
a
n
= lim
n
2
n+1
(x +3)
n+1
3(n +1)
3n
2
n
(x +3)
n
= lim
n
2(x +3)
3n
3n +3
= lim
n
2(x +3)
1
1+1/n
= |2(x +3)|
Then < 1when|2(x + 3)| < 1, sowhen|x + 3| <
1
2
. Thustheradiusof convergenceis
1
2
, andtheseriesconverges
absolutely on theinterval |x + 3| <
1
2
, or
7
2
< x <
5
2
. For theendpoint x =
5
2
, theseries becomes

n=1
1
3n
,
whichdivergesbecauseit isamultipleof thedivergent harmonic series. For theendpoint x =
7
2
, theseriesbecomes
n=1
(1)
n
3n
, whichconverges by theLeibniz Test. Thus, theseries

n=1
2
n
3n
(x +3)
n
converges for
7
2
x <
5
2
and
divergeselsewhere.
n=0
(x 4)
n
n!
31.

n=0
(5)
n
n!
(x +10)
n
solution Witha
n
=
(5)
n
n!
(x +10)
n
,
= lim
n
a
n+1
a
n
= lim
n
(5)
n+1
(x +10)
n+1
(n +1)!

n!
(5)
n
(x +10)
n
= lim
n
5(x +10)
1
n
= 0
Thus < 1for all x, sotheradiusof convergenceisinnite, and

n=0
(5)
n
n!
(x +10)
n
convergesfor all x.
n=10
n! (x +5)
n
33.

n=12
e
n
(x 2)
n
solution Witha
n
= e
n
(x 2)
n
,
= lim
n
a
n+1
a
n
= lim
n
e
n+1
(x 2)
n+1
e
n
(x 2)
n
= lim
n
|e(x 2)| = |e(x 2)|
Thus < 1when|e(x 2)| < 1, sowhen|x 2| < e
1
. Thustheradiusof convergenceise
1
, andtheseriesconverges
absolutelyontheinterval |x 2| < e
1
, or 2e
1
< x < 2+e
1
. For theendpoint x = 2+e
1
, theseriesbecomes
n=1
1, whichdivergesbytheDivergenceTest. For theendpoint x = 2e
1
, theseriesbecomes

n=1
(1)
n
, whichalso
divergesbytheDivergenceTest. Thus, theseries

n=12
e
n
(x 2)
n
convergesfor 2 e
1
< x < 2+ e
1
anddiverges
elsewhere.
n=2
(x +4)
n
(n lnn)
2
In Exercises 3540, use Eq. (2) to expand the function in a power series with center c = 0and determine the interval of
convergence.
35. f (x) =
1
13x
solution Substituting3x for x inEq. (2), weobtain
1
13x
=
n=0
(3x)
n
=
n=0
3
n
x
n
.
Thisseriesisvalidfor |3x| < 1, or |x| <
1
3
.
f (x) =
1
1+3x
37. f (x) =
1
3x
solution First write
1
3x
=
1
3
1
1
x
3
.
Substituting
x
3
for x inEq. (2), weobtain
1
1
x
3
=
n=0
_
x
3
_
n
=
n=0
x
n
3
n
;
Thus,
1
3x
=
1
3
n=0
x
n
3
n
=
n=0
x
n
3
n+1
.
Thisseriesisvalidfor |x/3| < 1, or |x| < 3.
f (x) =
1
4+3x
39. f (x) =
1
1+x
2
solution Substitutingx
2
1
1+x
2
=
n=0
(x
2
)
n
=
n=0
(1)
n
x
2n
Thisseriesisvalidfor |x| < 1.
f (x) =
1
16+2x
3
41. Usetheequalities
1
1x
=
1
3(x 4)
=

1
3
1+
_
x4
3
_
toshowthat for |x 4| < 3,
1
1x
=
n=0
(1)
n+1
(x 4)
n
3
n+1
solution Substituting
x4
3
1
1+
_
x4
3
_ =
n=0
_
x 4
3
_
n
=
n=0
(1)
n
(x 4)
n
3
n
.
Thus,
1
1x
=
1
3
n=0
(1)
n
(x 4)
n
3
n
=
n=0
(1)
n+1
(x 4)
n
3
n+1
.
Thisseriesisvalidfor |
x4
3
| < 1, or |x 4| < 3.
Usethemethodof Exercise41toexpand1/(1x) inpower serieswithcentersc = 2andc = 2. Determine
theinterval of convergence.
43. Usethemethodof Exercise41toexpand1/(4x) inapower serieswithcenter c = 5. Determinetheinterval of
convergence.
solution First write
1
4x
=
1
1(x 5)
=
1
1+(x 5)
.
Substituting(x 5) for x inEq. (2), weobtain
1
1+(x 5)
=
n=0
((x 5))
n
=
n=0
(1)
n
(x 5)
n
.
Thus,
1
4x
=
n=0
(1)
n
(x 5)
n
=
n=0
(1)
n+1
(x 5)
n
.
Thisseriesisvalidfor | (x 5)| < 1, or |x 5| < 1.
Findapower seriesthat convergesonlyfor x in[2, 6).
45. Applyintegrationtotheexpansion
1
1+x
=
n=0
(1)
n
x
n
= 1x +x
2
x
3
+
toprovethat for 1< x < 1,
ln(1+x) =
n=1
(1)
n1
x
n
n
= x
x
2
2
+
x
3
3

x
4
4
+
solution Toobtaintherst expansion, substitutex for x inEq. (2):
1
1+x
=
n=0
(x)
n
=
n=0
(1)
n
x
n
.
Thisexpansionisvalidfor | x| < 1, or 1< x < 1.
Uponintegratingbothsidesof theaboveequation, wend
ln(1+x) =
_
dx
1+x
=
_

n=0
(1)
n
x
n
dx.
Integratingtheseriesterm-by-termthenyields
ln(1+x) = C +
n=0
(1)
n
x
n+1
n +1
.
Todeterminetheconstant C, set x = 0. Then0= ln(1+0) = C. Finally,
ln(1+x) =
n=0
(1)
n
x
n+1
n +1
=
n=1
(1)
n1
x
n
n
.
Usetheresult of Exercise45toprovethat
ln
3
2
=
1
2
1
2 2
2
+
1
3 2
3
1
4 2
4
+
Useyour knowledgeof alternatingseriestondanN suchthat thepartial sumS
N
approximatesln
3
2
towithinan
error of at most 10
3
. Conrmusingacalculator tocomputebothS
N
andln
3
2
.
47. Let F(x) = (x +1) ln(1+x) x.
(a) Applyintegrationtotheresult of Exercise45toprovethat for 1< x < 1,
F(x) =
n=1
(1)
n+1
x
n+1
n(n +1)
(b) Evaluateat x =
1
2
toprove
3
2
ln
3
2
1
2
=
1
1 2 2
2
1
2 3 2
3
+
1
3 4 2
4
1
4 5 2
5
+
(c) Useacalculator to verify that thepartial sumS
4
approximates theleft-handsidewithanerror no greater thanthe
terma
5
of theseries.
solution
(a) Notethat
_
ln(x +1) dx = (x +1) ln(x +1) x +C
Thenintegratingbothsidesof theresult of Exercise45gives
(x +1) ln(x +1) x =
_
ln(x +1) dx =
_
n=1
(1)
n1
x
n
n
dx
For 1 < x < 1, whichistheinterval of convergenceof theseriesinExercise45, therefore, wecanintegratetermby
termtoget
(x +1) ln(x +1) x =
n=1
(1)
n1
n
_
x
n
dx =
n=1
(1)
n1
n
x
n+1
n +1
+C =
n=1
(1)
n+1
x
n+1
n(n +1)
+C
(notingthat (1)
n1
= (1)
n+1
). TodetermineC, evaluatebothsidesat x = 0toget
0= ln10= 0+C
sothat C = 0andweget nally
(x +1) ln(x +1) x =
n=1
(1)
n+1
x
n+1
n(n +1)
(b) Evaluatingtheresult of part(a) at x =
1
2
gives
3
2
ln
3
2
1
2
=
n=1
(1)
n+1
1
n(n +1)2
n+1
=
1
1 2 2
2
1
2 3 2
3
+
1
3 4 2
4
1
4 5 2
5
+. . .
(c)
S
4
=
1
1 2 2
2
1
2 3 2
3
+
1
3 4 2
4
1
4 5 2
5
= 0.1078125
a
5
=
1
5 6 2
6
0.0005208
3
2
ln
3
2
1
2
0.10819766
and
S
4
3
2
ln
3
2
1
2
0.0003852< a
5
Provethat for |x| < 1,
_
dx
x
4
+1
= x
x
5
5
+
x
9
9

Usethersttwotermstoapproximate
_
1/2
0
dx/(x
4
+1) numerically. Usethefactthatyouhaveanalternatingseries
toshowthat theerror inthisapproximationisat most 0.00022.
49. Usetheresult of Example7toshowthat
F(x) =
x
2
1 2
x
4
3 4
+
x
6
5 6
x
8
7 8
+
isanantiderivativeof f (x) = tan
1
x satisfyingF(0) = 0. What istheradiusof convergenceof thispower series?
solution For 1< x < 1, whichistheinterval of convergencefor thepower seriesfor arctangent, wecanintegrate
term-by-term, sointegratethat power seriestoget
F(x) =
_
tan
1
x dx =
n=0
_
(1)
n
x
2n+1
2n +1
dx =
n=0
(1)
n
x
2n+2
(2n +1)(2n +2)
=
x
2
1 2
x
4
3 4
+
x
6
5 6
x
8
7 8
+ +C
If weassumeF(0) = 0, thenwehaveC = 0. Theradiusof convergenceof thispower seriesisthesameasthat of the
original power series, whichis1.
Verify that function F(x) = x tan
1
x
1
2
log(x
2
+ 1) is an antiderivative of f (x) = tan
1
x satisfying
F(0) = 0. Thenusetheresult of Exercise49withx =
1
3
toshowthat
1
2
ln
4
3
=
1
1 2(3)
1
3 4(3
2
)
+
1
5 6(3
3
)
1
7 8(3
4
)
+
Useacalculator tocomparethevalueof theleft-handsidewiththepartial sumS
4
of theseriesontheright.
51. Evaluate

n=1
n
2
n
. Hint: Usedifferentiationtoshowthat
(1x)
2
=
n=1
nx
n1
(for |x| < 1)
solution Differentiatebothsidesof Eq. (2) toobtain
1
(1x)
2
=
n=1
nx
n1
.
Settingx =
1
2
thenyields
n=1
n
2
n1
=
1
_
1
1
2
_
2
= 4.
Dividethisequationby2toobtain
n=1
n
2
n
= 2.
Usethepower seriesfor (1+x
2
)
1
anddifferentiationtoprovethat for |x| < 1,
2x
(x
2
+1)
2
=
n=1
(1)
n1
(2n)x
2n1
53. Showthat thefollowingseriesconvergesabsolutelyfor |x| < 1andcomputeitssum:
F(x) = 1x x
2
+x
3
x
4
x
5
+x
6
x
7
x
8
+
Hint: WriteF(x) asasumof threegeometricserieswithcommonratiox
3
.
solution Becausethecoefcientsinthepower seriesareall 1, wend
r = lim
n
a
n+1
a
n
= 1.
Theradiusof convergenceisthereforeR = r
1
= 1, andtheseriesconvergesabsolutelyfor |x| < 1.
By Exercise43 of Section 11.4, any rearrangement of theterms of an absolutely convergent series yields another
absolutelyconvergent serieswiththesamesumastheoriginal series. Followingthehint, wenowrearrangethetermsof
F(x) asthesumof threegeometricseries:
F(x) =
_
1+x
3
+x
6
+
_
_
x +x
4
+x
7
+
_
_
x
2
+x
5
+x
8
+
_
=
n=0
(x
3
)
n
n=0
x(x
3
)
n
n=0
x
2
(x
3
)
n
=
1
1x
3
x
1x
3
x
2
1x
3
=
1x x
2
1x
3
.
Showthat for |x| < 1,
1+2x
1+x +x
2
= 1+x 2x
2
+x
3
+x
4
2x
5
+x
6
+x
7
2x
8
+
55. Findall valuesof x suchthat

n=1
x
n
2
n!
converges.
solution Witha
n
=
x
n
2
n!
,
a
n+1
a
n
=
|x|
(n+1)
2
(n +1)!

n!
|x|
n
2
=
|x|
2n+1
n +1
.
if |x| 1, then
lim
n
|x|
2n+1
n +1
= 0,
andtheseriesconvergesabsolutely. Ontheother hand, if |x| > 1, then
lim
n
|x|
2n+1
n +1
= ,
andtheseriesdiverges. Thus,

n=1
x
n
2
n!
convergesfor 1 x 1anddivergeselsewhere.
Findall valuesof x suchthat thefollowingseriesconverges:
F(x) = 1+3x +x
2
+27x
3
+x
4
+243x
5
+
57. Findapower seriesP(x) =
n=0
a
n
x
n
satisfyingthedifferential equationy
= y withinitial conditiony(0) = 1.
ThenuseTheorem1of Section5.8toconcludethat P(x) = e
x
.
solution LetP(x) =
n=0
a
n
x
n
andnotethatP(0) = a
0
; thus, tosatisfytheinitial conditionP(0) = 1, wemusttake
a
0
= 1. Now,
P
(x) =
n=1
na
n
x
n1
,
so
P
(x) +P(x) =
n=1
na
n
x
n1
+
n=0
a
n
x
n
=
n=0
_
(n +1)a
n+1
+a
n
_
x
n
.
Inorder for thisseriestobeequal tozero, thecoefcient of x
n
must beequal tozerofor eachn; thus
(n +1)a
n+1
+a
n
= 0 or a
n+1
=
a
n
n +1
.
Startingfroma
0
= 1, wethencalculate
a
1
=
a
0
1
= 1;
a
2
=
a
1
2
=
1
2
;
a
3
=
a
2
3
=
1
6
=
1
3!
;
and, ingeneral,
a
n
= (1)
n
1
n!
.
Hence,
P(x) =
n=0
(1)
n
x
n
n!
.
Thesolutiontotheinitial valueproblemy
= y, y(0) = 1isy = e
x
. Becausethissolutionisunique, it followsthat
P(x) =
n=0
(1)
n
x
n
n!
= e
x
.
Let C(x) = 1
x
2
2!
+
x
4
4!

x
6
6!
+ .
(a) Showthat C(x) hasaninniteradiusof convergence.
(b) Provethat C(x) andf (x) = cosx arebothsolutionsof y
= y withinitial conditionsy(0) = 1, y
(0) = 0.
59. Usethepower seriesfor y = e
x
toshowthat
1
e
=
1
2!

1
3!
+
1
4!

Useyour knowledgeof alternatingseriestondanN suchthat thepartial sumS
N
approximatese
1
towithinanerror
of at most 10
3
. Conrmthisusingacalculator tocomputebothS
N
ande
1
.
solution Recall that theseriesfor e
x
is
n=0
x
n
n!
= 1+x +
x
2
2!
+
x
3
3!
+
x
4
4!
+ .
Settingx = 1yields
e
1
= 11+
1
2!

1
3!
+
1
4!
+ =
1
2!

1
3!
+
1
4!
+ .
This is an alternating series with a
n
=
1
(n+1)!
. Theerror in approximating e
1
with thepartial sumS
N
is therefore
boundedby
|S
N
e
1
| a
N+1
=
1
(N +2)!
.
Tomaketheerror at most 10
3
1
(N +2)!
10
3
or (N +2)! 1000.
For N = 4, (N + 2)! = 6! = 720 < 1000, but for N = 5, (N + 2)! = 7! = 5040; hence, N = 5isthesmallest value
that satisestheerror bound. Thecorrespondingapproximationis
S
5
=
1
2!

1
3!
+
1
4!

1
5!
+
1
6!
= 0.368055555
Now, e
1
= 0.367879441, so
|S
5
e
1
| = 1.76110
4
< 10
3
.
Let P(x) =
n=0
a
n
x
n
beapower seriessolutiontoy
= 2xy withinitial conditiony(0) = 1.

(a) Showthat theoddcoefcientsa
2k+1
areall zero.
(b) Provethat a
2k
= a
2k2
/k andusethisresult todeterminethecoefcientsa
2k
.
61. Findapower seriesP(x) satisfyingthedifferential equation
y
xy
+y = 0 9
withinitial conditiony(0) = 1, y
(0) = 0. What istheradiusof convergenceof thepower series?

solution Let P(x) =
n=0
a
n
x
n
. Then
P
(x) =
n=1
na
n
x
n1
and P
(x) =
n=2
n(n 1)a
n
x
n2
.
NotethatP(0) = a
0
andP
(0) = a
1
; inorder tosatisfytheinitial conditionsP(0) = 1, P
(0) = 0, wemusthavea
0
= 1
anda
1
= 0. Now,
P
(x) xP
(x) +P(x) =
n=2
n(n 1)a
n
x
n2
n=1
na
n
x
n
+
n=0
a
n
x
n
=
n=0
(n +2)(n +1)a
n+2
x
n
n=1
na
n
x
n
+
n=0
a
n
x
n
= 2a
2
+a
0
+
n=1
_
(n +2)(n +1)a
n+2
na
n
+a
n
_
x
n
.
Inorder for thisseriestobeequal tozero, thecoefcient of x
n
must beequal tozerofor eachn; thus, 2a
2
+a
0
= 0and
(n +2)(n +1)a
n+2
(n 1)a
n
= 0, or
a
2
=
1
2
a
0
and a
n+2
=
n 1
(n +2)(n +1)
a
n
.
Startingfroma
1
= 0, wecalculate
a
3
=
11
(3)(2)
a
1
= 0;
a
5
=
2
(5)(4)
a
3
= 0;
a
7
=
4
(7)(6)
a
5
= 0;
and, ingeneral, all of theoddcoefcientsarezero. Asfor theevencoefcients, wehavea
0
= 1, a
2
=
1
2
,
a
4
=
1
(4)(3)
a
2
=
1
4!
;
a
6
=
3
(6)(5)
a
4
=
3
6!
;
a
8
=
5
(8)(7)
a
6
=
15
8!
andsoon. Thus,
P(x) = 1
1
2
x
2
1
4!
x
4
3
6!
x
6
15
8!
x
8

Todeterminetheradiusof convergence, treat thisasaseriesinthevariablex
2
, andobservethat
r = lim
k
a
2k+2
a
2k
= lim
k
2k 1
(2k +2)(2k +1)
= 0.
Thus, theradiusof convergenceisR = r
1
= .
Findapower seriessatisfyingEq. (9) withinitial conditiony(0) = 0, y
(0) = 1.
63. Provethat
J
2
(x) =
k=0
(1)
k
2
2k+2
k! (k +3)!
x
2k+2
isasolutionof theBessel differential equationof order 2:
x
2
y
+xy
+(x
2
4)y = 0
solution Let J
2
(x) =
k=0
(1)
k
2
2k+2
k! (k +2)!
x
2k+2
. Then
J
2
(x) =
k=0
(1)
k
(k +1)
2
2k+1
k! (k +2)!
x
2k+1
J
2
(x) =
k=0
(1)
k
(k +1)(2k +1)
2
2k+1
k! (k +2)!
x
2k
and
x
2
J
2
(x) +xJ
2
(x) +(x
2
4)J
2
(x) =
k=0
(1)
k
(k +1)(2k +1)
2
2k+1
k! (k +2)!
x
2k+2
+
k=0
(1)
k
(k +1)
2
2k+1
k! (k +2)!
x
2k+2
k=0
(1)
k
2
2k+2
k! (k +2)!
x
2k+4
k=0
(1)
k
2
2k
k! (k +2)!
x
2k+2
=
k=0
(1)
k
k(k +2)
2
2k
k!(k +2)!
x
2k+2
+
k=1
(1)
k1
2
2k
(k 1)! (k +1)!
x
2k+2
=
k=1
(1)
k
2
2k
(k 1)!(k +1)!
x
2k+2
k=1
(1)
k
2
2k
(k 1)!(k +1)!
x
2k+2
= 0.
Why is it impossibleto expand f (x) = |x| as apower series that converges in an interval around x = 0?
ExplainusingTheorem2.
65. Supposethat thecoefcients of F(x) =
n=0
a
n
x
n
areperiodic; that is, for somewholenumber M > 0, wehave
a
M+n
= a
n
. Provethat F(x) convergesabsolutelyfor |x| < 1andthat
F(x) =
a
0
+a
1
x + +a
M1
x
M1
1x
M
Hint: Usethehint for Exercise53.
solution Supposethecoefcientsof F(x) areperiodic, witha
M+n
= a
n
for somewholenumber M andall n. The
F(x) canbewrittenasthesumof M geometricseries:
F(x) = a
0
_
1+x
M
+x
2M
+
_
+a
1
_
x +x
M+1
+x
2M+1
+
_
+
= a
2
_
x
2
+x
M+2
+x
2M+2
+
_
+ +a
M1
_
x
M1
+x
2M1
+x
3M1
+
_
=
a
0
1x
M
+
a
1
x
1x
M
+
a
2
x
2
1x
M
+ +
a
M1
x
M1
1x
M
=
a
0
+a
1
x +a
2
x
2
+ +a
M1
x
M1
1x
M
.
Aseachgeometricseriesconvergesabsolutelyfor |x| < 1, it followsthat F(x) alsoconvergesabsolutelyfor |x| < 1.
Continuityof Power Series Let F(x) =
n=0
a
n
x
n
beapower serieswithradiusof convergenceR > 0.
(a) Provetheinequality
|x
n
y
n
| n|x y|(|x|
n1
+|y|
n1
)
Hint: x
n
y
n
= (x y)(x
n1
+x
n2
y + +y
n1
).
(b) ChooseR
1
with 0 < R
1
< R. Showthat theinniteseries M =
n=0
2n|a
n
|R
n
1
converges. Hint: Showthat
n|a
n
|R
n
1
< |a
n
|x
n
for all n sufcientlylargeif R
1
< x < R.
(c) UseEq. (10) toshowthat if |x| < R
1
and|y| < R
1
, then|F(x) F(y)| M|x y|.
(d) Provethat if |x| < R, thenF(x) is continuous at x. Hint: ChooseR
1
suchthat |x| < R
1
< R. Showthat if
> 0isgiven, then|F(x) F(y)| for all y suchthat |x y| < , where isanypositivenumber that isless
than/M andR
1
|x| (seeFigure6).
11.7 Taylor Series
1. Determinef (0) andf
(0) for afunctionf (x) withMaclaurinseries

T (x) = 3+2x +12x
2
+5x
3
+
solution TheMaclaurinseriesfor afunctionf hastheform
f (0) +
f
(0)
1!
x +
f
(0)
2!
x
2
+
f
(0)
3!
x
3
+
Matchingthisgeneral expressionwiththegivenseries, wendf (0) = 3and
f
(0)
3!
= 5. Fromthislatter equation, it
followsthat f
(0) = 30.
2. Determinef (2) andf
(4)
(2) for afunctionwithTaylor series
T (x) = 3(x +2) +(x +2)
2
4(x +2)
3
+2(x +2)
4
+
solution TheTaylor seriesfor afunctionf centeredat x = 2hastheform
f (2) +
f
(2)
1!
(x +2) +
f
(2)
2!
(x +2)
2
+
f
(2)
3!
(x +2)
3
+
f
(4)
(2)
4!
(x +2)
4
+
Matchingthisgeneral expressionwiththegivenseries, wendf (2) = 0and
f
(4)
(2)
4!
= 2. Fromthislatter equation,
it followsthat f
(4)
(2) = 48.
3. What istheeasiest waytondtheMaclaurinseriesfor thefunctionf (x) = sin(x
2
)?
solution Theeasiest waytondtheMaclaurinseriesfor sin
_
x
2
_
istosubstitutex
2
for x intheMaclaurinseriesfor
sinx.
4. FindtheTaylor seriesfor f (x) centeredat c = 3if f (3) = 4andf
(x) hasaTaylor expansion

f
(x) =
n=1
(x 3)
n
n
solution Integratingtheseriesfor f
(x) term-by-termgives
f (x) = C +
n=1
(x 3)
n+1
n(n +1)
.
S E C T I ON 11.7 Taylor Series 711
Substitutingx = 3thenyields
f (3) = C = 4;
so
f (x) = 4+
n=1
(x 3)
n+1
n(n +1)
.
5. Let T (x) betheMaclaurinseriesof f (x). Whichof thefollowingguaranteesthat f (2) = T (2)?
(a) T (x) convergesfor x = 2.
(b) Theremainder R
k
(2) approachesalimit ask .
(c) Theremainder R
k
(2) approacheszeroask .
solution Thecorrect responseis(c): f (2) = T (2) if andonlyif theremainder R
k
(2) approacheszeroask .
Exercises
1. Writeout therst four termsof theMaclaurinseriesof f (x) if
f (0) = 2, f
(0) = 3, f
(0) = 4, f
(0) = 12
solution Therst four termsof theMaclaurinseriesof f (x) are
f (0) +f
(0)x +
f
(0)
2!
x
2
+
f
(0)
3!
x
3
= 2+3x +
4
2
x
2
+
12
6
x
3
= 2+3x +2x
2
+2x
3
.
Writeout therst four termsof theTaylor seriesof f (x) centeredat c = 3if
f (3) = 1, f
(3) = 2, f
(3) = 12, f
(3) = 3
In Exercises 318, nd the Maclaurin series and nd the interval on which the expansion is valid.
3. f (x) =
1
12x
solution Substituting2x for x intheMaclaurinseriesfor
1
1x
gives
1
12x
=
n=0
(2x)
n
=
n=0
2
n
x
n
.
1
2
.
f (x) =
x
1x
4
5. f (x) = cos3x
solution Substituting3x for x intheMaclaurinseriesfor cosx gives
cos3x =
n=0
(1)
n
(3x)
2n
(2n)!
=
n=0
(1)
n
9
n
x
2n
(2n)!
.
Thisseriesisvalidfor all x.
f (x) = sin(2x)
7. f (x) = sin(x
2
)
2
for x intheMaclaurinseriesfor sinx gives
sinx
2
=
n=0
(1)
n
(x
2
)
2n+1
(2n +1)!
=
n=0
(1)
n
x
4n+2
(2n +1)!
.
f (x) = e
4x
9. f (x) = ln(1x
2
)
2
for x intheMaclaurinseriesfor ln(1+x) gives
ln(1x
2
) =
n=1
(1)
n1
(x
2
)
n
n
=
n=1
(1)
2n1
x
2n
n
=
n=1
x
2n
n
.
Thisseriesisvalidfor |x| < 1.
f (x) = (1x)
1/2
11. f (x) = tan
1
(x
2
)
2
for x intheMaclaurinseriesfor tan
1
x gives
tan
1
(x
2
) =
n=0
(1)
n
(x
2
)
2n+1
2n +1
=
n=0
(1)
n
x
4n+2
2n +1
.
Thisseriesisvalidfor |x| 1.
f (x) = x
2
e
x
2
13. f (x) = e
x2
solution e
x2
= e
2
e
x
; thus,
e
x2
= e
2

n=0
x
n
n!
=
n=0
x
n
e
2
n!
.
f (x) =
1cosx
x
15. f (x) = ln(15x)
solution Substituting5x for x intheMaclaurinseriesfor ln(1+x) gives
ln(15x) =
n=1
(1)
n1
(5x)
n
n
=
n=1
(1)
2n1
5
n
x
n
n
=
n=1
5
n
x
n
n
.
1
5
, andfor x =
1
5
.
f (x) = (x
2
+2x)e
x
17. f (x) = sinhx
solution Recall that
sinhx =
1
2
(e
x
e
x
).
Therefore,
sinhx =
1
2
n=0
x
n
n!

n=0
(x)
n
n!
n=0
x
n
2(n!)
_
1(1)
n
_
.
Now,
1(1)
n
=
_
0, n even
2, n odd
so
sinhx =
k=0
2
x
2k+1
2(2k +1)!
=
k=0
x
2k+1
(2k +1)!
.
f (x) = coshx
In Exercises 1928, nd the terms through degree four of the Maclaurin series of f (x). Use multiplication and substitution
as necessary.
19. f (x) = e
x
sinx
solution Multiplythefourth-order Taylor Polynomialsfor e
x
andsinx:
_
1+x +
x
2
2
+
x
3
6
+
x
4
24
__
x
x
3
6
_
= x +x
2
x
3
6
+
x
3
2

x
4
6
+
x
4
6
+ higher-order terms
= x +x
2
+
x
3
3
+ higher-order terms.
Thetermsthroughdegreefour intheMaclaurinseriesfor f (x) = e
x
sinx aretherefore
x +x
2
+
x
3
3
.
f (x) = e
x
ln(1x) 21. f (x) =
sinx
1x
solution Multiplythefourthorder Taylor Polynomialsfor sinx and
1
1x
:
_
x
x
3
6
_
_
1+x +x
2
+x
3
+x
4
_
= x +x
2
x
3
6
+x
3
+x
4
x
4
6
+ higher-order terms
= x +x
2
+
5x
3
6
+
5x
4
6
+ higher-order terms.
Thetermsthroughorder four of theMaclaurinseriesfor f (x) =
sinx
1x
aretherefore
x +x
2
+
5x
3
6
+
5x
4
6
.
f (x) =
1
1+sinx
23. f (x) = (1+x)
1/4
solution Therst vegeneralizedbinomial coefcientsfor a =
1
4
are
1,
1
4
,
1
4
_
3
4
_
2!
=
3
32
,
1
4
_
3
4
_ _
7
4
_
3!
=
7
128
,
1
4
_
3
4
_ _
7
4
_ _
11
4
_
4!
=
77
2048
Therefore, therst four termsinthebinomial seriesfor (1+x)
1/4
are
1+
1
4
x
3
32
x
2
+
7
128
x
3
77
2048
x
4
f (x) = (1+x)
3/2
25. f (x) = e
x
tan
1
x
solution UsingtheMaclaurinseriesfor e
x
andtan
1
x, wend
e
x
tan
1
x =
_
1+x +
x
2
2
+
x
3
6
+
__
x
x
3
3
+
_
= x +x
2
x
3
3
+
x
3
2
+
x
4
6

x
4
3
+
= x +x
2
+
1
6
x
3
1
6
x
4
+ .
f (x) = sin(x
3
x)
27. f (x) = e
sinx
solution Substitutingsinx for x intheMaclaurinseriesfor e
x
andthenusingtheMaclaurinseriesfor sinx, wend
e
sinx
= 1+sinx +
sin
2
x
2
+
sin
3
x
6
+
sin
4
x
24
+
= 1+
_
x
x
3
6
+
_
+
1
2
_
x
x
3
6
+
_
2
+
1
6
(x )
3
+
1
24
(x )
4
= 1+x +
1
2
x
2
1
6
x
3
+
1
6
x
3
1
6
x
4
+
1
24
x
4
+
= 1+x +
1
2
x
2
1
8
x
4
+ .
f (x) = e
(e
x
)
In Exercises 2938, nd the Taylor series centered at c and nd the interval on which the expansion is valid.
29. f (x) =
1
x
, c = 1
solution Write
1
x
=
1
1+(x 1)
,
andthensubstitute(x 1) for x intheMaclaurinseriesfor
1
1x
toobtain
1
x
=
n=0
[(x 1)]
n
=
n=0
(1)
n
(x 1)
n
.
Thisseriesisvalidfor |x 1| < 1.
f (x) = e
3x
, c = 1
31. f (x) =
1
1x
, c = 5
solution Write
1
1x
=
1
4(x 5)
=
1
4
1
1+
x5
4
.
Substituting
x5
4
1
1x
yields
1
1+
x5
4
=
n=0
_
x 5
4
_
n
=
n=0
(1)
n
(x 5)
n
4
n
.
Thus,
1
1x
=
1
4
n=0
(1)
n
(x 5)
n
4
n
=
n=0
(1)
n+1
(x 5)
n
4
n+1
.
Thisseriesisvalidfor
x5
4

< 1, or |x 5| < 4.
f (x) = sinx, c =

2
33. f (x) = x
4
+3x 1, c = 2
solution TodeterminetheTaylor serieswithcenter c = 2, wecompute
f
(x) = 4x
3
+3, f
(x) = 12x
2
, f
(x) = 24x,
andf
(4)
(x) = 24. All derivativesof order veandhigher arezero. Now,
f (2) = 21, f
(2) = 35, f
(2) = 48, f
(2) = 48,
andf
(4)
(2) = 24. Therefore, theTaylor seriesis
21+35(x 2) +
48
2
(x 2)
2
+
48
6
(x 2)
3
+
24
24
(x 2)
4
,
or
21+35(x 2) +24(x 2)
2
+8(x 2)
3
+(x 2)
4
.
f (x) = x
4
+3x 1, c = 0
35. f (x) =
1
x
2
, c = 4
solution Wewill rst ndtheTaylor seriesfor
1
x
andthendifferentiatetoobtaintheseriesfor
1
x
2
. Write
1
x
=
1
4+(x 4)
=
1
4
1
1+
x4
4
.
Nowsubstitute
x4
4
1
1x
toobtain
1
x
=
1
4
n=
_
x 4
4
_
n
=
n=0
(1)
n
(x 4)
n
4
n+1
.
Differentiatingterm-by-termyields
1
x
2
=
n=1
(1)
n
n
(x 4)
n1
4
n+1
,
sothat
1
x
2
=
n=1
(1)
n1
n
(x 4)
n1
4
n+1
=
n=0
(1)
n
(n +1)
(x 4)
n
4
n+2
.
Thisseriesisvalidfor
x4
4

< 1, or |x 4| < 4.
f (x) =
x, c = 4
37. f (x) =
1
1x
2
, c = 3
solution Bypartial fractiondecomposition
1
1x
2
=
1
2
1x
+
1
2
1+x
,
so
1
1x
2
=
1
2
2(x 3)
+
1
2
4+(x 3)
=
1
4
1
1+
x3
2
+
1
8
1
1+
x3
4
.
Substituting
x3
2
1
1x
gives
1
1+
x3
2
=
n=0
_
x 3
2
_
n
=
n=0
(1)
n
2
n
(x 3)
n
,
whilesubstituting
x3
4
for x inthesameseriesgives
1
1+
x3
4
=
n=0
_
x 3
4
_
n
=
n=0
(1)
n
4
n
(x 3)
n
.
Thus,
1
1x
2
=
1
4
n=0
(1)
n
2
n
(x 3)
n
+
1
8
n=0
(1)
n
4
n
(x 3)
n
=
n=0
(1)
n+1
2
n+2
(x 3)
n
+
n=0
(1)
n
2
2n+3
(x 3)
n
=
n=0
_
(1)
n+1
2
n+2
+
(1)
n
2
2n+3
_
(x 3)
n
=
n=0
(1)
n+1
(2
n+1
1)
2
2n+3
(x 3)
n
.
f (x) =
1
3x 2
, c = 1
39. Usetheidentitycos
2
x =
1
2
(1+cos2x) tondtheMaclaurinseriesfor cos
2
x.
solution TheMaclaurinseriesfor cos2x is
n=0
(1)
n
(2x)
2n
(2n)!
=
n=0
(1)
n
2
2n
x
2n
(2n)!
sotheMaclaurinseriesfor cos
2
x =
1
2
(1+cos2x) is
1+
_
1+
n=1
(1)
n 2
2n
x
2n
(2n)!
_
2
= 1+
n=1
(1)
n
2
2n1
x
2n
(2n)!
Showthat for |x| < 1,
tanh
1
x = x +
x
3
3
+
x
5
5
+
Hint: Recall that
d
dx
tanh
1
x =
1
1x
2
.
41. UsetheMaclaurinseriesfor ln(1+x) andln(1x) toshowthat
1
2
ln
_
1+x
1x
_
= x +
x
3
3
+
x
5
5
+
for |x| < 1. What canyouconcludebycomparingthisresult withthat of Exercise40?
solution UsingtheMaclaurinseriesfor ln(1+x) andln(1x), wehavefor |x| < 1
ln(1+x) ln(1x) =
n=1
(1)
n1
n
x
n
n=1
(1)
n1
n
(x)
n
=
n=1
(1)
n1
n
x
n
+
n=1
x
n
n
=
n=1
1+(1)
n1
n
x
n
.
Since1+(1)
n1
= 0for evenn and1+(1)
n1
= 2for oddn,
ln(1+x) ln(1x) =
k=0
2
2k +1
x
2k+1
.
Thus,
1
2
ln
_
1+x
1x
_
=
1
2
(ln(1+x) ln(1x)) =
1
2
k=0
2
2k +1
x
2k+1
=
k=0
x
2k+1
2k +1
.
Observethat thisisthesameserieswefoundinExercise40; therefore,
1
2
ln
_
1+x
1x
_
= tanh
1
x.
DifferentiatetheMaclaurinseriesfor
1
1x
twicetondtheMaclaurinseriesof
1
(1x)
3
.
43. Show, byintegratingtheMaclaurinseriesfor f (x) =
1
_
1x
2
, that for |x| < 1,
sin
1
x = x +
n=1
1 3 5 (2n 1)
2 4 6 (2n)
x
2n+1
2n +1
solution FromExample10, weknowthat for |x| < 1
1
_
1x
2
=
n=0
1 3 5 (2n 1)
2 4 6 (2n)
x
2n
= 1+
n=1
1 3 5 (2n 1)
2 4 6 (2n)
x
2n
,
so, for |x| < 1,
sin
1
x =
_
dx
_
1x
2
= C +x +
n=1
1 3 5 (2n 1)
2 4 6 (2n)
x
2n+1
2n +1
.
Sincesin
1
0= 0, wendthat C = 0. Thus,
sin
1
x = x +
n=1
1 3 5 (2n 1)
2 4 6 (2n)
x
2n+1
2n +1
.
Usetherst vetermsof theMaclaurinseriesinExercise43toapproximatesin
1 1
2
. Comparetheresult with
thecalculator value.
45. Howmanytermsof theMaclaurinseriesof f (x) = ln(1+x) areneededtocomputeln1.2towithinanerror of at
most 0.0001? Makethecomputationandcomparetheresult withthecalculator value.
solution Substitutex = 0.2intotheMaclaurinseriesfor ln(1+x) toobtain:
ln1.2=
n=1
(1)
n1
(0.2)
n
n
=
n=1
(1)
n1
1
5
n
n
.
Thisisanalternatingserieswitha
n
=
1
n 5
n
. Usingtheerror boundfor alternatingseries
|ln1.2S
N
| a
N+1
=
1
(N +1)5
N+1
,
sowemust chooseN sothat
1
(N +1)5
N+1
< 0.0001 or (N +1)5
N+1
> 10,000.
For N = 3, (N +1)5
N+1
= 4 5
4
= 2500< 10, 000, and for N = 4, (N +1)5
N+1
= 5 5
5
= 15, 625> 10, 000;
thus, thesmallest acceptablevaluefor N isN = 4. Thecorrespondingapproximationis:
S
4
=
4
n=1
(1)
n1
5
n
n
=
1
5
1
5
2
2
+
1
5
3
3
1
5
4
4
= 0.182266666.
Now, ln1.2= 0.182321556, so
|ln1.2S
4
| = 5.48910
5
< 0.0001.
Showthat

3
3!
+

5
5!

7
7!
+
convergestozero. Howmanytermsmust becomputedtoget within0.01of zero?
47. UsetheMaclaurinexpansionfor e
t
2
toexpressthefunctionF(x) =
_
x
0
e
t
2
dt asanalternatingpower seriesinx
(Figure4).
(a) Howmany terms of theMaclaurinseries areneededtoapproximatetheintegral for x = 1towithinanerror of at
most 0.001?
(b) Carryout thecomputationandcheckyour answer usingacomputer algebrasystem.
F(x)
T
15
(x)
1 2
y
x
FIGURE 4 TheMaclaurinpolynomial T
15
(x) for F(t ) =
_
x
0
e
t
2
dt.
solution Substitutingt
2
for t intheMaclaurinseriesfor e
t
yields
e
t
2
=
n=0
(t
2
)
n
n!
=
n=0
(1)
n
t
2n
n!
;
thus,
_
x
0
e
t
2
dt =
n=0
(1)
n
x
2n+1
n!(2n +1)
.
(a) For x = 1,
_
1
0
e
t
2
dt =
n=0
(1)
n
1
n!(2n +1)
.
n
=
1
n!(2n+1)
; therefore, theerror incurredby usingS
N
toapproximatethevalueof
thedeniteintegral isboundedby
_
1
0
e
t
2
dt S
N
a
N+1
=
1
(N +1)!(2N +3)
.
Toguaranteetheerror isat most 0.001, wemust chooseN sothat
1
(N +1)!(2N +3)
< 0.001 or (N +1)!(2N +3) > 1000.
For N = 3, (N + 1)!(2N + 3) = 4! 9 = 216 < 1000andfor N = 4, (N + 1)!(2N + 3) = 5! 11 = 1320 > 1000;
thus, thesmallest acceptablevaluefor N isN = 4. Thecorrespondingapproximationis
S
4
=
4
n=0
(1)
n
n!(2n +1)
= 1
1
3
+
1
2! 5
1
3! 7
+
1
4! 9
= 0.747486772.
(b) Usingacomputer algebrasystem, wend
_
1
0
e
t
2
dt = 0.746824133;
therefore
_
1
0
e
t
2
dt S
4
= 6.62610
4
< 10
3
.
Let F(x) =
_
x
0
sint dt
t
. Showthat
F(x) = x
x
3
3 3!
+
x
5
5 5!

x
7
7 7!
+
EvaluateF(1) tothreedecimal places.
In Exercises 4952, express the denite integral as an innite series and nd its value to within an error of at most 10
4
.
49.
_
1
0
cos(x
2
) dx
2
for x intheMaclaurinseriesfor cosx yields
cos(x
2
) =
n=0
(1)
n
(x
2
)
2n
(2n)!
=
n=0
(1)
n
x
4n
(2n)!
;
therefore,
_
1
0
cos(x
2
) dx =
n=0
(1)
n
x
4n+1
(2n)!(4n +1)
1
0
=
n=0
(1)
n
(2n)!(4n +1)
.
n
=
1
(2n)!(4n+1)
; therefore, theerror incurredbyusingS
N
_
1
0
cos(x
2
) dx S
N
a
N+1
=
1
(2N +2)!(4N +5)
.
1
(2N +2)!(4N +5)
< 0.0001 or (2N +2)!(4N +5) > 10,000.
For N = 2, (2N +2)!(4N +5) = 6! 13= 9360< 10,000andfor N = 3, (2N +2)!(4N +5) = 8! 17= 685,440>
10,000; thus, thesmallest acceptablevaluefor N isN = 3. Thecorrespondingapproximationis
S
3
=
3
n=0
(1)
n
(2n)!(4n +1)
= 1
1
5 2!
+
1
9 4!

1
13 6!
= 0.904522792.
_
1
0
tan
1
(x
2
) dx
51.
_
1
0
e
x
3
dx
3
for x intheMaclaurinseriesfor e
x
yields
e
x
3
=
n=0
(x
3
)
n
n!
=
n=0
(1)
n
x
3n
n!
;
therefore,
_
1
0
e
x
3
dx =
n=0
(1)
n
x
3n+1
n!(3n +1)
1
0
=
n=0
(1)
n
n!(3n +1)
.
n
=
1
n!(3n+1)
; therefore, theerror incurredby usingS
N
_
1
0
e
x
3
dx S
N
a
N+1
=
1
(N +1)!(3N +4)
.
1
(N +1)!(3N +4)
< 0.0001 or (N +1)!(3N +4) > 10,000.
For N = 4, (N + 1)!(3N + 4) = 5! 16 = 1920 < 10,000andfor N = 5, (N + 1)!(3N + 4) = 6! 19 = 13,680 >
10,000; thus, thesmallest acceptablevaluefor N isN = 5. Thecorrespondingapproximationis
S
5
=
5
n=0
(1)
n
n!(3n +1)
= 0.807446200.
_
1
0
dx
_
x
4
+1
In Exercises 5356, express the integral as an innite series.
53.
_
x
0
1cos(t )
t
dt , for all x
solution TheMaclaurinseriesfor cost is
cost =
n=0
(1)
n
t
2n
(2n)!
= 1+
n=1
(1)
n
t
2n
(2n)!
,
so
1cost =
n=1
(1)
n
t
2n
(2n)!
=
n=1
(1)
n+1
t
2n
(2n)!
,
and
1cost
t
=
1
t
n=1
(1)
n+1
t
2n
(2n)!
=
n=1
(1)
n+1
t
2n1
(2n)!
.
Thus,
_
x
0
1cos(t )
t
dt =
n=1
(1)
n+1
t
2n
(2n)!2n
x
0
=
n=1
(1)
n+1
x
2n
(2n)!2n
.
_
x
0
t sint
t
dt , for all x
55.
_
x
0
ln(1+t
2
) dt , for |x| < 1
solution Substitutingt
2
for t intheMaclaurinseriesfor ln(1+t ) yields
ln(1+t
2
) =
n=1
(1)
n1
(t
2
)
n
n
=
n=1
(1)
n
t
2n
n
.
Thus,
_
x
0
ln(1+t
2
) dt =
n=1
(1)
n
t
2n+1
n(2n +1)
x
0
=
n=1
(1)
n
x
2n+1
n(2n +1)
.
_
x
0
dt
_
1t
4
, for |x| < 1
57. WhichfunctionhasMaclaurinseries

n=0
(1)
n
2
n
x
n
?
solution Werecognizethat
n=0
(1)
n
2
n
x
n
=
n=0
(2x)
n
istheMaclaurinseriesfor
1
1x
withx replacedby2x. Therefore,
n=0
(1)
n
2
n
x
n
=
1
1(2x)
=
1
1+2x
.
WhichfunctionhasMaclaurinseries
k=0
(1)
k
3
k+1
(x 3)
k
?
For whichvaluesof x istheexpansionvalid?
In Exercises 5962, use Theorem 2 to prove that the f (x) is represented by its Maclaurin series for all x.
59. f (x) = sin
_
x
2
_
+cos
_
x
3
_
,
solution All derivatives of f (x) consist of sin or cos appliedto eachof x/2andx/3andaddedtogether, so each
summandisboundedby1. Thus
f
(n)
(x)
2for all n andx. ByTheorem2, f (x) isrepresentedbyitsTaylor seriesfor

everyx.
f (x) = e
x
,
61. f (x) = sinhx,
solution Bydenition, sinhx =
1
2
(e
x
e
x
), soif bothe
x
ande
x
arerepresentedbytheir Taylor seriescentered
at c, thensoissinhx. But thepreviousexerciseshowsthat e
x
issorepresented, andthetext showsthat e
x
is.
f (x) = (1+x)
100
In Exercises 6366, nd the functions with the following Maclaurin series (refer to Table 1 on page 599).
63. 1+x
3
+
x
6
2!
+
x
9
3!
+
x
12
4!
+
solution Werecognize
1+x
3
+
x
6
2!
+
x
9
3!
+
x
12
4!
+ =
n=0
x
3n
n!
=
n=0
(x
3
)
n
n!
astheMaclaurinseriesfor e
x
withx replacedbyx
3
. Therefore,
1+x
3
+
x
6
2!
+
x
9
3!
+
x
12
4!
+ = e
x
3
.
14x +4
2
x
2
4
3
x
3
+4
4
x
4
4
5
x
5
+ 65. 1
5
3
x
3
3!
+
5
5
x
5
5!

5
7
x
7
7!
+
solution Note
1
5
3
x
3
3!
+
5
5
x
5
5!

5
7
x
7
7!
+ = 15x +
_
5x
5
3
x
3
3!
+
5
5
x
5
5!

5
7
x
7
7!
+
_
= 15x +
n=0
(1)
n
(5x)
2n+1
(2n +1)!
.
TheseriesistheMaclaurinseriesfor sinx withx replacedby5x, so
1
5
3
x
3
3!
+
5
5
x
5
5!

5
7
x
7
7!
+ = 15x +sin(5x).
x
4
x
12
3
+
x
20
5

x
28
7
+
In Exercises 67 and 68, let
f (x) =
1
(1x)(12x)
67. FindtheMaclaurinseriesof f (x) usingtheidentity
f (x) =
2
12x
1
1x
solution Substituting2x for x intheMaclaurinseriesfor
1
1x
gives
1
12x
=
n=0
(2x)
n
=
n=0
2
n
x
n
which is valid for |2x| < 1, or |x| <
1
2
. BecausetheMaclaurin series for
1
1x
is valid for |x| < 1, thetwo series
together arevalidfor |x| <
1
2
. Thus, for |x| <
1
2
,
1
(12x)(1x)
=
2
12x
1
1x
= 2

n=0
2
n
x
n
n=0
x
n
=
n=0
2
n+1
x
n
n=0
x
n
=
n=0
_
2
n+1
1
_
x
n
.
FindtheTaylor seriesfor f (x) at c = 2. Hint: Rewritetheidentityof Exercise67as
f (x) =
2
32(x 2)
1
1(x 2)
69. WhenavoltageV isappliedtoaseriescircuit consistingof aresistor R andaninductor L, thecurrent at timet is
I (t ) =
_
V
R
_
_
1e
Rt /L
_
ExpandI (t ) inaMaclaurinseries. Showthat I (t )
Vt
L
for small t .
solution Substituting
Rt
L
for t intheMaclaurinseriesfor e
t
gives
e
Rt /L
=
n=0
_
Rt
L
_
n
n!
=
n=0
(1)
n
n!
_
R
L
_
n
t
n
= 1+
n=1
(1)
n
n!
_
R
L
_
n
t
n
Thus,
1e
Rt /L
= 1
1+
n=1
(1)
n
n!
_
R
L
_
n
t
n
n=1
(1)
n+1
n!
_
Rt
L
_
n
,
and
I (t ) =
V
R
n=1
(1)
n+1
n!
_
Rt
L
_
n
=
Vt
L
+
V
R
n=2
(1)
n+1
n!
_
Rt
L
_
n
.
If t issmall, thenwecanapproximateI (t ) by therst (linear) term, andignoretermswithhigher powersof t ; thenwe
nd
V(t )
Vt
L
.
Usetheresult of Exercise69andyour knowledgeof alternatingseriestoshowthat
Vt
L
_
1
R
2L
t
_
I (t )
Vt
L
(for all t )
71. FindtheMaclaurinseriesfor f (x) = cos(x
3
) anduseit todeterminef
(6)
(0).
solution TheMaclaurinseriesfor cosx is
cosx =
n=0
(1)
n
x
2n
(2n)!
Substitutingx
3
for x gives
cos(x
3
) =
n=0
(1)
n
x
6n
(2n)!
Now, thecoefcient of x
6
inthisseriesis
1
2!
=
1
2
=
f
(6)
(0)
6!
so
f
(6)
(0) =
6!
2
= 360
Findf
(7)
(0) andf
(8)
(0) for f (x) = tan
1
x usingtheMaclaurinseries.
73. Usesubstitutionto ndtherst threeterms of theMaclaurinseries for f (x) = e
x
20
. Howdoes theresult
showthat f
(k)
(0) = 0for 1 k 19?
20
x
yields
e
x
20
=
n=0
(x
20
)
n
n!
=
n=0
x
20n
n!
;
therst threetermsintheseriesarethen
1+x
20
+
1
2
x
40
.
Recall that thecoefcient of x
k
intheMaclaurinseries for f is
f
(k)
(0)
k!
. For 1 k 19, thecoefcient of x
k
inthe
Maclaurinseriesfor f (x) = e
x
20
iszero; it thereforefollowsthat
f
(k)
(0)
k!
= 0 or f
(k)
(0) = 0
for 1 k 19.
Usethebinomial seriestondf
(8)
(0) for f (x) =
_
1x
2
.
75. DoestheMaclaurinseriesfor f (x) = (1+ x)
3/4
convergetof (x) at x = 2? Givenumerical evidencetosupport
your answer.
solution TheTaylor seriesfor f (x) = (1+x)
3/4
convergestof (x) for |x| < 1; becausex = 2isnot containedon
thisinterval, theseriesdoesnot convergetof (x) at x = 2. Thegraphbelowdisplays
S
N
=
N
n=0
_
3
4
n
_
2
n
for 0 N 14. Thedivergent natureof thesequenceof partial sumsisclear.
0
2 14 10 6 8 12 4
5
10
15
20
15
10
5
S
N
N
Explainthestepsrequiredtoverifythat theMaclaurinseriesfor f (x) = e
x
convergestof (x) for all x.
77. Let f (x) =
1+x.
(a) Useagraphingcalculator tocomparethegraphof f withthegraphsof therst veTaylor polynomialsfor f . What
dotheysuggest about theinterval of convergenceof theTaylor series?
(b) Investigatenumericallywhether or not theTaylor expansionfor f isvalidfor x = 1andx = 1.
solution
(a) Theverst termsof theBinomial serieswitha =
1
2
are
1+x = 1+
1
2
x +
1
2
_
1
2
1
_
2!
x
2
+
1
2
_
1
2
1
_ _
1
2
2
_
3!
x
3
+
1
2
_
1
2
1
_ _
1
2
2
_ _
1
2
3
_
4!
x
4
+
= 1+
1
2
x
1
8
x
2
+
9
4
x
3
45
2
x
4
+
Therefore, therst veTaylor polynomialsare
T
0
(x) = 1;
T
1
(x) = 1+
1
2
x;
T
2
(x) = 1+
1
2
x
1
8
x
2
;
T
3
(x) = 1+
1
2
x
1
8
x
2
+
1
8
x
3
;
T
4
(x) = 1+
1
2
x
1
8
x
2
+
1
8
x
3
5
128
x
4
.
Theguredisplaysthegraphsof theseTaylor polynomials, alongwiththegraphof thefunctionf (x) =
1+x, which
isshowninred.
1 0.5 0.5 1
1.5
1
1.5
Thegraphssuggest that theinterval of convergencefor theTaylor seriesis1< x < 1.
(b) Usingacomputer algebrasystemtocalculateS
N
=
N
n=0
_
1
2
n
_
x
n
for x = 1wend
S
10
= 1.409931183, S
100
= 1.414073048, S
1000
= 1.414209104,
whichappearstobeconvergingto
2asexpected. At x = 1wecalculateS
N
=
N
n=0
_
1
2
n
_
(1)
n
, andnd
S
10
= 0.176197052, S
100
= 0.056348479, S
1000
= 0.017839011,
whichappearstobeconvergingtozero, thoughslowly.
Usetherst vetermsof theMaclaurinseriesfor theellipticfunctionE(k) toestimatetheperiodT of a1-meter
pendulumreleasedat anangle =

4
(seeExample11).
79. UseExample11andtheapproximationsinx x toshowthattheperiodT of apendulumreleasedatanangle has
thefollowingsecond-order approximation:
T 2
_
L
g
_
1+

2
16
_
solution TheperiodT of apendulumof lengthL releasedfromanangle is
T = 4
_
L
g
E(k),
whereg 9.8m/s
2
istheaccelerationduetogravity, E(k) istheellipticfunctionof therst kindandk = sin
2
. From
Example11, weknowthat
E(k) =

2
n=0
_
1 3 5 (2n 1)
2 4 6 (2n)
_
2
k
2n
.
Usingtheapproximationsinx x, wehave
k = sin

2
;
moreover, usingtherst twotermsof theseriesfor E(k), wend
E(k)

2
_
1+
_
1
2
_
2
_
2
_
2
_
=

2
_
1+

2
16
_
.
Therefore,
T = 4
_
L
g
E(k) 2
_
L
g
_
1+

2
16
_
.
In Exercises 8083, nd the Maclaurin series of the function and use it to calculate the limit.
lim
x0
cosx 1+
x
2
2
x
4
81. lim
x0
sinx x +
x
3
6
x
5
solution UsingtheMaclaurinseriesfor sinx, wend
sinx =
n=0
(1)
n
x
2n+1
(2n +1)!
= x
x
3
6
+
x
5
120
+
n=3
(1)
n
x
2n+1
(2n +1)!
.
Thus,
sinx x +
x
3
6
=
x
5
120
+
n=3
(1)
n
x
2n+1
(2n +1)!
and
sinx x +
x
3
6
x
5
=
1
120
+
n=3
(1)
n
x
2n4
(2n +1)!
Notethattheradiusof convergencefor thisseriesisinnite, andrecall fromtheprevioussectionthataconvergentpower
seriesiscontinuouswithinitsradiusof convergence. Thustocalculatethelimit of thispower seriesasx 0it sufces
toevaluateit at x = 0:
lim
x0
sinx x +
x
3
6
x
5
= lim
x0
1
120
+
n=3
(1)
n
x
2n4
(2n +1)!
=
1
120
+0=
1
120
lim
x0
tan
1
x x cosx
1
6
x
3
x
5
83. lim
x0
_
sin(x
2
)
x
4

cosx
x
2
_
solution Westart with
sinx =
n=0
(1)
n
x
2n+1
(2n +1)!
cosx =
n=0
(1)
n
x
2n
(2n)!
sothat
sin(x
2
)
x
4
=
n=0
(1)
n
x
4n+2
(2n +1)!x
4
=
n=0
(1)
n
x
4n2
(2n +1)!
cosx
x
2
=
n=0
(1)
n
x
2n2
(2n)!
Expandingtherst fewtermsgives
sin(x
2
)
x
4
=
1
x
2
n=1
(1)
n
x
4n2
(2n +1)!
cosx
x
2
=
1
x
2
1
2
+
n=2
(1)
n
x
2n2
(2n)!
sothat
sin(x
2
)
x
4

cosx
x
2
=
1
2
n=1
(1)
n
x
4n2
(2n +1)!

n=2
(1)
n
x
2n2
(2n)!
Notethat all termsunder thesummationsignshavepositivepowersof x. Now, theradiusof convergenceof theseries
for bothsinandcosisinnite, sotheradiusof convergenceof thisseriesisinnite. Recall fromtheprevioussectionthat
aconvergent power seriesiscontinuouswithinitsradiusof convergence. Thustocalculatethelimit of thispower series
asx 0it sufcestoevaluateit at x = 0:
lim
x0
_
sin(x
2
)
x
4

cosx
x
2
_
= lim
x0
1
2
n=1
(1)
n
x
4n2
(2n +1)!

n=2
(1)
n
x
2n2
(2n)!
=
1
2
+0=
1
2
Inthisexerciseweshowthat theMaclaurinexpansionof f (x) = ln(1+x) isvalidfor x = 1.
(a) Showthat for all x = 1,
1
1+x
=
N
n=0
(1)
n
x
n
+
(1)
N+1
x
N+1
1+x
(b) Integratefrom0to1toobtain
ln2=
N
n=1
(1)
n1
n
+(1)
N+1
_
1
0
x
N+1
dx
1+x
(c) Verifythat theintegral ontheright tendstozeroasN byshowingthat it issmaller than
_
1
0
x
N+1
dx.
(d) Provetheformula
ln2= 1
1
2
+
1
3
1
4
+
85. Let g(t ) =
1
1+t
2
t
1+t
2
.
(a) Showthat
_
1
0
g(t ) dt =

4

1
2
ln2.
(b) Showthat g(t ) = 1t t
2
+t
3
t
4
t
5
t
6
+
(c) EvaluateS = 1
1
2
1
3
+
1
4
1
5
1
6
1
7
+
solution
(a)
_
1
0
g(t ) dt =
_
tan
1
t
1
2
ln(t
2
+1)
_
1
0
= tan
1
1
1
2
ln2=

4

1
2
ln2
(b) Start withtheTaylor seriesfor
1
1+t
:
1
1+t
=
n=0
(1)
n
t
n
andsubstitutet
2
for t toget
1
1+t
2
=
n=0
(1)
n
t
2n
= 1t
2
+t
4
t
6
+. . .
sothat
t
1+t
2
=
n=0
(1)
n
t
2n+1
= t t
3
+t
5
t
7
+. . .
Finally,
g(t ) =
1
1+t
2
t
1+t
2
= 1t t
2
+t
3
+t
4
t
5
t
6
+t
7
+. . .
(c) Wehave
_
g(t ) dt =
_
(1t t
2
+t
3
+t
4
t
5
. . . ) dt = t
1
2
t
2
1
3
t
3
+
1
4
t
4
+
1
5
t
5
1
6
t
6
+C
Theradiusof convergenceof theseriesfor g(t ) is1, sotheradiusof convergenceof thisseriesisalso1. However, this
seriesconvergesat theright endpoint, t = 1, since
_
1
1
2
_
_
1
3
1
4
_
+
_
1
5
1
6
_
. . .
isanalternatingserieswithgeneral termdecreasingtozero. Thusbypart (a),
1
1
2
1
3
+
1
4
+
1
5
1
6
=

4

1
2
ln2
In Exercises 86 and 87, we investigate the convergence of the binomial series
T
a
(x) =
n=0
_
a
n
_
x
n
Provethat T
a
(x) hasradiusof convergenceR = 1if a isnot awholenumber. What istheradiusof convergence
if a isawholenumber?
87. ByExercise86, T
a
(x) convergesfor |x| < 1, but wedonot yet knowwhether T
a
(x) = (1+x)
a
.
(a) Verifytheidentity
a
_
a
n
_
= n
_
a
n
_
+(n +1)
_
a
n +1
_
(b) Use(a) toshowthat y = T
a
(x) satisesthedifferential equation(1+x)y
= ay withinitial conditiony(0) = 1.
(c) Provethat T
a
(x) = (1+x)
a
for |x| < 1byshowingthat thederivativeof theratio
T
a
(x)
(1+x)
a
iszero.
solution
(a)
n
_
a
n
_
+(n +1)
_
a
n +1
_
= n
a (a 1) (a n +1)
n!
+(n +1)
a (a 1) (a n +1) (a n)
(n +1)!
=
a (a 1) (a n +1)
(n 1)!
+
a (a 1) (a n +1) (a n)
n!
=
a (a 1) (a n +1) (n +(a n))
n!
= a
_
a
n
_
(b) DifferentiatingT
a
(x) term-by-termyields
T
a
(x) =
n=1
n
_
a
n
_
x
n1
.
Thus,
(1+x)T
a
(x) =
n=1
n
_
a
n
_
x
n1
+
n=1
n
_
a
n
_
x
n
=
n=0
(n +1)
_
a
n +1
_
x
n
+
n=0
n
_
a
n
_
x
n
=
n=0
_
(n +1)
_
a
n +1
_
+n
_
a
n
__
x
n
= a
n=0
_
a
n
_
x
n
= aT
a
(x).
Moreover,
T
a
(0) =
_
a
0
_
= 1.
(c)
d
dx
_
T
a
(x)
(1+x)
a
_
=
(1+x)
a
T
a
(x) a(1+x)
a1
T
a
(x)
(1+x)
2a
=
(1+x)T
a
(x) aT
a
(x)
(1+x)
a+1
= 0.
Thus,
T
a
(x)
(1+x)
a
= C,
for someconstant C. For x = 0,
T
a
(0)
(1+0)
a
=
1
1
= 1, soC = 1.
Finally, T
a
(x) = (1+x)
a
.
Thefunction G(k) =
_
/2
0
_
1k
2
sin
2
t dt is called an elliptic functionof thesecondkind. Provethat for
|k| < 1,
G(k) =

_
1 3 (2n 1)
_
2
k
2n
89. Assumethat a < b andlet L bethearc length(circumference) of theellipse
_
x
a
_
2
+
_
y
b
_
2
= 1showninFigure5.
Thereisnoexplicit formulafor L, but it isknownthat L = 4bG(k), withG(k) asinExercise88andk =
_
1a
2
/b
2
.
Usetherst threetermsof theexpansionof Exercise88toestimateL whena = 4andb = 5.
a
b
y
x
FIGURE 5 Theellipse
_
x
a
_
2
+
_
y
b
_
2
= 1.
solution Witha = 4andb = 5,
k =
_
1
4
2
5
2
=
3
5
,
andthearclengthof theellipse
_
x
4
_
2
+
_
y
5
_
2
= 1is
L = 20G
_
3
5
_
= 20
2

2
n=1
_
1 3 (2n 1)
2 4 (2n)
_
2
_
3
5
_
2n
2n 1
.
Usingtherst threetermsintheseriesfor G(k) gives
L 10 10
_
_
1
2
_
2
(3/5)
2
1
+
_
1 3
2 4
_
2
(3/5)
4
3
_
= 10
_
1
9
100
243
40,000
_
=
36,157
4000
28.398.
UseExercise88toprovethat if a < b anda/b isnear 1(anearlycircular ellipse), then
L

2
_
3b +
a
2
b
_
Hint: Usetherst twotermsof theseriesfor G(k).
91. Irrationalityofe Provethate isanirrational number usingthefollowingargumentbycontradiction. Supposethat
e = M/N, whereM, N arenonzerointegers.
(a) Showthat M! e
1
isawholenumber.
(b) Usethepower seriesfor e
x
at x = 1toshowthat thereisaninteger B suchthat M! e
1
equals
B +(1)
M+1
_
1
M +1
1
(M +1)(M +2)
+
_
(c) Useyour knowledgeof alternatingserieswithdecreasingtermstoconcludethat0< |M! e
1
B| < 1andobserve
that thiscontradicts(a). Hence, e isnot equal toM/N.
solution Supposethat e = M/N, whereM, N arenonzerointegers.
(a) Withe = M/N,
M!e
1
= M!
N
M
= (M 1)!N,
whichisawholenumber.
(b) Substitutingx = 1intotheMaclaurinseriesfor e
x
andmultiplyingtheresultingseriesbyM! yields
M!e
1
= M!
_
11+
1
2!

1
3!
+ +
(1)
k
k!
+
_
.
For all k M,
M!
k!
isawholenumber, so
M!
_
11+
1
2!

1
3!
+ +
(1)
k
M!
_
isaninteger. Denotethisinteger byB. Thus,
M! e
1
= B +M!
_
(1)
M+1
(M +1)!
+
(1)
M+2
(M +2)!
+
_
= B +(1)
M+1
_
1
M +1
1
(M +1)(M +2)
+
_
.
(c) The series for M! e
1
obtained in part (b) is an alternating series with a
n
=
M!
n!
. Using the error bound for an
alternatingseriesandnotingthat B = S
M
, wehave
M! e
1
B
a
M+1
=
1
M +1
< 1.
Thisinequality impliesthat M! e
1
B isnot awholenumber; however, B isawholenumber soM! e
1
cannot bea
wholenumber. Wegetacontradictiontotheresultinpart(a), whichprovesthattheoriginal assumptionthate isarational
number isfalse.
Usetheresult of Exercise69inSection7.7toshowthat theMaclaurinseriesof thefunction
f (x) =
_
e
1/x
2
for x = 0
0 for x = 0
isT (x) = 0. Thisprovidesanexampleof afunctionf (x) whoseMaclaurinseriesconvergesbut doesnot converge
tof (x) (except at x = 0).
1. Let a
n
=
n 3
n!
andb
n
= a
n+3
. Calculatetherst threetermsineachsequence.
(a) a
2
n
(b) b
n
(c) a
n
b
n
(d) 2a
n+1
3a
n
solution
(a)
a
2
1
=
_
13
1!
_
2
= (2)
2
= 4;
a
2
2
=
_
23
2!
_
2
=
_
1
2
_
2
=
1
4
;
a
2
3
=
_
33
3!
_
2
= 0.
(b)
b
1
= a
4
=
43
4!
=
1
24
;
b
2
= a
5
=
53
5!
=
1
60
;
b
3
= a
6
=
63
6!
=
1
240
.
(c) Usingtheformulafor a
n
andthevaluesin(b) weobtain:
a
1
b
1
=
13
1!

1
24
=
1
12
;
a
2
b
2
=
23
2!

1
60
=
1
120
;
a
3
b
3
=
33
3!

1
240
= 0.
(d)
2a
2
3a
1
= 2
_
1
2
_
3(2) = 5;
2a
3
3a
2
= 2 03
_
1
2
_
=
3
2
;
2a
4
3a
3
= 2
1
24
3 0=
1
12
.
Provethat lim
n
2n 1
3n +2
=
2
3
usingthelimit denition.
In Exercises 38, compute the limit (or state that it does not exist) assuming that lim
n
a
n
= 2.
3. lim
n
(5a
n
2a
2
n
)
solution
lim
n
_
5a
n
2a
2
n
_
= 5 lim
n
a
n
2 lim
n
a
2
n
= 5 lim
n
a
n
2
_
lim
n
a
n
_
2
= 5 22 2
2
= 2.
lim
n
1
a
n
5. lim
n
e
a
n
solution Thefunctionf (x) = e
x
iscontinuous, hence:
lim
n
e
a
n
= e
lim
n
a
n
= e
2
.
lim
n
cos(a
n
)
7. lim
n
(1)
n
a
n
n
a
n
= 0, it followsthat lim
n
(1)
n
a
n
doesnot exist.
lim
n
a
n
+n
a
n
+n
2
In Exercises 922, determine the limit of the sequence or show that the sequence diverges.
9. a
n
=
n +5
n +2
solution First rewritea
n
asfollows:
a
n
=
_
n +5
n +2
_ _
n +5+
n +2
_
n +5+
n +2
=
(n +5) (n +2)
n +5+
n +2
=
3
n +5+
n +2
.
Thus,
lim
n
a
n
= lim
n
3
n +5+
n +2
= 0.
a
n
=
3n
3
n
12n
3
11. a
n
= 2
1/n
2
solution Thefunctionf (x) = 2
x
iscontinuous, so
lim
n
a
n
= lim
n
2
1/n
2
= 2
lim
n
(1/n
2
)
= 2
0
= 1.
a
n
=
10
n
n!
13. b
m
= 1+(1)
m
solution Because1+(1)
m
isequal to0formoddandisequal to2formeven, thesequence{b
m
} doesnotapproach
onelimit; hencethissequencediverges.
b
m
=
1+(1)
m
m
15. b
n
= tan
1
_
n +2
n +5
_
solution Thefunctiontan
1
x iscontinuous, so
lim
n
b
n
= lim
n
tan
1
_
n +2
n +5
_
= tan
1
_
lim
n
n +2
n +5
_
= tan
1
1=

4
.
a
n
=
100
n
n!

3+
n
5
n
17. b
n
=
_
n
2
+n
_
n
2
+1
solution Rewriteb
n
as
b
n
=
__
n
2
+n
_
n
2
+1
_ __
n
2
+n +
_
n
2
+1
_
_
n
2
+n +
_
n
2
+1
=
_
n
2
+n
_
_
n
2
+1
_
_
n
2
+n +
_
n
2
+1
=
n 1
_
n
2
+n +
_
n
2
+1
.
Then
lim
n
b
n
= lim
n
n
n

1
n
_
n
2
n
2
+
n
n
2
+
_
n
2
n
2
+
1
n
2
= lim
n
1
1
n
_
1+
1
n
+
_
1+
1
n
2
=
10
1+0+
1+0
=
1
2
.
c
n
=
_
n
2
+n
_
n
2
n
19. b
m
=
_
1+
1
m
_
3m
solution lim
m
b
m
= lim
m
_
1+
1
m
_
m
= e
3
.
c
n
=
_
1+
3
n
_
n
21. b
n
= n
_
ln(n +1) lnn
_
solution Write
b
n
= n ln
_
n +1
n
_
=
ln
_
1+
1
n
_
1
n
.
UsingLHpitalsRule, wend
lim
n
b
n
= lim
n
ln
_
1+
1
n
_
1
n
= lim
x
ln
_
1+
1
x
_
1
x
= lim
x
_
1+
1
x
_
1
1
x
2
_
1
x
2
= lim
x
_
1+
1
x
_
1
= 1.
c
n
=
ln(n
2
+1)
ln(n
3
+1)
23. UsetheSqueezeTheoremtoshowthat lim
n
arctan(n
2
)
n
= 0.
solution For all x,
2
< arctanx <
2
,
so
/2
n
<
arctan(n
2
)
n
<
/2
n
,
for all n. Because
lim
n
_
/2
n
_
= lim
n
/2
n
= 0,
lim
n
arctan(n
2
)
n
= 0.
Giveanexampleof adivergent sequence{a
n
} suchthat {sina
n
} isconvergent. 25. Calculate lim
n
a
n+1
a
n
, wherea
n
=
1
2
3
n
1
3
2
n
.
solution Because
1
2
3
n
1
3
2
n
1
2
3
n
1
3
3
n
=
3
n
6
and
lim
n
3
n
6
= ,
weconcludethat lim
n
a
n
= , soLHpitalsrulemaybeused:
lim
n
a
n+1
a
n
= lim
n
1
2
3
n+1
1
3
2
n+1
1
2
3
n
1
3
2
n
= lim
n
3
n+2
2
n+2
3
n+1
2
n+1
= lim
n
32
_
2
3
_
n+1
1
_
2
3
_
n+1
=
30
10
= 3.
Denea
n+1
=
_
a
n
+6witha
1
= 2.
(a) Computea
n
for n = 2, 3, 4, 5.
(b) Showthat {a
n
} isincreasingandisboundedby3.
(c) Provethat lim
n
a
n
existsandnditsvalue.
27. Calculatethepartial sumsS
4
andS
7
of theseries

n=1
n 2
n
2
+2n
.
solution
S
4
=
1
3
+0+
1
15
+
2
24
=
11
60
= 0.183333;
S
7
=
1
3
+0+
1
15
+
2
24
+
3
35
+
4
48
+
5
63
=
287
4410
= 0.065079.
Findthesum1
1
4
+
1
4
2
1
4
3
+ .
29. Findthesum
4
9
+
8
27
+
16
81
+
32
243
+ .
solution Thisisageometricserieswithcommonratior =
2
3
. Therefore,
4
9
+
8
27
+
16
81
+
32
243
+ =
4
9
1
2
3
=
4
3
.
Findthesum

n=2
_
2
e
_
n
.
31. Findthesum

n=1
2
n+3
3
n
.
solution Note
n=1
2
n+3
3
n
= 2
3

n=1
2
n
3
n
= 8

n=1
_
2
3
_
n
;
therefore,
n=1
2
n+3
3
n
= 8
3
2
1
1
2
3
= 36.
Showthat

n=1
_
b tan
1
n
2
_
divergesif b =

2
.
33. Giveanexampleof divergent series

n=1
a
n
and

n=1
b
n
suchthat

n=1
(a
n
+b
n
) = 1.
solution Let a
n
=
_
1
2
_
n
+1, b
n
= 1. ThecorrespondingseriesdivergebytheDivergenceTest; however,
n=1
(a
n
+b
n
) =
n=1
_
1
2
_
n
=
1
2
1
1
2
= 1.
Let S =
n=1
_
1
n
1
n +2
_
. ComputeS
N
for N = 1, 2, 3, 4. FindS byshowingthat
S
N
=
3
2
1
N +1
1
N +2
35. EvaluateS =
n=3
1
n(n +3)
.
solution Notethat
1
n(n +3)
=
1
3
_
1
n
1
n +3
_
sothat
N
n=3
1
n(n +3)
=
1
3
N
n=3
_
1
n
1
n +3
_
=
1
3
__
1
3
1
6
_
+
_
1
4
1
7
_
+
_
1
5
1
8
_
_
1
6
1
9
_
+ +
_
1
N 1
1
N +2
_
+
_
1
N
1
N +3
__
=
1
3
_
1
3
+
1
4
+
1
5
1
N +1
1
N +2
1
N +3
_
Thus
n=3
1
n(n +3)
=
1
3
lim
N
N
n=3
_
1
n
1
n +3
_
=
1
3
_
1
3
+
1
4
+
1
5
1
N +1
1
N +2
1
N +3
_
=
1
3
_
1
3
+
1
4
+
1
5
_
=
47
180
Findthetotal areaof theinnitelymanycirclesontheinterval [0, 1] inFigure1.
In Exercises 3740, use the Integral Test to determine whether the innite series converges.
37.

n=1
n
2
n
3
+1
x
2
x
3
+1
. Thisfunctioniscontinuousandpositivefor x 1. Because
f
(x) =
(x
3
+1)(2x) x
2
(3x
2
)
(x
3
+1)
2
=
x(2x
3
)
(x
3
+1)
2
,
weseethatf
(x) < 0andf isdecreasingontheinterval x 2. Therefore, theIntegral Testappliesontheinterval x 2.

Now,
_

2
x
2
x
3
+1
dx = lim
R
_
R
2
x
2
x
3
+1
dx =
1
3
lim
R
_
ln(R
3
+1) ln9
_
= .

n=2
n
2
n
3
+1
diverges, asdoestheseries

n=1
n
2
n
3
+1
.
n=1
n
2
(n
3
+1)
1.01
39.

n=1
1
(n +2)(ln(n +2))
3
1
(x+2) ln
3
(x+2)
. Usingthesubstitutionu = ln(x +2), sothat du =
1
x+2
dx, wehave
_

0
f (x) dx =
_

ln2
1
u
3
du = lim
R
_

ln2
1
u
3
du = lim
R
_
1
2u
2
R
ln2
_
= lim
R
_
1
2(ln2)
2
1
2(lnR)
2
_
=
1
2(ln2)
2
Sincetheintegral of f (x) converges, sodoestheseries.
n=1
n
3
e
n
4
In Exercises 4148, use the Comparison or Limit Comparison Test to determine whether the innite series converges.
41.

n=1
1
(n +1)
2
solution For all n 1,
0<
1
n +1
<
1
n
so
1
(n +1)
2
<
1
n
2
.
Theseries

n=1
1
n
2
isaconvergent p-series, sotheseries

n=1
1
(n +1)
2
convergesbytheComparisonTest.
n=1
1
n +n
43.

n=2
n
2
+1
n
3.5
2
n
=
n
2
+1
n
3.5
2
andb
n
=
1
n
1.5
. Now,
L = lim
n
n
2
+1
n
3.5
2
1
n
1.5
= lim
n
n
3.5
+n
1.5
n
3.5
2
= 1.
Because L exists and

n=1
1
n
1.5
is a convergent p-series, we conclude by the Limit Comparison Test that the series
n=2
n
2
+1
n
3.5
2
alsoconverges.
n=1
1
n lnn
45.

n=2
n
_
n
5
+5
solution For all n 2,
n
_
n
5
+5
<
n
n
5/2
=
1
n
3/2
.
Theseries

n=2
1
n
3/2

n=2
n
_
n
5
+5
n=1
1
3
n
2
n
47.

n=1
n
10
+10
n
n
11
+11
n
n
=
n
10
+10
n
n
11
+11
n
andb
n
=
_
10
11
_
n
. Then,
L = lim
n
a
n
b
n
= lim
n
n
10
+10
n
n
11
+11
n
_
10
11
_
n
= lim
n
n
10
+10
n
10
n
n
11
+11
n
11
n
= lim
n
n
10
10
n
+1
n
11
11
n
+1
= 1.
The series

n=1
_
10
11
_
n
is a convergent geometric series; because L exists, we may therefore conclude by the Limit
ComparisonTest that theseries

n=1
n
10
+10
n
n
11
+11
n
alsoconverges.
n=1
n
20
+21
n
n
21
+20
n
49. Determinetheconvergenceof

n=1
2
n
+n
3
n
2
usingtheLimit ComparisonTest withb
n
=
_
2
3
_
n
.
solution Witha
n
=
2
n
+n
3
n
2
, wehave
L = lim
n
a
n
b
n
= lim
n
2
n
+n
3
n
2
3
n
2
n
= lim
n
6
n
+n3
n
6
n
2
n+1
= lim
n
1+n
_
1
2
_
n
12
_
1
3
_
n
= 1
SinceL = 1, thetwoserieseither bothconvergeor bothdiverge. Since

n=1
_
2
3
_
n
isaconvergent geometricseries, the
Limit ComparisonTest tellsusthat

n=1
2
n
+n
3
n
2
alsoconverges.
Determinetheconvergenceof

n=1
lnn
1.5
n
usingtheLimit ComparisonTest withb
n
=
1
1.4
n
.
51. Let a
n
= 1
_
1
1
n
. Showthat lim
n
a
n
= 0andthat

n=1
a
n
diverges. Hint: Showthat a
n

1
2n
.
solution
1
_
1
1
n
= 1
_
n 1
n
=
n 1
n
=
n (n 1)
n(
n +
n 1)
=
1
n +
_
n
2
n
1
n +
n
2
=
1
2n
.
Theseries

n=2
1
2n
diverges, sotheseries
n=2
_
1
_
1
1
n
_
alsodivergesbytheComparisonTest.
Determinewhether

n=2
_
1
_
1
1
n
2
_
converges.
53. Let S =
n=1
n
(n
2
+1)
2
.
(a) Showthat S converges.
(b) UseEq. (4) inExercise83of Section11.3withM = 99toapproximateS. Whatisthemaximumsizeof the
error?
solution
(a) For n 1,
n
(n
2
+1)
2
<
n
(n
2
)
2
=
1
n
3
.
Theseries

n=1
1
n
3

n=1
n
(n
2
+1)
2
(b) Witha
n
=
n
(n
2
+1)
2
, f (x) =
x
(x
2
+1)
2
andM = 99, Eq. (4) inExercise83of Section11.3becomes
99
n=1
n
(n
2
+1)
2
+
_

100
x
(x
2
+1)
2
dx S
100
n=1
n
(n
2
+1)
2
+
_

100
x
(x
2
+1)
2
dx,
or
0 S
99
n=1
n
(n
2
+1)
2
+
_

100
x
(x
2
+1)
2
dx
100
(100
2
+1)
2
.
Now,
99
n=1
n
(n
2
+1)
2
= 0.397066274; and
_

100
x
(x
2
+1)
2
dx = lim
R
_
R
100
x
(x
2
+1)
2
dx =
1
2
lim
R
_
1
R
2
+1
+
1
100
2
+1
_
=
1
20002
= 0.000049995;
thus,
S 0.397066274+0.000049995= 0.397116269.
Theboundontheerror inthisapproximationis
100
(100
2
+1)
2
= 9.99810
7
.
In Exercises 5457, determine whether the series converges absolutely. If it does not, determine whether it converges
conditionally.
n=1
(1)
n
3
n +2n
55.

n=1
(1)
n
n
1.1
ln(n +1)
solution Consider thecorrespondingpositiveseries

n=1
1
n
1.1
ln(n +1)
. Because
1
n
1.1
ln(n +1)
<
1
n
1.1
and

n=1
1
n
1.1
isaconvergent p-series, wecanconcludebytheComparisonTest that

n=1
(1)
n
n
1.1
ln(n +1)
alsoconverges.
Thus,

n=1
(1)
n
n
1.1
ln(n +1)
n=1
cos
_
4
+n
_
n
57.

n=1
cos
_
4
+2n
_
n
solution cos
_
4
+2n
_
= cos
4
=
2
2
, so
n=1
cos
_
4
+2n
_
n
=
2
2
n=1
1
n
.
Thisisadivergent p-series, sotheseries

n=1
cos
_
4
+2n
_
n
diverges.
Useacomputer algebrasystemtoapproximate

n=1
(1)
n
n
3
+
n
towithinanerror of at most 10
5
.
59. Catalansconstant isdenedbyK =
k=0
(1)
k
(2k +1)
2
.
(a) Howmanytermsof theseriesareneededtocalculateK withanerror of lessthan10
6
?
(b) Carryout thecalculation.
solution Usingtheerror boundfor analternatingseries, wehave
|S
N
K|
1
(2(N +1) +1)
2
=
1
(2N +3)
2
.
For accuracytothreedecimal places, wemust chooseN sothat
1
(2N +3)
2
< 510
3
or (2N +3)
2
> 2000.
Solvingfor N yields
N >
1
2
_
20003
_
20.9.
Thus,
K
21
k=0
(1)
k
(2k +1)
2
= 0.915707728.
Giveanexampleof conditionally convergent series

n=1
a
n
and

n=1
b
n
suchthat

n=1
(a
n
+b
n
) convergesabso-
lutely.
61. Let

n=1
a
n
beanabsolutelyconvergentseries. Determinewhether thefollowingseriesareconvergentor divergent:
(a)

n=1
_
a
n
+
1
n
2
_
(b)

n=1
(1)
n
a
n
(c)

n=1
1
1+a
2
n
(d)

n=1
|a
n
|
n
solution Because

n=1
a
n
convergesabsolutely, weknowthat

n=1
a
n
convergesandthat

n=1
|a
n
| converges.
(a) Becauseweknowthat

n=1
a
n
convergesandtheseries

n=1
1
n
2
isaconvergent p-series, thesumof thesetwoseries,
n=1
_
a
n
+
1
n
2
_
alsoconverges.
(b) Wehave,
n=1
(1)
n
a
n
n=1
|a
n
|
Because

n=1
|a
n
| converges, itfollowsthat

n=1
(1)
n
a
n
convergesabsolutely, whichimpliesthat

n=1
(1)
n
a
n
converges.
(c) Because

n=1
a
n
converges, lim
n
a
n
= 0. Therefore,
lim
n
1
1+a
2
n
=
1
1+0
2
= 1= 0,
andtheseries

n=1
1
1+a
2
n
divergesbytheDivergenceTest.
(d)
|a
n
|
n
|a
n
| andtheseries

n=1
|a
n
| converges, sotheseries

n=1
|a
n
|
n
Let {a
n
} beapositivesequencesuchthat lim
n
n
a
n
=
1
2
. Determinewhether thefollowingseriesconvergeor
diverge:
(a)

n=1
2a
n
(b)

n=1
3
n
a
n
(c)

n=1
a
n
In Exercises 6370, apply the Ratio Test to determine convergence or divergence, or state that the Ratio Test is inconclusive.
63.

n=1
n
5
5
n
solution Witha
n
=
n
5
5
n
,
a
n+1
a
n
=
(n +1)
5
5
n+1

5
n
n
5
=
1
5
_
1+
1
n
_
5
,
and
= lim
n
a
n+1
a
n
=
1
5
lim
n
_
1+
1
n
_
5
=
1
5
1=
1
5
.
Because < 1, theseriesconvergesbytheRatioTest.
n=1
n +1
n
8
65.

n=1
1
n2
n
+n
3
solution Witha
n
=
1
n2
n
+n
3
,
a
n+1
a
n
=
n2
n
+n
3
(n +1)2
n+1
+(n +1)
3
=
n2
n
_
1+
n
2
2
n
_
(n +1)2
n+1
_
1+
(n+1)
2
2
n+1
_ =
1
2
n
n +1
1+
n
2
2
n
1+
(n+1)
2
2
n+1
,
and
= lim
n
a
n+1
a
n
=
1
2
1 1=
1
2
.
Because < 1, theseriesconvergesbytheRatioTest.
n=1
n
4
n!
67.

n=1
2
n
2
n!
solution Witha
n
=
2
n
2
n!
,
a
n+1
a
n
=
2
(n+1)
2
(n +1)!

n!
2
n
2
=
2
2n+1
n +1
and = lim
n
a
n+1
a
n
= .
Because > 1, theseriesdivergesbytheRatioTest.
n=4
lnn
n
3/2
69.

n=1
_
n
2
_
n
1
n!
solution Witha
n
=
_
n
2
_
n
1
n!
,
a
n+1
a
n
=
_
n +1
2
_
n+1
1
(n +1)!

_
2
n
_
n
n! =
1
2
_
n +1
n
_
n
=
1
2
_
1+
1
n
_
n
,
and
= lim
n
a
n+1
a
n
=
1
2
e.
Because =
e
2
> 1, theseriesdivergesbytheRatioTest.
n=1
_
n
4
_
n
1
n!
In Exercises 7174, apply the Root Test to determine convergence or divergence, or state that the Root Test is inconclusive.
71.

n=1
1
4
n
solution Witha
n
=
1
4
n
,
L = lim
n
n
a
n
= lim
n
n
_
1
4
n
=
1
4
.
BecauseL < 1, theseriesconvergesbytheRoot Test.
n=1
_
2
n
_
n
73.

n=1
_
3
4n
_
n
solution Witha
n
=
_
3
4n
_
n
,
L = lim
n
n
a
n
= lim
n
n
_
_
3
4n
_
n
= lim
n
3
4n
= 0.
BecauseL < 1, theseriesconvergesbytheRoot Test.
n=1
_
cos
1
n
_
n
3
In Exercises 7592, determine convergence or divergence using any method covered in the text.
75.

n=1
_
2
3
_
n
solution Thisisageometricserieswithratior =
2
3
< 1; hence, theseriesconverges.
n=1
7n
e
8n
77.

n=1
e
0.02n
solution Thisisageometricserieswithcommonratior =
1
e
0.02
0.98< 1; hence, theseriesconverges.
n=1
ne
0.02n
79.

n=1
(1)
n1
n +
n +1
solution Inthisalternatingseries, a
n
=
1
n+
n+1
. Thesequence{a
n
} isdecreasing, and
lim
n
a
n
= 0;
thereforetheseriesconvergesbytheLeibnizTest.
n=10
1
n(lnn)
3/2
81.

n=2
(1)
n
lnn
solution Thesequencea
n
=
1
lnn
isdecreasingfor n 10and
lim
n
a
n
= 0;
therefore, theseriesconvergesbytheLeibnizTest.
n=1
e
n
n!
83.

n=1
1
n
n +lnn
solution For n 1,
1
n
n +lnn
1
n
n
=
1
n
3/2
.
Theseries

n=1
1
n
3/2

n=1
1
n
n +lnn
n=1
1
3
n(1+
n)
85.

n=1
_
1
n +1
_
solution Thisseriestelescopes:
n=1
_
1
n +1
_
=
_
1
1
2
_
+
_
1
3
_
+
_
1
4
_
+. . .
sothat then
th
partial sumS
n
is
S
n
=
_
1
1
2
_
+
_
1
3
_
+
_
1
4
_
+ +
_
1
n +1
_
= 1
1
n +1
andthen
n=1
_
1
n +1
_
= lim
n
S
n
= 1 lim
n
1
n +1
= 1
n=1
_
lnn ln(n +1)
_ 87.

n=1
1
n +
n
solution For n 1,
n n, sothat
n=1
1
n +
n=1
1
2n
whichdivergessinceitisaconstantmultipleof theharmonicseries.Thus

n=1
1
n +
n
divergesaswell,bytheComparison
Test.
n=2
cos(n)
n
2/3
89.

n=2
1
n
lnn
solution For n N largeenough, lnn 2sothat
n=N
1
n
lnn

n=N
1
n
2
whichis aconvergent p-series. Thus by theComparisonTest,

n=N
1
n
lnn
also converges; addingback intheterms for
n < N doesnot affect convergence.
n=2
1
ln
3
n
91.

n=1
sin
2

n
solution For all x > 0, sinx < x. Therefore, sin
2
x < x
2
, andfor x =

n
,
sin
2

n
<
2
n
2
=
2
1
n
2
.
Theseries

n=1
1
n
2

n=1
sin
2

n
n=0
2
2n
n!
In Exercises 9398, nd the interval of convergence of the power series.
93.

n=0
2
n
x
n
n!
solution Witha
n
=
2
n
x
n
n!
,
= lim
n
a
n+1
a
n
= lim
n
2
n+1
x
n+1
(n +1)!

n!
2
n
x
n
= lim
n
x
2
n
= 0
Then < 1for all x, sothat theradiusof convergenceisR = , andtheseriesconvergesfor all x.
n=0
x
n
n +1
95.

n=0
n
6
n
8
+1
(x 3)
n
solution Witha
n
=
n
6
(x3)
n
n
8
+1
,
= lim
n
a
n+1
a
n
= lim
n
(n +1)
6
(x 3)
n+1
(n +1)
8
1

n
8
+1
n
6
(x 3)
n
= lim
n
(x 3)
(n +1)
6
(n
8
+1)
n
6
((n +1)
8
+1)
= lim
n
(x 3)
n
14
+termsof lower degree
n
14
+termsof lower degree
= |x 3|
Then < 1when|x 3| < 1, sotheradiusof convergenceis1, andtheseriesconvergesabsolutelyfor |x 3| < 1, or
2< x < 4. For theendpointx = 4, theseriesbecomes

n=0
n
6
n
8
+1
, whichconvergesbytheComparisonTestcomparing
withtheconvergent p-series

n=1
1
n
2
. For theendpoint x = 2, theseriesbecomes

n=0
n
6
(1)
n
n
8
+1
, whichconvergesbythe
LeibnizTest. Theseries

n=0
n
6
(x 3)
n
n
8
+1
thereforeconvergesfor 2 x 4.
n=0
nx
n
97.

n=0
(nx)
n
solution Witha
n
= n
n
x
n
, andassumingx = 0,
= lim
n
a
n+1
a
n
= lim
n
(n +1)
n+1
x
n+1
n
n
x
n
= lim
n
x(n +1)
_
n +1
n
_
n
=
since
_
n+1
n
_
n
=
_
1+
1
n
_
n
converges toe andthe(n + 1) termdiverges to. Thus < 1only whenx = 0, sothe
seriesconvergesonlyfor x = 0.
n=0
(2x 3)
n
n lnn
99. Expandf (x) =
2
43x
asapowerseriescenteredatc = 0.Determinethevaluesof x forwhichtheseriesconverges.
solution Write
2
43x
=
1
2
1
1
3
4
x
.
Substituting
3
4
x for x intheMaclaurinseriesfor
1
1x
, weobtain
1
1
3
4
x
=
n=0
_
3
4
_
n
x
n
.
Thisseriesconvergesfor
3
4
x
< 1, or |x| <

4
3
. Hence, for |x| <
4
3
,
2
43x
=
1
2
n=0
_
3
4
_
n
x
n
.
Provethat
n=0
ne
nx
=
e
x
(1e
x
)
2
Hint: Expresstheleft-handsideasthederivativeof ageometricseries.
101. Let F(x) =
k=0
x
2k
2
k
k!
.
(a) Showthat F(x) hasinniteradiusof convergence.
(b) Showthat y = F(x) isasolutionof
y
= xy
+y, y(0) = 1, y
(0) = 0
(c) Plot thepartial sumsS
N
for N = 1, 3, 5, 7onthesameset of axes.
solution
(a) Witha
k
=
x
2k
2
k
k!
,
a
k+1
a
k
=
|x|
2k+2
2
k+1
(k +1)!

2
k
k!
|x|
2k
=
x
2
2(k +1)
,
and
= lim
k
a
k+1
a
k
= x
2
0= 0.
Because < 1for all x, weconcludethat theseriesconvergesfor all x; that is, R = .
(b) Let
y = F(x) =
k=0
x
2k
2
k
k!
.
Then
y
k=1
2kx
2k1
2
k
k!
=
k=1
x
2k1
2
k1
(k 1)!
,
y
k=1
(2k 1)x
2k2
2
k1
(k 1)!
,
and
xy
+y = x
k=1
x
2k1
2
k1
(k 1)!
+
k=0
x
2k
2
k
k!
=
k=1
x
2k
2
k1
(k 1)!
+1+
k=1
x
2k
2
k
k!
= 1+
k=1
(2k +1)x
2k
2
k
k!
=
k=0
(2k +1)x
2k
2
k
k!
=
k=1
(2k 1)x
2k2
2
k1
(k 1)!
= y
.
Moreover,
y(0) = 1+
k=1
0
2k
2
k
k!
= 1 and y
(0) =
k=1
0
2k1
2
k1
(k 1)!
= 0.
Thus,

k=0
x
2k
2
k
k!
isthesolutiontotheequationy
= xy
+y satisfyingy(0) = 1, y
(0) = 0.
(c) Thepartial sumsS
1
, S
3
, S
5
andS
7
areplottedinthegurebelow.
y
x
1 2 1
1
2
3
4
5
6
7
2
Findapower seriesP(x) =
n=0
a
n
x
n
that satisestheLaguerredifferential equation
xy
+(1x)y
y = 0
withinitial conditionsatisfyingP(0) = 1.
In Exercises 103112, nd the Taylor series centered at c.
103. f (x) = e
4x
, c = 0
solution Substituting4x for x intheMaclaurinseriesfor e
x
yields
e
4x
=
n=0
(4x)
n
n!
=
n=0
4
n
n!
x
n
.
f (x) = e
2x
, c = 1
105. f (x) = x
4
, c = 2
solution Wehave
f
(x) = 4x
3
f
(x) = 12x
2
f
(x) = 24x f
(4)
(x) = 24
andall higher derivativesarezero, sothat
f (2) = 2
4
= 16 f
(2) = 4 2
3
= 32 f
(2) = 12 2
2
= 48 f
(2) = 24 2= 48 f
(4)
(2) = 24
ThustheTaylor seriescenteredat c = 2is
4
n=0
f
(n)
(2)
n!
(x 2)
n
= 16+
32
1!
(x 2) +
48
2!
(x 2)
2
+
48
3!
(x 2)
3
+
24
4!
(x 2)
4
= 16+32(x 2) +24(x 2)
2
+8(x 2)
3
+(x 2)
4
f (x) = x
3
x, c = 2
107. f (x) = sinx, c =
solution Wehave
f
(4n)
(x) = sinx f
(4n+1)
(x) = cosx f
(4n+2)
(x) = sinx f
(4n+3)
(x) = cosx
sothat
f
(4n)
() = sin = 0 f
(4n+1)
() = cos = 1 f
(4n+2)
() = sin = 0 f
(4n+3)
() = cos = 1
ThentheTaylor seriescenteredat c = is
n=0
f
(n)
()
n!
(x )
n
=
1
1!
(x ) +
1
3!
(x )
3
+
1
5!
(x )
5
+
1
7!
(x )
7
. . .
= (x ) +
1
6
(x )
3
1
120
(x )
5
+
1
5040
(x )
7
. . .
f (x) = e
x1
, c = 1
109. f (x) =
1
12x
, c = 2
solution Write
1
12x
=
1
52(x +2)
=
1
5
1
1
2
5
(x +2)
.
Substituting
2
5
(x +2) for x intheMaclaurinseriesfor
1
1x
yields
1
1
2
5
(x +2)
=
n=0
2
n
5
n
(x +2)
n
;
hence,
1
12x
=
1
5
n=0
2
n
5
n
(x +2)
n
=
n=0
2
n
5
n+1
(x +2)
n
.
f (x) =
1
(12x)
2
, c = 2
111. f (x) = ln
x
2
, c = 2
solution Write
ln
x
2
= ln
_
(x 2) +2
2
_
= ln
_
1+
x 2
2
_
.
Substituting
x2
2
for x intheMaclaurinseriesfor ln(1+x) yields
ln
x
2
=
n=1
(1)
n+1
_
x2
2
_
n
n
=
n=1
(1)
n+1
(x 2)
n
n 2
n
.
f (x) = x ln
_
1+
x
2
_
, c = 0
In Exercises 113116, nd the rst three terms of the Maclaurin series of f (x) and use it to calculate f
(3)
(0).
113. f (x) = (x
2
x)e
x
2
solution Substitutex
2
x
toget
e
x
2
= 1+x
2
+
1
2
x
4
+
1
6
x
6
+. . .
sothat theMaclaurinseriesfor f (x) is
(x
2
x)e
x
2
= x
2
+x
4
+
1
2
x
6
+ x x
3
1
2
x
5
= x +x
2
x
3
+x
4
+. . .
Thecoefcient of x
3
is
f
(0)
3!
= 1
sothat f
(0) = 6.
f (x) = tan
1
(x
2
x)
115. f (x) =
1
1+tanx
solution Substitutetanx intheMaclaurinseriesfor
1
1x
toget
1
1+tanx
= 1tanx +(tanx)
2
(tanx)
3
+. . .
Wehavenot yet encounteredtheMaclaurinseriesfor tanx. Weneedonlythetermsupthroughx
3
, socompute
tan
(x) = sec
2
x tan
(x) = 2(tanx) sec

2
x tan
(x) = 2(1+tan
2
x) sec
2
x +4(tan
2
x) sec
2
x
sothat
tan
(0) = 1 tan
(0) = 0 tan
(0) = 2
ThentheMaclaurinseriesfor tanx is
tanx = tan0+
tan
(0)
1!
x +
tan
(0)
2!
x
2
+
tan
(0)
3!
x
3
+ = x +
1
3
x
3
+. . .
Substitutetheseintotheseriesabovetoget
1
1+tanx
= 1
_
x +
1
3
x
3
_
+
_
x +
1
3
x
3
_
2
_
x +
1
3
x
3
_
3
+. . .
= 1x
1
3
x
3
+x
2
x
3
+higher degreeterms
= 1x +x
2
4
3
x
3
+higher degreeterms
Thecoefcient of x
3
is
f
(0)
3!
=
4
3
sothat
f
(0) = 6
4
3
= 8
f (x) = (sinx)
1+x 117. Calculate
2

3
2
3
3!
+

5
2
5
5!

7
2
7
7!
+ .
solution Werecognizethat
2

3
2
3
3!
+

5
2
5
5!

7
2
7
7!
+ =
n=0
(1)
n
(/2)
2n+1
(2n +1)!
istheMaclaurinseriesfor sinx withx replacedby/2. Therefore,
2

3
2
3
3!
+

5
2
5
5!

7
2
7
7!
+ = sin
2
= 1.
FindtheMaclaurinseriesof thefunctionF(x) =
_
x
0
e
t
1
t
dt .
12 PARAMETRIC EQUATIONS,
POLAR COORDINATES,
AND CONIC SECTIONS
12.1 Parametric Equations
1. Describetheshapeof thecurvex = 3cost, y = 3sint .
solution For all t ,
x
2
+y
2
= (3cost )
2
+(3sint )
2
= 9(cos
2
t +sin
2
t ) = 9 1= 9,
thereforethecurveis onthecirclex
2
+ y
2
= 9. Also, eachpoint onthecirclex
2
+ y
2
= 9canberepresentedinthe
form(3cost, 3sint ) for somevalueof t . Weconcludethat thecurvex = 3cost , y = 3sint is thecircleof radius 3
centeredat theorigin.
2. Howdoesx = 4+3cost, y = 5+3sint differ fromthecurveinthepreviousquestion?
solution Inthiscasewehave
(x 4)
2
+(y 5)
2
= (3cost )
2
+(3sint )
2
= 9(cos
2
t +sin
2
t ) = 9 1= 9
Therefore, thegivenequationsparametrizethecircleof radius3centeredat thepoint (4, 5).
3. What isthemaximumheight of aparticlewhosepathhasparametricequationsx = t
9
, y = 4t
2
?
solution Theparticles height is y = 4 t
2
. To ndthemaximumheight weset thederivativeequal to zero and
solve:
dy
dt
=
d
dt
(4t
2
) = 2t = 0 or t = 0
Themaximumheight isy(0) = 40
2
= 4.
4. Cantheparametriccurve(t, sint ) berepresentedasagraphy = f (x)?What about (sint, t )?
solution Intheparametriccurve(t, sint ) wehavex = t andy = sint , therefore, y = sinx. Thatis, thecurvecanbe
representedasagraphof afunction. Intheparametriccurve(sint, t ) wehavex = sint , y = t , thereforex = siny. This
equationdoesnot deney asafunctionof x, thereforetheparametriccurve(sint, t ) cannot berepresentedasagraphof
afunctiony = f (x).
5. Matchthederivativeswithaverbal description:
(a)
dx
dt
(b)
dy
dt
(c)
dy
dx
(i) Slopeof thetangent linetothecurve
(ii) Vertical rateof changewithrespect totime
(iii) Horizontal rateof changewithrespect totime
solution
(a) Thederivative
dx
dt
isthehorizontal rateof changewithrespect totime.
(b) Thederivative
dy
dt
isthevertical rateof changewithrespect totime.
(c) Thederivative
dy
dx
istheslopeof thetangent linetothecurve.
Hence, (a) (iii), (b) (ii), (c) (i)
742
S E C T I ON 12.1 Parametric Equations 743
Exercises
1. Findthecoordinatesat timest = 0, 2, 4of aparticlefollowingthepathx = 1+t
3
, y = 93t
2
.
solution Substitutingt = 0, t = 2, andt = 4intox = 1+ t
3
, y = 9 3t
2
givesthecoordinatesof theparticleat
thesetimesrespectively. That is,
(t = 0) x = 1+0
3
= 1, y = 93 0
2
= 9 (1, 9)
(t = 2) x = 1+2
3
= 9, y = 93 2
2
= 3 (9, 3)
(t = 4) x = 1+4
3
= 65, y = 93 4
2
= 39 (65, 39).
Findthecoordinatesat t = 0,

4
, of aparticlemovingalongthepathc(t ) = (cos2t, sin
2
t ).
3. Showthat thepathtracedbythebullet inExample3isaparabolabyeliminatingtheparameter.
solution Thepathtracedbythebullet isgivenbythefollowingparametricequations:
x = 200t, y = 400t 16t
2
Weeliminatetheparameter. Sincex = 200t , wehavet =
x
200
. Substitutingintotheequationfor y weobtain:
y = 400t 16t
2
= 400
x
200
16
_
x
200
_
2
= 2x
x
2
2500
Theequationy =
x
2
2500
+2x istheequationof aparabola.
Usethetableof valuestosketchtheparametriccurve(x(t ), y(t )), indicatingthedirectionof motion.
t 3 2 1 0 1 2 3
x 15 0 3 0 3 0 15
y 5 0 3 4 3 0 5
5. Graphtheparametriccurves. Includearrowsindicatingthedirectionof motion.
(a) (t, t ), < t < (b) (sint, sint ), 0 t 2
(c) (e
t
, e
t
), < t < (d) (t
3
, t
3
), 1 t 1
solution
(a) For thetrajectory c(t ) = (t, t ), < t < wehavey = x. Also thetwo coordinates tendto and as
t andt respectively. Thegraphisshownnext:
x
y
(b) For thecurvec(t ) = (sint, sint ), 0 t 2, wehavey = x. sint is increasingfor 0 t

2
, decreasingfor
2
t
3
2
andincreasingagainfor
3
2
t 2. Hencetheparticlemovesfromc(0) = (0, 0) toc(
2
) = (1, 1), then
movesbacktoc(
3
2
) = (1, 1) andthenreturnstoc(2) = (0, 0). Weobtainthefollowingtrajectory:
x
y
t = (1,1)
2
t = 0
x
y
t = (1,1)
2
t = (1,1)
3
2
x
y
t = (1,1)
3
2
0< t

2

2
t
3
2
3
2
t < 2
744 C HA P T E R 12 PARAMETRIC EQUATIONS, POLAR COORDINATES, AND CONIC SECTIONS
Thesethreepartsof thetrajectoryareshowntogether inthenext gure:
x
y
t = (1,1)
3
2
t = (1,1)
2
t = 0
t = 2
(c) For thetrajectoryc(t ) = (e
t
, e
t
), < t < , wehavey = x. However since lim
t
e
t
= 0and lim
t
e
t
= ,
thetrajectoryisthepart of theliney = x, 0< x.
x
y
(d) For thetrajectoryc(t ) = (t
3
, t
3
), 1 t 1, wehaveagainy = x. Sincethefunctiont
3
isincreasingtheparticle
movesinonedirectionstartingat ((1)
3
, (1)
3
) = (1, 1) andendingat (1
3
, 1
3
) = (1, 1). Thetrajectory isshown
next:
x
y
t = 1(1,1)
t = 1 (1,1)
Givetwodifferent parametrizationsof thelinethrough(4, 1) withslope2.
In Exercises 714, express in the form y = f (x) by eliminating the parameter.
7. x = t +3, y = 4t
solution Weeliminatetheparameter. Sincex = t +3, wehavet = x 3. Substitutingintoy = 4t weobtain
y = 4t = 4(x 3) y = 4x 12
x = t
1
, y = t
2
9. x = t , y = tan
1
(t
3
+e
t
)
solution Replacingt byx intheequationfor y weobtainy = tan
1
(x
3
+e
x
).
x = t
2
, y = t
3
+1
11. x = e
2t
, y = 6e
4t
solution Weeliminatetheparameter. Sincex = e
2t
, wehave2t = lnx or t =
1
2
lnx. Substitutinginy = 6e
4t
weget
y = 6e
4t
= 6e
4(
1
2
lnx)
= 6e
2lnx
= 6e
lnx
2
= 6x
2
y =
6
x
2
, x > 0.
x = 1+t
1
, y = t
2
13. x = lnt , y = 2t
solution Sincex = lnt wehavet = e
x
. Substitutinginy = 2t weobtainy = 2e
x
.
x = cost , y = tant
In Exercises 1518, graph the curve and draw an arrow specifying the direction corresponding to motion.
15. x =
1
2
t , y = 2t
2
solution Let c(t ) = (x(t ), y(t )) = (
1
2
t, 2t
2
). Thenc(t ) = (x(t ), y(t )) sothecurveissymmetricwithrespect to
they-axis.Also, thefunction
1
2
t isincreasing. Hencethereisonlyonedirectionof motiononthecurve. Thecorresponding
functionistheparabolay = 2 (2x)
2
= 8x
2
. Weobtainthefollowingtrajectory:
x
y
t = 0
x = 2+4t , y = 3+2t
17. x = t , y = sint
solution Wendthefunctionbyeliminatingt . Sincex = t , wehavet =
x
. Substitutingt =
x
intoy = sint we
get y = sin
x
. Weobtainthefollowingcurve:
x
y
(4
2
,0)
(2
2
,0)
x = t
2
, y = t
3
19. Matchtheparametrizations(a)(d) belowwiththeir plotsinFigure14, anddrawanarrowindicatingthedirectionof
motion.
2
x x
y y
1 5 5 5
(II) (III) (I)
x x
10
20
1
5
y y
(IV)
FIGURE 14
(a) c(t ) = (sint, t ) (b) c(t ) = (t
2
9, 8t t
3
)
(c) c(t ) = (1t, t
2
9) (d) c(t ) = (4t +2, 53t )
solution
(a) Inthecurvec(t ) = (sint, t ) thex-coordinateisvaryingbetween1and1sothiscurvecorrespondstoplot IV. As
t increases, they-coordinatey = t isdecreasingsothedirectionof motionisdownward.
x
y
1
2
2
1
(IV) c(t ) = (sint, t )
(b) Thecurvec(t ) = (t
2
9, t
3
8) intersects thex-axis wherey = t
3
8 = 0, or t = 2. Thex-intercept is
(5, 0). They-interceptsareobtainedwherex = t
2
9= 0, or t = 3. They-interceptsare(0, 35) and(0, 19). As
t increasesfromto0, x andy decrease, andast increasesfrom0to, x increasesandy decreases. Weobtainthe
followingtrajectory:
x
y
t = 0, (9,8) 5
19
(II)
(c) Thecurvec(t ) = (1t, t
2
9) intersectsthey-axiswherex = 1t = 0, or t = 1. They-interceptis(0, 8). The
x-interceptsareobtainedwheret
2
9= 0or t = 3. Thesearethepoints(2, 0) and(4, 0). Settingt = 1x weget
y = t
2
9= (1x)
2
9= x
2
2x 8.
Ast increasesthex coordinatedecreasesandweobtainthefollowingtrajectory:
x
y
2 4 5
10
(III)
(d) Thecurvec(t ) = (4t +2, 53t ) isastraight line, sinceeliminatingt inx = 4t +2andsubstitutinginy = 53t
givesy = 53
x2
4
=
3
4
x +
13
2
whichistheequationof aline.Ast increases, thex coordinatex = 4t +2increases
andthey-coordinatey = 53t decreases. Weobtainthefollowingtrajectory:
x
y
5
5
(I)
A particlefollowsthetrajectory
x(t ) =
1
4
t
3
+2t, y(t ) = 20t t
2
witht insecondsanddistanceincentimeters.
(a) What istheparticlesmaximumheight?
(b) Whendoestheparticlehit thegroundandhowfar fromtheorigindoesit land?
21. Findaninterval of t -valuessuchthat c(t ) = (cost, sint ) tracesthelower half of theunit circle.
solution For t = , wehavec() = (1, 0). Ast increasesfrom to2, thex-coordinateof c(t ) increasesfrom
1to 1, andthey-coordinatedecreases from0to 1(at t = 3/2) andthenreturns to 0. Thus, for t in[, 2], the
equationtracesthelower part of thecircle.
Findaninterval of t -valuessuchthat c(t ) = (2t +1, 4t 5) parametrizesthesegment from(0, 7) to(7, 7).
In Exercises 2338, nd parametric equations for the given curve.
23. y = 94x
solution ThisisalinethroughP = (0, 9) withslopem = 4. Usingtheparametricrepresentationof aline, asgiven
inExample3, weobtainc(t ) = (t, 94t ).
y = 8x
2
3x
25. 4x y
2
= 5
solution We dene the parameter t = y. Then, x =
5+y
2
4
=
5+t
2
4
, giving us the parametrization c(t ) =
_
5+t
2
4
, t
_
.
x
2
+y
2
= 49
27. (x +9)
2
+(y 4)
2
= 49
solution This is a circle of radius 7 centered at (9, 4). Using the parametric representation of a circle we get
c(t ) = (9+7cost, 4+7sint ).
_
x
5
_
2
+
_
y
12
_
2
= 1
29. Lineof slope8through(4, 9)
solution Usingtheparametricrepresentationof alinegiveninExample3, weget theparametrizationc(t ) = (4+
t, 9+8t ).
Linethrough(2, 5) perpendicular toy = 3x
31. Linethrough(3, 1) and(5, 4)
solution Weusethetwo-point parametrizationof alinewithP = (a, b) = (3, 1) andQ = (c, d) = (5, 4). Then
c(t ) = (38t, 1+3t ) for < t < .
Linethrough
_
1
3
,
1
6
_
and
_
7
6
,
5
3
_
33. Segment joining(1, 1) and(2, 3)
solution Weusethetwo-point parametrizationof alinewithP = (a, b) = (1, 1) andQ = (c, d) = (2, 3). Then
c(t ) = (1+t, 1+2t ); sincewewant onlythesegment joiningthetwopoints, wewant 0 t 1.
Segment joining(3, 0) and(0, 4)
35. Circleof radius4withcenter (3, 9)
solution Substituting(a, b) = (3, 9) andR = 4intheparametricequationof thecirclewegetc(t ) = (3+4cost, 9+
4sint ).
Ellipseof Exercise28, withitscenter translatedto(7, 4)
37. y = x
2
, translatedsothat theminimumoccursat (4, 8)
solution Wemay parametrizey = x
2
by (t, t
2
) for < t < . Theminimumof y = x
2
occurs at (0, 0),
so the desired curve is translated by (4, 8) fromy = x
2
. Thus a parametrization of the desired curve is c(t ) =
(4+t, 8+t
2
).
y = cosx translatedsothat amaximumoccursat (3, 5)
In Exercises 3942, nd a parametrization c(t ) of the curve satisfying the given condition.
39. y = 3x 4, c(0) = (2, 2)
solution Let x(t ) = t +a andy(t ) = 3x 4= 3(t +a) 4. Wewant x(0) = 2, thuswemust usea = 2. Our line
isc(t ) = (x(t ), y(t )) = (t +2, 3(t +2) 4) = (t +2, 3t +2).
y = 3x 4, c(3) = (2, 2)
41. y = x
2
, c(0) = (3, 9)
solution Let x(t ) = t + a andy(t ) = x
2
= (t + a)
2
. Wewant x(0) = 3, thus wemust usea = 3. Our curveis
c(t ) = (x(t ), y(t )) = (t +3, (t +3)
2
) = (t +3, t
2
+6t +9).
x
2
+y
2
= 4, c(0) = (1,
3)
43. Describec(t ) = (sect, tant ) for 0 t <

2
intheformy = f (x). Specifythedomainof x.
solution Thefunction x = sect has period 2 and y = tant has period . Thegraphs of thesefunctions in the
interval t , areshownbelow:
p p
p
2
p
2
y
x
p p p
2
p
2
y
x
x = sect y = tant
x = sect x
2
= sec
2
t
y = tant y
2
= tan
2
t =
sin
2
t
cos
2
t
=
1cos
2
t
cos
2
t
= sec
2
t 1= x
2
1
Hencethegraphof thecurveisthehyperbolax
2
y
2
= 1. Thefunctionx = sect isanevenfunctionwhiley = tant is
odd. Alsox hasperiod2 andy hasperiod. Itfollowsthattheintervals t <
2
,

2
< t <

2
and
2
< t <
tracethecurveexactlyonce. Thecorrespondingcurveisshownnext:
y
x
p
2
t =
p
2
t =
p
2
+ t =
p
2
+ t =
t = 0
(1, 0) (1, 0)
t = p
c(t ) = (sect, tant )
Findaparametrizationof theright branch(x > 0) of thehyperbola
_
x
a
_
2
_
y
b
_
2
= 1
usingthefunctionscosht andsinht . Howcanyouparametrizethebranchx < 0?
45. Thegraphs of x(t ) andy(t ) as functions of t areshown in Figure15(A). Which of (I)(III) is theplot of c(t ) =
(x(t ), y(t ))? Explain.
y y y y
x(t)
y(t)
x x x t
(A) (III) (II) (I)
FIGURE 15
solution As seeninFigure15(A), thex-coordinateis anincreasingfunctionof t , whiley(t ) is rst increasingand
thendecreasing. InFigureI, x andy arebothincreasingor bothdecreasing(dependingonthedirectiononthecurve).
InFigureII, x does not maintainonetendency, rather, it is decreasingandincreasingfor certainvalues of t . Theplot
c(t ) = (x(t ), y(t )) isplot III.
Whichgraph, (I) or (II), is thegraphof x(t ) andwhichis thegraphof y(t ) for theparametric curveinFigure
16(A)?
47. Sketchc(t ) = (t
3
4t, t
2
) followingthestepsinExample7.
solution Wenotethat x(t ) = t
3
4t isoddandy(t ) = t
2
iseven, hencec(t ) = (x(t ), y(t )) = (x(t ), y(t )).
Itfollowsthatc(t ) isthereectionof c(t ) acrossy-axis. Thatis, c(t ) andc(t ) aresymmetricwithrespecttothey-axis;
thus, it sufcestographthecurvefor t 0. For t = 0, wehavec(0) = (0, 0) andthey-coordinatey(t ) = t
2
tendsto
ast . Toanalyzethex-coordinate, wegraphx(t ) = t
3
4t for t 0:
x
3 4 1 2
4
2
8
6
4
2
y
x = t
3
4t
Weseethat x(t ) < 0anddecreasingfor 0 < t < 2/
3, x(t ) < 0andincreasingfor 2/
3 < t < 2andx(t ) > 0and

increasingfor t > 2. Alsox(t ) tendstoast . Therefore, startingat theorigin, thecurverst directstotheleft of
they-axis, thenat t = 2/
3it turnstotheright, alwayskeepinganupwarddirection. Thepart of thepathfor t 0is

obtainedbyreectingacrossthey-axis. Wealsousethepointsc(0) = (0, 0), c(1) = (3, 1), c(2) = (0, 4) toobtainthe
followinggraphfor c(t ):
x
y
t = 0
t = 1
t = 2
(3, 1)
(0, 4)
y
x
t = 1
t = 0
t = 2
t = 1
t = 2
Graphof c(t ) for t 0. Graphof c(t ) for all t .
Sketchc(t ) = (t
2
4t, 9t
2
) for 4 t 10.
In Exercises 4952, use Eq. (7) to nd dy/dx at the given point.
49. (t
3
, t
2
1), t = 4
solution ByEq. (7) wehave
dy
dx
=
y
(t )
x
(t )
=
(t
2
1)
(t
3
)
=
2t
3t
2
=
2
3t
Substitutingt = 4weget
dy
dx
=
2
3t
t =4
=
2
3 (4)
=
1
6
.
(2t +9, 7t 9), t = 1
51. (s
1
3s, s
3
), s = 1
solution UsingEq. (7) weget
dy
dx
=
y
(s)
x
(s)
=
(s
3
)
(s
1
3s)
=
3s
2
s
2
3
=
3s
4
13s
2
Substitutings = 1weobtain
dy
dx
=
3s
4
13s
2
s=1
=
3 (1)
4
13 (1)
2
=
3
4
.
(sin2, cos3), =

6
In Exercises 5356, nd an equation y = f (x) for the parametric curve and compute dy/dx in two ways: using Eq. (7)
and by differentiating f (x).
53. c(t ) = (2t +1, 19t )
solution Sincex = 2t +1, wehavet =
x 1
2
. Substitutinginy = 19t wehave
y = 19
_
x 1
2
_
=
9
2
x +
11
2
Differentiatingy =
9
2
x +
11
2
gives
dy
dx
=
9
2
. Wenownd
dy
dx
usingEq. (7):
dy
dx
=
y
(t )
x
(t )
=
(19t )
(2t +1)
=
9
2
c(t ) =
_
1
2
t,
1
4
t
2
t
_ 55. x = s
3
, y = s
6
+s
3
solution Wendy asafunctionof x:
y = s
6
+s
3
=
_
s
3
_
2
+
_
s
3
_
1
= x
2
+x
1
.
Wenowdifferentiatey = x
2
+x
1
. Thisgives
dy
dx
= 2x x
2
.
Alternatively, wecanuseEq. (7) toobtainthefollowingderivative:
dy
dx
=
y
(s)
x
(s)
=
_
s
6
+s
3
_
_
s
3
_
=
6s
5
3s
4
3s
2
= 2s
3
s
6
.
Hence, sincex = s
3
,
dy
dx
= 2x x
2
.
x = cos, y = cos +sin
2
57. Findthepointsonthecurvec(t ) = (3t

2
2t, t
3
6t ) wherethetangent linehasslope3.
solution Wesolve
dy
dx
=
3t
2
6
6t 2
= 3
or 3t
2
6= 18t 6, or t
2
6t = 0, sotheslopeis3at t = 0, 6andthepointsare(0, 0) and(96, 180)
Findtheequationof thetangent linetothecycloidgeneratedbyacircleof radius4at t =

2
.
In Exercises 5962, let c(t ) = (t
2
9, t
2
8t ) (see Figure 17).
60
40
20
60
40
20
x
y
FIGURE 17 Plot of c(t ) = (t
2
9, t
2
8t ).
59. Drawanarrowindicatingthedirectionof motion, anddeterminetheinterval of t -valuescorrespondingtotheportion
of thecurveineachof thefour quadrants.
solution Weplot thefunctionsx(t ) = t
2
9andy(t ) = t
2
8t :
t
x
3 3
t
y
1 2 3 4 5 6 7 8 9 321
x = t
2
9 y = t
2
8t
Wenotecarefully whereeach of thesegraphs arepositiveor negative, increasing or decreasing. In particular, x(t ) is
decreasingfor t < 0, increasingfor t > 0, positivefor |t | > 3, andnegativefor |t | < 3. Likewise, y(t ) isdecreasingfor
t < 4, increasingfor t > 4, positivefor t > 8or t < 0, andnegativefor 0 < t < 8. Wenowdrawarrowsonthepath
followingthedecreasing/increasingbehavior of thecoordinatesasindicatedabove. Weobtain:
x
y
20
t = 0
(9,0)
t = 8
(55,0)
t = 3
(0,15)
t = 3 (0,33)
t = 4 (7,16)
40 60 20
20
20
40
60
Thisplot alsoshowsthat:
Thegraphisintherst quadrant for t < 3or t > 8.
Thegraphisinthesecondquadrant for 3< t < 0.
Thegraphisinthethirdquadrant for 0< t < 3.
Thegraphisinthefourthquadrant for 3< t < 8.
Findtheequationof thetangent lineat t = 4.
61. Findthepointswherethetangent hasslope
1
2
.
solution Theslopeof thetangent at t is
dy
dx
=
_
t
2
8t
_
_
t
2
9
_
=
2t 8
2t
= 1
4
t
Thepoint wherethetangent hasslope
1
2
correspondstothevalueof t that satises
dy
dx
= 1
4
t
=
1
2
4
t
=
1
2
t = 8.
Wesubstitutet = 8inx(t ) = t
2
9andy(t ) = t
2
8t toobtainthefollowingpoint:
x(8) = 8
2
9= 55
y(8) = 8
2
8 8= 0
(55, 0)
Findthepointswherethetangent ishorizontal or vertical.
63. Let A andB bethepointswheretheray of angle intersectsthetwoconcentric circlesof radii r < R centeredat
theorigin(Figure18). Let P bethepoint of intersectionof thehorizontal linethroughA andthevertical linethroughB.
Expressthecoordinatesof P asafunctionof anddescribethecurvetracedbyP for 0 2.
x
y
B
P
R r
A
FIGURE 18
solution Weusetheparametricrepresentationof acircletodeterminethecoordinatesof thepointsA andB. That is,
A = (r cos, r sin), B = (Rcos, Rsin)
Thecoordinatesof P aretherefore
P = (Rcos, r sin)
Inorder toidentifythecurvetracedbyP, wenoticethat thex andy coordinatesof P satisfy
x
R
= cos and
y
r
= sin.
Hence
_
x
R
_
2
+
_
y
r
_
2
= cos
2
+sin
2
= 1.
Theequation
_
x
R
_
2
+
_
y
r
_
2
= 1
istheequationof ellipse. Hence, thecoordinatesof P, (Rcos, r sin) describeanellipsefor 0 2.
A 10-ft ladder slidesdownawall asitsbottomB ispulledawayfromthewall (Figure19). Usingtheangle as
parameter, ndtheparametric equations for thepathfollowedby (a) thetopof theladder A, (b) thebottomof the
ladder B, and(c) thepoint P located4ft fromthetopof theladder. Showthat P describesanellipse.
In Exercises 6568, refer to the Bzier curve dened by Eqs. (8) and (9).
65. Showthat theBzier curvewithcontrol points
P
0
= (1, 4), P
1
= (3, 12), P
2
= (6, 15), P
3
= (7, 4)
hasparametrization
c(t ) = (1+6t +3t
2
3t
3
, 4+24t 15t
2
9t
3
)
Verifythat theslopeat t = 0isequal totheslopeof thesegment P
0
P
1
.
solution For thegivenBzier curvewehavea
0
= 1, a
1
= 3, a
2
= 6, a
3
= 7, andb
0
= 4, b
1
= 12, b
2
= 15, b
3
= 4.
SubstitutingthesevaluesinEq. (8)(9) andsimplifyinggives
x(t ) = (1t )
3
+9t (1t )
2
+18t
2
(1t ) +7t
3
= 13t +3t
2
t
3
+9t (12t +t
2
) +18t
2
18t
3
+7t
3
= 13t +3t
2
t
3
+9t 18t
2
+9t
3
+18t
2
18t
3
+7t
3
= 3t
3
+3t
2
+6t +1
y(t ) = 4(1t )
3
+36t (1t )
2
+45t
2
(1t ) +4t
3
= 4(13t +3t
2
t
3
) +36t (12t +t
2
) +45t
2
45t
3
+4t
3
= 412t +12t
2
4t
3
+36t 72t
2
+36t
3
+45t
2
45t
3
+4t
3
= 4+24t 15t
2
9t
3
Then
c(t ) = (1+6t +3t
2
3t
3
, 4+24t 15t
2
9t
3
), 0 t 1.
Wendtheslopeat t = 0. Usingtheformulafor slopeof thetangent lineweget
dy
dx
=
(4+24t 15t
2
9t
3
)
(1+6t +3t
2
3t
3
)
=
2430t 27t
2
6+6t 9t
2

dy
dx
t =0
=
24
6
= 4.
Theslopeof thesegment P
0
P
1
istheslopeof thelinedeterminedby thepointsP
0
= (1, 4) andP
1
= (3, 12). That is,
124
31
=
8
2
= 4. Weseethat theslopeof thetangent lineat t = 0isequal totheslopeof thesegment P
0
P
1
, asexpected.
Findanequationof thetangent linetotheBzier curveinExercise65at t =
1
3
.
67. Findandplot theBzier curvec(t ) passingthroughthecontrol points
P
0
= (3, 2), P
1
= (0, 2), P
2
= (5, 4), P
3
= (2, 4)
solution Setting a
0
= 3, a
1
= 0, a
2
= 5, a
3
= 2, and b
0
= 2, b
1
= 2, b
2
= 4, b
3
= 4 into Eq. (8)(9) and
simplifyinggives
x(t ) = 3(1t )
3
+0+15t
2
(1t ) +2t
3
= 3(13t +3t
2
t
3
) +15t
2
15t
3
+2t
3
= 39t +24t
2
16t
3
y(t ) = 2(1t )
3
+6t (1t )
2
+12t
2
(1t ) +4t
3
= 2(13t +3t
2
t
3
) +6t (12t +t
2
) +12t
2
12t
3
+4t
3
= 26t +6t
2
2t
3
+6t 12t
2
+6t
3
+12t
2
12t
3
+4t
3
= 2+6t
2
4t
3
Weobtainthefollowingequation
c(t ) = (39t +24t
2
16t
3
, 2+6t
2
4t
3
), 0 t 1.
Thegraphof theBzier curveisshowninthefollowinggure:
x
y
1 2 3
1
2
3
4
Showthat acubicBzier curveistangent tothesegment P
2
P
3
at P
3
.
69. A bullet redfromagunfollowsthetrajectory
x = at, y = bt 16t
2
(a, b > 0)
Showthat thebullet leavesthegunat anangle = tan
1
_
b
a
_
andlandsat adistanceab/16fromtheorigin.
solution The height of the bullet equals the value of the y-coordinate. When the bullet leaves the gun, y(t ) =
t (b 16t ) = 0. Thesolutionstothisequationaret = 0andt =
b
16
, witht = 0correspondingtothemoment thebullet
leavesthegun. Wendtheslopem of thetangent lineat t = 0:
dy
dx
=
y
(t )
x
(t )
=
b 32t
a
m =
b 32t
a
t =0
=
b
a
Itfollowsthattan =
b
a
or = tan
1
_
b
a
_
. Thebulletlandsatt =
b
16
. Wendthedistanceof thebulletfromtheorigin
at thistime, bysubstitutingt =
b
16
inx(t ) = at . Thisgives
x
_
b
16
_
=
ab
16
Plot c(t ) = (t
3
4t, t
4
12t
2
+ 48) for 3 t 3. Findthepointswherethetangent lineishorizontal or
vertical.
71. Plot theastroidx = cos
3
, y = sin
3
andndtheequationof thetangent lineat =

3
.
solution Thegraphof theastroidx = cos
3
, y = sin
3
isshowninthefollowinggure:
x
y
= (0, 1)

2
= (0, 1)
3
2
= 0
(1, 0)
=
(1, 0)
Theslopeof thetangent lineat =

3
is
m =
dy
dx
=/3
=
(sin
3
)
(cos
3
)
=/3
=
3sin
2
cos
3cos
2
(sin)
=/3
=
sin
cos
=/3
= tan
/3
=
3
Wendthepoint of tangency:
_
x
_
3
_
, y
_
3
__
=
_
cos
3

3
, sin
3

3
_
=
_
1
8
,
3
3
8
_
Theequationof thetangent lineat =

3
is, thus,
y
3
3
8
=
3
_
x
1
8
_
y =
3x +
3
2
Findtheequationof thetangentlineatt =

4
tothecycloidgeneratedbytheunitcirclewithparametricequation
(5).
73. Findthepointswithhorizontal tangent lineonthecycloidwithparametricequation(5).
solution Theparametricequationsof thecycloidare
x = t sint, y = 1cost
Wendtheslopeof thetangent lineat t :
dy
dx
=
(1cost )
(t sint )
=
sint
1cost
Thetangent lineishorizontal whereit hasslopezero. That is,
dy
dx
=
sint
1cost
= 0
sint = 0
cost = 1
t = (2k 1), k = 0, 1, 2, . . .
Wendthecoordinatesof thepointswithhorizontal tangent line, by substitutingt = (2k 1) inx(t ) andy(t ). This
gives
x = (2k 1) sin(2k 1) = (2k 1)
y = 1cos((2k 1)) = 1(1) = 2
Therequiredpointsare
((2k 1), 2), k = 0, 1, 2, . . .
Propertyof theCycloid Provethat thetangent lineat apoint P onthecycloidalwayspassesthroughthetop
point ontherollingcircleasindicatedinFigure20. Assumethegeneratingcircleof thecycloidhasradius1.
75. A curtate cycloid (Figure21) is thecurvetracedby apoint at adistanceh fromthecenter of acircleof radius R
rollingalongthex-axiswhereh < R. Showthat thiscurvehasparametricequationsx = Rt hsint , y = R hcost .
y
h
R
x
4 2
FIGURE 21 Curtatecycloid.
solution Let P beapoint at adistanceh fromthecenter C of thecircle. Assumethat at t = 0, thelineof CP is
passingthroughtheorigin. WhenthecirclerollsadistanceRt alongthex-axis, thelengthof thearc
SQ (seegure) is
alsoRt andtheangle
SCQ hasradianmeasuret . Wecomputethecoordinatesx andy of P.

0
C C
R
S
Rt
A
P
h
t
Q
x = Rt PA = Rt hsin( t ) = Rt hsint
y = R +AC = R +hcos( t ) = R hcost
Weobtainthefollowingparametrization:
x = Rt hsint, y = R hcost.
Useacomputer algebrasystemtoexplorewhat happenswhenh > R intheparametricequationsof Exercise
75. Describetheresult.
77. Showthat thelineof slopet through(1, 0) intersectstheunit circleinthepoint withcoordinates
x =
1t
2
t
2
+1
, y =
2t
t
2
+1
10
Concludethat theseequationsparametrizetheunit circlewiththepoint (1, 0) excluded(Figure22). Showfurther that
t = y/(x +1).
(x, y)
(1, 0)
Slopet
y
x
FIGURE 22 Unit circle.
solution The equation of the line of slope t through (1, 0) is y = t (x + 1). The equation of the unit circle is
x
2
+y
2
= 1. Hence, thelineintersectstheunit circleat thepoints(x, y) that satisfytheequations:
y = t (x +1) (1)
x
2
+y
2
= 1 (2)
Substitutingy fromequation(1) intoequation(2) andsolvingfor x weobtain
x
2
+t
2
(x +1)
2
= 1
x
2
+t
2
x
2
+2t
2
x +t
2
= 1
(1+t
2
)x
2
+2t
2
x +(t
2
1) = 0
Thisgives
x
1,2
=
2t
2
_
4t
4
4(t
2
+1)(t
2
1)
2(1+t
2
)
=
2t
2
2
2(1+t
2
)
=
1t
2
1+t
2
So x
1
= 1andx
2
=
1t
2
t
2
+1
. Thesolutionx = 1corresponds to thepoint (1, 0). Weareinterestedinthesecond
point of intersectionthat isvaryingast varies. Hencetheappropriatesolutionis
x =
1t
2
t
2
+1
Wendthey-coordinatebysubstitutingx inequation(1). Thisgives
y = t (x +1) = t
_
1t
2
t
2
+1
+1
_
= t
1t
2
+t
2
+1
t
2
+1
=
2t
t
2
+1
Weconcludethat thelineandtheunit circleintersect, besidesat (1, 0), at thepoint withthefollowingcoordinates:
x =
1t
2
t
2
+1
, y =
2t
t
2
+1
(3)
Sincethesepointsdetermineall thepointsontheunit circleexcept for (1, 0) andnoother points, theequationsin(3)
parametrizetheunit circlewiththepoint (1, 0) excluded.
Weshowthat t =
y
x +1
. Using(3) wehave
y
x +1
=
2t
t
2
+1
1t
2
t
2
+1
+1
=
2t
t
2
+1
1t
2
+t
2
+1
t
2
+1
=
2t
t
2
+1
2
t
2
+1
=
2t
2
= t.
Thefoliumof Descartesisthecurvewithequationx
3
+y
3
= 3axy, wherea = 0isaconstant (Figure23).
(a) Showthattheliney = t x intersectsthefoliumattheoriginandatoneotherpointP forall t = 1, 0. Expressthe
coordinatesof P intermsof t toobtainaparametrizationof thefolium. Indicatethedirectionof theparametrization
onthegraph.
(b) Describetheinterval of t -valuesparametrizingthepartsof thecurveinquadrantsI, II, andIV. Notethat t = 1
isapoint of discontinuityof theparametrization.
(c) Calculatedy/dx asafunctionof t andndthepointswithhorizontal or vertical tangent.
79. Usetheresults of Exercise78to showthat theasymptoteof thefoliumis thelinex + y = a. Hint: Showthat
lim
t 1
(x +y) = a.
solution Wemust showthat as x or x thegraphof thefoliumis gettingarbitrarily closeto theline
x +y = a, andthederivative
dy
dx
isapproachingtheslope1of theline.
InExercise78weshowedthatx whent (1
) andx whent (1
+
).Werstshowthatthegraph
isapproachingthelinex +y = a asx or x , byshowingthat lim
t 1
x +y = lim
t 1+
x +y = a.
For x(t ) =
3at
1+t
3
, y(t ) =
3at
2
1+t
3
, a > 0, calculatedinExercise78, weobtainusingLHpitalsRule:
lim
t 1
(x +y) = lim
t 1
3at +3at
2
1+t
3
= lim
t 1
3a +6at
3t
2
=
3a 6a
3
= a
lim
t 1+
(x +y) = lim
t 1+
3at +3at
2
1+t
3
= lim
t 1+
3a +6at
3t
2
=
3a 6a
3
= a
Wenowshowthat
dy
dx
isapproaching1ast 1andast 1+.Weuse
dy
dx
=
6at 3at
4
3a 6at
3
computedinExercise
78toobtain
lim
t 1
dy
dx
= lim
t 1
6at 3at
4
3a 6at
3
=
9a
9a
= 1
lim
t 1+
dy
dx
= lim
t 1+
6at 3at
4
3a 6at
3
=
9a
9a
= 1
Weconcludethat thelinex +y = a isanasymptoteof thefoliumasx andasx .
Findaparametrizationof x
2n+1
+y
2n+1
= ax
n
y
n
, wherea andn areconstants.
81. SecondDerivativefor aParametrizedCurve Givenaparametrizedcurvec(t ) = (x(t ), y(t )), showthat
d
dt
_
dy
dx
_
=
x
(t )y
(t ) y
(t )x
(t )
x
(t )
2
Usethistoprovetheformula
d
2
y
dx
2
=
x
(t )y
(t ) y
(t )x
(t )
x
(t )
3
11
solution Bytheformulafor theslopeof thetangent linewehave
dy
dx
=
y
(t )
x
(t )
Differentiatingwithrespect tot , usingtheQuotient Rule, gives
d
dt
_
dy
dx
_
=
d
dt
_
y
(t )
x
(t )
_
=
x
(t )y
(t ) y
(t )x
(t )
x
(t )
2
BytheChainRulewehave
d
2
y
dx
2
=
d
dx
_
dy
dx
_
=
d
dt
_
dy
dx
_
dt
dx
Substitutingintotheaboveequation
_
andusing
dt
dx
=
1
dx/dt
=
1
x
(t )
_
gives
d
2
y
dx
2
=
x
(t )y
(t ) y
(t )x
(t )
x
(t )
2

1
x
(t )
=
x
(t )y
(t ) y
(t )x
(t )
x
(t )
3
Thesecondderivativeof y = x
2
isdy
2
/d
2
x = 2.VerifythatEq.(11)appliedtoc(t ) = (t, t
2
) yieldsdy
2
/d
2
x = 2.
Infact,anyparametrizationmaybeused.Checkthatc(t ) = (t
3
, t
6
) andc(t ) = (tant, tan
2
t ) alsoyielddy
2
/d
2
x = 2.
In Exercises 8386, use Eq. (11) to nd d
2
y/dx
2
.
83. x = t
3
+t
2
, y = 7t
2
4, t = 2
solution Wendtherst andsecondderivativesof x(t ) andy(t ):
x
(t ) = 3t
2
+2t x
(2) = 3 2
2
+2 2= 16
x
(t ) = 6t +2 x
(2) = 6 2+2= 14
y
(t ) = 14t y
(2) = 14 2= 28
y
(t ) = 14 y
(2) = 14
UsingEq. (11) weget
d
2
y
dx
2
t =2
=
x
(t )y
(t ) y
(t )x
(t )
x
(t )
3
t =2
=
16 1428 14
16
3
=
21
512
x = s
1
+s, y = 4s
2
, s = 1
85. x = 8t +9, y = 14t , t = 3
solution Wecomputetherst andsecondderivativesof x(t ) andy(t ):
x
(t ) = 8 x
(3) = 8
x
(t ) = 0 x
(3) = 0
y
(t ) = 4 y
(3) = 4
y
(t ) = 0 y
(3) = 0
UsingEq. (11) weget
d
2
y
dx
2
t =3
=
x
(3)y
(3) y
(3)x
(3)
x
(3)
3
=
8 0(4) 0
8
3
= 0
x = cos, y = sin, =

4
87. UseEq. (11) tondthet -intervalsonwhichc(t ) = (t
2
, t
3
4t ) isconcaveup.
solution Thecurveisconcaveupwhere
d
2
y
dx
2
> 0. Thus,
x
(t )y
(t ) y
(t )x
(t )
x
(t )
3
> 0 (1)
Wecomputetherst andsecondderivatives:
x
(t ) = 2t, x
(t ) = 2
y
(t ) = 3t
2
4, y
(t ) = 6t
Substitutingin(1) andsolvingfor t gives
12t
2
(6t
2
8)
8t
3
=
6t
2
+8
8t
3
Since6t
2
+8> 0for all t , thequotient ispositiveif 8t
3
> 0. Weconcludethat thecurveisconcaveupfor t > 0.
UseEq. (11) tondthet -intervalsonwhichc(t ) = (t
2
, t
4
4t ) isconcaveup.
89. AreaUnder aParametrizedCurve Letc(t ) = (x(t ), y(t )), wherey(t ) > 0andx
(t ) > 0(Figure24). Showthat

theareaA under c(t ) for t
0
t t
1
is
A =
_
t
1
t
0
y(t )x
(t ) dt 12
Hint: Becauseit isincreasing, thefunctionx(t ) hasaninverset = g(x) andc(t ) isthegraphof y = y(g(x)). Applythe
change-of-variablesformulatoA =
_
x(t
1
)
x(t
0
)
y(g(x)) dx.
y
c(t)
x(t
1
) x(t
0
)
xx
FIGURE 24
solution Let x
0
= x(t
0
) andx
1
= x(t
1
). Wearegiven that x
(t ) > 0, hencex = x(t ) is an increasingfunction of

t , so it has an inversefunction t = g(x). TheareaA is given by
_
x
1
x
0
y(g(x)) dx. Recall that y is afunction of t and
t = g(x), sotheheight y at anypoint x isgivenbyy = y(g(x)). Wendthenewlimitsof integration. Sincex
0
= x(t
0
)
andx
1
= x(t
1
), thelimitsfor t aret
0
andt
1
, respectively. Alsosincex
(t ) =
dx
dt
, wehavedx = x
(t )dt . Performingthis
substitutiongives
A =
_
x
1
x
0
y(g(x)) dx =
_
t
1
t
0
y(g(x))x
(t ) dt.
Sinceg(x) = t , wehaveA =
_
t
1
t
0
y(t )x
(t ) dt .
Calculatetheareaunder y = x
2
over [0, 1] usingEq. (12) withtheparametrizations(t
3
, t
6
) and(t
2
, t
4
).
91. What doesEq. (12) sayif c(t ) = (t, f (t ))?
solution In the parametrization x(t ) = t , y(t ) = f (t ) we have x
(t ) = 1, t
0
= x(t
0
), t
1
= x(t
1
). Hence Eq. (12)
becomes
A =
_
t
1
t
0
y(t )x
(t ) dt =
_
x(t
1
)
x(t
0
)
f (t ) dt
Weseethat inthisparametrizationEq. (12) isthefamiliar formulafor theareaunder thegraphof apositivefunction.
Sketchthegraphof c(t ) = (lnt, 2t ) for 1 t 2andcomputetheareaunder thegraphusingEq. (12).
93. Galileotriedunsuccessfully tondtheareaunder acycloid. Around1630, GillesdeRoberval provedthat thearea
under onearchof thecycloidc(t ) = (Rt Rsint, R Rcost ) generatedbyacircleof radiusR isequal tothreetimes
theareaof thecircle(Figure25). VerifyRobervalsresult usingEq. (12).
x
R
R 2R
y
FIGURE 25 Theareaof onearchof thecycloidequalsthreetimestheareaof thegeneratingcircle.
solution Thisreducesto
_
2
0
(R Rcost )(Rt Rsint )
dt =
_
2
0
R
2
(1cost )
2
dt = 3R
2
.
Provethefollowinggeneralizationof Exercise93: For all t > 0, theareaof thecycloidal sector OPC isequal to
threetimestheareaof thecircular segment cut bythechordPC inFigure26.
95. Derivetheformulafor theslopeof thetangent linetoaparametriccurvec(t ) = (x(t ), y(t )) usingamethod
different fromthat presentedinthetext. Assumethat x
(t
0
) andy
(t
0
) exist andthat x
(t
0
) = 0. Showthat
lim
h0
y(t
0
+h) y(t
0
)
x(t
0
+h) x(t
0
)
=
y
(t
0
)
x
(t
0
)
Thenexplainwhy thislimit isequal totheslopedy/dx. Drawadiagramshowingthat theratiointhelimit istheslope
of asecant line.
solution Sincey
(t
0
) andx
(t
0
) exist, wehavethefollowinglimits:
lim
h0
y(t
0
+h) y(t
0
)
h
= y
(t
0
), lim
h0
x(t
0
+h) x(t
0
)
h
= x
(t
0
) (1)
WeuseBasicLimit Laws, thelimitsin(1) andthegivendatax
(t
0
) = 0, towrite
lim
h0
y(t
0
+h) y(t
0
)
x(t
0
+h) x(t
0
)
= lim
h0
y(t
0
+h)y(t
0
)
h
x(t
0
+h)x(t
0
)
h
=
lim
h0
y(t
0
+h)y(t
0
)
h
lim
h0
x(t
0
+h)x(t
0
)
h
=
y
(t
0
)
x
(t
0
)
Noticethatthequotient
y(t
0
+h) y(t
0
)
x(t
0
+h) x(t
0
)
istheslopeof thesecantlinedeterminedbythepointsP = (x(t
0
), y(t
0
)) and
Q = (x(t
0
+h), y(t
0
+h)). Hence, thelimit of thequotient as h 0is theslopeof thetangent lineat P, that is the
derivative
dy
dx
.
x
y
x(t
0
+ h) x(t
0
)
y(t
0
)
y(t
0
, h)
P
Q
Verifythat thetractrixcurve( > 0)
c(t ) =
_
t tanh
t
, sech
t
_
hasthefollowingproperty: For all t , thesegment fromc(t ) to(t, 0) istangent tothecurveandhaslength (Figure
27).
97. InExercise54of Section10.1(ET Exercise54of Section9.1), wedescribedthetractrixbythedifferential equation
dy
dx
=
y
_
2
y
2
Showthatthecurvec(t ) identiedasthetractrixinExercise96satisesthisdifferential equation. Notethatthederivative
ontheleft istakenwithrespect tox, not t .
solution Notethat dx/dt = 1sech
2
(t /) = tanh
2
(t /) anddy/dt = sech(t /) tanh(t /). Thus,
dy
dx
=
dy/dt
dx/dt
=
sech(t /)
tanh(t /)
=
y/
_
1y
2
/
2
Multiplyingtopandbottomby/ gives
dy
dx
=
y
_
2
y
2
q
y
(x, y)
x
FIGURE 28 Theparameter ontheellipse
_
x
a
_
2
+
_
y
b
_
2
= 1.
In the parametrization c(t ) = (a cost, b sint ) of an ellipse, t is not an angular parameter unless a = b (in
whichcasetheellipseis acircle). However, t canbeinterpretedinterms of area: Showthat if c(t ) = (x, y), then
t = (2/ab)A, whereA istheareaof theshadedregioninFigure29. Hint: UseEq. (12).
99. Showthat theparametrizationof theellipsebytheangle is
x =
ab cos
_
a
2
sin
2
+b
2
cos
2
y =
ab sin
_
a
2
sin
2
+b
2
cos
2
solution Weconsider theellipse

x
2
a
2
+
y
2
b
2
= 1.
For theangle wehavetan =
y
x
, hence,
y = x tan (1)
Substitutingintheequationof theellipseandsolvingfor x weobtain
x
2
a
2
+
x
2
tan
2
b
2
= 1
b
2
x
2
+a
2
x
2
tan
2
= a
2
b
2
(a
2
tan
2
+b
2
)x
2
= a
2
b
2
x
2
=
a
2
b
2
a
2
tan
2
+b
2
=
a
2
b
2
cos
2
a
2
sin
2
+b
2
cos
2
Wenowtakethesquareroot. Sincethesignof thex-coordinateisthesameasthesignof cos, wetakethepositiveroot,

obtaining
x =
ab cos
_
a
2
sin
2
+b
2
cos
2
(2)
Henceby(1), they-coordinateis
y = x tan =
ab cos tan
_
a
2
sin
2
+b
2
cos
2
=
ab sin
_
a
2
sin
2
+b
2
cos
2
(3)
Equalities(2) and(3) givethefollowingparametrizationfor theellipse:
c
1
() =
_
ab cos
_
a
2
sin
2
+b
2
cos
2
,
ab sin
_
a
2
sin
2
+b
2
cos
2
_
S E C T I ON 12.2 Arc Length and Speed 759
12.2 Arc Length and Speed
1. What isthedenitionof arclength?
solution A curvecanbeapproximatedbyapolygonal pathobtainedbyconnectingpoints
p
0
= c(t
0
), p
1
= c(t
1
), . . . , p
N
= c(t
N
)
onthepathwithsegments. Onegets anapproximationby summingthelengths of thesegments. Thedenitionof arc
lengthisthelimitof thatapproximationwhenincreasingthenumberof pointssothatthelengthsof thesegmentsapproach
zero. Indoingso, weobtainthefollowingtheoremfor thearclength:
S =
_
b
a
_
x
(t )
2
+y
(t )
2
dt,
whichisthelengthof thecurvec(t ) = (x(t ), y(t )) for a t b.
2. What istheinterpretationof
_
x
(t )
2
+y
(t )
2
for aparticlefollowingthetrajectory(x(t ), y(t ))?
solution Theexpression
_
x
(t )
2
+y
(t )
2
denotesthespeedattimet of aparticlefollowingthetrajectory(x(t ), y(t )).
3. Aparticletravelsalongapathfrom(0, 0) to(3, 4). Whatisthedisplacement?Canthedistancetraveledbedetermined
fromtheinformationgiven?
solution Thenet displacement isthedistancebetweentheinitial point (0, 0) andtheendpoint (3, 4). That is
_
(30)
2
+(40)
2
=
25= 5.
Thedistancetraveledcanbedeterminedonlyif thetrajectoryc(t ) = (x(t ), y(t )) of theparticleisknown.
4. A particletraverses theparabolay = x
2
withconstant speed3cm/s. What is thedistancetraveledduringtherst
minute? Hint: Nocomputationisnecessary.
solution Sincethespeedisconstant, thedistancetraveledisthefollowingproduct: L = st = 3 60= 180cm.
Exercises
In Exercises 110, use Eq. (3) to nd the length of the path over the given interval.
1. (3t +1, 94t ), 0 t 2
solution Sincex = 3t +1andy = 94t wehavex
= 3andy
= 4. Hence, thelengthof thepathis

S =
_
2
0
_
3
2
+(4)
2
dt = 5
_
2
0
dt = 10.
(1+2t, 2+4t ), 1 t 4
3. (2t
2
, 3t
2
1), 0 t 4
solution Sincex = 2t
2
andy = 3t
2
1, wehavex
= 4t andy
= 6t . Bytheformulafor thearclengthweget
S =
_
4
0
_
x
(t )
2
+y
(t )
2
dt =
_
4
0
_
16t
2
+36t
2
dt =
52
_
4
0
t dt =
52
t
2
2
4
0
= 16
13
(3t, 4t
3/2
), 0 t 1
5. (3t
2
, 4t
3
), 1 t 4
solution Wehavex = 3t
2
andy = 4t
3
. Hencex
= 6t andy
= 12t
2
. Bytheformulafor thearclengthweget
S =
_
4
1
_
x
(t )
2
+y
(t )
2
dt =
_
4
1
_
36t
2
+144t
4
dt = 6
_
4
1
_
1+4t
2
t dt.
Usingthesubstitutionu = 1+4t
2
, du = 8t dt weobtain
S =
6
8
_
65
5
u du =
3
4
2
3
u
3/2
65
5
=
1
2
(65
3/2
5
3/2
) 256.43
(t
3
+1, t
2
3), 0 t 1
7. (sin3t, cos3t ), 0 t
solution Wehavex = sin3t , y = cos3t , hencex
= 3cos3t andy
= 3sin3t . Bytheformulafor thearclength

weobtain:
S =
_

0
_
x
(t )
2
+y
(t )
2
dt =
_

0
_
9cos
2
3t +9sin
2
3t dt =
_

0
9dt = 3
(sin cos, cos + sin), 0 2
In Exercises 9 and 10, use the identity
1cost
2
= sin
2
t
2
9. (2cost cos2t, 2sint sin2t ), 0 t

2
solution Wehavex = 2cost cos2t ,y = 2sint sin2t .Thus,x
= 2sint +2sin2t andy
= 2cost 2cos2t .
Weget
x
(t )
2
+y
(t )
2
= (2sint +2sin2t )
2
+(2cost 2cos2t )
2
= 4sin
2
t 8sint sin2t +4sin
2
2t +4cos
2
t 8cost cos2t +4cos
2
2t
= 4(sin
2
t +cos
2
t ) +4(sin
2
2t +cos
2
2t ) 8(sint sin2t +cost cos2t )
= 4+48cos(2t t ) = 88cost = 8(1cost )
Wenowusetheformulafor thearclengthtoobtain
S =
_
/2
0
_
x
(t )
2
+y
(t )
2
=
_
/2
0
_
8(1cost ) dt =
_
/2
0
_
16sin
2
t
2
dt = 4
_
/2
0
sin
t
2
dt
= 8cos
t
2
/2
0
= 8
_
cos
4
cos0
_
= 8
_
2
2
1
_
2.34
(5( sin), 5(1cos)), 0 2
11. Showthat onearchof acycloidgeneratedbyacircleof radiusR haslength8R.
solution Recall fromearlier that thecycloidgeneratedby acircleof radius R has parametric equations x = Rt
Rsint , y = R Rcost . Hence, x
= R Rcost , y
= Rsint . Usingtheidentitysin
2
t
2
=
1cost
2
, weget
x
(t )
2
+y
(t )
2
= R
2
(1cost )
2
+R
2
sin
2
t = R
2
(12cost +cos
2
t +sin
2
t )
= R
2
(12cost +1) = 2R
2
(1cost ) = 4R
2
sin
2
t
2
Onearchof thecycloidistracedast variesfrom0to2. Hence, usingtheformulafor thearclengthweobtain:
S =
_
2
0
_
x
(t )
2
+y
(t )
2
dt =
_
2
0
_
4R
2
sin
2
t
2
dt = 2R
_
2
0
sin
t
2
dt = 4R
_

0
sinu du
= 4Rcosu
0
= 4R(cos cos0) = 8R
Findthelengthof thespiral c(t ) = (t cost, t sint ) for 0 t 2 tothreedecimal places(Figure7). Hint: Use
theformula
_
_
1+t
2
dt =
1
2
t
_
1+t
2
+
1
2
ln
_
t +
_
1+t
2
_
13. Findthelengthof thetractrix(seeFigure6)
c(t ) = (t tanh(t ), sech(t )), 0 t A
solution Sincex = t tanh(t ) andy = sech(t ) wehavex
= 1sech
2
(t ) andy
= sech(t ) tanh(t ). Hence,

x
(t )
2
+y
(t )
2
= (1sech
2
(t ))
2
+sech
2
(t )tanh
2
(t )
= 12sech
2
(t ) +sech
4
(t ) +sech
2
(t )tanh
2
(t )
= 12sech
2
(t ) +sech
2
(t )(sech
2
(t ) +tanh
2
(t ))
= 12sech
2
(t ) +sech
2
(t ) = 1sech
2
(t ) = tanh
2
(t )
Hence, usingtheformulafor thearclengthweget:
S =
_
A
0
_
x
(t )
2
+y
(t )
2
dt =
_
A
0
_
tanh
2
(t ) dt =
_
A
0
tanh(t ) dt = ln(cosh(t ))
A
0
= ln(cosh(A)) ln(cosh(0)) = ln(cosh(A)) ln1= ln(cosh(A))
Findanumerical approximationtothelengthof c(t ) = (cos5t, sin3t ) for 0 t 2 (Figure8).
In Exercises 1518, determine the speed s at time t (assume units of meters and seconds).
15. (t
3
, t
2
), t = 2
solution We have x(t ) = t
3
, y(t ) = t
2
hence x
(t ) = 3t
2
, y
(t ) = 2t . The speed of the particle at time t is thus,

ds
dt
=
_
x
(t )
2
+y
(t )
2
=
_
9t
4
+4t
2
= t
_
9t
2
+4. At timet = 2thespeedis
ds
dt
t =2
= 2
_
9 2
2
+4= 2
40= 4
10 12.65m/s.
(3sin5t, 8cos5t ), t =

4
17. (5t +1, 4t 3), t = 9
solution Sincex = 5t +1, y = 4t 3, wehavex
= 5andy
= 4. Thespeedof theparticleat timet is

ds
dt
=
_
x
(t ) +y
(t ) =
_
5
2
+4
2
=
41 6.4m/s.
Weconcludethat theparticlehasconstant speedof 6.4m/s.
(ln(t
2
+1), t
3
), t = 1
19. Findtheminimumspeedof aparticlewithtrajectoryc(t ) = (t
3
4t, t
2
+1) for t 0. Hint: It iseasier tondthe
minimumof thesquareof thespeed.
solution Werst ndthespeedof theparticle. Wehavex(t ) = t
3
4t , y(t ) = t
2
+ 1, hencex
(t ) = 3t
2
4and
y
(t ) = 2t . Thespeedisthus
ds
dt
=
_
(3t
2
4)
2
+(2t )
2
=
_
9t
4
24t
2
+16+4t
2
=
_
9t
4
20t
2
+16.
Thesquareroot functionisanincreasingfunction, hencetheminimumspeedoccursat thevalueof t wherethefunction
f (t ) = 9t
4
20t
2
+16hasminimumvalue. Since lim
t
f (t ) = , f hasaminimumvalueontheinterval 0 t < ,
andit occursat acritical point or at theendpoint t = 0. Wendthecritical point of f ont 0:
f
(t ) = 36t
3
40t = 4t (9t
2
10) = 0t = 0, t =
_
10
9
.
Wecomputethevaluesof f at thesepoints:
f (0) = 9 0
4
20 0
2
+16= 16
f
__
10
9
_
= 9
__
10
9
_
4
20
__
10
9
_
2
+16=
44
9
4.89
Weconcludethat theminimumvalueof f ont 0is4.89. Theminimumspeedistherefore
_
ds
dt
_
min
4.89 2.21.
Findtheminimumspeedof aparticlewithtrajectoryc(t ) = (t
3
, t
2
) for t 0.5.
21. Findthespeedof thecycloidc(t ) = (4t 4sint, 44cost ) at pointswherethetangent lineishorizontal.
solution Werst ndthepoints wherethetangent lineis horizontal. Theslopeof thetangent lineis thefollowing
quotient:
dy
dx
=
dy/dt
dx/dt
=
4sint
44cost
=
sint
1cost
.
Tondthepointswherethetangent lineishorizontal wesolvethefollowingequationfor t 0:
dy
dx
= 0,
sint
1cost
= 0sint = 0 and cost = 1.
Now, sint = 0andt 0at thepoints t = k, k = 0, 1, 2, . . . . Sincecosk = (1)
k
, thepoints wherecost = 1are
t = k for k odd. Thepointswherethetangent lineishorizontal are, therefore:
t = (2k 1), k = 1, 2, 3, . . .
Thespeedat timet isgivenbythefollowingexpression:
ds
dt
=
_
x
(t )
2
+y
(t )
2
=
_
(44cost )
2
+(4sint )
2
=
_
1632cost +16cos
2
t +16sin
2
t =
1632cost +16
=
_
32(1cost ) =
_
32 2sin
2
t
2
= 8
sin
t
2

That is, thespeedof thecycloidat timet is
ds
dt
= 8
sin
t
2
.
Wenowsubstitute
t = (2k 1), k = 1, 2, 3, . . .
toobtain
ds
dt
= 8
sin
(2k 1)
2
= 8|(1)
k+1
| = 8
Calculatethearclengthintegral s(t ) for thelogarithmic spiral c(t ) = (e
t
cost, e
t
sint ).
In Exercises 2326, plot the curve and use the Midpoint Rule with N = 10, 20, 30, and 50 to approximate its
length.
23. c(t ) = (cost, e
sint
) for 0 t 2
solution Thecurveof c(t ) = (cost, e
sint
) for 0 t 2 isshowninthegurebelow:
y
t = 0, t = 2, (1, 1) t = , (1, 1)
x
t = (0, e)

2
t = (0, )
3
2
1
e
c(t ) = (cost, e
sint
), 0 t 2.
Thelengthof thecurveisgivenbythefollowingintegral:
S =
_
2
0
_
x
(t )
2
+y
(t )
2
dt =
_
2
0
_
(sint )
2
+(cost e
sint
)
2
dt.
That is, S =
_
2
0
_
sin
2
t +cos
2
t e
2sint
dt . Weapproximatetheintegral usingtheMid-Point RulewithN = 10, 20,
30, 50. For f (t ) =
_
sin
2
t +cos
2
t e
2sint
weobtain
(N = 10): x =
2
10
=

5
, c
i
=
_
i
1
2
_

5
M
10
=

5
10
i=1
f (c
i
) = 6.903734
(N = 20): x =
2
20
=

10
, c
i
=
_
i
1
2
_

10
M
20
=

10
20
i=1
f (c
i
) = 6.915035
(N = 30): x =
2
30
=

15
, c
i
=
_
i
1
2
_

15
M
30
=

15
30
i=1
f (c
i
) = 6.914949
(N = 50): x =
2
50
=

25
, c
i
=
_
i
1
2
_

25
M
50
=

25
50
i=1
f (c
i
) = 6.914951
c(t ) = (t sin2t, 1cos2t ) for 0 t 2 25. Theellipse
_
x
5
_
2
+
_
y
3
_
2
= 1
solution WeusetheparametrizationgiveninExample4, section12.1, that is, c(t ) = (5cost, 3sint ), 0 t 2.
Thecurveisshowninthegurebelow:
y
t = 0
t = 2
x
c(t ) = (5cost, 3sint ), 0 t 2.
Thelengthof thecurveisgivenbythefollowingintegral:
S =
_
2
0
_
x
(t )
2
+y
(t )
2
dt =
_
2
0
_
(5sint )
2
+(3cost )
2
dt
=
_
2
0
_
25sin
2
t +9cos
2
t dt =
_
2
0
_
9(sin
2
t +cos
2
t ) +16sin
2
t dt =
_
2
0
_
9+16sin
2
t dt.
That is,
S =
_
2
0
_
9+16sin
2
t dt.
Weapproximatetheintegral usingtheMid-Point RulewithN = 10, 20, 30, 50, for f (t ) =
_
9+16sin
2
t . Weobtain
(N = 10): x =
2
10
=

5
, c
i
=
_
i
1
2
_

5
M
10
=

5
10
i=1
f (c
i
) = 25.528309
(N = 20): x =
2
20
=

10
, c
i
=
_
i
1
2
_

10
M
20
=

10
20
i=1
f (c
i
) = 25.526999
(N = 30): x =
2
30
=

15
, c
i
=
_
i
1
2
_

15
M
30
=

15
30
i=1
f (c
i
) = 25.526999
(N = 50): x =
2
50
=

25
, c
i
=
_
i
1
2
_

25
M
50
=

25
50
i=1
f (c
i
) = 25.526999
x = sin2t , y = sin3t for 0 t 2
27. If youunwindthreadfromastationarycircular spool, keepingthethreadtaut at all times, thentheendpoint tracesa
curveC calledtheinvoluteof thecircle(Figure9). Observethat PQ haslengthR. Showthat C isparametrizedby
c() =
_
R(cos + sin), R(sin cos)
_
Thenndthelengthof theinvolutefor 0 2.
P = (x, y)
y
q
x
R
Q
FIGURE 9 Involuteof acircle.
solution Supposethat thearc

QT corresponding to theangle is unwound. Then thelength of thesegment QP
equalsthelengthof thisarc. That is, QP = R. Withthehelpof thegurewecanseethat
x = OA +AB = OA +EP = Rcos +QP sin = Rcos +R sin = R(cos + sin).
Furthermore,
y = QA QE = Rsin QP cos = Rsin R cos = R(sin cos)
Thecoordinatesof P withrespect totheparameter formthefollowingparametrizationof thecurve:
c() = (R(cos + sin), R(sin cos)), 0 2.
Wendthelengthof theinvolutefor 0 2, usingtheformulafor thearclength:
S =
_
2
0
_
x
()
2
+y
()
2
d.
Wecomputetheintegrand:
x
() =
d
d
(R(cos + sin)) = R(sin +sin + cos) = R cos
y
() =
d
d
(R(sin cos)) = R(cos (cos sin)) = R sin
_
x
()
2
+y
()
2
=
_
(R cos)
2
+(R sin)
2
=
_
R
2
2
(cos
2
+sin
2
) =
_
R
2
2
= R
Wenowcomputethearclength:
S =
_
2
0
R d =
R
2
2
2
0
=
R (2)
2
2
= 2
2
R.
Let a > b andset
k =
_
1
b
2
a
2
Useaparametricrepresentationtoshowthat theellipse
_
x
a
_
2
+
_
y
b
_
2
= 1haslengthL = 4aG
_
2
, k
_
, where
G(, k) =
_

0
_
1k
2
sin
2
t dt
istheelliptic integral of the second kind.
In Exercises 2932, use Eq. (4) to compute the surface area of the given surface.
29. Theconegeneratedbyrevolvingc(t ) = (t, mt ) about thex-axisfor 0 t A
solution Substitutingy(t ) = mt , y
(t ) = m, x
(t ) = 1, a = 0, andb = 0intheformulafor thesurfacearea, weget

S = 2
_
A
0
mt
_
1+m
2
dt = 2
_
1+m
2
m
_
A
0
t dt = 2m
_
1+m
2
t
2
2
A
0
= m
_
1+m
2
A
2
A sphereof radiusR
31. Thesurfacegeneratedbyrevolvingonearchof thecycloidc(t ) = (t sint, 1cost ) about thex-axis
solution Onearchof thecycloidistracedast variesfrom0to2. Sincex(t ) = t sint andy(t ) = 1cost , we
havex
(t ) = 1cost andy
(t ) = sint . Hence, usingtheidentity1cost = 2sin

2 t
2
, weget
x
(t )
2
+y
(t )
2
= (1cost )
2
+sin
2
t = 12cost +cos
2
t +sin
2
t = 22cost = 4sin
2
t
2
Bytheformulafor thesurfaceareaweobtain:
S = 2
_
2
0
y(t )
_
x
(t )
2
+y
(t )
2
dt = 2
_
2
0
(1cost ) 2sin
t
2
dt
= 2
_
2
0
2sin
2
t
2
2sin
t
2
dt = 8
_
2
0
sin
3
t
2
dt = 16
_

0
sin
3
u du
Weuseareductionformulatocomputethisintegral, obtaining
S = 16
_
1
3
cos
3
u cosu
_
0
= 16
_
4
3
_
=
64
3
Thesurfacegeneratedbyrevolvingtheastroidc(t ) = (cos
3
t, sin
3
t ) about thex-axisfor 0 t

2
33. Let b(t ) betheButteryCurve:
x(t ) = sint
_
e
cost
2cos4t sin
_
t
12
_
5
_
y(t ) = cost
_
e
cost
2cos4t sin
_
t
12
_
5
_
(a) Useacomputer algebrasystemtoplot b(t ) andthespeeds
(t ) for 0 t 12.
(b) Approximatethelengthb(t ) for 0 t 10.
solution
(a) Let f (t ) = e
cost
2cos4t sin
_
t
12
_
5
, then
x(t ) = sintf (t )
y(t ) = costf (t )
andso
(x
(t ))
2
+(y
(t ))
2
= [sintf
(t ) +costf (t )]
2
+[costf
(t ) sintf (t )]
2
Usingtheidentitysin
2
t +cos
2
t = 1, weget
(x
(t ))
2
+(y
(t ))
2
= (f
(t ))
2
+(f (t ))
2
.
Thus, s
(t ) isthefollowing:
_
_
e
cost
2cos4t sin
_
t
12
_
5
_
2
+
_
sint e
cost
+8sin4t
5
12
_
t
12
_
4
cos
_
t
12
_
5
_
2
.
Thefollowingguresshowthecurvesof b(t ) andthespeeds
(t ) for 0 t 10:
y
x
t = 10p
t = 0
30 20 10
15
20
10
5
x
y
TheButteryCurve b(t ), 0 t 10 s
(t ), 0 t 10
Looking at the graph, we see it would be difcult to compute the length using numeric integration; due to the high
frequencyoscillations, verysmall stepswouldbeneeded.
(b) Thelengthof b(t ) for 0 t 10 is givenby theintegral: L =
_
10
0
s
(t ) dt wheres
(t ) is giveninpart (a). We
approximate the length using the Midpoint Rule with N = 30. The numerical methods in Mathematica approximate
theanswer by 211.952. UsingtheMidpoint RulewithN = 50, weget 204.48; withN = 500, weget 211.6; andwith
N = 5000, weget 212.09.
Let a b > 0andset k =
2
ab
a b
. Showthat thetrochoid
x = at b sint, y = a b cost, 0 t T
haslength2(a b)G
_
T
2
, k
_
withG(, k) asinExercise28.
35. A satelliteorbitingat adistanceR fromthecenter of theearthfollowsthecircular pathx = Rcost , y = Rsint .
(a) Showthat theperiodT (thetimeof onerevolution) isT = 2/.
(b) AccordingtoNewtonslawsof motionandgravity,
x
(t ) = Gm
e
x
R
3
, y
(t ) = Gm
e
y
R
3
whereG is theuniversal gravitational constant andm
e
is themass of theearth. Provethat R
3
/T
2
= Gm
e
/4
2
. Thus,
R
3
/T
2
hasthesamevaluefor all orbits(aspecial caseof KeplersThirdLaw).
solution
(a) AsshowninExample4, thecircular pathhasconstantspeedof
ds
dt
= R. Sincethelengthof onerevolutionis2R,
theperiodT is
T =
2R
R
=
2
.
(b) Differentiatingx = Rcost twicewithrespect tot gives
x
(t ) = Rsint
x
(t ) = R
2
cost
Substitutingx(t ) andx
(t ) intheequationx
(t ) = Gm
e
x
R
3
andsimplifying, weobtain
R
2
cost = Gm
e
Rcost
R
3
R
2
=
Gm
e
R
2
R
3
=
Gm
e
2
Bypart (a), T =
2
. Hence, =
2
T
. Substitutingyields
R
3
=
Gm
e
4
2
T
2
=
T
2
Gm
e
4
2

R
3
T
2
=
Gm
e
4
2
Theaccelerationduetogravityonthesurfaceof theearthis
g =
Gm
e
R
2
e
= 9.8m/s
2
, whereR
e
= 6378km
UseExercise35(b) to showthat asatelliteorbiting at theearths surfacewould haveperiod T
e
= 2
R
e
/g
84.5min. ThenestimatethedistanceR
m
fromthemoonto thecenter of theearth. Assumethat theperiodof the
moon(sidereal month) isT
m
27.43days.
12.3 Polar Coordinates
1. PointsP andQ withthesameradial coordinate(choosethecorrect answer):
(a) Lieonthesamecirclewiththecenter at theorigin.
(b) Lieonthesameraybasedat theorigin.
solution Twopoints withthesameradial coordinateareequidistant fromtheorigin, thereforethey lieonthesame
circlecenteredattheorigin. Theangular coordinatedenesaraybasedattheorigin. Therefore, if thetwopointshavethe
sameangular coordinate, theylieonthesameraybasedat theorigin.
2. Givetwopolar representationsfor thepoint (x, y) = (0, 1), onewithnegativer andonewithpositiver.
solution Thepoint(0, 1) isonthey-axis, distantoneunitfromtheorigin, hencethepolarrepresentationwithpositive
r is (r, ) =
_
1,

2
_
. Thepoint (r, ) =
_
1,

2
_
is thereectionof (r, ) =
_
1,

2
_
throughtheorigin, hencewemust
add toreturntotheoriginal point.
Weobtainthefollowingpolar representationof (0, 1) withnegativer:
(r, ) =
_
1,
2
+
_
=
_
1,
3
2
_
.
3. Describeeachof thefollowingcurves:
(a) r = 2 (b) r
2
= 2 (c) r cos = 2
solution
(a) Convertingtorectangular coordinatesweget
_
x
2
+y
2
= 2 or x
2
+y
2
= 2
2
.
Thisistheequationof thecircleof radius2centeredat theorigin.
(b) Weconverttorectangularcoordinates, obtainingx
2
+y
2
= 2.Thisistheequationof thecircleof radius
2, centered
at theorigin.
(c) Weconverttorectangularcoordinates. Sincex = r cos weobtainthefollowingequation: x = 2.Thisistheequation
of thevertical linethroughthepoint (2, 0).
4. If f () = f (), thenthecurver = f () issymmetricwithrespect tothe(choosethecorrect answer):
(a) x-axis (b) y-axis (c) origin
solution Theequality f () = f () for all implies that whenever apoint (r, ) is on thecurve, also thepoint
(r, ) isonthecurve. Sincethepoint (r, ) isthereectionof (r, ) withrespect tothex-axis, weconcludethat the
curveissymmetricwithrespect tothex-axis.
S E C T I ON 12.3 Polar Coordinates 767
Exercises
1. Findpolar coordinatesfor eachof thesevenpointsplottedinFigure16.
x
(x, y) = (23, 2)
y
4
4
A
B
C D
G
E F
FIGURE 16
solution Wemarkthepointsasshowninthegure.
x
A
y
F(2 3, 2)
G(2 3, 2)
B
C D
E
Usingthedatagiveninthegurefor thex andy coordinatesandthequadrantsinwhichthepointarelocated, weobtain:
(A), withrectangular coordinates(3, 3):
r =
_
(3)
2
+3
2
=
18
=

4
=
3
4
(r, ) =
_
3
2,
3
4
_
x
A
y
3 2
3
4
(B), withrectangular coordinates(3, 0):
r = 3
=
(r, ) = (3, )
x
y
3 B
(C), withrectangular coordinates(2, 1):

r =
_
2
2
+1
2
=
5 2.2
= tan
1
_
1
2
_
= tan
1
_
1
2
_
= +0.46 3.6
(r, )
_
5, 3.6
_
x
y
C
3.6
2.2
(D), withrectangular coordinates(1, 1):
r =
_
1
2
+1
2
=
2 1.4
= +

4
=
5
4
(r, )
_
2,
5
4
_
x
y
D
5
4
1.4
(E), withrectangular coordinates(1, 1):
r =
_
1
2
+1
2
=
2 1.4
= tan
1
_
1
1
_
=

4
(r, )
_
2,

4
_
x
y
E

4
1.4
(F), withrectangular coordinates(2
3, 2):
r =
_
_
2
3
_
2
+2
2
=
16= 4
= tan
1
_
2
2
3
_
= tan
1
_
1
3
_
=

6
(r, ) =
_
4,

6
_
x
y
F(2 3, 2)

6
4
(G), withrectangularcoordinates(2
3, 2): Gisthereectionof F aboutthex axis, hencethetwopointshaveequal

radial coordinates, andtheangularcoordinateof Gisobtainedfromtheangularcoordinateof F: = 2

6
=
11
6
.
Hence, thepolar coordinatesof G are
_
4,
11
6
_
.
Plot thepointswithpolar coordinates:
(a)
_
2,

6
_
(b)
_
4,
3
4
_
(c)
_
3,
2
_
(d)
_
0,

6
_
3. Convert fromrectangular topolar coordinates.
(a) (1, 0) (b) (3,
3) (c) (2, 2) (d) (1,
3)
solution
(a) The point (1, 0) is on the positive x axis distanced one unit fromthe origin. Hence, r = 1 and = 0. Thus,
(r, ) = (1, 0).
(b) The point
_
3,
3
_
is in the rst quadrant so = tan
1
_
3
3
_
=

6
. Also, r =
_
3
2
+
_
3
_
2
=
12. Hence,
(r, ) =
_
12,

6
_
.
(c) Thepoint (2, 2) isinthesecondquadrant. Hence,
= tan
1
_
2
2
_
= tan
1
(1) =

4
=
3
4
.
Also, r =
_
(2)
2
+2
2
=
8. Hence, (r, ) =
_
8,
3
4
_
.
(d) Thepoint
_
1,
3
_
isinthesecondquadrant, hence,
= tan
1
_
3
1
_
= tan
1
_
3
_
=

3
=
2
3
.
Also, r =
_
(1)
2
+
_
3
_
2
=
4= 2. Hence, (r, ) =
_
2,
2
3
_
.
Convert fromrectangular to polar coordinates usingacalculator (makesureyour choiceof gives thecorrect
quadrant).
(a) (2, 3) (b) (4, 7) (c) (3, 8) (d) (5, 2)
5. Convert frompolar torectangular coordinates:
(a)
_
3,

6
_
(b)
_
6,
3
4
_
(c)
_
0,

5
_
(d)
_
5,
2
_
solution
(a) Sincer = 3and =

6
, wehave:
x = r cos = 3cos
6
= 3
3
2
2.6
y = r sin = 3sin
6
= 3
1
2
= 1.5
(x, y) (2.6, 1.5) .
(b) For
_
6,
3
4
_
wehaver = 6and =
3
4
. Hence,
x = r cos = 6cos
3
4
4.24
y = r sin = 6sin
3
4
4.24
(x, y) (4.24, 4.24) .
(c) For
_
0,

5
_
, wehaver = 0, sothat therectangular coordinatesare(x, y) = (0, 0).
(d) Sincer = 5and =
2
wehave
x = r cos = 5cos
_
2
_
= 5 0= 0
y = r sin = 5sin
_
2
_
= 5 (1) = 5
(x, y) = (0, 5)
Whichof thefollowingarepossiblepolar coordinatesfor thepoint P withrectangular coordinates(0, 2)?
(a)
_
2,
2
_
(b)
_
2,
7
2
_
(c)
_
2,
3
2
_
(d)
_
2,
7
2
_
(e)
_
2,
2
_
(f)
_
2,
7
2
_
7. Describeeachshadedsector inFigure17byinequalitiesinr and.
(A) (B) (C)
x x x
y y y
3 5 3 5 3 5
45
FIGURE 17
solution
(a) Inthesector shownbelowr is varyingbetween0and3and is varyingbetween and2. Hencethefollowing
inequalitiesdescribethesector:
0 r 3
2
(b) Inthesector shownbelowr isvaryingbetween0and3and isvaryingbetween

4
and

2
. Hence, theinequalities
for thesector are:
0 r 3
4

2
(c) Inthesector shownbelowr isvaryingbetween3and5and isvaryingbetween
3
4
and. Hence, theinequalities
are:
3 r 5
3
4

Findtheequationinpolar coordinatesof thelinethroughtheoriginwithslope
1
2
.
9. What istheslopeof theline =
3
5
?
solution Thislinemakesanangle
0
=
3
5
withthepositivex-axis, hencetheslopeof thelineism = tan
3
5
3.1.
Whichof r = 2sec andr = 2csc denesahorizontal line?
In Exercises 1116, convert to an equation in rectangular coordinates.
11. r = 7
solution r = 7describesthepointshavingdistance7fromtheorigin, that is, thecirclewithradius7centeredat the
origin. Theequationof thecircleinrectangular coordinatesis
x
2
+y
2
= 7
2
= 49.
r = sin
13. r = 2sin
solution Wemultiplytheequationbyr andsubstituter
2
= x
2
+y
2
, r sin = y. Thisgives
r
2
= 2r sin
x
2
+y
2
= 2y
Movingthe2y andcompletingthesquareyield: x
2
+ y
2
2y = 0andx
2
+ (y 1)
2
= 1. Thus, r = 2sin is the
equationof acircleof radius1centeredat (0, 1).
r = 2csc 15. r =
1
cos sin
solution Wemultiplytheequationbycos sin andsubstitutey = r sin, x = r cos. Thisgives
r (cos sin) = 1
r cos r sin = 1
x y = 1y = x 1. Thus,
r =
1
cos sin
istheequationof theliney = x 1.
r =
1
2cos
In Exercises 1720, convert to an equation in polar coordinates.
17. x
2
+y
2
= 5
solution Wemakethesubstitutionx
2
+y
2
= r
2
toobtain; r
2
= 5or r =
5.
x = 5
19. y = x
2
solution Substitutingy = r sin andx = r cos yields
r sin = r
2
cos
2
.
Then, dividingbyr cos
2
weobtain,
sin
cos
2
= r so r = tan sec
xy = 1
21. Matcheachequationwithitsdescription.
(a) r = 2 (i) Vertical line
(b) = 2 (ii) Horizontal line
(c) r = 2sec (iii) Circle
(d) r = 2csc (iv) Linethroughorigin
solution
(a) r = 2describesthepoints2unitsfromtheorigin. Hence, it istheequationof acircle.
(b) = 2describesthepointsP sothatOP makesanangleof
0
= 2withthepositivex-axis. Hence, itistheequation
of alinethroughtheorigin.
(c) Thisisr cos = 2, whichisx = 2, avertical line.
(d) Converting to rectangular coordinates, we get r = 2csc, so r sin = 2 and y = 2. This is the equation of a
horizontal line.
Findthevaluesof intheplot of r = 4cos correspondingtopointsA, B, C, D inFigure18. Thenindicatethe
portionof thegraphtracedout as variesinthefollowingintervals:
(a) 0

2
(b)

2
(c)
3
2
23. Supposethat P = (x, y) haspolar coordinates(r, ). Findthepolar coordinatesfor thepoints:
(a) (x, y) (b) (x, y) (c) (x, y) (d) (y, x)
solution
(a) (x, y) is the symmetric point of (x, y) with respect to the x-axis, hence the two points have the same radial
coordinate, andtheangular coordinateof (x, y) is2 . Hence, (x, y) = (r, 2 ).
y
x
2p q
q
q
(x, y)
(x, y)
(b) (x, y) isthesymmetricpoint of (x, y) withrespect totheorigin. Hence, (x, y) = (r, +).
y
x
p +q
q
(x, y)
(x, y)
(c) (x, y) is the symmetric point of (x, y) with respect to the y-axis. Hence the two points have the same radial
coordinatesandtheangular coordinateof (x, y) is . Hence, (x, y) = (r, ).
q q
p q
y
x
(x, y) (x, y)
(d) Let (r
1
,
1
) denotethepolar coordinatesof (y, x). Hence,
r
1
=
_
y
2
+x
2
=
_
x
2
+y
2
= r
tan
1
=
x
y
=
1
y/x
=
1
tan
= cot = tan
_
2

_
Sincethepoints(x, y) and(y, x) areinthesamequadrant, thesolutionfor
1
is
1
=

2
. Weobtainthefollowing
polar coordinates: (y, x) =
_
r,

2

_
.
q
q
p
2
q
y
x
(x, y)
(y, x)
Matcheachequationinrectangular coordinateswithitsequationinpolar coordinates.
(a) x
2
+y
2
= 4 (i) r
2
(12sin
2
) = 4
(b) x
2
+(y 1)
2
= 1 (ii) r(cos +sin) = 4
25. What arethepolar equationsof thelinesparallel totheliner cos
_

3
_
= 1?
solution Theliner cos
_

3
_
= 1, or r = sec
_

3
_
, is perpendicular to theray =

3
andat distanced = 1
fromtheorigin. Hence, thelines parallel tothis linearealsoperpendicular totheray =

3
, sothepolar equations of
theselinesarer = d sec
_

3
_
or r cos
_

3
_
= d.
Showthat thecirclewithcenter at
_
1
2
,
1
2
_
inFigure19haspolar equationr = sin + cos andndthevalues
of between0and correspondingtopointsA, B, C, andD.
27. Sketch the curve r =
1
2
(the spiral of Archimedes) for between 0 and 2 by plotting the points for =
0,

4
,

2
, . . . , 2.
solution Werst plot thefollowingpoints(r, ) onthespiral:
O = (0, 0) , A =
_
8
,
4
_
, B =
_
4
,
2
_
, C =
_
3
8
,
3
4
_
, D =
_
2
,
_
,
E =
_
5
8
,
5
4
_
, F =
_
3
4
,
3
2
_
, G =
_
7
8
,
7
4
_
, H = (, 2) .
p
4
3p
4
3p
2
5p
4
7p
4
p
2
O
D
E
A
G
C
B
0
2p
p
H
F
Sincer(0) =
0
2
= 0, thegraphbeginsat theoriginandmovestowardthepointsA, B, C, D, E, F, G andH as varies
from = 0to theother values statedabove. Connectingthepoints inthis directionweobtainthefollowinggraphfor
0 2:
p
4
3p
4
3p
2
5p
4
7p
4
p
2
O
D
E
A
G
C
B
0
2p
p
H
F
Sketchr = 3cos 1(seeExample8).
29. Sketchthecardioidcurver = 1+cos.
solution Sincecos isperiodwithperiod2, theentirecurvewill betracedoutas variesfrom0to2.Additionally,
sincecos(2 ) = cos(), wecansketchthecurvefor between0and andreect theresult throughthex axisto
obtainthewholecurve. Usethevalues = 0,

6
,

4
,

3
,

2
,
2
3
,
3
4
,
5
6
, and:
r point
0 1+cos0= 2 (2, 0)
6
1+cos
6
=
2+
3
2
_
2+
3
2
,

6
_
4
1+cos
4
=
2+
2
2
_
2+
2
2
,

4
_
3
1+cos
3
=
3
2
_
3
2
,

3
_
2
1+cos
2
= 1
_
1,

2
_
2
3
1+cos
2
3
=
1
2
_
1
2
,
2
3
_
3
4
1+cos
3
4
=
2
2
2
_
2
2
2
,
3
4
_
5
6
1+cos
5
6
=
2
3
2
_
2
3
2
,
5
6
_
= 0correspondstothepoint (2, 0), andthegraphmovesclockwiseas increasesfrom0to. Thusthegraphis
5
6
3
4
2
3
3

4

6
0
Reectingthroughthex axisgivestheother half of thecurve:
1
1
1
2
y
x
2
1 2
Showthat thecardioidof Exercise29hasequation
(x
2
+y
2
x)
2
= x
2
+y
2
inrectangular coordinates.
31. Figure20displaysthegraphsof r = sin2 inrectangular coordinatesandinpolar coordinates, whereit isarose
withfour petals. Identify:
(a) Thepointsin(B) correspondingtopointsAI in(A).
(b) Thepartsof thecurvein(B) correspondingtotheangleintervals
_
0,

2
_
,
_
2
,
_
,
_
,
3
2
_
, and
_
3
2
, 2
_
.
A C E I G
B F
D H
x
r
y
(A) Graph of r as a function
of , where r = sin 2
(B) Graph of r = sin 2
in polar coordinates

2
3 2
2

FIGURE 20
solution
(a) Thegraph(A) givesthefollowingpolar coordinatesof thelabeledpoints:
A: = 0, r = 0
B: =

4
, r = sin
2
4
= 1
C: =

2
, r = 0
D: =
3
4
, r = sin
2 3
4
= 1
E: = , r = 0
F: =
5
4
, r = 1
G: =
3
2
, r = 0
H: =
7
4
, r = 1
I: = 2, r = 0.
Sincethemaximal valueof |r| is1, thepointswithr = 1or r = 1arethefurthest pointsfromtheorigin. Thecorre-
spondingquadrant isdeterminedbythevalueof andthesignof r. If r
0
< 0, thepoint (r
0
,
0
) isontheray =
0
.
Theseconsiderationsleadtothefollowingidenticationof thepointsinthexy plane. Noticethat A, C, G, E, andI are
thesamepoint.
x
y
2
r = 1 r = 1
B

4

2
5
4
r = 1
7
4
3
4
r = 1
=
= =
=

4
3
4
7
4
3
2
5
4
H
F D
A,C,E,G,I
(b) We use the graph (A) to nd the sign of r = sin2 : 0

2
r 0 (r, ) is in the rst quadrant.
2
r 0 (r, ) is in the fourth quadrant.
3
2
r 0 (r, ) is in the third quadrant.
3
2
2 r 0(r, ) isinthesecondquadrant. That is,
x
y

3
2
2
3
2
0

2

2
Sketchthecurver = sin3. Firstll inthetableof r-valuesbelowandplotthecorrespondingpointsof thecurve.
Noticethat thethreepetals of thecurvecorrespondto theangleintervals
_
0,

3
_
,
_
3
,
2
3
_
, and
_
3
,
_
. Thenplot
r = sin3 inrectangular coordinatesandlabel thepointsonthisgraphcorrespondingto(r, ) inthetable.
0

12

6

4

3
5
12

11
12

r
33. Plot thecissoidr = 2sin tan andshowthat itsequationinrectangular coordinatesis
y
2
=
x
3
2x
solution UsingaCASweobtainthefollowingcurveof thecissoid:
x
y
0

2
3 1 2
3
2
Wesubstitutesin =
y
r
andtan =
y
x
inr = 2sin tan toobtain
r = 2
y
r
y
x
.
Multiplyingbyrx, settingr
2
= x
2
+y
2
andsimplifying, yields
r
2
x = 2y
2
(x
2
+y
2
)x = 2y
2
x
3
+y
2
x = 2y
2
y
2
(2x) = x
3
so
y
2
=
x
3
2x
Provethat r = 2a cos istheequationof thecircleinFigure21usingonlythefact that atriangleinscribedina
circlewithonesideadiameter isaright triangle.
35. Showthat
r = a cos +b sin
istheequationof acirclepassingthroughtheorigin. Expresstheradiusandcenter (inrectangular coordinates) interms
of a andb.
solution Wemultiply theequationby r andthenmakethesubstitutionx = r cos, y = r sin, andr
2
= x
2
+y
2
.
Thisgives
r
2
= ar cos +br sin
x
2
+y
2
= ax +by
Transferringsidesandcompletingthesquareyields
x
2
ax +y
2
by = 0
_
x
2
2
a
2
x +
_
a
2
_
2
_
+
_
y
2
2
b
2
y +
_
b
2
_
2
_
=
_
a
2
_
2
+
_
b
2
_
2
_
x
a
2
_
2
+
_
y
b
2
_
2
=
a
2
+b
2
4
Thisistheequationof thecirclewithradius
a
2
+b
2
2
centeredat thepoint
_
a
2
,
b
2
_
. Byplugginginx = 0andy = 0it is
clear that thecirclepassesthroughtheorigin.
Use the previous exercise to write the equation of the circle of radius 5 and center (3, 4) in the formr =
a cos +b sin.
37. Usetheidentitycos2 = cos
2
sin
2
tondapolar equationof thehyperbolax
2
y
2
= 1.
solution Wesubstitutex = r cos, y = r sin inx
2
y
2
= 1toobtain
r
2
cos
2
r
2
sin
2
= 1
r
2
(cos
2
sin
2
) = 1
Usingtheidentitycos2 = cos
2
sin
2
weobtainthefollowingequationof thehyperbola:
r
2
cos2 = 1 or r
2
= sec2.
Findanequationinrectangular coordinatesfor thecurver
2
= cos2.
39. Showthatcos3 = cos
3
3cos sin
2
andusethisidentitytondanequationinrectangular coordinatesfor the
curver = cos3.
solution We use the identities cos( + ) = cos cos sin sin, cos2 = cos
2
sin
2
, and sin2 =
2sin cos towrite
cos3 = cos(2 +) = cos2 cos sin2 sin
= (cos
2
sin
2
) cos 2sin cos sin
= cos
3
sin
2
cos 2sin
2
cos
= cos
3
3sin
2
cos
Usingthisidentitywemayrewritetheequationr = cos3 asfollows:
r = cos
3
3sin
2
cos (1)
Sincex = r cos andy = r sin, wehavecos =
x
r
andsin =
y
r
. Substitutinginto(1) gives:
r =
_
x
r
_
3
3
_
y
r
_
2
_
x
r
_
r =
x
3
r
3

3y
2
x
r
3
Wenowmultiplybyr
3
andmakethesubstitutionr
2
= x
2
+y
2
toobtainthefollowingequationfor thecurve:
r
4
= x
3
3y
2
x
(x
2
+y
2
)
2
= x
3
3y
2
x
Usetheaddition formulafor thecosineto showthat thelineL with polar equation r cos( ) = d has the
equationinrectangular coordinates (cos)x + (sin)y = d. Showthat L has slopem = cot andy-intercept
d/sin.
In Exercises 4144, nd an equation in polar coordinates of the line L with the given description.
41. Thepoint onL closest totheoriginhaspolar coordinates
_
2,

9
_
.
solution InExample5, it isshownthat thepolar equationof thelinewhere(r, ) isthepoint onthelineclosest to
theoriginisr = d sec( ). Setting(d, ) =
_
2,

9
_
weobtainthefollowingequationof theline:
r = 2sec
_

9
_
.
Thepoint onL closest totheoriginhasrectangular coordinates(2, 2).
43. L istangent tothecircler = 2
10at thepoint withrectangular coordinates(2, 6).

solution
x
y
(2, 6)
SinceL istangent tothecircleat thepoint (2, 6), thisisthepoint onL closest tothecenter of thecirclewhichisat
theorigin. Therefore, wemayusethepolar coordinates(d, ) of thispoint intheequationof theline:
r = d sec( ) (1)
Wethusmust convert thecoordinates(2, 6) topolar coordinates. Thispoint isinthethirdquadrant so < <
3
2
.
Weget
d =
_
(2)
2
+(6)
2
=
40= 2
10
= tan
1
_
6
2
_
= tan
1
3 +1.25 4.39
Substitutingin(1) yieldsthefollowingequationof theline:
r = 2
10sec( 4.39) .
L hasslope3andistangent totheunit circleinthefourthquadrant.
45. Showthat everylinethat doesnot passthroughtheoriginhasapolar equationof theform
r =
b
sin a cos
whereb = 0.
solution Writetheequationof thelineinrectangular coordinatesasy = ax +b. Sincethelinedoesnotpassthrough
theorigin, wehaveb = 0. Substitutefor y andx toconvert topolar coordinates, andsimplify:
y = ax +b
r sin = ar cos +b
r(sin a cos) = b
r =
b
sin a cos
BytheLawof Cosines, thedistanced betweentwopoints(Figure22) withpolar coordinates(r, ) and(r
0
,
0
) is
d
2
= r
2
+r
2
0
2rr
0
cos(
0
)
Usethisdistanceformulatoshowthat
r
2
10r cos
_

4
_
= 56
istheequationof thecircleof radius9whosecenter haspolar coordinates
_
5,

4
_
.
47. For a > 0, alemniscatecurveis theset of points P suchthat theproduct of thedistances fromP to (a, 0) and
(a, 0) isa
2
. Showthat theequationof thelemniscateis
(x
2
+y
2
)
2
= 2a
2
(x
2
y
2
)
Then nd the equation in polar coordinates. To obtain the simplest formof the equation, use the identity cos2 =
cos
2
sin
2
. Plot thelemniscatefor a = 2if youhaveacomputer algebrasystem.
solution Wecomputethedistancesd
1
andd
2
of P(x, y) fromthepoints(a, 0) and(a, 0) respectively. Weobtain:
d
1
=
_
(x a)
2
+(y 0)
2
=
_
(x a)
2
+y
2
d
2
=
_
(x +a)
2
+(y 0)
2
=
_
(x +a)
2
+y
2
For thepointsP(x, y) onthelemniscatewehaved
1
d
2
= a
2
. That is,
a
2
=
_
(x a)
2
+y
2
_
(x +a)
2
+y
2
=
_
_
(x a)
2
+y
2
_ _
(x +a)
2
+y
2
_
=
_
(x a)
2
(x +a)
2
+y
2
(x a)
2
+y
2
(x +a)
2
+y
4
=
_
(x
2
a
2
)
2
+y
2
_
(x a)
2
+(x +a)
2
_
+y
4
=
_
x
4
2a
2
x
2
+a
4
+y
2
_
x
2
2xa +a
2
+x
2
+2xa +a
2
_
+y
4
=
_
x
4
2a
2
x
2
+a
4
+2y
2
x
2
+2y
2
a
2
+y
4
=
_
x
4
+2x
2
y
2
+y
4
+2a
2
(y
2
x
2
) +a
4
=
_
(x
2
+y
2
)
2
+2a
2
(y
2
x
2
) +a
4
.
Squaringbothsidesandsimplifyingyields
a
4
= (x
2
+y
2
)
2
+2a
2
(y
2
x
2
) +a
4
0= (x
2
+y
2
)
2
+2a
2
(y
2
x
2
)
so
(x
2
+y
2
)
2
= 2a
2
(x
2
y
2
)
Wenowndtheequationinpolar coordinates. Wesubstitutex = r cos, y = r sin andx
2
+y
2
= r
2
intotheequation
of thelemniscate. Thisgives
(r
2
)
2
= 2a
2
(r
2
cos
2
r
2
sin
2
) = 2a
2
r
2
(cos
2
sin
2
) = 2a
2
r
2
cos2
r
4
= 2a
2
r
2
cos2
r = 0isasolution, hencetheoriginisonthecurve. For r = 0wedividetheequationby r
2
toobtainr
2
= 2a
2
cos2.
This curvealso includes theorigin(r = 0is obtainedfor =

4
for example), hencethis is thepolar equationof the
lemniscate. Settinga = 2weget r
2
= 8cos2.
r
2
= 8 cos 2q
3p
2
p
2
p 0
Let c beaxedconstant. Explaintherelationshipbetweenthegraphsof:
(a) y = f (x +c) andy = f (x) (rectangular)
(b) r = f ( +c) andr = f () (polar)
(c) y = f (x) +c andy = f (x) (rectangular)
(d) r = f () +c andr = f () (polar)
49. TheDerivativeinPolar Coordinates Showthat apolar curver = f () hasparametricequations
x = f () cos, y = f () sin
ThenapplyTheorem2of Section12.1toprove
dy
dx
=
f () cos +f
() sin
f () sin +f
() cos
2
wheref
() = df /d.
solution Multiplyingbothsides of thegivenequationby cos yields r cos = f () cos; multiplyingbothsides
bysin yieldsr sin = f () sin. Theleft-handsidesof thesetwoequationsarethex andy coordinatesinrectangular
coordinates, so for any wehavex = f () cos and y = f () sin, showing that theparametric equations areas
claimed. Now, bytheformulafor thederivativewehave
dy
dx
=
y
()
x
()
(1)
Wedifferentiatethefunctionsx = f () cos andy = f () sin usingtheProduct Rulefor differentiation. Thisgives
y
() = f
() sin +f () cos
x
() = f
() cos f () sin
Substitutingin(1) gives
dy
dx
=
f
() sin +f () cos
f
() cos f () sin
=
f () cos +f
() sin
f () sin +f
() cos
.
UseEq. (2) tondtheslopeof thetangent linetor = sin at =

3
.
51. UseEq. (2) tondtheslopeof thetangent linetor = at =

2
and = .
solution Inthegivencurvewehaver = f () = . UsingEq. (2) weobtainthefollowingderivative, whichisthe
slopeof thetangent lineat (r, ).
dy
dx
=
f () cos +f
() sin
f () sin +f
() cos
=
cos +1 sin
sin +1 cos
(1)
Theslope, m, of thetangent lineat =

2
and = isobtainedbysubstitutingthesevaluesin(1). Weget ( =

2
):
m =
2
cos
2
+sin
2
sin
2
+cos
2
=
2
0+1
2
1+0
=
1
2
=
2
.
( = ):
m =
cos +sin
sin +cos
=

1
= .
Findtheequationinrectangular coordinatesof thetangent linetor = 4cos3 at =

6
.
53. Findthepolarcoordinatesof thepointsonthelemniscater
2
= cos2t inFigure23wherethetangentlineishorizontal.
y
x
1 1
r
2
= cos (2t)
FIGURE 23
solution Thiscurveisdenedfor
2
2t

2
(wherecos2t 0), sofor
4
t

4
. For each inthat range,
therearetwovaluesof r satisfyingtheequation(
cos2t ). Bysymmetry, weneedonlycalculatethecoordinatesof the

pointscorrespondingtothepositivesquareroot (i.e. totheright of they axis). Thentheequationbecomesr =
cos2t .
Now, byEq. (2), withf (t ) =
cos(2t ) andf
(t ) = sin(2t )(cos(2t ))
1/2
, wehave
dy
dx
=
f (t ) cost +f
(t ) sint
f (t ) sint +f
(t ) cost
=
cost
cos(2t ) sin(2t ) sint (cos(2t ))

1/2
sint
cos(2t ) sin(2t ) cost (cos(2t ))

1/2
Thetangentlineishorizontal whenthisderivativeiszero, whichoccurswhenthenumerator of thefractioniszeroandthe
denominator isnot. Multiplytopandbottomof thefractionby
cos(2t ), andusetheidentitiescos2t = cos

2
t sin
2
t ,
sin2t = 2sint cost toget
cost cos2t sint sin2t

sint cos2t +cost sin2t
=
cost (cos
2
t 3sin
2
t )
sint cos2t +cost sin2t
Thenumerator iszerowhencost = 0, sowhent =

2
or t =
3
2
, or whentant =
1
3
, sowhent =
6
or t =
5
6
.
Of thesepossibilities, onlyt =
6
lieintherange
4
t

4
. Notethat thedenominator isnonzerofor t =
6
, so
thesearethetwovaluesof t for whichthetangent lineishorizontal. Thecorrespondingvaluesof r aresolutionsto
r
2
= cos
_
2

6
_
= cos
_
3
_
=
1
2
r
2
= cos
_
2

6
_
= cos
_
3
_
=
1
2
Finally, thefour pointsare(r, t ) =
_
1
2
,
6
_
,
_
2
,
6
_
,
_
1
2
,
pi
6
_
,
_
2
,
6
_
If desired, wecanchangethesecondandfourthpointsbyadding totheangleandmakingr positive, toget
_
1
2
,
6
_
,
_
1
2
,
7
6
_
,
_
1
2
,
pi
6
_
,
_
1
2
,
5
6
_
Findthepolar coordinates of thepoints onthecardioidr = 1+ cos wherethetangent lineis horizontal (see
Figure24).
55. UseEq. (2) toshowthat for r = sin +cos,
dy
dx
=
cos2 +sin2
cos2 sin2
Thencalculatetheslopesof thetangent linesat pointsA, B, C inFigure19.
solution InExercise49weprovedthat for apolar curver = f () thefollowingformulaholds:
dy
dx
=
f () cos +f
() sin
f () sin +f
() cos
(1)
For thegivencirclewehaver = f () = sin +cos, hencef
() = cos sin. Substitutingin(1) wehave

dy
dx
=
(sin +cos) cos +(cos sin) sin
(sin +cos) sin +(cos sin) cos
=
sin cos +cos
2
+cos sin sin
2
sin
2
cos sin +cos
2
sin cos
=
cos
2
sin
2
+2sin cos
cos
2
sin
2
2sin cos
Weusetheidentitiescos
2
sin
2
= cos2 and2sin cos = sin2 toobtain
dy
dx
=
cos2 +sin2
cos2 sin2
(2)
Thederivative
dy
dx
istheslopeof thetangentlineat(r, ).Theslopesof thetangentlinesatthepointswithpolarcoordinates
A =
_
1,

2
_
B =
_
0,
3
4
_
C = (1, 0) arecomputedbysubstitutingthevaluesof in(2). Thisgives
dy
dx
A
=
cos
_
2

2
_
+sin
_
2

2
_
cos
_
2

2
_
sin
_
2

2
_ =
cos +sin
cos sin
=
1+0
10
= 1
dy
dx
B
=
cos
_
2
3
4
_
+sin
_
2
3
4
_
cos
_
2
3
4
_
sin
_
2
3
4
_ =
cos
3
2
+sin
3
2
cos
3
2
sin
3
2
=
01
0+1
= 1
dy
dx
C
=
cos(2 0) +sin(2 0)
cos(2 0) sin(2 0)
=
cos0+sin0
cos0sin0
=
1+0
10
= 1
Let f (x) beaperiodic function of period 2that is, f (x) = f (x + 2). Explain howthis periodicity is
reectedinthegraphof:
(a) y = f (x) inrectangular coordinates
(b) r = f () inpolar coordinates
57. Useagraphingutilitytoconvinceyourself thatthepolarequationsr = f
1
() = 2cos 1andr = f
2
() =
2cos + 1 have the same graph. Then explain why. Hint: Show that the points (f
1
( + ), + ) and (f
2
(), )
coincide.
solution Thegraphsof r = 2cos 1andr = 2cos +1inthexy -planecoincideasshowninthegraphobtained
usingaCAS.
x
y
2
2
2 2
x
y
0

2
3 1 2
3
2
Recall that (r, ) and(r, + ) represent thesamepoint. Replacing by + andr by(r) inr = 2cos 1we
obtain
r = 2cos( +) 1
r = 2cos 1
r = 2cos +1
Thus, thetwo equations denethesamegraph. (Onecouldalso convert bothequations to rectangular coordinates and
notethat theycomeout identical.)
Weinvestigatehowtheshapeof thelimaoncurver = b +cos dependsontheconstant b (seeFigure24).
(a) ArgueasinExercise57toshowthat theconstantsb andb yieldthesamecurve.
(b) Plot thelimaonfor b = 0, 0.2, 0.5, 0.8, 1anddescribehowthecurvechanges.
(c) Plot thelimaonfor b = 1.2, 1.5, 1.8, 2, 2.4anddescribehowthecurvechanges.
(d) UseEq. (2) toshowthat
dy
dx
=
_
b cos +cos2
b +2cos
_
csc
(e) Findthepointswherethetangent lineisvertical. Notethat therearethreecases: 0 b < 2, b = 1, andb > 2.
Dotheplotsconstructedin(b) and(c) reect your results?
12.4 Area and Arc Length in Polar Coordinates
1. Polar coordinatesaresuitedtondingthearea(chooseone):
(a) Under acurvebetweenx = a andx = b.
(b) Boundedbyacurveandtworaysthroughtheorigin.
solution Polar coordinatesarebestsuitedtondingtheareaboundedbyacurveandtworaysthroughtheorigin. The
formulafor theareainpolar coordinatesgivestheareaof thisregion.
2. Istheformulafor areainpolar coordinatesvalidif f () takesnegativevalues?
solution Theformulafor thearea
1
2
_

f ()
2
d
alwaysgivestheactual (positive) area, evenif f () takesonnegativevalues.
3. Thehorizontal liney = 1haspolar equationr = csc. Whichareaisrepresentedbytheintegral
1
2
_
/2
/6
csc
2
d
(Figure12)?
(a) ABCD (b) ABC (c) ACD
y
x
A
D
B
C
y = 1
1
3
FIGURE 12
solution Thisintegral representsanareatakenfrom = /6to = /2, whichcanonlybethetriangleACD, as
seeninpart (c).
Exercises
1. Sketchtheareaboundedby thecircler = 5andtherays =

2
and = , andcomputeitsareaasanintegral in
polar coordinates.
solution Theregionboundedby thecircler = 5andtherays =

2
and = is theshadedregioninthegure.
Theareaof theregionisgivenbythefollowingintegral:
1
2
_

/2
r
2
d =
1
2
_

/2
5
2
d =
25
2
_

2
_
=
25
4
x
y
=
=

2
S E C T I ON 12.4 Area and Arc Length in Polar Coordinates 781
Sketchtheregionboundedby theliner = sec andtherays = 0and =

3
. Computeitsareaintwoways:
asanintegral inpolar coordinatesandusinggeometry.
3. Calculatetheareaof thecircler = 4sin asanintegral inpolar coordinates(seeFigure4). Becareful tochoosethe
correct limitsof integration.
solution Theequationr = 4sin denesacircleof radius2tangenttothex-axisattheoriginasshowninthegure:
=

2

3
2
3

6
5
6
x
y
= =
Thecircleistracedas variesfrom0to. Weusetheareainpolar coordinatesandtheidentity
sin
2
=
1
2
(1cos2)
toobtainthefollowingarea:
A =
1
2
_

0
r
2
d =
1
2
_

0
(4sin)
2
d = 8
_

0
sin
2
d = 4
_

0
(1cos2) d = 4
_

sin2
2
_
0
= 4
__

sin2
2
_
0
_
= 4.
Findtheareaof theshadedtriangleinFigure13as anintegral inpolar coordinates. Thenndtherectangular
coordinatesof P andQ andcomputetheareaviageometry.
5. Findtheareaof theshadedregioninFigure14. Notethat variesfrom0to

2
.
x
y
r =
2
+ 4
8
1 2
FIGURE 14
solution Since variesfrom0to

2
, theareais
1
2
_
/2
0
r
2
d =
1
2
_
/2
0
(
2
+4)
2
d =
1
2
_
/2
0

4
+8
3
+16
2
d
=
1
2
_
1
5
5
+2
4
+
16
3

3
_
/2
0
=

5
320
+

4
16
+

2
3
Whichinterval of -valuescorrespondstothetheshadedregioninFigure15? Findtheareaof theregion.
7. Findthetotal areaenclosedbythecardioidinFigure16.
y
x
1 2
FIGURE 16 Thecardioidr = 1cos.
solution Wegraphr = 1cos inr and (cartesian, not polar, thistime):
r
1
2
2
2
3
2
Weseethat as variesfrom0to, theradiusr increasesfrom0to2, soweget theupper half of thecardioid(thelower
half isobtainedas variesfrom to2 andconsequentlyr decreasesfrom2to0). Sincethecardioidissymmetricwith
respect tothex-axiswemaycomputetheupper areaanddoubletheresult. Using
cos
2
=
cos2 +1
2
weget
A = 2
1
2
_

0
r
2
d =
_

0
(1cos)
2
d =
_

0
_
12cos +cos
2
_
d
=
_

0
_
12cos +
cos2 +1
2
_
d =
_

0
_
3
2
2cos +
1
2
cos2
_
d
=
3
2
2sin +
1
4
sin2
0
=
3
2
Thetotal areaenclosedbythecardioidisA =
3
2
.
Findtheareaof theshadedregioninFigure16.
9. Findtheareaof oneleaf of thefour-petaledrose r = sin2 (Figure17). Thenprovethat thetotal areaof therose
isequal toone-half theareaof thecircumscribedcircle.
y
x
r = sin 2
FIGURE 17 Four-petaledroser = sin2.
solution Weconsider thegraphof r = sin2 incartesianandinpolar coordinates:
r
A
1
1

4

2
3
4
y
A
x
r = 1, =

4
Weseethatas variesfrom0to

4
theradiusr isincreasingfrom0to1, andwhen variesfrom
4
to

2
, r isdecreasing
back tozero. Hence, theleaf intherst quadrant istracedas variesfrom0to

2
. Theareaof theleaf (thefour leaves
haveequal areas) isthus
A =
1
2
_
/2
0
r
2
d =
1
2
_
/2
0
sin
2
2 d.
Usingtheidentity
sin
2
2 =
1cos4
2
weget
A =
1
2
_
/2
0
_
1
2
cos4
2
_
d =
1
2
_
sin4
8
_
/2
0
=
1
2
__
4

sin2
8
_
0
_
=

8
Theareaof oneleaf isA =

8
0.39. It followsthat theareaof theentireroseis

2
. Sincetheradius of therose(the
pointwhere =

4
) is1, andthecircumscribedcircleistangentthere, thecircumscribedcirclehasradius1andthusarea
. Hencetheareaof theroseishalf that of thecircumscribedcircle.
Findtheareaenclosedbyoneloopof thelemniscatewithequationr
2
= cos2 (Figure18). Chooseyour limits
of integrationcarefully.
11. Sketchthespiral r = for 0 2 andndtheareaboundedbythecurveandtherst quadrant.
solution Thespiral r = for 0 2 isshowninthefollowinggureinthexy-plane:
x
y
q = 2p,
r = 2p
q = p,
r = p
q = p/2,
r = p/2
q = 0,
r = 0
Thespiral r =
Wemust computetheareaof theshadedregion. Thisregionistracedas variesfrom0to

2
. Usingtheformulafor the
areainpolar coordinatesweget
A =
1
2
_
/2
0
r
2
d =
1
2
_
/2
0

2
d =
1
2
3
3
/2
0
=
1
6
_
2
_
3
=

3
48
Findtheareaof theintersectionof thecirclesr = sin andr = cos.
13. Findtheareaof regionA inFigure19.
y
x
1 4 1 2
A
r = 4 cos
r = 1
FIGURE 19
solution Werstndthevaluesof atthepointsof intersectionof thetwocircles, bysolvingthefollowingequation
for
2
x

2
:
4cos = 1cos =
1
4
1
= cos
1
_
1
4
_
y
x
r = 4 cos
= 1.32
= 1.32
r = 1
Wenowcomputetheareausingtheformulafor theareabetweentwocurves:
A =
1
2
_

1
1
_
(4cos)
2
1
2
_
d =
1
2
_

1
1
_
16cos
2
1
_
d
Usingtheidentitycos
2
=
cos2+1
2
weget
A =
1
2
_

1
1
_
16(cos2 +1)
2
1
_
d =
1
2
_

1
1
(8cos2 +7) d =
1
2
(4sin2 +7)
1
= 4sin2
1
+7
1
= 8sin
1
cos
1
+7
1
= 8
_
1cos
2
1
cos
1
+7
1
Usingthefact that cos
1
=
1
4
weget
A =
15
2
+7cos
1
_
1
4
_
11.163
Findtheareaof theshadedregioninFigure20, enclosedbythecircler =
1
2
andapetal of thecurver = cos3.
Hint: Computetheareaof boththepetal andtheregioninsidethepetal andoutsidethecircle.
15. Findtheareaof theinner loopof thelimaonwithpolar equationr = 2cos 1(Figure21).
2 1
1
1
y
x
FIGURE 21 Thelimaonr = 2cos 1.
solution Weconsider thegraphof r = 2cos 1incartesianandinpolar, for
2
x

2
:
r
1
1

2

3

3

2
y
x

3

3
r = 2cos 1
As varies from
3
to 0, r increases from0to 1. As varies from0to

3
, r decreases from1back to 0. Hence, the
inner loopof thelimaonistracedas variesfrom
3
to

3
. Theareaof theinner loopisthus
A =
1
2
_
/3
/3
r
2
d =
1
2
_
/3
/3
(2cos 1)
2
d =
1
2
_
/3
/3
_
4cos
2
4cos +1
_
d
=
1
2
_
/3
/3
(2(cos2 +1) 4cos +1) d =
1
2
_
/3
/3
(2cos2 4cos +3) d
=
1
2
(sin2 4sin +3)
/3
/3
=
1
2
__
sin
2
3
4sin
3
+
_
_
sin
_
2
3
_
4sin
_
3
_
__
=
3
2

4
3
2
+ =
3
3
2
0.54
Findtheareaof theshadedregioninFigure21betweentheinner andouter loopof thelimaonr = 2cos 1.
17. Findtheareaof thepart of thecircler = sin +cos inthefourthquadrant (seeExercise26inSection12.3).
solution Thevalueof correspondingtothepoint B isthesolutionof r = sin +cos = 0for .
y
x
B A C
That is,
sin +cos = 0sin = cos tan = 1 =
4
At thepoint C, wehave = 0. Thepart of thecircleinthefourthquadrant istracedif variesbetween
4
and0. This
leadstothefollowingarea:
A =
1
2
_
0
/4
r
2
d =
1
2
_
0
/4
(sin +cos)
2
d =
1
2
_
0
/4
_
sin
2
+2sin cos +cos
2
_
d
Usingtheidentitiessin
2
+cos
2
= 1and2sin cos = sin2 weget:
A =
1
2
_
0
/4
(1+sin2) d =
1
2
_

cos2
2
_
0
/4
=
1
2
_
_
0
1
2
_
4

cos
_
2
_
2
__
=
1
2
_
4

1
2
_
=

8

1
4
0.14.
Findtheareaof theregioninsidethecircler = 2sin
_
+

4
_
andabovetheliner = sec
_

4
_
.
19. FindtheareabetweenthetwocurvesinFigure22(A).
y y
x x
r = 2 + cos 2q
r = 2 + sin 2q
r = sin 2q
r = sin 2q
(A) (B)
FIGURE 22
solution WecomputetheareaAbetweenthetwocurvesasthedifferencebetweentheareaA
1
of theregionenclosed
intheouter curver = 2+cos2 andtheareaA
2
of theregionenclosedintheinner curver = sin2. That is,
A = A
1
A
2
.
y
x
A
A
2
r = 2 + 2cos
r = sin
InExercise9weshowedthat A
2
=

2
, hence,
A = A
1

2
(1)
WecomputetheareaA
1
.
y
x
A
1
Usingsymmetry, theareaisfour timestheareaenclosedintherst quadrant. That is,
A
1
= 4
1
2
_
/2
0
r
2
d = 2
_
/2
0
(2+cos2)
2
d = 2
_
/2
0
_
4+4cos2 +cos
2
2
_
d
Usingtheidentitycos
2
2 =
1
2
cos4 +
1
2
weget
A
1
= 2
_
/2
0
_
4+4cos2 +
1
2
cos4 +
1
2
_
d = 2
_
/2
0
_
9
2
+
1
2
cos4 +4cos2
_
d
= 2
_
9
2
+
sin4
8
+2sin2
_
/2
0
= 2
__
9
4
+
sin2
8
+2sin
_
0
_
=
9
2
(2)
Combining(1) and(2) weobtain
A =
9
2

2
= 4.
FindtheareabetweenthetwocurvesinFigure22(B).
21. FindtheareainsidebothcurvesinFigure23.
y
x
2 + sin 2q
2 + cos 2q
FIGURE 23
solution Theareaweneedtondistheareaof theshadedregioninthegure.
y
x
A
D
C
B
r = 2 + sin 2
r = 2 + cos 2
Werst nd thevalues of at thepoints of intersection A, B, C, and D of thetwo curves, by solving thefollowing
equationfor :
2+cos2 = 2+sin2
cos2 = sin2
tan2 = 12 =

4
+k =

8
+
k
2
Thesolutionsfor are
A: =

8
.
B: =
3
8
.
C: =
7
8
.
D: =
5
8
.
Usingsymmetry, wecomputetheshadedareainthegurebelowandmultiplyit by4:
r = 2 + cos 2
0

8

2

2
5
8
A
1
A = 4 A
1
= 4
1
2
_
5/8
/8
(2+cos2)
2
d = 2
_
5/8
/8
_
4+4cos2 +cos
2
2
_
d
= 2
_
5/8
/8
_
4+4cos2 +
1+cos4
2
_
d =
_
5/8
/8
(9+8cos2 +cos4) d
= 9 +4sin2 +
sin4
4
5/8
/8
= 9
_
5
8

8
_
+4
_
sin
5
4
sin
4
_
+
1
4
_
sin
5
2
sin
2
_
=
9
2
4
2
Findtheareaof theregionthat liesinsideonebut not bothof thecurvesinFigure23.
23. Calculatethetotal lengthof thecircler = 4sin asanintegral inpolar coordinates.
solution Weusetheformulafor thearclength:
S =
_

_
f ()
2
+f
()
2
d (1)
Inthiscase, f () = 4sin andf
() = 4cos, hence
_
f ()
2
+f
()
2
=
_
(4sin)
2
+(4cos)
2
=
16= 4
Thecircleistracedas isvariedfrom0to. Substituting = 0, = in(1) yieldsS =
_
0
4d = 4.
2
y
x
Thecircler = 4sin
Sketch thesegment r = sec for 0 A. Then computeits length in two ways: as an integral in polar
coordinatesandusingtrigonometry.
In Exercises 2530, compute the length of the polar curve.
25. Thelengthof r =
2
for 0
solution Weusetheformulafor thearclength. Inthiscasef () =
2
, f
() = 2, soweobtain
S =
_

0
_
_
2
_
2
+(2)
2
d =
_

0
_
4
+4
2
d =
_

0

_
2
+4d
Wecomputetheintegral usingthesubstitutionu =
2
+4, du = 2 d. Thisgives
S =
1
2
_

2
+4
4
u du =
1
2
2
3
u
3/2
2
+4
4
=
1
3
_
_
2
+4
_
3/2
4
3/2
_
=
1
3
_
_
2
+4
_
3/2
8
_
14.55
Thespiral r = for 0 A
27. Theequiangular spiral r = e
for 0 2
solution Sincef () = e
, bytheformulafor thearclengthwehave:
L =
_
2
0
_
f
()
2
+f () d +
_
2
0
_
_
e
_
2
+
_
e
_
2
d =
_
2
0
_
2e
2
d
=
2
_
2
0
e
d =
2e
2
0
=
2
_
e
2
e
0
_
=
2
_
e
2
1
_
755.9
Theinner loopof r = 2cos 1inFigure21
29. Thecardioidr = 1cos inFigure16
solution Intheequationof thecardioid, f () = 1cos. Usingtheformulafor arclengthinpolar coordinateswe
have:
L =
_

_
f ()
2
+f
()
2
d (1)
Wecomputetheintegrand:
_
f ()
2
+f
()
2
=
_
(1cos)
2
+(sin)
2
=
_
12cos +cos
2
+sin
2
=
_
2(1cos)
Weidentifytheinterval of . Since1 cos 1, every0 2 correspondstoanonnegativevalueof r. Hence,
variesfrom0to2. By(1) weobtain
L =
_
2
0
_
2(1cos) d
Now, 1 cos = 2sin
2
(/2), and on the interval 0 , sin(/2) is nonnegative, so that
2(1cos) =
_
4sin
2
(/2) = 2sin(/2) there. Thegraphissymmetric, soit sufcestocomputetheintegral for 0 , andwe
have
L = 2
_

0
_
2(1cos) d = 2
_

0
2sin(/2) d = 8sin
0
= 8
r = cos
2

In Exercises 31 and 32, express the length of the curve as an integral but do not evaluate it.
31. r = (2cos)
1
, 0 2
solution Wehavef () = (2cos)
1
, f
() = (2cos)
2
sin, hence,
_
f
2
() +f
()
2
=
_
(2cos)
2
+(2cos)
4
sin
2
=
_
(2cos)
4
_
(2cos)
2
+sin
2
_
= (2cos)
2
_
44cos +cos
2
+sin
2
= (2cos)
2
54cos
Usingtheintegral for thearclengthweget
L =
_
2
0
54cos(2cos)
2
d.
r = sin
3
t , 0 2
In Exercises 3336, use a computer algebra system to calculate the total length to two decimal places.
33. Thethree-petal roser = cos3 inFigure20
solution Wehavef () = cos3, f
() = 3sin3, sothat
_
f ()
2
+f
()
2
=
_
cos
2
3 +9sin
2
3 =
_
cos
2
3 +sin
2
3 +8sin
2
3 =
_
1+8sin
2
3
Notethatthecurveistraversedcompletelyfor 0 . UsingthearclengthformulaandevaluatingwithMaplegives
L =
_

0
_
f ()
2
+f
()
2
d =
_

0
_
1+8sin
2
3 d 6.682446608
Thecurver = 2+sin2 inFigure23
35. Thecurver = sin inFigure24for 0 4
y
x
5 5
5
10
FIGURE 24 r = sin for 0 4.
solution Wehavef () = sin, f
() = sin + cos, sothat

_
f ()
2
+f
()
2
=
_
2
sin
2
+(sin + cos)
2
=
_
2
sin
2
+sin
2
+2 sin cos +
2
cos
2
=
_
2
+sin
2
+ sin2
usingtheidentitiessin
2
+ cos
2
= 1and2sin cos = sin2. Thusby thearc lengthformulaandevaluatingwith
Maple, wehave
L =
_
4
0
_
f ()
2
+f
()
2
d =
_
4
0
_
2
+sin
2
+ sin2 d 79.56423976
r =
, 0 4
37. Supposethat thepolar coordinates of amovingparticleat timet are(r(t ), (t )). Provethat theparticles speedis
equal to
_
(dr/dt )
2
+r
2
(d/dt )
2
.
solution Thespeedof theparticleinrectangular coordinatesis:
ds
dt
=
_
x
(t )
2
+y
(t )
2
(1)
Weneedtoexpressthespeedinpolar coordinates. Thex andy coordinatesof themovingparticlesasfunctionsof t are
x(t ) = r(t ) cos(t ), y(t ) = r(t ) sin(t )
S E C T I ON 12.5 Conic Sections 789
Wedifferentiatex(t ) andy(t ), usingtheProductRulefor differentiation. Weobtain(omittingtheindependentvariablet )
x
= r
cos r (sin)
= r
sin r (cos)
Hence,
x
2
+y
2
=
_
r
cos r
sin
_
2
+
_
r
sin +r
cos
_
2
= r
2
cos
2
2r
cos sin +r
2
2
sin
2
+r
2
sin
2
+2r
sin
2
cos +r
2
2
cos
2
= r
2
_
cos
2
+sin
2
_
+r
2
2
_
sin
2
+cos
2
_
= r
2
+r
2
2
(2)
Substituting(2) into(1) weget
ds
dt
=
_
r
2
+r
2
2
=
_
_
dr
dt
_
2
+r
2
_
d
dt
_
2
Computethespeedat timet = 1of aparticlewhosepolar coordinatesat timet arer = t , = t (useExercise
37). Whatwouldthespeedbeif theparticlesrectangularcoordinateswerex = t , y = t ?Whyisthespeedincreasing
inonecaseandconstant intheother? 12.5 Conic Sections
1. Whichof thefollowingequationsdenesanellipse?Whichdoesnot deneaconicsection?
(a) 4x
2
9y
2
= 12 (b) 4x +9y
2
= 0
(c) 4y
2
+9x
2
= 12 (d) 4x
3
+9y
3
= 12
solution
(a) Thisistheequationof thehyperbola
_
x
3
_
2
_
y
2
3
_
2
= 1, whichisaconicsection.
(b) Theequation4x +9y
2
= 0canberewrittenasx =
9
4
y
2
, whichdenesaparabola. Thisisaconicsection.
(c) Theequation 4y
2
+9x
2
= 12can berewritten in theform
_
y
3
_
2
+
_
x
2
3
_
2
= 1, henceit is theequation of an
ellipse, whichisaconicsection.
(d) Thisisnot theequationof aconicsection, sinceit isnot anequationof degreetwoinx andy.
2. For whichconicsectionsdotheverticesliebetweenthefoci?
solution If theverticesliebetweenthefoci, theconicsectionisahyperbola.
y
x
Vertex
Vertex
Vertex
Vertex Focus Focus
F
1
F
2
Vertex Vertex Focus Focus
x
y
F
2
F
1
ellipse: foci betweenvertices hyperbola: verticesbetweenfoci
3. What arethefoci of
_
x
a
_
2
+
_
y
b
_
2
= 1 if a < b?
solution If a < b thefoci of theellipse
_
x
a
_
2
+
_
y
b
_
2
= 1areat thepoints (0, c) and(0, c) onthey-axis, where
c =
_
b
2
a
2
.
F
1
= (0, c)
F
2
= (0, c)
y
x
b
a
_
x
a
_
2
+
_
y
b
_
2
= 1; a < b
4. What isthegeometricinterpretationof b/a intheequationof ahyperbolainstandardposition?
solution Thevertices, i.e., thepointswherethefocal axisintersectsthehyperbola, areatthepoints(a, 0) and(a, 0).
Thevalues
b
a
aretheslopesof thetwoasymptotesof thehyperbola.
x
y
y = x
(a, 0) (a, 0)
b
a
y = x
b
b
b
a
Hyperbolainstandardposition
Exercises
In Exercises 16, nd the vertices and foci of the conic section.
1.
_
x
9
_
2
+
_
y
4
_
2
= 1
solution Thisisanellipseinstandardpositionwitha = 9andb = 4. Hence, c =
_
9
2
4
2
=
65 8.06. Thefoci
areat F
1
= (8.06, 0) andF
2
= (8.06, 0), andtheverticesare(9, 0) , (9, 0), (0, 4) , (0, 4).
x
2
9
+
y
2
4
= 1
3.
_
x
4
_
2
_
y
9
_
2
= 1
solution Thisisahyperbolainstandardpositionwitha = 4andb = 9. Hence, c =
_
a
2
+b
2
=
97 9.85. The
foci areat (
97, 0) andtheverticesare(2, 0).

x
2
4

y
2
9
= 36
5.
_
x 3
7
_
2
_
y +1
4
_
2
= 1
solution Werst consider thehyperbola
_
x
7
_
2
_
y
4
_
2
= 1. For thishyperbola, a = 7, b = 4andc =
_
7
2
+4
2
8.06. Hence, thefoci areat (8.06, 0) and(8.06, 0) andtheverticesareat (7, 0) and(7, 0). Sincethegivenhyperbola
is obtained by translating the center of the hyperbola
_
x
7
_
2
_
y
4
_
2
= 1 to the point (3, 1), the foci are at F
1
=
(8.06+3, 01) = (11.06, 1) andF
2
= (8.06+3, 01) = (5.06, 1) andtheverticesareA = (7+3, 01) =
(10, 1) andA
= (7+3, 01) = (4, 1).

_
x 3
4
_
2
+
_
y +1
7
_
2
= 1
In Exercises 710, nd the equation of the ellipse obtained by translating (as indicated) the ellipse
_
x 8
6
_
2
+
_
y +4
3
_
2
= 1
7. Translatedwithcenter at theorigin
solution Recall that theequation
(x h)
2
a
2
+
(y k)
2
b
2
= 1
describesanellipsewithcenter (h, k). Thus, for our ellipsetobelocatedat theorigin, it must haveequation
x
2
6
2
+
y
2
3
2
= 1
Translatedwithcenter at (2, 12)
9. Translatedtotheright sixunits
solution Recall that theequation
(x h)
2
a
2
+
(y k)
2
b
2
= 1
describesanellipsewithcenter(h, k).Theoriginal ellipsehascenterat(8, 4), sowewantanellipsewithcenter(14, 4).
Thusitsequationis
(x 14)
2
6
2
+
(y +4)
2
3
2
= 1
Translateddownfour units
In Exercises 1114, nd the equation of the given ellipse.
11. Vertices(5, 0) and(0, 7)
solution Sincebothsets of vertices aresymmetric aroundtheorigin, thecenter of theellipseis at (0, 0). Wehave
a = 5andb = 7, sotheequationof theellipseis
_
x
5
_
2
+
_
y
7
_
2
= 1
Foci (6, 0) andfocal vertices(10, 0)
13. Foci (0, 10) andeccentricitye =
3
5
solution Sincethefoci areonthey axis, thisellipsehasavertical major axiswithcenter (0, 0), soitsequationis
_
x
b
_
2
+
_
y
a
_
2
= 1
Wehavea =
c
e
=
10
3/5
=
50
3
and
b =
_
a
2
c
2
=
_
2500
9
100=
1
3
2500900=
40
3
Thustheequationof theellipseis
_
x
40/3
_
2
+
_
y
50/3
_
2
= 1
Vertices(4, 0), (28, 0) andeccentricitye =
2
3
In Exercises 1520, nd the equation of the given hyperbola.
15. Vertices(3, 0) andfoci (5, 0)
solution Theequationis
_
x
a
_
2
_
y
b
_
2
= 1. Theverticesare(a, 0) witha = 3andthefoci (c, 0) withc = 5. We
usetherelationc =
_
a
2
+b
2
tondb:
b =
_
c
2
a
2
=
_
5
2
3
2
=
16= 4
Therefore, theequationof thehyperbolais
_
x
3
_
2
_
y
4
_
2
= 1.
Vertices(3, 0) andasymptotesy =
1
2
x
17. Foci (4, 0) andeccentricitye = 2
solution Wehavec = 4ande = 2; fromc = ae weget a = 2, andthen
b =
_
c
2
a
2
=
_
4
2
2
2
= 2
3
Thehyperbolahascenter at (0, 0) andhorizontal axis, soitsequationis
_
x
2
_
2
_
y
2
3
_
2
= 1
Vertices(0, 6) andeccentricitye = 3
19. Vertices(3, 0), (7, 0) andeccentricitye = 3
solution Thecenter isat
3+7
2
= 2withahorizontal focal axis, sotheequationis
_
x 2
a
_
2
_
y
b
_
2
= 1.
Thena = 72= 5, andc = ae = 5 3= 15. Finally,
b =
_
c
2
a
2
=
_
15
2
5
2
= 10
2
sothat theequationof thehyperbolais
_
x 2
5
_
2
_
y
10
2
_
2
= 1
Vertices(0, 6), (0, 4) andfoci (0, 9), (0, 7)
In Exercises 2128, nd the equation of the parabola with the given properties.
21. Vertex(0, 0), focus
_
1
12
, 0
_
solution Sincethefocus is onthex-axis rather thanthey-axis, andthevertex is (0, 0), theequationis x =
1
4c
y
2
.
Thefocusis(0, c) withc =
1
12
, sotheequationis
x =
1
4
1
12
y
2
= 3y
2
Vertex(0, 0), focus(0, 2)
23. Vertex(0, 0), directrixy = 5
solution Theequationisy =
1
4c
x
2
. Thedirectrixisy = c withc = 5, hencey =
1
20
x
2
.
Vertex(3, 4), directrixy = 2
25. Focus(0, 4), directrixy = 4
solution Thefocusis(0, c) withc = 4andthedirectrixisy = c withc = 4, hencetheequationof theparabolais
y =
1
4c
x
2
=
x
2
16
.
Focus(0, 4), directrixy = 4
27. Focus(2, 0), directrixx = 2
solution Thefocusisonthex-axisrather thanonthey-axisandthedirectrixisavertical linerather thanhorizontal
as intheparabolainstandardposition. Therefore, theequationof theparabolais obtainedby interchangingx andy in
y =
1
4c
x
2
. Also, bythegiveninformationc = 2. Hence, x =
1
4c
y
2
=
1
42
y
2
or x =
y
2
8
.
Focus(2, 0), vertex(2, 0)
In Exercises 2938, nd the vertices, foci, center (if an ellipse or a hyperbola), and asymptotes (if a hyperbola).
29. x
2
+4y
2
= 16
solution Werst dividetheequationby16toconvert it totheequationinstandardform:
x
2
16
+
4y
2
16
= 1
x
2
16
+
y
2
4
= 1
_
x
4
_
2
+
_
y
2
_
2
= 1
For thisellipse, a = 4andb = 2hencec =
_
4
2
2
2
=
12 3.5. Sincea > b wehave:

Theverticesareat (4, 0), (0, 2).
Thefoci areF
1
= (3.5, 0) andF
2
= (3.5, 0).
Thefocal axisisthex-axisandtheconjugateaxisisthey-axis.
Theellipseiscenteredat theorigin.
4x
2
+y
2
= 16 31.
_
x 3
4
_
2
_
y +5
7
_
2
= 1
solution For this hyperbola a = 4 and b = 7 so c =
_
4
2
+7
2
=

65 8.06. For the standard hyperbola
_
x
4
_
2
_
y
7
_
2
= 1, wehave
TheverticesareA = (4, 0) andA
= (4, 0).
Thefoci areF = (
65, 0) andF
= (
65, 0).
Thefocal axisisthex-axisy = 0, andtheconjugateaxisisthey-axisx = 0.
Thecenter isat themidpoint of FF
; that is, at theorigin.

Theasymptotesy =
b
a
x arey =
7
4
x.
Thegivenhyperbolaisatranslationof thestandardhyperbola, 3unitstotheright and5unitsdownward. Hencethe
followingholds:
Theverticesareat A = (7, 5) andA
= (1, 5).
Thefoci areat F = (3+
65, 5) andF
= (3
65, 5).
Thefocal axisisy = 5andtheconjugateaxisisx = 3.
Thecenter isat (3, 5).
Theasymptotesarey +5=
7
4
(x 3).
3x
2
27y
2
= 12
33. 4x
2
3y
2
+8x +30y = 215
solution Sincethereisnocrossterm, wecompletethesquareof thetermsinvolvingx andy separately:
4x
2
3y
2
+8x +30y = 4
_
x
2
+2x
_
3
_
y
2
10y
_
= 4(x +1)
2
43(y 5)
2
+75= 215
Hence,
4(x +1)
2
3(y 5)
2
= 144
4(x +1)
2
144

3(y 5)
2
144
= 1
_
x +1
6
_
2
_
y 5
48
_
2
= 1
Thisistheequationof thehyperbolaobtainedbytranslatingthehyperbola
_
x
6
_
2
_
y
48
_
2
= 1oneunit totheleft and
veunitsupwards. Sincea = 6, b =
48, wehavec =
36+48=
84 9.2. Weobtainthefollowingtable:
Standardposition Translatedhyperbola
vertices (6, 0), (6, 0) (5, 5), (7, 5)
foci (9.2, 0) (8.2, 5), (10.2, 5)
focal axis Thex-axis y = 5
conjugateaxis They-axis x = 1
center Theorigin (1, 5)
asymptotes y = 1.15x
y = 1.15x +3.85
y = 1.15x +6.15
y = 4x
2
35. y = 4(x 4)
2
solution By Exercise34, theparabolay = 4x
2
has thevertex at theorigin, thefocus at
_
0,
1
16
_
andits axis is the
y-axis. Our parabolaisatranslationof thestandardparabolafour unitstotheright. Henceitsvertexisat(4, 0), thefocus
isat
_
4,
1
16
_
anditsaxisisthevertical linex = 4.
8y
2
+6x
2
36x 64y +134= 0
37. 4x
2
+25y
2
8x 10y = 20
solution Sincetherearenocrosstermsthisconicsectionisobtainedbytranslatingaconicsectioninstandardposition.
Toidentifytheconicsectionwecompletethesquareof thetermsinvolvingx andy separately:
4x
2
+25y
2
8x 10y = 4
_
x
2
2x
_
+25
_
y
2
2
5
y
_
= 4(x 1)
2
4+25
_
y
1
5
_
2
1
= 4(x 1)
2
+25
_
y
1
5
_
2
5= 20
Hence,
4(x 1)
2
+25
_
y
1
5
_
2
= 25
4
25
(x 1)
2
+
_
y
1
5
_
2
= 1
_
x 1
5
2
_
2
+
_
y
1
5
_
2
= 1
Thisistheequationof theellipseobtainedbytranslatingtheellipseinstandardposition
_
x
5
2
_
2
+y
2
= 1oneunit tothe
right and
1
5
unit upward. Sincea =
5
2
, b = 1wehavec =
_
_
5
2
_
2
1 2.3, soweobtainthefollowingtable:
Standardposition Translatedellipse
Vertices
_
5
2
, 0
_
, (0, 1)
_
1
5
2
,
1
5
_
,
_
1,
1
5
1
_
Foci (2.3, 0) , (2.3, 0)
_
1.3,
1
5
_
,
_
3.3,
1
5
_
Focal axis Thex-axis y =
1
5
Conjugateaxis They-axis x = 1
Center Theorigin
_
1,
1
5
_
16x
2
+25y
2
64x 200y +64= 0
In Exercises 3942, use the Discriminant Test to determine the type of the conic section (in each case, the equation is
nondegenerate). Plot the curve if you have a computer algebra system.
39. 4x
2
+5xy +7y
2
= 24
solution Here, D = 254 4 7= 87, sotheconicsectionisanellipse.
x
2
2xy +y
2
+24x 8= 0
41. 2x
2
8xy +3y
2
4= 0
solution Here, D = 644 2 3= 40, givingusahyperbola.
2x
2
3xy +5y
2
4= 0
43. Showthat theconic x
2
+3y
2
6x +12y +23= 0hasnopoints.
solution Completethesquareineachvariableseparately:
23= x
2
6x +3y
2
+12y = (x
2
6x +9) +(3y
2
+12y +12) 912= (x 3)
2
+3(y +2)
2
21
Collectingconstantsandreversingsidesgives
(x 3)
2
+3(y +2)
2
= 2
whichhasnosolutionssincetheleft-handsideisasumof twosquaressoisalwaysnonnegative.
For whichvaluesof a doestheconic3x
2
+2y
2
16y +12x = a haveat least onepoint?
45. Showthat
b
a
=
_
1e
2
for astandardellipseof eccentricitye.
solution Bythedenitionof eccentricity:
e =
c
a
(1)
For theellipseinstandardposition, c =
_
a
2
b
2
. Substitutinginto(1) andsimplifyingyields
e =
_
a
2
b
2
a
=
_
a
2
b
2
a
2
=
_
1
_
b
a
_
2
Wesquarethetwosidesandsolvefor
b
a
:
e
2
= 1
_
b
a
_
2
_
b
a
_
2
= 1e
2
b
a
=
_
1e
2
Showthat theeccentricity of ahyperbolainstandardpositionise =
_
1+m
2
, wherem aretheslopesof the
asymptotes.
47. ExplainwhythedotsinFigure23lieonaparabola. Wherearethefocusanddirectrixlocated?
y = c
y = c
y = 2c
y = 3c
y
x
FIGURE 23
solution All thecirclesarecenteredat(0, c) andthekthcirclehasradiuskc. HencetheindicatedpointP
k
onthekth
circlehasadistancekc fromthepoint F = (0, c). Thepoint P
k
alsohasdistancekc fromtheliney = c. That is, the
indicatedpoint oneachcircleisequidistant fromthepoint F = (0, c) andtheliney = c, henceit liesontheparabola
withfocusat F = (0, c) anddirectrixy = c.
y = c
(0, c) 2c
2c
3c
3c
P
2
P
3
P
1
y
x
Find theequation of theellipseconsisting of points P such that PF
1
+ PF
2
= 12, whereF
1
= (4, 0) and
F
2
= (2, 0).
49. A latusrectumof aconic sectionisachordthroughafocusparallel tothedirectrix. Findtheareaboundedby the
parabolay = x
2
/(4c) anditslatusrectum(refer toFigure8).
solution Thedirectrix isy = c, andthefocusis(0, c). Thechordthroughthefocusparallel toy = c isclearly
y = c; thislineintersectstheparabolawhenc = x
2
/(4c) or 4c
2
= x
2
, sowhenx = 2c. Thedesiredareaisthen
_
2c
2c
c
1
4c
x
2
dx =
_
c x
1
12c
x
3
_
2c
2c
= 2c
2
8c
3
12c
_
2c
2
(2c)
3
12c
_
= 4c
2
4
3
c
2
=
8
3
c
2
Showthat thetangent lineat apoint P = (x
0
, y
0
) onthehyperbola
_
x
a
_
2
_
y
b
_
2
= 1hasequation
Ax By = 1
whereA =
x
0
a
2
andB =
y
0
b
2
.
In Exercises 5154, nd the polar equation of the conic with the given eccentricity and directrix, and focus at the origin.
51. e =
1
2
, x = 3
solution Substitutinge =
1
2
andd = 3inthepolar equationof aconicsectionweobtain
r =
ed
1+e cos
=
1
2
3
1+
1
2
cos
=
3
2+cos
r =
3
2+cos
e =
1
2
, x = 3
53. e = 1, x = 4
solution Wesubstitutee = 1andd = 4inthepolar equationof aconicsectiontoobtain
r =
ed
1+e cos
=
1 4
1+1 cos
=
4
1+cos
r =
4
1+cos
e =
3
2
, x = 4
In Exercises 5558, identify the type of conic, the eccentricity, and the equation of the directrix.
55. r =
8
1+4cos
solution Matchingwiththepolar equationr =
ed
1+e cos
weget ed = 8ande = 4yieldingd = 2. Sincee > 1, the
conicsectionisahyperbola, havingeccentricitye = 4anddirectrixx = 2(referringtothefocus-directrixdenition(11)).
r =
8
4+cos
57. r =
8
4+3cos
solution Werst rewritetheequationintheformr =
ed
1+e cos
, obtaining
r =
2
1+
3
4
cos
Hence, ed = 2ande =
3
4
yieldingd =
8
3
. Sincee < 1, theconic sectionis anellipse, havingeccentricity e =
3
4
and
directrixx =
8
3
.
r =
12
4+3cos
59. Findapolar equationfor thehyperbolawithfocusat theorigin, directrixx = 2, andeccentricitye = 1.2.
solution Wesubstituted = 2ande = 1.2inthepolar equationr =
ed
1+e cos
anduseExercise40toobtain
r =
1.2 (2)
1+1.2cos
=
2.4
1+1.2cos
=
12
5+6cos
=
12
56cos
Let C betheellipser = de/(1+e cos), wheree < 1. Showthat thex-coordinates of thepoints inFigure24
areasfollows:
Point A C F
2
A
de de
2
2de
2
de
61. Findanequationinrectangular coordinatesof theconic
r =
16
5+3cos
Hint: Usetheresultsof Exercise60.
solution Put thisequationintheformof thereferencedexercise:
16
5+3cos
=
16
5
1+
3
5
cos
=
16
3

3
5
1+
3
5
cos
sothat e =
3
5
andd =
16
3
. Thenthecenter of theellipsehasx-coordinate
de
2
1e
2
=
16
3

9
25
1
9
25
=
16
3

9
25
25
16
= 3
andy-coordinate0, andA
hasx-coordinate
de
1e
=
16
3

3
5
1
3
5
=
16
3

3
5
5
2
= 8
andy-coordinate0, soa = 3(8) = 5, andtheequationis
_
x +3
5
_
2
+
_
y
b
_
2
= 1
Tondb, set =

2
; thenr =
16
5
. But thepoint correspondingto =

2
liesonthey-axis, sohascoordinates
_
0,
16
5
_
.
Thispoint isontheellipse, sothat
_
0+3
5
_
2
+
_
16
5
b
_
2
= 1
256
25 b
2
=
16
25

256
b
2
= 16 b = 4
andtheequationis
_
x +3
5
_
2
+
_
y
4
_
2
= 1
Let e > 1. Showthat theverticesof thehyperbolar =
de
1+e cos
havex-coordinates
ed
e +1
and
ed
e 1
.
63. KeplersFirstLawstatesthatplanetaryorbitsareellipseswiththesunatonefocus. Theorbitof Plutohaseccentricity
e 0.25. Its perihelion(closest distanceto thesun) is approximately 2.7 billion miles. Find theaphelion(farthest
distancefromthesun).
solution Wedeneanxy-coordinatesystemsothattheorbitisanellipseinstandardposition, asshowninthegure.
y
x
Sun
F
1
(c, 0)
A(a, 0) A' (a, 0)
Theaphelionisthelengthof A
F
1
, that isa +c. Bythegivendata, wehave
0.25= e =
c
a
c = 0.25a
a c = 2.7c = a 2.7
Equatingthetwoexpressionsfor c weget
0.25a = a 2.7
0.75a = 2.7a =
2.7
0.75
= 3.6, c = 3.62.7= 0.9
Theaphelionisthus
A
F
0
= a +c = 3.6+0.9= 4.5billionmiles.
KeplersThirdLawstates that theratioT/a
3/2
is equal toaconstant C for all planetary orbits aroundthesun,
whereT istheperiod(timefor acompleteorbit) anda isthesemimajor axis.
(a) ComputeC inunitsof daysandkilometers, giventhat thesemimajor axisof theearthsorbit is15010
6
km.
(b) Computetheperiodof Saturnsorbit, giventhat itssemimajor axisisapproximately1.4310
9
km.
65. VerifyTheorem2.
solution Let F
1
= (c, 0) andF
2
= (c, 0) andlet P (x, y) beanarbitrary point onthehyperbola. Thenfor some
constant a,
PF
1
PF
2
= 2a
y
x
F
2
= (c, 0) F
1
= (c, 0)
P = (x, y)
Usingthedistanceformulawewritethisas
_
(x c)
2
+y
2
_
(x +c)
2
+y
2
= 2a.
Movingthesecondtermtotheright andsquaringbothsidesgives
_
(x c)
2
+y
2
=
_
(x +c)
2
+y
2
2a
(x c)
2
+y
2
= (x +c)
2
+y
2
4a
_
(x +c)
2
+y
2
+4a
2
(x c)
2
(x +c)
2
4a
2
= 4a
_
(x +c)
2
+y
2
xc +a
2
= a
_
(x +c)
2
+y
2
Wesquareandsimplifytoobtain
x
2
c
2
+2xca
2
+a
4
= a
2
_
(x +c)
2
+y
2
_
= a
2
x
2
+2a
2
xc +a
2
c
2
+a
2
y
2
_
c
2
a
2
_
x
2
a
2
y
2
= a
2
_
c
2
a
2
_
x
2
a
2

y
2
c
2
a
2
= 1
For b =
_
c
2
a
2
(or c =
_
a
2
+b
2
) weget
x
2
a
2

y
2
b
2
= 1
_
x
a
_
2
_
y
b
_
2
= 1.
VerifyTheorem5inthecase0< e < 1. Hint: Repeat theproof of Theorem5, but set c = d/(e
2
1).
67. Verify that if e > 1, thenEq. (11) denesahyperbolaof eccentricity e, withitsfocusat theoriginanddirectrix at
x = d.
solution ThepointsP = (r, ) onthehyperbolasatisfyPF = ePD, e > 1. Referringtothegureweseethat
PF = r, PD = d r cos (1)
Hence
r = e(d r cos)
r = ed er cos
r(1+e cos) = ed r =
ed
1+e cos
F
r
P
q
D
y
x
x = d
D
d rcosq
rcosq
Remark: Equality(1) holdsalsofor >

2
. For example, inthefollowinggure, wehave
PD = d +r cos( ) = d r cos
y
x
P
r
d r cos (p q )
q
x = d
D
ReectivePropertyof theEllipse In Exercises 6870, we prove that the focal radii at a point on an ellipse make equal
angles with the tangent line L. Let P = (x
0
, y
0
) be a point on the ellipse in Figure 25 with foci F
1
= (c, 0) and
F
2
= (c, 0), and eccentricity e = c/a.
R
2
= (
2
,
2
)
R
1
= (
1
,
1
)
1
2
y
x
P = (x
0
, y
0
)
F
1
= (c, 0) F
2
= (c, 0)
L
FIGURE 25 Theellipse
_
x
a
_
2
+
_
y
b
_
2
= 1.
Showthat theequationof thetangent lineat P isAx +By = 1, whereA =
x
0
a
2
andB =
y
0
b
2
.
69. PointsR
1
andR
2
inFigure25aredenedsothat F
1
R
1
andF
2
R
2
areperpendicular tothetangent line.
(a) Show, withA andB asinExercise68, that
1
+c
1
=

2
c
2
=
A
B
(b) Use(a) andthedistanceformulatoshowthat
F
1
R
1
F
2
R
2
=

1
2
(c) Use(a) andtheequationof thetangent lineinExercise68toshowthat
1
=
B(1+Ac)
A
2
+B
2
,
2
=
B(1Ac)
A
2
+B
2
solution
(a) SinceR
1
= (
1
,
1
) andR
2
= (
2
,
2
) lieonthetangent lineat P, that isonthelineAx +By = 1, wehave
A
1
+B
1
= 1 and A
2
+B
2
= 1
Theslopeof thelineR
1
F
1
is

1
1
+c
andit isperpendicular tothetangent linehavingslope
A
B
. Similarly, theslopeof
thelineR
2
F
2
is

2
2
c
andit isalsoperpendicular tothetangent line. Hence,
1
+c
1
=
A
B
and

2
c
2
=
A
B
.
(b) Usingthedistanceformula, wehave
R
1
F
1
2
= (
1
+c)
2
+
2
1
Thus,
R
1
F
1
2
=
2
1
_
_
1
+c
1
_
2
+1
_
(1)
Bypart (a),

1
+c
1
=
A
B
. Substitutingin(1) gives
R
1
F
1
2
=
2
1
_
A
2
B
2
+1
_
(2)
Likewise,
R
2
F
2
2
= (
2
c)
2
+
2
2
=
2
2
_
_
2
c
2
_
2
+1
_
(3)
but since
2
c
2
=
A
B
, substitutingin(3) gives
R
2
F
2
2
=
2
2
_
A
2
B
2
+1
_
. (4)
Dividing, wendthat
R
1
F
1
2
R
2
F
2
2
=

2
1
2
2
so
R
1
F
1
R
2
F
2
=

1
2
,
asdesired.
(c) Inpart (a) weshowedthat
A
1
+B
1
= 1
1
+c
=
B
A
Eliminating
1
andsolvingfor
1
gives
1
=
B(1+Ac)
A
2
+B
2
. (5)
Similarly, wehave
A
2
+B
2
= 1
2
c
=
B
A
Eliminating
2
andsolvingfor
2
yields
2
=
B (1Ac)
A
2
+B
2
(6)
(a) Provethat PF
1
= a +x
0
e andPF
2
= a x
0
e. Hint: Showthat PF
1
2
PF
2
2
= 4x
0
c. Thenusethedening
propertyPF
1
+PF
2
= 2a andtherelatione = c/a.
(b) Verifythat
F
1
R
1
PF
1
=
F
2
R
2
PF
2
.
(c) Showthat sin
1
= sin
2
. Concludethat
1
=
2
.
71. Hereisanother proof of theReectiveProperty.
(a) Figure25suggeststhat L istheuniquelinethat intersectstheellipseonlyinthepoint P. Assumingthis, provethat
QF
1
+QF
2
> PF
1
+PF
2
for all pointsQ onthetangent lineother thanP.
(b) UsethePrincipleof Least Distance(Example6inSection4.7) toprovethat
1
=
2
.
solution
(a) Consider apoint Q = P onthelineL (seegure). SinceL intersectstheellipseinonlyonepoint, theremainder of
thelineliesoutsidetheellipse, sothat QR doesnot havezerolength, andF
2
QR isatriangle. Thus
QF
1
+QF
2
= QR +RF
1
+QF
2
= RF
1
+(QR +QF
2
) > RF
1
+RF
2
sincethesumof lengthsof twosidesof atriangleexceedsthelengthof thethirdside. ButsincepointR liesontheellipse,
RF
2
+RF
2
= PF
1
+PF
2
, andwearedone.
y
Q
R
P
x
F
1
F
2
(b) Consider abeamof light travelingfromF
1
toF
2
by reectionoff of thelineL. By theprincipleof least distance,
thelight takestheshortest path, whichbypart (a) isthepaththroughP. ByExample6inSection4.6, thisshortest path
hasthepropertythat theangleof incidence(
1
) isequal totheangleof reection(
2
).
Showthat thelengthQR inFigure26isindependent of thepoint P.
73. Showthat y = x
2
/4c istheequationof aparabolawithdirectrixy = c, focus(0, c), andthevertexat theorigin,
asstatedinTheorem3.
solution ThepointsP = (x, y) ontheparabolaareequidistant fromF = (0, c) andtheliney = c.
y
x
y = c
P(x, y)
F(0, c)
That is, bythedistanceformula, wehave
PF = PD
_
x
2
+(y c)
2
= |y +c|
Squaringandsimplifyingyields
x
2
+(y c)
2
= (y +c)
2
x
2
+y
2
2yc +c
2
= y
2
+2yc +c
2
x
2
2yc = 2yc
x
2
= 4yc y =
x
2
4c
Thus, we showed that the points that are equidistant fromthe focus F = (0, c) and the directrix y = c satisfy the
equationy =
x
2
4c
.
Consider twoellipsesinstandardposition:
E
1
:
_
x
a
1
_
2
+
_
y
b
1
_
2
= 1
E
2
:
_
x
a
2
_
2
+
_
y
b
2
_
2
= 1
Wesay that E
1
issimilar toE
2
under scalingif thereexistsafactor r > 0suchthat for all (x, y) onE
1
, thepoint
(rx, ry) lies onE
2
. Showthat E
1
andE
2
aresimilar under scalingif andonly if they havethesameeccentricity.
Showthat anytwocirclesaresimilar under scaling.
75. DeriveEqs. (13) and(14) inthetext asfollows. Writethecoordinatesof P withrespect totherotatedaxes
inFigure21inpolar formx
= r cos, y
= r sin. ExplainwhyP haspolar coordinates(r, +) withrespect tothe

standardx andy-axesandderiveEqs. (13) and(14) usingtheadditionformulasfor cosineandsine.
solution If thepolar coordinates of P withrespect totherotatedaxes are(r, ), thenthelinefromtheorigintoP
haslengthr andmakesanangleof withtherotatedx-axis(thex
-axis). Sincethex
-axismakesanangleof withthe
x-axis, it followsthat thelinefromtheorigintoP makesanangleof + withthex-axis, sothat thepolar coordinates
of P withrespect tothestandardaxesare(r, + ). Write(x
, y
) for therectangular coordinatesof P withrespect to

therotatedaxesand(x, y) for therectangular coordinatesof P withrespect tothestandardaxes. Then
x = r cos( +) = (r cos) cos (r sin) sin = x
cos y
sin
y = r sin( +) = r sin cos +r cos sin = (r cos) sin +(r sin) cos = x
sin +y
cos
If werewritethegeneral equationof degree2(Eq. 12) intermsof variablesx
andy
thatarerelatedtox andy by
Eqs. (13) and(14), weobtainanewequationof degree2inx
andy
of thesameformbutwithdifferentcoefcients:
1. Whichof thefollowingcurvespassthroughthepoint (1, 4)?
(a) c(t ) = (t
2
, t +3) (b) c(t ) = (t
2
, t 3)
(c) c(t ) = (t
2
, 3t ) (d) c(t ) = (t 3, t
2
)
solution Tocheckwhetheritpassesthroughthepoint(1, 4), wesolvetheequationsc(t ) = (1, 4) forthegivencurves.
(a) Comparingthesecondcoordinateof thecurveandthepoint yields:
t +3= 4
t = 1
Wesubstitutet = 1intherst coordinate, toobtain
t
2
= 1
2
= 1
Hencethecurvepassesthrough(1, 4).
(b) Comparingthesecondcoordinateof thecurveandthepoint yields:
t 3= 4
t = 7
Wesubstitutet = 7intherst coordinatetoobtain
t
2
= 7
2
= 49= 1
Hencethecurvedoesnot passthrough(1, 4).
(c) Comparingthesecondcoordinateof thecurveandthepoint yields
3t = 4
t = 1
Wesubstitutet = 1intherst coordinate, toobtain
t
2
= (1)
2
= 1
Hencethecurvepassesthrough(1, 4).
(d) Comparingtherst coordinateof thecurveandthepoint yields
t 3= 1
t = 4
Wesubstitutet = 4inthesecondcoordinate, toobtain:
t
2
= 4
2
= 16= 4
Hencethecurvedoesnot passthrough(1, 4).
Findparametricequationsfor thelinethroughP = (2, 5) perpendicular totheliney = 4x 3.
3. Findparametricequationsforthecircleof radius2withcenter(1, 1).Usetheequationstondthepointsof intersection
of thecirclewiththex- andy-axes.
solution Usingthestandardtechniquefor parametricequationsof curves, weobtain
c(t ) = (1+2cost, 1+2sint )
Wecomparethex coordinateof c(t ) to0:
1+2cost = 0
cost =
1
2
t =
2
3
Substitutinginthey coordinateyields
1+2sin
_
2
3
_
= 12
3
2
= 1
3
Hence, theintersectionpointswiththey-axisare(0, 1
3). Wecomparethey coordinateof c(t ) to0:

1+2sint = 0
sint =
1
2
t =
6
or
7
6
Substitutinginthex coordinatesyields
1+2cos
_
6
_
= 1+2
3
2
= 1+
3
1+2cos
_
7
6
_
= 12cos
_
6
_
= 12
3
2
= 1
3
Hence, theintersectionpointswiththex-axisare(1
3, 0).
Findaparametrizationc(t ) of theliney = 52x suchthat c(0) = (2, 1).
5. Findaparametrizationc() of theunit circlesuchthat c(0) = (1, 0).
solution Theunit circlehastheparametrization
c(t ) = (cost, sint )
Thisparametrizationdoesnot satisfyc(0) = (1, 0). Wereplacetheparameter t byaparameter sothat t = +, to
obtainanother parametrizationfor thecircle:
c
() = (cos( +), sin( +)) (1)

Weneedthat c
(0) = (1, 0), that is,

c
(0) = (cos, sin) = (1, 0)

Hence
cos = 1
sin = 0
=
Substitutingin(1) weobtainthefollowingparametrization:
c
() = (cos( +), sin( +))

Findapathc(t ) that tracestheparabolicarcy = x
2
from(0, 0) to(3, 9) for 0 t 1.
7. Findapathc(t ) that tracestheliney = 2x +1from(1, 3) to(3, 7) for 0 t 1.
solution Solution 1: By one of the examples in section 12.1, the line through P = (1, 3) with slope 2 has the
parametrization
c(t ) = (1+t, 3+2t )
But thisparametrizationdoesnot satisfyc(1) = (3, 7). Wereplacetheparameter t byaparameter s sothat t = s +.
Weget
c
(s) = (1+s +, 3+2(s +)) = (s + +1, 2s +2 +3)

Weneedthat c
(0) = (1, 3) andc
(1) = (3, 7). Hence,

c
(0) = (1+, 3+2) = (1, 3)

c
(1) = ( + +1, 2 +2 +3) = (3, 7)

Weobtaintheequations
1+ = 1
3+2 = 3
+ +1= 3
2 +2 +3= 7
= 0, = 2
Substitutingin(1) gives
c
(s) = (2s +1, 4s +3)

Solution2: Thesegment from(1, 3) to(3, 7) hasthefollowingvector parametrization:
(1t ) 1, 3 +t 3, 7 = 1t +3t, 3(1t ) +7t = 1+2t, 3+4t
Theparametrizationisthus
c(t ) = (1+2t, 3+4t )
Sketchthegraphc(t ) = (1+cost, sin2t ) for 0 t 2 anddrawarrowsspecifyingthedirectionof motion.
In Exercises 912, express the parametric curve in the form y = f (x).
9. c(t ) = (4t 3, 10t )
solution Weusethegivenequationtoexpresst intermsof x.
x = 4t 3
4t = x +3
t =
x +3
4
Substitutingintheequationof y yields
y = 10t = 10
x +3
4
=
x
4
+
37
4
That is,
y =
x
4
+
37
4
c(t ) = (t
3
+1, t
2
4)
11. c(t ) =
_
3
2
t
, t
3
+
1
t
_
solution Weusethegivenequationtoexpresst intermsof x:
x = 3
2
t
2
t
= 3x
t =
2
3x
Substitutingintheequationof y yields
y =
_
2
3x
_
3
+
1
2/(3x)
=
8
(3x)
3
+
3x
2
x = tant , y = sect
In Exercises 1316, calculate dy/dx at the point indicated.
13. c(t ) = (t
3
+t, t
2
1), t = 3
solution Theparametric equationsarex = t
3
+ t andy = t
2
1. Weusethetheoremontheslopeof thetangent
linetond
dy
dx
:
dy
dx
=
dy
dt
dx
dt
=
2t
3t
2
+1
Wenowsubstitutet = 3toobtain
dy
dx
t =3
=
2 3
3 3
2
+1
=
3
14
c() = (tan
2
, cos), =

4
15. c(t ) = (e
t
1, sint ), t = 20
solution Weusethetheoremfor theslopeof thetangent linetond
dy
dx
:
dy
dx
=
dy
dt
dx
dt
=
(sint )
(e
t
1)
=
cost
e
t
Wenowsubstitutet = 20:
dy
dx
t =0
=
cos20
e
20
c(t ) = (lnt, 3t
2
t ), P = (0, 2)
17. Findthepoint onthecycloidc(t ) = (t sint, 1cost ) wherethetangent linehasslope
1
2
.
solution Sincex = t sint andy = 1cost , thetheoremontheslopeof thetangent linegives
dy
dx
=
dy
dt
dx
dt
=
sint
1cost
Thepointswherethetangent linehasslope
1
2
arethosewhere
dy
dx
=
1
2
. Wesolvefor t :
dy
dx
=
1
2
sint
1cost
=
1
2
(1)
2sint = 1cost
Welet u = sint . Thencost =
_
1sin
2
t =
_
1u
2
. Hence
2u = 1
_
1u
2
Wetransfer sidesandsquaretoobtain
_
1u
2
= 2u 1
1u
2
= 4u
2
4u +1
5u
2
4u = u(5u 4) = 0
u = 0, u =
4
5
Wendt bytherelationu = sint :
u = 0: sint = 0t = 0, t =
u =
4
5
: sint =
4
5
t 0.93, t 2.21
Thesecorrespondtothepoints(0, 1), (, 2), (0.13, 0.40), and(1.41, 1.60), respectively, for 0< t < 2.
Findthepointson(t +sint, t 2sint ) wherethetangent isvertical or horizontal.
19. Findtheequationof theBzier curvewithcontrol points
P
0
= (1, 1), P
1
= (1, 1), P
2
= (1, 1), P
3
(1, 1)
solution Wesubstitutethegivenpointsintheappropriateformulasinthetext tondtheparametricequationsof the
Bzier curve. Weobtain
x(t ) = (1t )
3
3t (1t )
2
+t
2
(1t ) +t
3
= (13t +3t
2
t
3
) (3t 6t
2
+3t
3
) +(t
2
t
3
) +t
3
= (2t
3
+4t
2
1)
y(t ) = (1t )
3
+3t (1t )
2
+t
2
(1t ) t
3
= (13t +3t
2
t
3
) +(3t 6t
2
+3t
3
) +(t
2
t
3
) t
3
= (2t
3
8t
2
+6t 1)
Findthespeedat t =

4
of aparticlewhosepositionat timet secondsisc(t ) = (sin4t, cos3t ).
21. Findthespeed(asafunctionof t ) of aparticlewhosepositionattimet secondsisc(t ) = (sint +t, cost +t ). What
istheparticlesmaximal speed?
solution Weusetheparametricdenitiontondthespeed. Weobtain
ds
dt
=
_
((sint +t )
)
2
+((cost +t )
)
2
=
_
(cost +1)
2
+(1sint )
2
=
_
cos
2
t +2cost +1+12sint +sin
2
t =
_
3+2(cost sint )
Wenowdifferentiatethespeedfunctiontonditsmaximum:
d
2
s
dt
2
=
_
_
3+2(cost sint )
_
=
sint cost
3+2(cost sint )
Weequatethederivativetozero, toobtainthemaximumpoint:
d
2
s
dt
2
= 0
sint cost
3+2(cost sint )
= 0
sint cost = 0
sint = cost
sin(t ) = cos(t )
t =

4
+k
t =
4
+k
Substitutingt inthefunctionof speedweobtainthevalueof themaximal speed:
_
3+2
_
cos
4
sin
4
_
=
_
3+2
_
2
2

_
2
2
__
=
_
3+2
2
Findthelengthof (3e
t
3, 4e
t
+7) for 0 t 1.
In Exercises 23 and 24, let c(t ) = (e
t
cost, e
t
sint ).
23. Showthat c(t ) for 0 t < hasnitelengthandcalculateitsvalue.
solution Weusetheformulafor arclength, toobtain:
s =
_

0
_
((e
t
cost )
)
2
+((e
t
sint )
)
2
dt
=
_

0
_
(e
t
cost e
t
sint )
2
+(e
t
sint +e
t
cost )
2
dt
=
_

0
_
e
2t
(cost +sint )
2
+e
2t
(cost sint )
2
dt
=
_

0
e
t
_
cos
2
t +2sint cost +sin
2
t +cos
2
t 2sint cost +sin
2
t dt
=
_

0
e
t
2dt =
2(e
t
)
0
=
2
_
lim
t
e
t
e
0
_
=
2(01) =
2
Findtherst positivevalueof t
0
suchthat thetangent linetoc(t
0
) isvertical, andcalculatethespeedat t = t
0
.
25. Plot c(t ) = (sin2t, 2cost ) for 0 t . Express the length of the curve as a denite integral, and
approximateit usingacomputer algebrasystem.
solution WeuseaCAStoplot thecurve. Theresultinggraphisshownhere.
x
y
2
1
2
1
2 1 2 1
Plot of thecurve(sin2t, 2cost )
Tocalculatethearclengthweusetheformulafor thearclengthtoobtain
s =
_

0
_
(2cos2t )
2
+(2sint )
2
dt = 2
_

0
_
cos
2
2t +sin
2
t dt
WeuseaCAStoobtains = 6.0972.
Convert thepoints(x, y) = (1, 3), (3, 1) fromrectangular topolar coordinates.
27. Convert thepoints(r, ) =
_
1,

6
_
,
_
3,
5
4
_
frompolar torectangular coordinates.
solution Weconvert thepointsfrompolar coordinatestocartesiancoordinates. For therst point wehave
x = r cos = 1 cos
6
=
3
2
y = r sin = 1 sin
6
=
1
2
For thesecondpoint wehave
x = r cos = 3cos
5
4
=
3
2
2
y = r sin = 3sin
5
4
=
3
2
2
Write(x +y)
2
= xy +6asanequationinpolar coordinates.
29. Writer =
2cos
cos sin
asanequationinrectangular coordinates.
solution Weusetheformulafor convertingfrompolar coordinatestocartesiancoordinatestosubstitutex andy for
r and:
r =
2cos
cos sin
_
x
2
+y
2
=
2r cos
r cos r sin
_
x
2
+y
2
=
2x
x y
Showthat r =
4
7cos sin
isthepolar equationof aline.
31. Convert theequation
9(x
2
+y
2
) = (x
2
+y
2
2y)
2
topolar coordinates, andplot it withagraphingutility.
solution Weusetheformulafor convertingfromcartesiancoordinatestopolar coordinatestosubstituter and for
x andy:
9(x
2
+y
2
) = (x
2
+y
2
2y)
2
9r
2
= (r
2
2r sin)
2
3r = r
2
2r sin
3= r 2sin
r = 3+2sin
Theplot of r = 3+2sin isshownhere:
r = 3+ 2sin
5
4 0 3 1 4 2 1 2 3
2
4
3
2
1
0
1
Plot of r = 3+2sin
Calculatetheareaof thecircler = 3sin boundedbytherays =

3
and =
2
3
.
33. Calculatetheareaof onepetal of r = sin4 (seeFigure1).
y
x
n = 2 (4 petals)
y
x
n = 4 (8 petals)
y
x
n = 6 (12 petals)
FIGURE 1 Plot of r = sin(n).
solution WeuseaCAStogeneratetheplot, asshownhere.
r = 4sin
0.8 0.4
1
0.8
0.6
0.4
0.2
0
0.2
0.4
0.6
0.8
1
1 0 0.8 0.4 1
Plot of r = sin4
Wecanseethat oneleaf liesbetweentherays = 0and =

4
. Wenowusetheformulafor areainpolar coordinatesto
obtain
A =
1
2
_
/4
0
sin
2
4 d =
1
4
_
/4
0
(1cos8) d =
1
4
_

sin8
8
/4
0
_
=

16
1
32
(sin2 sin0) =

16
Theequationr = sin(n), wheren 2iseven, isarose of 2n petals(Figure1). Computethetotal areaof the
ower, andshowthat it doesnot dependonn.
35. Calculatethetotal areaenclosedbythecurver
2
= cose
sin
(Figure2).
y
x
1
1 1
FIGURE 2 Graphof r
2
= cose
sin
.
solution Notethatthisisdenedonlyfor between/2and/2. Weusetheformulafor areainpolar coordinates
toobtain:
A =
1
2
_
/2
/2
r
2
d =
1
2
_
/2
/2
cose
sin
d
Weevaluatetheintegral bymakingthesubstitutionx = sin dx = cos d:
A =
1
2
_
/2
/2
cose
sin
d =
1
2
e
x
1
1
=
1
2
_
e e
1
_
FindtheshadedareainFigure3.
37. Findtheareaenclosedbythecardioidr = a(1+cos), wherea > 0.
solution Thegraphof r = a (1+cos) inther-planefor 0 2 andthecardioidinthexy-planeareshown
inthefollowinggures:
r
a
2a
2
2
3
2
y
x
= 0
r = 2a
= , r = a
3
2
= , r = a
2
= , r = 0
r = a (1+cos) Thecardioidr = a (1+cos), a > 0
As variesfrom0to theradiusr decreasesfrom2a to0, andthisgivestheupper part of thecardioid.
Thelower part is traced as varies from to 2 and consequently r increases from0 back to 2a. Wecompute
the area enclosed by the upper part of the cardioid and the x-axis, using the following integral (we use the identity
cos
2
=
1
2
+
1
2
cos2):
1
2
_

0
r
2
d =
1
2
_

0
a
2
(1+cos)
2
d =
a
2
2
_

0
_
1+2cos +cos
2
_
d
=
a
2
2
_

0
_
1+2cos +
1
2
+
1
2
cos2
_
d =
a
2
2
_

0
_
3
2
+2cos +
1
2
cos2
_
d
=
a
2
2
_
3
2
+2sin +
1
4
sin2
_
0
=
a
2
2
_
3
2
+2sin +
1
4
sin2 0
_
=
3a
2
4
Usingsymmetry, thetotal areaA enclosedbythecardioidis
A = 2
3a
2
4
=
3a
2
2
Calculatethelengthof thecurvewithpolar equationr = inFigure4.
39. Figure5showsthegraphof r = e
0.5
sin for 0 2. Useacomputer algebrasystemtoapproximate
thedifferenceinlengthbetweentheouter andinner loops.
y
x
5
10
3 6
FIGURE 5
solution Wenotethat theinner loopisthecurvefor [0, ], andtheouter loopisthecurvefor [, 2]. We
expressthelengthof theseloopsusingtheformulafor thearclength. Thelengthof theinner loopis
s
1
=
_

0
_
(e
0.5
sin)
2
+((e
0.5
sin)
)
2
d =
_

0
_
e
sin
2
+
_
e
0.5
sin
2
+e
0.5
cos
_
2
d
andthelengthof theouter loopis
s
2
=
_
2
_
e
sin
2
+
_
e
0.5
sin
2
+e
0.5
cos
_
2
d
WenowusetheCAStocalculatethearclengthof eachof theloops. Weobtainthatthelengthof theinner loopis7.5087
andthelengthof theouter loopis36.121, hencetheouter oneis4.81timeslonger thantheinner one.
Showthat r = f
1
() andr = f
2
() denethesamecurvesinpolar coordinatesif f
1
() = f
2
( +). Use
thistoshowthat thefollowingdenethesameconicsection:
r =
de
1e cos
, r =
de
1+e cos
In Exercises 4144, identify the conic section. Find the vertices and foci.
41.
_
x
3
_
2
+
_
y
2
_
2
= 1
solution This is an ellipse in standard position. Its foci are (
_
3
2
2
2
, 0) = (
5, 0) and its vertices are

(3, 0), (0, 2).
x
2
2y
2
= 4
43.
_
2x +
1
2
y
_
2
= 4(x y)
2
solution Wesimplifytheequation:
_
2x +
1
2
y
_
2
= 4(x y)
2
4x
2
+2xy +
1
4
y
2
= 4x
2
+2xy y
2
5x
2
+
5
4
y
2
= 4
5x
2
4
+
5y
2
16
= 1
x
2
2
+
y
4
2
= 1
This is an ellipsein standard position, with foci
_
0,
_
_
4
5
_
2
_
2
5
_
2
_
=
_
0,
_
12
5
_
and vertices
_
5
, 0
_
,
_
0,
4
5
_
.
(y 3)
2
= 2x
2
1
In Exercises 4550, nd the equation of the conic section indicated.
45. Ellipsewithvertices(8, 0) andfoci (
3, 0)
solution Sincethefoci of thedesiredellipseareonthex-axis, weconcludethat a > b. Wearegiventhat thepoints
(8, 0) areverticesof theellipse, andsincethey areonthex-axis, a = 8. Wearegiventhat thefoci are(
3, 0) and
wehaveshownthat a > b, hencewehavethat
_
a
2
b
2
=
3. Solvingfor b yields
_
a
2
b
2
=
3
a
2
b
2
= 3
8
2
b
2
= 3
b
2
= 61
b =
61
Next weusea andb toconstruct theequationof theellipse:
_
x
8
_
2
+
_
y
61
_
2
= 1.
Ellipsewithfoci (8, 0), eccentricity
1
8
47. Hyperbolawithvertices(8, 0), asymptotesy =
3
4
x
solution Since the asymptotes of the hyperbola are y =
3
4
x, and the equation of the asymptotes for a general
hyperbolainstandardpositionis y =
b
a
x, weconcludethat
b
a
=
3
4
. Wearegiventhat thevertices are(8, 0), thus
a = 8. Wesubstituteandsolvefor b:
b
a
=
3
4
b
8
=
3
4
b = 6
Next weusea andb toconstruct theequationof thehyperbola:
_
x
8
_
2
_
y
6
_
2
= 1.
Hyperbolawithfoci (2, 0) and(10, 0), eccentricitye = 4
49. Parabolawithfocus(8, 0), directrixx = 8
solution Thisissimilar totheusual equationof aparabola, but wemust usey asx, andx asy, toobtain
x =
1
32
y
2
.
Parabolawithvertex(4, 1), directrixx = 15
51. Findtheasymptotesof thehyperbola3x
2
+6x y
2
10y = 1.
solution Wecompletethesquaresandsimplify:
3x
2
+6x y
2
10y = 1
3(x
2
+2x) (y
2
+10y) = 1
3(x
2
+2x +11) (y
2
+10y +2525) = 1
3(x +1)
2
3(y +5)
2
+25= 1
3(x +1)
2
(y +5)
2
= 21
_
y +5
21
_
2
_
x +1
7
_
2
= 1
Weobtainedahyperbolawithfocal axisthat isparallel tothey-axis, andisshifted5unitsonthey-axis, and1units
inthex-axis. Therefore, theasymptotesare
x +1=
21
(y +5) or y +5=
3(x +1).
Showthat theconicsection withequationx
2
4x +y
2
+5= 0hasnopoints.
53. Showthat therelation
dy
dx
= (e
2
1)
x
y
holdsonastandardellipseor hyperbolaof eccentricitye.
solution Wedifferentiatetheequationsof thestandardellipseandthehyperbolawithrespect tox:
Ellipse: Hyperbola:
x
2
a
2
+
y
2
b
2
= 1
2x
a
2
+
2y
b
2
dy
dx
= 0
dy
dx
=
b
2
a
2
x
y
x
2
a
2

y
2
b
2
= 1
2x
a
2

2y
b
2
dy
dx
= 0
dy
dx
=
b
2
a
2
x
y
Theeccentricityof theellipseise =
a
2
b
2
a
, hencee
2
a
2
= a
2
b
2
or e
2
= 1
b
2
a
2
yielding
b
2
a
2
= 1e
2
.
Theeccentricity of thehyperbolais e =
a
2
+b
2
a
, hencee
2
a
2
= a
2
+ b
2
or e
2
= 1+
b
2
a
2
, giving
b
2
a
2
= e
2
1.
Combiningwiththeexpressionsfor
dy
dx
weget:
Ellipse: Hyperbola:
dy
dx
= (1e
2
)
x
y
= (e
2
1)
x
y
dy
dx
= (e
2
1)
x
y
We, thus, provedthat therelation
dy
dx
= (e
2
1)
x
y
holdsonastandardellipseor hyperbolaof eccentricitye.
Theorbit of J upiter isanellipsewiththesunat afocus. Findtheeccentricityof theorbit if theperihelion(closest
distancetothesun) equals74010
6
kmandtheaphelion(farthest distancefromthesun) equals81610
6
km.
55. Refer toFigure25inSection12.5. Provethat theproduct of theperpendicular distancesF
1
R
1
andF
2
R
2
fromthe
foci toatangent lineof anellipseisequal tothesquareb
2
of thesemiminor axes.
solution Werst consider theellipseinstandardposition:
x
2
a
2
+
y
2
b
2
= 1
Theequationof thetangent lineat P = (x
0
, y
0
) is
x
0
x
a
2
+
y
0
y
b
2
= 1
or
b
2
x
0
x +a
2
y
0
y a
2
b
2
= 0
Thedistancesof thefoci F
1
= (c, 0) andF
2
= (c, 0) fromthetangent lineare
F
1
R
1
=
|b
2
x
0
c a
2
b
2
|
_
b
4
x
2
0
+a
4
y
2
0
; F
2
R
2
=
|b
2
x
0
c +a
2
b
2
|
_
b
4
x
2
0
+a
4
y
2
0
Wecomputetheproduct of thedistances:
F
1
R
1
F
2
R
2
=
_
b
2
x
0
c a
2
b
2
_ _
b
2
x
0
c +a
2
b
2
_
b
4
x
2
0
+a
4
y
2
0
b
4
x
2
0
c
2
a
4
b
4
b
4
x
2
0
+a
4
y
2
0
(1)
Thepoint P = (x
0
, y
0
) liesontheellipse, hence:
x
2
0
a
2
+
y
2
0
b
2
= 1a
4
y
2
0
= a
4
b
2
a
2
b
2
x
2
0
Wesubstitutein(1) toobtain(noticethat b
2
a
2
= c
2
)
F
1
R
1
F
2
R
2
=
|b
4
x
2
0
c
2
a
4
b
4
|
|b
4
x
2
0
+a
4
b
2
a
2
b
2
x
2
0
|
=
|b
4
x
2
0
c
2
a
4
b
4
|
|b
2
(b
2
a
2
)x
2
0
+a
4
b
2
|
=
|b
4
x
2
0
c
2
a
4
b
4
|
| b
2
x
2
0
c
2
+a
4
b
2
|
=
|b
2
(x
2
0
c
2
a
4
)|
| (x
2
0
c
2
a
4
)|
= | b
2
| = b
2
Theproduct F
1
R
1
F
2
R
2
remainsunchangedif wetranslatethestandardellipse.

Calculus 2nd Edition - Single Variable Solutions

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Calculus 2nd Edition - Single Variable Solutions

Uploaded by

Copyright:

Available Formats

This page intentionally left blank

Students Solutions Manual

< 5 (d) |a| > 5

5. Findthelengthsof thearcssubtendedbytheangles and radiansinFigure20.

T asinExample2. Estimatetheinstantaneousrateof changeof v withrespect toT whenT = 300K.

C) ataltitudeh metersaboveacertainpointonearthisT = 150.0065h

T asinExample2. Istherateof changeof v withrespecttoT greater atlowtemperaturesor high

L asinExercise21. Thenumbersinthesecondcolumnof Table4areincreasing, andthoseinthe

numericallyfor several valuesof n. Thenguessthevalueingeneral.

2.955202 2.999550 2.999996 2.999996 2.999550 2.955202

4.794255 4.997917 4.999979 4.999979 4.997917 4.794255

x 0.99 0.9999 1.0001 1.01

x 0.99 0.9999 1.0001 1.01

x andsinx arecontinuous, sois

0.10x for 0 x < 8350

1+h andrewriteasalimit asx 1.

1+h andrewriteasalimit asx 1.

andl(x) = sinx onthesameset of axes. What canyousay about

1, whichholds for x = 1, by |x 1| yields

C at somemoment of timebetween6am andnoon?

= |x 2||x +2| |x 2| (|x| +2) < 5|x 2| < 0.05.

= |x 2||x +2| |x 2| (|x| +2) < 4.0002|x 2| < 0.00080004< 0.0009.

5|x 2|. Let

5|x 2| < 510

< if 0< |x 2| <

2x 13| < 0.1if 0< |x 5| < .

2x 13| < 0.1whenever |x 5| .

< . Repeat for = 0.2and0.1.

< if 0< |x 2| <

< 0.01if 0< |x 2| < .

< 0.08 if 0<

= |(x +c)(x c)| = |x +c| |x c| .

< 7|x 1| < 7

< 1andthat |cos| 1, andmakingthesubstitutionh = x a, weseethat thislast relation

(roomtemperature)? Estimatetherateof changeof averagevelocity with

. What aretheunitsof thisrate?

sec = and lim

tanx = and lim

51. Findtheleft- andright-handlimits of thefunctionf (x) inFigure2at x = 0, 2, 4. Statewhether f (x) is left- or

< 0.01for |x 2| < .

(0.9) representstheslopeof thetangent linetothegraphat x = 0.9.

(3) bytakingthelimit ash 0.

(a) in two ways, using Eq. (1) and Eq. (2).

(2) = 0. Ontheinterval 3 x 5, thegraphof

(5.5) islarger thanf

(3) doesnot exist.

(3), assumingthat thetangent linetoy = f (x) at a = 3hasequationy = 5x +2.

(a) and nd an equation of the tangent line.

(3)(x 3) +f (3) = 22(x 3) +48= 22x 18.

(3)(t 3) +f (3) = 11(t 3) 15= 11t +18.

(2)(x +2) +f (2) = 1(x +2) +1= x 1.

(0) obtainedinpart (a), theapproximateequationof thetangent lineis

(0)(x 0) +f (0) = 0.68x +2.

(a). Find f (x) and a.

(1) islarger or smaller thantheslopeof thesecant line

(1) islarger than

(1) issmaller than

(x) tofour decimal placesfor x = 1, 2, 3, 4.

(0) for x = 1, 2, 3, 4. Canyouguessanapproximateformulafor f

(4) tofour decimal places.

(4). Likewise, if h < 0, theslopeof thesecant

(4) obtainedinpart (b), theequationof thelinetangent tof (x) = x

(4)(x 4) +f (4) = 20(x 4) +32= 20x 48.

(c) = 0. Findc tosufcient accuracysothat