You are on page 1of 466

'

"75

gale bicentennial publications

VECTOR ANALYSIS

gale 'Bicentennial publications


With

the

approval of the

President

and Fellows

of Tale University, a series of volumes has been


prepared by a number of the Professors and Instructors,

to

tion

be

issued

in

connection

with

the

a partial indicathe
the
character
studies
in which the
of
of

Bicentennial

Anniversary, as

University teachers are engaged.

This

series

of volumes

is

of

respectfully

tijc

dedicated

to

VECTOR ANALYSIS
A TEXT-BOOK FOR THE USE OF STUDENTS
OF MATHEMATICS AND PHYSICS

FOUNDED UPON THE LECTURES OF

J.

WILLARD

GIBBS,

PH.D., LL.D.

Professor of Mathematical Physics in Yale University

BY

EDWIN BIDWELL WILSON,

PH.D.

Instructor in Mathematics in Yale University

NEW YORK CHARLES SCRIBNER'S SONS


LONDON: EDWARD ARNOLD
:

1901

Copyright, 1901,

BY YALE UNIVERSITY.
Published, December, 1901.

UNIVERSITY PRESS

AND SON

JOHN WILSON

CAMBRIDGE,

U.S.A.

PKEFACE BY PROFESSOR GIBBS


SINCE the printing of a short pamphlet on the Elements of
Vector Analysis in the years 188184,
never published, but
somewhat widely circulated among those who were known to
be interested in the subject,
the desire has been expressed
in more than one quarter, that the substance of that treatise,

perhaps in fuller form, should be made accessible to

the public.

As, however, the years passed without my finding the


meet this want, which seemed a real one, I was
very glad to have one of the hearers of my course on Vector
leisure to

Analysis in the year 1899-1900 undertake the preparation of


a text-book on the subject.
I

have not desired that Dr. Wilson should aim simply

at the reproduction of my lectures, but rather that he should


use his own judgment in all respects for the production of a

text-book in which the subject should be so illustrated by an


adequate number of examples as to meet the wants of students of geometry and physics.
J.

YALE UNIVERSITY,

WILLARD

September, 1901.

G47442
BHGINKEKING LIBRARY

GIBBS.

GENERAL PREFACE
WHEN I undertook to adapt the lectures of Professor Gibbs
on VECTOR ANALYSIS for publication in the Yale Bicentennial Series, Professor Gibbs himself was already so fully
engaged upon his work to appear in the same series, Elementary
Principles in Statistical Mechanics, that it was understood no
material assistance in the composition of this book could be
expected from him. For this reason he wished me to feel
entirely free to use my own discretion alike in the selection
of the topics to be treated and in the mode of treatment.
It has been

my

only in so far as

endeavor to use the freedom thus granted

was necessary

for presenting his

method

in

text-book form.
far the greater part of the material used in the following pages has been taken from the course of lectures on
Vector Analysis delivered annually at the University by

By

Professor Gibbs.

Some

use,

however, has been made of the

chapters on Vector Analysis in Mr. Oliver Heaviside's Elec~


tromagnetic Theory (Electrician Series, 1893) and in Professor

Foppl's lectures on Die MaxwelVsche Theorie der Electricitdt


(Teubner, 1894). My previous study of Quaternions has
also

been of great assistance.

The

material thus obtained has been arranged in the

way

which seems best suited to easy mastery of the subject.


Those Arts, which it seemed best to incorporate in the
text but which for various reasons may well be omitted at
the first reading have been marked with an asterisk (*). Numerous illustrative examples have been drawn from geometry,
mechanics, and physics. Indeed, a large part of the text has
to do with applications of the method.
These applications
have not been set apart in chapters by themselves, but have

GENERAL PREFACE

been distributed throughout the body of the book as fast as


the analysis has been developed sufficiently for their adequate
treatment. It is hoped that by this means the reader may be
Great care
better enabled to make practical use of the book.
has been taken in avoiding the introduction of unnecessary
ideas, and in so illustrating each idea that is introduced as

make its necessity evident and its meaning easy to grasp.


Thus the book is not intended as a complete exposition of

to

the theory of Vector Analysis, but as a text-book from which


so much of the subject as may be required for practical appli-

may be learned. Hence a summary, including a list


more important formulae, and a number of exercises,
have been placed at the end of each chapter, and many less
essential points in the text have been indicated rather than
fully worked out, in the hope that the reader will supply the
The summary may be found useful in reviews and
details.

cations
of the

for reference.

The subject of Vector Analysis naturally divides itself into


three distinct parts.
First, that which concerns addition and
the scalar and vector products of vectors.
Second, that which
concerns the differential and integral calculus in its relations

and vector functions. Third, that which contains


the theory of the linear vector function.
The first part is
a necessary introduction to both other parts. The second
to scalar

and third are mutually independent. Either may be taken


up first. For practical purposes in mathematical physics the
second must be regarded as more elementary than the third.
But a student not primarily interested in physics would naturally pass from the first part to the third, which he would
probably find more attractive and easy than the second.
Following this division of the subject, the main body of
the book is divided into six chapters of which two deal with
each of the three parts in the order named. Chapters I. and
II. treat of addition, subtraction, scalar
multiplication, and
the scalar and vector products of vectors. The exposition
has been

made

stood by and

knowledge

is

quite elementary. It can readily be underespecially suited for such readers as have a

of only the elements of

Trigonometry and Ana-

GENERAL PREFACE

xi

Those who are well versed in Quaternions


lytic Geometry.
or allied subjects may perhaps need to read only the summaries.

III.

Chapters

and IV. contain the treatment

of

those topics in Vector Analysis which, though of less value


to the students of pure mathematics, are of the utmost importance to students of physics. Chapters V. and VI. deal with
the linear vector function.

vector function

is

To

students of physics the linear

of particular importance in the

mathemati-

treatment of phenomena connected with non-isotropic


media and to the student of pure mathematics this part of
the book will probably be the most interesting of all, owing
cal

to the fact that it leads to Multiple

of Matrices.

Algebra or the Theory


which contains the

A concluding chapter, VII.,

development of certain higher parts of the theory, a number


of applications, and a short sketch of imaginary or complex
vectors, has been added.
In the treatment of the integral calculus, Chapter IV.,
questions of mathematical rigor arise. Although modern
theorists are devoting much time and thought to rigor, and
although they will doubtless criticise this portion of the book
it

adversely,

has been deemed best to give but

to the discussion of this subject.


And the
reason that whatever system of notation be

little

attention

more so

for the

employed ques-

tions of rigor are indissolubly associated with the calculus


and occasion no new difficulty to the student of Vector

Analysis,

who must

first

learn

what the

facts are

and may

postpone until later the detailed consideration of the restrictions that are put upon those facts.

Notwithstanding the efforts which have been made during


half a century to introduce Quaternions into
the
fact remains that they have not found wide favor.
physics

more than

On

hand there has been a growing tendency espedecade toward the adoption of some form of
Vector Analysis. The works of Heaviside and Foppl referred to before may be cited in evidence.
As yet however
no system of Vector Analysis which makes any claim to
the other

cially in the last

completeness has been published. In fact Heaviside says


"I am in hopes that the chapter which I now finish may
:

GENERAL PREFACE

x ii

serve as a stopgap till regular vectorial treatises come to be


written suitable for physicists, based upon the vectorial treat"
ment of vectors (Electromagnetic Theory, Vol. I., p. 305).
Elsewhere in the same chapter Heaviside has set forth the

claims of vector analysis as against Quaternions, and others


have expressed similar views.
then, to any system of vector analysis must
This, I feel confident, was Professor
practical utility.
He uses it
Gibbs's point of view in building up his system.

The keynote,

be

its

on Electricity and Magnetism and on


Electromagnetic Theory of Light. In writing this book I
have tried to present the subject from this practical standpoint, and keep clearly before the reader's mind the quesentirely in his courses

tions:

What

combinations or functions of vectors occur in


And how may these be represented

physics and geometry ?

symbolically in the way best suited to facile analytic manipulation ? The treatment of these questions in modern books
on physics has been too much confined to the addition and

This is scarcely enough. It has


to
aim
here
been the
give also an exposition of scalar and
f divergence and curl
vector products, of the operator
which have gained such universal recognition since the appearance of Maxwell's Treatise on Electricity and Magnetism,
of slope, potential, linear vector function, etc., such as shall

subtraction of vectors.

be adequate for the needs of students of physics at the


present day and adapted to them.
It has

been asserted by some that Quaternions, Vector

Analysis, and all such algebras are of little value for investigating questions in mathematical physics. Whether this
assertion shall prove true or not, one
vectors are to mathematical physics

may

still

maintain that

what invariants are to

geometry. As every geometer must be thoroughly conversant with the ideas of invariants, so every student of physics
should be able to think in terms of vectors. And there is

no way

in

which

he, especially at the beginning of his scican come to so true an appreciation of the
importance of vectors and of the ideas connected with them
as by working in Vector
Analysis and dealing directly with

entific studies,

GENERAL PREFACE
the vectors themselves.

success

of

Professor

To

xiii

those that hold these views the

Vorlesungen uber

Fb'ppl's

Technische

Mechanik (four volumes, Teubner, 1897-1900, already in a


second edition), in which the theory of mechanics is developed by means of a vector analysis, can be but an encouraging sign.
I take pleasure in

thanking

script.

The good

colleagues, Dr. M. B. Porter


assisting me with the manu-

my

and Prof. H. A. Bumstead, for

services of the latter have been particularly

valuable in arranging Chapters III. and IV. in their present


form and in suggesting many of the illustrations used in the

am

also under obligations to my father, Mr. Edwin


for help in connection both with the proofs and
the manuscript. Finally, I wish to express my deep indebt-

work.

H. Wilson,

For although he has been so


preoccupied as to be unable to read either manuscript or
proof, he has always been ready to talk matters over with
me, and it is he who has furnished me with inspiration sufedness to Professor Gibbs.

ficient to

cany through the work.

EDWIN BIDWELL WILSON.


YALE UNIVERSITY,

October, 1901.

TABLE OF CONTENTS
PAGE

PREFACE BY PROFESSOR GIBBS

vii

GENERAL PREFACE

ix

CHAPTER

ADDITION AND SCALAR MULTIPLICATION


ARTS.

1-3

4
5

6-7

8-10

SCALARS AND VECTORS


EQUAL AND NULL VECTORS
THE POINT OF VIEW OF THIS CHAPTER
SCALAR MULTIPLICATION. THE NEGATIVE SIGN
ADDITION. THE PARALLELOGRAM LAW

11

SUBTRACTION

12

LAWS GOVERNING THE FOREGOING OPERATIONS


COMPONENTS OF VECTORS. VECTOR EQUATIONS
THE THREE UNIT VECTORS 1, j, k

13-16
17

18-19

20-22
23-24
25

4
6

....

8
11

....
....

14
18

APPLICATIONS TO SUNDRY PROBLEMS IN GEOMETRY.


VECTOR RELATIONS INDEPENDENT OF THE ORIGIN
CENTERS OF GRAVITY. BARYCENTRIC COORDINATES
THE USE OF VECTORS TO DENOTE AREAS
SUMMARY OF CHAPTER i
EXERCISES ON CHAPTER i

CHAPTER

12

21

...

27

39

46
51

52

II

DIRECT AND SKEW PRODUCTS OF VECTORS


27-28

29-30

31-33
34-35
36

THE DIRECT, SCALAR, OR DOT PRODUCT OF TWO VECTORS


THE DISTRIBUTIVE LAW AND APPLICATIONS
THE SKEW, VECTOR, OR CROSS PRODUCT OF TWO VECTORS
THE DISTRIBUTIVE LAW AND APPLICATIONS
THE TRIPLE PRODUCT A* B C

55
58

60
63

67

CONTENTS

XVI
ARTS.

37-38
39-40
41-42

43-45
46-47

48-50
51

52
53

54

THE SCALAR
THE VECTOR

TRIPLE PRODUCT A* B X COR [ABC]


TRIPLE PRODUCT A X (B X C)

68

71

PRODUCTS OF MORE THAN THREE VECTORS WITH APPLICATIONS TO TRIGONOMETRY


RECIPROCAL SYSTEMS OF THREE VECTORS
SOLUTION OF SCALAR AND VECTOR EQUATIONS LINEAR IN
AN UNKNOWN VECTOR
SYSTEMS OF FORCES ACTING ON A RIGID BODY
KINEMATICS OF A RIGID BODY
CONDITIONS FOR EQUILIBRIUM OF A RIGID BODY
RELATIONS_JBETWEEN TWO RIGHT-HANDED SYSTEMS OF
TIIKKK I'KKI'I.M>I<TLAU T N IT VKCTOKS
PROBLEMS IN GEOMETRY. PLANAR COORDINATES
SUMMARY OF CHAPTER n
EXERCISES ON CHAPTER n

....
.

75

81

87
92
97
101
104

106
109
113

CHAPTER HI
THE DIFFERENTIAL CALCULUS OF VECTORS
DERIVATIVES AND DIFFERENTIALS OF VECTOR FUNCTIONS
WITH RESPECT TO A SCALAR VARIABLE
115
57 CURVATURE AND TORSION OF GAUCHE CURVES
120
58-59 KINEMATICS OF A PARTICLE. THE HODOGRAPH
125
60 THE INSTANTANEOUS AXIS OF ROTATION
131
61 INTEGRATION WITH APPLICATIONS TO KINEMATICS
133
62 SCALAR FUNCTIONS OF POSITION IN SPACE
136
63-67 THE YECTORJDIFFEUENTIATING OPERATOR
138
68 THE SCALAR OPERATOR A * 'v7
147
69 VECTOR FUNCTIONS OF POSITION IN SPACE
149
70 THE DIVERGENCE
AND THE CURL
150
X
71
INTERPRETATION OF THE DIVERGENCE S7
152
72 INTERPRETATION OF THE CURL
155
X
*
73
LAWS OF OPERATION OF V>
157'
X
74-76 THE PARTIAL APPLICATION OF V- EXPANSION OF A VECTOR FUNCTION ANALOGOUS TO TAYLOR'S THEOREM.
159
APPLICATION TO HYDROMECHANICS
77
THE DIFFERENTIATING OPERATORS OF THE SECOND ORDER 166
78 GEOMETRIC INTERPRETATION OF LAPLACE'S OPERATOR
As THE DISPERSION
170
V*
SUMMARY OF CHAPTER in
172
EXERCISES ON CHAPTER in
177
55-56

....

V
>

CONTENTS

xvii

CHAPTER IV
THE INTEGRAL CALCULUS OF VECTORS
PAGE

ARTS.

79-80
81

82
83
84

LINE INTEGRALS OF VECTOR FUNCTIONS WITH APPLICATIONS


GAUSS'S THEOREM
STOKKS'S THEOREM
CONVERSE OF STOKES'S THEOREM WITH APPLICATIONS
TRANSFORMATIONS OF LINE, SURFACE, AND VOLUME INTEGRALS. GREEN'S THEOREM
.

91

REMARKS ON MULTIPLE-VALUED FUNCTIONS


POTENTIAL. THE INTEGRATING OPERATOR " POT "
COMMUTATIVE PROPERTY OF POT AND ^7
REMARKS UPON THE FOREGOING
THE INTEGRATING OPERATORS "NEW," "LAP," " MAX "
RELATIONS BETWEEN THE INTEGRATING AND DIFFER-

92

THE POTENTIAL

85

86-87
88
89
90

ENTIATING OPERATORS
"

POT "

MAXWELLIANS

193

197
200
205
211

215
222

230
234
240

CERTAIN BOUNDARY VALUE THEOREMS


SUMMARY OF CHAPTER iv
EXERCISES ON CHAPTER iv

96

187

A SOLUTION OF POISSON'S

95

184

228
is

EQUATION
SOLENOIDAL AND IRROTATIONAL PARTS OF A VECTOR
FUNCTION. CERTAIN OPERATORS AND THEIR INVERSE
MUTUAL POTENTIALS, NEWTONIANS, LAPLACIANS, AND

93-94

179

243
.

249
255

CHAPTER V
LINEAR VECTOR FUNCTIONS
97-98
99

100

LINEAR VECTOR FUNCTIONS DEFINED


DYADICS DEFINED
ANY LINEAR VECTOR FUNCTION MAY BE REPRESENTED
BY A DYADIC. PROPERTIES OF DYADICS
THE NONION FORM OF A DYADIC
THE DYAD OR INDETERMINATE PRODUCT OF TWO VECTORS IS THE MOST GENERAL. FUNCTIONAL PROPERTY
OF THE SCALAR AND VECTOR PRODUCTS
PRODUCTS OF DYADICS
DEGREES OF NULLITY OF DYADICS
THE IDEMF ACTOR

....

101

102

103-104

105-107
108

...

...

260
264
266

269

271

276

282

288

XV111

CONTENTS
PAGE

ARTS.

109-110
111

112-114
115-116
117

118-119
120
121

122

POWERS AND ROOTS OF DYADICS


SELF-CONJUGATE AND ANTICONJUGATE DYADICS.
SELF-CONJUGATE PARTS OF A DYADIC

RECIPROCAL DYADICS.

ANTI-SELF-CONJUGATE

DYADICS.

THE VECTOR

290
294

PROD-

UCT. QUADRANTAL VERSORS


REDUCTION OF DYADICS TO NORMAL FORM
DOUBLE MULTIPLICATION OF DYADICS
THE SECOND AND THIRD OF A DYADIC
CONDITIONS FOR DIFFERENT DEGREES OF NULLITY
NONION FORM. DETERMINANTS
INVARIANTS OF A DYADIC. THE HAMILTON-CAYLEY
EQUATION
SUMMARY OF CHAPTER v
EXERCISES ON CHAPTER v

....
.

297

302
306

310
313
315

319
321

329

CHAPTER VI
ROTATIONS AND STRAINS
123-124
125-126
127
128

129

130
131

132

HOMOGENEOUS STRAIN REPRESENTED BY A DYADIC


ROTATIONS ABOUT A FIXED POINT. VERSORS
THE VECTOR SEMI-TANGENT OF VERSION
BlQUADRANTAL VERSORS AND THEIR PRODUCTS
CYCLIC DYADICS
RIGHT TENSORS
TONICS AND CYCLOTONICS
REDUCTION OF DYADICS TO CANONICAL FORMS, TONICS,
CYCLOTONICS, SIMPLE AND COMPLEX SHEARERS
SUMMARY OF CHAPTER vi

332

334

339

CHAPTER

343

347
351
353
356

368

VII

MISCELLANEOUS APPLICATIONS
136-142
143-146

147-148
149-157
158-162

QUADRIC SURFACES
THE PROPAGATION OF LIGHT IN CRYSTALS
VARIABLE DYADICS
CURVATURE OF SURFACES
HARMONIC VIBRATIONS AND BIVECTORS

....

372
392

403
411

426

YECTOft ANALYSIS

VECTOR ANALYSIS
CHAPTER

ADDITION AND SCALAR MULTIPLICATION


1.]

IN mathematics and especially

two

in physics

different kinds of quantity present themselves.

very-

Consider, for

example, mass, time, density, temperature, force, displacement


of a

point, velocity,

and

Of

acceleration.

these quantities

some can be represented adequately by a single number


temperature, by degrees on a thermometric scale time, by
mass and density, by numerical val3^ears, days, or seconds
;

ues which are wholly determined when the unit of the scale
On the other hand the remaining quantities are not
is fixed.
capable of such representation. Force to be sure is said to be
of so many pounds or grams weight; velocity, of so many
But in addition to this each
feet or centimeters per second.
-of

them must be considered

as

having direction as well as

A force points North, South, East, West, up,


magnitude.
down, or in some intermediate direction. The same is true
of displacement, velocity, and acceleration.
No scale of numbers can represent

them adequately.

It

can represent only

their magnitude, not their direction.


2.]

Definition

vector is a quantity

which

is

considered

as possessing direction as well as magnitude.

Definition

A scalar is a quantity which is considered as pos-

sessing magnitude but no direction.

VECTOR ANALYSIS
The positive and negative numbers of ordinary algebra are the
For this reason the ordinary algebra is called
typical scalars.

when

necessary to distinguish it from the vector


algebra or analysis which is the subject of this book.
The typical vector is the displacement of translation in space.
scalar algebra

Consider

first

a point

(Fig. 1).

straight line

Let

and take a

This change of position


line

PP. The

ment
it is

P'.

be displaced in a

new

position

P'.

represented by the
magnitude of the displaceis

is the length of PP' ; the direction of


the direction of the line PP' from P to

Next consider a displacement not

of one,

the points in space. Let all the


but
points move in straight lines in the same direction and for the
same distance D. This is equivalent to shifting space as a
of all

body

rigid

in that direction through the distance

without

Such a displacement is called a translation.


It
and
direction
When
possesses
magnitude.
space undergoes
rotation.

a translation T, each point of space undergoes a displacement

equal to

in

magnitude and direction; and conversely if


PP' which any one particular point P suf-

the displacement

T is known, then that of any other


Q is also known for Q Q' must be equal and parallel
toPP'.
The translation T is represented geometrically or graphically
by an arrow T (Fig. 1) of which the magnitude and direction
fers in the translation

point

are equal to those of the translation.


The absolute position
of this arrow in space is
immaterial.
entirely
Technically the
arrow is called a stroke. Its tail or initial point is its origin ;

and

its

head or

is
designated by
metric quantity, a stroke,

origin

In the figure the


and the terminus by T. This geo-

final point, its terminus.

is

used as the mathematical symbol

for all vectors, just as the


ordinary positive and negative
bers are used as the
symbols for all scalars.

num-

ADDITION AND SCALAR MULTIPLICATION


*

As examples

3.]
sity,

of scalar quantities

and temperature have been mentioned.

mass, time, denOthers are dis-

moment of inertia, work, etc. Magnitude,


means the sole property of these quantities.
no
by

tance, volume,

however,

is

Each implies something besides magnitude.

Each has its


an example of which

own

distinguishing characteristics, as
dimensions in the sense well known to physicists may
distance 3, a time 3, a work 3, etc., are verybe cited.
The magnitude 3 is, however, a property common
different.
its

to

them

tities

perhaps the only one.

all

pure number

the simplest.

is

Of

all scalar

It implies

quanti-

nothing but

magnitude. It is the scalar par excellence and consequently


used as the mathematical symbol for all scalars.

it is

As examples
city,

of vector quantities force, displacement, veloacceleration have been given.


Each of these has

and

other characteristics than those which belong to a vector pure


and simple. The concept of vector involves two ideas and

two alone
vector.
,to

But

a rigid

body the

consideration
fice.

magnitude of the vector and direction of the


force is more complicated.
When it is applied

And

line in

which

it

acts

must be taken

into

magnitude and direction alone do not suf-

in case

it is

applied to a non-rigid body the point


is as important as the magnitude or

of application of the force


direction.

than force.

Such is frequently true for vector quantities other


Moreover the question of dimensions is present

as in the case of scalar quantities.

the stroke, which

The mathematical

vector,

the primary object of consideration in


this book, abstracts from all directed quantities their magniis

tude and direction and nothing but these

just as the mathe-

the magnitude and


matical scalar, pure number,
Hence one must be on his guard lest from
that alone.
abstracts

analogy he attribute some properties to the mathematical


vector which do not belong to it and he must be even more
;

careful lest he obtain erroneous results

by considering the

VECTOR ANALYSIS

vector quantities of physics as possessing no properties other

For example

than those of the mathematical vector.

never do to consider force

and

would

it

effects as unaltered

its

by

This warning may not be


parallel
shifting
it may possibly save some confusion.
necessary, yet
Inasmuch as, taken in its entirety, a vector or stroke
4.]
to

it

itself.

it may appropriately be designated


by
to the fundamental difference
however
Owing
between scalars and vectors, it is necessary to distinguish
Sometimes, as in mathecarefully the one from the other.
is

but a single concept,

one

letter.

matical physics, the distinction is furnished by the physical


Thus if n be the index of refraction it
interpretation.

must be
scalars

scalar;

but

/,

the time, are also


m, the mass, and
are
force, and
a, the acceleration,
,

the

When, however,

vectors.

the letters are regarded merely

with no particular physical significance some


typographical difference must be relied upon to distinguish

as symbols

vectors from scalars.

Hence

book Clarendon type

in this

is

used for setting up vectors and ordinary type for scalars.


This permits the use of the same letter differently printed
to represent the vector

and

C be the electric current in

its

Thus

scalar magnitude. 1

magnitude and

direction,

if

C may

be used to represent the magnitude of that current if g be


the vector acceleration due to gravity, g may be the scalar
;

value of that acceleration


mass, v

may be

if

v be the velocity of a moving

the magnitude of that velocity.

Clarendons to denote vectors makes

it

The use

possible to pass

of

from

quantities to their scalar magnitudes by a mere


change in the appearance of a letter without any confusing
change in the letter itself.

directed

Definition

Two vectors

are said to be equal

when they have

the same magnitude and the same direction.


1

This convention, however,

instances

it

would prove

is

by no means invariably followed.

just as undesirable as

it

is

In some

convenient in others.

chiefly valuable in the application of vectors to physics.

It is

ADDITION AND SCALAR MULTIPLICATION


The equality of two vectors A and
usual sign =. Thus
A B

is

denoted by the

Evidently a vector or stroke is not altered by shifting it


about parallel to itself in space. Hence any vector A = PP'
(Fig. 1)
for the

as origin
be drawn from any assigned point
until
PP'
to
itself
be
moved
parallel
segment
may

may

the point

P falls

A=
In this way

all

and P' upon some point

upon the point

PP = OT= T.

vectors in space

be replaced by directed

may

segments radiating from one fixed point


in space will of course coincide,

Equal vectors

0.

when placed with

Thus (Fig. 1) A = PP'

mini at the same point 0.


both fall upon T = ~OT.

are necessary.

with

its

their ter-

and B =

For the numerical determination of a vector


If r, $,

T.

Q~Q',

three scalars

These may be chosen in a variety of ways.

be polar coordinates in space any vector r drawn


origin at the origin of coordinates may be represented

by the three

scalars

r,

<, 9

the vector.

which determine the terminus of

(r,

/,,

$, V)

Or if z, y, z be Cartesian coordinates in space a vector r may


be considered as given by the differences of the coordinates x\
y , z' of its terminus and those #, y, z of its origin.
f

(x

- x, y'

y,z'

z).

If in particular the origin of the vector coincide

with the

origin of coordinates, the vector will be represented

three coordinates of

its

r~(x',

When two

by the

terminus
y', z').

vectors are equal the three scalars which repre-

sent them must be equal respectively each to each.


one vector equality implies three scalar equalities.

Hence

VECTOR ANALYSIS

A vector A

Definition

magnitude A is
Such a vector
equal to

is

said to be equal to zero

when

its

zero.

is

called a null or zero vector

in the usual manner.

naught

A=

if

and

is

written

Thus

A = 0.

All null vectors are regarded as equal to each other without


of direction.

any considerations

In fact a null vector from a geometrical standpoint would


that is to
be represented by a linear segment of length zero
say,

by a

It consequently

point.

terminate direction
all.

If,

direction

which

finite vector,

would have a wholly indeto the same thing, none at

what amounts

be regarded as the limit approached by a


length, it might be considered to have that

however,

vector of finite

or,

is

it

the limit approached by the direction of the


the length decreases indefinitely and ap-

when

proaches zero as a limit.

The

justification for disregarding

and looking upon all null vectors as equal is


that when they are added (Art. 8) to other vectors no change
occurs and when multiplied (Arts. 27, 31) by other vectors

this direction

the product

is

zero.

In extending to vectors the fundamental operations

5.]

and arithmetic, namely, addition, subtraction, and


multiplication, care must be exercised not only to avoid selfof algebra

contradictory definitions but also to lay down useful ones.


Both these ends may be accomplished most naturally and
easily

by looking to physics (for in that science vectors conand by observing how such

tinually present themselves)

If then A be a given displaceor


what
is two, three, or in general x
ment, force,
velocity,
times A? What, the negative of A? And if B be another,

quantities are treated there.

what

is

the

sum

equivalent of

of A and B ?
That
and B taken together ?

is

to say,

what

is

the

The obvious answers

to these questions suggest


immediately the desired definitions.

ADDITION AND SCALAR MULTIPLICATION

Scalar Multiplication
Definition:

6.]

positive scalar

and

Thus
v

when

direction

its

vector

its

is left

said to be multiplied

is

magnitude

is

by a

multiplied by that scalar

unaltered.

v be a velocity of nine knots East by North, 2 J times


a velocity of twenty-one knots with the direction still

is

if

East by North.

Or

if f

be the force exerted upon the scale-

gram weight, 1000 times f is the force exerted by a


kilogram. The direction in both cases is vertically down-

pan by a
ward.

A be

the vector and x the scalar the product of x and


denoted as usual by
x A or A x.
If

A is

however, more customary to place the scalar multiplier


before the multiplicand A. This multiplication by a scalar
It

is

is,

and

called scalar multiplication,

x (y A)
as in ordinary algebra

mediately obvious

(x y)

it

follows the associative law

A=

y (x A)

and arithmetic.

when the

fact is

This statement

is

im-

taken into consideration

that scalar multiplication does not alter direction but merely


multiplies the length.
Definition

Any

vector

vector a in

its

A unit vector is
A may

one whose magnitude

is

unity.

be looked upon as the product of a unit

direction

by the

positive scalar

A,

its

magni-

tude.

A=A

= a A.

The unit vector

A by \/A

a may similarly be written as the product of


or as the quotient of A and A.
1

VECTOR ANALYSIS

The negative

Definition

7.]

reverses its direction

For example

- A

if

but leaves

sign,

prefixed to a vector

its

magnitude unchanged.
be a displacement for two feet to the right,

a displacement for two feet to the left. Again if the


B be A, the stroke B A, which is of the same length
stroke
is

A
A B but which is in
from A to B, will be
as

of the negative sign


of motion.
" reaction."

If

The

B to A

the direction from

A.

may

Another

instead of

illustration of the use

be taken from Newton's third law

denote an "action,"
positive sign,

+ may
,

will denote the

be prefixed to a vec-

tor to call particular attention to the fact that the direction

The two

has not been reversed.

signs

and

when used

in connection with scalar multiplication of vectors follow the

same laws

of operation as in ordinary algebra.

These are

symbolically

++=+

+ - = -; - + = -;

(ra

The

interpretation

is

= ra

A)

-- = +

A).

obvious.

Addition and Subtraction

The

8.]

addition of two vectors or strokes

may

be treated

most simply by regarding them as defining translations in


Let S be one vector and T the other. Let P

space (Art. 2).

be a point of space (Fig. 2). The translation S carries P into P' such that the
line

PP'

is

equal to S in magnitude and

The transformation T will then


the line P'P" being
carry P' into P"
parallel to T and equal to it in magnitude.
direction.

FIG.

2.

Consequently the result of S followed by

T
P".

If

is

now Q be any

into Q' such that

to carry the point

QQ =
1

into the point


other point in space, S will carry Q
S and T will then carry Q' into Q"

ADDITION AND SCALAR MULTIPLICATION

= T.

such that Q' Q"

Thus S followed by T carries Q into Q".


For
Q Q' Q" is equal to PP'P".

Moreover, the triangle

two

the

sides

Q'

and

Q' Q", being equal

and T respectively, must be likewise

and

parallel to S

P P'

parallel to

and

P' P" respectively which are also parallel to S and T. Hence


the third sides of the triangles must be equal and parallel.

That

is

Q Q"

As Q

is

equal and parallel to

PP".

any point in space this is equivalent to saying that


by means of S followed by T all points of space are displaced
the same amount and in the same direction. This displacement is therefore a translation.
Consequently the two
is

translations S

and T are equivalent

to a single translation

B,.

Moreover
S

if

The

= PP' and

stroke

strokes S

B,

is

= P' P",

called the

and T to which

it

then

resultant

is

E = PP".
or

sum

equivalent.

of the

two

sum

de-

This

is

noted in the usual manner by

From analogy with

the

R=

sum

or resultant of

T.

two translations

the following definition for the addition of any


laid

two vectors is

down.

Definition

The sum

or resultant of

two vectors

is

found

by placing the origin of the second upon the terminus of the


first and drawing the vector from the origin of the first to the
terminus of the second.
9.]

Theorem.

added does not

The order
affect the

in

which two vectors S and T are

sum.

S followed by T gives precisely the same result as


by S. For let S carry P into P' (Fig. 3) ; and T,

T followed

into P".

+ T then carries P into P". Suppose now that T carries P


into P'". The line PP'" is equal and parallel to P'P".
Con-

VECTOR ANALYSIS

10

and P'" lie at the vertices of


sequently the points P, P', P",
Hence
a parallelogram.
i
j
ua
an(
pin pn j s e q
p ar_
allel

to

PP'.

Hence S
T fol-

carries P'" into P".

lowed by S therefore

car-

P into P" through P',


whereas S followed by T

ries

carries
P'".

This

either case the same.

may

into

The

P" through

final result is in

be designated symbolically

by writing
It is to

be noticed that S = PP' and T

of the parallelogram

= P P'" are the two sides


the point P as

PP' P" P'" which have

R=

PP" is the diagonal drawn


origin ; and that
through P. This leads to another very common way of
stating the definition of the sum of two vectors.
common

two vectors be drawn from the same origin and a parallelogram be constructed upon them as sides, their sum will be that
If

diagonal which passes through their common origin.


"
This is the well-known " parallelogram law according to

which the physical vector quantities force, acceleration, velocIt is important to


ity, and angular velocity are compounded.
note that in case the vectors lie along the same line vector
addition becomes equivalent to algebraic scalar addition.
The
if
the
of
to
be
added
are
added
vectors
the
two
vectors
lengths

have the same direction


site directions.

but subtracted

if

they have oppo-

In either case the sum has the same direction

as that of the greater vector.


10.]

After the definition of the

been laid down, the

sum

of several

sum

of

may

be found by adding

two vectors has

together the first two, to this sum the third, to this the fourth,
and so on until all the vectors have been combined into a sin-

ADDITION AND SCALAR MULTIPLICATION

11

gle one. The final result is the same as that obtained by placing
the origin of each succeeding vector upon the terminus of the
preceding one and then drawing at once the vector from

In case
the origin of the first to the terminus of the last.
these two points coincide the vectors form a closed polygon

and
that

their
if

strokes

sum

is

zero.

Interpreted geometrically this states

are such that the


number of displacements R, S, T
T
the
sides
a
closed
form
of
R, S,
polygon taken in

order, then the effect of carrying out the displacements

Each point

of space

is

brought back to

terpreted in mechanics it states that

if

its

is nil.

starting point.

any number

In-

of forces

act at a point and if they form the sides of a closed polygon


taken in order, then the resultant force is zero and the point
is

in equilibrium under the action of the forces.


The order of sequence of the vectors in a sum

sequence.
cent vectors

This

may

is

of

no con-

may be shown by proving that any two adjabe interchanged without affecting the result.

To show

A+B+C+D+E=A+B+D+C+R
A = ~0~A, B = Zi?, C = ~BC, D = (TlT, E = WE.
Then
Ofi = A + B + C + D + E.
Let

B C' = D. Then C' B C D


consequently C' D = C. Hence
0~E = A + B + D +
Let now

is

a parallelogram and

E,

which proves the statement. Since any two adjacent vectors


may be interchanged, and since the sum may be arranged in
any order by successive interchanges of adjacent vectors, the
order in which the vectors occur in the sum is immaterial.
11.]
is

Definition:

added

vector

is

said to be subtracted

after reversal of direction.

when

it

Symbolically,

A - B = A + (- B).
By

this

means subtraction

is

reduced to addition and needs

VECTOR ANALYSIS

12

is however an interesting and


the
difference of two vectors
of representing

no special consideration.
important

way

geometrically.

Let

There

A = OA, B = OB

Complete
(Fig. 4).
the parallelogram of which A and B
are the sides.
Then the diagonal

=C

is

the

sum A + B

of the

Next complete the


of
which A and
B
parallelogram
OB' are the sides. Then the ditwo

vectors.

agonal

OD = D will

be the

sum

of

But the
and
OD
is
parallel
equal
segment
Hence BA may be taken as the difference to the two
to EA.
This leads to the following rule The differvectors A and B.
ence of two vectors which are drawn from the same origin is
the vector drawn from the terminus of the vector to be subtracted to the terminus of the vector from which it is subThus the two diagonals of the parallelogram, which
tracted.
is constructed upon A and B as sides, give the sum and difference of A and B.
In the foregoing paragraphs addition, subtraction, and
12.]
scalar multiplication of vectors have been defined and interand the negative of

B.

To make

the development of vector algebra matheexact


and
matically
systematic it would now become necessary
to demonstrate that these three fundamental operations follow
the same formal laws as in the ordinary scalar algebra, alpreted.

though from the standpoint of the physical and geometrical


These
interpretation of vectors this may seem superfluous.
laws are
Ia
I

II

III a

III 6
III,

= (m ri) A,
=
A+ (B-f C),
(A + B) + C
A B = B + A,
(m, + n) A = m A + n A,
m (A + B) = m A + m B,
- (A + B) = - A - B.
(n A)

-I-

n (m A)

ADDITION AND SCALAR MULTIPLICATION

13

and commutation

of

the scalar factors in scalar multiplication.


I 6 is the law of association for vectors in vector addition.

It

I a is the so-called

law

of association

states that in adding vectors parentheses


any points without altering the result.

II

is

III a

be inserted at

may

the commutative law of vector addition.


is

the distributive law for scalars in scalar multipli-

is

the distributive law for vectors in scalar multipli-

cation.

IIIj
cation.

III C

The

the distributive law for the negative sign.


proofs of these laws of operation depend upon those

is

propositions in elementary geometry which have to deal with


the first properties of the parallelogram and similar triangles.

They will not be given here; but it is suggested that the


reader work them out for the sake of fixing the fundamental
ideas of addition, subtraction,

The

clearly in mind.
in the statement

and

scalar multiplication

result of the laws

may

be

more

summed up

The laws which govern addition, subtraction, and scalar


multiplication of vectors are identical with those governing these
operations in ordinary scalar algebra.
It is precisely this identity of formal laws

the extension of the use of the familiar signs


of arithmetic to the algebra of vectors and

which ensures the correctness of

which

justifies

=, +, and
it is

results obtained

also this

by operat-

ing with those signs in the usual manner. One caution only
need be mentioned. Scalars and vectors are entirely different

For this reason they can never be equated


except perhaps in the trivial case where each is
For the same reason they are not to be added together.

sorts of quantity.

to each other
zero.

So long as this is borne in mind no difficulty need be anticipated from dealing with vectors much as if they were scalars.
Thus from equations in which the vectors enter linearly with

VECTOR ANALYSIS

14

unknown

coefficients

scalar

vectors

may

be eliminated or

same way and with the same limitafound by


tions as in ordinary algebra; for the eliminations and solutions depend solely on the scalar coefficients of the equations
If for
and not at all on what the variables represent.
solution in the

instance

aA + &B +
then A, B,

C,

or

D may

cC

+ dD =

0,

be expressed in terms of the other

three

=-

as

(a

A + bB+

C).

(M

And two

vector equations such as

and
yield

A+4B=E

A+

=F

by the usual processes the solutions

A=3E-4F
B

and

=3F-

2 E.

Components of Vectors
13.]

Definition

Vectors are said to be collinear when

they are parallel to the same line; coplanar, when parallel


to the

same plane.

Two

or

more vectors

to

which no

line

can be drawn parallel are said to be non-collinear. Three or


more vectors to which no plane can be drawn parallel are
said to be non-coplanar.

Obviously any two vectors are

coplanar.

vector b collinear with a

may be expressed as the


product of a and a positive or negative scalar which is the
ratio of the magnitude of b to that of a.
The sign is positive

Any

when b and

a have the same direction

have opposite directions.

If

then

OA =

negative, when they


a, the vector r drawn

ADDITION AND SCALAR MULTIPLICATION


from the origin
either direction

any point of the

to

line

OA

15

produced in

is

= x a.

(1)

x be a variable scalar parameter this equation may therefore be regarded as the (vector) equation of all points in the
If

OA.

line

upon

now B be
OAor that

Let

the line

any point not


line produced

in either direction (Fig. 5).

Let

B = b. The

not of the form x


a line parallel to

point upon

it.

vector b

OA
The

and

let

vector

to

0~R=
r

This equation
all

FlG 5

be any

a.

collinear with a and is


Hence the vector drawn

is

E is

or

may

CTB

+ B~R

=b+

a.

(2)

be regarded as the (vector) equation of

the points in the line which

B is

surely

BE

consequently expressible as x

from

is

Draw through B

a.

is

parallel to a

and of which

one point.

vector r coplanar with two non-collinear vectors


a and b may be resolved into two components parallel to a
and b respectively. This resolution may
14.]

Any

be accomplished by constructing the parallelogram (Fig. 6) of which the sides are


parallel to a and b and of which the di-

agonal is r. Of these components one is


x a the other, y b.
x and y are respec;

tively the

scalar ratios (taken with the

proper sign) of the lengths of these components to the lengths


of a and b.
Hence
r
x a + yb
(2)'

If
a typical form for any vector coplanar with a and b.
as
in
form
be
this
several vectors r v r 2 , r 3
expressed
may
is

VECTOR ANALYSIS

16

=x a+
r2 = z 2 a +
r = ^3 a +
rx

their

sum

r is

sum

b,
b,

2/3

then

+
This

yl
2/2

(2/i

y2

2/3

---) b

the well-known theorem that the components of a


of vectors are the sums of the components of those
is

vectors.

each of

If the vector r is zero

its

components must

Consequently the one vector equation r


equivalent to the two scalar equations

be zero.

2/i

2/2

2/3

is

(3)

vector r in space may be resolved into three


components parallel to any three given non-coplanar vectors.
Let the vectors be a, b,
15.]

Any

and

The

c.

may

resolution

then be

accom-

plished by constructing
the parallelepiped (Fig.

which the edges


are parallel to a, b, and
7) of

and of which the

agonal

p IG

is r.

di-

This par-

allelopiped may be
drawn easily by passing

three planes parallel respectively to a and b, b and c,


of the vector r ; and a similar set of three planes through its
terminus R. These six planes will then be parallel in pairs
c

and a through the origin

ADDITION AND SCALAR MULTIPLICATION


and hence form a

parallelepiped.

That the

17

intersections of

the planes are lines which are parallel to a, or b, or c is


The three components of r are x a, y b, and z c ;
obvious.
where #, y, and z are respectively the scalar ratios (taken with
the proper sign) of the lengths of these components to the

length of

a, b,

and

c.

Hence

= #a +

2/b

zc

(4)

a typical form for any vector whatsoever in space. Several


vectors i v r 2 , r 3 . may be expressed in this form as

is

^ = x a + y b + z c,
r 2 = a a + y^ b + z 2
r 3 = # 3 a + yB b + * 3 c,
l

aj

Their

sum

c,

r is then

+
+ 2/2 + 2/3 +
+ (2J+Z2 + 23+

'

(.l/i

If the vector r is zero

each of

its

*1

+ 2 + 2/3 +
+ *2 + 2 3 +

'

'

'

'

'

'

2/

)<>

three components

Consequently the one vector equation r


the three scalar equations

yi

=
= '

is

- 0.

is zero.

equivalent to

(5)

Should the vectors all be coplanar with a and b, all the components parallel to c vanish. In this case therefore the above
equations reduce to those given before.
If two equal vectors are expressed in terms of the
16.]

same three non-coplanar

vectors, the corresponding scalar co-

efficients are equal.


2

VECTOR ANALYSIS

18

Let

r',

= x' a + y' b + z' c,


= x', y = y', z = z
- x') a + (y - y'} b + (z - z')
=
=
r
r'
For
(x
= 0, z z' = 0.
x' = 0,
x
Hence
y
y'
r'

Then

c.

But this would not be true if a, b, and c were coplanar. In


that case one of the three vectors could be expressed in terms
of the other

two

as

= m a + n b.
r =rca + 2/b + 2c = (ic + m2)a+(2/ + 7i2)b,
c = (x' + m z') a + (y' + n z') b,
r' = x' a + y b +
=
m
r
r'
z)
[(x +
(#' + m 2')] a,
+ [(y + nz)-(y' + nz')'] b = 0.
c

Then

Hence

z'

the individual components of r

in the directions

r'

a and b (supposed different) are zero.


r-i
I

o H
i"i

O"
*/

/*

L\ _.

T^

vfl
//t

v
*

/
y*
tX/

J^

^^

^ + n z = y' +

But

this

tively to

by no means
#', y', z'.

necessitates

/>ri

//(>

a?,

*y'

z'.

be equal respeca and b were col-

y, z to

In a similar manner

if

impossible to infer that their coefficients vanish


The theorem may perhaps be stated as follows
individually.
linear it

is

In

case two equal vectors are expressed in terms of one vector,

or two non-collinear vectors, or three non-coplanar vectors, the

corresponding scalar coefficients are equal. But this is not necessarily true if the two vectors be collinear ; or the three vectors,
coplanar.
(Arts. 18

This principle will be used in the

applications

et seq.).

The Three Unit Vectors

i, j,

k.

In the foregoing paragraphs the method of expressin terms of three given non-coplanar ones has been
vectors
ing
The simplest set of three such vectors is the rectexplained.
17.]

ADDITION AND SCALAR MULTIPLICATION

19

Cartesian Geometry. This


angular system familiar in Solid
be either of two very distinct
rectangular system may however
l

-axis
lies upon
In one case (Fig. 8, first part) the
rotation
which
Y- plane on
that side of the
through a right
F-axis appears counterclockwise
angle from the X-axis to the
or positive according to the convention adopted in Trigonome-

types.

This relation

try.

may

be stated in another form.

axis be directed to the right

and the F-axis

the Z'-axis will point upward.

the

X-

right,
state-

Left-handed

FIG.

common

if

Right-handed

is

Or

and the F-axis to the


Still another method of

ment

X-

vertically, the

^-axis will be .directed toward the observer.


axis point toward the observer

If the

8.

in mathematical physics

and engineering.

If

a right-handed screw be turned from the X-axis to the Faxis it will advance along the (positive) Z-axis. Such a sys-

tem

of axes is called right-handed, positive, or counterclockwise. 2


It is easy to see that the F-axis lies upon that side of

Z JT-plane

the
axis

is

on which rotation from the .Z-axis to the Xand the X-axis, upon that side of

counterclockwise

the X-, Y-, or Z-axis the positive half of that axis is meant.
YThe
0.
plane means the plane which contains the X- and Y-axis, f. e., the plane z
2
convenient right-handed system and one which is always available consists
1

By

of the
first

thumb,

first finger,

and second finger of the right hand. If the thumb and


from the palm perpendicular to each other, and if the

finger be stretched out

second finger be bent over toward the palm at right angles to first finger, a righthanded system is formed by the fingers taken in the order thumb, first finger,
second finger.

VECTOR ANALYSIS

20

the YZ-plaue on which rotation from the F-axis to the ZThus it appears that the relation
axis is counterclockwise.

between the three axes

same

cyclic order

screw

is

is

perfectly symmetrical so long as the


If a right-handed
is observed.

X YZX Y

turned from one axis toward the next

it

advances

along the third.


In the other case (Fig. 8, second part) the Z'-axis lies upon
that side of the
F-plane on which rotation through a right
the
JT-axis
to the F-axis appears clockwise, or negfrom
angle

The

ative.

F-axis then lies

upon that

on which rotation from the

side of the ^"J^-plane

the JT-axis appears


made concerning
be
may
In this case, too,
relation to the F^-plane.
to

^"-axis

clockwise and a similar statement


the X-axis in

its

the relation between the three axes

symmetrical so long

is

X YZX Y

is preserved but it is just


cyclic order
the opposite of that in the former case. If a left-handed screw

as the

is

same

turned from one axis toward the next

the third.

Hence

this

system

is

it

advances along

called left-handed, negative,

or clockwise. 1

The two systems


metric.

One

mirror.

If the

is

are not superposable.

the

image

X- and F-axes

the

of

of the

They

other

two

as

are sym-

seen in a

different systems be

superimposed, the ^"-axes will point in opposite directions.


Thus one system may be obtained from the other by reversing
the direction of one of the axes.

little

thought will show

two of the axes be reversed in direction the system will


not be altered, but if all three be so reversed it will be.
that

if

Which
asmuch

of the

two systems be used, matters little. But ingeometry and mechanics differ

as the formulae of

slightly in the matter of sign, it is advisable to settle once for


all which shall be adopted.
In this book the right-handed or

counterclockwise system will be invariably employed.

1
left-handed system may be formed by the
one was formed by the right.

left

hand

just as

a right-handed

ADDITION AND SCALAR MULTIPLICATION


Definition

The

k will be reserved to deunit length drawn respectively in the


F-, and Z- axes of a right-handed rectan-

three letters

note three vectors of


directions of the X-,

i,

j,

gular system.
In terms of these vectors, any vector

= x'\ +

The

21

y\

may

be expressed as

ak.

(6)

coefficients x, y, z are the ordinary Cartesian coordinates

of the terminus of r
coordinates.

if its

origin be situated at the origin of


of r parallel to the JT-, F-, and

The components

.Z-axes are respectively

The

rotations about

about k from

By means

to

i,

from

z k.

j,

to k, about

from k to

i,

and

are all positive.

of these vectors

i,

j,

k such a correspondence

is

established between vector analysis and the analysis in Cartesian coordinates that it becomes possible to pass at will

from either one to the other.

There

is

nothing contradic-

On the contrary it is often desirable


tory between them.
or even necessary to translate the formulae obtained by
vector methods into

Cartesian coordinates for the sake of

comparing them with results already known and it is


more frequently convenient to pass from Cartesian

still

analysis to vectors both

on account of the brevity thereby

obtained and because the vector expressions show forth the


meaning of the formulae.

intrinsic

Applications
*18.] Problems in plane geometry
easily

by vector methods.

the plane

solved
in

be taken as the fundamental ones in terms of

others in that plane may be expressed.


The origin
also be selected at pleasure.
Often it is possible to

which

may

may

may frequently be

Any two non-collinear vectors

all

VECTOR ANALYSIS

22

choice of the origin and fundathat


the
mental vectors
analytic work of solution is materially
The adaptability of the vector method is about
simplified.

make such an advantageous

the same as that of oblique Cartesian coordinates with different scales upon the two axes.

(Example?^} The
gram
O

to the

onal (Fig.

Let

line

which

one vertex of a parallelo-

joins

side
middle point of an opposite
ii
-t

trisects the diagO

9).

A BCD

BE

be the parallelogram,
the line joining the
of the side
vertex B to the middle point

^^J

Z^^-^/

AC

is

the point in which this line cuts the

diagonal

A C.

A R is one third of
origin, A B and A D as the

To show

Choose A as
two fundamental vectors S and T.
Then
the sum of S and T. Let A~R = R. To show
'

AD, R
AC.

EB = A~R =
where x

is

is

(S

T).

AE + ER =| T + x (S - | T),

the ratio of

ER to EB

an unknown

E=y

(S

the scalar ratio of

jl(

And
where y

scalar.

T),
to

AG

to be

shown equal

toj.

Hence
or

x (S

-i

T)

= y (S +

T)

Sl

Hence, equating corresponding

coefficients (Art. 16),

ADDITION AND SCALAR MULTIPLICATION


From which

Inasmuch as x

also

is

well as the diagonal

g-

23

-.

EB

the line

must be

trisected as

A C.

Example %j) If through any point within a triangle lines


be drawn parallel to the sides the sum of the ratios of these
lines to their corresponding sides is 2.

Let

A B C be the triangle, R the point within


A B and A C as the two fundamental

and

Choose

it.

as origin,

vectors S

Let

T.

2TB

= R = mS+

wT.

(a)

m S is the fraction of A B which is cut off by the line through


R parallel to A C. The remainder of A B must be the fracm S. Consequently by similar triangles the ratio of
tion 1
the line parallel to A G to the line A C itself is (1
m).
Similarly the ratio of the line parallel to A B to the line A B
itself is

n ).

(1

Next express

third side of the triangle.

E = (m +
Hence (m
line

+ n) S

through

is

angles the ratio of this


the three ratios

and the theorem

and T

S the

Evidently from (a)


n) S

+n

(T

- S).

A B which is cut off by the


B C. Consequently by similar triline to B C itself is (m -f n).
Adding

w) +
is

in terms of S

the fraction of

R parallel

(1

to

(1

TO)

+ (m +

n)

= 2,

proved.

If from any point within a parallelogram lines


Example 3
be drawn parallel to the sides, the diagonals of the parallelo:

grams thus formed

intersect

upon the diagonal

of the given

parallelogram.

Let

A B CD

and Z^Vtwo

be a parallelogram,

lines

through

R parallel

?
N

a point within
respectively to

it,

KM

AB and

VECTOR ANALYSIS

24

AD,

the points K, Z,

M>

N lying

B C, CD respectively. To

LM

show

upon the

KEND

two parallelograms
on A C. Choose A as origin, A B and
mental vectors S and T. Let
of the

let

P be the

point of intersection of

KN=KR + RN = m S +

Then

Zlf =

m) S

(1

?i

and

A D as

and

sides

DA,AB,

that the diagonals

KN and

LBMR

meet

the two funda-

KN with LM.
(1

- n)

T,

T,

P = w T + x [m S + (1 - n) T],
P = mS + y [(l-m)S + wT],

Hence
and
Equating

coefficients,

By

solution,

=
=

m+

n
n

m
m+

7i

Substituting either of these solutions in the expression for P,


the result is

m+
which shows that P
*
19.]

solved in
sions.

is

collinear with

A G.

Problems in three dimensional geometry may be


essentially the same manner as those in two dimen-

In this case there are three fundamental vectors in


others can be expressed. The method of
analogous to that in the simpler case. Two

terms of which
solution

is

all

ADDITION AND SCALAR MULTIPLICATION

25

expressions for the same vector are usually found. The cocorresponding terms are equated. In this way

efficients of the

unknown

the equations between three

from which those

scalars

scalars are obtained

be determined by solution and

may

then substituted in either of the expressions for the required


vector.
The vector method has the same degree of adapta-

method

bility as the Cartesian

different scales are employed.

those in the foregoing section

which oblique axes with

in

The following examples like


are worked out not so much for

their intrinsic value as for gaining a familiarity with vectors.

Example 1
point within

Let

A B CD

be

a tetrahedron

Join the vertices to

it.

and

P and produce

until they intersect the opposite faces in A',

B\

any

the lines

C', D'.

To

show

PA'

PB'

PC'

A~A'

~B'

~GG'

PD' _
~

"

'

LTD'

AD

and the edges A B, AC,


three fundamental vectors B, C, D.
Let the vector
Choose

as origin,

= AP=l"B + mC + D,
A' = A A' = k P = k (I B + m C +
P

The

A'

vector
B.

BA'

Hence
A'

Equating

it

= AA' = AB+

=B+

x l (C

coefficients

Jc

kl
-

*-

D).

BA'.

BC =

- B) +
m = xv

=
I

B and
C
coplanar with
may be expressed in terms of them,

is

Hence
and

A P be

?i

Also

as the

~
n

-i

- B).

+m + n

y l (D

~4

SI

(7

+m+

N.

n).

BD =

VECTOR ANALYSIS

26

A B' = # 2 C + yz D

In like manner

~AB'

and

Hence

= ~AB '+

y2

and

&'= B + k

(P

- B).

= B + & 2 (/B + mC + wD= 1 -f k z (I - 1),


x z =Jc z m,

Hence

7/2

=&

&

l-l

&-:

PB'

and

n.

"2

In the same

way

may

be shown that

PC'

,PZ>'

it

Adding the four


1

ratios the result is


(l

+m+

n)

+l+

m+n = l.

Example 2 : To find a line which passes through a given


point and cuts two given lines in space.
Let the two lines be fixed respectively by two points A
be the given point. Choose
and B,
and D on each. Let
it as origin and let

V=OD.
Any

point

Any

point

P of A B may be

expressed as

= OP= OA + x TB = A + x
Q

If the points

are.collinear.

of

CD may likewise be

P and
That

lie

is

in the

same

(B

- A).

written

line

through

0,

P and

ADDITION AND SCALAR MULTIPLICATION

possible to equate coefficients one of the four


of the other three.

Before

it is

vectors

must be expressed in terms

= lA. + m'B + nC.


P = A + x (B - A)
= z [C + y (IA.+ ra B + nC
1
x=zy
=
x
zym,
= z [1 + y (n - 1)J.
m
x =

Let

27

J)

Then
Hence

0)].

I,

Hence

+m
1

Substituting in

P and Q

+m

_ A+ m B
+m
I

Either of these

and cutting

may be taken

as defining a line

drawn from

A B and

Vector Relations independent of the Origin

Example 1: To divide a

20.]

find the vector

the terminus
ratio

in a given ratio

is,

P =~OP

divides

of

AB

as

'&.

which
in the

A^*^^ P

&/ Px^^Jv^

f^^^

FlQ 10

n.

P= OP=
That

AB

(Fig. 10).

Choose any arbitrary point


Let OA = L and OB
origin.

To

line

Oil

-^ ~AB = &+ ^
m+
m+n
n A + m B
P =
m+n

n (B-A).
/7 ,
(<)

VECTOR ANALYSIS

28

The components

of

AP

to the segments
divided by the point P.

the line

AB

parallel to

and

PB

If it

and B are in inverse ratio


into which the line A B is

should so happen that

externally, the ratio

AP PB
/

P divided

would be nega-

and the signs of m, and n would be opposite, but the


formula would hold without change if this difference of sign
in m and n be taken into account.
tive,

Example 2

To find the

point of intersection of the medians

of a triangle.
at random.
Let A BC be the given
Choose the origin
=
'=
and
~OC = C. Let A\ B\ 0'
OA
OB
Let
B,
A,
triangle.
be respectively the middle points of the sides opposite the
vertices

A, B,

medians and

G.

Let

M be

the point of intersection of the

M = OM the vector drawn to

it.

Then

and

Assuming

that

has been chosen outside of the plane of the

triangle so that A, B, C are non-coplanar, corresponding coefficients

may be

equated.

Hence
Hence

=y=

ADDITION AND SCALAR MULTIPLICATION


The

29

vector drawn to the median point of a triangle is equal


sum of the vectors drawn to trie vertices.

to one third of the

In the problems of which the solution has just been given


the origin could be chosen arbitrarily and the result is independent of that choice. Hence it is even possible to disre-

gard the origin entirely and replace the vectors A, B, C, etc.,


by their termini A,B,C, etc. Thus the points themselves

become the subjects of analysis and the formulae read

nA

+mB

m+

M=\(A + B+C).

and
This

typical of a whole class of problems soluble

by vector
methods. In fact any purely geometric relation between the
different parts of a figure must necessarily be independent
of the origin assumed for the analytic demonstration.
In

some

is

cases,

such as those in Arts. 18, 19, the position of the

origin may be specialized with regard to some crucial point


but in many
of the figure so as to facilitate the computation
;

other cases the generality obtained by leaving the origin unspecialized and undetermined leads to a symmetry which

renders the results just as easy to compute and more easy


to

remember.

: The
necessary and sufficient condition that a
vector equation represent a relation independent of the origin
is that the sum of the scalar coefficients of the vectors on

Theorem

one side of the sign of equality


coefficients of the vectors

upon

is

sum
Or if

of the

equal to the

the other side.

all

the

terms of a vector equation be transposed to one side leaving


zero on the other, the

sum

of the scalar coefficients

be zero.

Let the equation written in the

latter

form be

must

VECTOR ANALYSIS

30

'

to
by adding a constant vector
Change the origin from
=
E 00' to each of the vectors A, B, C, D ---- The equation

then becomes

+ E) + b (B + B) + c
= aA + 5B + cC + dD +

a (A

If this is to be

must vanish.

That
is

(C
...

+ E) +

d (D

+B

j,

+ R) -f
+c+d+

...).

independent of the origin the coefficient of

Hence

this condition is fulfilled in the

two examples

cited

obvious.

m+

_*_.+
m+n m+

If

+ B +

M=|(A
i

*
21.]

The

_1
~
^
3
,

,1

7" *."

C),

necessary and sufficient condition that two

vectors satisfy an equation, in which the sum of the scalar


coefficients is zero, is that the vectors be equal in magnitude
and in direction.

First let

and

It is of course

vanish.

=
0.

coefficients a and I
would mean nothing. Sub-

assumed that not both the

If they did the equation

stitute the value of

the

A+

a obtained from the second equation into

first.

= B.

ADDITION AND SCALAR MULTIPLICATION


Secondly

if

and B are equal

the equation

subsists

in

31

magnitude and direction

A-B =
The sum

between them.

The necessary and

of the coefficients is zero.

sufficient condition that three vectors

satisfy an equation, in which the sum of the scalar coefficients


is zero, is that when drawn from a common origin they termi-

nate in the same straight

line. 1

A + &B + cC =
a + b + c = 0.

First let

and

Not all the coefficients a, b, c, vanish or the equations


would be meaningless. Let c be a non-vanishing coefficient.
Substitute the value of a obtained from the second equation
into the first.

-(& + c)A+&B + cC =
c (C-A) = &(A-B).

or

Hence the vector which


collinear with that

0,

joins the extremities of C

which

joins the extremities of

Hence those three points A, B, C lie on a


suppose three vectors A = OA, B = OB,C=
the same origin
terminate in a straight

line.

00

and

and

is

B.

Secondly

drawn from

line.

Then

the

vectors

A~B
are collinear.

= B - A and AC = C - A
r

Hence

the equation

(B
subsists.

The sum

- A) = x

(C

- A)

of the coefficients

on the two sides

is

the same.

in

The necessary and


which the sum of

sufficient condition that

an equation,

the scalar coefficients

zero, subsist

Vectors which hare a common origin and terminate


Hamilton " termino-collinear."
1

in

is

one

line are called

by

VECTOR ANALYSIS

32

between four vectors, is that


1
they terminate in one plane.

if

drawn from a common

origin

aA + 6B + cC + dD =
a + b + c + d = Q.

First let

and

Let d be a non-vanishing

Substitute the value

coefficient.

of a obtained from the last equation into the

first.

d(D-A) = 6 (A-B) + c (A - C).


The line A D is coplanar with A B and A C. Hence all four
D of A, B, C, D lie in one plane. Secondly
termini A, B,
or

(7,

suppose that the termini of A, B, C, D do lie in one plane.


Then A~D = D - A, ~AC = C - A, and
B - A are covectors.
of
One
them
be
planar
expressed in terms of
may

AB

the other two.


I

where

Z,

TO,

(B

This leads to the equation

- A) + m

- A) +

and n are certain

cients in this equation

of

(C

is

B, C, D,

- A) = 0,

The sum

scalars.

of the coeffi-

zero.

Between any five vectors there


whose coefficients is zero.
Let A,

n (D

be the

five

exists one equation the

sum

Form

the

given

vectors.

differences

E-A, E-B, E-C, E-D.


One

of these

may

be expressed in terms of the other three

what amounts to the same thing there must


equation between them.
or

k (E

- A) +

The sum

(E

- B) + m

(E

C)

+n

of the coefficients of this equation

1
Vectors which have a common origin and terminate
"
by Hamilton termino-complanar.
1'

(E

exist

- D) ==

an

0.

is zero.

in

one plane are called

ADDITION AND SCALAR MULTIPLICATION

33

The results of the foregoing section afford simple


*22.]
solutions of many problems connected solely with the geometric properties of figures. Special theorems, the vector
equations of lines and planes, and geometric nets in
three dimensions are taken up in order.

Example 1:

If a line

two and

be drawn parallel to the base of a

triangle, the line which joins the opposite vertex to the intersection of the diagonals of the

trapezoid thus formed bisects the


base (Fig. 11).
Let
be the triangle,

ABC

ED

the line parallel to the base CB,


G the point of intersection of the

and DC of the trapezoid CBDE, and jP'the intersection of A G with CB.


To show
that F bisects CB.
Choose the
diagonals

EB

rIG. 11.

Let the vectors drawn from it to the


origin at random.
various points of the figure be denoted by the corresponding

Then

Clarendons as usual.
lel to

-D=n

E
The sum

holds true.
it

since

ED is by hypothesis

paral-

CB, the equation

should be.

(C

- B)

is evidently zero as
that the equation
terms
so
the
Rearrange

of the coefficients

takes on the form

WC

=D

w B.

n C is coplanar with E and C. It must cut


n B is coplanar with D
The equal vector D
and B. It must cut the line DB. Consequently the vector
represented by either side of this equation must pass through
the point A. Hence

The

vector

the line

E C.

7iC

=D

wB =

a;A.

VECTOR ANALYSIS

34

However the points E, C, and A lie upon the same straight


Hence the equation which connects the vectors E, C,
line.
and A must be such that the sum of its coefficients is zero.
This determines x as 1

By

nC = D -

Hence

n.

n~B

(1

- w) A.

another rearrangement and similar reasoning

E+
Subtract the

first

wB=D + wC= (1 + w) G.

equation from the second

n (B + C)
This

vector

cuts

BO

(1

n)

and AG.

(1

It

- n) A.

must therefore be a

that the
multiple of F and such a multiple
cients of the equations which connect B,

and F

Hence

sum
C,

of the coeffi-

and F or

G, A,

shall be zero.

n (B + C)

(1

n)

Hence

(1

=B+

A=

2 nJf.

C
>

SB

The proof has covered


considerable space because each detail of the reasoning has
been given. In reality, however, the actual analysis has con-

and the theorem has been proved.

sisted of just four equations obtained simply

Example 2 :

To determine

from the

first.

the equations of the line and

plane.

Let the line be fixed by two points A and B upon it. Let
be any point of the line. Choose an arbitrary origin.
The vectors A, B, and P terminate in the same line. Hence

and
Therefore

+ p = 0.
aA + 6B
P=
a + 6
a

ADDITION AND SCALAR MULTIPLICATION


For

different

points

the scalars a and I have different

They may be replaced by x and y, which


Then
to represent variables.

values.

35

are used

more generally

_ xA. + yE
x

Let a plane be determined by three points A,B, and C.


Let P be any point of the plane. Choose an arbitrary origin.

The

vectors A, B,

and P terminate in one plane.

C,

aA+
Therefore

a,

Z>,

c,

SB + cC

+ c+p = Q.
aA. + JB + cC
P=
a + b + c
a

and

As

Hence

vary for different points of the plane,

customary to write in their stead x, y,

The

Example 3 ;

line

which

it is

more

z.

joins

one vertex of a com-

plete quadrilateral to the intersection of


divides the opposite sides har-

two diagonals

monically (Fig. 12).


Let A, B, C, D be four vertices

Let

of a quadrilateral.

CD

A B meet
and A D

in a fifth vertex E,

BC

meet

in the sixth vertex F.

Let the two diagonals

AC

BD

intersect in

To show

that

FG

G.

AB

intersects

AB

such that the lines

E' and

by E.

E"

in the

That

is

(A

to

same

and

in a point

and

CD

E' and

CD

in a point

1'

are divided internally at

ratio as they are divided externally

show that the

cross ratios

B EE') = (CD -EE") = - 1.


.

VECTOR ANALYSIS

36

Choose the origin at random. The four vectors A, B, C, D


drawn from it to the points A, B, C\ D terminate in one

Hence

plane.

and

aA + &B + cC + dD =
a-f-& + c + ^ = 0.

Separate the equations by transposing two terms

A + cC = (Z>B
a + c =
(b + d).
= A + cC 5B +
Divide
a + c
b + d
6B + cC
A + d"D
In like manner
F=
a + d
b + c
cG
(a + c)0
(a + d)?
Form:
(a + c)
(a + d)
(a + c)
(a +

= cC-dJ> =

or
c

Separate the equations again and divide

A + JB

cC
c

+ dD =
+d

d)

E.

(&)

E divides A B in the ratio a b and CD in the ratio


But equation (a) shows that J" divides CD in the
c : d.
Hence E and E" divide CD internally and
ratio
Which of the two divisions is
externally in the same ratio.
internal and which external depends upon the relative signs
of c and d.
If they have the same sign the internal point
Hence

<2.

E; if opposite signs, it is E". In a similar way


E' and E may be shown to divide A B harmonically.
Example 4 To discuss geometric nets.

of division

is

'

a geometric net in a plane is meant a figure composed


of points and straight lines obtained in the following manner.
Start with a certain number of points all of which lie in one

By

ADDITION AND SCALAR MULTIPLICATION


Draw

plane.

These

37

the lines joining these points in pairs.

all

lines will intersect each other in a

number

of points.

the lines which connect these points in pairs.


This second set of lines will determine a still greater number

Next draw
of points

all

which may in turn be joined in

pairs

and so

on.

The construction may be kept up indefinitely. At each step


the number of points and lines in the figure increases.
Probably the most interesting case of a plane geometric net is
that in which four points are given to commence with.
Joining these there are six lines which intersect in three
points different from the given four. Three new lines may
now be drawn in the figure. These cut out six new points.

From these more lines may be obtained and so on.


To treat this net analytically write down the equations

aA + 6B + cC + dD =
a + b + c + d = Q

and

(c)

which subsist between the four vectors drawn from an unde-

From

termined origin to the four given points.


possible to obtain

E
F

= A+
a

=o

cC

b"B

+_c_

these

it is

= dD
+

+d
"B + cG
b + c

splitting the equations into two parts and dividing. Next


four vectors such as A, D, E, F may be chosen and the equa-

by

tion the

sum

of

whose

coefficients is zero

may

be determined.

This would be
aA.

d'D

(a

b)'E+

treating this equation as (c)


be obtained.

By

(a

c)

= 0.

was treated new points may

VECTOR ANALYSIS

38

aA

+ dD
+d

(a

a
1

- a A + (a

&)

6) E + (a +
2a + b + c

_ dD +

+ c) F
+c+d
(a

dD + (a +
a + b +

c)

6)E
d

Equations between other sets of four vectors selected from


A, B,C, D, E, F, G may be found; and from these more points

The

process of finding more points goes forward


fuller account of geometric nets may be
indefinitely.
" Elements
Hamilton's
found in
of Quaternions," Book I.
obtained.

As

regards geometric nets in space just a word may be


Five points are given. From these new points may be
obtained by finding the intersections of planes passed through

said.

sets of three of the given points

The

with lines connecting the


may then be carried for-

construction

remaining pairs.
ward with the points thus obtained.
is

The

analytic treatment
similar to that in the case of plane nets.
There are

drawn from an undetermined origin to the given


Between these vectors there exists an equation
the sum of whose coefficients is zero. This equation may be
separated into parts as before and the new points may thus
five vectors

five points.

be obtained.

aA + b"B
a + b

then

a
are

two

way.

of the points

cG

+ dD + E
+d+e

and others may be found

in the

Nets in space are also discussed by Hamilton,

same

loc. cit.

ADDITION AND SCALAR MULTIPLICATION

39

Centers of Gravity
*

The center of gravity of a system of particles may


23.]
be found very easily by vector methods. The two laws of
physics which will be assumed are the following:

1.

The

center of gravity of two masses (considered as


situated at points) lies on the line connecting the two masses

and divides

it

into

two segments which are inversely pro-

portional to the masses at the extremities.

2.

In finding the center of gravity of two systems of


may be replaced by a single mass equal

masses each system


in

magnitude to the sum of the masses in the system and

situated at the center of gravity of the system.


Given two masses a and b situated at two points

Their center of gravity

is

and B.

given by

where the vectors are referred

any origin whatsoever.

to

This follows immediately from law 1 and the formula (7)


for division of a line in a given ratio.

The

center of gravity of three masses

three points A, B,

masses a and

mass a

-f b

C may

be found by means of law

= (a +

A + &B
a + b

6)

a
a

Hence

situated at the
2.

The

may be considered as equivalent to a single


situated at the point
b

Then

a, o, c

+b+

= a A + &B +
a

cC

VECTOR ANALYSIS

40

Evidently the center of gravity of any number of masses


a, 6,

c,

d,

. . .

situated at the points

The

be found in a similar manner.

A,

C, Z>, ...

JB,

may

result is

_ aA + &B + cC + ^D +
a + l + c + d+ ...
Theorem 1 :

The

lines

which

join the center of gravity of a

triangle to the vertices divide it into three triangles

which

are proportional to the masses at the op-

Let

posite vertices (Fig. 13).

A,B,C

be the vertices of a triangle weighted


with masses a, Z>, c. Let G be the center of gravity.

Join A, B,

to

G and

produce the lines until they intersect


the opposite sides hi A', 2?', C" respectively.
To show that
the areas

The

GBC:GCA:GAB:ABC=a,:l:c:a + l +
last proportion between ABC and a + I + c

from compounding the

first three.

It

is,

c.

comes

however, useful in

the demonstration.

ABC ~ A A'
GBC

Hence

GA' _b +
=

~~

ABC
GBC

In a similar manner

and

AG

G A'
a

BCA a + J 4- c
GCA
b
CAB _
a + I +
=
GAB
~c

Hence the proportion is proved.


Theorem 2 : The lines which

'

join the center of gravity of


a tetrahedron to the vertices divide the tetrahedron into four

ADDITION AND SCALAR MULTIPLICATION

41

tetrahedra which are proportional to the masses at the opposite vertices.

C, D be the vertices of the tetrahedron weighted


with
respectively
weights a, b, c, d. Let G be the center of
Join A, B, C, D to G and produce the lines until
gravity.

Let A, B,

they meet the opposite faces in A',


the volumes

To show

D'.

C\

',

that

BCDG:CDAG:DABG:ABCG:ABGD
=a

BCD A ~ A A'
BCDG ~G~A7

d a
:

AG

+c+

d.

G A'

+c+

"

~GA'

~GAJ
b

+c+

a
In like manner

and

CD A G _ a + b + c + d
"'

CDAB~
b
DABG _a
+b+
=

d
~

DABC
ABCG _ a+b+c+d
"'

'

and

A BCD'

~d~

which proves the proportion.


*

24.]
By a suitable choice of the three masses, a, 6, c located at the vertices A, B, C, the center of gravity G may
be made to coincide with any given point
of the triangle.

not obvious from physical considerations it cerbecomes


so in the light of the foregoing theorems.
tainly
If this be

For in order that the center of gravity

fall at P, it is

necessary to choose the masses a,

proportional to the

areas of the triangles

Thus not merely one


an infinite number of

PBC,

PC A,

and

set of masses a,

which

&,

&,

PAB
c

only

respectively.

may be

found, but

from each other only


a
common
of
factor
These quantities
by
proportionality.
sets

differ

VECTOR ANALYSIS

42
a, 5, c

points

may

therefore be looked

upon

as coordinates of the

ABC. To

inside of the triangle

each set there

corresponds a definite point P, and to each point

there

corresponds an infinite number of sets of quantities, which


however do not differ from one another except for a factor
of proportionality.

To

obtain the points

of the triangle

ABC

of the plane

one

ABC which lie outside

resort to the conception of

may

negative weights or masses. The center of gravity of the


and
1 situated at the points
masses 2 and
respectively
the
line
a
G
would be
externally in the
point
dividing

A
AB

ratio 1

That

2.

is

B produced may be represented by


point of the line
a suitable set of masses a, b which differ in sign.
Similarly

Any

any point

of the

plane

suitable set of masses a,

from the other two

ABC.

Z>,

ABO
c of

may be

represented by a

which one

will differ in sign

the point
lies outside of the triangle
as only the ratios of a, Z>, and c are imif

Inasmuch
two
of the quantities may always be taken positive.
portant
The idea of employing the masses situated at the vertices
as coordinates of the center of gravity is due to Mobius and
was published by him in his book entitled " Barycentrische
This

Calcul" in 1826.
point of

modern

may be

fairly

regarded as the starting

analytic geometry.
of negative masses which have

The conception

no existence

in nature
vertices

may be avoided by replacing the masses at the


by the areas of the triangles GBC, GCA, and

GAB

to which they are proportional.


The coordinates of
a point
would then be three numbers proportional to the
areas of the three triangles of which
is the common vertex ;

and the

sides of a given triangle

of these areas

is

ABC,

the bases.

The

determined by the following definition.

sign

ADDITION AND SCALAR MULTIPLICATION


The area

Definition:
positive

when

ABC

positive or counterclockwise

a triangle

of

C follow

the vertices A, B,

direction

is

43

said to be

each other in the

upon the

circle de-

scribed through them. The area is said to be negative when


the points follow in the negative or clockwise direction.
not alter
Cyclic permutation of the letters therefore does

the sign of the area.

ABG = BCA = CAB.


Interchange of two letters which amounts to a reversal of
the cyclic order changes the sign.

If

P be

any point within the triangle the equation

PA B + PB C + PGA = A B C
must

The same

hold.

will also

hold

if

be outside of the

triangle provided the signs of the areas be taken into conThe areas or three quantities proportional to
sideration.

them may be regarded as coordinates of the point P.


"
The extension of the idea of " larycentric coordinates
space is immediate. The four points A, B, C, D situated
the vertices of a tetrahedron are weighted with mass a, &,
The center of gravity Gr is represented
respectively.
these quantities or four others proportional to them.
obtain

points

outside

to
at

c,

by

To

the tetrahedron negative masses


in the light of theorem 2, page 40,

of

be employed. Or
may be replaced by the four tetrahedra which
are proportional to them.
Then the idea of negative volumes takes the place of that of negative weights. As this

may

the masses

idea
it

is

here

of considerable importance later, a brief treatment of

Definition
to

not be out of place.

may
:

be positive

The volume A B CD of a tetrahedron is said


when the triangle ABC appears positive to

VECTOR ANALYSIS

44

The volume is negative


the eye situated at the point D.
if the area of the triangle appear negative.
To make the discussion of the signs of the various
tetrahedra perfectly clear
solid model.

A plane

difficult to see

from

it

almost necessary to have a

it is

drawing is scarcely sufficient. It is


which triangles appear positive and

which negative. The following relations will be seen to


hold if a model be examined.
The interchange of two letters in the tetrahedron A B CD
changes the sign.

ACBD=CBAD=BACD=DBCA
The
sible

sign of the tetrahedron for any given one of the postwenty-four arrangements of the letters may be obtained

D by
by reducing that arrangement to the order A B
means of a number of successive interchanges of two letters.
If the number of interchanges is even the
sign is the same
as that of

AB CD ;

if

Thus

odd, opposite.

CADB = -CABD = + A CBD = -ABCD.


If

is

any point inside of the tetrahedron

ABCD

the

equation

ABCP-BCDP+ CDAP-DABP=ABCD
P

holds good. It still is true if


be without the tetrahedron
provided the signs of the volumes be taken into considera-

The equation may be put into a form more symmetriand more easily remembered by transposing all the terms
to one number.
Then

tion.

cal

ABCD + BCDP + CDPA + DPAB+PABC=Q.


if

The proportion in theorem 2, page 40, does not hold true


the signs of the tetrahedra be
regarded. It should read

BCDG:CDGA:DOAB:GABC:ABCD
=a

-f b

d.

ADDITION AND SCALAR MULTIPLICATION

45

If the point G- lies inside the tetrahedron a, 5, c, d represent quantities proportional to the masses which must be
located at the vertices A,B,C,D respectively if G is to be the

G lies outside of the tetrahedron they may


be regarded as masses some of which are negative
or
as
four
better
numbers
whose
ratios determine
perhaps
merely
center of gravity. If

still

the position of the point G-. In this manner a set of "Jary"


centric
coordinates is established for space.

The vector P drawn from an indeterminate


point of the plane

A B G is
_

Comparing

this

origin to

any

(page 35)

+ yB + aC
+y+

a?A

with the expression

_ A+
a

6B + cC
b

05, y z are in reality nothing


than the barycentric coordinates of the point
with respect to the triangle ABO. In like manner from

it

will be seen that the quantities

more nor

less

equation

_xA. +
x

y'B

+ zG + w'D
+z+w

which expresses any vector P drawn from an indeterminate


origin in terms of four given vectors A, B, C, D drawn from
the same origin, it may be seen by comparison with
aA.

l'B
b

+ c C + d"D
+c+d
z, w are precisely

the barythat the four quantities #, y,


of
centric coordinates of P, the terminus
P, with respect to
CD. Thus the vector methods in which
the tetrahedron
the origin is undetermined and the methods of the " Bary"
are practically co-extensive.
centric Calculus

AB

It

was mentioned before and

it

may be

well to repeat here

VECTOR ANALYSIS

46
that the origin

may

be

left

wholly out of consideration and

The

the vectors replaced by their termini.

vector equations

then become point equations

xA + y B + zC
x

P=

and

xA + yB + z C + wD
x

+ z + w.

This step brings in the points themselves as the objects of


"
"
Barycentrische Calcul
analysis and leads still nearer to the
"
"
of Grassmann.
of Mb'bius and the Ausdehnungslehre

The Use of Vectors


Definition

25.]

An

denote

area lying in one

bounded by a continuous curve


itself

to

PQR

Areas
plane

PQR

letters

At

and

when

said to appear positive from the point

is

MN

which nowhere cuts


follow

the

each

other in the counterclockwise


or positive order; negative,
when they follow in the

negative or clockwise order


(Fig. 14).
It is evident that an area

can have no determined sign


per se^ but only in reference

Cf

FIG.

14.

to that direction in

boundary
side of

PQR;

is

which

supposed to be traced and to some point

For the area P


plane.
and an area viewed from

its

is

its

out-

negative relative to
negative relative to the
is

same area viewed from a point 0' upon the side of the plane
opposite to 0. A circle lying in the JTF-plane and described
in the positive trigonometric order appears positive from every
point on that side of the plane on which the positive Z-axis
lies,

but negative from

all

points on the side

upon which

ADDITION AND SCALAR MULTIPLICATION


the negative ^-axis

and the direction

lies.

For

47

view
must
be
boundary
kept

this reason the point of

of description of the

clearly in mind.

Another method

of stating the definition is as follows

If

a person walking upon a plane traces out a closed curve, the


is said to be positive if it lies upon his left-

area enclosed

hand

negative if upon his right. It is clear that if two


persons be considered to trace out together the same curve by
walking upon opposite sides of the plane the area enclosed
side,

will lie

upon the right hand

other.

To one

it

seems positive

which

is

and the

left

hand

will consequently appear positive

That

other, negative.

of one

side of the plane

called

the positive

of the
to the

upon which the area


side

the side

upon

appears negative, the negative side. This idea is


familiar to students of electricity and magnetism.
If an
it

around a closed plane curve the lines of


magnetic force through the circuit pass from the negative to
electric current flow

the positive side of the plane.


positive magnetic pole
placed upon the positive side of the plane will be repelled by
the circuit.

A plane

area

may be looked upon

as possessing

more than

or negative magnitude. It may be considered to


possess direction, namely, the direction of the normal to the
Hence a plane
positive side of the plane in which it lies.
positive

area is a vector quantity. The following theorems concerning


areas when looked upon as vectors are important.

Theorem 1

magnitude
direction

is

is

If a plane area be denoted by a vector whose


the numerical value of that area and whose

the normal upon the positive side of the plane,

then the orthogonal projection of that area upon a plane


will be represented by the component of that vector in the
direction normal to the plane of projection (Fig. 15).
lie in the plane MN.
Let it be projected
Let the area
M'
N'.
the
Let
and M' N' interplane
orthogonally upon

MN

VECTOR ANALYSIS

48
sect in the line

and

let the diedral

angle between these

MN

two planes be x. Consider first a rectangle PQRS in


whose sides, PQ, RS and QR, SP are respectively parallel
and perpendicular to the line I. This will project into a
rectangle P'Q'R'S' in M'N'. The sides P' Q' and R' S'
will be equal to PQ and RS; but the sides Q' R' and S'P'
will be equal to QR and SP multiplied by the cosine of #,
the angle between the planes.

P'Q'R'S'

Consequently the rectangle

= PQRS coax.

FIG. 15.

Hence
parallel

two

rectangles, of

which the

and perpendicular to

I,

sides

are

respectively
the line of intersection of the

planes, project into rectangles

whose

sides are likewise

and whose area is


respectively parallel
equal to the area of the original rectangles multiplied by the
cosine of the angle between the planes.
and perpendicular

From

to

projected into an
equal to the given area multiplied by the cosine
of the angle between the planes.
For any area
may be divided up into a large number of small rectangles by drawing a
this it follows that

area which

any area

is

is

series of lines in

MN parallel and perpendicular to the line

/.

ADDITION AND SCALAR MULTIPLICATION

49

Each of these rectangles when projected is multiplied by the


cosine of the angle between the planes and hence the total
On the
area is also multiplied by the cosine of that angle.
other hand the component A' of the vector A, which represents the given area, in the direction normal to the plane
M'N' of projection is equal to the total vector A multiplied

by the cosine of the angle between its direction which is


and the normal to M'N'. This
the normal to the plane
x
for
the
between
the normals to two planes
is
;
angle
angle

MN

is

the same as the angle between the planes.


A and A' is therefore

The

relation

between the magnitudes of

A'

=A

cos

ic,

which proves the theorem.


Definition

26.]

said to be

Two

plane areas regarded as vectors are

added when the vectors which represent them are

added.

A vector
ponents

area

is

obtainable

consequently the

by orthogonal

sum

of its three

projection

upon

comthree

Moreover in adding two


mutually perpendicular planes.
areas each may be resolved into its three components, the
corresponding components added as scalar quantities, and
these

sums compounded

as vectors into the resultant area.

A generalization of this statement to the case where the three


planes are not mutually orthogonal and where the projection

oblique exists.
surface made up of several plane areas may be represented by the vector which is the sum of all the vectors
is

representing those areas.

In case the surface be looked upon

as forming the boundary of a portion of the boundary of a


solid, those sides of the bounding planes which lie outside of

the body are conventionally taken to be positive.


The vectors which represent the faces of solids are always directed

out from the

solid,

not into

it.

VECTOR ANALYSIS

50
Theorem 2
surface

The vector which represents

a closed polyhedral

is zero.

be proved by means of certain considerations of


Suppose the polyhedron drawn in a body of
hydrostatics.

This

fluid

may

assumed

cluded.

pressures.

surface

to be free

from

all

external forces, gravity in-

The fluid is in equilibrium under its own internal


The portion of the fluid bounded by the closed

moves neither one way nor the

other.

Upon each face

of the surface the fluid exerts a definite force proportional


to the area of the face

these forces

and normal

must be zero,

to

it.

The resultant of all


Hence

as the fluid is in equilibrium.

sum of all the vector areas in the closed surface is zero.


The proof may be given in a purely geometric manner.

the

Consider the orthogonal projection of the closed surface upon


any plane. This consists of a double area. The part of the
surface farthest from the plane projects into positive area ;
the part nearest the plane, into negative area. Thus the
surface projects into a certain portion of the plane which is
covered twice, once with positive area and once with negative.

Hence the total projection of a


a
upon plane (if taken with regard to sign) is
But
zero.
by theorem 1 the projection of an area upon a
plane is equal to the component of the vector representing
that area in the direction perpendicular to that plane. Hence
These cancel each other.
closed surface

the vector which represents a closed surface has no component


along the line perpendicular to the plane of projection. This,

however, was any plane whatsoever.

Hence the vector

is

zero.

The theorem has been proved


closed surface consists of planes.

for the case in

which the

In case that surface be

1
Such a state of affairs is realized to all practical purposes in the case of a
polyhedron suspended in the atmosphere and consequently subjected to atmospheric pressure. The force of gravity acts but is counterbalanced by the tension

in the suspending string.

ADDITION AND SCALAR MULTIPLICATION


curved

it

may

51

be regarded as the limit of a polyhedral surface

whose number

of faces increases without limit.

Hence the

vector which represents any closed surface polyhedral or


curved is zero. If the surface be not closed but be curved it

be represented by a vector just as

were polyhedral.
approached by the vector which
represents that polyhedral surface of which the curved surface
is the limit when the number of faces becomes
indefinitely

may

That vector

is

the limit

if it

great.

SUMMARY OF CHAPTER

A vector is a quantity considered as possessing magnitude


and direction. Equal vectors possess the same magnitude
and the same direction. A vector is not altered by shifting it
A null or zero vector is one whose magparallel to itself.
nitude

is zero.

multiply

its

unchanged.

To

multiply a vector by a positive scalar

length by that scalar and leave its direction


To multiply a vector by a negative scalar mul-

length by that scalar and reverse its direction.


Vectors add according to the parallelogram law. To subtract

tiply its

its direction and add.


Addition, subtracand multiplication of vectors by a scalar follow the same
laws as addition, subtraction, and multiplication in ordinary-

a vector reverse
tion,

algebra.

vector

may be

resolved into three components

any three non-coplanar vectors.


can be accomplished in only one way.

parallel to

The components

= xa, +

y'b

This resolution

zc.

(4)

of equal vectors, parallel to three given

non-coplanar vectors, are equal, and conversely if the components are equal the vectors are equal. The three unit
vectors

i, j,

k form a right-handed rectangular system.

1 This limit exists and is


unique. It is independent of the
the polyhedral surface approaches the curved surface.

method

in

In
which

VECTOR ANALYSIS

52

terms of them any vector may be expressed by means of the


Cartesian coordinates x, y, z.
r

is

y\

zk.

(6)

The

point which divides a line in a given


given by the formula

Applications.
ratio

= xi +

m+

(7)

The necessary and

sufficient condition that a vector equation


a
relation
independent of the origin is that the sum
represent
Between
of the scalar coefficients in the equation be zero.

any four vectors there


If the

cients.

sum

an equation with scalar

exists

of the coefficients

is

coeffi-

zero the vectors are

termino-coplanar. If an equation the sum of whose scalar


coefficients is zero exists between three vectors they are
termino-collinear.

masses

a,

Z>,

The center
situated

at

of

the

gravity of a number of
termini of the vectors

common

supposed to be drawn from a


given by the formula
A, B, C

origin

is

.
-

vector

may

-\

----

be used to denote an area.

If the area is

plane the magnitude of the vector is equal to the magnitude


of the area, and the direction of the vector is the direction of
the normal

upon the

positive side of the plane.

representing a closed surface

is

The

vector

zero.

EXEECISES ON CHAPTER I
1.

Demonstrate the laws stated in Art.

2.

triangle

and equal

to the

may

12.

be constructed whose sides are parallel

medians of any given triangle.

ADDITION AND SCALAR MULTIPLICATION


3.

The

six points in

53

which the three diagonals of a com-

meet the pairs of opposite sides lie three


four
three
upon
straight lines.
by
If
two
4.
triangles are so situated in space that the three
plete quadrangle

points of intersection of corresponding sides lie on a line, then


the lines joining the corresponding vertices pass through a

common

point and conversely.


Given a quadrilateral in space. Find the middle point
5.
of the line which joins the middle points of the diagonals.

Find the middle point of the line which joins the middle
Show that these two points are
points of two opposite sides.
the same and coincide with the center of gravity of a system
of equal masses placed at the vertices of the quadrilateral.
If two opposite sides of a quadrilateral in space be
6.
divided proportionally and if two quadrilaterals be formed by
joining the two points of division, then the centers of gravity

of these

two quadrilaterals

lie

on a

gravity of the original quadrilateral.


is

meant the center

the vertices.

Can

line with the center of

By the center of gravity

of gravity of four equal masses placed at


this

theorem be generalized to the case

where the masses are not equal?


7.

The

bisectors of the angles of a triangle

meet in a

point.
8.

If the edges of a

hexahedron meet four by four in three

points, the four diagonals of the hexahedron meet in a point.


In the special case in which the hexahedron is a parallelepiped

the three points are at an infinite distance.


9.
Prove that the three straight lines through the middle
points of the sides of

any face

of a tetrahedron, each parallel

to the straight line connecting a fixed point


with the middle point of the opposite edge of the tetrahedron, meet in a

A complete quadrangle consists of the six straight lines which may be passed
through four points no three of which are collinear. The diagonals are the lines
which join the points of intersection of pairs of sides.
1

VECTOR ANALYSIS

54
point

and

is

10.

PE

E and

that this point is such that


passes through
by the center of gravity of the tetrahedron.
Show that without exception there exists one vector

bisected

equation with scalar coefficients between any four given


vectors A, B,

C,

L.

Discuss the conditions, imposed upon three, four, or


five vectors if they satisfy two equations the sum of the co11.

efficients in

each of which

is zero.

CHAPTER
DIRECT AND

II

SKEW PEODUCTS OF VECTORS

Products of

Two

Vectors

THE

operations of addition, subtraction, and scalar


multiplication have been defined for vectors in the way
27.]

suggested by physics and have been employed in a few


It now becomes necessary to introduce two
applications.

new

combinations of vectors. These will be called products


because they obey the fundamental law of products i. e., the
distributive law which states that the product of A into the
;

sum of B and C is equal to the sum of the products of A into


B and A into C.
The direct product of two vectors A and B is
Definition :
the scalar quantity obtained by multiplying the product of
the magnitudes of the vectors by the cosine of the angle be-

tween them.

The direct product is denoted by writing the two vectors


with a dot between them as
A-B.
This

is

read

dot

B and

therefore

may

often be called the

dot product instead of the direct product.


the scalar product owing to the fact that
lar.

If

definition

be the magnitude of

A-B = ^cos

and

(A,B).

It
its

is

also called

value

is

sca-

that of B, then

(1)

Obviously the direct product follows the commutative law

A B=B

A.

(2)

by

VECTOR ANALYSIS

56
If either vector

be multiplied by a scalar the product

That

multiplied by that scalar.

(x A)

=A

In case the two vectors

is

is

(x B)

= x (A

and B are

B).

collinear the angle be-

tween them becomes zero or one hundred and eighty degrees


and its cosine is therefore equal to unity with the positive or

Hence the scalar product of two parallel


negative sign.
vectors is numerically equal to the product of their lengths.
The sign of the product is positive when the directions of the
vectors are the same, negative when they are opposite.
The
of
a
is
itself
therefore
to
the
vector
product
equal
by
square
of its length

A-A=^

if

Consequently
vector

is

2.

(3)

the product of a vector by itself vanish the

a null vector.

In case the two vectors

and B are perpendicular the

angle between them becomes plus or minus ninety degrees


and the cosine vanishes. Hence the product A B vanishes.

Conversely

if

the scalar product

AB
Hence

A B

cos (A, B)

vanishes, then
0.

or cos (A, B) is zero, and either the


vectors are perpendicular or one of them is null. Thus the
condition for the perpendicularity of two vectors, neither of
either

which vanishes,
28.]

vectors

The
k

i, j,

is

or

A B=

scalar products of the three fundamental unit

are evidently

i.i

If

0.

= j.j = k-k = l,

(4)

more generally a and b are any two unit vectors the

product
a

= cos

(a, b).

DIRECT AND SKEW PRODUCTS OF VECTORS


Thus the

57

scalar product determines the cosine of the angle


is in a certain sense equivalent to

between two vectors and

For

it.

this reason

might be better to give a purely


the product rather than one which

it

geometric definition of

depends upon trigonometry. This is easily accomplished as


If a and b are two unit vectors, a b is the length

follows

upon the other. If more generally


and B are any two vectors A B is the product of the length

of the projection of either

by the length

of either

From

of projection of the other

upon

it.

these definitions the facts that the product of a vector

the square of its length and the product of two


perpendicular vectors is zero follow immediately. The trigo-

by

itself is

nometric definition can also readily be deduced.

The

scalar product of

two vectors

cosine of the included angle

may be

examples
vector

The

whenever the

will appear

The following
a vector B upon a

of importance.

projection of

is
'

A A
where a

cited.

is

is

A=

AA

a cos (A, B)

= B cos

(A,

a unit vector in the direction of A.

B)
If

a,

(5)

is itself

unit vector the formula reduces to

If

(A B) A = B cos (A, B) A.
be a constant force and B a displacement .the work done

by the force

during the displacement

is

B.

If

repre-

sent a plane area (Art. 25), and if B be a


vector inclined to that plane, the scalar product A B will be the volume of the cylinder
of

which the area

is

the

base and of

which B

For
is the directed slant height.
the volume (Fig. 16) is equal to the base
multiplied by the altitude h. This is

FlG

the projection of

B upon A

or

v = A h = A B cos

cos (A, B).

(A, B)

=A

B.

Hence

VECTOR ANALYSIS

58

The

29.]

scalar or direct product follows the distributive

That

law of multiplication.

(A

is

B)

=A

+B

C.

(6)

may be proved by means of projections. Let C be


magnitude C multiplied by a unit vector c in its
To show

This

equal

its

direc-

to

tion.

(A
or

B)

(A

=A
c = A

(G c)
B)

c is the projection of

(A

B)

c,

that of

(A

is

+B

c.

B upon

that of

of the projections.

ft+---)

scalar product

nary algebra.
If

three unit vectors

and

no peculiar

A and B
k

i, j,

A=A

2 j

k)

+ A Bz i k
k
j + A 2 #, j
2 j
^ 3 8 k.k.
2 k.j +

of (4) this reduces to

A B = A l B + 4, 2 + A 8 JSS
particular A and B are unit vectors, their
.

A ^A Z ^A 8
lines

l j

terms of the

i + A
+ A k,
z
= B i + #2 + Bz k,

If in

as

= A B^ i i + A
+ A2 B i + A2
+ ^ 3 k.j + ,4 3
By means

...

difficulties.

are expressed in

A.B^^i + ^j + ^gk) .(^1 +

then

Hence

be used just as the product in ordi-

may

It has

two vectors

= A.P + A.a+...
+ B.P + B.Q +

The

the projection of the

By an immediate generalization

proved.

+ B+...)-(P +

sum

B. (Co)

c,

equal to the

is

the relation (6)

A upon c B
A + B upon c. But

sum A + B

(tfc)

and JS^Szt

A and B

&

are

(7)

components

the direction cosines of the

referred to X, Y, Z.

DIRECT AND SKEW PRODUCTS OF VECTORS

A =
1

^ = cos
Moreover

A z = cos (A,
= cos (B,
3

cos (A, JT),


(B, JT),

A B

is

59

A = cos (A,
^3 = cos (B <&)

F),

F),

"),

Hence

the cosine of the included angle.

the equation becomes


cos (A, B)

= cos

(A,

Z)

cos (B,

Z) +

+
A and B

In case

known

cos (A,

cos

F) cos

(A,)

(B,

F)

cos (B,^).

are perpendicular this reduces to the well-

relation

= cos

(A,

Z)

cos (B,

Z) +

cos (A,
cos

F) cos
cos

A,

(B,

F)

B^

between the direction cosines of the

A and

line

If

30.]

the line B.

A and B

are

two

OB

side

^isC = B-A (Fig. 17).

C C

= (B - A)
C*

or

That

sum

of a triangle

is,

(B

OAB,

sides

and

the third

- A) = B

+ A A - 2 A B cos

= A* + *-2Acos

(AB).

the square of one side of a triangle

of the squares of the other

two

(A, B) ,

is

equal to the
by twice

sides diminished

their product times the cosine of the angle

between them.

Or, the square of one side of a triangle is equal to the sum of


the squares of the other two sides diminished bj twice the
projection of either of those sides

upon the

other, the

theorem

sometimes known as the generalized Pythagorean theorem.


If A and B are two sides of a parallelogram, C = A +

= A B are the diagonals. Then


C.C = (A + B).(A + B)=A.A + 2A.B + BB,
DD = (A-B).(A-B)=A.A-2A-B + B.B,

and D

or

=2

VECTOR ANALYSIS

60

That

is,

gram

is

sum of

the squares of the diagonals of a paralleloequal to twice the sum of the squares of two sides.
the

In like manner also

C-C-D.D = 4 A-B
C*-D* = AB cos (A, B).

or

That

is,

sides

by the projection of the other upon

the difference of the squares of the diagonals of a


parallelogram is equal to four times the product of one of the
If

is

any vector expressed

it.

in terms of

i,

j,

k as

A = A i -f A 2 j + A B k,
A A = A 2 = A? + A* + A*.
l

then

But

if

vectors

(8)

be expressed in terms of any three non-coplanar unit


as

a, b, c

A 2 = a? +
This formula

is

ft

2 a b cos

a,

b)

b c

cos

(b, c)

a cos

(c,

a)

analogous to the one in Cartesian geometry

which gives the distance between two points referred to


If the points be x^ y v z v and x v yz z z the
oblique axes.
,

distance squared

D2

is

- zj) 2 + (y, - yi ) + (*, - Zi) 2


+ 2 O 2 - xj (2/ 2 - yj) cos (JT, F)
- 2/i) (a ~ *i) cos (^ ^)
+ 2
+ 2 2 -zj (x z - ajj) cos (^,X).
(a,

(2/2

The skew product of the vector A into


the vector quantity C whose direction is the
normal upon that side of the plane of A and B on which
31.]

Definition:

the vector

is

DIRECT AND SKEW PRODUCTS OF VECTORS

61

rotation from A to B through an angle of less than one


hundred and eighty degrees appears positive or counterclockwise and whose magnitude is obtained by multiplying
;

the product of the magnitudes of

angle from

The

A X B may

of

direction

which an

and B by the

sine of the

denned

as that in

to B.

also be

ordinary right-handed
it turns so as

screw advances as

AXB

toward B (Fig. 18).


The skew product is denoted by
a cross as the direct product was
to carry

by a

It is written

dot.

FIG. 18.

and read

B*

cross B.

For

= AxB

this reason it is often called the cross

product. More frequently, however, it is called the vector product, owing to the fact that it is a vector quantity and in contrast with the direct or scalar product whose value is scalar.
The vector product is by definition

when

where

and

c is a

=A

B are the

X B

= A B sin

(A, B)

c,

(9)

magnitudes of A and B respectively and


In case A and
C.

unit vector in the direction of

are unit vectors the

AXB

skew product

reduces to the

unit vector c multiplied by the sine of the angle from A to B.


Obviously also if either vector A or B is multiplied by a scalar

x their product

is

multiplied by that scalar.

(x A)
If

x B

=Ax

(zB)

= xC.

A and B are parallel the angle between them is either zero

or one hundred and eighty degrees. In either case the sine


vanishes and consequently the vector product A X B is a null
vector.

And

conversely

if

A X B is zero

A B sin

(A, B)

= 0.

VECTOR ANALYSIS

^62

Hence

or

or sin (A, B)

is

Thus

zero.

the condition

for

is A X B
parallelism of
As a corollary the vector product of any vector into
0.

two vectors neither

of which vanishes

itself vanishes.

The vector product

32.]

of

two vectors

will appear wher-

ever the sine of the included angle is of importance, just as


the scalar product did in the case of the cosine. The two products are in a certain sense complementary. They have been

denoted by the two common signs of multiplication, the dot


In vector analysis they occupy the place held
cross.

and the

by the trigonometric functions of scalar analysis. They are


at the same time amenable to algebraic treatment, as will be
seen later. At present a few uses of the vector product may
be cited.
If

and B

(Fig. 18) are the

two adjacent sides

of a parallel-

ogram the vector product


C

=A

x B

= A B sin

(A,

B)

represents the area of that parallelogram in magnitude


direction (Art. 25).
This geometric representation of A
is

of such

common occurrence and importance

that

it

and

x B

might

well be taken as the definition of the product. From it the


trigonometric definition follows at once. The vector product

appears in mechanics in connection with couples. If A and


A are two forces forming a couple, the moment of the

couple

any

is

A x B

point of

provided only that B


to

any

point of

A.

is

a vector drawn from

The product makes

its

appearance again in considering the velocities of the individual particles of a body which is rotating with an angular veIf B, be the
locity given in magnitude and direction by A.

drawn from any point of the axis of rotation A


the product A x E will give the velocity of the extremity of
radius vector

This velocity is perpendicular alike to the axis


(Art. 51).
of rotation and to the radius vector R.

DIRECT AND SKEW PRODUCTS OF VECTORS

63

The vector products A X B and B x A are not the


same. They are in fact the negatives of each other. For if
rotation from A to B appear positive on one side of the plane
33.]

of

and

from B

B, rotation

Hence A x B

to

A will

appear positive on the

the normal to the plane of A and B


upon that side opposite to the one upon which B x A is the
normal. The magnitudes of A X B and B X A are the same.
other.

Hence

is

AxB = -BxA.

The factors in a
if the sign

of

the

vector product

product

can

be

(10)

interchanged if

and only

be reversed.

instance in which the laws of operation in


vector analysis differ essentially from those of scalar analysis.
It may be that at first this change of sign which must

This

is

the

first

accompany the interchange of factors in a vector product will


give rise to some difficulty and confusion. Changes similar to
this are, however, very familiar. No one would think of interchanging the order of x and y in the expression sin (x
without prefixing the negative sign to the result. Thus
sin (y

although the sign

is

x)

if

sin (x

y),

not required for the case of the cosine,

cos (y

Again

y)

x)

= cos

the cyclic order of the letters

y~).

A B C in the

area of a

triangle be changed, the area will be changed in sign (Art.


25).

A B C = -A

CB.

In the same manner this reversal of sign, which occurs

when

the order of the factors in a vector product is reversed,


will appear after a little practice and acquaintance just as

natural and convenient as


34.]

The

distributive

it is

necessary.

law of multiplication holds in the

case of vector products just as in ordinary algebra

except

VECTOR ANALYSIS

64

order of the factors must

that the

carefully maintained

be

when expanding.

A very simple proof may be given by making use of the ideas


developed in Art. 26. Suppose that C
Let A
is not coplanar with A and B.

and B be two

sides of a triangle taken

(A + B)

Then

in order.

Form

third side (Fig. 19).

of

which

of

which C

the prism

this triangle is the base

and

the slant height or edge.


areas of the lateral faces of this

The

FIG. 19.

will be the

is

prism are

A X
The

C,

(A

B)

C.

areas of the bases are

s
ii

But the sum


is

prism

A x

B X

C,

(A X B) and

x~C

B).

Hence

(A

X C

B)

AxC + BxC
A x

or

The

(A x

of all the faces of the prism is zero; for the

a closed surface.

+B

+B x

\ (A X B)
(A + B) x C

= (A 4-

- (A
\
= 0,

B) x

X B)

C.

0,

(11)

therefore proved in case C is non-coplanar


Should C be coplanar with A and B, choose D,
any vector out of that plane. Then C + D also will lie out of

with

relation

and

that plane.

Ax

is

B.

Hence by

(C

D)

(11)

+B x

(C

D)

(A

B)

(C

D).

Since the three vectors in each set A, C, D, and B, C, D, and


A + B, C, D will be non-coplanar if D is properly chosen, the
products

may

be expanded.

DIRECT AND SKEW PRODUCTS OF VECTORS

65

AxC+AxD+BxC+BxD
= (A +

+ B) x D.
A^-D + BxD=(A + B)xD.
AxC + BxC (A + B)xC.

But by (11)
Hence

B) x C

(A

The

This completes the demonstration.


for a vector product.

The

(A + B +

<J

x (P +

distributive

generalization

is

law holds

immediate.

= AxP + Axft+---

(11)'

+B xP+BXQ+
The

vector products of the three unit vectors


seen
easily
by means of Art. 17 to be
35.]

ixi = jxj = kxk =


i

= k,
j x i
k x j = i,

ixk

xk

k x

i, j,

k are

0,

(12)

j.

The skew product of two equal 1 vectors of the system i, j, k


The product of two unequal vectors is the third taken

is zero.

with the positive sign if the vectors follow in the cyclic order
k but with the negative sign if they do not.
If two vectors A and B are expressed in terms of i, j, k,

i j

may be found by expanding according


law and substituting.

their vector product


to the distributive

vi4->4
-*X
1

7? iA
-i-'o

^.
,i

j4

T)

+ A3 B

Hence

A x B

1.

|^

i -4-

k x

-I

>4

v
/'X

+ A 3 BZ k x

= (^4 2 ^3

J!

-4-

K
'

j 4-

A 3 BJ) i +
+

This follows also from the fact that the sign

factors

is

reversed.

Hence

jXj =

X j = 0.

>4
7?
*-'
-^*-1
I
>4

-4 3 J?3

is

kx

lr

k.

A ^3 ) J
BZ -A,B^*.

(A 3 B^
(A 1

v k

O i* -^ *^
O

??

changed when the order of

MifliU

-*V*r*

-/
s

VECTOR ANALYSIS

66
This

may be

written in the form of a determinant as

A x B
The

sum

formulas for the sine and cosine of the

or dif-

ference of two angles follow immediately from the dot and


Let a and b be two unit vectors lying in the
cross products.
i j -plane.

If

x be the angle that a makes with


i, then

i,

and y the

angle b makes with

cos x

cos y

cos (a, b)

a-b
Hence

cos (y

x)
b'

If

a-b'
Hence

cos (y

x)

x b
a x b
a

Hence

sin (y

If

Z,

x)

j,

= cos
+
+

sin y

= k sin

(sin y cos x

x)

sin ^ cos

sin

(sin y cos x

x).

{y

#),

sin x cos y).

=k

sin (a, b')

(y

sin y sin x.

sin y cos x

m\n

j,

= cos

sin (a, b)

sin (y

x),

sin y sin x.

cos y cos x

k
k

(y

sin x sin y.

cos y

cos (a, b')

b'

j,"

x
x

m, w and l\

sin

cos y cos

b'

sin

cos x cos y

k
k

Hence

+
+

a;

cos y.

sin (y

sin x cos

#)
?/).

sin x cos y.

are the direction cosines of

unit vectors a and a' referred to

JL,

two

F, Z, then

w'k,
a

a' = cos

(a, a')

+ m m' +

as has already been shown in Art. 29.


for the square of the sine of the angle

be found.

The

familiar formula

between a and

a'

may

DIRECT AND SKEW PRODUCTS OF VECTORS


a

a'

= sin

(a,

(7m'

-f

where
(a
sin

m' n)

(run'

I'm)

a')

(a

a')

= sin

(a,a')

= (mn'-m'n) +
z

n' I)

(nl'

k,

e is a unit vector perpendicular to a

67

(a, a') e

(nl'-

n'

and a

f
.

= sin 2
2

Z)

(a, a').

+(lm'

I'

w)

2
.

This leads to an easy way of establishing the useful identity

(7

Products of More than

Two

Vectors

to this point nothing has been said concerning


products in which the number of vectors is greater than
36.]

two.

Up

If three vectors are

combined into a product the result

a triple product. Next to the simple products


and AxB the triple products are the most important.
All higher products may be reduced to them.
called

is

AB

The

simplest triple product is formed by multiplying the


scalar product of two vectors A and B into a third C as

(A-B)

C.

This in reality does not differ essentially from scalar multiplication (Art. 6).

The

scalar in this case merely happens to

be the scalar product of the two vectors A and B. Moreover


inasmuch as two vectors cannot stand side by side in the

form of a product as B C without either a dot or a cross to


unite them, the parenthesis in (A-B) C is superfluous. The
expression

cannot be interpreted in any other way


of the vector C by the scalar AB.

than as the product

Later (Chap. V.) the product BC, where no sign either dot or cross occurs,
But it will be seen there that (A.B) C and A(BC) are identical
and consequently no ambiguity can arise from the omission of the parenthesis.
i

will be defined.

VECTOR ANALYSIS

68
37.]

two

The second

vectors, of

triple

which one

the scalar product of


itself a vector product, as

product
is

is

A(BxC)or (AxB>C.
This sort of product has a scalar value and consequently

is

often called the scalar triple product. Its properties are perhaps most

deduced from

easily

its

commonest

geometrical interpretation. Let A, B,


and C be any three vectors drawn

from

same origin

the

Then BxC

B and C

allelogram of which
scalar

A-(BxC)
volume

will therefore be the

BxC

is

the base and

This volume v

is

positive

B C-plane

two adjacent

(14)

of the parallelepiped of

and

BxC

lie

which

See Art. 28.

upon the same

positive system of three vectors the scalar


tive; but
is

20).

The

sides.

the slant height or edge.


if

(Fig.

the area of the par-

but negative if they lie on opposite


In other words if A, B, C form a right-handed or

side of the
sides.

are

is

if

A(BxC)

is

posi-

they form a left-handed or negative system,

it

negative.

In case A, B, and C are coplanar this volume will be


neither positive nor negative but zero. And conversely if
the volume

is

piped must

lie

zero the three edges A, B, C of the parallelein one plane.


Hence the necessary and suffi-

for the coplanarity of three vectors A, B, C none


which
vanishes
is A(BxC) = 0.
As a corollary the scalar
of
cient condition

triple

product of three vectors of which two are equal or


must vanish for any two vectors are coplanar.

collinear

The two products A(BxC) and (AxB)C

are equal to the

same volume v of the parallelepiped whose concurrent edges


are A, B, C.
The sign of the volume is the same in both
cases.
Hence
(14)'
(AxB)-C = A.(BxC) = v.

DIRECT AND SKEW PRODUCTS Of VECTORS


This equality

and

may

the cross in

be stated as a rule of operation.

a scalar

triple product may


without altering the value of the product.
It may also be seen that the vectors A, B, C

muted

~be

69

The dot

interchanged

may

be per-

cyclicly without altering the product.

= B.(CxA) = C.(AxB).

A-(BxC)

(15)

For each of the expressions gives the volume of the same


parallelepiped and that volume will have in each case the

same

sign, because if

plane,

upon the

is

positive side of the

upon the

may

positive side of the

A B-plane.

The

C-

triple

product

therefore have any one of six equivalent forms

= B-(CxA) = i-(AxB)
= (AxB>C = (BxC).A = (CxA)-B.
A-(BxC)

If

on the positive side of the C A-plane and C

will be

however the

cyclic order of the letters

is

(15)'

changed the

product will change sign.

A-(BxC)
This

may

= - A<CxB).

be seen from the figure or from the fact that

BxC
Hence

(16)

CxB.

scalar triple product is not altered by interchanging

the dot or the cross or by

permuting

cyclicly the order

of the

in sign if the cyclic order be changed.


necessary upon the subject of parentheses

vectors, but it is reversed

38.]

in

this

biguity ?

word

triple

is

Can they be omitted without amThe expression

product.

They

can.

A-BxC
can have only the one interpretation

A<BxC).
For the expression (AB)xC is meaningless. It is impossible to form the skew product of a scalar A-B and a vector

VECTOR ANALYSIS

70

only one way in which ABxC may


be interpreted, no confusion can arise from omitting the
Furthermore owing to the fact that there are

Hence

C.

as there

is

parentheses.

and C which have the same


value and are consequently generally not worth distinguishit is often convenient to use the
ing the one from another,
six scalar triple products of A, B,

symbol

[ABC]
any one of the six equal products.

to denote

[A B C]
then

= A-BxC = B-CxA = C-AxB


= AxB.C = BxC-A =' CxA-B
- [A C B].
[A B C] =

(16)'

scalar triple products of the three unit vectors i, j, k


vanish except the two which contain the three different

The
all

vectors.

Hence

[ijk]=-[ikj] =

if

l.

(17)

three vectors A, B, C be expressed in terms of

i, j,

as

A=A

+ A z j + A 3 k,
B = B i + 2 j + #, k,
C=C i+ <72 j + C3 k,
= A l z C3 + Bl C2 A 3 + C, A t Bz
- S C3 A 2 - 0, A s Br
A, Bz <72
l

[A B C]

then

This
tions

be obtained by actually performing the multiplicawhich are indicated in the triple product. The result

may

may be

written in the form of a determinant. 1

A A A
--\ -^2 "*
[A B C]

(18)'

C1
1
This is the formula given in solid analytic geometry for the volume of a
tetrahedron one of whose vertices is at the origin. For a more general formula

see exercises.

DIRECT AND SKEW PRODUCTS OF VECTORS


If

more generally A,

non-coplanar vectors
vectors,

B, C are expressed in terms of

any three
which are not necessarily unit

a, b, c

A = &J

where a v

3 ;

2,

B=\a+
C =c a +

stants,

&2

c2

and

b v 62 , J3 ;

+ 3c
+ &3 c
+ c3 c

e?

x,

2,

are certain con-

then

[ABC]- (a

&2 c 3

ft,

c2

a2

[ABC]-

or

The

39.]

Cl

Co

[a

6 2 ) [a

c].

be]

(19)'

d.

third type of triple product

two vectors of which one

J3

a3

Cj

&1 &2 &


Ci

of

71

is itself

is

the vector product

a vector product.

Such

are

Ax(BxC) and (AxB)xC.

The vector Ax(BxC) is perpendicular


But (BxC) is perpendicular to the plane

Ax (BxC),
plane of

to

of

B and

and

being perpendicular to (BxC) must


take the form

Hence

lie

in the

B and C and thus

Ax(BxC)
where x and y are two

= x B + y C,

scalars.

In like manner also the

(AxB)xC, being perpendicular to (AxB) must


the plane of A and B. Hence it will be of the form

vector
in

to (BxC).
C.

(AxB)xC
where

and

v are

two

= mA + wB

scalars.

From

in general

(AxB)xC
The parentheses
changed.

It

is

is

not equal to

therefore
essential

to

lie

this it is evident that

Ax (BxC).

removed or interknow which cross product is

cannot be

VECTOR ANALYSIS

72

formed

first

and which second.

This product

is

termed the

vector triple product in contrast to the scalar triple product.


The vector triple product may be used to express that com-

ponent of a vector B which is perpendicular to a given vector


A. This geometric use of the product is valuable not only in
itself but for the light it sheds

AXB
AXB

upon the properties of the product.


Let A (Fig. 21) be a given vector
and B another vector whose components parallel and perpendicular

to

are

X
p

Let the
and per-

be found.

to

components of

parallel

B'and B" reDraw A and B from a


spectively.
common origin. The product AxB
The product
plane of A and B.
pendicular to

(AXB)
21

perpendicular to the
Ax (AxB) lies in the plane of
is

and

B.

be

It is

furthermore

Hence it is collinear with B". An


perpendicular to A.
examination of the figure will show that the direction of
Ax (AxB) is opposite to that of B". Hence

where

Now
but
if

c is

some

scalar constant.

Ax(AxB) =~A
- c B" = - c B

b" be a unit vector

B sin

(A,B) b"

sin (A, B) b",

in the direction of B".

Hence

Hence

B"

= A = AA.
z

Ax (AxB)
7-7

(20)

AfA

The component

of

B perpendicular

to

has been expressed

in terms of the vector triple product of A, A,

component B' parallel to

was found in Art. 28

and
to

B.

be

The

DIRECT AND SKEW PRODUCTS OF VECTORS

B'=?A

73

(21)

00
The vector triple product Ax (BxC) may be expressed
sum of two terms as

40.]

as the

AX (BXC) =

A.C

B-A-B

In the

first place consider the product


vectors are the same. By equation (22)

when two

= A-B A - Ax(AxB)
Ax(AxB) = A.B A - A- A B
A- A

or

of the

(22)
(23)

This proves the formula in case two vectors are the same.
To prove it in general express A in terms of the three
non-coplanar vectors B,

C,

and BxC.

A = 6B+cC +
where

a,

Z>,

are scalar constants.

Ax(BxC)

By

(23)

of

(I)

Then

= SBx(BxC) +
+

The vector product

(BxC),

cCx(BxC)

(II)

a (BxC)x(BxC).

any vector by

itself is zero.

Hence

(BxC)x(BxC) =
Ax(BxC) = JBx(BxC) + c Cx(BxC).
Bx(BxC) = B-C B - BB C

Cx(BxC) = - Cx(CxB) = - C.B C + CC B.


Hence Ax(BxC) = [(6B-C + cCC)B- (6B.B + cC-B)C].
But from (I) A-B = JB-B + cC-B + a (BxC)-B
and

By

= b B-C + c CC + a (BxC>C.
(BxC)-B = and (BxC>C = 0.
AB = JBB + cC-B,
A-C = &B-C + cC-C.

A-C
Art. 37

Hence

(II)'

(II)"

VECTOR ANALYSIS

74

Substituting these values in (II)",

Ax(BxC)

The

= A-C

A.B

C.

(24)

therefore proved for any three vectors A, B, C.


Another method of giving the demonstration is as follows.
relation

was shown that the vector

It

form

of the

Since
it

is

Ax (BxC) = #B +

Ax(AxC)

by

product

triple

is zero.

is

Ax (BxC) was

2/C.

perpendicular to A, the direct product of

Hence

= zA-B + 2/A-C =
x : y = AC
A-B.
=
Ax (BxC). n (A-C B AB C),

A.[Ax(BxC)]
and

Hence
where n

is

Multiply by

a scalar constant.

remains to show n

1.

B.

Ax(BxC>B
The

It

.'

= n (A-C

B-B

- A-B

C-B).

scalar triple product allows an interchange of dot and

cross.

Hence

Ax(BxC>B = A<BxC)xB = - A-[Bx(BxC)],


if

the order of the factors (BxC) and

B be

inverted.

-A.[Bx(BxC)] =-A.[B-C B-B-BC]


= -B.C A,B + B.B A.C.
Hence n

=1

and

Ax(BxC)

= A-C

- A^B

C.

(24)

From tne three letters A, B, C by


four allied products in each of which B and C are included in
parentheses may be formed. These are
different arrangements,

Ax(BxC),

As

Ax(CxB),

factors is

(BxC)xA.

sign whenever the order of


the
above products evidently
interchanged,

a vector product changes

two

(CxB)xA,
its

satisfy the equations

Ax (BxC) =-Ax(CxB) = (CxB)xA = -(BxC)xA.

DIRECT AND SKEW PRODUCTS OF VECTORS


The expansion

75

a vector triple product in which the


parenthesis comes first may therefore be obtained directly
from that already found when the parenthesis comes last
for

(AxB)xC

= - Cx(AxB) = - C.B A + CA

B.

The formulae then become

Ax (BxC) = A.C B - A.B


and

(AxB)xC

= A-C

(24)

C-B A.

(24)'

These reduction formulae are of such constant occurrence and


great importance that they should be committed to memory.
Their content may be stated in the following rule.
To expand

vector triple

product first multiply

the exterior factor into the

remoter term in the parenthesis to form a scalar coefficient for


the nearer one, then multiply the exterior factor into the nearer

term in the parenthesis


remoter one,
41.]

As

and

to

form a scalar coefficient for


from the first.

the

subtract this result

far as the practical applications of vector analysis

concerned, one can generally get along without any


formulae more complicated than that for the vector triple
are

product. But it is frequently more convenient to have at


hand other reduction formulae of which all may be derived

simply by making use of the expansion for the triple product


Ax (BxC) and of the rules of operation with the triple pro-

duct

ABxC.

To
is

reduce a scalar product of two vectors each of which


itself a vector product of two vectors, as

(AxB).(CxD).
Let

this

be regarded as a scalar triple product of the three


CxD thus

vectors A, B, and

AxB.(CxD).
Interchange the dot and the cross.

VECTOR ANALYSIS

76

AxB.(CxD)=A-Bx(CxD)

= B-D C - B-C D.
- A-D
(AxB).(CxD) = A-C B-D
Bx(CxD)

Hence
This

may

(25)

be written in determinantal form.

(AxB)-(CxD)
If

B.C.

A-C

A-D

B-C

B-D

and D be called the extremes

and C the antecedents; B and

B and C

(25)'

the means

the consequents in this

product according to the familiar usage in proportions, then


the expansion may be stated in words. The scalar product
of two vector products is equal to the (scalar) product of the
antecedents times the (scalar) product of the consequents
diminished by the (scalar) product of the means times the

product of the extremes.


reduce a vector product of two vectors each of which

(scalar)

To

is itself

a vector product of two vectors, as

(AxB)x(CxD).
Let

CxD

= E.

The product becomes

(AxB)xE
Substituting the value of

(AxB)x(CxD)
Let F

= AxB.

= A.E

E back

- B.E A.

into the equation

(A.CxD)B

(B-CxD)

By

A.

(26)

The product then becomes

Fx(CxD) = FD C-F.C D
(AxB)x(CxD) = (AxBD)C - (AxBC)
all

D.

(26)'

equating these two equivalent results and transposing

the terms to one side of the equation,

[B C D]

A-

[C

D A] B + [D A

B] C

[A B C] D

= 0.

(27)

an equation with scalar coefficients between the four


vectors A, B, C, D. There is in general only one such equaThis

is

DIRECT AND SKEW PRODUCTS OF VECTORS

77

tion, because any one of the vectors can be expressed in only


one way in terms of the other three thus the scalar coefficients of that equation which exists between four vectors are
:

found to be nothing but the four scalar


those vectors taken three at a time.

be written in the

[A B C D]

More examples

also

form

[B C D]

A+

[C

D] B

[A B D]

of reduction formulae, of

important, are given

among

In view of these

chapter.

triple products of

The equation may

C.

(27)'

which some are

the exercises at the end of the


it

becomes

fairly obvious that

any number of vectors connected in


any legitimate way by dots and crosses or the product of any
number of such combinations can be ultimately reduced to
the combination

of

sum of terms each of which contains only one cross at most.


The proof of this theorem depends solely upon analyzing the
a

and showing that they all


under the reduction formulae in such a way that the

possible combinations of vectors


fall

crosses

may

be removed two at a time until not more than

one remains.
*

The

formulae developed in the foregoing article have


interesting geometric interpretations.
They also afford a
simple means of deducing the formulae of Spherical Trigo42.]

nometry. These do not occur in the vector analysis proper.


Their place is taken by the two quadruple products,

= A.C B-D - B.C A.D


(AxB)x(CxD) = [ACD] B - [BCD] A
= [ABD] C - [ABC] D,
(AxB).(CxD)

and

(25)

(26)

which are now to be interpreted.


Let a unit sphere (Fig. 22) be given. Let the vectors
A, B, C, D be unit vectors drawn from a common origin, the
centre of the sphere, and terminating in the surface of the
sphere at the points A, B, C, D.

The great

circular arcs

VECTOR ANALYSIS

78

AB,

and

(7,

etc.,

give the angles between the vectors A and B,


determine a quadrilateral
points A, B, (7,

The

C, etc.

upon the sphere.

AC

and

BD are

one

A D and
the
A B and CD are the diagonals.
- A.D B.C
(AxB).(CxD) = A.C B.D
=
=
AxB sin (A,B), CxD sin(C, D).

pair of opposite sides

B C,

other.

The angle between AxB and CxD


FIG. 22.

the angle between the planes themselves.

(AxB). (CxD)

The

the

Let

it

be denoted

Then

x.

by

is

angle between the normals to the ABand CD-planes.


This is the same as

angles

(A, B), (C,

= sin

(A,B) sin (C,D) cos a;.

D) may be replaced by the great

AB, CD which measure them. Then


sin A B sin CD cos x,
(AxB).(CxD)
=
B.D A.D BC
cos AC cosBD
cos AD cos BC.

circular arcs

A.C

Hence
sin

A B sin CD cos x = cos A C cos B D

In words

The product

of the cosines of

cos

AD

cos

B C.

two opposite sides

of a spherical quadrilateral less the product of the cosines of

the other two opposite sides

is

equal to the product of the

of the

sines

diagonals multiplied
cosine of the angle between them.

theorem

is

by the
This

credited to Gauss.

C (Fig. 23) be a spherical trithe


sides
of which are arcs of great
angle,
circles.
Let the sides be denoted by a, &, c
Let A, B,

FIG. 23.

respectively.

drawn from the center


Furthermore

let

p^ p b p c
,

Let A, B, C be the unit vectors

of the sphere to the points A, B, C.

be the great circular arcs dropped

DIRECT AND SKEW PRODUCTS OF VECTORS


perpendicularly from the vertices A, B,
Interpret the formula

(AxB).(CxA)

= sin

(AxB)

Then

(A, B)

c,

is

the

sin b cos x,

AxB

between

angle

J, c.

<z,

B-A

sin

(AxB). (CxA)

where x

to the sides

- B-C A- A.
(Cx A) = sin (C, A) = sin b.

= A-C

= sin

79

and

CxA.

This

between the plane of A, B and the


angle
plane of C, A. It is, however, not the interior angle A which
is one of the angles of the triangle
but it is the exterior
equal to the angle

is

angle 180

as

A,

Hence

an examination of the figure will show.

= sin c sin b cos (180 A)


= sin c sin b cos A
BC A A = cos b cos c cos a 1.

(AxB)x(CxA)

AC BA
By

equating the results and transposing,


cos a

= cos b

cos

sin b sin

cos b =' cos c cos a

cos

The

last

letters or

= cos a

two may be obtained by


from the identities

(CxA)x(BxC)
cases in

(AxB)x(AxC)
BxC, CxA,

AxB

cyclic permutation of the

(AxB)x(CxD)

which one of the vectors

a, b, c

A
B

= B.A C.B - C.A,


= C-B A.C - B.C.

interpret the identity

Let the three vectors

cos

sin a sin b cos C.

cos b

(BxC)x(AxB)

Next

sin

sin a cos

is

in the special

repeated.

= [A B C] A.

be unit vectors in the direction of

respectively.

Then

AxB = c sin
AxC = b sin b
cxb sin c sin b
A sin c sin b sin A
(AxB)x(AxC) =
[A B C] = (AxB).C = c.C sin c = cos (90
p sin c
=
[ABC] A sin c sin p
c,

e)

VECTOR ANALYSIS

80

By

common

results and cancelling the


equating the

sin

The

last

letters.

triangle

pe

= sin b sin A
= sin c sin B
= sin a sin C.

two may be obtained by cyclic permutation of the


The formulae give the sines of the altitudes of the
Again
in terms of the sines of the angle and sides.

write

(AxB)x(AxC)
(BxC)x(BxA)
(CxA)x(CxB)
Hence

sin

= [ABC]A
= [BCA]B
= [C AB] C.
[A B C]

sin b sin A-

sin a sin c sin B =


=
sin b sin a sin

The

factor,

expressions

[B C A]
[C

A B].

[ABC], [BCA], [CAB]

are equal.

Equate

the results in pairs and the formulae

A = sin a sin B
sin c sin B = sin b sin C
sin a sin C *= sin c sin A

sin b sin

are obtained.

These may be written in a single


sin

A _ sin B _ sin C

sin a

The
obtain.

is

zero

for the

all

triangle

equation
a

is

sin

this

almost

sin b

formulae of Plane Trigonometry are even more easy to


If
B C be a triangle, the sum of the sides taken

as vectors

From

line.

is

a closed polygon.

the elementary formulae follow immediately.

to be hoticed that the


angles

from a to

b,

from b to

c,

It

from

DIRECT AND SKEW PRODUCTS OF VECTORS

81

a are not the interior angles A, B, C, but the exterior

c to

angles ISO

-A,

- B,

180

- C.

180

=b+c
+ c) = bb +

a
a.a
If a,

= (b +

c).(b

be the length of the sides

&, c

= J 2 -f c 2
b = c -f a 2
2
2
2
c = a + 6
a2
2

The

last

a, b, c. this

2 b

(7.

area of the triangle

to the first

letters.

is

-b

b sin

C=

b c sin

A = ^ c a sin B.

each of the last three equalities be divided by the product

| a

c,

the fundamental relation


sin

sin

a
is

A
cos 5

two are obtained in a manner similar

2
If

becomes

cos

2 a b cos

one or by cyclic permutation of the

The

ec + 2 bc.

obtained.

Another formula

the product

(bxc)-(bxc)

2 Area (b

sin

A) =

2 Area

sin

for the area

(c

a2

The problem
1

sin

Let

a, b, c

b sin

G)

B sin G

of expressing

three non-coplanar vectors

may be found from

(cxa)(axb)

a sin B} (a

Reciprocal Systems of Three Vectors.


43.]

Solution of Equations

any vector

may

r in terms of

be solved as follows.

VECTOR ANALYSIS

82

where

a,

constants to be determined.

are three scalar

6,

bx

Multiply by

c.
//

r-bxc

= a abxc +

&

b-bxc

cc-bxc

[rbc]=a[abo].

or

In like manner by multiplying the equation by


. a x b the coefficients & and c may be found.

X a and

= & [b c a]
= c [c a b]
[r a b]
[r c a]

Hence

c]
[r a]
,[r b] c.
b+
= -4.
a +
^-\
-^ a b
r^
J
[c
aj
[be
be]
[a

The denominators
equation
[a

c] r

are

all

Hence

equal.

- [b c r] a +

[c r a]

[r

which must exist between the four vectors

The equation may

rb
r

=r

gives the

this

a b]

r, a, b, c.

also be written

--

raxb

rcxa

[abc]
or

(28)

[abc]

bxc

[abc]

[a

cxa,b

+r

be]

axb
c.

[abc]

[abc]

_ __

The three vectors which appear here multiplied by r, namely

bxc

axb

cxa

[a

[a b c]

be]

[a b c]

are very important.

They are perpendicular respectively to


the planes of b and c, c and a, a and b.
They occur over and
over again in a large number of important relations. For
this reason

they merit a distinctive

Definition

The system
b x

[abc]'

name and

of three vectors

cxa

axb

[abc]'

[abc]

notation.

DIRECT AND SKEW PRODUCTS OF VECTORS

83

which are found by dividing the three vector products b x


c

a,

x b

product

of three non-coplanar vectors

a b

is

a, b, c

called the reciprocal system to

c,

by the scalar
a, b, c.

The word non-coplanar

If a, b, c were cois important.


the
scalar
b
would
vanish and
triple product [a
planar
c]
fractions
consequently the

bxc

cxa

axb

[a b c]

[a b c]

>

[a be]

would all become meaningless. Three coplanar vectors have


no reciprocal system. This must be carefully remembered.
Hereafter

when

the term reciprocal system

understood that the three vectors

The system

of three vectors

will be denoted

by primes

bxc

'

[a

The expression

used,

it

a,

b, c

as a', b', c'.

[a b

'

= axb

= r.a'a +

(29)

[a b c]

c]

for r reduces then to the very simple


r

will be

not coplanar.

reciprocal to system

= cxa

be]

is

are

b, c

a,

r.b'b

r.c'c.

form
(30)

may be expressed in terms of the reciprocal


c' instead of in terms of a, b, c.
In the first

Tl^e vector r

system
place

a', b',

it

is

a', b', c'


c

and

a,

necessary to note that

if a, b, c

are non-coplanar,

which are the normals to the planes of b and c,


a and b must also be non-coplanar. Hence r may

be expressed in terms of them by means of proper scalar


coefficients #, y,

z.

= x a' + 7/b' + z c'


[abc]r = zbxc + 2/cxa + zaxb.
r

or

Multiply successively by

c.

This gives

ra
rb
[a be]

= x [b c a],

rc

= z [a b c],

[a b c]

[a b c]

Hence

b,

^a,

[c

ab],

= raa' + rbb' +

ra

y = rb
z = rc

rc

c'.

(31)

VECTOR ANALYSIS

84

be the system reciprocal to a, b, c the


any vector of the reciprocal system into the

If a', b', c'

44.]

scalar product of

the given

corresponding vector of
the product of

system

is

unity;

two non-corresponding vectors is zero.

= b'-b = c'.c = l
a'.b = a'.c = b'.a = b'-c = c'a = c'-b =
a'.a

That

is

(32)
0.

be seen most easily by expressing a', b',


terms of themselves according to the formula (31)
This

but

may

c'

in

= ra a' + rb b' + rc c'.


a' = a'aa' + a'-bb' + a'"c c'
b' = b'.aa' + b'-bb' + b'-cc'
o' = c'aa' + c'-bb' + c'.cc'.
r

Hence

Since

non-coplanar the corresponding coefficients on the two sides of each of these three equations must
b', c'

a',

are

Hence from the

be equal.

= a'.a
= b'a

l=b'-b

O^c'.a

= c'.b

From

the second

From

the third

This

proves

the

first

= a'.b

relations.

= a' c.
b'.c.

= c'c.

They may

also

be

proved

directly from the definitions of a', b', c'.

a'a =
,

bxc
a

b x

[abc]
a

'

bxc
b =

[abc
.

=-[be a]- = 1

[a

be]

[abc]

=0
[abc]

forth.

Conversely

and

= bxc.b =

[abc]
and so

a, b, c,

if

two

sets of three vectors each,

satisfy the relations

Aa = B'b = Cc = 1
A.a = A.c = Ba = B.C = C-a = (M> =

say A, B,

C,

DIRECT AND SKEW PRODUCTS OF VECTORS


then the set A, B, C

By

is

the system reciprocal to

85

a, b, c.

reasoning similar to that before

A=
C

A-a

a'

+ Ab

= Ca a' + Cb

+ Ac

b'

b'

C-c

c'

c'.

Substituting in these equations the given relations the result

is

A=

a',

= b',

= c'.

Hence
Theorem

The necessary and

set of vectors

a', b', c'

sufficient conditions that the

be the reciprocals of

a, b, c is

they satisfy the equations

= b'.b = c'-c = l
= a'.c = b'.a = b'.c = c'.a = c'.b = 0.
a'.a

a'.b

that

(32)

As

these equations are perfectly symmetrical with respect


to a', b', c' and a, b, c it is evident that the system a, b, c may

be looked upon as the reciprocal of the system a', b', c' just
may be regarded as the reciprocal of

as the system a', b', c'


a, b, c.

That

is

Theorem: If
then

a, b, c will

b'
"

to say,

a', b', c'

be the reciprocal system of

be the reciprocal system of

a', b', c'.

a'

c'

c'

a'

[a'bV]'

"[a'b'c']'

b'

a, b, c,

(29)'

~[a'b'c']'

These relations may be demonstrated directly from


definitions of a', b', c'.

The demonstration

is

the.

straightfor-

ward, but rather long and tedious as it depends on complicated reduction formulae. The proof given above is as short
as could be desired.
The relations between a', b', c' and

symmetrical and hence if a', b', c' is the reciprocal


system of a, b, c, then a, b, c must be the reciprocal system of
a, b,

c are

a', b', c'.

VECTOR ANALYSIS

86
45.]

Theorem

If a

'

the scalar triple products

That

reciprocals.

'

and

'

be reciprocal systems
and [a b c] are numerical

a, b, c

[a'bV]

is

[a'bV] [abc]=l
,

,,

b x

~L[abc]

=
[abc]

Hence

Hence

[a'bV]

By means
is

[abc]

[abc]J

[bxc cxa axb].

(bxc) x (cxa)
[bxc cxa axb]

axbl

= (bxc)x(cxa>(axb).

[bxc cxa axb]

But

(33;

ex

[abc]c.

caxb

[abc]

[abc]

[abc]

2
.

(33y

[abc]

[abc]

between

of this relation

[a' b' c']

and

[a b c]

it

possible to prove an important reduction formula,

(P.QxE)(A-BxC)

P.A

P.B

Q.A
E.A

Q.B

Q.C

E.B

E-C

P-C
(34)

which replaces the two scalar triple products by a sum of


nine terms each of which is the product of three direct products.

Thus the two

products are removed.


expressed as

But

To

which occur in the two scalar

give the proof let P, Q,

= P.A

A'

E = E.A

A'

+ E.B

Then

crosses

FPQE1=

P.B B'
B'

p.c c

E.C

P.A

P.B

p.c

Q-A

Q.B

Q-C

E.A

E.B

E.C

[A'B'C'J

[ABC]

c'.

[A'B'C'].

be

DIRECT AND SKEW PRODUCTS OF VECTORS


Hence

[PftR] [ABC]

The system of

P.A

P.B

P.C

Q.A
S.A

Q.B

Q.C

B-B

R.C

three unit vectors

i, j,

--- --

is its

own

87

reciprocal

system.

.,_jxk _i_. ,,_kxi

k ,_ixj_

J'

For this reason the primes i', j', k' are not needed to denote
a system of vectors reciprocal to i, j, k. The primes will
therefore be used in the future to denote another set of rect-

',
F', Z are used to denote a
angular axes i, j, k just as
set of axes different from JT, Z, Z.
TJie only systems of three vectors which are their own reciprocals
,

and left-handed systems of three unit


k.
the system i, j, k and the system i, j,
Let A, B, C be a set of vectors which is its own reciprocal.

are

the right-handed

That

vectors.

is

Then by (32)
A.A
Hence the vectors

= B.B = c-c = i.

are all unit vectors.

A.B

Hence

is

perpendicular to

B.A

Hence B

is

= A.C =
B and

o.

C.

= B-C = o.

perpendicular to

A and

C.

=C-B = O.
to A and B.

C.A

Hence C is perpendicular
Hence A, B, C must be a system
*
46.]

A scalar equation

like

i,

j,

k or

like

i,j,

k.

of the first degree in a vector r

is

an equation in each term of which r occurs not more than


once.
The value of each term must be scalar. As an example of such

an equation the following may be given.

a a-bxr

&(cxd).(exr)

+ c fr

-f

= 0,

VECTOR ANALYSIS

88

where

are

a, b, c, d, e, f

known

vectors

and

a,

Z>,

c,

of the
Obviously any scalar equation
form
to
the
reduced
be
r
vector
an unknown
may

first

scalars.

r-A

where

complish

is

known

vector

d,

known

degree in

=a

and

a,

known

this result in the case of the

scalar.

To

ac-

given equation proceed

as follows.

a axbr
{a

+ b (cxd)xe-r + c fr + d
+ b (cxd)xe + c f}r =

axb

d.

be nec'essary to make use


of various reduction formulae before the equation can be made

In more complicated forms


to take the desired form,

As

may

it

rA =

a.

a vector has three degrees of freedom it is clear that one


Three
is insufficient to determine a vector.

scalar equation

scalar equations are necessary.

The geometric

interpretation of the equa-

tion

r-A

=a

(36)

'A

Let r be a variable vector


drawn
from a fixed origin. Let
(Fig. 24)
A be a fixed vector drawn from the same

is

origin.

interesting.

The equation then becomes


r

or

if

cos

(r,

r cos (r,A)
r be the magnitude of r

and

r cos

A)

= a,

=.
A

A
(r,

that of A.

The

expression

A)

the projection of r upon A.


The equation therefore states
that the projection of r upon a certain fixed vector A must
is

DIRECT AND SKEW PRODUCTS OF VECTORS

89

always be constant and equal to a!A. Consequently the terof r must trace out a plane perpendicular to the vector

minus

equal to a/A from the origin. The projecof any radius vector drawn from the origin to a

at a distance

tion

upon

point of this plane is constant and equal to a/A.


the following theorem.

Theorem

scalar equation in

This gives

an unknown vector may be


which is the locus of the

regarded as the equation of a plane,

unknown

terminus of the

vector

if its

origin be fixed.

It is easy to see why three scalar equations in an unknown


vector determine the vector completely. Each equation determines a plane in which the terminus of r must lie. The

common

three planes intersect in one


tor r

is

determined.

equations

is

The

point.

Hence one

vec-

analytic solution of three scalar


If the equations are

extremely easy.

rA

=I
rC =
r-B

(37)

c,

it is

only necessary to call to mind the formula

Hence

The

solution

is

= a A' +

6 B'

c C'.

therefore accomplished.

(38)
It is expressed in

which is the reciprocal system to A, B, C. One


be observed. The vectors A, B, C will
however
must
caution
terms

A', B', C'

have no reciprocal system


In this
solution will fail.

if

they are coplanar.

Hence the

however, the three planes dethree


the
termined by
equations will be parallel to a line.
They will therefore either not intersect (as in the case of the
case,

lateral faces of a triangular prism) or

common

line.

Hence there

they will intersect in a

will be either

there will be an infinite number.

no solution

for r or

VECTOR ANALYSIS

90

From

four scalar equations

rA = a

=b
rC = c

r.B

(39)

r.D=d
the vector r

may be

To accomplish

entirely eliminated.

this

solve three of the equations and substitute the value in the


fourth.

a [BCD]

or
*

47.]

[CAD] +c [ABD]

&

A vector equation

=d

[ABC].

of the first degree in an

(40)

unknown

an equation each term of which is a vector quantity


containing the unknown vector not more than once. Such
vector

is

an equation

is

(AxB)x(Cxr)
where A,

and

r the

B, C, D, E,

unknown

eral be solved for

+ F =0,

known vectors, n a known scalar,


One such equation may in gen-

are

vector.

That

one vector equation is in


to determine the unknown vector which is
r.

general sufficient
contained in it to the

The method

+ D ET + n

first

is

to say,

degree.

of solving a vector equation is to multiply it

with a dot successively by three


arbitrary
vectors.

may
first

three scalar equations are obtained.


be solved by the methods of the
foregoing article.

These
In the

place let the equation be

A
where A,

a*r

+B

B, C, D, a, b, c are

b-r

C cr

known

= D,

vectors.

cients are written in the terms, for


they
the vectors.
Multiply the equation
It is

known non-coplanar

Thus

may

No

scalar coeffi-

be incorporated in

successively by A', B',


understood of course that A, B, C are non-coplanar.

C'.

DIRECT AND SKEW PRODUCTS OF VECTORS

91

= D-A
br = D-B'

a.r

= a' ar + b' br + c' c-r.


r = D-A' a' + D-B' b' + D-C'

But

Hence

The

solution

therefore accomplished in case A, B, C are nonThe special cases in


a, b, c also non-coplanar.

is

coplanar and

which either

c'.

of these sets of three vectors

is

coplanar will not

be discussed here.

The most general vector equation of the first degree


unknown vector r contains terms of the types

A
That

is it

ar.

r,

D.

Exr,

contain terms

will

in an

which consist of a known

vector multiplied by the scalar product of another known vecterms which are scalar multitor and the unknown vector
;

unknown vector; terms which are the vector


ples
product of a known and the unknown vector and constant
The terms of the type A ar may always be reduced
terms.
of the

number.

to three in

multiplied into r

coplanar vectors.

may

For the vectors


all

Hence

a,

b,

c,

which are

be expressed in terms of three nonall the products ar, b-r, cr,

expressed in terms of three. The sum of all terms of


the type A ar therefore reduces to an expression of three

may be

terms, as

A ar + B
The terms

of the types

b-r

C cr.

and Exr may

also be expressed

in this form.

Exr
Adding
to the

all

= wa'ar +
==

Exa' ar

n b'br

+n

c'cr

+ Exb' b-r+Exc' cr.

these terms together the whole equation reduces

form

L ar +

M br + N cr = K.

VECTOR ANALYSIS

92

This has already been solved as


r

The

solution

is

= K.L' a' +

K-N'

c'.

in terms of three non-coplanar vectors a', b',

These form the system reciprocal to


the products containing the
*

K.M' b'

a,

unknown

b, c in terms of

c'.

which

vector r were expressed.

SUNDRY APPLICATIONS OF PRODUCTS


to

Applications

Mechanics

In the mechanics of a rigid body a force is not a


See Art. 3.

48.]

vector in the sense understood in this book.

magnitude and direction but it has also a line


of application.
Two forces which are alike in magnitude
and direction, but which lie upon different lines in the body
force has

do not produce the same

Nevertheless vectors are

effect.

sufficiently like forces to be useful in treating

If a

number

.same point

them.

act on a body at the


of the forces added as vectors is called

of forces f v f 2 , f g ,

0, the

sum

the resultant R.

R=f +

fa

In the same

way

if f , f
x

the term resultant

added just as

if

2,

...

applied to the

sum

of these forces

they were vectors.

R = f1 +
The

f,

do not act at the same point

f3

is still

fa

...

(41)

idea of the resultant therefore does not introduce the

As far as the resultant


a force does not differ from a vector.

line of action of a force.

is

concerned

The moment of a force f about the point is


Definition :
equal to the product of the force by the perpendicular distance from
to the line of action of the force.
The moment
however

is

nitude

as defined above.

is

best looked

upon

as a vector quantity.

Its direction is
usually

Its

mag-

taken to

DIRECT AND SKEW PRODUCTS OF VECTORS

93

be the normal on that side of the plane passed through the

and the

point

line f

upon which the force appears

to pro-

in the positive
duce a tendency to rotation about the point
Another
method
direction.
of
defining the
trigonometric
is as follows
moment of a force f = P Q about the point
:

The moment

of the force f

= PQ

to twice the area of the triangle

about the point


is equal
P Q. This includes at once

both the magnitude and direction of the moment (Art. 25).


The point P is supposed to be the origin and the point Q,
;

the terminus of the arrow which represents the force f. The


letter
will be used to denote the moment.
subscript will
be attached to designate the point about which the moment is

taken.

Mo

{t}

The moment of a number


sum of the moments of the

of forces f x , f 2 ,

= -Pi0i. ** = ?*Qi'~
- 2 (0 P! Q + OPZ
Mo {f n f a
}

If

the (vector)

*i

This
f

is

individual forces.

known
"

is

'

*2'

49.]

as the total or resultant

moment

..).

of the forces

'

If f

be a force acting on a body and

drawn from the point


the force, the

to

moment

dxf

d be the vector

line of action of
is

the

Mo
For

any point in the

if

of the force about the point

vector product of d into f

if e

Qz

W - dxf

= d/sin

(d,f)

(42)
e,

be a unit vector in the direction of dxf.

dxf

Now

sin (d,

f)

The magnitude

is

= d sin

(d,

f)/e.

the perpendicular distance from

dxf

to

f.

accordingly equal to this perpendicular distance multiplied by /, the magnitude of the force.
of

is

VECTOR ANALYSIS

94

The direction
the magnitude of the moment MO {f}.
moment.
Hence
the
of
direction
the
as
same
of dxf is the
This

is

is

the relation

proved.

Mo
The sum

= dxf.

{f}

moments about

of the

number

of a

of forces

moment

is equal to the
f r f 2,
acting at the same point
of the forces acting at that point.
of the resultant

d be the vector from

to P.

Mo
Mo

Mo

{fJ

+ Ko

{f a }

---

{f !>
{f a |

= dxf
- dxf 2
l

= dxf +

dxf a

moment about 0'

total

let

Then

of

...

H----

The

For

(43)

any number of forces

f x, f 2,

acting on a rigid body is equal to the total moment of those


forces about
increased by the moment about 0' of the
resultant

Mo-

Ro

considered as acting at 0.

{f !, f a

Let d 15 d 2

,} = Mo

{f p f jp

be vectors drawn from

f
Let d/, d 2
j,
2
respectively.
from 0' to the same points in f v f 2
be the vector from
to 0'.
Then

',

d1

Mo

{f !, f 2

#!,

'

+M

f2,

\.

'

any point in
be the vectors drawn
Let c
respectively.

'

-\

But
is

the

(44)

to

= d + c,
d 2 = d 2 + c,
= d xf + d 2 xf +
}
= d/xf + d 2 'xf 2 +
= (d - c)xf + (d 2 - c)xf 2 +
= djXf i+daxf 2 + ---- cx(f + f 2 +
.

(Ho

'

c is the vector

moment about

drawn from 0'

to 0.

Hence

f,

0' of a force equal in magnitude and


parallel hi direction to f 1 but situated at 0.
Hence

DIRECT AND SKEW PRODUCTS OF VECTORS


-cxCfj+fjj

Hence

{f 1? f a

95

+ ...)=- cxR = M {Ro}= M {f f 2 ,} + M {Ro |.


}
'

'

15

The theorem is therefore proved.


The resultant R is of course the same at all points. The
is attached
subscript
merely to show at what point it is
to
act
when
the
moment about 0' is taken. For
supposed
the point of application of R affects the value of that moment.
The scalar product of the total moment and the resultant
is

the same no matter about

what point the moment be taken.

In other words the product of the total moment, the resultant, and the cosine of the angle between them is invariant
for all points of space.

B-M
where 0' and

'{fi,f a

}=

Mo {fi,*2-"}

any two points in space. This important


from the equation

are

relation follows immediately

Me?

{fj, f a

But

moment

the

{fj

}=. Mo

ForE-Mo^f^fa,

the point

= Mo

of

is

is

about

A point
it

is

parallelism

0'

may

parallel

|f i

fa

{Ro}.

'

{R

}-

no matter what

The
of

variation in the

total

the point about which the

perpendicular to the resultant.


be found such that the total moment

to the

The condition

resultant.

is

'

'

=0

RxM

+M
+R

Hence

proved.

moment due to a variation


moment is taken is always
50.]

perpendicular to

of application be.

relation

.
,

{f^f-j,

R.Mo' IRo}

and the

f2

= RxM
+RX M

{ lf f a
'

|B

for

VECTOR ANALYSIS

96

where

by

its

random. Replace
any point chosen at
write the f v f 2
to
omit
for
value and
brevity

braces

is

Then

{ }.

= ExM

ExM(y
The problem

is

Ex(cxE)

to solve this equation for

ExM
Now E

'{E O }
in the

R'B. c

0.

c.

+ Ec E =

0.

quantity. MO is also supposed to be


Let c be chosen in the plane through
perpen-

known.

known

is

=
ExM

and the equation reduces

Then E-c

dicular to E.

to

BB c

E-E
chosen equal to this vector the total moment about
at a vector distance from
the point
', which is
equal to c,

If c be

Moreover, since the scalar product of


is constant and since the

will be parallel to E.

the total

moment and

resultant itself

is

the resultant

constant

it is

clear that in the case

they are parallel the numerical value of the total


will be a minimum.

The

total

about which

For
If c

about 0'

(Jf

is

is

moment
it is

'

is

moment

unchanged by displacing the point

taken in the direction of the resultant.

|fi,f a ,

}=M

parallel to E,

cxE

{f 1

,f a ,

parallel to the resultant

- cxB.

vanishes and the

Hence

equal to that about 0.

find not merely one point 0'


is

where

it is

moment

possible to

moment
moment about any

about which the total


but the total

point in the line drawn through


parallel to E is parallel
to E.
Furthermore the solution found in equation for c is
the only one which exists in the
plane perpendicular to E
unless the resultant E vanishes.
The results that have been
'

obtained

may be summed up

as follows

DIRECT AND SKEW PRODUCTS OF VECTORS

97

whose resultant is not


any system of forces f 1? f 2
zero act upon a rigid body, then there exists in space one
and only one line such that the total moment about any
If

point of

it is

The

parallel to the resultant.

of

it is

same and

the

This line

parallel to the resultant.

is

total

numerically

moment about
less

itself

is

all

points

than that about any

other point in space.

equivalent to the one which states that


upon a rigid body is equivalent
any system
to a single force (the resultant) acting in a definite line and
a couple of which the plane is perpendicular to the resultant

This theorem

is

of forces acting

and

may

which the moment is a minimum. A system of forces


be reduced to a single force (the resultant) acting at any

of

desired point

of space

and a couple the moment

(regarded as a vector quantity)

about

of the forces acting

is

equal to the total

on the body.

But

of

which

moment

in general the

plane of this couple will not be perpendicular to the resultant, nor will its moment be a minimum.

Those who would pursue the study of systems of forces


acting on a rigid body further and more thoroughly may
consult the Traite de Mechanique Rationelle 1 by P. APPELL.
The first chapter of the first volume is entirely devoted to
the discussion of systems of forces. Appell defines a vector
as a quantity possessing magnitude, direction, and point of
application.

His vectors are consequently not the same as

those used in this book.


carried through in the

however may be
analysis.

The treatment

of his vectors

Cartesian coordinates.

Each

is

step

easily converted into the notation of vector

number

of

exercises

is

given at the close of

the chapter.

Suppose a body be rotating about an axis with a con51.]


stant angular velocity a.
The points in the body describe
circles concentric with the axes in planes perpendicular to
1

Paris, Gauthier-Villars et Fils, 1893.

VECTOR ANALYSIS

98

in its circle is equal


velocity of any point
and
the radius of the
the
of
to the product
angular velocity
the
to
It is therefore equal
product of the angular
circle.

the axis.

The

and the perpendicular distance from the point to the axis.

velocity

The

direction

of

the velocity

to the

axis

is

and to

perpendicular
the radius of the circle described

by the

point.

Let a (Fig. 25) be a vector drawn


along the axis of rotation in that
direction in which a right-handed

screw would advance

if

turned in

the direction in which the body is


Let the magnitude of a
rotating.

FIG. 25.

The vector a may be taken to


the angular velocity.
Let r be a radius vector
represent the rotation of the body.
drawn from any point of the axis of rotation to a point in the

be

a,

The

body.

vector product

axr

= a r sin (a, r)

equal in magnitude and direction to the velocity v of the


terminus of r. For its direction is perpendicular to a and r
and its magnitude is the product of a and the perpendicular

is

distance r sin

(a, r)

from the point to the line

v
If the

a 2< a
*3a
case

of

a.

= axr.

That

is

(45)

body be rotating simultaneously about several axes


which pass through the same point as in the
the gyroscope

the

rotations are
v.

velocities

due

to

the various

DIRECT AND SKEW PRODUCTS OF VECTORS


where

99

drawn from points


same point of the body. Let
be drawn from the common point of

are the radii vectores

r x, r 2 , r g,

on the axis a 1} a 2 a 3
,

to the

the vectors r 1? r 2 , r 3 ,

Then

intersection of the axes.


rl

= r2 = r3 =

v3

=r

and
v

= vx +

v2

= a xr + a xr + a
= (a + a 2 + a +
2

This shows that the body moves as


angular velocity which

is

velocities a 15 a 2 , a 3 ,

This theorem

the vector

if

3
.

xr

-)xr.

rotating with the


of the angular

sum
is

sometimes known

law of angular velocities.


It will be shown later (Art.) 60 that the motion of any
rigid body one point of which is fixed is at each instant of
time a rotation about some axis drawn through that point.
as the parallelogram

This axis
axis

is

position.

fixed

is

called the instantaneous axis of rotation.

not the same for

is

The

but constantly changes its


The motion of a rigid body one point of which is
all time,

therefore represented

by

= axr

(45)

the instantaneous angular velocity; and r, the


radius vector drawn from the fixed point to any point of the

where a

is

body.

The most general motion

body no point of which


is fixed may be treated as follows.
Choose an arbitrary
0.
At
this
instant
will
have a velocity v
point
any
point
Eelative to the point
the body will have a motion of rotation
about some axis drawn through 0. Hence the velocity v of
of a rigid

any point of the body may be represented by the sum of


and axr the velocity of that point
v the velocity of
relative to 0.

=v +

axr.

(46)

VECTOR ANALYSIS

100
In case v

is

body moves around a and


precisely the motion of a

parallel to a, the

This is
along a simultaneously.
In case v is perpendicular to a, it
a.
screw advancing along
such that
is possible to find a point, given by the vector r,
its

velocity

is

That

zero.

is

axr
This

may

Multiply by xa.

be done as follows.

v xa

(axr)xa

ar a

be chosen perpendicular to

a.

aa

or

Let

aa r =
~
-

v xa.

Then ar

is

zero

and

v x a
v x a
a*a

The

point

r,

a,

thus determined, has the property that its velocbe drawn through this point parallel to

If a line

ity is zero.

the motion of the

about

this

new

In case v

is

one of instantaneous rotation

is

body

axis.

neither parallel nor perpendicular to a

it

may

be resolved into two components

= V*o

Vv o

'

4Vv o
i

"

which are respectively parallel and perpendicular to


v = v

point

'

4-

may now be found such


v

"

"

a.

+ axr

that

axr.

Let the different points of the body referred to this point be


denoted by

r'.

Then

the equation becomes

=v +
'

axr'.

(46)'

The motion here expressed consists of rotation about an axis


a and translation
along that axis. It is therefore seen that
the most general motion of a
is at
instant
rigid

body

any

DIRECT AND SKEW PRODUCTS OF VECTORS

101

the motion of a screw advancing at a certain rate along a


The axis of the screw and its rate
definite axis a in space.
of advancing per unit of rotation

(i, e. its

pitch) change from

instant to instant.

The

52.]

of

principle

methods.

conditions for equilibrium as obtained by the


virtual velocities may be treated by vector

Suppose any system of forces

f x , f 2,

act

on a

body be displaced through a vector disrigid body.


tance D whether this distance be finite or infinitesimal the
If the

work done by the

forces is
...
D-fj, D.fa ,

The

total

work done

therefore

is

W = D.f
If the

the

D.f 2

body be in equilibrium under the action of the forces

work done must be

W = D-f

D-f 2

zero.
.

= D(f +

f2

)= D.E = 0.

The work done by

the forces is equal to the work done by


This must be zero for every displacement

their resultant.

D.

The equation
D.R

holds for

all

=O

Hence

vectors D.

The total resultant must be zero if the body be in equilibrium.


The work done by a force f when the rigid body is displaced by a rotation of angular velocity a for an infinitesimal
time t is approximately

adxf
where d

is

a vector

drawn from any point

tion a to any point of

components

f ', f

t,

To prove

f.

"

parallel

a-dxf

this

of the axis of rota-

break up

f into

and perpendicular respectively

= a-dxf +
'

a-dxf ".

two
to

a.

VECTOR ANALYSIS

102

As

'

is

scalar
parallel to a the

adxf

On

the other hand the

done by

product [a d

']

vanishes.

= a.dxf ".

work done by

"

For

during the displacement.

equal to the work

is

'

being parallel to

If h be the common
perpendicular to its line of action.
a
to
the
force t", the work
the
line
from
vector perpendicular
"
a for time
a
of

is

done by
t is

rotation

during

angular velocity

approximately

W=hf" at = *hxt"t.
The vector d drawn from any point of a to any point of f may
be broken up into three components of which one is h, another
In the scalar
is parallel to a, and the third is parallel to f ".
only that component of d which
"
perpendicular alike to a and f has any effect. Hence
triple

product [a d

f "]

W= ahxf"

= a.dxf f = adxf

is

t.

act be disbody upon which the forces t lt f 2


placed by an angular velocity a for an infinitesimal time t
and if d 15 d 2
be the vectors drawn from any point
of
If a rigid

a to any points of f 1?
by the forces f v f 2 ,

2,

respectively, then the

work done

will be
approximately

W= (a^xfj + a.d

xf 2

= a-CdjXfj + d 2 xf 2 +
= a.M {fj.fj,, }

+
..-)

)*
t

t.

If the

body be

work must be

in equilibrium this

Hence

a .M

|flf f2 ,

.} t

zero.

= 0.

The scalar product of the


angular velocity a and the total
moment of the forces tv f
about any point
must be
a
.

zero.

As

may

be any vector whatsoever the

must vanish.

Mo

{t v t v

.}

= 0.

moment

itself

DIRECT AND SKEW PRODUCTS OF VECTORS

103

The necessary

conditions that a rigid body be in equilibrium under the action of a system of forces is that the result-

ant of those forces and the total

moment about any

point in

space shall vanish.


if

Conversely

the resultant of a system of forces

and the

moment

of those forces about any one particular point in space


vanish simultaneously, the body will be in equilibrium.
= 0, then for any displacement of translation D
If

D.R

and the

total

work done

is

= 0.

zero,

when

the body suffers any

displacement of translation.

Let

MO

{fp

f 2,

be zero for a given point 0.

Then

for

any other point 0'


MO-

{fi? f 2 >

But by hypothesis

'I

is

= MO

f: , f 2 ,

also zero.

MO' (fp

Hence

aMo'

MO* {BO}.

Hence

f2

0.

{fp f a

=Q

any vector whatsoever. But this expression is


equal to the work done by the forces when the body is rotated
for a time t with an angular velocity a about the line a

where a

is

passing through the point 0'. This work is zero.


Any displacement of a rigid body may be regarded as a
translation through a distance D combined with a rotation
for a time t with angular velocity a about a suitable line a in
space.

It has

been proved that the

forces during this displacement

must be

in equilibrium.

is

total

zero.

The theorem

is

work done by the


Hence the forces
proved.

VECTOR ANALYSIS

104

Applications

to

Geometry

Relations between two right-handed systems of three


Let i, j, k and i', j', k'
mutually perpendicular unit vectors.
53.]

be two such systems.

They form

own reciprocal systems.

their

Hence
and

r = r.ii + r.jj + r-kk


= r.i'i' + r.j'j' + r.k'k'.

From

this

=
+
+ i'.k k = a +
= j'.i + j'.j + j'.k k = b, i +
1 k' = k'.i
+ k'-j j + k'.k k = Cj +
i'.i i

i'

i'.j j

j'

The

scalarsaj,

2,

a3

direction cosines of

That

bv

i'; j';

Z>

2,

2 j

J2 j
c2 j

+
+
+

53

k
k

(47')

c 3 k.

b 3 ; c r c2 , c 3 are
respectively the

k' with respect to

i,

j, k.

is

= cos
= cos
&!
= cos
cj

ax

= cos
= cos
2
c = cos
2

a2

(i', i)

Z>

(j', i)

(k', i)

(i',

j)

(j',

j)

= cos
= cos
3
c = cos
3
3

Z>

(k', j)

(i',

k)

(j',

k)

(48)

(k', k).

In the same manner


s
]
(

== i.i'i'

i-j'j'

i.k'

k'=

= j-i'i' + j-j'J' + j-k'k' = aa


k = k.i'i' + k-j'j' + k-k'k' = a

and

i'

i'

i'

+ ^ j' + Cj k'
+ &a j' + c2 k'
+ 6, j' + c 3 k'

- 1 = a22 + 622 + C22


= 1 = a 32 + &3 + C32
= = ai 6j + a 62 + a 8 J3
r
= 6 lCl +62C2 + 5 3C3
j'-k' =
I k'.i' =
= cj a, + c2 a2 + c as

(47)"

j-j

)
C k.k

i'.j'

(50)

DIRECT AND SKEW PRODUCTS OF VECTORS

= =a
= 2
j.k =
i.j

and

+
+

&j 6 2 -f Cj c 2

&2 b s

(50)'

and

(51)
Cj

k'

= i'xj'= (a a

But

k'

&g

= Cj +
i

8 6j

Cg

ax

68 ) j

c 2 j 4- c 3 k.

Hence

(52)

Or

Co

and similar relations may be found


a x,

2,

orthogonal axes JT',


important and well
ease with

for the other six quantities

All these scalar relations between the

J p J2 , J3.

coefficients of a transformation

The

105

which expresses one

set of

Y\ Z in terms of another set X, I Z are


known to students of Cartesian methods.
7

",

which they are obtained here may be note-

worthy.

number

which are perhaps not so well


known, but nevertheless important, may be found by multiof vector relations,

plying the equations


i'

in vector multiplication.

a2

k.

(53)

The quantity on either side of this equality is a vector. From


its form upon the right it is seen to possess no component in

VECTOR ANALYSIS

106

direction but to lie wholly in the jk-plane ; and from


form upon the left it is seen to lie in the j'k '-plane.

the

its

Hence
Its

it

must be the

magnitude

is

line of intersection of those


2
a

or

V bf + c^.

two planes.

This gives the

scalar relations

the square of the sine of the angle


and i'. Hence the vector

The magnitude 1

af

between the vectors

is

k'-c

&1

j'

= a,j-a s

(53)

and jk-planes, and


its magnitude is the sine of the angle between the planes.
Eight other similar vectors may be found, each of which gives
the line of intersection of the j'k'-

is

one of the nine lines of intersection of the two sets of mutually orthogonal planes.

The magnitude

of the vector

is

each case the sine of the angle between the planes.


Various examples in Plane and Solid Geometry
54.]

in

may

be solved by means of products.

The perpendiculars from the vertices of a trianmeet in a point. Let A B C be the


Let
the
triangle.
perpendiculars from A to B C and from B
to CA meet in the point 0.
To show C is perpendicular
to A B.
Choose
as origin and let OA
A, OB = B, and
Example 1

gle to the opposite sides

~C. Then

By

hypothesis

- B) =
B-(A - C) = 0.
C.(B-A) = 0,
A.(C

aQ d
Subtract;

which proves the theorem.

Example 2:

To

through the point

find the vector


equation of a line
parallel to a given vector A.

drawn

DIRECT AND SKEW PRODUCTS OF VECTORS


be the origin and

Let

dius vector from

is

the vector

OB.

any point of the required

to

be the ra-

line.

Hence the vector product

parallel to A.

Ax(R-B) = 0.
This

Let

107

Then

vanishes.

^\

the desired equation. It is a vector equation in the


unknown vector R. The equation of a plane was seen (page
88) to be a scalar equation such as
is

EC = c
in the

unknown

The point

vector B.

of intersection of a line

found at once.

The equations

and a plane may be

are

Ax(B - B) =
B.C = c
AxB = AxB
(AxR)xC = (AxB)xC
A.C B - C.R A = (AxB)xC
A.C B - c A = (AxB)xC
(

(AxB)xC +

Hence

A
,

A.C

In this case howsolution evidently fails when A C


0.
ever the line is parallel to the plane and there is no solution ;

The

or, if it lies in the plane, there are

an

infinite

number

of solu-

tions.

Example 3: The introduction of vectors

to represent planes.
Heretofore vectors have been used to denote plane areas of
The direction of the vector was normal to
definite extent.

the plane and the magnitude was equal to the area to be represented. But it is possible to use vectors to denote not a

plane area but the entire plane itself, just as a vector represents
a point. The result is analogous to the plane coordinates of
analytic geometry.
a plane in space.

MN

be an assumed origin. Let


be
is to be denoted by a vector
plane

Let

The

MN

VECTOR ANALYSIS

108

whose direction is the direction of the perpendicular dropped


and whose magnitude is the
upon the plane from the origin

Thus the nearer


that perpendicular.
reciprocal of the length of
is to the origin the longer will be the vector which
a
plane

represents

it.

any radius vector drawn from the origin to a point


the plane and if p be the vector which denotes the plane,

If r be

in

then
r.p
is

the equation of the plane.


r.p

=l
For

= r cos

(r,

p) p.

Now PI

the length of p is the reciprocal of the perpendicular


to the plane.
distance from
On the other hand r cos (r, p)
is that perpendicular distance.
Hence rp must be unity.

and p be expressed

If r

in terms of

i, j,

= ui + vj + wk
rp = xu + yv + zw = l.
p

Hence

quantities u, v, w are the reciprocals of the intercepts of


the plane p upon the axes.

The

The

relation

tion of duality.

between

and p

is

symmetrical.

It

is

a rela-

If in the equation

r be regarded as variable, the


equation represents a plane p
which is the locus of all points given
If however p be
r.

by

regarded as variable and r as constant, the equation represents a point r


through which all the planes p pass. The
development of the idea of duality will not be carried out
It

is

familiar to all students of


geometry.

tors to denote
planes will scarcely

Chapter VII.

The use

of vec-

be alluded to again until

DIRECT AND SKEW PRODUCTS OF VECTORS

SUMMARY OF CHAPTER
The

two vectors

scalar products of

A.B

II

equal to the product

by the cosine of the angle between

of their lengths multiplied

them.

is

= A B cos (A, B)
A.B = B.A
AA = ^ 2

(1)
(2)

The necessary and


of

(3)

sufficient condition for the perpendicularity

two vectors neither of which vanishes

The

product vanishes.
are

109

is

that their scalar

scalar products of the vectors

M=

= kk = 1
=j.k = k.i =
A.B = A B + A 2 B2 + A s B3
A-A = A 2 = A? + A* + A*.

i,

j,

j'j

i.j

If the projection of a vector

The

(7)

B upon

vector product of two vectors

a vector

is

(8)

is B',

equal in magnitude to

the product of their lengths multiplied by the sine of the anThe direction of the vector product is the
gle between them.

normal to the plane of the two vectors on that side on which


a rotation of less than 180 from the first vector to the second
appears positive.

AxB = A B

The vector product

is

sin (A,

B)

c.

(9)

equal in magnitude and direction to the

vector which represents the parallelogram of which A and B


are the two adjacent sides.
The necessary and sufficient condition for the parallelism of

two vectors neither of which

VECTOR ANALYSIS

110
vanishes

is

that their vector product vanishes.

The com-

mutative laws do not hold.

AxB=-BxA
ixi
ixj

jxk
kxi

Bs -

AxB =

a)

(10)

= jxj = kxk =
= jxi = k
= kxj =
= ixk =
+ (^3 B - A

(12)

(13)
j

AxB =

The
to the

A,

(13)'

scalar triple product of three vectors

[A B

C] is equal

volume of the parallelepiped of which A, B, C are three

edges which meet in a point.

[AB C]

= A-BxC = B.CxA = CAxB


= AxBC = BxC-A = CxA-B
[ABC]

The dot and the

=-

(15)'

[ACB].

cross in a scalar triple product

(16)'

may

be inter-

changed and the order of the letters may be permuted cyclicly


without altering the value of the product ; but a change of
cyclic order

changes the sign.

[ABC]

[ABC]

(18)'

[a

be]

(19)'

DIRECT AND SKEW PRODUCTS OF VECTORS


If the

component of B perpendicular

to

A be

B",

>_

Ax (BxC) = A.C B -

(20)

AB C

(24)

(AxB)xC = A-C B - C.B A


(AxB).(CxD) = A.C B.D - A.D B.C
(AxB)x(CxD) = [A CD] B - [BCD] A
= [ABD] C-[ABC] D.
The equation which

subsists

111

(24)'

(25)

(26)

between four vectors A, B,

C,

is

[BCD]

A-[CDA]B+

[DAB]

C-

[ABC] D

= 0.

(27)

Application of formulae of vector analysis to obtain the formulae of Plane and Spherical Trigonometry.

The system

of vectors

a',

V,

c' is

said to be reciprocal to the

system of three non-coplanar vectors

bxc

a'= ,..-,

when

[a

be]

cxa
V = p-

a, b, c

r'

c'

= axb

[a b c]

[a

A vector r may be expressed in terms of a set


its

reciprocal in

(29)

c]

of vectors

and

two similar ways

= r.a' a + r.b' b + r.c' c


r = r.aa' + r.bb' + r.cc'.

(30)

(31)

The necessary and sufficient conditions that the two systems of


non-coplanar vectors a, b, c and a', b', c' be reciprocals is that

= b'.b = c'c = 1
= b'.c = b'.a = c'-a = c'-b = 0.
a'.a

a'.b

a'.c

b', c' form a system reciprocal to


form a system reciprocal to a', b', c'.

If a

a, b, c

then

a, b, c

will

VECTOR ANALYSIS

112

[PftR]

The system
system be

its

i, j,

own

[ABC]is

its

own

reciprocal

P.A

P.B

P.C

Q.A
E.A

a-B
R.B

R.C

reciprocal

a-c

and

must be a right

it

if

(34)

conversely a

or left

handed

mutually perpendicular unit vectors. Application of the theory of reciprocal systems to the solution of

system of three
scalar

and vector equations of the

vector.

The

first

degree in an

vector equation of a plane

r-A

unknown

is

= a.

(36)

Applications of the methods developed in Chapter II., to the


treatment of a system of forces acting on a rigid body and in
particular to the reduction of

any system of forces to a single


and a couple of which the plane is perpendicular to that
force.
Application of the methods to the treatment of

force

instantaneous motion of a rigid body obtaining

where v

=v +

(46)

the velocity of any point, v a translational velocity in the direction a, and a the vector angular velocity of rotation.
Further application of the methods to obtain the
is

conditions for equilibrium


virtual velocities.

by making use

of the principle of

Applications of the method to obtain

the relations which exist between the nine direction cosines


of the angles between two
of

systems
mutually orthogonal
Application to special problems in geometry including
the form under which
plane coordinates make their appearance in vector analysis and the method
by which planes (as
distinguished from finite plane areas) may be represented
axes.

by

vectors.

DIRECT AND SKEW PRODUCTS OF VECTORS


EXERCISES ON CHAPTER

113

II

Prove the following reduction formulae

= [ACD]B- A-B CxD


= B.D AxC - B-C AxD.

1.

Ax{Bx(CxD)}

2.

[AxB CxD ExF] = [A B D]

- [A B C]
= [ABE] [FCD] - [ABF]
= [CD A] [BEF] - [CDB]

BxC

3.

[AxB

4.

[PQR] (AxB)

CxA]

P.A

E F]

[C

[ABC]

[D

E F]

[E C D]

[AEFj.

2
.

P.B

Q-A

Q.B

R-A

R.B

5.

Ax(BxC) + Bx(CxA) + Cx(AxB)

6.

[AxP

7.

Obtain formula

= 0.

BxQ CxR] + [AxQ BxR CxP]


+ [AxR BxP CxQ] = 0.
(34) in the text

by expanding

[(AxB)xP].[Cx(CtxR)]
in

two

different

ways and equating the

results.

Demonstrate directly by the above formulae that if


b', c' form a reciprocal system to a, b, c; then a, b, c form

8.
a',

a system reciprocal to

Show

9.

tors

a',

b', c'.

the connection between reciprocal systems of vecObtain some of the


triangles upon a sphere.

and polar

geometrical formulae connected with polar triangles by interpreting vector formulae such as (3) in the above list.
10.

The perpendicular

meet in a

bisectors of the sides of a triangle

point.

Find an expression for the common perpendicular


two lines not lying in the same plane.
11.

to

VECTOR ANALYSIS

114

Show by

12.

ume

vector methods that the formulae for the vol-

of a tetrahedron

whose four vertices are

is

1/3

13.

Making use

of formula (34) of the text

1
[a b c]

where

a, b, c
I

14.

which

= abc

show that

are the lengths of a, b, c respectively

= cos

m = cos

(b, c),

(c, a),

= cos

and where

(a, b).

Determine the perpendicular (as a vector quantity)


is dropped from the origin upon a plane determined by

the termini of the vectors

a, b, c.

Use the method of solution

given in Art. 46.


15.

Show

that the

volume of a tetrahedron

is

equal to one

two opposite edges by the perpendicudistance between them and the sine of the included angle.

sixth of the product of


lar

16.

If a line is

drawn in each face plane

of any triedral angle

through the vertex and perpendicular to the third edge, the


three lines thus obtained lie in a plane.

CHAPTER

III

THE DIFFEBENTIAL CALCULUS OF VECTOES


Differentiation of Functions of One Scalar Variable

IF a vector varies and changes from r to

55.]

ment

of r will be the difference

denoted as usual

between

r'

and

r'

Ar

(1)

must be a vector quantity.

unrestricted the increment


it

will be

by A r.

Ar = r'-r,
where

the incre-

and

Ar

is

If the variable r

be

of course also unrestricted

may have any magnitude and any

direction.

If,

however,

the vector r be regarded as a function (a vector function) of


a single scalar variable t the value of A r will be completely

determined when the two values

two values

and

r',

are

and

t'

of

which give the

known.

To
it

obtain a clearer conception of the quantities involved


will be advantageous to think of the vector r as drawn

from a fixed origin

(Fig. 26).

When

the independent variable t changes its


value the vector r will change, and as t
possesses one degree of freedom r will

vary in such a way that its terminus


describes a curve in space, r will be

of
the radius vector of one point
the curve; r', of a neighboring point P'.
P' of the curve. The ratio
chord

Ar
Al

FIG. 26.

r will be the

VECTOR ANALYSIS
chord

will be a vector collinear with the

in

the ratio 1

but magnified

When A t approaches zero P' will ap7 will


approach the tangent at P, and

t.

proach P, the chord


the vector

PP'

PP
ArAt

...

dT

will approach

dt

directed in that
a vector tangent to the curve at
sense in which the variable t increases along the curve.

which
If r

is

be expressed in terms of

i, j,

as

the components r v r 2 , rz will be functions of the scalar


r'

(>!

+ A rt ) i +

Ar

l i.

t?

+ A ra ) j +

+ A r3 ) k

(r 8

A r2$
An
4T A*
A J J ^ A.I
A*

A r,
A*

t.

V
*

'

and

Hence the components


spect to

of the first derivative of r

are the first derivatives with respect to

components of

The same

r.

derivatives.

dn
d

~ d* r

__ __
k
dt*
.

t*

In a similar manner

if r

non-coplanar vectors

a, b, c

|_

d n r2

dt*

of the

true for the second and higher

is

_i
_

dt*

dt*

with re-

7r

d n r3

'

(2)'

'

~dr'

be expressed in terms of any three


as

aa + 5b +

d^r_d^
"~

cc

<Z*J

dn c
c'

THE DIFFERENTIAL CALCULUS OF VECTORS


Example

The
are

Let

= a cos +
t

b sin

t.

vector r will then describe an ellipse of which a and b


This may be seen by assum-

two conjugate diameters.

ing a set of oblique Cartesian axes X,

and

117

Then

b.

X = a cos

Y=b sin

t,

coincident with a

t,

the equation of an ellipse referred to a pair of conjugate diameters of lengths a and 5 respectively.

which

is

dr
a

Hence

= a cos

( t

CL t

The tangent

+ 90) +

t.

b sin

(t

90).

=
The second

+ b cos

to the curve is parallel to the radius vector

+ 90).

for

a sin

derivative

is

(a cos

b sin

the negative of

t).

Hence

r.

_
~
is

evidently a differential equation satisfied by the ellipse.

Example

The

vector

Let

r will

=a

cosh

b sinh

t.

then describe an hyperbola of which a and

b are two conjugate diameters.

dr

=a

b cosh

= a cosh +

b sinh

sinh

t,

CL b

and
TT
Hence

is

dt 2

t.

a differential equation satisfied by the hyperbola.

VECTOR ANALYSIS

118
56

A combination

same

scalar variable

of vectors all of

may be

which depend on the

differentiated very

much

as in

ordinary calculus.

For
(a

A (a

b)

= (a + A a)

+ Ab) -

(b

Aa

Ab

A(a.b)A2

= a TT + ~T7
A
A
Hence in the limit when A = 0,

'

Aa-Ab
A~7
Ac

db

(ab) = a-r

/ d b\

<^

da.

,o,

f-

cZ

a\

(o)
.

//<x
<4 >

-(axb) = ax l77J+(T7) xb

d
dt

(abxc) = a.bx
v

'

fdv\
-\ +

a( ^-Ixc
(5)

[bx.].
The

last three of these formulae

may be demonstrated

(6)

exactly

as the first was.

The formal process

of

differentiation in vector analysis

no way from that in scalar analysis except in this


one point in which vector analysis always differs from scalar

differs in

analysis,

namely

The order

of the factors in a vector product

THE DIFFERENTIAL CALCULUS OF VECTORS

119

cannot be changed without changing the sign of the product.

Hence

of the

two formulae

and

first is
evidently incorrect, but the second correct. In
other words, scalar differentiation must take place without
The
altering the order of the factors of a vector product.

the

factors

must be

differentiated in situ.

This of course was to

be expected.
In case the vectors depend upon more than one variable
the results are practically the same. In place of total derivatives

with respect to the scalar variables, partial derivatives


Suppose a and b are two vectors which depend on

occur.

three scalar variables

depend upon

#, y,

z.

The

these three variables,

scalar product

and

it

ab

will

will have three

partial derivatives of the first order.

CT)

The second

partial derivatives are

formed in the same way.

5x9y

VECTOR ANALYSIS

120

Often

it is

more convenient to use not the derivatives but


particularly true when dealing with
formulae (3), (4) become

This

the differentials.

is

The

first differentials.

= da.b + a
= da. x b + a

d (a-b)
d (a X b)
and so

forth.

example.

The

As an

(3)'

x db,

(4)'

consider the following

illustration

If r be a unit vector

locus of the terminus of r

is

a spherical surface of unit

radius described about the origin,


ables.

db,

depends upon two

vari-

Differentiate the equation.

(d r)

= 2r
d r = 0.

(d r)

(d r)

Hence

Hence the increment dr

of a unit vector

the vector.

= 0.

perpendicular to
This can be seen geometrically. If r traces a
is

sphere the variation d r must be at each point in the tangent

plane and hence perpendicular to r.


Vector methods may be employed advantageously
*57.]
in the discussion of curvature and torsion of curves.
Let r

denote the radius vector of a curve

where

f is some vector function of the scalar t.


In most applications in physics and mechanics t
Let
represents the time.
s be the
length of arc measured from some definite point of

the curve as origin.


curve.
Hence A r / A
to unity

The increment A
s is

chord of the

approximately equal in magnitude

and approaches unity

infinitesimal.

r is the

Hence dr/ds

as its limit

when A

becomes

will be a unit vector tangent

the curve and will be directed toward that


portion of the

THE DIFFERENTIAL CALCULUS OF VECTORS


curve along which

is

increasing (Fig. 27).

Let

121

be the

unit tangent

=t
The curvature

(8)

of the curve is the

limit of the ratio of the angle through


which the tangent turns^ to tile length
The tangent changes by the increment At.
of the arc.
is

of unit length, the length of

At

is

As

approximately the angle

through which the tangent has turned measured in circular


measure. Hence the directed curvature C is
C

The vector C
t

to

is

At

dt

d*r

As=0 As

ds

ds*

LIM

collinear with

inasmuch

for

At

and hence perpendicular to

as t is a unit vector

At

is

perpendicular

t.

The

tortuosity of a curve is the limit of the ratio of the


angle through which the osculating plane turns to the length
of the arc.

vector

The

osculating plane is the plane of the tangent


and the curvature vector C. The normal to this

N = txC.

planeis

If c be a unit vector collinear

n
will be a unit

with C

=t

normal (Fig. 28) to the osculating plane and


t, c, n form an i, j, k system,

the three vectors

a right-handed rectangular system.


the angle through which the osculating
plane turns will be given approximately by
A n and hence the tortuosity is by definition
that

is,

Then

d n

FIG. 28.

s.

From
vectors

the fact that

t, c,

n form an

i,

j,

k system of unit

122

VECTOR ANALYSIS

and

= cc = nn = l
tc = cn = nt = 0.
t.t

Differentiating the

first

t.dt

set

= cdc = nd!n = 0,

and the second

do

But d t

is

parallel to c

and consequently perpendicular

dn

Hence

The increment
n

is

of n

is

0.

to n.

= 0.
t = 0.

dt

of

dcn = ndt + dnt =

+ dt'C = cdn +

perpendicular to

also perpendicular to n.

It

As the tortuosity is T = dn/ds, it

is

is

But the increment

t.

therefore parallel to

parallel to

dn and

c.

hence

too.

The

tortuosity

is

ds (txc)

dr

d*r d

T~

ds z ds

The

first

term of

d fdr
d*r
x
(
ds*2

dsds

^c

1
.

(11)

this expression vanishes.

been seen to be parallel to C

= d z i/d s 2

T moreover has
Consequently the

magnitude of T is the scalar product of T by the unit vector c in the direction of C.


It is desirable however to have
the tortuosity positive when the normal n appears to turn in
the positive or counterclockwise direction if viewed from
that side of the n c-plane upon which t or the positive part
of the curve lies.
With this convention d n appears to move
in the direction

when

the tortuosity

is

positive, that

is,

away from c. The scalar value of the tortuosity will


therefore be given by
c . T.
turns

THE DIFFERENTIAL CALCULUS OF VECTORS

cT =
But

c.

dr d B r
x
ds ds 3

c is parallel to the

And

dz r d

^s

vector d r/d

ft ^

T=

ur

The

tortuosity

may

somewhat shorter

dt

n=

if

ct

^s

VCC

Hence

Hence

*?

(13)

be obtained by another method which

not quite so straightforward.

tc = cn = nt = 0.
dtc = dc t
dc n =
dnc
dnt = dt*n.

Hence

Now

s2

= 0.

-^

r=-c.T = --as'^x-^
2
as ds* C

Hence

dt

Y/cC
2

-T-

dr
c

c is a unit vector in the direction C.

is

123

is

0.

parallel to c

hence perpendicular to

Hence dnt = 0.

Hence d n must be

parallel to

n.

Hence

But<Znis perpendicular to
c.

The

tortuosity

is

n.

the mag-

taken however with the negative sign


n
because d appears clockwise from the positive direction of
the curve. Hence the scalar tortuosity T may be given by
nitude of

dn/ds

r=-

dn
.c

ds

= n. dc
ds

(14)

dp

T = txc-^>
as

(14)'

VECTOR ANALYSIS

124

>

dC
do
ds
dc
.

=t

c.c
x

dC
,-.
- -

ds

But
t

VC -C

r=

= 0.

dV /^
VC

^1

c.c
t

x C

s
'

^TC
dr

T=

(13)

ds z
In Cartesian coordinates this becomes

T=

dx

dy

dz

ds

ds

ds 2

ds 2

ds z

ds x
ds*

ds y

dB

ds

ds 9

(13)'

'

ds 2
Those who would pursue the study of twisted curves and
surfaces in space further from the standpoint of vectors will
find the book " Application de la Methode Vectorielle de Grass-

mann d

la

Geomttrie Infinitesimale"
1

Paris,

Cam?

et

by FEHB extremely

Naud, 1899.

THE DIFFERENTIAL CALCULUS OF VECTORS


He works

helpful.
is

The

elegant.

The treatment

with vectors constantly.

notation used

however

is

from that used by the present writer.

125

slightly different

The fundamental

points of difference are exhibited in this table

EJ

ct

One used

-I

Clo
Jj

to either

Clo

~ [j

<* 1

L.

J.

<*

o
i

Q
O_|

/N^

CZl
Li
I

method need have no

2]

$oi ^Qo_|

difficulty

with the

All the important elementary properties of curves


and surfaces are there treated. They will not be taken
other.

up

here.
* Kinematics

58.]

Let

be a radius vector drawn from a fixed origin to


Let t be the time. The equation

a moving point or particle.


of the path is then

The

velocity of the particle is its rate of

This

is

the limit of the increment

LIM
* as

rArn

change of position.
increment A t.

r to the

di

A*-*.-nrTTL*jFr

( 15 )

This velocity is a vector quantity.


Its direction is the
direction of the tangent of the curve described by the particle.
The term speed is used frequently to denote merely
the scalar value of the velocity.
followed here. Then

This convention will be

be the length of the arc measured from some fixed point


of the curve.
It is found convenient in mechanics to denote

if s

differentiations with respect to the time by dots placed over


the quantity differentiated. This is the oldfluxional notation

VECTOR ANALYSIS

126

introduced by Newton.

It will also be convenient to denote

the unit tangent to the curve by

The equations become

t.

dr

T=r= j*
dS
= s=-

The
is

acceleration

is

a vector quantity.

=v

rtR\
(16)

t.

(17)

the rate of change

Let

it

of velocity.

be denoted by A.

It

Then by

definition

LIM

Av

= d v =_

and
v

Differentiate the expression

_dv_d(vt)
=
=
d7

~d~T

dv

=v

t.

dt

dv
i

'dt

di*

dz s

dt _dt d
dt
ds d

s
t

where C

is the (vector) curvature of the curve and v is the


in
the
curve. Substituting these values in the equation
speed
the result is
g? ~
+ oZ

'

The

A=st+

acceleration of a particle

C.

moving

%*
in a curve has there-

fore been
lel

broken up into two components of which one is paralto the


tangent t and of which the other is parallel to the

curvature C, that

is,

That this
would be unimportant were

perpendicular to the tangent.

resolution has been


accomplished

THE DIFFERENTIAL CALCULUS OF VECTORS


it

not for the remarkable fact which

component of the acceleration

127

The

brings to light.

it

parallel to the tangent is equal

magnitude to the rate of change of speed. It is entirely


independent of what sort of curve the particle is describing.
It would be the same if the particle described a right line
in

with the same speed as

hand the component

it

On

describes the curve.

of the acceleration

the other

normal to the tangent

equal in magnitude to the product of the square of the


speed of the particle and the curvature of the curve. The
is

sharper the curve, the greater this component. The greater


the speed of the particle, the greater the component. But the
of change of speed in path has no
normal component of the acceleration.
If r be expressed in terms of i, j, k as

effect at all

rate

= # + yj + 2 k,
v = T = xi + j/j + zk,
r

V x* +

A=v=r=

A = v=s=
^t

change
is

y*

(16)'

+ ^j+sk,
x x + yy y + 2

(15)'

Jr

(18)'
z

apparent.

A=f

and

Just

when

\V*v\

T==g- -^-

these formulae the difference between


of speed,

this

From

on

s>

the rate of

the rate of change of velocity,

this difference first

became clearly

But certain it is that


recognized would be hard to say.
Newton must have had it in mind when he stated his second
law of motion. The rate of change of velocity is proportional
to the impressed force
59.]

but rate of change of speed

is

not.

The hodograph was introduced by Hamilton

as

an

aid to the study of the curvilinear motion of a particle.


With any assumed origin the vector velocity f is laid off.
The locus of its terminus is the hodograph. In other words,

the radius vector in the hodograph gives the velocity of the

VECTOR ANALYSIS

128

It is
at any instant.
magnitude and direction
one step further' and construct the hodopossible to proceed
This is done by laying off the
graph of the hodograph.
The
=
an assumed origin.
from
r
vector acceleration A
particle in

radius vector in the hodograph of the hodograph therefore


at each instant.
gives the acceleration
in a circle (Fig. 29)
a
particle revolve
Example 1 : Let

of radius r with a uniform


a.

angular velocity

The

speed of the particle will then


be equal to
v

Let
velocity v

is

perpendicular to
r

The

vector v

r.

be the radius vector

=v=a

r.

always perpendicular and of constant magnia r.


is therefore a circle of radius v

radius vector f in this circle

just ninety degrees in

is

advance of the radius vector r in

its

which

circle,

and

it

conse-

same angular velocity

quently describes the circle with the

The

The

drawn to the particle.


It is
r and to a.

The hodograph

tude.

The

is

=a

the rate of change of v


always perpendicular to v and equal in magnitude to
a.

acceleration

is

A = a v = a2
The

acceleration

A may

r.

be given by the formula

r = A = axv = ax(axr) = ar

But

as a

is

Hence

The

=A=

acceleration

a circle

is

aa

perpendicular to the plane in which r


r

is

aa

a2

r.

lies,

= 0.

r.

due to the uniform motion of a particle in


and is equal in magni-

directed toward the centre

tude to the square of the angular velocity multiplied by the


radius of the circle.

THE DIFFERENTIAL CALCULUS OF VECTORS


Example 2:

Consider the motion of a projectile.

acceleration in this case

The hodograph of
vector.
The curve
the

hodograph reduces to a constant

the
is

any given

=v +

Thus the hodograph


ing through the

At

instant.

will be

a later instant the velocity


t

g.

a straight line parallel to g and passThe hodograph of a


extremity of v
is

particle moving under the influence of gravity

The path

straight line.

Example 3

The

the acceleration g due to gravity.

is

merely a point. It is easy to find


Let v be the velocity of the projectile

hodograph.

in path at

129

well

is

known

is

hence a

to be a parabola.

In case a particle move under any central

acceleration

= A = f(r).

The tangents to the hodograph of r are


But these tangents are approximately

the accelerations

r.

collinear with the

chords between two successive values r and f of the radius

That

vector in the hodograph.

is

approximately

A*
Multiply by rx.

=rx

Since r and r are parallel


r

Hence

But ^

r
r

x
x

- r ) = 0.
r = r x f

(r

r is the rate of description of area.

when

Hence the

a particle moves under an acequation


celeration directed towards the centre, equal areas are swept
states

that

over in equal times by the radius vector.


9

VECTOR ANALYSIS

130
it

Perhaps

would be well

this question.

go a

to

little

more carefully

If r be the radius vector of the

into

particle in

path at one instant, the radius vector at the next instant


r + A r.
The area of the vector of which r and r + A r are

its
is

the bounding radii is approximately equal to the area of the


This
triangle enclosed by r, r + A r, and the chord A r.
area

is

|rx(r + Ar)=irxr + ^r
The

rate

of description

xAr=|rxAr.

of area

by the radius vector

consequently

Lm

y r-

irx(r*-Ar)
A*

A*=02
Let

Lm
~A*=0

r <

Ar_i.
A t~*

and

r be two values of the


velocity at two points
which are near together. The acceleration r at P

P and P
is

is

the limit of
r

_ Af

r'

A*
Break up the vector
parallel

= Lz

'

into

two components one

and the other perpendicular to the acceleration f

Af
if

n be a normal to the vector

i?

The quantity x apThe quantity y

proaches unity when A t approaches zero.


approaches zero when A t
zero.

approaches

Hence

(r

(r

-r)=rxi-

r'

= x A*

r'

(r

A*

^A A x

n.

THE DIFFERENTIAL CALCULUS OF VECTORS

131

Hence

rxf

xi =

But each

Ar
X
At

upon the right-hand side is an


Hence the rates of descripinfinitesimal of the second order.
tion of area at P and P differ by an infinitesimal of the
of the three terms

This

second order with respect to the time.

Hence the

is

true for any

must be exactly equal


This proves the theorem.
The motion of a rigid body one point of which is
at any instant a rotation about an instantaneous axis

point of the curve.

rates

at all points.
60.]

fixed

is

passing through the fixed point.


Let i, j, k be three axes fixed in the body but moving in
space. Let the radius vector r be drawn from the fixed point
to any point of the body.
Then

But

= (d r

i) i

Substituting the values of


the second equation

dr

But
Hence

j) j

c?

i,

j,

di+

yi

dj

+ (# j di +
+ (#kdi +

yj

rf

=j

irfj+j <! =
jdk + kdj =
k.di + idk =

Moreover

Hence

= (xi

(d r

()

(d r

d r k obtained from

+ 2i^k)i
+ z j d k) j

ykcj +

=k
or

k) k.

krfk)k.

= 0.

j(H =

or k.dj =
or i dk =
ii=jj = kk = l.
i^i = d = kc?k = 0.

irfj
j.tfk
'k

di.

VECTOR ANALYSIS

132

in the expression for


Substituting these values

= (2

dk

(y k

+
This

dj

is

a vector product.

= (Wj i + idk j +

Let
a

dj

=k

-^

Then

(# j

d k)

k)x(>

dk

di=

d j)

k.

yj

z k).

+ l -d7 J+
'-dl

i) i

r.

axr

This shows that the instantaneous motion of the body is one


the angular velocity a about the line a.

of rotation with

This angular velocity changes from instant to instant. The


proof of this theorem fills the lacuna in the work in Art. 51.

Two
Let

aj

due to

infinitesimal

rotations

and a 2 be two angular


them are

may be added like vectors.


velocities. The displacements

= aj
dz r = aa
dj r

If r be displaced

T
If it

d1

Hence

+dr=r+

d1

ftl

If the infinitesimals

(d

t,

t.

it

a.

xidt.

becomes

+ % x [r + (& X r) d f\ d
+ a a x r d t + a 2 x (a x r)

=T+

then be displaced by a a ,
r

becomes

a, it

by

t.

t)

(d t)\

of order higher than the first be

neglected,

= &1

a2

which proves the theorem.


.

= dr =

If

( &1

a 2)

t,

both sides be divided by d

a1

a a )xr.

THE DIFFERENTIAL CALCULUS OF VECTORS

133

the parallelogram law for angular velocities.


was obtained before (Art. 51) in a different way.

This

It

is

In case the direction of

a,

the instantaneous axis,

is

con-

stant, the motion reduces to one of steady rotation about a.

=a

The

As

acceleration

r.

r=axr+axr=axr+ax

a does not change

a and hence a x r

On

is

its

direction a

parallel to a

the other

hand

r.

ax

(axr).

must be collinear with


That is, it is perpen-

dicular to

r.

Inasmuch

as all points of the rotating

(a

r) is parallel to

r.

body move in con-

about a in planes perpendicular to a, it is


unnecessary to consider more than one such plane.
The part of the acceleration of a particle toward the centre
centric circles

of the circle in

This

which

it

moves

is

a x (a

r).

equal in magnitude to the square of the angular


It does not
velocity multiplied by the radius of the circle.
is

depend upon the angular acceleration a


to what is known as centrifugal force.

at all.

On

It corresponds

the other hand

the acceleration normal to the radius of the circle

is

axr.
This

equal in magnitude to the rate of change of angular


velocity multiplied by the radius of the circle. It does not
is

depend in any way upon the angular velocity itself but only
upon its rate of change.
61.] The subject of integration of vector equations in which
the differentials depend upon scalar variables needs but a
word. It is precisely like integration in ordinary calculus.
If

then

d
r

r = d
= s + C,
s,

VECTOR ANALYSIS

134

where C

is

some constant

To accomplish

vector.

the integra-

tion in

any particular case may be a matter of some

just as

it is

Example

difficulty

in the case of ordinary integration of scalars.


1: Integrate the equation of motion of

projectile.

The equation

of motion

is

simply

r=

g,

which expresses the fact that the acceleration


to gravity.
tically downward and due

is

a constant of integration.

It is

evidently the

= 0.

velocity at the time t

c is another constant of integration.

of the point at time

always ver-

= g + b,

where b

is

= 0.

It is the position vector

The path which

is given by this
That this is so may be seen by
x and y and eliminating t.

last equation is a parabola.

expressing

it

in terms of

Example 2 : The rate of description of areas when a parmoves under a central acceleration is constant.

ticle

Since the acceleration

But
For

(r

x
x

is

= f(r).
parallel to the radius,

f)

Hence
r

0.

(r

(r

j)

C,

cl t

and

which proves the statement.

r).

r.

THE DIFFERENTIAL CALCULUS OF VECTORS

135

Example 3

:
Integrate the equation of motion for a particle
with
an
acceleration toward the centre and equal to
moving

a constant multiple of the inverse square of the distance

from the centre.


2

Given

= --c

Then

3 r.

r3

Hence

0.

= C.

Multiply the equations together with X.


r

x C

-^2-

Differentiate.

7F

rx ( rxr )=

Then

=rf

c2

Each

r2

is

'
{r rr

-'"}.

r2

side of this equality

7F

121?

Hence

1
:

r.

a perfect differential.

= + e I,
r

Integrate.

Then

where

the vector constant of integration,

c2

e I is

tude and I a unit vector in


tion

by

x C

C
C*

x C

e r

e is its

magniMultiply the equa-

But

its direction.

I.

rxr.CCC

VECTOR ANALYSIS

136

p=

Let

and cos u

p=

Then

Or

P
6

axis.

COS

the equation of the ellipse of which e is the eccentriI is drawn in the direction of the major

is

city.

(r, I).

r cos u.

f=

This

= cos

The vector
The length

of this axis is

drr
1
e

It is possible to carry the integration further

So

the time.

far

merely the path has been found.

The Operator

Scalar Functions of Position in Space.


62.]

A function V

(a?,

y, 2)

which takes on a

value for each set of coordinates

#, y, z

definite scalar

in space

Such a

scalar function of position in space.

ample,

and obtain

is

called a

function, for ex-

is

(a?,

y, z)

=x 2 +

zz

= r2

This function gives the square of the distance of the point


will be supposed to
(, y, z) from the origin. The function

be in general continuous and single- valued. In physics scalar


functions of position are of constant occurrence. In the
theory of heat the temperature T at any point of a body is a
scalar function of the position of that point.
In mechanics

and theories of attraction the potential


function.

This, too,

If a scalar function

is

is the all-important
a scalar function of position.
be set equal to a constant, the equa-

tion

r(x,y,z)=c.

(20)

defines a surface in space such that at


every point of it the

function

V has the

same value

c.

In case

F be

the tempera-

THE DIFFERENTIAL CALCULUS OF VECTORS


ture, this is a surface of constant temperature.

In case

isothermal surface.

137

It is called

an

V be the potential, this surface of

constant potential is known as an equipotential surface. As


is a typical scalar function of position in
space,

the potential

and

most important of

as it is perhaps the

owing

all

such functions

to its manifold applications, the surface

V (x, y,z}=c
V equal to

obtained by setting

a constant

is

frequently spoken
an equipotential surface even in the case where V has
no connection with the potential, but is any scalar function
of as

of positions in space.

The

rate at

that

tion

constant

is,

which the function V increases in the


direcwhen x changes to x + A x and y and z remain

is

LIM

(a;

+ A a,

y, g)

- T (x, y, z)

"I

J*
This

the partial derivative of Fwith respect to x. Hence


increases in the directions of the three

is

the rates at which

axes

F,

Z are

respectively

9V
9 x

9V

9V_ '
9 y

'

Q z

Inasmuch as these are rates in a certain direction, they may


Let i, j, k be a system
be written appropriately as vectors.
unit
of
vectors coincident with the rectangular system of
axes X, F, Z. The rates of increase of V are
1
.

The sum

3V .9V
- 9V
9x

Ku

of these three vectors

9 z

9 y

would therefore appear

to be

a vector which represents both in magnitude and direction


the resultant or most rapid rate of increase of V. That this
is

actually the case will be

shown

later (Art. 64).

VECTOR ANALYSIS

138

The

63.]

F is

of

vector

denoted

V F represents

sum which is the resultant

a directed rate of change of

For

or vector derivative of V, so to speak.

V; and

will be called the derivative of

VF.

The terms

V F.

It is

a vector

gradient and

customary to regard

VFfrom

slope of

is

now

in

a directed

this reason

VV

F, the primitive of
also used for

F are

V as an operator which obtains

a scalar function

This symbolic operator

Hamilton and

rate of increase

by VF.

Fof

position in space.

was introduced by Sir W. R.


There
employment.

universal

1
seems, however, to be no universally recognized name for it,
although owing to the frequent occurrence of the symbol

some name

is a practical necessity.
It has been found by
that
the
del
is
so short and easy to
experience
monosyllable
occurs
pronounce that even in complicated formulae in which

a number of times no inconvenience to the speaker or hearer


" del F."
arises from the repetition.
is read simply as
has been defined as
Although this operator

VF

d
V = .9
+ --h
dx
.

3y

Some

use the term Nabla owing to

its

Qz

fancied resemblance to an Assyrian

A and have consequently


coined the none too euphonious name Atled
by inverting the order of the letters in
the word Delta. Fb'ppl in his
Einfuhrung in die MaxweWsche Theorie der Electricitat avoids
any special designation and refers to the symbol as "die Operation
V." How this is to be read is not
divulged. Indeed, for printing no particular
name is necessary, but for lecturing and purposes of instruction
something is required
something too that does not confuse the speaker or hearer even when
often repeated.
harp.

Others have noted

its

likeness to an inverted

THE DIFFERENTIAL CALCULUS OF VECTORS


so that

139

appears to depend upon the choice of the axes,

it

it

This would be surmised


in reality independent of them.
as the magnitude and direction
from the interpretation of
is

most rapid increase

of the

of V.

pendence take another set of axes,


variables a/, y', z' referred to them.
system

To demonstrate the indei', j', k' and a new set of


Then V referred to this

is

+*

< 22 >'

use of the formulae (47)' and (47)", Art. 53, page


104, for transformation of axes from i, j, k to i', j', k' and by
actually carrying out the differentiations and finally by

By making

taking into account the identities (49) and (50),


actually be transformed into V.

may

= V.

V'

The

details of the proof are omitted here, because another

shorter

method

of demonstration

is

to be given.

Consider two surfaces (Fig. 30)

64. ]

V O,

y,z)=c

V (#, y, z) = c + A

and

c,

upon which Vis constant and which are moreover infinitely


near together. Let #, y, z be a given point upon the surface

V=

Let

c.

dius

vector

denote the ra-

drawn

to

point from any fixed

this

origin.

Then any
the

point near by in
neighboring surface

=c+

may

be represented

by the radius vector

The actual
the
is

first

increase of

d r.

Ffrom

FIG. 30.

surface to the second

a fixed quantity dc.

Tlie rate of increase is a variable

VECTOR ANALYSIS

140

dr which is folquantity and depends upon the direction


to
the
other.
The rate
one
surface
from
lowed when passing
of increase will be^the^guptieir^of the actual increase d c and
dr between the surfaces at the point
tnV~fTifitapOft

Vdr

as,

y, 2 in

the direction d

Let n be a unit normal to the

r.

and d n the segment of that normal intercepted


between the surfaces, n d n will then be the least value for
surfaces

dr.

The quotient

yd r
will therefore

equal in

be a

dr

maximum when d r

magnitude of

d n.

is

n and

parallel to

The expression
n
^
an

(23)

which the direction is the direction of


most rapid increase of Fand of which the magnitude is the
is

therefore a vector of

rate of that increase.

the axes X, Y, Z.
is

the increment of

second.

Then

let

This vector

Fin

is

entirely independent of

be replaced by its equal d


which
passing from the first surface to the

Let d

V F"be

defined again as

VV

From

this definition,
is certainly the vector which
the
direction
of
most rapid increase of
and the rate
gives
in that direction.
Moreover VFis independent of the axes.

It remains to

show that

first

To do

given.

this definition is

this multiply

by

d n

equivalent to the one

d r.
.rfr.

(25)

n is a unit normal. Hence n d r is the


projection of d r on
n and must be equal to the
perpendicular distance d n between
the surfaces.

THE DIFFL ENTIAL CALCULUS OF VECTORS

VF.

= dV dn = dV.

dr

9V

where

(d #)

(d y)

9V
-^-dy

ay

(d z) 2

9V
-~-dz,
az

= dr

dr

takes on successively the values


equation (25)' takes on the values
If

VF

(25)'

dn

dx+
dV=^dx

But

dx

141

d r.
dx,

dy,

bdz the

= 9V
dx
TT

VV.)dy =

dx

9V

r dy
-

(26)

9V
W*ls.dz = ^dz.
9z
If the factors

d x, d y d z be cancelled these equations state


k of
in the
y

VF

VT

VF

that the components


i,
j,
i, j, k directions respectively are equal to

9V_
9 x*

VF= (VF. i)
Hence by(26)

The second

9V

9V

3y*

9z'

+ (VF- j) j + (VF.

VF=i 9 x + j+k
9
y

definition (24) has

and consequently

is

equivalent to

k.

(21)

been reduced to the

first

it.

The equation (25V found above is often taken as a


V F. Accordmg to ordinary calculus the deriv-

*65.]

definition of

9z

k)

11

ative -^- satisfies the equation

d x

dy

VECTOR ANALYSIS

142

Moreover
it is

down

the following definition.


of a scalar function of
derivative

possible to lay

Definition:

In a similar manner

dy / dx.

this equation defines

VF

The

position in space shall satisfy the equation

for all values of dr.

This definition

is

certainly the

most natural and important

But for practical purposes


the
before
either of
definitions
given seems to be better.
They are more tangible. The real significance of this last
from theoretical considerations.

definition cannot be appreciated until the subject of linear

vector functions has been treated.

The computation
frequently

See Chapter VII.

V of

of the derivative

a function

on by means of the ordinary

carried

is

most

partial

differentiation.

Example 1 :

Let

V (x, y,z)r = V
.

dr

3r

.dr

i^-+J^
r +k^ax
&z
dy
Vr=:i

Vx* +

Vx* +

z2

y'*

z*

Vx* +
Hence

vr =

Vr =

and

y'*

zz

The
This

is

derivative of r

is

a unit vector in the direction of

r.

evidently the direction of most rapid increase of r


and the rate of that increase.

THE DIFFERENTIAL CALCULUS OF VECTORS


Let

143

_i
(x

x2

-f

y
v2

22

z 2 )'

<>2

Hence

V-=
r

5
2
(a;

of

r,

The proof

Example 4

'

= -Y3

-T

(r

is

r)

a vector whose direction

that

equal to the reciprocal of the

is

1*

= 711*

to the reader.

Let

F(#,

y, 2)

= log y x 2 -f

drawn from the

V may be

(#, y, 2) of space, the function

= log Vr
ia? + jy = r

F(#,y, 2)

and

V log V # + # =
/

origin to the point

written as

(k r)

k kr.
r

&
2

u
2

+ j y).

(i

If r denote the vector

v3

a;

Hence

is

r.

Vr = nr*

is left

2)

and whose magnitude

Example 3 :

derivative of 1/r

square of the length

2 2) 1

i x - j y + 2 2^- (-

^_

The

-=2

V -r =

and

2
2/

k kr
.

^Krj*

k kr

(r-kk.r).(r-kk.r)'

VECTOR ANALYSIS

144

There

is

another method of computing

which

is

based

upon the identity

F = Vrr = r.

Example 1 : Let

YT

Vr

= -'
^7^=-^=
V rr r

Hence
Example 2 :

F= r

Let

where a

a,

a constant vector.

is

dF=rfr-a = <2r.VF.

VF=a.

Hence
Example 3:

Let

constant vectors.

F= (rxa)

(rxb), where a and b are

F = rr ab >ra rb.
r<<
*T
ab - dr-a r-b - drb ra = r V F.
,

Ar.rtub*

d F = 2 drr

i-.*-

'

la

<f

VF= 2rab ar.b bra


VF= (ra.b-ar.b) + (rab-br.a)

Hence

= bx(rxa) +
Which

of these

two methods

x (rxb).
for

computing

shall

be

applied in a particular case depends entirely upon their


The latter method is
relative ease of execution in that case.

independent of the cob'rdinate axes and


preferred.

may

therefore be

It is also shorter in case the function

Fcan

be

cannot be so
expressed easily in terms of r. But when
the
to.
former
be
resorted
method
has
to
expressed

The

in mathegreat importance of the operator


matical physics may be seen from a few illustrations.
Sup*66.]

pose

(#, y, 2)

be the temperature at the point #,

y, 2 of

THE DIFFERENTIAL CALCULUS OF VECTORS

145

That direction in which the temperature

heated body.

de-

most rapidly gives the direction of the flow of heat.


as has been seen, gives the direction of most rapid

creases

jP,

Hence the flow

increase of temperature.
f

where
body.

of heat f

is

= -k Vr,

a constant depending upon the material of the


Suppose again that V be the gravitational potential

Jc

is

The

due to a fixed body.


y, z) is in

the point (#,

force acting upon a unit mass at


the direction of most rapid increase of

potential and is in magnitude equal to the rate of increase


per unit length in that direction. Let F be the force per unit

Then

mass.

As

= VF.

different writers use different conventions as regards the

sign of the gravitational potential,

that the potential

V referred

the potential energy.


situated at

a mass

W denoted

If
#,

it

/,

2,

might be well

to state

to here has the opposite sign to

the potential energy of

the force acting upon that mass

would be

In case

represent the electric or magnetic potential due

to a definite electric charge or to a definite magnetic pole re-

spectively the force F acting


as the case might be is

The

force

is

in

upon a unit charge or unit pole

= - VF.

the direction of most rapid decrease of

In dealing with electricity and magnetism potenpotential.


whereas in
tial and potential energy have the same sign
;

attraction

problems they are generally considered to have

opposite signs. The direction of the force in either case


the direction of most rapid decrease of potential energy.
difference between potential

and potential energy


10

is

is

in

The
this.

VECTOR ANALYSIS

146

Potential in electricity or magnetism is the potential energy


and potential in attraction problems
per unit charge or pole ;
with the
is
energy per unit mass taken, however,
potential

negative sign.
It is

*67.]

often convenient to treat an operator as a


it obeys the same formal laws as that

quantity provided
Consider for example the partial differentiators
quantity.

JL _?__?_.
9x* 9y* 9z

As

far as combinations of these are concerned, the formal

are precisely

what they would be

laws

instead of differentiators

if

three true scalars


a,

were given.

&,

For instance

the commutative law

99

-r

ao

9y9x

9x9y

= o a,

the associative law

9/99\/99\9
=

;r-

;r- ;r~

9z\9y9zJ
and the

distributive

9/9

ir~ TT~

a (6 c)

= /(a w
6)c,

law

99 99
= --5zJ 9x9y 9x9z

----

9 \
-

9
9x\dy

\9x9yj9z

hold for the differentiators just as for scalars.

Of course such

formulae as

w
where u

is

^ = ai M

>

a function of x cannot hold on account of the

properties of differentiators.

A scalar

function u cannot be

placed under the influence of the sign of differentiators.


Such a patent error may be avoided by remembering that an

operand must be understood upon which 9/9 #

is

to operate.

THE DIFFERENTIAL CALCULUS OF VECTORS


In the same way a great advantage
looking upon

Sf\

V
as a vector.

4-

dz

dy

It is not a true vector, for the coefficients

are not true scalars.

a vector differentiator and of

It is

always implied with it As far as formal


concerned it behaves like a vector.
For

course an operand
operations
instance

obtained by

f\

4-

9x

may be

147

are

is

V (u 0) = (V u) + u (V 0),
c V u = V (c M),
i)

if

u and v are any two scalar functions of the scalar variables


and if c be a scalar independent of the variables with

#, y, z

regard to which the differentiations are performed.


68.] If A represent any vector the formal combination

A. Vis

=^

^ii;

A = Al i +

provided

-4 2

< 2T>

^a4-'

+ A 3 k.

V is a scalar differentiator. When


to a scalar function V (x, y, z) it gives a scalar.
V
V
V
+ A,- + A 3 -.
This operator

^r=A -

e$

Suppose for convenience that

applied

is

a unit vector

a.

-+ a3

C/ vC

(28)

(29)

VECTOR ANALYSIS

148

where a v

2,

to the axes

is

are the direction cosines of the line a referred

X, Y, Z.

well-known
This

Consequently (a V)
derivative of

directional

F appears

as the

in the direction

a.

often written

It expresses the

3x

5s

a
z

3y

magnitude of the rate of increase of

F in

In the particular case where -this direction is


the normal n to a surface of constant value of F, this relation
the direction

a.

becomes the normal derivative.

n v n v 3 be the direction cosines of the normal.


The operator a V applied to a scalar function of position
F yields the same result as the direct product of a and the

if

vector

V F.
(a.V)F=a.(VF).

For

this reason either operation

may

(30)

be denoted simply by

a.VF
without parentheses and no ambiguity can result from the
omission. The two different forms (a V) Fand a (V F)

may however be interpreted in an important


(a V) F is the directional derivative of F in the
.

a.

On

the other hand a

the direction

any direction

VFin that

is

(V F) is the component of V F in
The directional derivative of F in
component of the derivative
Fdenote gravitational potential the

equal to the

direction.

theorem becomes
in

Hence

a.

theorem.
direction

The

If

directional derivative of the potential

any direction gives the component of the force per unit


mass in that direction. In case Fbe electric or magnetic
potential a difference of sign must be observed.

THE DIFFERENTIAL CALCULUS OF VECTORS

149

Vector Functions of Position in Space

A vector

69.]

&

which

5>M J

v:

function of position in space

*-

* ^i^r

"*?

a function

(x
v z\'
v ' yj

with each point

associates

x, y,

space a definite

z in

The function may be broken up

vector.

is

into its three

com-

ponents

(x, y, z)

Examples

F!

(x,

y,z)i+

(as,

y, z)

of vector functions are very

VF has

Already the function


space

F2

(x, y, z) k.

numerous in physics.

At

occurred.

V F has in general a definite

+ F3

each point of

vector value.

In mechan-

each point of the body is a


vector function of the position of the point. Fluxes of heat,
ics of rigid bodies the velocity of

electricity,

vector functions

fluids, etc., are all

magnetic force,

of position in space.

The
tion

Let

= V

(x, y, z), i

and
a

(a.V)V
(

may

al

V=

04

(a-V)

a.V)V

+ F2

Then

and

scalar operator a

be applied to a vector func-

to yield another vector function.

(a,

2 j

Qx
i

(a

F3

y,z)j+

a3

-9
3y

V)

(as,

y, z)

k.

l~

5
a 3T
o

F2 j +( a .V)F3

9V

(31)

VECTOR ANALYSIS

150

This

may be

written in the f oral

Hence (a V)
function

the directional derivative of the vector

is

in the direction

without parentheses.

when

appliod to

= a VV

For the meaning of the vector symbol


yp.pt.nr fn notion V has not yet been

ft-

Hence from the present standpoint the expression


have but the one interpretation given to it by

dctiiu'd.

V) V

(a

It is possible to write

a.

VV can

(a.V)
70.]

V.

Although the operation

V V has

not been defined and

cannot be at present, 1 two formal combinations of the vector


and a vector function V may be treated. These
operator

and the (formal) vector prod-

are the (formal) scalar product


uct of
into V.
They are

and

VxV =

V V is read del dot V;


The

differentiators

by the dot and the

+j

and

+
x

500
-

=lx

That

< 32 >

xV

(33 )

V, del cross V.

cross.

being scalar operators, pass

is

+J x

+k

x.

These may be expressed in terms of the


components
of V.
1

definition of

AV

will

be given

in

Chapter

VTL

(33)'

Vv Vv Vz

T^^ DIFFERENTIAL CALCULUS OF VECTORS


Now.

PV_3F
= ^^
9x 9x
3V
_

3F,i

3F2

1i

3F3v

.
,

9x

9x
i

3F9
i

3F-2 V
9y

3y

3V

151

i
J -

(34)

9z

K*j-

3V_3Fj

Then

c) ZC

Q) tC

3V

Hence

v.v =

^ ^
+

+ ^^

Moreover

3V = i 3F23 k
-r.

3V = 3FJ _.
3s

3s

3s

Hence

\
This

may

3a;

3y J

be written in the form of a determinant

Vx V =

9 y

3s

999
i

(33)'"

C*K

VECTOR ANALYSIS

152

be applied to
the
determinant.
expanding

to
It is to be understood that the operators are

Vv Vv Vz when

the functions

forward
standpoints objections may be brought
V
as a symbolic vector and introducing
against treating
and
scalar
vector
the
as
x V respectively
and
symbolic
These objections may be avoided by
into V.
of
products
and
definition that the symbols
simply laying down the
new
as
looked
be
operators
entirely
upon
x, which may

From some

shall be
quite distinct from V,

= ix
But

for practical purposes

seems by

all

+k

+ jx
and

means advisable

9x

for

^
71.]

(33)'

remembering formulae

it

to regard

"

9z

Qy

This symbol obeys the

as a symbolic vector differentiator.

same laws as a vector

x.

just in so far as the differentiators

obey the same laws as ordinary scalar quantities.

That the two functions

V V

and

x V have very

important physical meanings in connection with the vector


function V may be easily recognized.
By the straight-

forward proof indicated in Art. 63


operator

V is

it

was seen that the

independent of the choice of axes.

fact the inference is

immediate that

intrinsic properties of

From

this

V V and V x V represent

invariant of choice of axes.

In order

to perceive these
properties it is convenient to attribute to the

function

some

meaning such as flux or


Let therefore the vector V denote

definite physical

flow of a fluid substance.

153
at each point of space the direction

flow of some fluid.

This

may

and the magnitude of the

be a material fluid as water

or gas, or a fictitious one as heat or electricity.


To obtain as
clearness
as
possible let- the fluid be material but not
great
necessarily restricted to incompressibility like water.

v.V =

Then
is

dx

called the divergence of

dz

9y

and

is

often written

VV=div V.
The reason

VV gives at each

term is that
volume per unit time

for this

rate per unit

at

which

point the

fluid is leaving

the rate of diminution of density. To prove


that point
this consider a small cube of matter (Fig. 31).
Let the edges
of the cube be dx, dy, and dz respectively.
Let

V O,

y, z)

=V

(x, y, z) i

V^

(x, y, z)

+ Vz (x, y

z) k.

Consider the amount of fluid which passes through those


faces of the cube which are parallel to the F^-plane, i. e.

perpendicular to the
axis.
The normal to the
face

whose x coordinate

the lesser, that

mal

is,

is

the nor-

to the left-hand face

of the cube

is

i.

The

-i

d.y

flux

of substance through this

face

is
i

(x, y, z)

face,

31

the face whose

x coordinate
through

dz.

to the oppo- z

The normal
site

dy

greater by the
therefore

is

it is

amount dx,

is

+ i and

the flux

VECTOR ANALYSIS

154

[3V
V(x,y,z)

= i V (x, y, z)

dy dz

#V dx

<y

The
faces
is

total
is

dy

"I

dx\ dydz

dz.

<X/

outward from the cube through these two


sum of these quantities. This

flux

therefore the algebraic

simply
i

dx dy dz

dx

-^

ox

dx dy

dz.

In like manner the fluxes through the other pairs of faces of


the cube are
V 7
,
,
3
j 1.3V ,
k = dx dy dz.
i .
dx dy dz and
&Z
&y

The

total flux

out from the cube

3V

/.
i

Tdx

This

is

3V
o~
9y

is

3V

the net quantity of fluid

unit time.

The quotient

therefore

of this

dx d

dz

which leaves the cube per


by the volume dx dy dz of

the cube gives the rate of diminution of density.

V.V=i.

3V
c/a;

3Vdy

3V =
dz

3F,
-i
ax

This

is

3F3
3F
+ __2^ + az-*.
ay

Because
V thus represents the diminution of density
or the rate at which matter is leaving a point per unit volume
per unit time, it is called the divergence. Maxwell employed
the term convergence to denote the rate at which fluid approaches a point per unit volume per unit time. This is the
negative of the divergence. In case the fluid is incompressible,
as much matter must leave the cube as enters it. The total

change of contents must therefore be

zero.

For

this reason

the characteristic differential equation which


any incompressible fluid

must

satisfy is

THE DIFFERENTIAL CALCULUS OF VECTORS


where

known

as the hydrodynamic equation.

is

This equation

the flux of the fluid.

flow of water, since water

is

is

It is satisfied

155
often

by any

The

practically incompressible.

great importance of the equation for work in electricity is due


to the fact that according to Maxwell's hypothesis electric dis-

placement obeys the same laws as an incompressible


then D be the electric displacement,
div

If

= V D = 0.

To

the operator
x Maxwell gave the
This nomenclature has become widely accepted.
72.]

fluid.

name

curl.

V x V = curl V.
The

curl of a vector function

As

of position in space.

the

is itself

name

a vector function

indicates, it is closely

connected with the angular velocity or spin of the flux at


each point. But the interpretation of the curl is neither so
easily obtained nor so simple as that of the divergence.
Consider as before that V represents the flux of a fluid.

Take

an infinitesimal sphere about any


the next instant what has become of the

at a definite instant

At

point (#, y, z).


sphere ? In the first place it may have moved off as a whole
in a certain direction by an amount d r. In other words it

may have a translational velocity of di/dt. In addition to


this it may have undergone such a deformation that it is no
longer a sphere. It may have been subjected to a strain by
virtue of which

Finally

it

may

axis through

it

becomes slightly

have been rotated

an angle dw.

That

jis_
is

in shape.
a whole about some
ellipsoidal

to say, it

may have an

An
angular velocity the magnitude of which is dw/dt.
infinitesimal sphere therefore may have any one of three
distinct types of

motion or

all of

translation with definite velocity.


definite rates of elongation

them combined.

Firsjb,

Second, a strain with three

along the axes of an

ellipsoid.

VECTOR ANALYSIS

156

Th0d, an angular velocity about a definite axis. It is this


which is given by the curl. In fact,
is a vector which has at each point of

third type of motion


the curl of the flux V

space the direction of the instantaneous axis of rotation at


that point and a magnitude equal to twice the instantaneous

angular velocity about that axis.


analytic discussion of the motion of a fluid presents
than it is necessary to introduce in treating
The motion of a rigid body is sufficiently complex
the curl.

The

more

difficulties

was seen (Art.


51) that the velocity of the particles of a rigid body at any
instant is given by the formula
to give

an adequate idea of the operation.

=v +a x
curl v = Vxv = Vxv + Vx
a = a i + a2 j + 3 k
v

Let

product of

x (a x r) formally
V, a, and r. Then

V
is

x v

=V

x v

is

as if it

(axr).

+ (V

r) a

were the vector

- (V

Hence the term

a constant vector.

v
As a

r.

expand

It

=_dx
^

a constant vector

dx
it

of the differential operator,

1-

9y
dy

\-

may be

a)

x v

triple

r.

vanishes.

dz_
= 6.
dz

placed upon the other side

=a

V.

f\

l^+ 2^+"3^
(r\
Hence

Vxv = 3a

= 2a.

Therefore in the case of the motion of a rigid body the curl


of the linear velocity at
any point is equal to twice the
angular velocity in magnitude and in direction.

THE DIFFERENTIAL CALCULUS OF VECTORS

157

V x v = curl v = 2
a =
gVxv = 2 curl v.
a,

=v +

(V x

v) x r

= Vo +

\ (curl

v)

r.

(34)

The expansion of V X (a x r) formally may be avoided by


multiplying a x r out and then applying the operator V X to
the result.
73.]

tion

It frequently happens, as in the case of the applica-

just cited, that the operators

V>V*>

V x,

have to be

applied to combinations of scalar functions, vector functions,


or both. The following rules of operation will be found
useful.
Let w, v be scalar functions and u, v vector functions of position in space.

Then

+ v) = Vu + Vv
(35)
=
V.u
V.v
+
V.(u + v)
(36)
=
Vxu
Vxv
+
Vx(u + v)
(37)
=
v
V
u
u
V
v
V (u v)
+
(38)
=
V.(wv) Vwv + wVv
(39)
B
Vx ( v) V * x .?'+ V X ?
(40)
=
v.
Vn + u Vv
V(uv)
(41)
+ v x (V x u) + n x (V x v) 1

(u

'

V(uxv)=v.Vxu u.Vxv
V

X (u x v)

A word

is

= v. Vu

vV-u

n.

Vv +

(42)

uV.v. 1

(43)

necessary upon the matter of the interpretation

of such expressions as

Vwv,
The

rule followed in this

to the nearest term only.


1

By

book

That

Art. 69 the expressions v

V) v

'Vu X

Vw.v,

is

v.

that the operator

V applies

is,

an(i

V v are

be interpreted as

VECTOR ANALYSIS

158

V u v (V u~) v
V u v = (V u) v
V u x v = ( V u) x v.
If
it,

to be applied to

is

more than the one term which follows

the terms to which

it is applied are enclosed in a parenside of the above equations.


the
left-hand
upon
the
formulae
of
may be given most naturally
proofs

thesis as

The

by expanding the expressions in terms of three assumed unit


The sign 2 of summation will be found coni, j, k.

vectors

venient.

By means

of it the operators

V*

Ax

take the

form

The summation extends over

To demonstrate

Vx

#, y, z.

wv

2.
Hence

Vx(wv

To demonstrate

(u

v)

=v Vu +
.

u.

Vv + v

x (V x u)

u x

(V x

v).

THE DIFFERENTIAL CALCULUS OF VECTORS

159

'

Pa

9 x)

Now

5u

v*

2v.^
2v ^
-*

or

Pu
v
= vx(Vxu) + 2v..-

In like manner

^n
^

Bx

x (V x u)

v Vu.

= ux (V xv) + uVv.

V(nv) =

Hence

+
The

=v

->

v x

(V x

+ u x (V x v).

u)

other formulae are demonstrated in a similar manner.

71]

The notation 1

V(n.v) u

(44)

will be used to denote that in applying the operator

product (u

That

is,

v),

is

to be regarded as constant.

carried out only partially upon


In general if
is to be carried out

the operation

the product (u
partially

the quantity

V to the

is

v).

upon any number

of functions

which occur

after

a parenthesis, those functions which are constant for the


differentiations are written after the parenthesis as subscripts.

it in

Let

= MJ +
i

w2j

+ w 3 k,

This idea and notation of a partial


so to speak may be avoided by means
But a certain amount of compactness and simplicity is
u ' v)n ' s surely no more complicated than u
v or
(
thereby. The idea of

of the formula 41.


lost

X (V X

u).

VECTOR ANALYSIS

160

=U

M.V

then

VI

+ u 2 v% + u s v 3

f\

and

(u

v)

0*i v i

+ u* v* +

But

and

<:t
Hence

V(u

= v Vw +

v)

and

Vw +
2

(44)'

-y

Hence

v2

Vv 1 + w a Vi? a + M 8 V l
V(u.v) v = Vw + v 2 Vw 2 + v 3 V^ 3
V (u v) = V (u v) n + V (u v) v

V(tt.) m =

But

(45)

This formula corresponds to the following one in the notation of differentials

d (u v)

d (u

or

= d (u
v) = u

v) n

dv

+ d (u v) T
+ d u v.

The formulae (35)-(43) given above (Art. 73) may be


written in the following manner, as is obvious from analogy
with the corresponding formulae in differentials
:

V (M +
V-

Vx

(u

(n

v)

v)

= V (u + v~) u + V (u +
= V. (u + v) u + V. (u +

+ v) = V

x (u + v) u

+Vx

(u

).

( 35 )'

v) y

(36)'

v) v

(37)'

THE DIFFERENTIAL CALCULUS OF VECTORS

161

V O v) = V (u v\ + V (u v).

(38)'

V.(wv)= V.(wv) + V-(wv) v

(39)'

V x (u v) = V x (u v) + V x (w v) v
V (u. v) = V (u. v) u + V (u. v) y
V (u x v) = V (u x v) + V (n x v) v
V x (u x v) = V x (u x v) u + V x (u x v) v

(40)'

tt

(41)'

'

(42)'
(43)'

This notation
scalar product

In almost

formally.

where

it

Take

Vx

(u

it

can be done away without

for instance (43)'.

v) u

= (V

which occur in V.

may

v) u
is

Then

Expand V

(V

(u

v) u

term the

Hence

n.

Vv + uV

v.

There are a number of important relations in which

V (u

v) u figures.

(Vx v) = V(n.v) n -uVv,


V (u.v) u = u. Vv + u x (V x v),
u V v = V (u v) u + (V x v) x n.
u x

or

loss of

v,

in the last

be placed before the sign V.

the partial operation

or

u)

constant for the differ-

V x (u x v) u =
In like manner V x (n x v) v = v V u
Hence
Vx(uxv)=vVu v V u
75.]

was introduced.

for this reason it

must be understood that n

entiations

factor u

and

other cases

all

simplicity.

particularly useful in the case of the

is

uv

(46)
(46)'

(46)"

The proof of this relation may be given by expanding in


terms of i, j, k.
method of remembering the result easily

is

as follows.

Expand

the product

u x

(V x
11

v)

VECTOR ANALYSIS

162
formally as

if

V,

u,

v were

all real vectors.

ux(Vxv) = uvV

Then

V v.

The second term is capable of interpretation as it stands.


The first term, however, is not. The operator V has nothing
upon which to operate. It therefore must be transposed so
But u being outside
that it shall have u v as an operand.
constant
for the differenis
in
u
x
x
of the parenthesis
(V v)
Hence
tiations.

V = V (n v) n
- u V v.
v) = V (u v) u

u.v
u x

and
If

(V x

u be a unit vector, say


a

a,

V v = V (a

(46)

the formula

v) a

+ (V x

v) x a

(47)

v of a
expresses the fact that the directional derivative a
vector function v in the direction a is equal to the derivative
of the projection of the vector v in that direction plus the

vector product of the curl of v into the direction a.


Consider the values of v at two neighboring points.

and

v (x

dv

= v (x +

Let

But by

=v

= dv

dy,

d y,
z

v (,

+ d v z j + d V B k.

v2

dv

= V (d r

=dr Vv
.

y, z).

v3 k

dv = dr. (V v t i + Vt> 2 j +
dv

d z)

dz)

= drVv
dv z = dr Vfl 2
dv 3 = dr Vv 8

Hence
(46)"

dv 1

(25)'

Hence

d #, y

dx, y

dv

By

(a, y, z)

v) dr

Vv

k).

+ (V x

v) x d r.

(48)

THE DIFFERENTIAL CALCULUS OF VECTORS


Or

if

v denote the value of v at the point (#,

y, z)

163

and v the

value at a neighboring point

= v + V (d r

v) at

+ (V x

This expression of v in terms of

its

and the displacement d

r is

the

dels,

v)

dr.

(49)

value v at a given point,


analogous to the expan-

sion of a scalar functor of one variable by Taylor's theorem,

/ (a) =/(*o) +/'()<* *The


That

derivative of (r

is

V (r

For

v) v

VV
v

v)

when

if

9x

Vr = v

V 2*

v2 j

T~+
9
y
+

is

r)

equal to v.

v,

V dB 0~

^3 k

9z

= v,

= 0.
v) v = v.

instead of the finite vector

vector d r be substituted, the result

V (d r
By

constant

999

V 1l TT-

V (r

In like manner

is

V (r v) v = v.
= v V r - (V

V
Hence

v) v

r,

an infinitesimal

still is

= v.

= v + V(dr.v) dr + (V x v) x rfr
V(dr-v) = V(rfr.v) d + V^r-vV
V (d r v) dr = V (d r v) v.

(47)

>

Hence
Substituting

v=4v + ^V(rfr.v) + !(Vxv)xdr.

(50)

This gives another form of (49) which is sometimes more


convenient It is also slightly more symmetrical.

VECTOR ANALYSIS

164
*

Consider a moving

76.]

Let v

fluid.

(#,

yt

at the point (#, y, z) at the


velocity of the fluid
round a
(z , y^ z ) with a small sphere.

z,

f)

time

t.

be the
Sur-

point

dr d

At each

=v +

In the increment of time 8

V v.

the points of this sphere will have

dr

V v) 8

point at the center will have

The

the distance

8t.

distance between the center

interval 8

1.

moved

and the points that were

upon the sphere of radius d r at the

To

is

the distance

(v

The

point of this sphere the velocity

moved

= c2

commencement

has become at the end of that interval 8

find the locus of the extremity of

d r'

it is

of the

necessary to

eliminate d r from the equations

dr'

The

first

equation

= dr + dr Vv
c 2 = d i d r.

may

St,

be solved for d r by the method of

Art. 47, page 90, and the solution substituted into the second.

The

result will

show that the

infinitesimal sphere

has been transformed into an


ellipsoid by the motion of the
fluid during the time 8 1.

A
may

more

account of the change that has taken place


be obtained by
making use of equation (50)
definite

THE DIFFERENTIAL CALCULUS OF VECTORS

165

or of the equation (49)

v
v

= v + V(dr.v) dr +(Vx v)x

dr,

=v +
[V(dr.v) dr +|(Vxv)x dr]+|(Vx v)xdr.
The

first

term v in these equations expresses the fact that


is moving as a whole with an instan-

the infinitesimal sphere

taneous velocity equal to v


This
of the motion.
The last term
.

is

the translational element

|(Vxv)xdr = ^ curl v x
shows that the sphere

dr

undergoing a rotation about an

is

instantaneous axis in the direction of curl v and with an angular velocity equal in magnitude to one half the magnitude of
curl

v.

or

The middle term

V(dr.v)

rfr

-(Vx v) x dr

expresses the fact that the sphere is undergoing a deformation known as homogeneous strain by virtue of which it

becomes

ellipsoidal.

dx

For

this

term

V v + dy Vv
1

is

2 -f

equal to

dz

3,

v v vv vs be respectively the components of v in the directions i, j, k.


It is fairly obvious that at any given point
z
a
of three mutually perpendicular axes i, j, k
set
(#< yi o)

if

may

be chosen such that at that point Vvj,

Vv2 Vv3
,

are re-

VECTOR ANALYSIS

166

Then

to them.
spectively parallel

the

expression above

becomes simply

The point whose coordinates

referred to the center of the

infinitesimal sphere are

d x,

dy,

dz

will

endowed with this, velocity.


have moved to a new position

The

totality of the points

is

therefore

In the time 8

it

upon the sphere


-2

the ellipsoid of
goes over into the totality of points upon
is
which the equation
~2

/2

The statements made before (Art. 72) concerning the three


types of motion which an infinitesimal sphere of fluid may
possess have therefore

now been

The symbolic operator

demonstrated.

V may be

applied several times


succession.
This will correspond in a general way to
forming derivatives of an order higher than the first. The
77.]

in

will all be independexpressions found by thus repeating


ent of the axes because
itself is.
There are six of these

dels of the second order.

Let

The

(x, y, z)

derivative

be a scalar function of position in space.


is a vector function and hence has a curl

VV

and a divergence.

Therefore

V.VF,

VxVF

THE DIFFERENTIAL CALCULUS OF VECTORS


are the

two derivatives

of the second order

167

which may be

obtained from V.

V.VF=divVF
VxVF=curlVF.
The second expression V x V V vanishes identically.
the derivative of

any

V possesses no
V V in terms of

scalar function

be seen by expanding
x
the terms cancel out. Later (Art. 83)

may

it

will be

if

a vector function

(52)

That

curl.
i, j,

k.

is,

This
All

shown con-

W possesses ho curl,
V x W= curl W = then W = VF,

versely that

(51)

i. e. if

0,

is

the derivative of some scalar function F.

The
i, j,

first

Symbolically,

The

V VF

expression

when expanded

in terms of

k becomes

operator

Laplace.

V V=
V V is

C/

C/

Cf

therefore the well-known operator of

Laplace's Equation

=
c/

ic~

ci

y~

& z-

(53)

becomes in the notation here employed

V.VF=0.

(53)'

V V

When applied to a scalar function F the operator


yields
a scalar function which is, moreover, the divergence of the
derivative.

Let

flow f

is

be the temperature in a body. Let c be the conThe


ductivity, p the density, and k the specific heat.

VECTOR ANALYSIS

168

The

rate at

unit time

is

which heat

f.

is

leaving a point per unit volume per

The increment

of temperature is

=-.
p K

=-,v-vr.
pk

dt
This

is

Fourier's equation for the rate of change of tempera-

ture.

Let

ponents.

Tv Vv V its three comof


V
V
Laplace may be applied to V.
operator

be a vector function, and

The

(54)

a vector function

Laplace's Equation, each of


its three scalar components does.
Other dels of the second
order may be obtained by considering the divergence and curl
If

V V has a derivative
VV.V = VdivV.

The divergence

of V.

The

curl

(55)

V X V has in turn a divergence and a curl,

V-VxV, VxVxV.
V V x V = div curl V
V x V x V = curl curl V.

and

and

Of

satisfies

these expressions

V VxV

vanishes identically.

(56)
(57)

That

is,

the divergence of the curl of any vector is zero. This may be


seen by expanding
x V in terms of i, j, k. Later (Art.

V V

it

will be

shown conversely that

83)
vector function

if

the divergence of a

W vanishes identically,
V W = div W = then .W = V x V = curl V,
i. e. if

0,

is

the curl of

some vector function V.

THE DIFFERENTIAL CALCULUS OF VECTORS

If the expression

169

x (V x V) were expanded formally

according to the law of the triple vector product,

Vx(VxV)=VVV-V.VV.
The terms

V VV

is

the beginning so that

V be

meaningless until

it

transposed to

operates upon V.

VxVxV = VV.V-VVV,
curl curl V = VdivV

or

(58)

V V V.

(58)'

This formula is very important. It expresses the curl of the


curl of a vector in terms of the derivative of the divergence

and the operator of Laplace.

Should the vector function

satisfy Laplace's Equation,

VVV = Oand
V=V

curl curl

Should the divergence of

curl curl

div V.

be zero,

V=

Should the curl of the curl of

V VV.
vanish,

V div V = V V V.
To sum

There are six of the

up.

dels of the

V-VF, VxVF,
V-VV, VV-V, V-VxV,
Of

these,

two vanish

second order.

VxVxV.

identically.

VV=

0,

V V

x V

0.

A third may be expressed in terms of two others.


VxVxV = VV.V-V.VV.
The

operator

V V is equivalent to the

(58)

operator of Laplace.

VECTOR ANALYSIS

170
*

The geometric

78.]

interpretation of

V Vw is interesting.

upon a geometric interpretation of the second


u of the one scalar variable x.
Let it be required
at
the
u
Let u i be the value of
point x
to find the second derivative of u with respect to x at the
Let x l and x2 be two points equidistant from x
point x
It depends

derivative of a scalar function

That

is,

let

=x

= a.

Then

the ratio of the difference between the average of u at the


points x1 and x2 and the value of u at x to the square of the
is

distance of the points xv x2 from x

dja u

LiM
a==0

That

a*

easily proved by Taylor's theorem.


Let u be a scalar function of position in space. Choose
three mutually orthogonal lines i, j, k and evaluate the
is

expressions

Let
x

and x 1 be two points on the line i at a distance a from


4 and xs two points on j at the same distance a from
xt and #6 two points on k at the same distance a from x
o!

a;

Lai

_ LlM
a=0

_J

'-,

THE DIFFERENTIAL CALCULUS OF VECTORS

171

Add:
2

LDI r

a=OL
As V and V

are independent of the particular axes chosen,

this expression

then for

still

may

be evaluated for a different set of axes,

a different one,

these results

ii>\

-w= a =

Let n become

infinite

and

etc.

By

+ w2 +

at the

adding together

all

n terms

same time

let the different

from x

sets of axes point in every direction issuing

The

fraction

u 1 + u2 +

'

Qn

terms

6n
then approaches the average value of u upon the surface of a
Denote this
sphere of radius a surrounding the point x
.

by

V V u is equal to six times the


of the excess of

u on the

limit approached

by the

ratio

surface of a sphere above the value

at the center to the square of the radius of the sphere.

The

same reasoning held in case u is a vector function.


If u be the temperature of a body
(except for a
constant factor which depends upon the material of the

VVw

VECTOR ANALYSIS

172
body)

is

77).

If

equal to the rate of increase of temperature (Art.


positive the average temperature upon a

VVwis

small sphere is greater than the temperature at the center.


The center of the sphere is growing warmer. In the case
of a steady flow the temperature at the center

constant.

flow

must remain

condition for a steady

the

Evidently therefore

is

V V u = 0.
That

the temperature is a solution of Laplace's Equation.


u whether
Maxwell gave the name concentration to
is,

u be a

scalar or vector function.

V V
Consequently V V u may

be called the dispersion of the function u whether it be scalar


or vector.
The dispersion is proportional to the excess of
the average value of the function on an infinitesimal surface

In case u

above the value at the center.

The

the average is a vector average.


vector additions.

SUMMARY OF CHAPTER
If a vector r is a function of a scalar

r with respect to

is

a vector function

additions in

it

are

III
t

the derivative of

a vector quantity whose direction is


that of the tangent to the curve described
by the terminus

of r

is

and whose magnitude

is

equal to the rate of advance of

that terminus along the curve


per unit change of t. The
derivatives of the components of a vector are the
components
of the derivatives.
n

d r2
d^r_d r
TT~~dl** ~Tr J +
l

d n rz

Tr

f
*

combination of vectors or of vectors and scalars

may

be

differentiated just as in
ordinary scalar analysis except that
the differentiations must be
performed in situ.

THE DIFFERENTIAL CALCULUS OF VECTORS


d

db

+ .._,

_(..),_.
d

da

d (a

or

b)

=da

(S)

db

b,

(3)'

d(axb) = daxb + axdb,


and so

forth.

The

173

(4)'

differential of a unit vector is perpendicu-

lar to that vector.

The

derivative of a vector r with respect to the arc s of

the curve which the terminus of the vector describes

is

the unit tangent to the curves directed toward that part of the

curve along which

s is

supposed to increase.

-*

<>

The

derivative of t with respect to the arc a is a vector whose


direction is normal to the curve on the concave side and

whose magnitude

is

equal to the curvature of the curve.

dt
C

d*r

-d~s~d^'

The

tortuosity of a curve in space is the derivative of the


unit normal n to the osculating plane with respect to the

arc

s.

-^

= ^r =
x
as
d s* v
ds\ds ^_i.

The magnitude

).

C/

(ii)

of the tortuosity is

,
cL

j-4
341
cL s
d s

70

/T

ff
(.to

7
/"/
(*

oC

(13)

VECTOR ANALYSIS

174
If r

denote the position of a moving particle,

v the velocity,

the time,

the acceleration,

'--*

< 15>

==*

< 16>

dv

d2

be broken up into two components of


which one is parallel to the tangent and depends upon the
rate of change of the scalar velocity v of the particle in its

The

acceleration

path,

and

of

may

which the other

is

perpendicular to the tangent

and depends upon the velocity of the particle and the curvature of the path.
A s t + v z C.
(19)

Applications to the hodograph, in particular motion in a


circle, parabola, or under a central acceleration.
Application
to the proof of the theorem that the motion of a rigid body

one point of which is fixed is an instantaneous rotation about


an axis through the fixed point.
Integration with respect to a scalar
of differentiation.

is

merely the inverse


due to

Application to finding the paths

given accelerations.

The operator
applied to a scalar function of position in
space gives a vector whose direction is that of most rapid
increase of that function and whose
magnitude is equal to
the rate of that increase per unit
change of position in that
direction

< 22 >

THE DIFFERENTIAL CALCULUS OF VECTORS

is invariant of the axes


The operator
defined by the equation

i,

j,

k.

may be
(24)

VV.dr = dV.

or

It

175

(25)'

Computation of the derivative V V by two methods depending upon equations (21) and (25)'. Illustration of the occurrence of

may

V in mathematical physics.

be looked upon as a fictitious vector, a vector


It obeys the formal laws of vectors just in

differentiator.

so far as the scalar differentiators of 9/ 9

9 / 5 y, 9[ dz obey

#,

the formal laws of scalar quantities

If a

be a unit vector a

VV

is

the directional derivative of

in the direction a.

If

a.VF=(aV) F= a (VF).
V a vector function a V V is the directional
is

(30)
derivative

of that vector function in the direction a.

d x

=ix

+ J.

d X

+ *.
9 y

+ jx

(32)'

9 z

kx,
d
Z

(33)'

VECTOR ANALYSIS

176
Proof that

V V is

of V.

the divergence of

V x V, the curl

and

V V = div V,
V x V = curl V.
V (w 4- v) = V u + V0,

(35)

V (u + v) = V. u + V- v,
V x (u + v) = V x u + V x v,
v V u + w V v,
V (u
V (w v) = V w v + u V v,
Vx(wv) = Vwxv + wVxv,

(36)

(37)
(38)

v~)

(39)
(40)

V(u.v)=vVu + iiVv + vx (Vxu)


+ ux (Vxv),

- u.Vxv,
V- u-n -Vv + uV

V(uxv)=vVxn
Vx (u x v)=vVu

Introduction of the partial


ferentiations are

del,

V (u

v) n

in

(41)
(42)

v.

(43)

which the

dif-

performed upon the hypothesis that u

is

constant.

nx (Vxv)=V(u
If a

-nVv.

(46)

be a unit vector the directional derivative


a

The expansion
of a point

or

v) u

(a;,,,

Vv=V

(a

v) a

+ (V x

v)

a.

(47)

any vector function v in the neighborhood


y# ) at which it takes on the value of v is
of

=v +V

v)

= |v + V(^r.v)+^(Vxv)

(d r

v) dr

+ (V x

rfr,

(49)

dt.

(50)

Application to hydrodynamics.
The dels of the second order are six in number.

THE DIFFERENTIAL CALCULUS OF VECTORS

VxVF=curlVF=0,
T 2 F"

V V

is

The

place's Equation.

operator

VV

V=

3 2 F"

l^

(51)

V^VF=0, V satisfies

If

Laplace's operator.

(52)

)2V

V.vrdivVF-f^+f^
may be

V div V,

(55)
(7,

interpretation of

La-

applied to a vector.

V V x V = div curl V =
V x V x V = curl curl V = VV.V-V.VV.
The geometric

177

V V as giving

(56)
(58)

the disper-

sion of a function.

EXERCISES ON CHAPTER III


1.

Given a

particle

moving

in a plane curve.

Let the

plane be the ij-plane. Obtain the formulae for the components of the velocity parallel and perpendicular to the radius
vector r. These are
r-,

6 k

r,

where 6 is the angle the radius vector


is the normal to the plane.

makes with

i,

and k

Obtain the accelerations of the particle parallel and


perpendicular to the radius vector. These are
2.

(r_r0 2)-,
r

(r

0+

2r 0) k X-r

Express these formulae in the usual manner in terms of x

and

y.

12

VECTOR ANALYSIS

178

Obtain the accelerations of a moving particle parallel


and perpendicular to the tangent to the path and reduce the
3.

results to the usual form.


If r, <, 6 be a

4.

where r

system of polar coordinates in space,


the distance of a point from the origin, <f> the

is

meridianal angle, and 6 the polar angle ; obtain the expressions


for the components of the velocity and acceleration along the
radius vector, a meridian, and a parallel of latitude. Reduce
these expressions to the ordinary
5.

Show

By

is

V of a triple product

a function of

a =

where
7.

8.

d,

*e,

a;,

(r

f are

#, z in

case

r) r,

in Art. 63 that the

(r

for

computing

find

[a b c] each term of which

a)

e,

=r

t,

constant vectors.

Compute

V V Fwhen

Compute

equal to r and

z.

axes.

method given

the second

the derivative

in terms of #, y,

method suggested

the direct

operator V is independent of the


6.

form

VV

VV,

when

V is

Fis r 2
.

V,

and

r,

-,

or

r2

V x V x V when V

is

and show that in these


equal to -?
r3

cases the formula (58) holds.

VF

and V V x
Expand V x
show that they vanish (Art. 77).
9.

10.

Show by expanding

in terms of

in terms of

i, j,

i, j,

k that

Vx VxV=VV.V-V. VV.
11.

Prove

A.V(VW) = VA. VW+WA- VV,

and

(VxV)

W = Vx

(Vx W) w

k and

CHAPTER IV
THE INTEGRAL CALCULUS OF VECTORS
79.]

Let

space.

Let

C be any

(z, y, z)

be a vector function of position in

curve in space, and r the radius vector

drawn from some fixed

origin to the points of the curve.


Divide the curve into infinitesimal elements dr. From the

sum

of the scalar product of these elements

of the function

at

thus

The
in

limit of this

d r and the value

some point of the element

sum when

d r.

the elements dr become infinite

number, each approaching zero, is called the


and is written
along the curve
-

line integral of

dr.

If

and

dr=
fw
c

J
The

C
Jo

(1)

definition of the line integral therefore coincides

with

the definition usually given.


It is however necessary to
in
which
radius
direction
the
vector r is supposed to
specify
describe the curve during the integration.

d r have opposite signs when the curve

is

For the elements


described in oppo-

VECTOR ANALYSIS

180

C and

one method of description be denoted by

If

site directions.

the other by

(7,

/W 'dr =

JI o

-o

W*

dr.

In case the curve C is a closed curve bounding a portion of


surface the curve will always be regarded as described in
such a direction that the enclosed area appears positive
(Art. 25).

denote the force which

If f

may

point to point along the curve

when
r

its

point of application

of the curve

C to

work done by the force


moved from the initial point

(7,

is

be supposed to vary from

the

point r

its final

is

Cf.dr = Ff.
Jr
JO
Theorem

The

scalar function

line integral of

F(z,

y,

the line integral

dr.

the derivative

z)

r to the point r is
equal to the difference

of the function

That

V"(#,

y, z) at

F(r)

r*

dV= F(r) -

F(r.)

y,

-V

yw

O-

V F= d V

F(a,y,)

The

of the

line integral

F(* ,y

,<).

derivative

single valued scalar function of position


closed curve vanishes.

(2)

V F of

taken around a

The
is

r
o

Theorem,
ifi^

between the values

the point r and at the point r

- F(r ) = F(*
dr.

definition

is,

=
By

V V of

along any curve from the point

fact that the


integral is taken around a closed curve
denoted by writing a circle at the foot of the
integral sign.

To show
r

= 0.

(3)

THE INTEGRAL CALCULUS OF VECTORS


The

initial

point r and the final point r coincide.

F(r)

= F(r ).

Jo
if

Conversely

closed curve vanishes,

function

Hence

Hence by (2)
Theorem :

181

F (x, y, z)

the line integral of


about every
is the derivative of some scalar

of position in space.

Given

*/

= 0.

dr

W = V F.

To show

be any fixed point in space and r a variable point.

Let

The

line integral

f W-rfr
o

independent of the path of integration


paths G and G' be drawn between r and
is

consists of the path

G from

to r is a closed curve.

to r

For

G.
r.

let

any two

The curve which

and the path

G' from r

Hence by hypothesis

/F
v

G'

*J c'

-G'

Hence

iw*dr = fwdr.

c'

Hence the value of the integral is independent of the path


of integration and depends only upon the final point r.

VECTOR ANALYSIS

182

The value of the integral


the position of the point r

is

dr

f.

therefore a scalar function of

whose coordinates are

a:,

y,

z.

= V O, y, 2).

Let the integral be taken between two points infinitely near


together.

But by

V V* d r = d V.

definition

Hence

The theorem
80.]

Let

is

therefore demonstrated.

be the force which acts upon a unit mass near

under the influence of gravity.

the surface of the earth

a system of axes

k be chosen so that k

i, j,

The work done by

the force

moves from the position

fdr =

its

Let

Then

point of application

r to the position r is

Jf r

w = - g (z

Hence

when

vertical.

is

gkdr =
z )

= g (ZQ

f g dz.

/ r

z).

The force f is said to be derivable from a force-function V


when there exists a scalar function of position V such that
the force

Evidently

is

equal at each

F is one
to V any

if

point of the derivative VF".


may be obtained

force-function, another

by adding
arbitrary constant.
the
force-function
is
ample

V=w=g(z
Or more simply
The force is

V
f

-e).
z.

= VF= -

k.

In the above ex-

THE INTEGRAL CALCULUS OF VECTORS


The necessary and

V (x, y, z) exist, is
point of application

sufficient condition that a force-function

work done by the force when


moves around a closed circuit be zero.
that the

The work done by

the force

w=
If this

f'di.

if

the integral vanishes. The force-function and the


differ only by a constant.

V= w +

In case there

is

work done by
closed circuit

friction

friction

is

closed

= VF==VM;
f = VV

conversely

its

is

when taken around every

integral vanishes

contour

And

183

const.

no force-function can

when a

work done

particle is

exist.

For the

moved around

in a

never zero.

upon
by a fixed mass
a unit mass is directed toward the fixed mass and is proportional to the inverse square of the distance between the

The

force of attraction exerted

masses.
f

= - c Ms

r.

This

is

the law of universal gravitation as stated by Newton.

It is easy to see that this force is derivable

function V.

from a force-

Choose the origin of coordinates at the center


mass M. Then the work done is

of the attracting

w=

Jro
But

cM C

Jr

dr

M Tdr.

c -r-

rs

=r

r,

-cM

VECTOR ANALYSIS

184

a proper choice of units the constant c may be made


The force-function
may therefore be
equal to unity.

By

chosen as

If there had been several attracting bodies


the force-function would have been

M v Mv

y __
are the distances of the attracted unit
where r v r 2 r 3
mass from the attracting masses
s
v z
,

M M M

The law of the conservation of mechanical energy requires


work done by the forces when a point is moved

that the

This is on the assumpnone of the mechanical energy has been converted


The law of
into other forms of energy during the motion.

around a closed curve shall be zero.

tion that

conservation of energy therefore requires the forces to be


derivable from a force-function.
Conversely if a forcefunction exists the work done by the forces when a point is
carried around a closed curve is zero and consequently there

no loss of energy.
mechanical system for which a forcefunction exists is called a conservative system.
From the
is

example just cited above it is clear that bodies moving under


the law of universal gravitation form a conservative
system
at least so
long as they
81.]

Let

space.

Let

finitesimal

do not

collide.

W (x,
S

y, 2) be any vector function of position in


be any surface. Divide this surface into in-

These elements may be regarded as


and
be
plane
may
represented by infinitesimal vectors of
which the direction is at each
the direction of the
elements.

point

normal

to the surface at that


point

tude

equal to the

is

magnitude

and of which the magni-

of the area of the infinitesimal

THE INTEGRAL CALCULUS OF VECTORS


Let

element.

this infinitesimal vector

185

which represents the

element of surface in magnitude and direction be denoted by

d a.

Form

which

is

sum

the

the

sum

of the scalar products of the value of

each element of surface and the

at

surface.

The

sum when

limit of this

face approach zero


the surface S, and

""

of the integral

W= W

or

(d a

da,

i) i

and d& be excomponents parallel to i, j, k


If

scalar.

is

pressed in terms of their three

da

W over

written

is

//
The value

(vector) element of
the elements of sur-

called the surface integral of

is

z j

(d a

j)

z k,

(d a

k)

k,

= dy dz i + dzdx + dxdy k,
j

(5)

The

surface integral therefore has been defined as

is

cus-

tomary in ordinary analysis. It is however necessary to


determine with the greatest care which normal to the surface
da

is.

integral

That
is

is,

which

taken over.

side of the surface (so to

speak) the

For the normals upon the two sides

are the negatives of each other.

Hence the

surface integrals
taken over the two sides will differ in sign. In case the

upon as bounding a portion of space d a


considered
to be the exterior normal.
always
If f denote the flux of any substance the surface integral

surface be looked
is

f-i
8

VECTOR ANALYSIS

186

which is passing through


gives the amount of that substance
the surface per unit time. It was seen before (Art. 71) that
the rate at which matter was leaving a point per unit

volume per unit time was V f. The total amount of matwhich leaves a closed space bounded by a surface
per

ter

unit time

is

the ordinary triple integral

V*tdv.
Hence the very important

(6)

relation connecting a surface in-

tegral of a flux taken over a closed surface

and the volume

integral of the divergence of the flux taken over the space

enclosed by the surface

= fff

(7)

Written out in the notation of the ordinary calculus this


becomes

\f[Xdydz + Tdzdx + Zdxdy]

where X, F, Z are the three components of the flux f


The
theorem is perhaps still more familiar when each of the three
.

components

is

treated separately.

ff X dx d y =
This

is

known

fff *^ dxdy dz.

as Gauss's Theorem.

(8)'

It states that the surface

integral (taken over a closed surface) of the product of a


function
and the cosine of the angle which the exterior

normal to that surface makes with the X-axis


the

volume integral

is

equal to

of the partial derivative of that function

THE INTEGRAL CALCULUS OF VECTORS

187

with respect to x taken throughout the volume enclosed by


that surface.

If the surface

be the surface bounding an infinitesimal

sphere or cube

CC f.d& = V.f dv
where d v

is

the volume of that sphere or cube.

dv J J
This equation

may be taken

The divergence

f.

(9)

as a definition of the divergence

of a vector function f

equal to the
taken over a suris

by the surface integral of f


an
infinitesimal body divided by that volume
bounding

limit approached

face

Hence

when

the volume approaches zero as

From

this definition

which

its limit.

That

is

evidently independent of the


the properties of the divergence may be deduced. In
order to make use of this definition it is necessary to develop
at least the elements of the integral calculus of vectors before

axes

is

all

the differentiating operators can be treated.


of

is

consequently
than from a practical standpoint.

Theorem

82.]

function

is

The

This definition

interesting more from a

theoretical

surface integral of the curl of a vector

equal to the line integral of that vector function

taken around the closed curve bounding that surface.

W-rfa=
f/Vx
g

/ /

This

is

Cw-dr.
*J

(11)

the celebrated theorem of Stokes.

On

account of

its

great importance in all branches of mathematical physics a


number of different proofs will be given.

VECTOR ANALYSIS

188
First Proof

(Fig. 32).

Then by

Consider a small triangle 1 23 upon the surface


at the vertex 1 be

Let the value of

at
(50), Chap. III., the value

W = \\ W

+ V (W

B r)

any neighboring point

+ (V x W) x

8 r

is

where the symbol S r has been introduced for the sake of distinguishing it from d r which is to be used as the element of
integration.

The

integral of

W taken

around the triangle

FIG. 32.

Cw-di=l

t/A

fw
A

f (V x W) x Sr-dr.
A

*J

Thefirstterm

Cw

\ /
A

.dr

f dr
t/

vanishes because the


integral of d r around a closed figure, in
this case a small
The second term
triangle, is zero.

vanishes by virtue of
(3) page 180.

Hence

THE INTEGRAL CALCULUS OF VECTORS

l-

189

f
JA

Interchange the dot and the cross in this triple product

Jfwdr=4
A

When d r

is

/VxW'Srxdr.

/ A

equal to the side 1 2 of the triangle, B r

is

also

Hence the product

equal to this side.

BT x di
vanishes because 8 r and d r are collinear.

when di

is

In like manner

is the same side.7#, but taken


Hence the vector product vanishes.
side 23, S r is a line drawn from the vertex
to this side 23.
Hence the product 8 r x d r

the side 31, Sr

in the opposite direction.

When d r

is

1 at which

W= W

the

twice the area of the triangle.


Hence
positive area 123.
is

2$r x di

This area, moreover,

is

the

= d a,

where d a denotes the positive area of the triangular element


For the infinitesimal triangle therefore the
of surface.
relation

/
I

W*dr = V

JA

d*

holds.

Let the surface

For convenience

S be

let

divided into

elementary triangles.

the curve which bounds the surface

be made up of the sides of these triangles.

Perform the

integration

around each of these triangles and add the results together.

VECTOR ANALYSIS

190

The second member

2^x^

da

-8

is

the surface integral of the curl of

W.

In adding together the line integrals which occur in the

first

necessary to notice that all the sides of the elementary triangles except those which lie along the bounding
curve of the surface are traced twice in opposite directions.

member

Hence

it is

all

the terms in the

sum

from those sides of the triangles lying within the


cancel out, leaving in the sum only the terms
which arise from those sides which make up the bounding
which

arise

surface

Hence the sum reduces to the line


along the curve which bounds the surface S.

curve of the surface.


tegral of

Hence

in-

= 1 W.dT.

=/w
t/o
Second Proof

Let

C be any closed

contour drawn upon the surface


It will be assumed that
(Fig. 33).
is

continuous and does not cut

S
C

itself.

Let C' be another such contour near


to C.

Consider the variation S which

FIG. 33.

takes place in the line integral of


in passing from the contour C to the

contour

C".

THE INTEGRAL CALCULUS OF VECTORS

c
J C'

/r
W

dr

Hence

dr=

The expression d
The value of the

c
Jc

(W cr)= J

and

dr

c
I oW
J

S r) is

dr.

= d S r.

/WdSr=/d(W8r)

(W

191

by

form a perfect

its

dW *ST.

differential.

integral of that expression will therefore be


d r at the end and at
the difference between the values of

the beginning of the path of integration. In this case the


Hence
integral is taken around the closed contour C.
f*

//
Jc

and

But

dr

r
I

W*di= r SW dr
I

iw di=
d

C\

SWdr

I^W8r,
dW*Sil.

9W d
9W d
dx+
W = 9W
y +
dx
-=

-z

-=

ay

or

r,

W = 9W
dx
-=

dr

9W
-^

dy

dr

-^
<y

^
k d r,
z
^

9 W^

and

9 W^

2,

<y

VECTOR ANALYSIS

192

Substituting these values

/r

i-Sr-W-dr=t \-^--dr
ax
J 3x
+

similar terms in y

and

z.

But by (25) page 111


3

(3
i

W\

x d

dxJ

r)

= 3-= W

/C

or

In Fig. 33

3r

3x

and

similar terms in y

it

will be seen that

5r

-^

r.

Srxdr

W'di=J jix-^
+

dr

3x

the element of arc

is

and S r is the distance from the curve C to


along the curve
the curve C".
Hence S r x d r is equal to the area of an elementary parallelogram included between
That is

C and

C' upon the

surface S.

Br X di
S

= d&,

Cw*dr= TV

Wda.

# starting at a point in $ expand until


coincides with the contour bounding S.
The line integral
Let the curve

fW'di
will vary

from the value o

at the point

IW-dr
Jo

to the value

it

THE INTEGRAL CALCULUS OF VECTORS

193

taken around the contour which bounds the surface S.

This

sum

of the

total variation of the integral will

be equal to the

variations 8

Or
Stokes's theorem that the surface integral of the curl
of a vector function is equal to the line integral of the func83.]

tion taken along the closed curve

The converse

has been proved.

integral of a vector function

U is

which bounds the surface

is also true.

equal

If

to the line
integral

function W taken around the curve bounding the

the curl

f fu.rfa = Cw'di, thenTJ=Vx W.

(12)

for

Form
x W.

W)da = f Wdr -

ff

or

of

The

f Wrfr = 0,

the integration

is

performed be

-V

W)

da

= 0.

equation holds for any element of surface d

factor vanishes.

Hence

U-Vx
The converse

W=

0.

U = V x W.

Hence
is

and

integral reduces to merely a single term

(U
this

TT

(U-Vx W)-da = 0.

S over which

Let the surface


infinitesimal.

first

in space, then

the surface integral of the difference between

f C (TJ-Vx

As

all surfaces

That is

if

of the

surface and if

U is

this relation holds

W.

the surface

therefore demonstrated.
13

a,

the

VECTOR ANALYSIS

194

VxW

which

independent of the axes


be obtained by applying Stokes's theorem to an ini, j, k may
Consider a point P. Pass a plane
finitesimal plane area.
definition of

through
area d

P and

draw

in

it,

is

concentric with P, a small circle of

a.

Vx W-da =Jo
f Wdr.
When d a

has the same direction as

line integral will be a

between

maximum,

(13)

W the

value of the

for the cosine of the angle

and d a will be equal

to unity.

For

this

value of da,

r
Jo
Hence the

curl

VxW

of a vector function

W has

at each

point of space the direction of the normal to that plane in


which the line integral of
taken about a small circle conThe magcentric with the point in question is a maximum.

nitude of the curl at the point

is

equal to the magnitude of

that line integral of maximum value divided by the area of


the circle about which it is taken.
This definition like the

one given in Art. 81 for the divergence is interesting more


from theoretical than from practical considerations.
Stokes's theorem or rather its converse

may

be used to de-

duce Maxwell's equations of the electro-magnetic field in a


simple manner. Let E be the electric force, B the magnetic
induction, H the magnetic force, and C the flux of electricity
per unit area per unit time (i. e. the current density).
It is a fact learned from experiment that the total electromotive force around a closed circuit is equal to the negative
of the rate of

the circuit.

change of

The

total

magnetic induction through

total electromotive force is the line integral

of the electric force taken

around the

circuit.

That

is

THE INTEGRAL CALCULUS OF VECTORS


The

195

magnetic induction through the circuit is the surB taken over a surface

total

face integral of the magnetic induction

bounded by the

That

circuit.

is

Experiment therefore shows that

/d
E.rfr

or

E*dr=

Jo
Hence by the converse

x E

C C
J J a B.rfa,
I

at

B d a.

JJ a

of Stokes's theorem

curl

B,

B.

experiment that the work done in carrya


unit
positive
ing
magnetic pole around a closed circuit is
to
4?r
total electric flux through the circuit.
times
the
equal
It is also a fact of

The work done

in carrying a unit pole

the line integral of

The

electricity

total

flux

of

around the

around a circuit

circuit.

through the

surface integral of C taken over a surface


circuit.

That

That

circuit

Experiment therefore teaches that

= 4?r

IJ

is

the

bounded by the

is

rH.rfr

is

is

(NcZa.

VECTOR ANALYSIS

196

By

the converse of Stokes's theorem

V
With a proper

H = 4 TT C.

interpretation of the current C, as the dis-

placement current in addition to the conduction current,

an interpretation depending upon one of Maxwell's primary


hypotheses, this relation and the preceding one are the fundamental equations of Maxwell's theory, in the form used by
Heaviside and Hertz.

The theorems

of Stokes

and Gauss may be used

to

demon-

strate the identities.

V V
.

W=

div curl

0,

VxVF=0,

W=

0.

curlVF=0.

According to Gauss's theorem

fff V.Vx W<20 = ffvxW.da.


J s

JJJ

According to Stokes's theorem

CCC V

Hence

Apply

this to

the sphere
point

is

Wtd * = f W-dr.

V x W d? = Cw*dr.

an infinitesimal sphere.

closed.

Hence

its

and the integral around

V.VxWdv=

The

surface bounding

bounding curve reduces to a

it,

to zero.

fw.<fr = 0,
Jo

VV x W =

0.

THE INTEGRAL CALCULUS OF VECTORS


Again according

theorem

to Stokes's

ffvxVF.
da=
3

JJ
to

Apply this
bounding

Hence the

this surface is closed.

The curve

line integral of

VF" vanishes.

V
As

fvF-dr.
J

any infinitesimal portion of surface.

the derivative

197

this equation holds for

any d a,

it

follows that

Vx VF=0.
In

similar

demonstrated.

is

If the

everywhere

If the curl

is

84.]

U=

V x U of

By making

x W.

a vector function

the derivative of

between the
dels, viz.

U of a vector function
divergence
then U is the curl of some vector

zero,

W.

function

then

manner the converse theorems may be

some

is

everywhere zero,

scalar function

F",

use of the three fundamental relations

line, surface,

and volume

integrals,

and the

f VF.rfr = F(r)-F(r ),

(2)

''o

ffvxWrfa =

it is

possible to obtain a large

transformation of integrals.

fw.rfr,

number

(11)

of formulae for the

These formulae correspond to

VECTOR ANALYSIS

198

"
"
in ordinary
those connected with
integration by parts
both
sides of the
are
obtained
calculus.
by integrating
They
formulse, page 161, for differentiating.

First

Jc

i?)

= u V v + v V u.

I
wVvdr= J c vVu'dr.
V(uv)'di = Jo
I

Hence

The

(u

J/ uVvdr=[uv'] r
[u

expression

v~\

lvVudi.

"

(14)

ro

represents the difference between the value of (u v~) at r, the


end of the path, and the value at r , the beginning of the path.
If the path be closed

f wV vdi =

J
Second

ff V

Vx

(wv)

C vV udr.
Jo

(14)'

= %Vxv + Vwxv.

(uv)d*= C C

wVxvcU+

C C Vwxvrfa.

Hence

f uvdr
rfvwxv^a=
a
JQ

J J

JfJ

uV xvda,

(15)

or

rfwVxvda=
Third

But
Hence

fwvc?r

fTVwxv.^a,

Vx(>Vfl)^ttVxVv + VwxVtf.

VxVv =
v x (u V v) = V u x V

v,

(15)'

THE INTEGRAL CALCULUS OF VECTORS

199

(wvv)^a= cr^ux^v'dB..

Hence

V M x Vv

JI J s

da.= i

Jo

Fourth

liiv

(u v)

(uv)dv

uV v

=uV

dr =
v

+V

C vV u* dr,
Jo
u

rrruvvdv +

(16)

v.

r r

v^

<z

v.

Hence
u

III

V*vdv=[

uv da,

C i i ^U'vdv,

(17)

or

v^v dv=

In

f f \?

all

surface

u x v

wv^a

vVwxv

Fifth

Hence

rrTwVv^v,

= VXV^*v

da,

(17)'

Vwvxv.

C i C ^ u*\7 x

v dv.

(18)

these formulae which contain a triple integral the


is the closed surface bounding the body throughout

which the integration is performed.


Examples of integration by parts

like those

above can be

Only one more will be


given here. It is known as Green's Theorem and is perhaps
the most important of all. If u and v are any two scalar
multiplied almost without

functions of position,

limit.

VECTOR ANALYSIS

200

= /J

IV(V*)rf

/V^*V^i?=/

(v

V w) ^ v

u^'Vvdv,

TV

V V w,

V V w d 0.

Hence

/V^Vv^v =

w V?> d a

= r A>vwda
By
I

V Vvd

r f fvv-vwdv.

(i9>

subtracting these equalities the formula

(wVVv v^7^u)dv=

P,

(20)

(u^7 v

V w)

cZ

a.

obtained. By expanding the expression in terms of i, j, k


the ordinary form of Green's theorem may be obtained.
further generalization due to Thomson (Lord Kelvin) is the
is

following
/

lV*V*<fr=i

=
where

The

vwVU'd*

uw^7V'da>

[w

u^7[w^7v\dv,

V w]

c?

v,

(21)

a third scalar function of position.


element of volume dv has nothing to do with the scalar

t0 is

function v in these equations or in those that go before. The


use of v in these two different senses can hardly give rise to

any misunderstanding.
*

In the preceding articles the scalar and vector funcwhich have been subject to treatment have been sup-

85.]

tions

THE INTEGRAL CALCULUS OF VECTORS

201

posed to be continuous, single-valued, possessing derivatives


of the first two orders at every point of space under consideration.
When the functions are discontinuous or multiplevalued, or fail to possess derivatives of the first two orders
in certain regions of space, some caution must be exercised in

applying the results obtained.

Suppose for instance

VK =
The

ar

x*

line integral

Introducing polar coordinates


=*

***

xdy
*

Form

(1,

the line integral from the point (+1,0) to the point


Let one path be a semi0) along two different paths.

circle lying

above the X-axis

lying below that


first

path

axis.

and the

The value

other, a semicircle

of the integral along the

is

along the second path,

From

i r~"
I
d6

fj

ir.

not depend merely


upon the limits of integration, but upon the path chosen,
this it appears that the integral does

'

VECTOR ANALYSIS

2Q2

the negative of the value


the value along one path being
The integral around the circle which is a
along the other.
2 TT.
but is equal to
closed curve does not vanish,
were false
Art.
79
of
results
the
therefore
seem

It might
which
and that consequently the entire bottom of the work
This however is not so. The difficulty is
follows fell out.

that the function

F=tan

-i y
-

not single-valued. At the point (!,!)


function V takes on not only the value

is

F= tan

f r instance,

the

=^
4

but a whole series of values

where k

is

any positive or negative integer. Furthermore at


was included between the two semicircular

the origin, which

becomes wholly indepaths of integration, the function


terminate and fails to possess a derivative. It will be seen
therefore that the origin is a peculiar or singular point of the
function V.

If the

two paths of integration from

(1,0) had not included


would not have

(4- 1, 0)

to

the origin the values of the integral


In other words the value of the

differed.

integral around a closed curve

which

does not include the

origin vanishes as it should.

Inasmuch

as

the

origin

appears to be the point which


be considered as marked

vitiates the results obtained, let it

by an impassable

barrier.

Any

closed curve

C which

does

not contain the origin may be shrunk up or expanded at will ;


but a closed curve C which surrounds the origin cannot be
so distorted as no
longer to enclose that point without breaking its continuity. The curve C not surrounding the origin

THE INTEGRAL CALCULUS OF VECTORS


may
but

203

shrink up to nothing without a break in its continuity ;


can only shrink down and fit closer and closer about

the origin.

It cannot be

shrunk down to nothing.

It

must

always remain encircling the origin. The curve C is said to


be reducible ; C, irreducible. In case of the function F", then,
it is

true that the integral taken around any reducible circuit


but the integral around any irreducible circuit G
;

C vanishes

does not vanish.

Suppose next that


the points at which

tinuous

first

V is

V fails

any function whatsoever. Let all


to be continuous or to have con-

Then any circuit C


may be shrunk up

barriers.

be marked as impassable
which contains within it no

partial derivatives

to nothing and is said to be


but
a
which
circuit
contains one or more such
reducible;
points cannot be so shrunk up without breaking its continuity
and it is said to be irreducible. The theorem may then be

such point

stated: The line integral of the derivative


vanishes around any reducible circuit 0.

vanish around an irreducible circuit


circuit

C may be

irreducible

W
It

of any function
not

may or may

In case one irreducible

distorted so as to coincide with another

circuit

of

without passing through any of the


and without breaking its continuity,

singular points
the two circuits are said to be reconcilable and the values of

VV

the line integral of


about them are the same.
that
such
any closed curve C within
region

shrunk up
point of

it

may

be

without passing through any singular


and without breaking its continuity, that is, a

to nothing

region every closed curve in which is reducible, is said to be


All other regions are cyclic.
acyclic.
By means of a simple device any cyclic region may be renConsider, for instance, the region (Fig. 34) enclosed between the surface of a cylinder and the surface of a
cube which contains the cylinder and whose bases coincide

dered acyclic.

with those of the cylinder.

Such a region

is

realized in a

room

VECTOR ANALYSIS

204
in
is

which a column reaches from the


evident that this region

around the column


x

A circuit which

irreducible.

It

passes

cannot be contracted to

It

nothing without breaking its continuity. If


now a diaphragm be inserted reaching from

-v

is

floor to the ceiling.

is cyclic.

"^^

the surface of the cylinder or column to the


surface of the cube the region thus formed

bounded by the surface of the cylinder, the


surface of the cube, and the two sides of the
diaphragm

FIG. 84.

is

to

draw a

circuit

which

shall

Owing

acyclic.

to the inser-

no longer possible
pass completely around the cyl-

tion of the diaphragm

it is

Hence every closed cirthe diaphragm prevents it.


cuit which may be drawn in the region is reducible and the
inder

region

is acyclic.

In like manner any region

may be

rendered acyclic by

The bounding
inserting a sufficient number of diaphragms.
surfaces of the new region consist of the bounding surfaces of
the given cyclic region and the two faces of each diaphragm.
In acyclic regions or regions rendered acyclic by the foregoing device all the results contained in Arts. 79 et seq.

hold true.
true.

To

For cyclic regions they may or may not hold


enter further into these questions at this point is

unnecessary.

Indeed, even as

much

discussion as has been

given them already may be superfluous. For they are questions which do not concern vector methods any more than the
corresponding Cartesian ones.

They belong properly

to the

subject of integration itself, rather than to the particular


notation which may be employed in connection with it and
which is the primary object of exposition here. In this
respect these questions are similar to questions of rigor.

THE INTEGRAL CALCULUS OF VECTORS

205

The Potential

The Integrating Operators.

Hitherto there have been considered line, surface,


86.]
and volume integrals of functions both scalar and vector.
There exist, however, certain special volume integrals which,

owing

operators V,

V,

occurrence and
consideration.

is

with the differentiating


to
their
Vx,
especially frequent
great importance in physics, merit especial

to their intimate connection

and owing

Suppose that

a scalar function of the position in space of the point

For the sake of defmiteness

may

density of matter at the point (x 2 , yv 22 ).

body

V is

In those portions of space in which no


identically zero. In non-homogeneous dis-

constant.

V is

matter exists

tributions of matter

each point

The

be regarded as the
In a homogeneous

varies

from point to point; but at

has a definite value.

it

vector

drawn from any assumed


the point (# 2 yv z 2 ). Let

origin,

may

be used to designate

On Vi> *i)
be any other fixed point of space, represented by the vector
T1

drawn from the same


ra
is

Vv

origin.

y1

z1

Then

~ r = Oa - *i) + Oa - yi) J + Oa - *i)


i

the vector

0&2>

=x

i?

a)

<

drawn from the point (x v yv j) to the point


this vector occurs a large number of times

As

in the sections immediately following,


12

it

will be denoted

by

VECTOR ANALYSIS

206

The length

of r 12 is then r 12

and

will

be assumed to be

positive.
12

=Vr

r 12

ia

= V

O 2

Consider the triple integral

The
xv

integration

y?fl

zi

tt*3

performed with respect to the variables


k* with respect to the body of which V
is

represents the density (Fig. 35). During


the integration the point (x v y v z^) re-

mains
,

j,^

FIG. 35.

fixed.

The

integral

/ has

a definite

value at each definite point (x v y v Zj).


It is a function of that point.
The interpretation of this integral

is easy, if

V be regarded as the density of matter in space.


mass dm at (x v yv 2 ) is
dm = V (x v yv z 2 ) dx z dy^ dz% = Vdv.

the function

The element

of

The integral / is therefore the sum of the elements of mass


in a body, each divided by its distance from a fixed point
VII

3 19 ^1 J *

r dm
m

J r*
This

what

termed the potential at the point (x v y v


due to the body whose density is
is

The
in

is

limits of integration in the


integral
of two ways.
In the first

I may be looked

at

either

place they may be


regarded as coincident with the limits of the body of which
V is the density. This indeed might seem the most natural
set of limits.

On

the other

hand the

integral

/ may

be

THE INTEGRAL CALCULUS OF VECTORS

207

regarded as taken over all space. The value of the integral


is the same in both cases.
For when the limits are infinite

the function
vanishes identically at every point (# 2 , yv z 2 )
situated outside of the body and hence does not augment
the value of the integral at all.
It is found most convenient
to consider the limits as infinite

and the integral as extended

This saves the trouble of writing in special


limits for each particular case.
The function Fof itself then
over

all space.

practically determines the limits


tically at all points unoccupied

owing to its vanishing idenby matter.

The

operation of finding the potential is of such


that a special symbol, Pot, is used for it.
occurrence
frequent
87.]

Pot

V= ill
JJJ

d&z

f la

^y*i

d%y

(22)

The potential,
is read "the potential of V"
a function not of the variables x v yv z 2 with

The symbol
Pot V,

is

performed but of the point


fixed during the integration.
These

regard to which the integration


(*!

y\) 2 i)

is

The function
enter in the expression for r 12
therefore have different sets of variables.

variables

and Pot

which

is

It may be necessary to note that although V has hitherto


been regarded as the density of matter in space, such an

interpretation for

Whenever

of

it

V is

becomes

entirely too restricted for convenience.


necessary to form the integral

any scalar function V, no matter what

integral

is

V represents,

that

The reason for calling


even in cases in which it has

called the potential of V.

such an integral the potential


no connection with physical potential
according to the

same formal law

is

that

it

is

formed

as the true potential

and

VECTOR ANALYSIS

208

by virtue of that formation has certain simple rules


tion which other types of integrals do not possess.

of opera-

this idea the potential of a vector function

Pursuant to

a,
\
Wfrf.
\*"S> yy *27
*,

may

be written down.

/W (Xn,

2/o,

Za)

dx% dyz dz v

(23)

In this case the integral is the sum of vector quantities


and is consequently itself a vector. Thus the potential of a
vector function

tion in space.

If

a vector function, just as the potential


of a scalar function F'was seen to be a scalar function of posi-

W (z

2,

yv

is

W be resolved into three components


yv ) + Y (xv yv )
) = X (#
its

2,

Pot

W=

PotX + j

Pot

The

potential of a vector function

sum

of the potentials of its three

The

(#i y\t *i>)

taken

over

for instance,

all

z 2)

Y+ k Pot Z.

is

(24)

equal to the vector

7
components X, I Z.
",

potential of a scalar function

when and only when

k Z <> 2 yv

exists

a point

at

the integral

If,
space converges to a definite value.
were everywhere constant in space the in-

would become greater and greater without limit as


the limits of integration were extended farther and farther

tegral

out into space.

Evidently therefore

if

the potential

is

to exist

must approach zero as its limit as the point (#2 yv za)


recedes indefinitely. A few important sufficient conditions
for the

convergence of the potential may be obtained by


transforming to polar cob'rdinates. Let

THE INTEGRAL CALCULUS OF VECTORS


x

= r sin 6

r sin
z

dv

cos <,
sin

= r cos

= r 2 sin

rf

<f>,

0,

is

fixed for the integration

T*

aw0

Let the point (a^, y v 2j) which


be chosen at the origin. Then
r i2

209

d <.

=r

and the integral becomes

dV * =

fff

the

remains

d0

dfr (22)

d<j>.

V decrease so rapidly that the product

function

finite as r increases indefinitely, then the integral con-

verges as

For

dr

PotF= CCfVrsmO dr dO

or simply

If

fff

far as

distant regions of space are concerned.

the

let

= GO

= 00

me

8m0 dr d0

dr d0

r-R

=R
r

Hence the

triple integral

sphere of radius
tity) is less

= oo

than 4

as regions distant

(where
TT

taken over

is

all

space outside of a

supposed to be a large quan-

K /R, and consequently converges as far

from the origin are concerned.

VECTOR ANALYSIS

210

remain
If the function
weakly that the product

finite or if it become infinite so

Vr
remains finite when r approaches zero, then the integral converges
as far as regions near to the origin are concerned. For let

Vr<K
r=R

r=R

III ^ r s n & dr

d<f>

<

=Q

Kd r

dd

d<j>.

r=0
r

=R

C C C Kdr dO
Hence the

triple integral

dd>

taken over

all

space inside a sphere

R (where E is now supposed to be a small quantity)


than 4 TT K E and consequently converges as far as

of radius
is less

regions near to the origin which


concerned.

If at any point (x 2 , y2 ,
i.

e.

the point

is

the point (x v y v zj are

z 2 ) not coincident

z x ), the

with the origin,

becomes infinite so

(x v y x ,
function
weakly that the product of the value 0/V at a point near
(x 2 , y2
(x 2 , y2

z2)
z 2)

by the square

remains

of

the distance

Jinite as that distance

the integral converges as far as regions

are concerned.

The

near

of that point
approaches

to the

to

from

zero, then

point (x 2 , y 2 , z 2 )

proof of this statement is like those given

These three conditions for the convergence of the


integral Pot V are sufficient.
They are by no means necesThe integral may converge when they do not hold.
sary.

before.

however indispensable to know whether or not an integral


Unless the tests given above
show the convergence, more stringent ones must be resorted

It is

under discussion converges.


to.

Such, however, will not be discussed here.

They belong

to the theory of integration in general rather than to the

THE INTEGRAL CALCULUS OF VECTORS

211

theory of the integrating operator Pot. The discussion of


rethe convergence of the potential of a vector function
duces at once to that of its three components which are scalar

functions and
88.]

The

may

be treated as above.

potential

is

a function of the variables x v y^ z1


to the integration.
Let the

which are constant with respect

value of the potential at the point (x v y v zj be denoted by

The
is

first partial

derivative of the potential with respect to x 1

therefore

The value

of this limit

may be determined by

a simple

Consider

device

(Fig. 36).
the potential at the point

due to a certain body T. This


is the same as the potential at
the point

due to the same body T displaced in the negative direction by


the amount A x r For in finding the potential at a point P
due to a body T the absolute positions in space of the body

T and the

point

P are

immaterial.

It is

only their positions

which determines the value of the potenboth body and point be translated by the same

relative to each other


tial.

If

amount

same direction the value of the potential is unBut


now if T be displaced in the negative direction
changed.
by the amount A #, the value of V at each point of space is
in the

changed from
to

where

A xz = A x r

VECTOR ANALYSIS

212

Hence
[Pot

FOa,^,^)]^ + A*, ,M * = [Pot


LlM

Hence

[Pot

FO

H****.*.*-

[Pot

It will be

integration.

body
body

found convenient to introduce the limits of


Let the portion of space originally filled by the
; and let the portion filled by the

be denoted by

after its translation in the negative direction

the distance

x l be denoted by M'.

The

regions

Let the region common to both be


remainder of
be m; the remainder of M' m
overlap.

//*/* V
Av
* fr
_ err
wtffj
-4-

r
.

through

M and M'

and

let the

Then

THE INTEGRAL CALCULUS OF VECTORS


f
(

LlM

<

(*(*(*( VT

l^

f*
f* T^V/n
y^v y
_ rrr

A f2 ,y2 ,z92 \;
+Ag

JJJfi

f*

II

-I-

Tja

Az =

Aa;2 =o(

213

ni

JJJ M

ru

a
la

or,
12

A
A

a;

CCC

LIM

A*=0t/

+Ag

rrr

LIM

XQjJJx

*> - ^c^a^ y^

>

when A

Xj approaches zero as its limit the regions mand


which are at no point thicker than A a, approach zero ;
as a limit.
and
both approach

'

TO',

There are cases in which

this reversal of the order in

which the two limits

are taken gives incorrect results. This is a question of double limits and leads to
the mazes of modern mathematical rigor.
}

If the derivative of

Fis

to exist at the surface

bounding

T the values of the

V must

diminish continuously to zero upon the surface. If Fchanged


suddenly from a finite value within the surface to a zero value outside the derivative 5 VI 3 *i would not exist and the triple integral would be meaningless.

function

For the same reason


within the region T.

V is

supposed to be

finite

and continuous at every point

VECTOR ANALYSIS

214

be assumed that the region Tis finite and that


vanishes upon the surface bounding T

Then

if it

111

LlM

5.//L

v^fl

*"a

#2'

~ay

=v

12

'

Consequently the expression for the derivative of the potential

reduces to merely

(26)

5%e partial derivative of the potential of a scalar function


The derivative
of

the derivative

V of the potential of^fis equal to the potential


V V.
VPotF=PotVr:

(27)

This statement follows immediately from the former.


the

upon the left-hand

ables x v

V upon

yv

2j,

it

may

side applies to the

be written

the right-hand side

tion to the fact that

Then

it

Vj

To demonstrate

Vr

may be

set of vari-

written

sc

to call atten-

v yv

PotF= Pot V2 F!

V may be expanded

.SPotF

*L J

L.

t
V

As

In like manner the

applies to the variables

this identity

.3PotF

V.

is equal to the potential of the partial derivative of

zz of V.

(27)'

in terms of

THE INTEGRAL CALCULUS OF VECTORS


As

i,

j,

k are constant vectors they may be placed under

the sign of integration and the terms

by means

may

Then

be collected.

of (26)

The curl
function

215

VX

and

divergence

are equal respectively

of the potential of a vector

to the

potential of the curl

and

divergence of that function.

W = Pot V x W,
curl Pot W = Pot curl W
Vj Pot W = Pot V W,
div Pot W = Pot div W.

Vj x Pot
or

and

or

These relations may be proved in a manner analogous to the


above. It is even possible to go further and form the dels
of higher order

V V Pot F= Pot V VF,


V V Pot W = Pot V V W,

VV Pot W = Pot V V W,
VxV x Pot W= Pot Vx VxW.

(30)
(31)
(32)
(33)

The dels upon the left might have a subscript 1 attached to


show that the differentiations are performed with respect to
the variables x v y, z v and for a similar reason the dels upon
the right might have been written with a subscript 2. The
results of this article

may

be

summed up

Theorem: The differentiating operator


operator Pot are commutative.

as follows

V and the integrating

In the foregoing work it has been assumed that the


was
finite and that the function Fwas everywhere
region
finite and continuous inside of the
region T and moreover
*89.]

decreased so as to approach zero continuously at the surface


bounding that region. These restrictions are inconvenient

VECTOR ANALYSIS

216

and may be removed by making use of a surface integral.


The derivative of the potential was obtained (page 213) in
form

essentially the

PPotF
otF
9 xl

sv
-M

CCC
= 111
J J J

r 12 *>x z

Lra

LIM

9
Let d a be a directed element of the surface
region M. The element of volume
therefore equal to

Hence

_L f f f
A* J J J

in the region

^('

the
r

m>

dv z

S bounding

12

= r r v<jx*
J J

The element

of

T 12

volume d v z in the region

///
ill

rlenco

'

-7

A a;, J J J m

xz

TT'f'v

m is equal to

a.

&v
v ti

<*

is j

A Vn

r 12

Consequently
1

9V
-r

r lz dx^

r r

dvi+l
J J

a r 12

i.rfa.

(34)

is

THE INTEGRAL CALCULUS OF VECTORS

217

The volume

with
integral is taken throughout the region
the understanding that the value of the derivative of
at

the surface
derivative
of

M.

shall be equal to the limit of the value of that

when

the surface

approached from the interior

is

This convention avoids the difficulty that arises in

connection with the existence of the derivative at the surface

S where V

becomes discontinuous.

The

surface integral

is

taken over the surface S which bounds the region.


becomes infinite. By virtue of
Suppose that the region

the conditions imposed

V to

upon

insure the convergence of

the potential

<

K.

Let the bounding surface S be a sphere of radius


tity which is large.
i

d& <

r2

dO

-B,

a quan-

d<f>.

The surface integral becomes smaller and smaller and


proaches

zero as its limits

when

the region

M becomes

ap-

infinite.

Moreover the volume integral

rev wn
tn

remains

V is

finite as

M becomes

\s IP

infinite.

such a function that Pot

dv,
Consequently provided

V exists as

far as the infinite

regions of space are concerned, then the equation

SPotF =
^

Pot

9V

holds as far as those regions of space are concerned.


ceases to be continuous or becomes infinite
Suppose that

at a single point (xv y v Zj) within the region T.

Surround

VECTOR ANALYSIS

218
this point

with a small sphere of radius

surface of this

within the sphere.

-- =

3PotF
~

9*i

By

sphere and

all

Let

JR.

the region

S denote

T not

the

included

Then

r r r 1 5F dv + C If
2
/ /
J Js
J J J*ru9xs
I

the conditions imposed

upon

-i.

r 12

Vr<K
f f

i.rfa

ff K dd

<

Consequently when the sphere of radius

becomes smaller

and smaller the surface integral may or may not become


Moreover the volume integral
f* f* f*

J J JM
may

or

may not approach

and smaller.

PF

r !2

5x Z
when

a limit

zero.

becomes smaller

Hence the equation

PPotF
_

l-*f\+

^
dx
l

J.

5F
_

\Jv

dx%

has not always a definite meaning at a point of the region


T at which becomes infinite in such a manner that the

product
If,

Fr

remains

however,

finite.

F remains

that the product

Fr

the point in question so


is zero
approaches zero, the constant
finite at

and the surface integral becomes smaller and smaller


approaches zero. Moreover the volume integral

///.

as

THE INTEGRAL CALCULUS OF VECTORS


approaches a definite limit as

becomes

infinitesimal.

219

Con-

sequently the equation

5PotFPot5T
holds in the neighborhood of all isolated points at which
remains finite even though it be discontinuous.

becomes infinite at some single point


Suppose that
z
coincident
with (a^, y, Zj). According to the
( 2 y 2 2 ) not
,

conditions laid upon

<

VI*
where

near to

is
it.

K,

the distance of the point (xv yv


Then the surface integral

2)

from a point

r !2

need not become zero and consequently the equation

PPotF
^

= Pot =5V

O #j

C'S'a

need not hold for any point (x^ yv z^) of the region.
z in such a manner that
if V becomes infinite at #
2 yv 2

But

VI

<K,

then the surface integral will approach zero as


the equation will hold.
Finally suppose the function
surface

In

S bounding

the region

V remains

jP,

its

finite

limit

and

upon the

but does not vanish there.

this case there exists a surface of discontinuities of V.

Within

this

surface

V is

finite

without,

surface integral
i

da

it is

zero.

The

VECTOR ANALYSIS

220

Hence the equation

does not vanish in general.

9V
PPotF =
Pot 5
jr

d xl

dx%

cannot hold.

may be

Similar reasoning

applied to each of the three

with respect to x v
partial derivatives
the results it is seen that in general

Vj

ff

PotF= Pot V2 F +

By combining

yv z r

J-

d&.

(35)

V be any function in
F exists. Surround

space, and let it be granted that


each point of space at which
Let the surface of the
ceases to be finite by a small sphere.
sphere be denoted by S. Draw in space all those surfaces

Let

Pot

which are surfaces of discontinuity of V. Let these surThen the formula (35) holds
faces also be denoted by 8.
where the surface integral is taken over all the surfaces
which have been designated by S.
If the integral taken
over

all

these surfaces vanishes

when

the radii of the spheres

above mentioned become infinitesimal, then

V PotF=PotV F

(27)'

This formula

V PotF=PotV

will surely hold at

F.

a point (x 1? y x z x ) if
remains always
at a point (x^ y2 z 3 ) so that the
,

finite or becomes infinite

product

remains

discontinuity

and

finite,

and

if

possesses no surfaces of

if furthermore the product

as r becomes infinite.

Vr

remains finite

In other cases special tests must be


applied to ascertain whether the formula (27) can be used
or the more complicated one
(35) must be resorted to.
'

For extensions and modifications of

this theorem, see exercises.

THE INTEGRAL CALCULUS OF VECTORS

221

The

relation (27) is so simple and so amenable to transformation that


will in general be assumed to be such a
function that (27) holds. In cases in which
possesses a

S of

frequently found convenient


replaced by another function V which has
in general the same values as F'but which instead of possessing a discontinuity at S merely changes very rapidly from
surface

to consider

one value

discontinuity

it is

V as

(#2 y v 2 2 ) passes from one


Such a device renders the potential
of V simpler to treat analytically and probably conforms to
actual physical states more closely than the more exact
side of

to another as the point

to the other.

conception of a surface of discontinuity. This device practically amounts to including the surface integral in the

symbol Pot V V.
In fact from the standpoint of pure mathematics it is
better to state that where there exist surfaces at which the
function

V becomes

discontinuous, the full value of Pot

VV

should always be understood as including the surface integral

'

in addition to the

12

volume integral

VF

fff
In like manner Pot

W,

Pot

12

VX

W, New

V W and other

similar expressions to be met in the future must be regarded


as consisting not only of a volume integral but of a surface
possesses
integral in addition, whenever the vector function

a surface of discontinuities.

convention in the interpretation of


formulse which permits such simple formulae as (27) to hold
in general, and which gives to the treatment of the integratIt is

precisely this

ing operators an elegance of treatment otherwise unobtainable.

VECTOR ANALYSIS

222

The

irregularities

which may

thrown into the

arise are

inter-

pretation, not into the analytic appearance of the formulae.


This is the essence of Professor Gibbs's method of treatment.
90.]

The

first

partial derivatives of the potential

may

also

be obtained by differentiating under the sign of integration. 1

"

V)

gotr

*- *2

In

like

manner

for a vector function

W
'

c>Potr

\jr

&X 1

= III

and

^
g

V Pot F=

But
1

If

i (a?
a

-ajj)

dv v

(38)'

12

+j

^^H-

(y a

-yj) + k

^ +k

- Zj ) = r 12

an attempt were made to obtain the second


partial derivatives in the same
it would be seen that the volume
integrals no longer converged.

manner,

THE INTEGRAL CALCULUS OF VECTORS

V Pot F= fff Tjf~ * *r

Hence

223

( 39)

In like manner

VxPotW= III

dv v

"^"3

V.PotW= JfJfJf ^f
r 12

and

(40)

dv *

(41 )

These three integrals obtained from the potential by the


differentiating operators are of great importance in mathe-

Each has

matical physics.

its

own

interpretation.

Conse-

quently although obtained so simply from the potential each


is given a separate name.
Moreover inasmuch as these
integrals

may

exist even

when

the potential

is

divergent,

they must be considered independent of it.


They are to
be looked upon as three new integrating operators defined
each upon its own merits as the potential was defined.
Let, therefore,

(42)
'

r 12

12

W (#
"

2>

'

AO
r 12

y 2> z 2 ) Cv
^ 30n

'

r 3 12

If

the potential exists,

n Cv %n
m Cv *r/ m
m

dx z dy z dz z = Ma,x.Vf.

first is

written

(44)

then

V Pot F= New F
V X Pot W = Lap W
V Pot W = Max W.
The

New F and

(45)

read " The Newtonian of F."

VECTOR ANALYSIS

224

reason for calling this integral the Newtonian is that if


represent the density of a body the integral gives the force

The

of attraction at the point (x v y v z^) due to the body. This


The second is written Lap
and
will be proved later.
"
This integral was used to a
the Laplacian of W."
read
considerable extent by Laplace. It is of frequent occurrence

and magnetism. If
represent the current
C in space the Laplacian of C gives the magnetic force at the
point (x v y v zj due to the current. The third is written
in electricity

Max

W and read " the Maxwellian of W."

used by Maxwell.

and magnetism.

It,

This integral was

too, occurs frequently in electricity

For instance

if
represent the intensity
the
Maxwellian
of
I gives the magnetic
I,
potential at the point (xv yv z-^) due to the magnetization.
To show that the Newtonian gives the force of attraction

of magnetization

according to the law of the inverse square of the distance.


Let dmz be any element of mass situated at the point
(#2*

y-K>

zz)'

The

force at (xv yv

z^)

due to

dm is equal to

d mn
12

in

magnitude and has the direction of the vector r12 from the

point (x v y v zj to the point (xv yv 2 2 ).

Hence the

force

is

12

Integrating over the entire body, or over all space according


to the convention here adopted, the total force is

where

V denotes the

density of matter.

THE INTEGRAL CALCULUS OF VECTORS


The

integral

may

be expanded in terms of

i,

j,

225

k,

12

The

three components

may be

expressed in terms of the po-

tential (if it exists) as

(42)'

It is in this

form that the Newtonian

is

generally found in

books.

To show that the Laplacian gives the magnetic force per


unit positive pole at the point (x v y v 2j) due to a distribution
z
of electric flux.
The magnetic force at (xv yv z x )
( 2 2/ 2 , 2 )

due to an element of current d C 2

is

equal in magnitude to

the magnitude d O z of that element of current divided by the


square of the distance r 12 ; that is

The

element of current

The

perpendicular both to the vector


and to the line r 12 joining the points.

direction of the force

dC 2

is

direction of the force

vector product of r 12

is

and dG z

therefore the direction of the

The

12

15

force

is

therefore

VECTOR ANALYSIS

226
Integrating over
the point (x v y v

all space,

zj

the total magnetic force acting at

upon a unit positive pole

is

.3

12

may be expanded

This integral

WO
'12

The

i,

j,

za)

2 , 2/ 2 ,

!>

k components

-Xfyv

of

(2/2

Lap

in terms of

+ J Y (x v

*a)

2/2

/i)

i, j,

k.

2/2'

Let
^2)

+ (a - *

W are respectively
(43)'

(200

k Lap

W=

In terms of the potential

Lap

(if

W=

one exists) this

Pot

9x l

To show

that

if

be writt

9z l

= 5PotF
-^

point (x v

9 Pot

may

dy l

I be the intensity of magnetization at the

^2, z 2 ), that

is, if

equal to the magnetic

I be a vector

moment

whose magnitude

is

per unit volume and whose

THE INTEGRAL CALCULUS OF VECTORS

227

the direction of magnetization of the element d v2


from south pole to north pole, then the Maxwellian of I is the
direction

is

magnetic potential due to the distribution of magnetization.


The magnetic moment of the element of volume d v% is I d vz
.

The potential at (% v y v Zj) due to this element is equal to its


magnetic moment divided by the square of the distance r 12
and multiplied by the cosine of the angle between the direction of magnetization I

and the vector

r la

The

potential

is

therefore

r 12

Integrating, the total magnetic potential


'i 2

'

1
->?,,,

is

= Max

seen to be

I.

*>

12

This integral

may

also be written out in terms of x, y,

z.

Let

JA +
If instead of

xv
oil

Vy

22

xv yv

z1

tne variables

(2/2

the variables x, y, z; and instead of


l
the expression takes
f, ?/, ^ be used

the form given by Maxwell.

Max

/// {<a<f-) + i(f-f)+ C(S-z)

^dv.

According to the notation employed for the Laplacian

Max

W - f f f (X

~ y } Y+ fa ~ ''

(44)'

Maxwell

Electricity

and Magnetism, Vol.

II. p. 9.

VECTOR ANALYSIS

228

The Maxwellian
It

may

of a vector function

is

a scalar quantity.

be written in terms of the potential

Max

W=

-- + c>Potr
5- +
-

This form of expression is


upon mathematical physics.

much used

(if it exists) as

---

(44)"

in ordinary treatises

The Newtonian, Laplacian, and Maxwellian, however, should


not be associated indissolubly with the particular physical
They should be looked
interpretations given to them above.
which may be applied, as the
of
functions
position in space. The Newany
and yields a vector
scalar
function
to
a
applied

as integrating operators

upon

potential

tonian

is

is,

to

The Laplacian

is applied to a vector function


a
of
the
same sort. The Maxwellian
function
yields
and yields a scalar function.
a
function
to
vector
applied

function.

and
is

Moreover, these integrals should not be looked upon as the


derivatives of the potential.

If

the potential exists they


exist when the

But they frequently

are its derivatives.

potential fails to converge.

be such functions that their potentials


Let V and
and have in general definite values. Then by (27) and

9L]
exist

(29)

VVPotF=V.PotVF=Pot V.VF.
But by (45)
V Pot V = New F,
and
V.Pot VF=Max W.
Hence
V. V PotF = V NewF = Max VF
= Pot V.VF
By (27) and (29)
But by

(45)

and by

(45)

(46)

VV .Pot W = VPot V. W = Pot V V. W.

W = Max W,
V Pot V W = New V W.
V

Pot

THE INTEGRAL CALCULUS OF VECTORS

W = V Max W = New V W
= Pot VV W
V x V x Pot W = V x Pot V x W

VV

Hence

By

(28)

But by (45)
and

Hence

By

(56),

229

(47)

= Pot V x V x W.
V x Pot W = Lap W,
V x Pot V x W = Lap V x W.
V x Pot W = V x Lap W = Lap V x W
= Pot V x V x W.
(48)

V V x Pot W = 0,
V Pot V x W =
V Lap W = Max V x W = 0.
(49)
Chap. III. V x V PotF= 0,
VxPotVF=0.
V x New V = Lap V V = 0.
(50)
Chap. III. V x V x W = VV- W- V- VW,

Chap.

or

Pot

III.

0.

Hence

And

by (52),

or

Hence

And by
Hence
or

(58),

V.VW = VV.W-VxVxW.
V V W = New V W - Lap V x W,
V V W = V Max W V x Lap W.

(51)

These formulae may be written out in terms of curl and


div

if

desired.

Thus
div

New V = Max

V F,

V Max W = New div W


curl Lap W = Lap curl W
div Lap W = Max curl W =
curl New V = Lap V V =
V V W = New div W - Lap curl W.

(46)'

(47)'

(48)'

(49)'
(50)'
(51)'

VECTOR ANALYSIS

230

Poisson's

Let

92.]

Equation

any function in space such that

"be

the potential

PotF
has in general a

Then

definite value.

V.VPotF=-47rF,
c>

PotF

or

P2

dx{
This equation

PotF

02
dy-f

22

(52)

PotF
02 = -47rF.

9 9f

known

as Poisson's Equation.
has
been defined as the potential
integral which

The

is

solution of Poisson's Equation.

Pot F= Vj

The

proof

is

is

as follows.

New F= Max V2 ^=
12

The

and # have been attached to designate


what are variables with respect to which the differen-

subscripts 1

clearly

tiations are performed.

= Vj NewF= C C
But

and

F V2

= - V2

Vj

C^7 l --

=V

1-

V F+ F V V
2

THE INTEGRAL CALCULUS OF VECTORS


Hence

-V
1

or

Vt1

V2 F = T V

v2 K

'12

Integrate

V
22

i - V /F V

1
I-

231

12

V V

But

That

is

Hence

to say

satisfies

V V
x

Pot

= 0.
*H

Laplace's Equation.

F= T

T TVj -- V2

= f f F YI
J Jj

And by (8)

T< v 2

(53)

<*.
r 12

The

surface integral is taken over the surface which bounds


This is
region of integration of the volume integral.
taken " over all space." Hence the surface integral must be
taken over a sphere of radius R, a large quantity, and
must

the

be allowed to increase without limit.

At

the point (x v y^z^),

however, the integrand of the surface integral becomes


finite owing to the presence of the term

12

in-

VECTOR ANALYSIS

232

S must

include not only the surface of the


of a sphere of radius
sphere of radius R> but also the surface
the point (x v y v z^) and
jK', a small quantity, surrounding

Hence

the surface

must be allowed

As

to

approach zero as

'

its limit.

has been assumed that the potential of F" exists, it is


assumed that the conditions given (Art. 87) for the existence
it

That

of the potential hold.

F"r 3

<

is

K, when r

Vr < K, when

is

is

large

small.

polar coordinates with the origin at the point


y v Zj). Then r 12 becomes simply r

Introduce
(xv

and

Then

= -V

for the large sphere of radius

V,

r 12

r8

--.da =

Hence the surface


as its limit.
For

Hence when

Hence the

sin0d0 d&.

jrr

infinite the surface integral

large sphere approaches zero as


For the small sphere

da
r la

integral over that sphere approaches zero

^ becomes

V,

its

limit.

= --r3 r 2 sin 6
r

integral over that sphere

becomes

Fsin0 de

d<f>.

dA.

over the

THE INTEGRAL CALCULUS OF VECTORS

V be

Let
(#!,

yv

2X )

233

supposed to be finite and continuous at the point


which has been selected as origin. Then for the

surface integral

is

practically constant

and equal to

its

value

(ajj,

y lt

!>

at the point in question.

f fssin 6 de

- f /Vsi
sin Q

Hence

when

<

=4

TT.

the radius R' of the sphere of integration approaches

zero as

its limit.

Hence

V- VPotF=-47rF.

and
In

like

manner

if

is

(52)

a vector function which has in

Poisgeneral a definite potential, then that potential satisfies


son's Equation.

V V Pot W = - 4

TT

W.

(52)'

W into

The proof

of this consists in resolving

ponents.

For each component the equation

its

three com-

holds.

V VPot^r=-47r X,
V VPotF=-47r F,
V V Pot Z - 4 Z.
TT

Consequently

V- VPot(JTi+

Fj + ^k)

= -47r(Xi+

Fj

Let

VECTOR ANALYSIS

234

Theorem

IfVandW are

such functions of position in space

that their potentials exist in general, then for all points at which

and

are finite and continuous those potentials

satisfy

Poisson's Equation,

V. VPot F=-47rF,
V V Pot W = - 4 TT W.
The

modifications in this theorem

points at

V and W become

which

(52)
(52)'

which are

to be

made

at

discontinuous will not be

taken up here.
It

93.]

was seen

(46) Art. 91 that

V- VPotF=V.NewF=Max VF.

V New V = - 4 V
MaxVF= 4?rF.

Hence

TT

or

In a similar manner

Hence
or

it

(53)

was seen (51) Art. 91 that

V V Pot W = V Max W - V x Lap W


= New V W Lap V x W.
V Max W V x Lap W = 4 W,
New V W Lap V x W = 4 W.
TT

By

?r

virtue of this equality

W=

-r

4-7T

and

is

LapVxW

1
-r

4?r

W =2

Lap

New V

4?T

NewV.W.

(55)

2,

1
W = 4?r
Lap curl W
-.

W=

(54)'

divided into two parts.

W = Wj + W

Let

where

(54)

New

div

W.

(56)

(57)

THE INTEGRAL CALCULUS OF VECTORS

235

Equation (55) states that any vector function


multiplied
4
TT is equal to the difference of the
by
Laplacian of its curl

and the Newtonian of

W,

its

V.LapV

4?r

Furthermore

divergence.

W = -rV-VxLapW,.
4?r

But the divergence

of the curl of a vector function is zero.

Hence

V.W = divW =

x Wo

But

-r

4?r

(58)

V x New V

-r

VxV Max W

4-n-

the curl of the derivative of a scalar function

Hence

2.

is zero.

= curl W 2 = 0.

(59)

which has a potential


Consequently any vector function
divided
into
two
which
one has no divergence
be
of
parts
may

and

of

which the other has no

two such parts

W into

This division of

curl.

is

unique.
In case a vector function has no potential but both
and divergence possess potentials, the vector function

its

curl

may be

divided into three parts of which the first has no divergence


the second, no curl; the third, neither divergence nor curl.

As

LapVxW-^-NewVW + W

W=

Let

(55)'

before

4?r

V
1

and
4-7T

Lap V x

V V

Vx

W=

V Pot V

W=

x Pot

4-7T

V x New V W = ^4?r V

The divergence
part of

W=

of the first part

W are therefore zero.

0.

and the curl of the second

VECTOR ANALYSIS

236

=
VxVxPotVxW
JLvxLapVxW
47T
47T
- VV Pot V x

T- V V

x W.

VV

Pot

4-7T

V X W = 4-7T V Pot V V X

W=

-:

Hence
4?r

Hence
47T
curl of

VPot

is

equal to the curl of the

which

V x W = V X W.

V x Lap VxW = VxW = VxWr

into

0,

V V x W = 0.

for

The

Pot

4?r

4?r

is

7T

divided.

Lap

Hence

as the second part has no

Moreover

V New VW = VW = VW

Thus the divergence of


the second part
-

4"7T

which

part

VxW

curl, the third part can have none.

4?r

first

is

New

equal to the divergence of

W.

is divided.
Hence as the first part has no
the
third
can
have
none. Consequently the third
divergence
has
neither
curl nor divergence.
This proves the
part
8
statement.

into

By means of Art. 96 it may be seen that any function 3


which possesses neither curl nor divergence, must either

THE INTEGRAL CALCULUS OF VECTORS

237

vanish throughout all space or must not become zero at


In physics functions generally vanish at infinity.
infinity.

Hence functions which represent actual phenomena may be


divided into two parts, of which one has no divergence and
the other no curl.

A vector function the

divergence of which
vanishes at every point of space is said to be solenoidal.
vector function the curl of which vanishes at every point of
space is said to be irrotational.
94.]

Definition

In general a vector function is neither solenoidal nor irrotaBut it has been shown that any vector function which
possesses a potential may be divided in one and only one
tional.

W W

way into two parts


the other irrotational.

ls

of

which one

solenoidal

is

The following theorems may be

and

stated.

been proved in the foregoing sections.


With respect to a solenoidal function Wj, the operators

They have

all

That

are inverse operators.

4?r

Lap

Applied
tors

to

gives zero.

is

irrotational function

That

respect to

an

_ JL New V

New

either

(60)
of

these opera-

= 0.

irrotational function

are inverse operators.

= Wr

is

and

4-7T

4 7T

or curl

Lap
With

Wj = V x ?4?r Lap

Vx
an

Lap and

4?r

That

or

2,

the operators

div

is

=-V

~ New W
4:

7T

= W2

(62)

VECTOR ANALYSIS

238

With

respect to

V
and

V the operators

a scalar function

div and

or

- New,

4?T

also

Max and

~r

4-7T

That

are inverse operators.

is

=F

(63)

4-7T

-J-MaxVT=F.

and

4?r

JFtYA respect to a solenoidal function

Pot

awo*

Wj

tf^e

operators

or curl curl

4?r

That

are inverse operators.

Pot

4-7T

fF^A

x Wj

respect to

an

=V

is

Pot

That

are inverse operators.

- Pot

With

respect

VV W
.

to

Pot
7T

irrotational function

aw(i

4"7T

the operators

is

= - VV

Pot

= W2

any scalar or vector function V,

Pot and

4?r

mwrse

operators.

That

(64)

VV

operators

are

Wi = W r

is

V V

(65)

the

THE INTEGRAL CALCULUS OF VECTORS

_ -Pot V- VF=- V. V
4?r

and

Pot

V VW=-V V

respect to

PotF=F

Pot

4-7T

With

4?r

47T

a solenoidal function

239

Wj

W = W. (66)

the differentiating

operators of the second order

V V

- V V Wj = V

and

are equivalent

With

respect to

an

Wr
irrotational function W
V

(67)

2 the differentiat-

ing operators of the second order

V V
That

are equivalent.

VV

is

VVW
By

and

= VV

(68)

2.

integrating the equations

47rF=- V.NewF
4 TT W = V x Lap W
V Max W

and

by means
4

TT

of the potential integral Pot

Pot

F= - Pot V

New F= - Max New

W = Pot V x Lap W
Pot W = Lap Lap W

47T Pot

TT

(69)

- Pot V Max W
New Max W.

(70)

Hence for scalar functions and irrotational vector functions

New Max

4?r
is

an operator which

is

functions the operator

equivalent to Pot.
..

7T

Lap Lap

For solenoidal

vector

VECTOR ANALYSIS

240

For any

gives the potential.

the

the potential

vector function the first operator

irrotational

of
potential of the solenoidal part.
*95.] There are a number of double
gives

the second, the

part;

volume integrals which


are of such frequent occurrence in mathematical physics as

to merit a passing mention, although the theory of them will


not be developed to any considerable extent. These double
They are not scalar funcintegrals are all scalar quantities.
tions of position in space.

The

They have but a

integrations in the expressions

convenience as extended over

may

all space.

single value.

be considered for

The

functions by

vanishing identically outside of certain finite limits determine for all practical purposes the limits of integration in
case they are finite.

Given two

scalar functions

of position in space.
?7,
or potential product, as it may be called,
of the two functions is the sextuple integral

The mutual potential

Pot

One

(71)
of the integrations

Pot

( U,

F)

may be performed

=fff PX*!,
(* 2 ,

yv

yi, *i)

*2)

Pot

Pot

Vd Vl

Udv2

(72)

In a similar manner the mutual potential or


potential product
of two vector functions
is

W, W"

(71)'

This

is

ried out

also a scalar quantity.

One

integration

may

be car-

THE INTEGRAL CALCULUS OF VECTORS


Pot (W, W")

The mutual Laplacian


vector functions W',

W"

'0*i' 2/i' *i)

or

'

Laplacian

Pot

"

dv i

of

product

of position in space

is

241

two

the sextuple

integral

Lap(W',W")

=ffffffw

'

(*i ?!

i)-^

w" (*

y* 2 2)

rfi ^^2(73)

One

integration

may

Lap (W, W")

be performed.

=f f fw" (x v yv

Lap

,)

Wd

vz
(74)

=fffw

'

The Newtonian product


function

fa* y

^i)

LaP

"

of a scalar function

W of position in space

is

dv

F",

and a vector

the sextuple integral

(75)

By

performing one integration


vz

(76)

In like manner the Maxwellian product of a vector function

and a

scalar function

of position in

space

is

the

integral

Max (W,F) =
(77)
16

VECTOR ANALYSIS

242

One

integration yields

Max(W, r)

=
(78)

By

(53) Art. 93.

U PotF= - (V New *7) PotF.


V [NewZ7PotF] =(V New U) Pot F + (New Z7) V Pot F.
- (V New U) PotF= - V [New Z7PotF]+ New U> New V.
4 TT

Integrate

47r

f f CuPotYdv=:- f f fv- [Ne

+ f f fNewZ7 NewF
4-7T

Pot

(Z7,

v.

F) = T r fNewZ
-

The surface

ci

integral

is

f f Pot F New CT d a.
.

(79)

to be taken over the entire surface

bounding the region of integration of the volume integral.


As this region of integration is " all space," the surface S may
be looked upon as the surface of a large sphere of radius E.
If the functions

U and V vanish identically for all points

side of certain finite limits, the surface integral

must

out-

vanish.

Hence
4

By

TT

Pot

U,

F)

= f f fNew U New Vdv.

(54) Art. 93,

47r

W". PotW

= V x Lap W" Pot W


- V Max W" Pot W'.

(79)'

THE INTEGRAL CALCULUS OF VECTORS

243

W] = Pot W V x Lap W"


- Lap W" V x Pot W',
and V [Max W" Pot W] = Pot W V Max W"
+ Max W" V Pot W.
Hence V x Lap W" Pot W = V [Lap W" x Pot W]

But

[Lap

W" x

Pot

+ Lap W"

V Max W"

and

Hence
4

TT

Pot W'

Lap

W,

=V

[Max W" Pot


Max W" Max W.

W]

substituting:

W"

Pot

W = Lap W
+V

[Lap

-V-

Lap

W + Max W Max W"

W" X

[Max W"

Pot

Pot

W]

W].

Integrating:

TT

Pot (W, W")

= f f fLap W'. Lap W'dv


+ f f fMaxW'MaxW"rfv

(80)

- f f PotW' x Lap W" da- f f Max W" Pot WW a.


If

now W' and

W"

exist only in finite space these surface

integrals taken over a large sphere of radius

R must

vanish

and then
4

TT

Pot (W', W")

=ff /Lap W'

Lap W" d v

+ f f |*Max W' Max W" d v.


*
96.]

(80)'

There are a number of useful theorems of a function-

theoretic nature

which may perhaps be mentioned here owing

VECTOR ANALYSIS

244

the integral calculus of


to their intimate connection with
of them will in some instances be given
The
vectors.
proofs

and in some

The theorems

not.

are often useful in practical

of vector analysis to physics as well as in purely


applications

mathematical work.

Theorem : If

V (x,

y, z)

be a scalar function of position

VV

in space which possesses in general a definite derivative


and if in any portion of space, finite or infinite but necessarily

continuous, that derivative vanishes, then the function


constant throughout that portion of space.

is

VF=0.

Given

F= const.

To show

By

Choose a fixed point (xv y v zj in the region.


180

xt

*u

**

V V* d r = Ffa

y, *)

- F fa,

yv

(2) page

*,)

Vi, *i

But
Hence

F(#,

Theorem

If

y, z)

= F (xv yv Zj) = const.

(#, y, 2)

be a scalar function of position

VV

'

in space

which possesses

in general a definite derivative

the divergence of that derivative exists and is zero throughout any region of space, 1 finite or infinite but necessarily
if

continuous

and

at every point of

if

furthermore the derivative

any

finite

volume or

surface in that region or bounding


vanishes throughout all that region

of
it,

V F vanishes

portion of
then the derivative

any

finite

and the function

F re-

duces to a constant by the preceding theorem.


The term throughout any region of space
boundaries of the region as well as the region
1

must be regarded as including the


itself.

THE INTEGRAL CALCULUS OF VECTORS

V VF"=

Given

VF=

and

F= const.

VF vanishes for the portion of surface S,

constant in S.

upon one side of

Fis certainly
S and in the

Suppose
were not constant.
The derivative V V upon
side of S has in the main the direction of the normal to

region T,
this

for a region T,

for a finite portion of surface S.

To show
Since

245

that,

Consider a sphere which lies for the most


outer
side of S but which projects a little
the
part upon
the
surface
The surface integral of VF" over
S.
through

the surface S.

the small portion of the sphere which projects through the


surface S cannot be zero. For, as
is in the main normal

VV

to $, it

must be nearly

of spherical surface

parallel to the

normal to the portion

under consideration.

VF.

Hence the terms

rfa

in the surface integral all have the same sign and cannot
cancel each other out. The surface integral of
over

VF

that portion of S which


face vanishes because

intercepted by the spherical surConsequently the surface

is

V F is zero.

integral of

V F taken over the entire surface

segment which projects through

But

S is

of the spherical

not zero.

f f VF.<a = f f fv VFd0 =

Hence

f fvF'da =

0.

0.

is not
appears that the supposition that
constant upon one side of S leads to results which contradict

It therefore

V VF=

the given relation


The supposition must there0.
fore have been incorrect and
must be constant not only in
S but in all portions of space near to $ in the region T. By

VECTOR ANALYSIS

246

an extension of the reasoning

is

seen to be constant

throughout the entire region T.


Theorem : If
(x, y, z) be a scalar function of position in
and if throughspace possessing in general a derivative

VV

out a certain region 1

T of

space, finite or infinite,

or discontinuous, the divergence


exists

and

is

zero,

and

a constant value

and V(x,

if

V VV

continuous

of that derivative

V possesses

furthermore the function

in all the surfaces

bounding the region

when

the point (x, y, z)


approaches
recedes to infinity, then throughout the entire region T the
function V has the same constant value c and the derivative
c

y, z)

as a limit

VF vanishes.
The proof does not

in the case of the last theorem.

eralized as follows

Tlieorem

from the one given


The theorem may be gen-

differ essentially

If V(x, y, z) be

any scalar function of position

in space possessing in general a derivative

V V;

if

U (#, y, z)

be any other scalar function of position which is either positive or negative throughout and upon the boundaries of a
region T, finite or infinite, continuous or discontinuous;
the divergence
and
[
F] of the product of

exists

and
to

and

upon

is

all

if

VF

and upon the boundaries of T


furthermore V be constant and equal

zero throughout

is

at infinity

function

UV

and

if

the boundaries of

and

at infinity

then the

constant throughout the entire region

is

equal to c.
Theorem : If

F (x,

y, z)

be any scalar function of position

in space possessing in
general a derivative

out any region

T and

V F;

if

through-

of space, finite or infinite, continuous or

V VF

of this derivative exists


discontinuous, the divergence
and is zero ; and if in all the bounding surfaces of the region

the normal

component of the derivative

at infinite distances in
1

The region

(if

V F vanishes

and

such there be) the product

includes

its

boundaries.

THE INTEGRAL CALCULUS OF VECTORS


rz 9

V/ 9

247

where r denotes the distance measured

r vanishes,

from any fixed origin

then throughout the entire region T


and in each continuous portion

VF vanishes

the derivative

T V is

constant, although for different continuous portions


this constant may not be the same.
of

This theorem may be generalized as the preceding one


was by the substitution of the relation V ( U V F) = for

V-VF=Oand Ur 9V/9r =
z

As

for r 2 9V/9r

= 0.

the foregoing theorems the following


be made. The language is not so precise

corollaries of

statements

may

as in the theorems themselves,

when they

stood

V U = V F,

If

but will perhaps be under-

are borne in mind.

then

and

differ

most by a

at

constant.

VV?7=V.VF

If

differs

If

V only by a

from

constant at most.

VVZ7= V.VF and

surfaces of the region

and

thereto), then at all points


If

equal and
zero, then

if

(if

in all the

if

from

Fare equal

F only by a
and

"

the bounding

bounding surfaces
V U and V F are

normal components of

and

all

the region extend

and Fare equal.

at infinite distances r z (9
27

U differs

U= V in

if

at infinity
7"

V V 7 = V V F and

of the region the

and

if
VZ7 = VFin any finite
V U = V V at all points and V

and

surface $, then

portion of

U /9 r

V/9 r)

is

at all points of the region

constant.

two vector functions of position


in space which in general possess curls and divergences ; if
Theorem

If

'

are

for any region T, finite or infinite but necessarily continuous,


and the divergence
the curl of W' is equal to the curl of
of W' is equal to the divergence of W"; and if moreover

W"

two functions W' and W"


every point of any finite volume
the

in

T or

bounding

of the region T.

it;

then

W'

is

are equal to each other at


in

T or of

equal to

any

finite

W" at

surface

every point

VECTOR ANALYSIS

248

Since

W = V x W", V x (W - W") =

whose curl vanishes

tor function

is

0.

vec-

equal to the derivative

VF=W'-

W".
Then V V V
owing to the equality of the divergences.
The theorem therefore becomes a corollary of a preceding one.

of a scalar function

Let

(page 197).

W' and W"

two vector functions of position which in general possess definite curls and divergences
2
if
region T, finite but not
throughout any aperiphractic
Theorem

If

are

necessarily continuous, the curl of

W'

equal to the curl of


W" and the divergence of W' is equal to the divergence of
W"; and if furthermore in all the bounding surfaces of the
is

W"

are equal;
region T the tangential components W' and
then W' is equal to W" throughout the aperiphractic region T.
Theorem,: If W' and
are two vector functions of posi-

W"

which in general possess

tion in space

definite

curls

and

if
;
throughout any acyclic region T, finite but not
the curl of W' is equal to the curl
continuous,
necessarily
W" and the divergence of W' is equal to the divergence of

divergences

if in all the
bounding surfaces of the region T the
normal components of W' and W" are equal then the functions W' and W" are equal throughout the region acyclic T.
The proofs of these two theorems are carried out by means

W"; and

of the device suggested before.

W' and W"

Theorem: If
that

V V
-

W'

and

V V W"

are

two vector functions such

have in general

definite values

in a certain region T, finite or infinite, continuous or discon-

tinuous

and

in all the bounding surfaces of the region


the
functions W' and
are equal ; then W'
infinity
equal to
throughout the entire region T.
The proof is given by treating separately the three com;

if

W"

and at
is

W"

ponents of

tic.

W' and W".


T

The

region
may hare to be made acyclic by the insertion of diaphragms.
region which encloses within itself another region is said to be periphrac-

If

it

encloses no region

it is

aperiphractic.

THE INTEGRAL CALCULUS OF VECTORS

249

SUMMARY OF CHAPTER IV
The

line integral of a vector function

W along a curve C

is

defined as

C
J
The

w<Zr=f
[Widx + Wtdy + Wtdz].
/ C

line integral of the derivative

V V of

(1)

a scalar function

along a curve C from r to r is equal to the difference


at the points r and r and hence the
between the values of

line integral

taken around a closed curve

is

zero

and con-

the line integral of a vector function


taken
versely
around any closed curve vanishes, then
is the derivative
if

V V of some scalar function

V.

= F(r)-F(r )

(2)

fvF.rfr
.rfr =
Jo
and

f W dr = 0,

if

Jo

Illustration of the

The

is

then

(3)

W = VF.

theorem by application to mechanics.

surface integral of a vector function

W over a surface

denned as

f f

d&= C C [W

Gauss's Theorem
tion taken over a

The

dy dz + W^dz dx +

dx

dy].

surface integral of a vector funcis equal to the volume

closed surface

integral of the divergence of that function taken throughout


the volume enclosed by that surface

ff

=
JV W dv C C W

da,

(7)

VECTOR ANALYSIS

250

'

'

* ~

dz

[X dy

'

3y

+ Y dz dx + Z dx

dy~],

(8)

^ be

the three components of the vector function W.


Stokes's Theorem: The surface integral of the curl of a
vector function taken over any surface is equal to the line
if

X,

integral of the function taken around the line bounding the


surface.
And conversely if the surface integral of a vector

function

taken over any surface is equal to the line integral


taken around the boundary, then U is the

TJ

of a function

curl of

W.

ff V x
JJ a
and

if

da

JJ JJ

f Wrfr,
W-da=Jo

=J

dr, then

(11)

U = V X W.

(12)

Application of the theorem of Stokes to deducing the


equations of the electro-magnetic field from two experimental
facts due to Faraday.
Application of the theorems of Stokes
and Gauss to the proof that the divergence of the curl of

a vector function
a scalar function

is

zero and the curl of the derivative of

is zero.

Formulae analogous to integration by parts

x v

Vv
'

d&

dT

[u v"fr

-Tv^U'dr,

(14)

=fo uv ff*-Jj^VXYrfl,(M)
'

dr,

(16)

THE INTEGRAL CALCULUS OF VECTORS


I

fw V

v dv

wv

d&

251

I^U'vdv,

(17)

f f VM x vrfa = - f f TV**- V x vdv.

(18)

Green's Theorem:

IV

Vv

dv

= If uVvd&

= CCv Vw da
C C C(uV'Vv

iu V

V v dv

f Cv V Vudv,

vV^Vw) dv=

MwVv

(19)

Vw)^a. (20)

Kelvin's generalization:

wVu*Vvdv=

uw*7vd&

= I fvw Vu*da,

Cv V

The integrating operator known

u* V[wVv]

[w VM] dv.

as the potential

is

c?v

(21)

defined

by the equation

Pot

Pot

V =f f f
J J J

W-

V(X* Vv

Zz)

a;2y2

dx z dy, dz v

(22)

<**2 ^2/2 ^^2-

VPotF=PotVF;
V x Pot W = Pot V x W,
V Pot W = Pot V W,
.

V V Pot V= Pot V

Vr,

(23)

(27)
(28)

(29)
(30)

VECTOR ANALYSIS

252

V V Pot W = Pot V V W,
VV Pot W = Pot
W,
V x V x Pot W = Pot V x V x W.

The

(31)
(32)
(33)

integrating operator Pot and the differentiating operator

V are

commutative.

The

three additional integrating operators


Newtonian, the Laplacian, and the Maxwellian.
r i2

New F=
T

W=

<**2

VT

TIT

Lap

*2)

i/2'

dy, dz2

2x j

-a

known

^12

as the

(42)

^^2 * i r

/-Af>\

( 43 )

MaxW =
If the potential exists these integrals are related to it as fol-

lows:

V Pot F= New F,
V x Pot W = Lap W;
V Pot W = Max W.
The interpretation of the physical meaning

(45)

of the

Newtonian

on the assumption that F is the density of an attracting


is electric
body, of the Laplacian on the assumption that
of
the
Maxwellian
on
the
that
is the
flux,
assumption

intensity of magnetization. The expression of these integrals


or their components in terms of #, y, z ; formulae (42)', (43)',

(44)'

and

(42)", (43)", (44)".

V New F= Max V F,
V Max W = New V W,
V x Lap W = Lap V x W,

(46)
(47)

(48)

THE INTEGRAL CALCULUS OF VECTORS

Lap

W = Max V x W =

x New

0,

(49)

V = Lap V V- 0,

(50)

V V W = New V W Lap V x W
= V Max W V x Lap W.
The

potential

and

(51)

That

a solution of Poisson's Equation.

is

(52)

V V Pot F= - 4

(52)'

TT

W.

(53)

IT

Lap

-.

4-7T

Hence

divided

is

is,

V- VPotF = -47rF,

F=^V.NewF,
4

W=

253

into

and the other

New V W.

-.

(55)

7T

two parts of which one

is

provided the potential


In case the potential does not exist a third term
exists.
3
must be added of which both the divergence and the curl
solenoidal

irrotational,

A list

vanish.

which follow immediately from


(55) and which state that certain

of theorems

equations (52), (52)', (53),


integrating operators are inverse to certain differentiating
Let
be a scalar function, W, a solenoidal vector
operators.

function, and

Lap

an irrotational vector function.

New
47T

Lap

= Vx

= 0, V
=-V

LapWj =

4 7T

Then

Wr

(60)

New W2 = W2

(61)

(62)

VECTOR ANALYSIS

254

V.

-^NewF= V
(63)

Max
47T

--

Pot

47T

VF =

Wi = V x V

F.

x Pot Wj

_-Lp
t VV.W = - V.V--PotW = W
4-7T
4
2

(64)

(65)

7T

_i Pot V
4-Tr

4-7T

_ -Pot V V W = - V V 1
4?r
1

Pot

4?r

-V.VW = VxVxW
V V W = VV W
1

4
4

TT

Pot

TT

Pot

W = W.

(67)

(68)

V = - Max New V

W = Lap Lap W

(66)

(69)

New Max W.

(70)

Mutual potentials Newtonians, Laplacians, and Maxwellians

may be formed. They are sextuple integrals. The integrations cannot all be performed immediately ; but the first three
may be. Formulae (71) to (80) inclusive deal with these integrals.

The chapter

of theorems of

closes with the enunciation of a

a function-theoretic nature.

these theorems certain facts

number

By means of

concerning functions

may

be

inferred from the conditions that satisfy Laplace's equation


and have certain boundary conditions.

Among
tion.

the exercises

The work done in

number 6
the text

is

worthy of especial atten-

has for the most part assumed

But many of the formulae connecting


Newtonians, Laplacians, and Maxwellians hold when the potential does not exist. These are taken
up in Exercise 6 referred to.
that the potential exists.

THE INTEGRAL CALCULUS OF VECTORS

255

EXEKCISES ON CHAPTER IV
I.

If F" is a scalar function of position in space the line

integral

/><"
is

Show

a vector quantity.

r
Jo
That

c c
/

is

that

c r
J tjs

i/ 5

the line integral of a scalar function around a


is equal to the skew surface integral of the deriv-

closed curve

ative of the function taken over

the contour of the curve.

Show

any surface spanned into

further that

if

V is

constant

the integral around any closed curve is zero and conversely


if the integral around any closed curve is zero the function V
is

constant.

Hint
it

Instead of treating the integral as it stands multiply


by an arbitrary constant unit vector and thus

(with a dot)

reduce
2.

it

If

to the line integral of a vector function.

is

a vector function the line integral

~JccW
a vector quantity. It
If c
of the function W.

is

X di

be called the skew line integral


any constant vector, show that if

may
is

the integral be taken around a closed curve

The

first

four exercises are taken from Foppl'a Einfiihrung in die


where they are worked out.

well'sche Theorie der Electricitat

Max-

VECTOR ANALYSIS

256

and

=o

|J j s V

W d a - J j V (W

a)

is taken over a plane curve and the


the portion of plane included by the curve

In case the integral


surface

is

=
Show
when

ff

-V

that the integral taken over a plane curve vanishes


is constant and conversely if the integral over any
must be constant.
vanishes
curve
plane
is
3. The surface integral of a scalar function

=
This

is

a vector quantity.

Show

that the surface integral

taken over any closed surface is equal to the volume


integral of VF" taken throughout the volume bounded by

of

that surface.

That

is

Hence conclude that the surface integral over a closed surif V be constant and
conversely if the surface
over
closed
surface
vanishes the function V must
integral
any
face vanishes

be constant.
4.

If

W be a vector function, the surface integral


=

may

I da.

be called the skew surface integral.

quantity.

Show

that the

skew

It

is

a vector

surface integral of a vector

THE INTEGRAL CALCULUS OF VECTORS


function taken over a closed surface
integral of the vector function

bounded by the

That

surface.

C C d&x

is

257

equal to the volume

taken throughout the volume


is

W=

x Wdv.

Hence conclude that the skew surface integral taken over


is an
any surface in space vanishes when and only when
That is, when and only when the line
irrotational function.

5.

for every closed circuit vanishes.


Obtain some formulae for these integrals which are

integral of

analogous to integrating by parts.


6. The work in the text assumes for the most part that the
potentials of

V and W exist. Many of

the relations, however,

may be demonstrated without that assumption. Assume that


the Newtonian, the Laplacian, the Maxwellian exist.
For
simplicity in writing let
12

-r
r 12

Then

New V

=fff ^i Pu ^(aj r y v

*,)

Lap

W =ViPu X W (x v y v

Max

W -Vtfu W

(x v

7/2,

d vv

a a)

a)

dv v

dv v

(81)

(82)

(83)

(84)

17

VECTOR ANALYSIS

258
exercise

By
It

<

can be shown that

if

F is

such a function that

New V

exists, then this surface integral taken over a large sphere of


radius R and a small sphere of radius R' approaches zero

when
small.

becomes indefinitely great; and R', indefinitely

Hence

NewF=PotVF.

or

(85)

Prove in a similar manner that

W = Pot V x W,
Max W = Pot V W.

Lap

By means

(86)
(87)

of (85), (86), (87) it is possible to prove that

x Lap

W = Lap V x W,

VNew F=Max VF,


V Max W = New V

W.

Then prove
/*

VxLapW=/

l.p la

t/ *J

t/

VV.W dv - /

VMaxW= f C

and

Hence

Hence
7.

/>

/"

An

x Lap
*

fp n

integral used

c/ t/

dv,v

rPi14 V

If

J J J

W V Max W = 4
by Helmholtz
o

s*

/j

VVW

W V Max W =

x Lap

s+

IT

is

Vdv a

-*

W.

VW

rf

(88)

*
-*

THE INTEGRAL CALCULUS OF VECTORS


or

if

W be a vector function
z

Show

that the integral converges

that

when

WdVr

if F"

(90)

diminishes so rapidly

K
r becomes indefinitely great.

V H ( F) = H (V F) = New (r F),
V # (W) = # (V W) = Max (r W),
V x # (W) = H (V x W) = Lap (r W),
2

VJ2" (

= JT (V VF)=Max(r 2 VF) = 2 Pot ^


V V # (W) = H(y V W) = 2 Pot W.

F)

H ( F) =JET

(W) = -

Pot Pot

F:

~ Pot

(91)

(92)
(93)
C 94 )

(95)

(96)

7T

Pot W.

VV.
8.

259

(97)
5"

(W).

(98)

Give a proof of Gauss's Theorem which does not depend

upon the physical interpretation of a function as the flux of a


The reasoning is similar to that employed in Art. 51
fluid.
and in the first proof of Stokes's Theorem.
Show that the division of
9.
into two parts, page 235,

is

unique.
10.

Treat, in a

manner analogous

the case in which

to that

upon page

has curves of discontinuities.

220,

CHAPTER V
LINEAR VECTOR FUNCTIONS
97.]

AFTER

the definitions of products had been laid

down

and applied, two paths of advance were open. One was


the other, higher algebra
differential and integral calculus
;

in the sense of the theory of linear

The treatment

homogeneous substitutions.

of the first of these topics led to

new

ideas

and new symbols


to the derivative, divergence, curl, scalar
and vector potential, that is, to V, V, Vx, and Pot with the
auxiliaries, the

The treatment

Newtonian, the Laplacian, and the Maxwellian.


of the second topic will likewise introduce

the linear vector


novelty both in concept and in notation
function, the dyad, and the dyadic with their appropriate
symbolization.

The

simplest example of a linear vector function

The

product of a scalar constant and a vector.


r'

= cr

is

the

vector r'

(1)

r.
more general linear function
be
obtained
may
by considering the components of r individLet i, j, k be a system of axes. The components of
ually.

is

a linear function of

r are
i

r,

r,

r.

Let each of these be multiplied by a scalar constant which

may be

different for the different

Cji.r,

j.r,

components.
c3

r.

LINEAR VECTOR FUNCTIONS


Take these
r'

The

as the

components of a new vector

(CiiTj+j

vector r'

is

(c 2

+k

j.r)

then a linear function of

are always equal to the corresponding


multiplied by a definite scalar constant.

Such a
etry and

linear function has

physics.

r.

r'

r).

Its

components

(2)

components of

each

numerous applications in geomi, j, k be the axes of a

for instance,

If,

and c v

strain

homogeneous

(c 8

261

c 2 , c 3,

the elongations along these

axes, a point

r=ri#+J7/ + kz
becomes

r'

or

'

=ic

i c

-f j c 2

+j

<?

+k

c s 2,

c3

2 j

r.

This sort of linear function occurs in the theory of elasticity


and in hydrodynamics. In the theory of electricity and

magnetism, the

electric force

electric displacement

is

a linear function of the

in a dielectric.

For

isotropic bodies

the function becomes merely a constant

But

in case the

body be non-isotropic, the components of the

force along the different axes will be multiplied

constants k v k v k 3

E
The
the

i'D + j& 2 j .D + k&gk.D.

linear vector function

phenomena

different

Thus

= ik

by

is

of electricity,

indispensable in dealing with

magnetism, and optics in non-

isotropic bodies.
98.]

It

is

possible to define a linear vector function, as has

been done above, by means of the components of a vector.


The most general definition would be

VECTOR ANALYSIS

262
Definition

vector r'

tion of another vector r

is

said to be a linear vector func-

when

the components of r' along

three non-coplanar vectors are expressible linearly with scalar


coefficients in terms of the components of r along those same
vectors.

= #a + yb + zc, where [abc]


r' = x' & + y'b + z'c,
x' = a x + by + c^z*

If

and
and

if

0,

(3)

then
etc.,

allel

(The constants a 15
have no connection with the components of a, b,
to i, j, k.)
Another definition however is found
a linear function of

r' is

more convenient and from


Definition

r.

c par-

to be

the foregoing may be deduced.


continuous vector function of a vector is

when

the function of the

any two vectors is the sum of the functions


vectors.
That is, the function /is linear if
of

of those

(4)

Theorem

it

said to be a linear vector function

sum

Jj,

a be any positive or negative scalar and if /


be a linear function, then the function of a times r is a times
:

If

the function of

r.

/(ar)

= a/(r).

(5)

And hence

/Oi*i + 2 r 2 + a 3 r 3 +
)
= i/(*i) + 2 /(r 2 )+ 8 /(r 8 ) +
The proof
obvious

is

of this theorem
trifle

long.

use of relation (4).

(5)'

which appears more or

It depends

less

upon making repeated

LINEAR VECTOR FUNCTIONS


Hence

/(2r)

In like manner

/ (nr~)

where n

is

any positive

= 2/(r).
= w/(r)

integer.

Let

263

be any other positive integer.


just obtained

Then by

the relation

and

Hence
That
is

is,

equation (5) has been proved in case the constant a

a rational positive number.

To show

the relation for negative numbers note that

/(0)=/(0 + 0)=-2/(0).
Hence

But

/(O)

0.

/(O) =/(r-r) =/(r +(-r)) =/(r) +/(-r).

Hence

To

/(r)=-/(-r).

prove (5) for incommensurable values of the constant

becomes necessary to make use of the continuity of the


function /. That is

a, it

LIM

- x

x= a

'

_
~~

Let x approach the incommensurable number a by passing


through a suite of commensurable values. Then

Hence

LlM
CC

_.

CL

(x

r)

= a f (r)

VECTOR ANALYSIS

264

Lm
i- fm

/>>

&

Hence

(ar)=ar.
^
*

tt

= a/(r)

/(ar)

which proves the theorem.


Theorem :

mined when

linear vector function

its

value for three non-coplanar vectors

/ (r)

is

entirely detera, b, c

are

known.
Let

l=/(a),

m=/(b),

*=/()
Since r

is

any vector whatsoever,

Hence

(r)

it

may be

expressed as

= x 1 + y m + z~ft

In Art. 97 a particular case of a linear function was

99.]

expressed as
r'

Cj

c2 j

c3

r.

For the sake of brevity and to save repeating the vector r


which occurs in each of these terms in the same way this

may be

written in the symbolic form


r

= (ic

In like manner
b3

jc 2

+ kc 3

k).r.

if a
be any given vectors, and b r b 2
p a 2 a3
another set equal in number, the expression

r'

is

= aj bj

a 2 b 2 -r

a linear vector function of r

for

a3 b3

owing

to the distributive

(6)

character of the scalar product this function of r satisfies


relation (4).
For the sake of brevity r' may be written symin
the
form
bolically
r'

(a x bj

aa b 2

a3 ba

)' r.

(6)'

LINEAR VECTOR FUNCTIONS

265

No

particular physical or geometrical significance


attributed at present to the expression

(a 1 b 1

It should be regarded as

verts

the vector

r into

affords a convenient

aa ba

+ a,b a +

is

to be

(7)

...).

an operator or symbol which conthe vector r' and which merely

and quick way

of writing the relation

(6).

An expression a b formed by the juxtaposition


:
two vectors without the intervention of a dot or a cross is

Definition
of

The symbolic sum of two dyads is called a


of three, a dyadic trinomial ; of any numbinomial
dyadic
a
ber,
dyadic polynomial. For the sake of brevity dyadic
called a dyad.

binomials, trinomials, and polynomials will be called simply


The first vector in a dyad is called the antecedent ;
dyadics.

and the second

vector, the consequent.

The antecedents

of a

dyadic are the vectors which are the antecedents of the


individual dyads of which the dyadic is composed. In like
manner the consequents of a dyadic are the consequents of
are
the individual dyads. Thus in the dyadic (7) a-p a2 a3
the antecedents and b x , b 2 , b 3
the consequents.
Dyadics will be represented symbolically by the capital
Greek letters. When only one dyadic is present the letter
,

will generally be used.


In case several are under consideration other Greek capitals will be employed also. With
this notation (7)

becomes

=a
and

(6)'

may now

b1

aa ba

By

a8 b8

+...,

(7)'

be written briefly in the form


r'

definition

=&

bx

r.

r -f a 2

(8)

b2

a3 b 3

The symbol <Pr is read


dot r.
It is called the direct
into r because the consequents b x , b 2 , b 3
are
product of

VECTOR ANALYSIS

266

The
multiplied into r by direct or scalar multiplication.
The direct
and r is important.
order of the factors
product of r into
r.

is

= r-(si
+ a2 b + a b +
)
= r.a b + r.a b 2 + ra b + ...
'

<P

l)

Definition

When

Evidently the vectors


:

(9)

r_andjr^_i? are in general different.

the dyadic

is

multiplied into r as

When

r,

is

said to be a prefactor to

0,

said to be a post/actor to r.
used either as a prefactor or as a postfactor
dyadic

r.

r is multiplied in

as

is

to

a linear vector function of


The two linear
vector functions thus obtained are in general different from
one another. They are called conjugate linear vector func-

vector r determines

tions.

r.

The two dyadics

= a b + a b + a b8 +
#- W = bj ^ + b 2 a 2 + b 3 a 3 +
<P

and

each of which

obtained from the other by inter-

may be

changing the antecedents and consequents, are called conjuThe fact that one dyadic is the conjugate of
gate dyadics.
another is denoted by affixing a subscript C to either.

Thus
Theorem:

0=

dyadic used as a postfactor gives the same

result as its conjugate used as a prefactor.

r
100.]

Definition

Any two

That

is

r.

dyadics

(9)

and

are said to

be equal

when
or

when

or

when

r
B

= W r
=r W
r =s
W

for all values of

r,

for all values of

r,

for all values of s

and

(10)
r.

LINEAR VECTOR FUNCTIONS


The

third relation

vectors

vector
scalar

and
like

and

is

equivalent to the

first.

267
For,

the

if

r are equal, the scalar products of

any
them must be equal. And conversely if the
r
product of any and every vector s into the vectors
into

&

! are equal, then those vectors

must be equal. In
shown
that
the
third
relation is equivamay
the second.
Hence all three are equivalent.

manner

lent to

Theorem

be

it

completely determined

is

dyadic

when

the

values

*..,

</>.,

<z>.b,

where

a, b, c are any three non-coplanar vectors, are known.


This follows immediately from the fact that a dyadic defines

a linear vector function.

If

and W are equal provided equaconsequently two dyadics


tions (10) hold for three non-coplanar vectors r and three
non-coplanar vectors s.
Theorem : Any linear vector function /

may

be represented

to be used as a prefactor and by a dyadic


,
by a dyadic
which is the conjugate of 0, to be used as a postfactor.
The linear vector function is completely determined when
its

values for three non-coplanar vectors (say

known (page

Then

264).

i,

j,

k) are

Let

the linear function

is

=a +

givenby

to be used as a postfactor

equivalent to the dyadic

bj

and

to the dyadic

ck,

= ia + jb + kc,

to be used as a prefactor.

/ (r ) =

= r 0^

VECTOR ANALYSIS

268

The study

of linear vector functions therefore

identical

is

with the study of dyadics.


dyad a b is said to be multiplied by a scalar
Definition :
a when the antecedent or the consequent is multiplied by

when a

that scalar, or

a (a b)

any manner between

distributed in

is

the antecedent and the consequent.

(a a) b

a (a b)

If

= a' a"

(a'

a) (a" b).

is said to be multiplied by the scalar a when


dyadic
each of its dyads is multiplied by that scalar. The product
is

written

0a.

or

The dyadic a

applied to a vector r either as a prefactor or


as a postfactor yields a vector equal to a times the vector
obtained by applying
to r
that is

V Theorem
utive.

(a 0)

The combination

That

= a($.r).
of vectors in a

dyad

distrib-

is

is

and

a (b

+ c) =

This follows immediately from the definition of equality of

For

dyadics (10).
[(a

b) c]

= (a +

+ c)J

= a(b + c) .r = abr + acr =

b) c

= ac

r -f

(a c

b c)

and
[a(b

Hence

it

follows that a

each of which

is

the

dyad which

sum

of a

consists of

number

two

of vectors,

factors,

may

be

multiplied out according to the law of ordinary algebra


except that the order of the factors in the dyads must le

maintained.

LINEAR VECTOR FUNCTIONS

269

...

cl

+ cm + cn+

(11)'

...

The dyad therefore appears as a product of the two vectors of


which it is composed, inasmuch as it obeys the characteristic

law of products

the distributive law.

This

is

justifi-

cation for writing a dyad with the antecedent and consequent in juxtaposition as is customary in the case of products
in ordinary algebra.

The Nonion Form of a Dyadic


lOL]

From

the three unit vectors

i, j,

k nine dyads may

be obtained by combining two at a time.


ii,

ij,

ik,

ji>

jj,

jk,

ki,

kj,

If all the antecedents

pressed in terms of

i, j,

These are

(12)

kk.

and consequents in a dyadic


be exand
if
the
k,
resulting expression be

by performing the multiplications according to the


distributive law (11)' and if the terms be collected, the dyadic

simplified

may

be reduced to the

sum

of nine dyads each of

which

is

a scalar multiple of one of the nine fundamental dyads given

above.

a 12

ij

13

ik

+
ki +

a 22

jj

23

ik

a n ii

+
+
This

is

a 31

j i

a 82 kj

(13)

a 33 kk.

called the nonion form of 0.

Theorem
dyadics

a al

d>

The necessary and

and

be equal

sufficient condition that

is that,

when

two

expressed in nonion

VECTOR ANALYSIS

270

form, the scalar coefficients of the corresponding

(?ads

be

equal.
If the coefficients be equal, then obviously

for

and the dyadics by (10) must be


and W are equal, then by
the dyadics

any value of

Conversely,

if

for all values of


i,j,k.

and

=s

Let

r.

and

each take on the vlues

Then

14)

i.

#.i

= i.

gr.i

k.

<P.i

= k.

^.i,

<Z>.j

=i

?F.

i-<Z>.k

= k.

r.j,

k.^.k^k.^k.

= i..k

But

these quantities are precisely the nine coefficients in ;he


and W. Hence the corresponding
expansion of the dyadics
coefficients are equal and the theorem is proved. 1
Thu
analytic statement of the equality of two dyadics can sometimes be used to greater advantage than the more fundamental
definition (10) based

upon the conception of the dyadic as


a
linear
vector
function.
defining
Theorem :
dyadic
may be expressed as the sum of nine

dyads of which the antecedents

are

any three given non-

coplanar vectors, a, b, c and the consequents any three given


non-coplanar vectors 1, m, n.

Every antecedent may be expressed in terms of a, b, c


and every consequent, in terms of 1, m, n. The dyadic may
then be reduced to the form
;

+
+

21

a 31

+
8l+
bl

a 23

bm +

a 32

cm +

23

bn

(15)

a 33 en.

9^
1

As a

corollary of the theorem it is evident that the nine dyads (12) are inNone of them may be expressed linearly in terms of the others.

dependent.

LINEAR VECTOR FUNCTIONS

271

is more general than that given in


This expression of
It reduces to that expression when each set of vectors
(13).

a, b, c

and

1,

Theorem

m, n coincides with

Any

i, j,

may

dyadic

k.

be reduced to the

sum

of

three dyads of which either the antecedents or the consequents,


but not both, may be arbitrarily chosen provided they be non-

coplanar.

Let
of

it

which

be required to express
a, b, c

three non-coplanar vectors.


(15).

(P

sum

as the

of three dyads
m, n be any other
then be expressed as in

Let

are the consequents.

may

1,

Hence

= a (a u +
1

a la

m+

a ls n)

+ b (a 21 1 + 22 m + a 23 n)
+ c (a 31 1 + a 32 m + a 32 n),

= aA + bB +

or

cC.

(16)

be required to express
as the sum of
three dyads of which the three non-coplanar vectors 1, m, n are
the consequents

In like manner

if -it

= Ll +

= au a +
M = a l2 a +
a +
N=
L

where

18

Mm + Nn,
+
b +

a zl b
a Z2
23

a 31

(16)'
c,

32 c,
33 c.

are unique. Two equal


expressions (15), (16), (16)' for
dyadics which have the same three non-coplanar ante-

The

these
a, b, c, have the same consequents A, B, C
however need not be non-coplanar. And two equal dyadics
which have the same three non-coplanar consequents 1, m, n,
have the same three antecedents.
cedents,

102.] Definition:

position of

or a cross

a and

b.

The symbolic product formed by

two vectors

is

a,

the juxtab without the intervention of a dot

called the indeterminate product of the

two vectors

VECTOR ANALYSIS

272

The reason

for the

term indeterminate

a x b have
products a b and

definite

meanings.

On

certain scalar, the other a certain vector.

the product

ab

is

The two
One is a

is this.

the other

neither vector nor scalar

hand

it is

purely
symbolic and acquires a determinate physical meaning only
when used as an operator. The product a b does not obey
It does however obey the distributive
the commutative law.

law (11) and the associative law as far as scalar multiplication


is concerned (Art 100).
TJieorem : The indeterminate product a b of two vectors is
the most general product in which scalar multiplication

is

associative.

The most general product conceivable ought to have


property that when the product is known the two factors

the
are

known. Certainly no product could be more general.


Inasmuch as scalar multiplication is to be associative, that is
also

a(ab)
it

= (aa) b = a (ab) = (a'a)

(a"b),

will be impossible to completely determine the vectors a

and b when

may

their product a b is given.


Any scalar factor
be transferred from one vector to the other. Apart from

this possible transference of

posing the product are


other words

Theorem

a scalar factor, the vectors com-

known when

the product

is

known. In

two indeterminate products a b and a' b'


and a', b and b' must be collinear and
the lengths of a and b (taking into account the

If the

are equal, the vectors a

the product of

positive or negative sign according as a


tively equal or opposite directions to

product of the lengths of

Let

=a

a'

and

and b have respec-

a'

and

aa

as

k,

b'.

b') is

equal to the

LINEAR VECTOR FUNCTIONS

273

'*
Then

ab

=
+

a'V

and

Since

ab

= a'b'

&1

& 1 kj

ii

/&/

kj

a3 6

ij

a/Jg' ik

jj

a 2 & 3 jk

a 3 '6 2 kj

a 3 '6 3 kk.

Z>

2
'

a,'ft a

'V ji+a 2 '& 2


'

a s '& 1 ki

'

'

'

corresponding coefficients are equal.


i

!':/!

i':

which shows that the vectors a and

And

&i:&,:6 8

= &i'

al

b1

are collinear.

a'

W.

which shows that the vectors b and

But

Hence

b' are collinear.

= a^ &/.

This shows that the product of the lengths (including sign)


are equal and the theorem

The proof may be


Since ab is equal to

r.

proved.

carried out geometrically as follows.


a'b'

ab
for all values of

is

= a'b'

Let r be perpendicular

to b.

vanishes and consequently b'r also vanishes.


for any vector r in the plane perpendicular to b.

Then b

This

is

true

Hence b and

same plane and are collinear. In


b
as a postfactor a and a' are seen
like manner by using a
Also
to be parallel.
a'b' b,
ab b

b'

are perpendicular to the

which shows that the products of the lengths are the same.
18

VECTOR ANALYSIS

274

The indeterminate product a b imposes Jive conditions upon


the vectors a and b. The directions of a and b are fixed and
The scalar product
likewise the product of their lengths.
being a scalar quantity, imposes only one condition upon
a and b. The vector product a x b, being a vector quantity,
a

b,

to the plane of a and


imposes three conditions. The normal
the
of
b is fixed and also the area
parallelogram of which they

are the side.

The nine indeterminate products (12) of i, j, k


The nine scalar products

into themselves are independent.

Only two

are not independent.

and

=j

=j

them are

different.

= kj=ki = ik = 0.

The nine vector products

of

are not independent either

for

ixi = jxj = kxk=:0,


and

xj

xk

i,

k x j,

kx

k.

The two products a b and a x b obtained respectively from


the indeterminate product by inserting a dot and a cross between the factors are functions of the indeterminate product.
is to say, when ab is given, a b and a X b are determined.
For these products depend solely upon the directions of a and b
and upon the product of the length of a and b, all of which

That

are
if

known when ab
ab

= a'b',

is

known.

= a'

b'

That

is

and a x b

= a'

b'.

(17)

not hold conversely that if a b and a x b are known


for taken together a b and a X b impose upon the
;
vectors only four conditions, whereas a b
imposes five. Hence
a b appears not only as the most
general product but as the
most fundamental product. The others are merely functions
It does

ab

of

is

it.

fixed

Their functional nature

is

notation of the dot and the cross.

brought out clearly by the

LINEAR VECTOR FUNCTIONS


Definition

scalar

known

275

as the scalar of

be ob-

may

tained by inserting a dot between the antecedent and consequent of each dyad in a dyadic. This scalar will be denoted

by a subscript $ attached

=a
=a
S

If

b1
bx

to 0.

+ a 2 b 2 + a 3 b 3 + ...
+ a2 b 2 + a3 b 3 +
.

In like manner a vector

known

. .

(18)

as the vector of

may

be

obtained by inserting a cross between the antecedent and consequent of each dyad in 0. This vector will be denoted by
attaching a subscript cross to 0.
X

=a

x bj + a 2 x b 2

x b3 +

a3

be expanded in nonion form in terms of

If

= a n + a Z2 + a^
- a is) + ( a i2 ~
* +
(si

(19)
i,

j,

k,

(20)

- ^32)
<Px =
23
= i. 0-i +j.0.j + k
S
<p x = (j
k -k
j) i + (k
3

Or

a 2i) k

0-k,
i - i

+ (i.0.j-J-0'i)k-

( 21 )

(20)'

k)

(21)'

In equations (20) and (21) the scalar and vector of


are
in
terms
of
the
coefficients
of
when
expressed
expanded
and
in the nonion form.
Hence if
are two equal
is equal to the scalar of
dyadics, the scalar of
is equal to the vector of
.
vector of

If

From

this it

appears that

and 0*

3
S

and

and the

x.

are

functions of

(22)

is given.
uniquely determined when
They may sometimes
be obtained more conveniently from (20) and (21) than from

(18)
1

and

and

(19),

and sometimes

not.

A subscript dot might be used for the scalar of *


free

from

liability to misinterpretation.

if it

were sufficiently

distinct

VECTOR ANALYSIS

276

Products of Dyadics

In giving the definitions and proving the theorems


concerning products of dyadics, the dyad is made the under103.]

What

lying principle.

owing
law
the
distributive
obey
Definition:

dyad

is

The

true for the dyad

is

true for the

to the fact that dyads

and dyadics

is

dyadic in general

of multiplication.

written

,,

(a b)

and

is

by

into the

,.

(c d)

definition equal to the

(ab).(cd)

ab

product of the dyad

direct

dyad (b

aA

c)

= a(b.c)d = b-c

a4.

(23)

and the consequent of the


That is,
second dyad are taken for the antecedent and consequent
respectively of the product and the whole is multiplied by
the scalar product of the consequent of the first and the
the antecedent of the

first

antecedent of the second.

Thus the two vectors which stand together


(a b)

in the product

(c d)

The

are multiplied as they stand.

other two are left to form

new dyad. The direct product of two dyadics may be


defined as the formal expansion (according to the distributive
law) of the product into a sum of products of dyads. Thus
a

and
<P.

+a

b2

+a

^=(01*1 +c 2

d2

^=(a

W=(^

^l

(cjdj

=a
+
+
+
1

The

b1
1

a 3 b a .c 1 d 1

parentheses

+a
+a

may

2
3

b3

c3

d3

+
+

+a b +a
+c 2 d +c
2

b3

d3

...)
)

a1 b1

b 2 .c 2 d 2

a 2 b 2 .c 3 d 3

b3

a3 b3

d 1 4-a 1 b 1 -c 2 d 2

a 2 b 2 .c 1 d 1

c2

d2

C3

d3

+ ...
c8 d3 +

he omitted in each of these three expressions.

(23)'

LINEAR VECTOR FUNCTIONS


.

W=b
+

t>

b3

a x dj

cl

'

+b

+b

ax d2

c2

a2 d l

1>2' C 2

&2 d2

1>2

dx

b3

a3 d2

b3

Cj a 3

c2

c3

277

a 1 d3

'

C 3 *2

d3

c3 a3

d3

---

'

'

'

(23)"

The product
Theorem

regarded as

of two dyadics
The product

an operator

and

W of

to be

To show
or

?F).r

Let ab be any dyad of

and

(a b

ab

(c d

Hence

The theorem

=
c

equiva-

r),

</>.(r.r).

d any dyad of

= b'c (a d
r) = a b c (d
(a b c d) r =

c d)

is

W when

W.

(<Z>.

W.

and

by the operator 0.

Q=

Let

a dyadic

used as a prefactor

W followed

lent to the operator

W is

two dyadics

(24)
.

= (b c) (d r) a,
r) = (b c) (d r) a,

r)

ab

(c d

r).

true for dyads. Consequently by virtue of


the distributive law it holds true for dyadics in general.
If r

is

denote the position vector drawn from an assumed origin


P in space, r' W r will be the position vector of

to a point

another point P\ and r"

$(*)

will be the

position
That is to say, W defines a transvector of a third point P".
formation of space such that the points
go over into the
a
P'.
defines
transformation
of
space such that the
points

Hence
points P' go over into the points P".
The single operation
into P".
carries
into P".

followed by
also carries

Theorem: Direct multiplication of dyadics obeys the


tributive law.

That

is

dis-

VECTOR ANALYSIS

278

&(
and

($'

Hence

= <P.
= &'

'}

<P)

+ 0+

'

(25)

in general the product

(0+

<p>

+ 0" +...).

(+

'+

"+..)

be expanded formally according to the distributive law.


Theorem : The product of three dyadics <P,
Q is associa-

may

That

tive,

^.^. Q=0

is

.^

(26)

and consequently either product may be written without


parentheses, as

The proof

$.?.&

(26 )'

theorem for

consists in the demonstration of the

three dyads ab, cd, ef taken respectively from the three


Q.
,
dyadics 0,

= (bc) ad* ef = (bc) (d e) af,


= (d-e) abcf = (de) (bc) af.

(ab.cd) ef

ab.(cd.ef)
The proof may

also be given

as operators

{(<?

V) . Q\

Let

= (0
.r = r'
r

{($)}.* = (&.
Let

by considering
.

W) (Q

).i'=<l>.

r).

(!').

= r",
\(0. W)*Q\ r = 0.i" = T'".
{(?.(?T. )} r =
[(?T. J2) r ].
.

r'
.

Again

Hence

{((^. ?F)

for all values of

r.

j2|

<^>.

{#.

Consequently

(?F.

<#,

.r

and

LINEAR VECTOR FUNCTIONS

279

The theorem may be extended by mathematical induction


any number of dyadics. The direct product
of any number of dyadics is associative.
Parentheses may
to the case of

be inserted or omitted at pleasure without altering the


It was shown above (24) that

(0. T).r

= &.(.T)=

0. W*r.

result.

(24)'

Hence the product of two dyadics and a vector is associative.


The theorem is true in case the vector precedes the dyadics
and also when the number of dyadics is greater than two.
But the theorem is untrue when the vector occurs between
the dyadics. The product of a dyadic, a vector, and another
dyadic

is

not associative.

(0.r).
Let a b be a dyad of
(a b

ab

r)
(r

c d)

and

<P.(r.

= b r (a c d) = (b
= ab d (r c) = b

results of this article

Theorem: The

may

(27)

).

d a dyad of W.
c) d,

r) (a

(r

c)

ab (rcd).

(abr) cd

Hence

The

<P,

W*

be

summed up

direct product of

as follows

any number

of dyadics

or of any number of dyadics with a vector factor at either


end or at both ends obeys the distributive and associative
laws of multiplication
parentheses may be inserted or

omitted at pleasure. But the direct product of any number


of dyadics with a vector factor at some other position than at
either

end

is

not associative

parentheses are necessary to

give the expression a definite meaning.


Later it will be seen that by making use of the conjugate
dyadics a vector factor which occurs between other dyadics

may
made

be placed at the end and hence the product


to assume a form in which it is associative.

may

be

VECTOR ANALYSIS

280
104.]

and

a vector r
respectively

dyad ab into
dyad ab are denned

The skew products

Definition:

a vector r into a

of

of a

by the equations

= a(b
rx(ab) = (r x

(ab) x

r),

a)b.

The skew product of a dyad and a vector at


dyad. The obvious extension to dyadics is

=a

b1 X r

=r

a2 b2

x ajbj +

xr +

x a2 b2

a3 b3

either

end

is

r H

a s bg

...

Theorem: The direct product of any number of dyadics


multiplied at either end or at both ends by a vector whether
the multiplication be performed with a cross or a dot is

But

associative.

in case the vector

position than the end the product

(rx $)

but

occurs at any other

not associative.

= r x (0-s) =r
(0 x s) = (r 0) x s = r
(0 x s) = (r x 0) x s = r
.(TX 0) * (?T.r)x

and

can have no other meaning than


s

d>

x <Ps,
x

x
V.

Furthermore the expressions


s

That

is,

= rx(0?r) = rx <!>,

x 0).s

(r
r

?F

is

= s (r x $),
s = (0 x r)
s,

<P

(29)

s,

a,

LINEAR VECTOR FUNCTIONS


with a cross into a scalar

since the product of a dyadic


is

281
s

Moreover since the dot and the cross may

meaningless.

be interchanged in the scalar triple product of three vectors


it

appears that

x
and

(r

The parentheses

in

d>

(B

r)
(r

= (0

x W)

<P,

(31)

s),

r)

these expressions

cannot be omitted

without incurring ambiguity.

$.(r X
(s

(0

s)

(<Z>.r)

&

r)

r)

x W#

The formal skew product


(ab) x

of

(c d)

0),

(r

(r

s,

(31)'

).

two dyads a b and

=a

(b

In

this expression three vectors a,

by

side with

d would be

c) d.

b x

c,

d are placed side

no sign of multiplication uniting them.

Such

an expression
rst
is

called a triad

The theory

and a sum

(32)

of such expressions, a triadic.

intimately connected with the theory


of linear dyadic functions of a vector, just as the theory of
dyadics is connected with the theory of linear vector functions
of triadics

is

In a similar manner by going a step higher


tetrads and tetradics may be formed, and finally polyads and
of a vector.

But the theory of these higher combinations of


The dyadic
vectors will not be taken up in this book.

polyadics.

furnishes about as great a generality as is ever called for in


practical applications of vector methods.

VECTOR ANALYSIS

282

Degrees of Nullity of Dyadics


105.]

It

was shown

be reduced to a

sum

(Art. 101) that a dyadic could always


of three terms at most, and this reduction

can be accomplished in only one

way when

or the consequents are specified.

In particular cases

the antecedents
it

may

be possible to reduce the dyadic further to a sum of two


terms or to a single term or to zero. Thus let

= al + bm + cn.
If

1,

m, n are coplanar one of the three

two

in terms of the other

may

be expressed

as

= x m + y n.

(P^azm + ayn + bm + cn,


<P = (a x 4- b) m + (a y + c) n.

Then

The dyadic has been reduced to two terms. If 1, m, n were


all collinear the dyadic would reduce to a single term and if
they all vanished the dyadic would vanish.
Theorem :

If a

be expressed as the

dyadic

sum

of three

terms

= al + bm +

d>

of

which the antecedents

a, b, c

are

en

known

to be non-coplanar,

sum of two dyads


when and only when the consequents are coplanar.
The proof of the first part of the theorem has just been
To prove the second part suppose that the dyadic
given.
then the dyadic

may

be reduced to the

could be reduced to a

sum
<P

of

two terms

= dp +

eq

and that the consequents

were non-coplanar.
1,
m, n of
This supposition leads to a contradiction. For let 1', m', n'
be the system reciprocal to 1, m, n. That is,
,

_" mx n
[Imn]'

n x

~jTmn]'

_
=

[Tmn]

g
'

LINEAR VECTOR FUNCTIONS

283

vectors 1', m', n' exist and are non-coplanar because


m, n have been assumed to be non-coplanar. Any vector r
may be expressed in terms of them as

The

1,

<p.r

But
and

(al
1

m'

= x\' +

+
1'

bin

=m

= 1' m = m

ym'

en)

(x\*

m' = n

n'

= m'

zn,'

= (dp +

This shows that

e q)

= d (p

= 0.

zc.

giving to r a suitable value the vector


equal to any vector in space.
r

1'

By

But

zn').

n' = 1,
= n'
n =n

0r = xa + yb +

Hence

+ ym' +

r)

may

e (q

must be coplanar with d and

be

made

r).

e.

Hence

can take on only those vector values which lie in the


plane of d and e. Thus the assumption that 1, m, n are noncoplanar leads to a contradiction. Hence 1, m, n must be
r

coplanar and the theorem


Theorem : If a dyadic
terms

is

proved.
be expressed as the

= &l + bm +

of

which the antecedents

a, b, c

are

sum

of three

en,

known

to be non-coplanar,

can be reduced to a single dyad when and only


the dyadic
when the consequents 1, m, n are collinear.

The proof of the first part was given above.


the second part suppose
could be expressed as

Let

W0 x

= dp

?F=al xp +

x p

= dO = 0,

bmxp + cnxp.

To prove

VECTOR ANALYSIS

284

From

the second equation


postfactor for any vector
r

where

a', b', c' is

=xa +
;

the

first

y b'

used as a

20',

the reciprocal system to

r. ?r =iclxp

From

evident that

it is

a, b, c

gives

+ 2/inxp + 2nxp.

expression
r

0.

zlxp+ymxp + znxp

Hence
must be zero
Hence
y, 2.

for every value of

lxp = 0,

1, m, and n are
been demonstrated.

Hence

x p

r,

that

= 0,

all parallel to

If the three consequents

1,

is,

for every value of #,

n x p

= 0.

p and the theorem has

m, n had been

known

to be non-

coplanar instead of the three antecedents, the statement of


the theorems would have to be altered by interchanging the

words antecedent and consequent throughout. There is a further theorem dealing with the case in which both antecedents

Then

and consequents of
are coplanar.
sum of two dyads.

is

reducible to

the

106.]

the

sum

Definition:

dyadic which cannot be reduced to


is said to be complete.

of fewer than three dyads

dyadic which may be reduced to the sum of two dyads, but

cannot be reduced to a single dyad is said to be planar. In


case the plane of the antecedents and the plane of the consequents coincide

when

the dyadic

is

expressed as the

sum

of

two dyads, the dyadic is said to be uniplanar.


A dyadic
which may be reduced to a single dyad is said to be linear.
In case the antecedent and
consequent of that dyad are col-

LINEAR VECTOR FUNCTIONS


linear, the

285

said to be unilinear.

If a dyadic
may be
terms vanish the dyadic is said to
In this case the nine coefficients of the
dyadic as

dyadic

is

so expressed that all of

be zero.

its

expressed in nonion form must vanish.

The properties of complete, planar, uniplanar, linear, and


unilinear dyadics when regarded as operators are as follows.
Let

and

=r

is complete s and t may be made


If
value by giving r a suitable value.

As

is

complete

reciprocal system
B

In like manner
t

take on

any

desired

m', n'.

1',

+ ym' +

a, b, c

= (xa. +

to

m, n are non-coplanar and hence have a

1,

(xl

0.

za')

= ica + yb +

20.

possess a system of reciprocals

y\>'

zc')

a', b', c'.

= x\ + y m + zn.

complete dyadic
applied to a vector r cannot give zero
unless the vector r itself is zero.
is

If

planar

the vector s

may

take on

any value in

the

plane
of the antecedents and t any value in the plane of the consequents
but no values out of those planes.
The dyadic
when
of
used as a prefactor reduces every vector r in space to a vector
;

In particular any vector r


in the plane of the antecedents.
is reduced
perpendicular to the plane of the consequents of
used as a postfactor reduces every
to zero.
The dyadic
vector r in space to a vector in the plane of the consequents
In particular a vector perpendicular to the plane of
of 0.
In case the dyadic
is reduced to zero.
the antecedents of

uniplanar the same statements hold.


is linear the vector s may take on any value collinear
If
and t any value collinear with the conwith the antecedent of

is

VECTOR ANALYSIS

286
<P

sequent of

The dyadic

but no other values.

<P

used as a

r to the line of the antecedent


prefactor reduces any vector
In particular any vectors perpendicular to the conof 0.

The dyadic

are reduced to zero.

sequent of

used as a

vector r to the line of the consequent

postfactor reduces

any
In particular any vectors perpendicular to the anteare thus reduced to zero.
cedent of

of 0.

If

is

zero dyadic the vectors s

matter what the value of r


Definition
nullity.
nullity.
lity

may

A planar dyadic

and

are loth zero no

be.

said to possess one degree of


said to possess two degrees of
is

A linear dyadic is
A zero dyadic is said

to possess three degrees of nul-

or complete nullity.

The

two complete dyadics


and
a planar dyadic,
dyadic
complete;
and
a
of
a
linear
planar ;
complete dyadic
dyadic, linear.
107.]

Theorem

direct product of

of a complete

is

Theorem

The product

of

two planar dyadics

is

planar
the plane of the consequent of the first dyadic
in the product is perpendicular to the plane of the antece-

except

when

dent of the second dyadic.

In this case the product reduces

and only in

to a linear dyadic

= &'b + ab

Let

Q=
The
and

The

vector

B'

s'

vector

r takes

W.

on

2,

all

values in the plane of

J* (bj -oj)

takes on the values

=x

+
B'

this case.

+y

(bj

X 0>2

(bj

c )
:

ax

y (bj

'

C l)

a2

2^

c 2 )} a x

(b 2

{x (b 2

C 2) a t
'

Cl )

C)

a 2>

y (b,

c 2 )}:

LINEAR VECTOR FUNCTIONS


Let

s'

=x
=
y'

where

x'

and

a?

+ y'a a
c^ + # (bj
c ) + y (b a

JT'E!

(b x
(b 2

287

c 2 ),
c 2 ).

These equations may always be solved for x and y when


that is, when s' has
any desired values x' and y' are given
unless the
any desired value in the plane of a^ and a2
determinant
b2

But by

(25),

Chap.

j.

vector bj

x b2

x b,)
a/

c,

merely the product

II., this is

(b!
x.

The

b2

Cj

(c,
\

j.

Co)
&/

0.

perpendicular to the plane of the conand c x X c 2 to the plane of the antecedents of


is

sequents of
Their scalar product vanishes
;

that

vectors are perpendicular

pendicular.
Consequently
a
and
of
a 2 and
plane
x

when and only when the


is, when the planes are per-

may
is

take on any value in the


therefore a planar dyadic

unless the planes of b x and b2 , Cj and c 2 are perpendicular.


If however bj and b2 , Cj and c 2 are perpendicular s' can take
on only values in a certain line of the plane of a x and a^, and

hence

(P

The theorem

is

therefore proved.

The product of two linear dyadics is linear


when the consequent of the first factor is perpen-

Tlieorem

except

is linear.

dicular to the antecedent of the second.

In this case the

and only in this case.


product is zero
Theorem : The product of a planar dyadic into a linear
linear except

planar dyadic
dyadic.

In

is

when
is

the plane of the consequents of the


perpendicular to the antecedent of the linear

this case the

product

is

zero

and only

in this

case.

Theorem: The product of a linear dyadic into a planar


dyadic

is

linear except

when

the consequent of the linear

VECTOR ANALYSIS

288

perpendicular to the plane of the antecedents of


and
the planar dyadic. In this case the product is zero

dyadic

is

only in this case.

immediately evident that in the cases mentioned the


products do reduce to zero. It is not quite so apparent that
It is

they can reduce to zero in only those cases.

The

similar to the one given above in the case of

They

dyadics.

are

left to

the reader.

theorem stated, page 286,

first

is

proofs are

two planar

The proof

of

the

also left to the reader.

The Idemfactor; 1 Reciprocals and Conjugates of Dyadics


108.] Definition : If a dyadic applied as a prefactor or as
a postfactor to any vector always yields that vector the
dyadic is said to be an idem/ actor. That is
d>

if

or

if

then

is

= r for all values of r,


<P = r for all values of r,
r

The capital I is used as the symThe idemfactor is a complete dyadic.

an idemfactor.

bol for an idemfactor.

For there can be no direction in which


Theorem :

When expressed

in nonion

r vanishes.

form the idemfactor is


(33)

Hence

idemfactors are equal.


that the idemfactor takes the form (33) it is
merely necessary to apply the idemfactor I to the vectors
all

To prove

i, j,

Let

respectively.

= a n ii +
+

a 21 ji

ij

13

ik

22 jj

23

jk

a 12

In the theory of dyadics the idemfactor I plays a role analogous to unity in


The notation is intended to suggest this analogy.

ordinary algebra.

LINEAR VECTOR FUNCTIONS

Ii = a n i+
I

If

au
In like manner

=1

may

it

vanish except a n ,

22 ,

21 j

and

21

31

k.

i,

= a 31 = 0.

be shown that
33

all

the coefficients

which are unity.

all of

= ii + jj +

289

kk.

Hence
(33)

Theorem : The direct product of any dyadic and the idemfactor

is

That

that dyadic.
d>

For

(0.1)

is,

T = 0.(lT) =#T,

no matter what the value of r


.

In like manner

Theorem:

If

it

may

a',

b', c'

& = d>.

and I

may
1

be.

Hence, page 266,

= 0.

be shown that I

and

a, b, c

= 0.

be two reciprocal systems

of vectors the expressions

= aa' + bb' + cc',


I = a'a + b'b + c'c
I

(34)

are idemfactors.

For by (30) and (31) Chap.

= raa + r-bb' + rcc',


r = ra'a + r*b'b + rc'c.
;

and

II.,

Hence the expressions must be idemfactors by


Theorem : Conversely
4>

is

an idemfactor

1,

if

definition.

the expression

= al +

bin

+ en

m, n must be the reciprocal system of

a, b, c.

19

VECTOR ANALYSIS

290

In the
dyadic.

is the idemfactor, it is a complete


place since
Hence the antecedents a, b, c are non-coplanar and

first

possess a set of reciprocals a', b',

By

hypothesis

Then

a;

a'

Let

c'.

+ yV +

c'.

= r.
= x\ + ym + zn = #a' + yV + zc
r

for all values of r, that

is,

corresponding coefficients
1

for all values of #, y,

must be equal.

m = b',

a',

Hence the

z.

That

is,

c'.

be any two dyadics, and if the product


1
is equal to the idemfactor;
then the product
0,
when the factors are taken in the reversed order, is also

Theorem

If

and

equal to the idemfactor.

= !.
= I.

Let

To show

r(<P.

(0

all

=f

all

0)

values of

If the

product of two dyadics


be taken in either order.

r.

*0,

(?)

As

Hence by

desired values.
.

may

=r.I = r,

r(0. F)
= (r
?T)

This relation holds for

must take on

=r
is

#.

complete r

definition

= I.
is

an idemfactor, that product

When two dyadics are so related that


equal to the idemfactor, they are said to be

109.] Definition:
their product is
1

of

This necessitates both the


dyadics

two incomplete dyadics

idemfactor.

is

* and V to be complete. For the product


incomplete and hence could not be equal to the

LINEAR VECTOR FUNCTIONS


The notation used

reciprocals^

algebra

is

291

for reciprocals in
ordinary

to denote reciprocal
dyadics.

employed

That

=0-i =
Theorem:

is,

(35)

Reciprocals of the same or equal dyadics are

equal.

and

Let

be two given equal dyadics,

their reciprocals as defined above.

0=
~i

and

hypothesis

= I.

= ~i.
0.0-^ = 1= W.

To show

0-i

0=

As

0-1

W,

~\

=0-1.0.

~\

0-1

I.0-i
Hence

~i.

0.0-i= 0.
0-1.0

The

By

0~l and W~l

= 1,

= 0-i = I. ~i =
0~i =
~\

~i.

the dyadic whose antecedents are the


and whose consereciprocal system to the consequents of
are
the
antecedents
to
the
of 0.
reciprocal system
quents
reciprocal of

is

be written in the form

If a complete dyadic
<P

its

reciprocal

For

(al

0~i

+ bm + cn)

Theorem
into

is

= al + bm + en,
= !'&' + m' b' + n'

(l'a'

+ n'V +

c'.

n'c')

(36)

=aa' + bV +

cc'.

If the direct products of a complete dyadic

two dyadics
1

and

An incomplete

are equal as dyadics then

dyadic has no

(finite) reciprocal.

and

VECTOR ANALYSIS

292
are equal.

If the

into

product of a dyadic

two vectors

r and s (whether the multiplication be performed with a dot


or a cross) are equal, then the vectors r and s are equal.

That

is,

=
then r =
then r =

= 0*8,
& r=
x
x r=

O.

if

and
and

if

s,

if

This

then

(37)

s,

s.

s,

may be

Q,

seen by multiplying each of the equations

through by the reciprocal of 0,

i .

To

= r = &~
r = I x r = 0"
.

<P

=
s = I
s,

reduce the last equation proceed as follows.

any vector,

Hence
t is

any

s.

Let

be

t-Ixr = tIxs,
t

As

vector, r

is

=
r = t x
I

equal to

Equations (37) give what

is

t.

s.

s.

equivalent to the law of can-

dyadics. Complete dyadics may be


canceled from either end of an expression just as if they
were scalar quantities. The cancelation of an incomplete
dyadic is not admissible. It corresponds to the cancelation

celation for complete

of a zero factor in ordinary algebra.

Theorem: The reciprocal of the product of any


of dyadics is equal to the product of the reciprocals
taken in the opposite order.
110.]

number

It will be sufficient to give the proof for the case in which


the product consists of two dyadics. To show

LINEAR VECTOR FUNCTIONS


(0.

W W~

0-

ST)-i

Hence

(0

2T)

Hence

W and

5F"1

3F-i. 0-i,

~l

ST"1

0-i

(
(

293

0- 1 )

0"1 must

0-1

= I.

I.

be reciprocals.

That

sr-i. 0-i.

(0. ?T)-i=

The proof for any number of dyadics may be given


same manner or obtained by mathematical induction.
Definition

of times,

The products

by

and so

taken any number


and are denoted in

powers of

the customary manner.

in the

of a dyadic 0,

itself are called

is,

02,

02

03^

forth.

TJieorem : The reciprocal of a power of


the reciprocal of 0.

(0)-i

= (0-i) =

is

0-.

the power of

(37)

The

proof follows immediately as a corollary of the precedtheorem.


The symbol 0~" may be interpreted as the
ing
nth power of the reciprocal of
or as the reciprocal of
the nth power of 0.
If
be interpreted as an operator determining a transformation of space, the positive powers of
correspond to

The negative powers of


The idemfactor
transformations.

repetitions of the transformation.

correspond to the inverse

that is, no transcorresponds to the identical transformation


formation at all. The fractional and irrational powers of
will not be denned.

single-valued.
square roots
infinite

They

are seldom used

For instance the idemfactor


I.

But

in addition to these

system of square roots of the

= -ii +

form

kk.

I
it

and are not


has the two
has a doubly

VECTOR ANALYSIS

294

Geometrically the transformation

r'=

<P-r

a reflection of space in the j k-plane. This transformation


replaces each figure by a symmetrical figure, symmetrically
The transsituated upon the opposite side of the j k-plane.
is

formation

is

sometimes called perversion.

The idemfactor

has also a doubly infinite system of square roots of the form

Geometrically the transformation


r'

is

= W

This transformation replaces each

a reflection in thei-axis.

figure by its equal rotated about the i-axis through an angle


of 180.
The idemfactor thus possesses not only two square

roots

roots

means

but in addition two doubly infinite systems of square


and it will be seen (Art. 129) that these are by no
all.

The conjugate of a dyadic has been defined (Art. 99)


111.]
as the dyadic obtained by interchanging the antecedents and
consequents of a given dyadic and the notation of a subscript
G has been employed. The equation
r

has been demonstrated.

(9)

The following theorems concerning

conjugates are useful.

Theorem
dyadics

is

Theorem
to the

:
The conjugate of the sum or difference of two
equal to the sum or difference of the conjugates,

The conjugate

of a product of dyadics is equal

product of the conjugates taken in the opposite order.

LINEAR VECTOR FUNCTIONS

295

It will be sufficient to demonstrate the theorem in case

To show

the product contains two factors.

(0.T)C =VC .0 C
8%.r = r-(0. W) = (r.Q).

(40)

(0.

Hence

(0

Theorem

power

)c

The conjugate

4>c>

of the

power

of a dyadic is the

of the conjugate of the dyadic.

(*%=
This

= Wc

V,

is

= **

(41)

a corollary of the foregoing theorem.

may

be interpreted in either of

Theorem

The conjugate

The

expression

two equal ways.

of the reciprocal of a dyadic

is

equal to the reciprocal of the conjugate of the dyadic.

=(0
(0-% *<P C = (0
(0-i) c

For

The idemfactor

its

is

own

)-i

= 0^.

(42)

conjugate as

may

be seen from

the nonion form.


I

=ii+

j j

kk

(0 c)-i.0 c =
4> c =
0~^
((P^"
=
((^c)(0-%
I.

Hence

Hence

~
The expression & c l may

therefore be interpreted in either

two equivalent ways

as the reciprocal of the conjugate

of

or as the conjugate of the reciprocal.


If a dyadic is equal to
Definition:
to be self-conjugate.

If

it is

its

conjugate,

it is

said

equal to the negative of its con-

VECTOR ANALYSIS

296
it is

jugate,

For se/-conjugate

said to be anti-self-conjugate.

dyadics.
r

r,

C.

For anti-self-conjugate dyadics


r

Theorem

way

into

Any

0.

dyadic

two parts

may

r,

00.

be divided in one and only one


is self-conjugate and the

which one

of

other anti-self-conjugate.

0=(0+0 C

For

But

(0 + 0c) c =

and

(0 -

C-)

+ (0-0 C\

=
- CC =
C=
C
C

Hence the part |(0 + ^c) ^s


C), anti-self-conjugate.
|(0

-0.

0,

self-conjugate

Thus the

and the part

division has been

Let

accomplished in one way.

(0 +

CC

(43)

<2>c)

= 0' +

*'

0".

in another
Suppose it were possible to decompose
into a self-conjugate and an anti-self-conjugate part.

way
Let

then

Where
Hence

(0'
if

(0'

- (0"
Hence

if

conjugate.

,12)

-Q)=
(0"

(0'

J2) c

= 0' c + Q c = 0' +

is self-con
jugate,

(0"
) is

)c

0" c

is

self-conjugate.
J2 e

= _ 0" -

anti-self-conjugate

J2

is

anti-self-

LINEAR VECTOR FUNCTIONS

Any
gate

dyadic which
is

both self-conjugate and anti-self-conju-

is

negative and consequently vanishes.


zero and the division of
into two parts is
to

equal

Hence

297

is

its

unique.

The Vector Product

Anti-self-conjugate Dyadics.

In case

112.]

is

any dyadic the expression

should be engives the anti-self-conjugate part of 0. If


is
to
0".
Let
therefore 0"
equal
tirely anti-self-conjugate
be any anti-self-conjugate dyadic,

0"

= al + bm +

Suppose

= &1
C
<?"

= 1(0-0^.

= al

+ bm
lar-f-bmr

But

mb +

la

lar =

al* r

cn

nc,

mbr + cnr
(a

bmr mbr

cn,

1)

ncr.

r,

(bxm)xr,

ncr =

0"

Hence
But by

cnr
(cxn)xr.
r =
^(axl + bxm + cxn)xr.

definition

<P x

0"

Hence
r

The

0"

= 0" c

results

Theorem

=:axl-|-bxm +
r

=-\

= - 0"

0* X

= \0 K

r,

may be stated in a theorem as


The direct product of any

dyadic and the vector r


minus one half the vector

cxii.

=- \ r

X.

follows.

anti-self-conjugate

equal to the vector product of


of that dyadic and the vector r.
is

VECTOR ANALYSIS

298

Theorem

Any

" = -3

(0-0*)

<2>xX

anti-self-conjugate dyadic
It

degree of nullity.

$" possesses one

a uniplanar dyadic the plane of

is

whose consequents and antecedents

is

perpendicular to

<P X ",

the vector of 0.

This theorem follows as a corollary from equations (44).


Theorem : Any dyadic
may be broken up into two parts
to
self-conjugate and the other equivalent
minus one half the vector of <P used in cross multiplication.

which one

of

is

d>

or symbolically

Any

113.]

it

0'

<P

r,

X.

(45)

vector c used in vector multiplication defines a

For

linear vector function.

Hence

0'

must be

possible to represent the operator c

as a

This dyadic will be uniplanar with plane of

dyadic.

antecedents and

consequents perpendicular to

c,

so that

be found as follows

(31)
(I

Hence
and
This

r =
=I

o)

may

( c
(I

o)

(c

I)

(I

c = r

I)

1}
.

c)

(I

be stated in words.

(I

r=

c)

{I

it

The dyadic may

will reduce all vectors parallel to c to zero.

By

its

I,

(o

= (c x I) r.
r = (c x I)
= r (c X I).
c)

I)}

r,

(46)

LINEAR VECTOR FUNCTIONS


Theorem : The vector

299

used in vector multiplication with


x c or c x I used in direct

a vector r is equal to the dyadic I

multiplication with

used as prefactors
I

and

If c precedes r the dyadics are to be

r.

follows

if c

r,

The dyadics

as postfactors.

I are anti-self-conjugate.

In case the vector

c is a unit vector the application of the


vector
r in a plane perpendicular to c is
operator
any
to
r
equivalent
turning through a positive right angle about
the axis c. The dyadic c X I or I x c where c is a unit vector
c

to

therefore turns any vector r perpendicular to c through a


If r were a vector
right angle about the line c as an axis.

lying out of a plane perpendicular to c the effect of the dyadic


I X c or c x I would be to annihilate that component of r which
is

and turn that component

parallel to c

of r

which

is

perpen-

dicular to c through a right angle about c as axis.


If the dyadic be applied twice the vectors perpendicular to
r are rotated

through two right angles.

They

are reversed in

If it be applied three times they are turned through


three right angles.
Applying the operator I x c or c x I four
times brings a vector perpendicular to c back to its original

direction.

position.

The powers
(I

(I

c)

c)

of the dyadic are therefore

(c

I)

4
(I x c)

(I

c)

I)

=-I

(c

(c

(c

x
5

I)

its

- c c),

(I

=-c X
4
- c c,
I) = I
x

=I

It thus appears that the dyadic I

law as far as

=-

=c

or c

I,

(47)

I.

I obeys the

same

powers are concerned as the scalar imaginary

1 in algebra.

The dyadic Ixc

or

cxlisa

vectors perpendicular to
as an annihilator.

To

c.

quadrantal versor only for

For vectors

avoid this

effect

parallel to c it acts

and obtain a true

VECTOR ANALYSIS

300

vectors r in space it is merely necessary to add the dyad c c to the dyadic I x c or c X I.

quadrantal versor for

all

If

(48)

The dyadic
idemfactor.

imaginary V
plete

and

conjugate

appears as a fourth root of the


The quadrantal versor
is analogous to the
1 of a scalar algebra.
The dyadic
is comtherefore

two parts

consists of
;

If

114.]

and
i,

j,

of

which

c is anti-self-

are three perpendicular unit vectors

I
I

may

c c, self -conjugate.

Ixi = ixl = kj

as

jk,

xj=j x I = ik-ki,
x k = k x I = j i i j,

(49)

be seen by multiplying the idemfactor


I

= ii + jj + kk

and k successively.

These expressions represent


quadrantal versors about the axis i, j, k respectively combined
with annihilators along those axes.
They are equivalent,
into

i, j,

when used in direct multiplication, to i

x, jx,

k x

respectively,

(IXk)=(kxI) = -(ii+jj),

The expression
j,

(I

k)

is

an idemfactor for the plane of i and


In a similar mank.

but an annihilator for the direction

ner the dyad k k

is

an idemfactor for the direction

k,

but an

LINEAR VECTOR FUNCTIONS

301

annihilator for the plane perpendicular to k.


These partial
idemfactors are frequently useful.
If a, b, c are any three vectors and a', V, c' the reciprocal

system,

aa'

bb'

used as a prefactor is an idemfactor for all vectors in the


plane of a and b, but an annihilator for vectors in the direction

Used

c.

as a postfactor it

in the plane of
direction

c'.

a'

and

b',

is

an idemfactor for

all vectors,

but an annihilator for vectors in the

In like manner the expression


cc'

used as a prefactor is an idemfactor for vectors in the direction


c, but for vectors in the plane of a and b it is an annihilator.

Used

as a postfactor it is

direction

and

b',

If a

c',

that

an idemfactor for vectors

in the

but an annihilator for vectors in the plane of a


is,

for vectors perpendicular of c.

and b are any two vectors


(a x b) x I

=I

x I}r =

x b) x

(a

x b)

= b a - ab.

(50)

For
{(a

b)

(a

= bar abT = (ba

ab)r.

The vector a X b in cross multiplication is therefore equal to


the dyadic (b a
a b) in direct multiplication. If the vector
is used as a prefactor the dyadic must be so used.
(a xb)
r

(a

= (b a a b)
- ab).
b) = r (b a
r

r,

(51)

This is a symmetrical and easy form in which to remember


the formula for expanding a triple vector product.

VECTOR ANALYSIS

302

Reduction of Dyadics

be any complete dyadic and let r be a unit


the vector r'

Let

115.]

Then

vector.

Normal Form

to

r'= 0-r
a linear function of

is

r.

When

on

r takes

all

values consis-

that is, when the terminus


being a unit vector
the vector r'
of r describes the surface of a unit sphere,
varies continuously and its terminus describes a surface. This
tent with

its

surface

closed.

is

Theorem
to a

sum

It is fact

an

ellipsoid.

always possible to reduce a complete dyadic


of three terms of which the antecedents among
It is

themselves and the consequents among themselves are mutuThis is called the normal form of 0.
ally perpendicular.
<P

To

= ai'i +

Jj'j

ck'k.

demonstrate the theorem consider the surface described

by

r'=0-r.

As

this is a closed surface there

which makes

must be some

direction of r

maximum

or at any rate gives r' as great


a value as it is possible for r' to take on. Let this direction
of r be called i, and let the corresponding direction of r'
r'

the direction in which r' takes on a value at least as great as


be called a. Consider next all the values of r which
any
lie in

a plane perpendicular to i. The corresponding values


r is a linear vector
owing to a fact that

of r' lie in a plane


1

This

may

be proved as follows

= *-L r'srUe-i.
Hence
rr=l= r'. (* c ~ 1 '* ~ l)'i' = r' V-r'.
By expressing V in nonion form, the equation r' V r' = is seen to be of the second
* = *>*

degree.
is

Hence

the ellipsoid.

r'

describes a quadric surface.

The

only closed quadric surface

LINEAR VECTOR FUNCTIONS

303

Of these values of r' one must be at least as


great
any other. Call this b and let the corresponding direction
r be called j.
Finally choose k perpendicular to i and j

function.
as

of

upon the
value of

changes

and j. Let c be the


which corresponds to r = k. Since the
dyadic
j, k into a, b, c it may be expressed in the form
positive side of plane of

r'
i,

= &i +

<P

show

It remains to

bj

ck.

that the vectors

a, b, c

as determined

above are mutually perpendicular.

= (ai +
dr' = (ai
r'

-f-

r'. dr'

When

r'

r is parallel to

If

dr

dr

further

r.

is

parallel to

i,

r is parallel to

'

r'

bj

dr

dr.

hence must be

a unit vector

is

Hence when

dr

is

always

r is parallel to i

perpendicular to

r'b

ck

r'

maximum and

Since r

perpendicular to k,

is

r' is

i,

perpendicular to dr'.
perpendicular to

bj

dr +

ai

+ck)r,
+ ck) .dr,

bj

j,

r'c

vanishes,

and

Hence when

vanishes.

if

r is

But when
perpendicular to both b and c.
r' is parallel to a.
Hence a is perpendicular

is
i,

Consider next the plane of j and k and the


plane of b and c. Let r be any vector in the plane of j and k.
to b

and

c.

= (bj + ck)r,
dr' = (bj + ck) dr,
dr + r' c
'dr' = r'''b
r'

When

r takes the value

hence

is

j,

r' is

perpendicular to dr'.

maximum
Since r

is

dr.

in this plane

a unit vector

and
it is

VECTOR ANALYSIS

304

Hence when
and

perpendicular to dr.
is

perpendicular to

Hence
value

c is zero.

r'

But when

Consequently b

b.

It has therefore
c,

j,

and that b

is

antecedents of

denoted by

i',

is

a,

J, c

is

parallel

r is parallel to

perpendicular to

been shown that a

perpendicular to

j', k'.

Then

r'

j,

dr

takes the

perpendicular to b and
Consequently the three

is

c.

They may be

takes the form

the dyadic

+6j'j

j,

to

c.

are mutually perpendicular.

= ai'i
where

ck'k,

(52)

are scalar constants positive or negative.

Theorem: The complete dyadic


may always be
116.]
reduced to a sum of three dyads whose antecedents and

whose consequents form a right-handed rectangular system


of unit vectors and whose scalar coefficients are either all
positive or all negative.

&=

(ai'i

&j'j

ck'k).

(53)

The proof of the theorem depends upon the statements


made on page 20 that if one or three vectors of a right-handed
system be reversed the resulting system is left-handed, but
if two be reversed the
system remains right-handed. If then
one of the coefficients in (52) is negative, the directions of the
other two axes may be reversed. Then all the coefficients
are negative. If two of the coefficients in
(52) are negative,
the directions of the two vectors to which
they belong may

be reversed and then the coefficients in

Hence

are all positive.

in

any case the reduction to the form in which all


the coefficients are positive or all are
negative has been
performed.
As a limiting case between that in which the coefficients
are all positive

and that in which they are

all

negative comes

LINEAR VECTOR FUNCTIONS


the case in which one of them
takes the form

and

= ai'i
The

The dyadic then

zero.

is

305

-f Jj'j

(54)

a and

J may always be taken


a proof similar to the one given above it is
possible to show that any planar dyadic may be reduced to
this form.
The vectors i'andj'are perpendicular, and the
is

planar.

positive.

vectors

coefficients

By

and j

are likewise perpendicular.

might be added that in case the three coefficients a, 6, c


in the reduction (53) are all different the reduction can be
It

performed in only one way. If two of the coefficients (say


a and &) are equal the reduction may be accomplished in an

number

ways in which the third vector k' is always


the same, but the two vectors i', j' to which the equal coefficients belong may be any two vectors in the plane perinfinite

of

In

pendicular to k.

all

these reductions the three scalar

have the same values as in any one of them.


If the three coefficients a, &, c are all equal when
is reduced
coefficients will

to the normal

form

may be accomplished
The three vectors
ways.

(53), the reduction

in a doubly infinite

number

of

space.

may be any right-handed rectangular system in


In all of these reductions the three scalar coefficients

are the

same

i',

j',

k'

as in

any one of them.

These statements will

not be proved. They correspond to the fact that the ellipsoid


which is the locus of the terminus of r' may have three
different principal axes or it

maybe an

ellipsoid of revolution,

or finally a sphere.

Theorem :

Any

self-conjugate

the form

where
Let

a, &,

=
and

ii

dyadic

&jj

may

be expressed in

+ c kk

(55)

are scalars, positive or negative.


<P

= ai'i +

fcj'j

6jj'
20

+ck'k,

ckk',

(52)

VECTOR ANALYSIS

306

= a*i'i' + & 2 j7 + c 2 k'k',


*0 = a ii + 6 2 jj + c kk.
<P =
m

4>.<P

<P

Since

<t>

+kk = i'i' +j'j + k'k',


= (6 2 -a j'j' + (e 2 -a 2)k'k',
j

-a 2 1

a2 I) would annihilate
were not parallel (<P 2
two vectors i and i' and hence every vector in their plane.
If

and

i'

2
a 2 I) would therefore possess two degrees of nullity
(0
and be linear. But it is apparent that if a, J, c are different

this dyadic is

not linear.

be parallel.

In like manner

k and

k' are parallel.

is

Hence i and i' must


may be shown that j and j

planar.

it

The dyadic

0=
where

It

aii

a, Z, c are positive or

Jjj

',

therefore takes the form

+ ckk

negative scalar constants.

Double Multiplication^
117.]

Definition

The

double dot product of

two dyads

is

the scalar quantity obtained by multiplying the scalar product


of the antecedents by the scalar product of the consequents.

The product

is

denoted by inserting two dots between the

ab:cd

= ac bd.

(56)

This product evidently obeys the commutative law

ab:cd
1

The

= cd:ab,

researches of Professor Gibbs upon Double Multiplication are here


first time.

printed for the

LINEAR VECTOR FUNCTIONS

307

and the

distributive law both with regard to the dyads and


with regard to the vectors in the dyads. The double dot

product of two dyadics is obtained by multiplying the product out formally according to the distributive law into the
sum of a number of double dot products of dyads.
If

<P

and

=a

=a
=c

b 1 :c 1 d 1

b1

^.

a2 b2

a3 b 3

c2

d2

c 3 d3

...

+ ajb^Oadg +

a 1 b 1 :o 8 d s

a 2 b 2 :c 1 d 1

a 2 b 2 :c 2 d 2

a 2 b 2 :c 3 d 3

...

agbgZCjdj

a 3 b 3 :c 2 d 2

a 3 b 3 :c 3 d 3

...............

Definition:

The

double cross product of

two dyads

(56)'

(56)"
is

the

dyad of which the antecedent is the vector product of the


antecedents of the two dyads and of which the consequent is
the vector product of the consequent of the two dyads. The
product

dyads

is

denoted by inserting two crosses between the

ab^cd^axc bxd.

(57)

This product also evidently obeys the commutative law

VECTOR ANALYSIS

308

and the distributive law both with regard to the dyads and
with regard to the vectors of which the dyads are composed.
The double cross product of two dyadics is therefore defined
as the formal expansion

distributive

of the product according to the

law into a sum of double cross products of

dyads.

a3 b3

LINEAR VECTOR FUNCTIONS


Theorem

The double dot product

of

two fundamental

unity or to zero according as the two

to

is

309

dyads
equal
dyads are equal or

different.

ij:ki

= ik

= 0.

Theorem: The double cross product of two fundamental


dyads (12)

is

equal to zero

either the antecedents or the

if

But

consequents are equal.

if

sequents are equal the product

mental dyads

That

neither antecedents nor con-

equal to one of the fundataken with a positive or a negative sign.


is

is
ij

=i

x k

ij*ki=ixk jxi = + jk.


There exists a scalar

product of three dyads in


which the multiplications are double. Let <P, W, Q be any
three dyadics.

triple

The expression
*

is

WiQ

a scalar quantity. The multiplication with the double


must be performed first. This product is entirely in-

cross

dependent of the order in which the factors are arranged or


Let ab, cd, and ef be

the position of the dot and crosses.


three dyads,

ab

cd:ef = [ace] [bdf].

(59)

the product of three dyads united by a double cross


and a double dot is equal to the product of the scalar triple
product of the three antecedents by the scalar triple product

That

is,

From this the statement made


of the three consequents.
above follows. For if the dots and crosses be interchanged
the order of the factors be permuted cyclicly the two
If the cyclic order of
scalar triple products are not altered.

or

if

VECTOR ANALYSIS

310

is reversed each scalar triple product changes


Their
product therefore is not altered.
sign.
may be multiplied by itself with double
dyadic
118.]

the factors

Let

cross.

$<P

= &l + bm +

= (al + bm + en) * (al + bm + en)


= a x a Ixl+axb Ixm+axc Ixn
+ bxa mxl + bxb m x m + b x c mxn
+ cxa nxl + cxb nxm + cxc nxn.

The products
equal in

en

The

main diagonal vanish.


Hence

in the

pairs.

= 2 (bxc mxn + cxa nxl + axb

<P*0

and

If a, b, c

1,

m, n are non-coplanar this

0*0=-__A
[a be]

The product

is

+b'm' +

a species of power of 0.

The

notation

Ixm).

(60)

be written

[Imn]

garded as a square of
to represent this

.(a'l'
v

may

others are

c'n'>

It

will be

(60)'

may be

re-

employed

product after the scalar factor 2 has been

stricken out.

0%

= 0*0
%- = (b x c mxn + cxa nxl + axb Ixm) (61)
3S

The

product of a dyadic
expressed as the
three dyads with itself twice repeated is

triple

(b

sum

of

0$0:0 = 3 0^:0
mxn + cxa nxl + axb Ixm)
:

(al

+ bm +

en).

In expanding this product


every term in which a letter is
repeated vanishes. For a scalar triple product of three vec-

LINEAR VECTOR FUNCTIONS


two of which are equal

tors

is

311

Hence the product

zero.

reduces to three terms only


2

:<P=[bca] [mnl] + [cab] [nlm] + [abc] [Imn]

or

<P

<P*0:<Z>

The

3 [abc]

product of a dyadic by

triple

[Imn]

6 [abc] [Imn].
itself

twice repeated

is

equal to six times the scalar triple product of its antecedents


multiplied by the scalar triple product of its consequents.
The product is a species of cube. It will be denoted by 3
after the scalar factor 6 has been stricken out.
d>
00 =
= ---

0.
119.]

If

(62)

[abc] [Imn].

be called the second of

and

3,

the third of

0, the following theorems may be stated concerning the


seconds and thirds of conjugates, reciprocals, and products.

Theorem

The second

of the conjugate of a dyadic

to the conjugate of the second of that dyadic.

the conjugate

is

The

is

equal

third of

equal to the third of the dyadic.

<*><=
4>,

(63)

= (*<-),

Theorem: The second and third of the reciprocal of a


dyadic are equal respectively to the reciprocals of the second

and

third.

<*-'),
1

(<^- ) 3

Let

=
=

(*,)-'=*,-*

= al
>-i = V
+
a.'

= <V

m'b'+n'c'

pi

[abc]

(36)

VECTOR ANALYSIS

812

(0a )-i

[abc] [Imn] (la


la
j

+ mb +

nc)

+ mb + nc

;2

~[a'b'c'] [1'm'n']'

=1
(0,)' =

But

[a'Vc'] [abc]

Hence

and

(0"

[1'm'n']
1

).

[Imn]

= 1.

= ^-

[abc] [Imn]
(0-i) 3 =[a'b'c'] [1'm'n'].

Hence
Theorem: The second and third of a product are equal
respectively to the product of the seconds and the product of
the thirds.

(0-^2

=<V^2

(*.*),*.*,.
of

Choose any three non-coplanar vectors 1, m, n as consequents


and let 1', m',n' be the antecedents of W.

= al +

(0.?P) 2

m+

cn,

=bxc exf + cxa fxd + axb

dxe,

= bxcmxn + cxa nxl + axb Ixm,


f x d +
x m' dxe.
^2 = m' x n' e x f + n' x
Hence ^2 ?T2 = bxc exf + cxa fxd + axb dxe.
Hence
(0 V)^ = ^ Vz
((P. F) 3 = [abc] [def]
cP

1'

1'

LINEAR VECTOR FUNCTIONS


3

=[abc] [Imn],

r3 =[l'm'n']
Hence

<P

Hence

(0

Theorem

[defj.

[a b c] [d e

r) 3

The second and

f].

rs

third of a

of a dyadic are

power

equal respectively to the powers of the second


the dyadic.

Theorem

The

W=0

(0") 2

(*")

= (*)*=*."

The second

313

n
2

of the idemfactor

third of the idemfactor

is

and third of

is

the idemfactor.

unity.

=I
1 8 = 1.
I2

(67)

Theorem: The product of the second and conjugate of


a dyadic is equal
idemfactor.

the product of

to

Z
<Z>

=b

[l

be non-coplanar.
(b

=0

and the
( 68 >

S I,

mxn-fcxa nxl + axb Ixm,


= la + mb + nc,

The antecedents

.4> C

the third

m n]

a, b, c

(b

x a b

of the dyadic

may

x b

c).

be assumed to

Then

X a b+

x b

c)

[abc]

[ab

c] (a'a

+Vb+

c'

c)

I.

Hence
120.]

sum

of

Let a dyadic
three dyads of

non-coplanar.

be given.

Let

it

be reduced to the

which the three antecedents are

VECTOR ANALYSIS

314

= al +
=b

<P

m+

cn,

mxn + cxa nxl + axb


3

x m,

[abc] [Imn].

Theorem: The necessary and sufficient condition that a


be different from
be complete is that the third of
dyadic
zero.

For

it

was shown (Art. 106) that both the antecedents and

consequents of

the

Hence the two

a complete

are

dyadic

scalar triple products

non-coplanar.

which occur in

cannot vanish.

Theorem: The necessary and sufficient condition that a


shall vanish but the
be planar is that the third of
dyadic
second of

shall not vanish.

was shown (Art. 106) that if a dyadic


be planar its conn
and
must
be
1,
m,
sequents
planar
conversely if the conseIt

quents be coplanar the dyadic


dyadic

must vanish.

But

is

planar.

<P

Hence

for a planar

cannot vanish.

Since

a,

have been assumed non-coplanar, the vectors b x c, c x a,


a x b are non-coplanar.
Hence if 0% vanishes each of the
b, c

vectors

mxn, nxl, Ixm

linear.

But

and not

linear.

vanishes

that

is, 1,

this is impossible since the dyadic

m, n are colis

planar

Theorem: The necessary and sufficient condition that a


non-vanishing dyadic be linear is that the second of 0, and
consequently the third of 0, vanishes.
For if
be linear the consequents

1,

m,

n,

are collinear.

Hence
Z

their vector products vanish and the consequents of


vanish.
If conversely
vanishes, each of its consequents
Z

must be zero and hence these consequents of


are collinear.
The vanishing of the third, unaccompanied by the vanishing of the second of a dyadic, implies one degree of nullity.
The vanishing of the second implies two degrees of nullity.

LINEAR VECTOR FUNCTIONS


The vanishing
results

may

of the dyadic itself

is

315

The

complete nullity.

be put in tabular form.


3

<P
3

= 0,

= 0,
3

</>

is

0,
2

complete.
is

0,

= 0,
2

<P

planar.

(69)

is linear.

0,

It follows immediately that the third of

any anti-self-conjugate
For any such

dyadic vanishes; but the second does not.


dyadic is planar but cannot be linear.

Nonion Form.

be expressed in nonion form

If

121.]

Invariants of a Dyadic

Determinants}-

<P

= OII ii +
+
+

The conjugate

of

a 21

ij

12

ji+a 22 jj +
+ a 32 kj +

a 31 ki

i3

ik

23

jk

(13)

a 33 k k.

has the same scalar coefficients as 0, but

they are arranged symmetrically with respect to the


diagonal.

main

Thus

+
+

+
ki +
13
i
12 j

22 j j
2 3J

+
+

(70)

32 j k,

kk
33

<P may be computed.


Take, for instance, one
be required to find the coefficient of ij in <P2
What terms in can yield a double cross product equal to
must be i and
ij? The vector product of the antecedents
Hence the
must
be
the
of
j.
the vector product
consequents
k
the
and
and
antecedents must be j
consequents, k and i.

The second of

term.

Let

it

These terms are


x

33

kk = -

aai a 33

iJ

the third order.


1 The results hold
only for determinants of
determinants of higher orders is through Multiple Algebra.

The extension

to

VECTOR ANALYSIS

316

Hence the term

in ij in
(

This

is

the

first

<P

is

a 3l a 23 ~~ a21 a 8s)iJ'

minor of a lz in the determinant


"13

'11

23

"88

This minor

is

coefficient of

taken with the negative sign. That is, the


in <P2 is what is termed the cofactor of the

ij

The cof actor is merely


minor taken with the positive or negative sign
according as the sum of the subscripts of the term whose
coefficient of ij in the determinant.

the

first

first

minor

is

under consideration

is

even or odd.

The

co-

any dyad in <P2 is easily seen to be the cofactor of


the corresponding term in 0.
The cofactors are denoted
efficient of

generally by large letters.

An =
*M

is

the cofactor of a 11(

is

the cofactor of a 12-

is

the cofactor of

*33
6

a21

With

this notation the

12

aX 223
second of

<P

a-,.
32'

becomes

(71)

+ A sl

ki

The value of the third of


as the sum of three
dyads

= (a n i+a 21 +
j

^32 k j

may

a 31 k)i

^33 kk.

be obtained by writing

LINEAR VECTOR FUNCTIONS

[Oil 1

21J

*81
j

This

is

k)

( a 2li

easily seen to be equal to the determinant

this reason

is

32

33

frequently called the determinant of

written
3

The

(72)

'22
(

is

22J

a 33 k)] [ijk]

'21

For
and

317

idea of the

determinant

(72)'

is

very natural when

is

regarded as expressed in nonion form. On the other hand


unless
be expressed in that form the conception of <P3 ,
the third of 0,

is

more natural.

The
most

reciprocal of a dyadic in nonion form


easily by making use of the identity

0^0o=^

s l

may

be found

(68)

or

or

Hence 0~l

=n
(73)
31

VECTOR ANALYSIS

318
If the

If

determinant be denoted by

is

a second dyadic given in nonion form as

ki

6 31

& 32

kj

6 33

kk,

two dyadics may readily be found


the
multiplication
by actually performing
of the

the product

a !3
a21 & 11

+
a 31

32

a 23
& 11

a 22

&21

& 32) J J

+ a 33 ^82) k J +

13

a 23

J 3l)

( a 21

+ a 32 J 21 +

a !2

ll

J 13

33 & 3l)

C a 31 & 12

J i

a 22

ki

a !3

23

a 21

J 23
(

& 12

tt

23

a 31 & 12

a 32 &28

33

a 83

a2

& 3s

a3

& 3s)

31

Since the third or determinant of a product is equal to the


product of the determinants, the law of multiplication of
determinants follows from (65) and (74).

LINEAR VECTOR FUNCTIONS

319

"12

22
31

'32

21
31

The

rule

33

+
+
+

"u

21

'23

+
+
+

a 12
a 22
a 32

may

#]

^22

31

+
+
+

13

23

& 31

a 33

^31

'13

a2

as

23

(76)

32

To

be stated in words.

multiply two deter-

minants form the determinant of which the element in the


with

row and nth column

is

the

sum

elements in the with row of the

column

of the products of the

first

determinant and nth

of the second.

If

<P

= al + bm + cn,

= bxc mxn + cxa nxl + axb Ixm.

Then

= (02 ) 3 =

[b

cxa axb]

xc

[mxn nxl Ixm]

Hence
Hence

-12

21
L

31

The determinant
the third order
122.]

is

22
32

is

13

an

a iz

23

a 21

*22

a<

33

a 31

a3

a 33

(7T)

of the cofactors of a given determinant of


equal to the square of the given determinant.

dyadic

has three scalar invariants

which are independent


expressed. These are

three scalar quantities

which

of the

the scalar of 0, the scalar of the second of 0,


or determinant of 0.
quantities are

If

that

is

form in

and the third

be expressed in nonion form these

VECTOR ANALYSIS

320

(78)

'31

No

22

''32

matter in terms of what right-handed rectangular system


may be expressed these quantities are

of these unit vectors

the same.
in the

sum

The

scalar of

The

main diagonal.

of the first

is

sum

the

of the three coefficients

scalar of the second of

is

the

minors or cofactors of the terms in the

main diagonal The third of


is the determinant of the
These three invariants are by far the most
coefficients.
important that a dyadic
Theorem : Any dyadic
the three invariants

By

S,

ZS,

-xl\

(68)

possesses.

a cubic equation of which


are the coefficients.

satisfies

(0

- xT) c =

-*

(0

(0-xl) s =

II
22

a;

Hence
as

may

- xl\

This equation
scalar x.

That

2S

x*

be seen by actually performing the expansion.

(0

dyadic

x I) 3

(0

is

(0

x I) c

23

an identity holding for

It therefore holds,

if

xz

all

xs

values of the

in place of the scalar

a?,

the

which depends upon nine scalars be substituted.

is

But the terms upon the


0*

left are identically zero.

8 0*

2S

& I

= 0.

Hence
(79)

LINEAR VECTOR FUNCTIONS

321

This equation may be called the Hamilton-Cayley equation.


Hamilton showed that a quaternion satisfied an equation
analogous to this one and Cayley gave the generalization to

matrices.

matrix of the

equation of the nth degree.


of dyadics

and the theory

Tith

order satisfies an algebraic


the theory

The analogy between

In

of matrices is very close.

fact,

be regarded as a matrix of the third order and


a
matrix
of the third order may be looked upon as
conversely
a dyadic. The addition and multiplication of matrices and
a dyadic

may

dyadics are then performed according to the same laws.


generalization of the idea of a dyadic to spaces of higher
dimensions than the third leads to Multiple Algebra and the

theory of matrices of orders higher than the third.

SUMMARY OF CHAPTER

A
when

vector r'

said to be a linear function of a vector r

is

the components of r' are linear homogeneous functions


components of r. Or a function of r is said to be a

of the

linear vector function of r

two vectors

is

sum

the

f (ri

when

the function of the

sum

of

of the functions of those vectors.

r 2)

= f (ri ) + f (r 2

(4)

).

These two ideas of a linear vector function are equivalent.


sum of a number of symbolic products of two vectors,

which are obtained by placing the vectors in juxtaposition


without intervention of a dot or cross and which are called
a Greek
dyads, is called a dyadic and is represented by

function of
dyadic determines a linear vector
vector
that
with
a vector by direct multiplication

capital.

= aj
r

=a

b1

!>!

a2 b 2

a2 b2
21

+
r

a3 b 3

CO

-f

a3 b3

(8)

VECTOR ANALYSIS

322

Two
upon
is,

dyadics are equal when they are equal as operators


vectors or upon three non-coplanar vectors. That

all

when
r

= W

r for all values or for three

coplanar values of
or

= r W for all values

g0r = s?

or

(10)

r,

or for three non-

coplanar values of
r

non-

r,

r for all values or for three non-

coplanar values of r and

s.

may be represented by a dyadic.


the
law of multiplication with
distributive
Dyads obey
to
the
two
vectors
composing the dyad
regard

Any

linear vector function

H---- )

= al + am + an +
+

bl

+ bm + bn +

cm + en +

f cl +

(11)'

Multiplication by a scalar is associative. In virtue of these


two laws a dyadic may be expanded into a sum of nine terms
by means of the fundamental dyads,
ii,

ij,

ik,

ji,

Jj,

jk,

ki,

kj,

kk,

(12)

= a n ii + a 12 ij + a 13 ik,
= 2 i + 22 j j + 2 k
= a 31 ki+ 32 kj + a 33 kk.

as

i J

If

3 J

>

( 13 )

two dyadics are equal the corresponding coefficients in


form are equal and conversely.

their expansions into nonion

LINEAR VECTOR FUNCTIONS

Any

dyadic

which the

323

may be expressed as the sum of three dyads of


antecedents or the consequents are any three
This expression of the dyadic

given non-coplanar vectors.

is

unique.

The symbolic product ab known

as a

dyad

is

the most

general product of two vectors in which multiplication by a


It is called the indeterminate product.
scalar is associative.

The product imposes


b.

five conditions

upon the vectors a and

Their directions and the product of their lengths are

The

determined by the product.

and vector products


A scalar and

scalar

are functions of the indeterminate product.

may be obtained from any dyadic by inserting a dot


and a cross between the vectors in each dyad. This scalar
and vector are functions of the dyadic.
a vector

= a bj + a b 2 + a
<P X =
aj x b + a x b + a
0j = i.0.i + j.0.j +
=a
a
a
a

~^~

<>23

32 )

(i.
i

'

b3

(18)

x b3 +

(19)

k.0k

(20)

'

0.j-j 0.i)k
.

<>31

- a 13> +
J

(21)

Ol2

2l)

The direct product of two dyads is the dyad whose antecedent and consequent are respectively the antecedent of the
first dyad and the consequent of the second multiplied by
the scalar product of the consequent of the
the antecedent of the second.

(ab)

(cd)

(b.c)ay.

first

dyad and

(23)

direct product of two dyadics is the formal expansion,


of the product into the
according to the distributive law,

The

VECTOR ANALYSIS

324

sum

Direct multiplication of dyadics


or of dyadics and a vector at either end or at both ends obeys
the distributive and associative laws of multiplication. Conof products of dyads.

sequently such expressions as

0.V.T,

&.<P.,

Q.W.Q

s.tf.r.r,

(24X26)

be written without parentheses; for parentheses may


be inserted at pleasure without altering the value of the

may

product. In case the vector occurs at other positions than


at the end the product is no longer associative.
The skew product of a dyad and a vector may be defined

by the equation

(ab) x

x (ab)

= a b x r,
= r x a b.

(28)

of a dyadic and a vector is equal to the


formal expansion of that product into a sum of products of
dyads and that vector. The statement made concerning the

The skew product

law for direct products holds when the vector is


connected with the dyadics in skew multiplication. The
associative

expressions
r

<P

4>.xr,

?F,

may be

<P

$ x

s,

s,

written without parentheses and parentheses

(29)

may

be

inserted at pleasure without altering the value of the product.

Moreover
s

(r

<P)

= (s
.

(r

r)
5T)

0,

(<P

d>

r)

r)

But the parentheses cannot be omitted.


The necessary and sufficient condition
be reduced to the

sum of two dyads or


when expressed as

to zero

is

dyads of

which the antecedents

that,

(r

s),

(31)'

that a dyadic

the

sum

may

dyad or

to a single

of

(or consequents) are

three

known

LINEAR VECTOR FUNCTIONS

325

to be non-coplanar, the consequents (or


antecedents) shall
be respectively coplanar or collinear or zero.
complete
dyadic is one which cannot be reduced to a sum of fewer
than three dyads.
planar dyadic is one which can be

reduced to a sum of just two dyads. A linear dyadic


which can be reduced to a single dyad.

A complete

dyadic possesses no degree of


no direction in space for which it is an

is

is

one

There

nullity.

annihilator.

planar dyadic possesses one degree of nullity. There is one


direction in space for which it is an annihilator when used as
a prefactor and one

when used

as a postfactor.

dyadic possesses two degrees of

nullity.

linear

There are two

independent directions in space for which it is an annihilator


when used as a prefactor and two directions when used as a
postfactor.

A zero dyadic

possesses three degrees of nullity


It annihilates every vector in space.

or complete nullity.
The products of a complete dyadic and a complete, planar,
or linear dyadic are respectively complete, planar, or linear.
The products of a planar dyadic with a planar or linear dyadic
are respectively planar or linear, except in certain cases where
between the consequents of the

relations of perpendicularity

dyadic and the antecedents of the second introduce one


The product of a
of nullity into the product.
linear dyadic by a linear dyadic is in general linear but in
first

more degree

case the consequent of the first

is

perpendicular to the ante-

cedent of the second the product vanishes.

any dyadic by a zero dyadic

The product

of

is zero.

dyadic which when applied to any vector in space reproduces that vector is called an idemfactor. All idemfactors
are equal

and reducible to the form

= ii +
I = aa' +
I

Or
The product

of

+ kk.
bb' + cc'.

(33)

jj

any dyadic and an idemfactor

(34)
is

that dyadic.

VECTOR ANALYSIS

326
If the

product of two complete dyadics

is

equal to the idem-

commutative and either

are

factor the

is
called
dyadics
the reciprocal of the other.
complete dyadic may be
of
a
from
either
end
canceled
product of dyadics and vectors
as in ordinary algebra ; for the cancelation is equivalent to

multiplication by the reciprocal of that dyadic.

dyadics possess no reciprocals.

They correspond

Incomplete
to zero in

ordinary algebra. The reciprocal of a product is equal to the


product of the reciprocals taken in inverse order.

(0.
The conjugate

gr)-i

of a dyadic

=
is

sr-i. 0-i.

(38)

the dyadic obtained by inter-

changing the order of the antecedents and consequents. The


conjugate of a product is equal to the product of the conjugates taken in the opposite order.

(0.

)=

4> c .

(40)

The conjugate of the reciprocal is equal to the reciprocal of


the conjugate.
dyadic may be divided in one and only
one way into the sum of two parts of which one is selfconjugate and the other anti-self-conjugate.

= ^(0+

0c)

+ 2L (0-0c)-

(43)

Any anti-self-conjugate dyadic or the anti-self-conjugate


part of any dyadic, used in direct multiplication, is equivalent
to minus one-half the vector of that
dyadic used in skew
multiplication.

]r.(0-0 )=-lrx0 x
ff

A dyadic of the

(44)

form c X I or I x c is anti-self-conjugate and


used in direct multiplication is
equivalent to the vector c
used in skew multiplication.

LINEAR VECTOR FUNCTIONS


Also

The dyadic

= (I x o)
4> = (I x
c)

I or I

c,

r == (o

where

c is

(c

I)

327
(46)

r,

I)

0.

a unit vector

rantal versor for vectors perpendicular to c

a quadand an annihilator
is

Ixc

The dyadic
+ cc is a true
parallel to c.
for
all
versor
vectors.
The
of
these dyadics
powers
quadrantal
behave like the powers of the imaginary unit
1, as may

for vectors

be seen from the geometric interpretation.


unit vectors

i, j,

Applied to the

k
I

= kj

k, etc.

The vector a x b in skew multiplication


(a X b) x I in direct multiplication.

=1

(a x b) x I
(a
r

b)

x (ax

=ba

x (a x b)

(b a

is

a b)

-(49)

equivalent to

ab

(50)

b)=r (ba-ab).

(51)

of three
complete dyadic may be reduced to a sum
and the
themselves
dyads of which the antecedents among

a right-handed
consequents among themselves each form
vectors and of which the
rectangular system of three unit
all negative.
or
scalar coefficients are all positive

0=
This

is

(ai'i

&j'j

called the normal form of the dyadic.

plete dyadic

may

(53)

ck'k).

An

incom-

be reduced to this form but one or more of

is unique in case
case
In
they are not
the constants a, 6, c are different.
more than
different the reduction may be accomplished in
one way. Any self-conjugate dyadic may be reduced to
the normal form
& = aii + 6jj + ckk,
(55)

the coefficients are zero.

in

which the constants

The reduction

a, &, c

are not necessarily positive.

VECTOR ANALYSIS

328

The double dot and double


is

cross multiplication of dyads

defined by the equations

= a c b.d,
abcd = axc bxd.
ab

(56)

(57)

The double dot and double

cross multiplication of dyadics


obtained by expanding the product formally, according to
The
the distributive law, into a sum of products of dyads.
double dot and double cross multiplication of dyadics is comis

mutative but not associative.

is

One-half the double cross product of a dyadic


If
called the second of 0.

by

= al + bm + cn,
= 10x0 = bxc mxn+cxa nxl + axb Ixm.
2

itself

<P

(61)

One-third of the double dot product of the second of


and
and is equal to the product of the
called the third of

is

scalar triple product of the antecedents of


triple

and the

scalar

product of the consequent of 0.


Z

The second
The third of

= \0* 0: 0=[abc]

of the conjugate

the conjugate

is
is

[1m n].

(62)

the conjugate of the second.


equal to the third of the

The second and

third of the reciprocal are


the reciprocals of the second and third of the second and
third of a dyadic.
The second and third of a product are the
original dyadic.

products of the seconds and thirds.

(0.3r),=

(*

.JTa

).!->>,

(65)

LINEAR VECTOR FUNCTIONS


The product of
to the

the second

product of the third

329

and conjugate of a dyadic


and the idemfactor.

is

0^'0 C =03 I

(68)

The

conditions for the various degrees of


nullity
expressed in terms of the second and third of 0.

=
3

is

0,

0,

= 0,
3

is

may

be

complete

0,

= 0,
2

equal

planar

(69)

is linear.

0,

The

closing sections of the chapter contain the expressions


(70)-(78) of a number of the results in nonion form and the

deduction therefrom of a number of theorems concerning


determinants. They also contain the cubic equation which is
satisfied

by a dyadic 0.
s

This

is

cients

0q

= 0.

(79)

The coefficalled the Hamilton-Cayley equation.


are the three fundamental scalar inS , <P2/S , and
3

variants of 0.

EXERCISES ON CHAPTER

Show

1.

that the

two

given in Art. 98 for

definitions

a linear vector function are equivalent

Show

that the reduction of a dyadic as in (15) can be


accomplished in only one way if a, b, c, 1, m, n, are given.
2.

3.

Show

4.

from

5.
r,

that

if

and

0xr =

a x

xr
SF

for

&&

any value of

unless both

r different

and

the line of their consequents is parallel to

Show
then

&) c

must equal

zero, then

linear

of

(0 X

Show

that

= 0.

if

for

W are

r.

any three non-coplanar values

VECTOR ANALYSIS

330
6.

Prove the statements made in Art. 106 and the con-

verse of the statements.


7.
<P

Show

and

that

if

is

complete and

Definition

then

Give the proof by means of theory

are equal.

developed prior to Art. 109.


8.

Q=

if <P

Two

&=

dyadics such that

that

are said to be
two dyadics that are commutative
if
Show
that
number
of
dyadics are homogehomologous.
any
neous to one another, any other dyadics which may be obtained
from them by addition, subtraction, and direct multiplication
is

to say,

are homologous to each other and to the given dyadics. Show


also that the reciprocals of homologous dyadics are homolo-

l
l
or
0,
Justify the statement that if
which are equal, be called the quotient of
then
the
3F,
by
rules governing addition, subtraction, multiplication and
division of homologous dyadics are identical with the rules

gous.

it being
governing these operations in ordinary algebra
understood that incomplete dyadics are analogous to zero,

and the idemfactor, to unity.

Hence the algebra and higher

analysis of homologous dyadics is practically identical with

that of scalar quantities.

=cx

9.

Show

that (I

10.

Show

that whether or not

c)

abxc+bcxa+c

a, b, c

x b

(c

I)

<P

=c x

0.

be coplanar

[a b c] I

bxca+cxab+axbc=[abc]I.

and
11.

and

If a, b, c are

coplanar use the above relation to prove

the law of sines for the triangle and to obtain the relation
with scalar coefficients which exists between three coplanar
vectors.

This

may

be done by multiplying the equation by a

unit normal to the plane of a, b,


12.

What

is

and

the condition which

coefficients in the

c.

must

subsist

between the

expansion of a dyadic into nonion form

if

LINEAR VECTOR FUNCTIONS


the dyadic be self-conjugate?

What,

if

331

the dyadic be anti-

self -conjugate ?
13.

Prove the statements made

number

of

ways

in

in Art. 116 concerning the

which a dyadic may be reduced

to

its

normal form.
14.

The necessary and

sufficient condition that

be zero

self-conjugate dyadic

an

anti-

that the vector of the

is

dyadic shall be zero.


15.

Show

that

be any dyadic the product

if

<P C is

self-conjugate.
16.

Show how

to

make

use of the relation

<P X

demonstrate that the antecedents and consequents


conjugate dyadic are the same (Art. 116).
17.

Show

that

and
18.

4> $
z

(<P

Show

that

+ W\ =

if

Z
<P

=
2

4>*

Show

that

20.

Show

21.

Show

changed by

that

= 4>3 +
(0 + ) 3 = 3 +
e f) 3

of a self-

V + Wy

the double dot product

(0 +

to

<P

by itself vanishes, the dyadic vanishes.


condition for a linear dyadic in the form
19.

e
2

of a dyadic

Hence obtain the


<P

<P
2
:

<P

<P

= 0.

f.

V+

Vz +

that the scalar of a product of dyadics is unThat is


cyclic permutation of the dyadics.

CHAPTER VI
ROTATIONS AND STRAINS
123.]

IN the foregoing chapter the analytical theory of


been dealt with and brought to a state of

dyadics has

completeness which

is

nearly final for practical purposes.

There are, however, a number of new questions which present


themselves and some old questions which present themselves
under a new form when the dyadic is applied to physics
or geometry. Moreover it was for the sake of the applica-

them was developed. It is


then the object of the present chapter to supply an extended

tions of dyadics that the theory of

application of dyadics to the theory of rotations

and

to develop, as far as

may

and

strains

appear necessary, the further

analytical theory of dyadics.

That the dyadic

$ may

of space has already

be used to denote a transformation

been mentioned.

knowledge

of the

precise nature of this transformation, however, was not needed


at the time.
Consider r as drawn from a fixed origin, and r'
as

drawn from the same

Let now

origin.
r'

= 0-r.

This equation therefore may be regarded as defining a transformation of the points P of space situated at the terminus of
r into the point P', situated at the terminus of r'. The origin
remains fixed.

Points in the

the finite regions of space.

becomes a point

regions of space remain in


point upon a line

finite

Any

r = b
r' =

4-

x a
b

-f

a.

ROTATIONS AND STRAINS


Hence

straight lines go over into


parallel to the same line a go over

same

333

straight lines

and

lines

by the transformation into

In like manner planes


and
the
into
of
over
planes
parallelism is invariant.
quality
go
Such a transformation is known as a homogeneous strain.
lines parallel to the

line

a.

strain is of frequent occurrence in physics. For


of the infinitesimal sphere in a fluid
the
deformation
instance,
(Art. 76) is a homogeneous strain. In geometry the homo-

Homogeneous

geneous strain is generally known by different names.


Or
called an affine collineation with the origin fixed.

known

as a linear

It is
it

is

The equa-

homogeneous transformation.

tions of such a transformation are

= an
y' =
z' = a si x + a^y +
Theorem

124.]

of the points of
<P ,
2

is

is

gives the transformation

If the dyadic

space which

due

to a

homogeneous

strain,

gives the transformation of plane areas


due to that strain and all volumes are magnified by

the second of

which

33

<P,

that strain in the ratio of

$3

the third or determinant of

to unity.

Let

The

<p

vectors

1',

= al + bm +

m', n' are changed by

en

into

and

the planes determined by m' and n', n'


transformed into the plane determined by b and

The dyadic which accomplishes

a and b.
(ft,
i

Hence
due to

=b

if s

Hence
and m' are

a, b, c.

1', 1'

c, c

and

a,

this result is

mxn + cxa nxl + axb

Ixm.

denote any plane area in space, the transformation


s by the area s' such that

replaces

s'

s.

VECTOR ANALYSIS

334

It is important to notice that the vector

area

be

is

if it

denoting a plane

not transformed into the same vector

denoted a

the latter case

upon

This

line.

acts

on

is

s'

as it

would

evident from the fact that in

whereas in the former case

acts

s.

To show

that volumes are magnified in the ratio of <# 3 to


choose
any three vectors d, e, f which determine the
unity
a
volume of
with the vecparallelepiped [d e f]. Express
tors

which form the reciprocal system

The dyadic
from the

to d,

e,

f as consequents.

changes d, e, f into a, b, c (which are different


above unless d, e, f are equal to 1', m', n').

a, b, c

Hence the volume [d

e f ] is

changed into the volume

[a

c].

[d'e'f]

Hence

The
But

[a

be]

[def]

3>

volume [a b cj to [d e f] is as 3 is to unity.
d, e, f were any three vectors which determine a parallelepiped. Hence all volumes are changed by
ratio of the

the vectors

the action of

in the

same

ratio

and

this ratio is as

dotations about a Fixed Point.

$3

is

to 1.

Versors

Theorem : The necessary and sufficient condition that


125.]
a dyadic represent a rotation about some axis is that it be
reducible to the form

= i'i+j'j +
where

i', j',

k'

and

i, j,

are

k'k

two right-handed rectangular

systems of unit vectors.

Let

(1)

= a>i +

+ 2k

ROTATIONS AND STRAINS


Hence

335

reducible to the given form the vectors i, j,


are changed into the vectors i', j', k' and any vector r
if

is

changed from

its

position relative to

change of shape

body

carries

suffers

effected.

is

i, j,

The

vectors

is

posi-

Conversely suppose the

into i',j', k'.

no change of shape

to a rotation.

k into the same

Hence by the transformation no


The strain reduces to a rotation

tion relative to i',j',k'.

which

i, j,

i,

that

suppose

is,

it

subjected

k must be carried into another

j,

right-handed rectangular system of unit vectors. Let these


be i', j', k'. The dyadic
may therefore be reduced to the

form

=
Definition

A dyadic

i'

+ j' j + k' k.

which
i'i

is

reducible to the form

+ j'j +

k'k

and which consequently represents a rotation

is

called a

versor.

Theorem: The conjugate and reciprocal of a versor are


equal,

and conversely

if

the conjugate

and

reciprocal of a

dyadic are equal the dyadic reduces to a versor or a versor


multiplied by the negative sign.

Let

<P

= i'i+j'j +

k'k,

0c = ii'+jj' +

0.

c = i'i' +

j'j'

0-*=
Hence the
second part

If

Hence

first

kk',

k'k'

0c.

part of the theorem

is

proved.

let

= ai + b j + ck,
<P c = ia + j b + kc,
0-0 c
aa + bb + cc = I.

fl-^00

=I
To

prove the

VECTOR ANALYSIS

336

Hence (Art. 108) the antecedents a, b, c and the consequents


Hence (page 87) they
a, b, c must be reciprocal systems.
must be either a right-handed or a left-handed rectangular
system of unit vectors. The left-handed system may be
changed

to a right-handed

one by prefixing the negative

Then

sign to each vector.

= i'i +
or

The

+ k'k,
j + k'k).

j' j

= -(i'i +

j'

third or determinant of a versor

is

evidently equal to

unity; that of the versor with a negative sign, to minus one.


Hence the criterion for a versor may be stated in the form

0.0 C = I,
Or inasmuch
if

0.

=I0I=1.

plus or minus one


only necessary to state that if

as the determinant of

C =I, it

is

(2)

0.0 C =I,
is

is

=I0I>0,

(2)'

a versor.

There are two geometric interpretations of the transformadue to a dyadic


such that

tion

=I

= -(i'i +
The transformation due
reflection in the origin.

rotation about a definite

to

is

j' j

=-1

(3)

k'k).

one of rotation combined with

The dyadic i'i+j'j + k'k causes a


The negative
axis
it is a versor.

sign then reverses the direction of every vector in space and


replaces each figure by a figure symmetrical to it with respect
to the origin.

By

k'

reversing the directions of

still

i'

and

= i'i+j'j
or

(i'i'

j'

the

remains right-handed and rectangular,

system
but the dyadic takes the form
i', j',

+ j'j'-k'k')

-k'k,
(i'i+j'

k'k).

ROTATIONS AND STRAINS


Hence

the transformation due to

337

a rotation due to

is

i'i+j j + k'k followed by a reflection in the plane of i' and


For the dyadic i'i' + j'j' k'k' causes such a transfor'

j'.

mation of space that each point goes over into a point symmetrically situated to it with respect to the plane of i' and j'.

Each

figure

Definition

is
:

therefore replaced by a symmetrical figure.


transformation that replaces each figure

by

a symmetrical figure is called a perversion and the dyadic


which gives the transformation is called a perversor.

The

criterion for a perversor

is

that the conjugate of a


and that the determi-

dyadic shall be equal to its reciprocal

nant of the dyadic shall be equal to minus one.

0.0 C =I,
Or inasmuch

as if

<P C

(3)

= I,

or minus one the criterion


.

I0I=-1.

the determinant must be plus


may take the form

= I,

\<

(3)'

0,

a perversor.

is

from geometrical considerations that the product of two versors is a versor of two perversors, a versor
It is evident

but of a versor and a perversor taken in either order, a


perversor.
126.]

If the axis of rotation

be the i-axis and

measured positive

of rotation be the angle q

if

the angle

in the positive

trigonometric direction, then by the rotation the vectors


that
i,j, k are changed into the vectors i',j', k' such
i'

j'

k'

The dyadic
tion

i'

cos q
j

= i'i+j'j +

+k

sin q

sin q,

+k

cos

q.

k'k which accomplishes

is

22

this rota-

VECTOR ANALYSIS

338
<P

i i

cos q

(j j

jk

kj

i i

sin q (k

k).

(4)

+ kk = I-ii,

jj

Hence

k k)

=I

cos q (I

i.

i)

sin q I

i.

(5)

more generally in place of the i-axis any axis denoted


unit vector a be taken as the axis of rotation and if as
the
by
before the angle of rotation about that axis be denoted by <?,
If

which accomplishes the rotation

the dyadic
<P

To show

=aa +

that this

rotation apply

it

a a)

cos q (I

is

sin q I

a.

(6)

dyadic actually does accomplish the


r.
The dyad a a is an idemfactor

to a vector

for all vectors parallel to a; but an annihilator for vectors

a a is an idemfactor
perpendicular to a. The dyadic I
in
all
vectors
the
for
plane perpendicular to a; but an
The dyadic I x a
annihilator for all vectors parallel to a.
is

a quadrantal versor (Art. 113) for vectors perpendicular


an annihilator for vectors parallel to a. If then

to a; but

r be parallel to a

<^r =

aar =

Hence

leaves unchanged
which
are parallel to
vectors)
d>

all
a.

vectors (or components of


If r is perpendicular to a

= cos q r +

r.

sin q a

r.

Hence the vector r has been rotated in its plane through the
angle q. If r were any vector in space its component parallel
to a suffers no change ; but its component perpendicular to a
is rotated about a
through an angle of q degrees. The whole
vector

is therefore rotated about a


through that angle.
Let a be given in terms of i, j, k as

aa=

=a
2

ii

a2

aa

ij

k,

ik

ROTATIONS AND STRAINS

+ 2a
+ az a

J*

ki

JJ+

a 3 a 2 kj

= ii+jj +

a = 0ii

3 ij

339

a 3 J*
2

kk,

kk,

ik,

3J +
JJ-iJk,
- 2 ki + ^kj + Okk.

Hence

= {a^ (1

cos j)

cos

> ii

+ {a 2 (1 cos q)
sin g-} i j
3
+ {a 1 a 3 (1 cos #) + 2 sin * k
+ {2 a (1 cos 2) + a 3 sing-} ji
+ (a22 (1 cos q) + cos q} j j
1

3.1

+ {#2 a s (1 ~ cos #) ~~ a si n ?} j k
+ {agffj (1 cos ^) a 2 sin q} ki
+ (a s a 2 (1 cos ^) + a sin q} kj
+ (a 32 (1 cos q) + cos q} k k.
i

-f-

(7)

be written as in equation (4) the vector of


and the scalar of
may be found.
If

127.]

+ cos q Q x j +

=
X
Qs = i

The

cos q

axis of rotation

k x k)

-f

2 sin q

sin q (k

i is

x k)

+ kk) + sin
a = 1 + 2 cos q.

(j

q (k

k),

seen to have the direction of

<P X ,

the negative of the vector of 0. This is true in general.


The direction of the axis of rotation of any versor is the

negative of the vector of 0. The proof of this statement


depends on the invariant property of <# x
Any versor
may be reduced to the form (4) by taking the direction of i
.

VECTOR ANALYSIS

340

coincident with the direction of the axis of rotation.


this reduction has

been made the direction of the axis

to be the negative of

But

<P X .

<P y

is

After
is

seen

not altered by the


nor is the axis of

any particular form


Hence the direction of
rotation altered by such a reduction.
<P X the directhe axis of rotation is always coincident with
to

reduction of

tion of the negative of the vector of

The tangent

<#.

of one-half the angle of version q

sm

+ cos q

is

+ 0s

..

The tangent of one-half the angle of version is therefore


determined when the values of X and @ s are known. The
vector

and the scalar

which are invariants

4> s,

mine completely the versor

Let

0.

ft

Q,

be a vector drawn

Let the magnitude

in the direction of the axis of rotation.


of

of $, deter-

be equal to the tangent of one-half the angle q of

version.

The vector

determines the versor

ft

completely.

ft

will

be

called the vector semi-tangent of version.

was expressed in terms of a unit vector


(6) a versor
to
the
axis
of
rotation.
parallel

By

= aa +
Hence

if ft

a a)

sin q I

a.

be the vector semi-tangent of version


cos

There

cos q (I

'

sin

a more compact expression for a versor


of the vector semi-tangent of version.
Let c be
is

in terms

any vector in

space.

The version represented by


c

ft

c into c

ft

carries
ft

c.

ROTATIONS AND STRAINS


show

It will be sufficient to
ft.

For

this in case c is perpendicular to

component of it) were parallel to ft the


multiplying by ft x would be zero and the statement

if c

result of

would be

(or any

that c

carried into

is

In the

c.

nitudes of the two vectors are equal.


(c

ft

Since

ft

and

c are

ft

first

place the

ftxc'ftxc

(1

c vanishes.

+
c

(1+

tan 2 i q)

Hence the

ft

(1

c)

ft

ftC.

tan
2

\ q}

tan 2

(1

q)

(1
(c

q)
2

is

In the

equal to

q.

ftxcftxc
+

(1

= COS

tan 2
\

q)

tan 2 i q)
40

+ ft x c)

tan 2 1

equality.

ftxc)_cc

'-

(c

by hypothesis perpendicular

ftxc).(c +
2

ftC

second place the angle between the two vectors


(c

mag-

For

= CC + ft.ft CC

ftxc.ftxc =

cc +
The term

cc +

ftxc)

(c

c)

+ ftXCftXC

C'C

341

tan 1 q

2 c x

_
c

(1

= sin

(ft

tan

c)

I q)

q.

Hence the cosine and sine of the angle between c ft x c


and c + ft x c are equal respectively to the cosine and sine of
the angle q and consequently the angle between the vectors
:

must equal the angle

q.

Now

VECTOR ANALYSIS

342
c

and
(i

(c

x a),

ft

= (I I x ft)
c) = (I + I x ft)

+Qx

-1

(i

x a)- 1

(i

-ixft)=i+ixft.

(c

-a

(I

-I

Multiply by c
(i

+ ixft)(i-ix

a)-

c)

=c+

ft

c.

Hence the dyadic


<P

(I

carries the vector c

what the value

of

ft
c.

ft)

(10)'

ft)-

x c into the vector c + ft x c no matter


Hence the dyadic <P determines the

version due to the vector semi-tangent of version


The dyadic I + 1 x ft carries the vector c
(I

(I

ft.

ft

into

+ ft-ft)c.

+ Ixft)(c
(I

=c+

ftxc)

x ft).(c-ft +

ftxc

c)=c +

ftxc

ft-ftc

ftx(ftxc)

(l

ftft)c.

Hence the dyadic

carries the vector c

dicular to

ft

ft

c into

the vector

as has been supposed.

(I

produces a rotation of

c, if c

Consequently the dyadic

+ Ixft) 2
1 + ft-ft

all vectors in the

plane perpendicular

be applied to a vector x
however,
the result is not equal to # ft.
to

ft.

(I

it

If,

ft)

(I

1+ft-ft

ft)

be perpen-

xQ,

= x -(I +

ft

ft)

1+ft-ft

parallel to

-ft

ft

xQ,

1+ft.ft

ROTATIONS AND STRAINS


To

obviate this difficulty the dyad

for all vectors perpendicular to

The

merator.

versor

may

aa +

^=

1
(i

(i

Hence

x
x

ft)

ft)

(i

(i

substituting

x a)

ft)

may

Q,

may

ft,

(i

is

an annihilator

be added to the nu-

ixtt)*

ft-ft

2 1

(i

which

then be written

=i+
ft)

x
ft

ft

=i

(i
ft ft

ft). (i

- ft

ft

x Q)

i.

This

Q,

343

ft.ft

be expanded in nonion form.

Let

-(11)

=
128. ]

form

If a is a unit vector a dyadic of the

= 2aa-I

(12)

a Mquadrantal versor. That is, the dyadic


turns the
a
of
about
the
axis
two
points
space
right angles.
through
is

This may be seen by setting q equal to


expression for a versor

=
or

it

may be

The dyadic

aa

cos q (I

a a)

sin

TT

in the general

a,

seen directly from geometrical considerations.


leaves a vector parallel to a unchanged but re-

verses every vector perpendicular to a in direction.


Theorem : The product of two biquadrantal versors

versor the axis of which

is

is

perpendicular to the axes of the

VECTOR ANALYSIS

344

biquadrantal versors and

the angle of which is twice the


angle from the axis of the second to the axis of the first.
Let a and b be the axes of two biquadrantal versors. The

product

=(2bb-I).(2aa-I)
is

is

certainly a versor;

for the product of

Consider the

a versor.

common

any two versors

perpendicular to a and

b.

The biquadrantal versor 2 a a I reverses this perpendicular


in direction. (2bb I) again reverses it in direction and consequently brings it back to its original position. Hence the
product

changed.

The
a

leaves the
is

common

perpendicular to a and b un-

therefore a rotation about this line as axis.

2 (b

cosine of the angle from a to

2 b

Hence the angle


from a to

a b

of the versor

is

is

a)

- 1 = cos

(b, a).

equal to twice the angle

b.

Theorem : Conversely any given versor may be expressed


as the product of two biquadrantal versors, of which the axes
lie in

the plane perpendicular to the axis of the given versor

and include between them an angle equal to one half the


angle of the given versor.
For let Q be the given versor.
perpendicular to the axis

J? x

Let a and b be unit vectors

of this versor.

Furthermore

angle from a to b be equal to one half the angle of


this versor.
Then by the foregoing theorem
let the

Q = (2bb The

I)

(2

aa

- I).

(14)

two biquadand simple method for


rotations about a fixed point
Let

resolution of versors into the product of

rantal versors affords an immediate

compounding two finite


and W be two given versors.

Let b be a unit vector per-

ROTATIONS AND STRAINS

345

and
Let a be a unit vector
pendicular to the axes of
and
axis
of
such
that the angle from
to
the
perpendicular
half
the
of
to
one
Let c be a unit
a to b is equal
0.
angle
.

vector perpendicular to the axis of W and such that the angle


from b to c is equal to one half the angle of W. Then
<P

?F.

But

(2

bb

I)

the fact that

it

(2

= (2bb-I).(2aa-I)
- I)

c c

(2 b b

I)

(2 a a

- I).

equal to the idemfactor, as may be seen from


represents a rotation through four right angles

is

or from the expansion


(2 b b

- I) = 4 b b bb-4bb + I = I.
= (2 cc - I) .(2 aa - I).

- I)

(2 b b

Hence

The product

of

perpendicular to a

into

and

is

one half the angle from a to


If

and

tangents of

a versor the axis of which

and the angle

is

is

equal to

c.

two versors of which the vector semiversion are respectively ft t and Q.J, the vector

Q3

of the product

q1 + a2

and

which

are

semi-tangent of version

Let

of

+Q xa
2

<P

15

i-ci^a,
- I)
d> = (2 bb
(2 aa
I)
V= (2 cc - I) (2 bb - I).
.

= (2cc-I)

(2aa-I).
*

ba

=4

2 aa
a

b b x

2 b b
a,

is

I,

VECTOR ANALYSIS

346

= 4(a.b) 2 -l,

F=4

=4c

?F X

?F

<P

-2bb -2cc +

b cb

=4

a ca

axb
HfillCfi

QJ

~^r -

(b

O A
y Hi
Q
Ho

Hence

Hence

c)

a)

I,

a,

1.

axe

bxc Q
(a

2 aa

fa b c] b

b)

D b

ft

But

~~

Q.

b,

2 cc

4 (c

0)^

=4c

$) x

(?F

I,

a -b b

= bxc ar + cxa br + axb cr,


[abc]b = bxc ab + cxabb + axbc.b.

[a be] r

ft 2

ft,

bxc

axb

ft 2

axe
a

Q,,

b b

c ft Q

~-

bb

a.e
a

b b>

_(axb)(bxc)_abb.c
a

Hence

b b

Q! x g 2

ft 2

b b

ft,

b b

ROTATIONS AND STRAINS

347

This formula gives the composition of two

finite rotations.

If the rotations be infinitesimal

ftj

and

Q, 2

are both infinitesi-

Neglecting infinitesimals of the second order the formula reduces to


mal.

Q, 3

The

= Q,j +

Q, 2 .

combine according to the law of


This demonstrates the parallelogram law for

infinitesimal rotations

vector addition.

The

angular velocities.

60.

Eight Tensors, Tonics, and Cyclotonics

Cyclics,

129.] If the

subject was treated from different

and

standpoints in Arts. 51

dyadic

be a versor

<P

it

may be

written in the

form (4)

i i

cos q

The

axis of rotation

axis

is q.

Let

(j j

sin q (k j

k).

and the angle of rotation about that


another versor with the same axis and

an angle of rotation equal to

Multiplying

is i

W be

W=ii +

k k)

cos q'

(j j

q'.

k k)

sin q' (k j

cos (2

?') (j j

k).

W= W

i i

+ kk)

kj-jk).

(16)

This

is the result which was to be


the product of
expected
two versors of which the axes are coincident is a versor with
the same axis and with an angle equal to the sum of the

angles of the two given versors.


If a versor be multiplied by itself, geometric and analytic
considerations alike make it evident that
2

and

= ii +

cos

2q

= ii +

cos

nq

(j j

(jj

+ kk) +

+ kk) +

sin 2 q (kj
sin

nq

(kj

-jk),
j

k).

VECTOR ANALYSIS

348

On

the other hand let

kj-jk.

4>*

The product of
i i.
Hence

(i i

= ii +

The dyadic
The dyadic
of $!

cos n

</>

<?

raised to

or

(cos q

$!*

cos q &!

ii into either

Q n = ii +

equal jj

+ kk; and

equal

Then

<P

<P

sin q

w cos"" 1

is

$ 2)n

zero and into itself

sin q

is

n
2)

sin ^ (Pj"- 1

*?

any power reproduces

raised to the second

itself.

+
0^

<P

r
power gives the negative

raised to the third power, the negative of <PZ ; raised


raised to the fifth power, <P2 and so
1 ;

to the fourth power,

on (Art. 114).
2

The dyadic

$2

multiplied by

is

equal to

Hence
n

= ii +

cos n q

- n(n~L)
0"

Equating

=i +
i

coefficients of

cos

nq
and

for <P n

cos

nq

= cos*

n cos n ~ 1 q sin

sm 2 $2 +

cos"- 2 q

2!

But

n (n
--

^-^-.

tf>

1)
-

AI

sin

71

in these

cos"" 2 q

<P
Z

<$

two expressions

sm 2

Thus the ordinary expansions for cos nq and sinnq are


obtained in a manner very similar to the manner in which
they are generally obtained.
The expression for a versor

may be generalized as follows.


be any three non-coplanar vectors ; and
a', V, c', the
Consider
the
reciprocal system.
dyadic
Let

a, b, c

cc')

sin q

(cb'-bc')- (IT)

ROTATIONS AND STRAINS

349

This dyadic leaves vectors parallel to a unchanged. Vectors


in the plane of b and c suffer a change similar to rotation.

Let
r

r'

= cos p b +

= cos

This transformation

(j?

sin

q) b

sin (p

may be given

interpretation as follows.

c,

q)

c.

a definite geometrical

The vector

r,

when p

is

regarded

as a variable scalar parameter, describes an ellipse of which


b and c are two conjugate semi-diameters (page 117). Let
this ellipse

be regarded as the parallel projection of the

unit circle
f

That

is,

the ellipse

= cos p +
i

and the

sin q

j.

circle are cut

The two semi-diameters

from the same

and j of the circle procylinder.


ject into the conjugate semi-diameters a and b of the ellipse.

The

radius vector r in the ellipse projects into the radius vector


The radius vector r' in the ellipse which

f in the unit circle.

equal to
such that
is

<#

r,

f'

projects into a radius vector

= cos (p + q) +
i

sin

(jp

r'

in the circle

+ q) j.

Thus

the vector r in the ellipse is so changed by the application of


as a prefactor that its projection f in the unit circle
is rotated through an
angle q.

This statement

be given a neater form by making use


of the fact that in parallel projection areas are changed in a

may

definite constant ratio.

The vector

f in the unit circle

may

be regarded as describing a sector of which the area is to the


area of the whole circle as q is to 2 TT. The radius vector f
then describes a sector of the ellipse. The area of this sector

whole ellipse as q is to 2 TT. Hence the


as
a
dyadic
applied
prefactor to a radius vector r in an ellipse
of which b and c are two conjugate semi-diameters advances
is

to the area of the

that vector through

sector the

area of which

is to the

area of

VECTOR ANALYSIS

350

the whole ellipse as q is to

radius vector r
sector q from
it is

of the

be called an

elliptic rotation through a


an ordinary rotation of which

similarity to

its

Such a displacement

the projection.

Definition

= a a' +
is

may

2-Tr.

of the form

dyadic
cos q (b

4- c c')

The

called a cyclic dyadic.

sin q (c

versor

is

c')

(17)

a special case of a

cyclic dyadic.
It is evident from geometric or analytic considerations that
the powers of a cyclic dyadic are formed, as the powers of a
versor were formed, by multiplying the scalar q by the power

to

which the dyadic


n

a'

cos

If the scalar q is

to be raised.

is

nq

(b

V+

c c')

sin

(c b'

nq

an integral sub-multiple of 2
2?r

TT,

that

c').

is, if

= m,

9.

possible to raise the dyadic

it is

namely, the

may then

power

and even

may

if

to such a

to raise

it

to

such an integral power,

becomes the idemfactor

be regarded as the rath root of the idemfactor.

In like manner
factor

m, that

if

and 2 TT are commensurable

power
and 2 TT

possible

it

be found which differs by as

the idemfactor.

it is

becomes equal to the idemare incommensurable a power of

that

little as

one pleases from

Hence any cyclic dyadic may be regarded as

a root of the idemfactor.


1

in

It is evident that
fixing

an

the result of the application of

ellipse practically fixes


vector in that plane may be

the ellipse.

4>

to all radii vectors

For any
for all vectors in the plane of b and c.
regarded as a scalar multiple of a radius vector of
it

ROTATIONS AND STRAINS

where

The

Definition:

130.]

d? adic

transformation represented by the

= aii + Jjj+ kk
C

a,

are positive scalars

&, c

351

is

(18)

called a pure strain.

The

a right tensor.
dyadic
right tensor may be factored into three factors
itself is called

0= (aii+jj
The order

in

kk>(-(ii

&j

+ kk) +(ii+jj + ckk).

which these factors occur

transformation

(ii

and

p'
is

such that the

+ jj + ckk)

immaterial.

The

components of a vector remain un-

altered but the k-component

The transformation may

is

is

altered in the ratio of c to 1.

therefore be described as a stretch or

elongation along the direction k. If the constant c is greater


than unity the elongation is a true elongation : but if c is less
than unity the elongation is really a compression, for the ratio
of elongation is less than unity.

Between these two cases


is unity.
The lengths

comes the case in which the constant

of the k-components are then not altered.

The transformation due

to the dyadic
may be regarded
or simultaneous elongation of the components of r parallel to i, j, and k respectively in the ratios
a to 1, b to 1, c to 1. If one or more of the constants a, &, c

as

the successive

than unity the elongation in that or those directions


becomes a compression. If one or more of the constants is

is less

unity,

The

components parallel to that direction are not altered.

directions

k are

called the principal axes of the strain.


Their directions are not altered by the strain whereas, if the

constants a,

The

i,

&, c

scalars a,

j,

be different, every other direction is altered.


&, c are known as the principal ratios of

elongation.

In Art. 115

it

was seen that any complete dyadic was

reducible to the normal form

0=

(ai'i

Jj'j

ck'k)'

VECTOR ANALYSIS

352

where

a,

Z>,

are positive constants.

This expression

may

be

factored into the product of two dyadics.

i/i/

(a

0=

or

The

& j'j'

(i'i

+j'j

+ k'k)

(i'i+j'j
(a ii

i'i+j'j

factor

k'k')

6jj

k'k),

+ ckk).

k'k

the same in either method of factoring


It turns the vectors i, j, k into the vectors i ', j

which

versor

is

be represented by

may

(19)

its

is
',

a versor.
k'.

This

vector semi-tangent of

version as
i/j

The other

v/

TT

+ J x ^ + k x k/
"l + i.i'+j.j' + k.k''

_i

*'

factor

i'i'

or

i i

6 j'j'

ck'k',

a right tensor and represents a pure strain. In the first


case the strain has the lines i', j', k' for principal axes: in

is

the second,

i, j,

k.

In both cases the ratios of elongation are

a to 1, & to 1, c to 1. If the negative sign occurs


the product the version and pure strain must have
associated with them a reversal of directions of all vectors in
the same,

before

a perversion. Hence
Any dyadic is reducible to the product of a
versor and a right tensor taken in either order and a positive
space

that

Theorem

is,

or negative sign. Hence the most general transformation


representable by a dyadic consists of the product of a rotation or version about a definite axis through a definite angle

accompanied by a pure strain either with or without perverThe rotation and strain may be performed in either
order.
In the two cases the rotation and the ratios of elongasion.

tion of the strain are the

same

strain differ according as it is

but the principal axes of the

performed before or after the

ROTATIONS AND STRAINS

353

system of axes being derivable from the other

rotation, either

by the application of the versor as a prefactor or postfactor


respectively.

be given the product of


and its conjugate
a right tensor the ratios of elongation of which are the
and the axes of which
squares of the ratios of elongation of
If a dyadic

is

are respectively the antecedents or consequents of


in the product.
C follows or precedes
ing as

d>=
C

0.

tf

(ai'i
(aii'

a ai'i'

6 j'j

ck'k),

J jj'

ckk'),

+
+

&

= a*ii +

The general problem

j'j'
2
Z>

jj

+
+

k'k',

accord-

(20)

kk.

of finding the principal ratios of elonga-

the antecedents, and consequents of a dyadic in its


normal form, therefore reduces to the simpler problem of findtion,

ing the principal ratios of elongation and the principal axes


of a pure strain.

The

131.]

natural and immediate generalization of the

right tensor

ckk,
is

=a

the dyadic

aa'

4- &

b'

c c c'

(21)

where a, 6, c are positive or negative scalars and where a, b, c


and a', b', c' are two reciprocal systems of vectors. Neces-

and

sarily a, b, c

Definition

a', b', c'

are each three non-coplanar.

dyadic that

may

be reduced to the form


ccc'

is

(21)

called a tonic.

The

effect of a tonic is to leave

coplanar directions
into

its

a, b, c in space.

components parallel to
23

unchanged three nonIf a vector be resolved

a, b, c

respectively these

VECTOR ANALYSIS

354

components are stretched in the ratios a to 1, & to 1, c to 1.


If one or more of the constants a, 6, c are negative the components parallel to the corresponding vector a, b, c are reversed in direction as well as changed in magnitude. The

may be

tonic

stretches the

factored into three

components parallel

factors

which each

of

one of the vectors

to

a, b, c

but leaves unchanged the components parallel to the other


two.

The value

of a tonic

is

not altered

if

in place of a, b, c

any three vectors respectively collinear with them be substituted,

provided of course that the corresponding changes

which are necessary be made in the reciprocal system

But with the exception

of this change, a dyadic

expressible in the form of a tonic

one

way

if

the constants a,

Z>,

a', b', c'.

which

is

so expressible in only
If two of the
are different.

is

constants say 5 and c are equal, any two vectors coplanar


with the corresponding vectors b and c may be substituted

and c. If all the constants are equal the tonic


reduces to a constant multiple of the idemfactor. Any three
non-coplanar vectors may be taken for a, b, c.
in place of b

The product
same

is

of

two

which the axes

tonics of

commutative and

is

a, b, c

are the

a tonic with these axes and

with scalar coefficients equal respectively to the products of


the corresponding coefficients of the two dyadics.

V=

&2

b'

c 2 c c'

ejCaCc'.

(22)

generalization of the cyclic dyadic

a
is

= V.0 = ai a z a&' + J^bb' +

0.
The

a z a a'

a'

cos q (b

V+

c c')

= aaa' + KbV +

cc')

sin q (c b'

- b c')

(cb'-W),

(23)

ROTATIONS AND STRAINS


where

355

which a

are three non-coplanar vectors of

a, b, c

b ',

',

'

the reciprocal system and where the quantities a, J, c, are


This dyadic may be changed
positive or negative scalars.
into a more convenient form by determining the positive
is

p and

the positive or negative scalar q (which


between the limits
be
chosen
TT) so that
always

scalar

and

=p

may

cos q

c=psinq.

That

(24)

p=+

is,

and

2=.

tan

(24)'

Then
(P

= a a a' + p cos
This

may

q (b

V+

The

sin q (c

b c').

(25)

be factored into the product of three dyadics

0= (aaa' + bb' + cc')


{a a'

+p

c c')

cos q (b b'

order of these factors

c c')

is

+p bb' +pcc')
+ sin q (c V b c')}.

(aa'

immaterial.

The

first is

a tonic

which leaves unchanged vectors

parallel to b and c but


stretches those parallel to a in the ratio of a to 1.
If a is

negative the stretching must be accompanied by reversal


in direction. The second factor is also a tonic. It leaves

unchanged vectors

parallel to a

but stretches

the plane of b and c in the ratio

to 1.

all

The

vectors in
third

is

Vectors parallel to a remain unchanged but


which b and c are conjugate
semi-diameters are rotated through a vector such that the
cyclic factor.

radii vectors in the ellipse of

area of the vector

TT.

is

to the area of the whole ellipse as q to

Other vectors in the plane of b and

may

be regarded

as scalar multiples of the radii vectors of the ellipse.

VECTOR ANALYSIS

356
Definition
<P

A dyadic which

= a a a' + p cos

V+

q (b

c c')

reducible to the form

+p

sin q (c

b c'),

(25)

combines the properties of the

to the fact that it

owing

is

and the tonic is called a cyclotonic.


of two cyclotonics which have the same three
b, c as antecedents and the reciprocal system

cyclic dyadic

The product
vectors,
a', b',

a,

c' for

consequents

is

a third cyclotonic and

is

com-

mutative.

= a a a' + p cos ^ (b V + c c') + p sin q


= a a a' + p cos q (b V + c + p2 sin q
W=
d> = a a aa' +
p^p cos (q + q
+ Pi PZ sin (ft + 2a (c b' - b o
l

c')

b'
z (c

z)

132.]

to

(bb'

c')

b c')

c c')

Reduction of Dyadics

(c b'

(26)

)-

Canonical Forms

may be reduced
The dyadics for which

Theorem : In general any dyadic

either to a tonic or to a cyclotonic.

the reduction

is

impossible

may

be regarded as limiting cases

which may be represented

to any desired degree of approximation by tonics or cyclotonics.


From this theorem the importance of the tonic and cyclo-

which have been treated as natural generalizations of


the right tensor and the cyclic dyadic may be seen.
The
of
all
the
the
discussion
of
a
theorem, including
proof

tonic

special cases that

The method
patent.
left

may

arise, is

long and somewhat tedious.

of proving the theorem in general

If three directions a, b, c

may

unchanged by the application of

however

is

be found which are


then

must be a

If only one such direction can be found, there exists


a plane in which the vectors suffer a
change such as that due
to the cyclotonic and the
dyadic indeed proves to be such.
tonic.

ROTATIONS AND STRAINS


The question

is

which are unchanged

to find the directions

of the dyadic 0.

by the application

If the direction a

is

unchanged, then
d>

or

(0

The dyadic

357

aI

is

=aa
al).a = 0.
a

(27)

therefore planar since

it

reduces vectors

In special cases, which are set


may be linear or zero. In

in the direction a to zero.

aside for the present, the dyadic

any case

if

the dyadic

0-al

reduces vectors collinear with a to zero

it

possesses at least

one degree of nullity and the third or determinant of

<P

vanishes.

(0-aI) 8 =
Now

(page 331)

Hence

(4>

(3>

- al) 3 =
I,

)3

<P

<P
3

- a &2

0.

(28)

<P
2

a2

W+
:

12

d>

a3 1 8

= I and I3 = 1.

But

ffi

Hence the equation becomes


a3

The value

of a

- a2

ffi

which

satisfies

a
is

zS -4>z

4>

= 0.

(29)

the condition that

= aa

a solution of a cubic equation.

Let x replace

a.

The

cubic equation becomes


(29)'

VECTOR ANALYSIS

358

Any

value of x which satisfies this equation will be such

that

That

(0-sI) 8 = 0.
is

to say, the dyadic

(28)'

I is planar.

vector per-

pendicular to its consequents is reduced to zero. Hence


The further discussion
leaves such a direction unchanged.
of the reduction of a dyadic to the form of a tonic or a cycle-

upon whether the cubic equation

tonic depends merely

in x

has one or three real roots.


133. J

Theorem
xs

If the cubic equation

- x*

&Z3 -0 S

has three real roots the dyadic


to a tonic.

For

let

= a,

The dyadics

be the three roots of the equation.

I,

<P

=c

&,

(29)'

in general be reduced

may

=0

bl,

cl

are in general planar.


Let a, b, c be respectively three
vectors drawn perpendicular to the planes of the
consequents
of these dyadics.

= 0,
(0-cI).c = 0.
$ a = a a,
0.b = Jb,
C = C C.
b

Then

If the roots a,

coplanar.

(30)

Z>,

are distinct the vectors

For suppose
c

= ma +

Ttb

(30)'

a,

b, c

are non-

ROTATIONS AND STRAINS

mca + n
0.a = a,

(P

a,

But

w (a
m (a
m=

Hence
and

Hence
Consequently

if

c)

= 0,
or a = c,

the vectors

a, b, c

?i

})

= 0.

<P.b=&b.
(&

c)

w (b

c)

= 0,
c) = 0.

or b

c.

are coplanar, the roots are


the roots are distinct, the

a, b, c

not distinct; and therefore


vectors

ft

<P

359

if

In case the roots

are necessarily non-coplanar.

it is still always possible to choose three


non-coplanar vectors a, b, c in such a manner that the equaThis being so, there exists a system a', b', c'
tions (30) hold.

are not distinct

reciprocal to

a a,

a, b, c

and the dyadic which

carries a, b, c into

b b, c c is the tonic

ccc.

Theorem :

If the cubic

xs

- x2

equation
tf>

has one real root the dyadic


a cyclotonic.

The cubic equation has one


tive or negative according as

<P

2S

may

<P

(29)'

in general be reduced to

This must be posiLet


positive or negative.

real root.
<P

is
8

the root be a.

Determine a perpendicular
the consequents of
a I.

to the plane of

(0-aI).a = 0,
Determine

and

a' also

so that

l.
lengths of a and a' be so adjusted that a' a
This cannot be accomplished in the special case in which a

let the

VECTOR ANALYSIS

360

and

a' are

Let b be any vector in

mutually perpendicular.

the plane perpendicular to a'.


a'

Hence (0

al)b

perpendicular to

0" 1

2
b, CP"

= 0.

perpendicular to

is

Hence

a'.

$2

In a similar manner

a'.

b, etc., will all

The

one plane.

- a I)

(<P

vectors

<P

0b

<0 3

b,

b,

is

and

be perpendicular to a' and lie in


b and b cannot be parallel or

would have the direction b as well as a unchanged and


thus the cubic would have more than one real root.

The dyadic
changes a,
The
volume
of
spectively.
[<P

The

a a

vectors

<0

b,

tf>

b,

Inasmuch

as a

x (0-b)

and

<P

=p

~l

and

b3

b_ 1

=^ ^2

or

The

Then

b3

b1

+ b_

b2

b2

= 2?ib 2
= 2wb

b1

b.

(31)'

(31)"

b,

b,

etc.

etc.

Let

2 n br

(33)

= 0.

b and b x are parallel.

b.

let

$*

=^ 0~

b) x bj

bj^bj x

b)

(31)

(32)

=p~*

b2 x

b2

b].

b^
+

0bxb.

-b

(b 2
vectors b 2

4>

re-

same plane. Their


and to each other. Hence

= a-i08

*
Let also

(<P

have the same sign,

in the

all lie

vector products are parallel to a'

b)

b, <P

the parallelepiped
b]

b)

(<P

Z
a, <P

4>

= <P8 [a
a = a a.
b) = 3 a

But
Hence

b into

b,

b2

(34)

27ib 3

+ b_r=2wb_

etc.,

(35)
etc.

ROTATIONS AND STRAINS


Lay

off

from a common origin the vectors


b,

Since

361

is

bp b2

etc.,

not a tonic, that

b_2

b_j,

is,

etc.

since there

no

is

direction in

the plane perpendicular to a' which is left unchanged by


these vectors b m pass round and round the origin as ra takes
all positive and negative values.
The value of
Let
therefore lie between plus one and minus one.

on

= cos
+ b = 2 cos
n

Then

b_j

Determine

(36)

q.

n must

q b.

from the equation

= cos q b + sin q c.
sin q c.
b_j = cos q b
bj

Then
Let

be the reciprocal system of


sible since a' was so determined that
a', b', c'

W=
W

Then

Hence

cos q (bb'

(aaa'
(a aa'

The dyadic a
the vectors

a.a!

The dyadic
is

cc')

sin q (c

+p
+ .p

+p

+p

a,

=a
=p

b^ = b.

=
b =
b_j =p b = <P
&

"b

a,

b,

b_ r

changes the vectors


b,

pos-

and since

V - be').

3P

SF.b^bj,

= (a a a' + p

root

0,

(a a,*'

=1

a'

is

Let

are non-coplanar.

a, b, c

This

a, b, c.

and

W)

=a

a,

b and b_j into

b_g respectively.
a

a'

+ p cos
+ p sin q

Hence

V + c c')
b c').
(c V
(b

in case the cubic equation has


only one real
in
cases
to
a cyclotonic.
except
special

reducible

The theorem

that a dyadic in general is reducible to a tonic


or cyclotonic has therefore been demonstrated.

VECTOR ANALYSIS

362

There remain two cases 1 in which the reduction

134.]

In
impossible, as can be seen by looking over the proof.
if
n
in
the
constant
the
used
reduction to cyclothe first place

is

tonic form be

place

if

In the second

1 the reduction falls through.

the plane of the antecedents of

0-al
and the plane of the consequents are perpendicular the
and a' used in the reduction to cyclotonic form are

vectors a

perpendicular and it
a a' shall be unity.

If

impossible to determine a' such that


The reduction falls through.

is

1,

Let

b_i

Choose

Hence

bj

c^bj

=a

Consider the dyadic

a a'

+p

The transformation due

bx

b.

= 2 b.

=b

=p G =p'b

a a a'

b_j

b_r

+p

J9

(b b'

(b

b'

c c')

c c')

to this dyadic

+p

-f-

c b'

C.

cb r

may be

(37)

seen best by

factoring it into three factors which are independent of the


order or arrangement

(a a a'

b'

c c')

{a a/

(a

a'

+ p (b b' +
+

bb'

c c')}

cc'

cb').

In these cases it will be seen that the cubic


equation has three real roots.
In one case two of them are equal and in the other case three of them. Thus
these dyadics may be regarded as
limiting cases lying between the cyclotonic in
which two of the roots are
imaginary and the tonic in which all the roots are real
and distinct. The limit may be
regarded as taking place either by the pure

imaginary part of the two imaginary roots of the cyclotonic becoming zero or by
two of the roots of the tonic
approaching each other.

ROTATIONS AND STRAINS


The

first

363

factor represents an elongation in the direction a in a


1.
The plane of b and c is undisturbed. The

ratio a to

second factor represents a stretching of the plane of b and


the ratio

to 1.

The

last factor takes the

(I

(I

+ c V)

c b')

(I

in

cb'.

x&

xb

+ c V)

form

= #a,

= x b + x c,
x c = x c.

A dyadic of the form I + cb' leaves vectors parallel to a and c


unaltered. A vector x b parallel to b
increased by the vecis

by the ratio of the vector x b to b. In other


words the transformation of points in space is such that the

tor c multiplied

plane of a and

remains fixed point for point but the points

in planes parallel to that plane are shifted in the direction c

by an amount proportional to the distance of the plane in


which they lie from the plane of a and c.
Definition : A dyadic reducible to the form
I

cb'

called a shearing dyadic or shearer and the geometrical


transformation which it causes is called a shear. The more

is

general dyadic
cc')

+ oV

(37)

The transwill also be called a shearing dyadic or shearer.


formation to which it gives rise is a shear combined with
elongations in the direction of a and is in the plane of b and c.
If n
1 instead of n
+1, the result is much the same.

The dyadic then becomes

cc')-cV

(a aa'

+ bV +

cc')

{aa'

-p

(bV +

cc')}

(37)'
(I

cV).

VECTOR ANALYSIS

364

The

factors are the

same except the second which now repre-

sents a stretching of the plane of b and c combined with a


reversal of all the vectors in that plane. The shearing dyadic
then represents an elongation in the direction a, an elonga-

combined with a reversal of direction in the plane of


c, and a shear.

tion

b and

Suppose that the plane of the antecedents and the plane of


the consequents of the dyadic <# al are perpendicular.
Let
these planes be taken respectively as the plane of j and k
the plane of i and j k. The dyadic then takes the form

The

is

coefficient

B must vanish.

and

For otherwise the dyadic

planar and the scalar a + B is a root of the cubic equation.


this root the reduction to the form of a tonic may be

With

carried on as before.

new

This

case occurs.

Nothing new
Let

may be reduced

as follows to the

ab'

where
Square

V=a

arises.

But

be'

= b e' =
and b
W* = A D ki = ac'.
e'

= A D k,

B vanishes

form

Hence a must be chosen parallel to k and


The dyadic
may then be transformed into

Then

if

b'

= Ci+Di
AD
O =

b'

c',

= 1.

parallel to

i.

With

ROTATIONS AND STRAINS

365

reduces to the

this choice of a, b, V,

c'

the dyadic

desired form ab'+ be' and hence the dyadic

= al +

= aaa' + abb'+

or

ab'

ace'

reduced to

is

bc'

(38)

+ aV +

be'.

factored into the product of two dyadics the


order of which is immaterial.

This

The

may be

factor al represents a stretching of space in all


The second factor

first

directions in the ratio a to 1.

represents

r'=
If r

is

what may be

IT +

ab'.r

parallel to a

it

is

r is parallel to b it is

which

is

if

bc'^r

+ ab'.r +

r.

unaltered by the dyadic Q. If


changed by the addition of a term

magnitude

and

in

of the vector

magnitude proporIn like manner

r.

changed by the addition of a term


equal to b and which in magnitude is

r is parallel to c it is

which in direction

is

proportional to the magnitude of the vector

= (I + ab' -f bc') #b =
a;c = (I + ab' + bc')#c =

.zb

Definition

dyadic which

= al +
is

bc'

left

in direction equal to a

tional to the

For

called a complex shear.

may
ab'

r.

#b + x&
xc

#b.

be reduced to the form

+ bc'

(38)

called a complex shearer.

The complex

shearer as well as the simple shearer mentioned before are limiting cases of the cyclotonic and tonic
dyadics.

VECTOR ANALYSIS

366

more systematic treatment of the various kinds


of dyadics which may arise may be given by means of the
Hamilton-Cayley equation
135.]

03

0z

zs

0_

(39)

= 0.

(29)'

and the cubic equation in x


x3

<P

z
a x

23 x

<P

are the roots of this cubic the Hamilton-Cayley


equation may be written as
If a,

&, c

(0-aI).(0-&I).(0-cI) =0.
If,

(40)

however, the cubic has only one root the Hamilton-Cayley

equation takes the form

a I).

(0 -2^?

cos

<P+p*I) =0.

(41)

In general the Hamilton-Cayley equation which is an equation of the third degree in


is the equation of lowest
degree

which

by (P. In general therefore one of the above


and
the
equations
corresponding reductions to the tonic or
cyclotonic form hold. In special cases, however, the dyadic
is satisfied

may

satisfy

an equation of lower degree. That equation


which may be satisfied by a dyadic is called

of lowest degree
its characteristic

II.

III.

IV.

V.
VI.
VII.

equation.

The following

possibilities occur.

(<P-aI

(0- a I).(<P-& I) 2 = 0.

(0- a !)(</>- 61)


3
(<P-aI) = 0.
(0-aI) 2 = 0.
0-aI = 0.

=0.

ROTATIONS AND STRAINS


In the
to the

first

case the dyadic

form

a tonic and

is

= aaa' +

Jbb'

In the second case the dyadic


reduced to the form

= aaa' + p

cos q (bb'

In the third case the dyadic


reduced to the form
<P

= aaa' +

is

cc')

367

may

be reduced

ccc'.

a cyclotonic and

may

+p

be').

sin q

(cV

a simple shearer and

is

(bb'

cc')

may

be

be

cb'.

In the fourth case the dyadic is again a tonic. Two of the


The following reduction
ratios of elongation are the same.

may

be accomplished in an infinite number of ways.


<P

In the

fifth

= aaa' +

case the dyadic

so expressed that

0=

al

&

(bb'

cc').

is

a complex shearer

ab'

In the sixth case the dyadic


may be reduced to the form

is

= al + cb'=a(aa'

and may be

bc'.

again a simple shearer which

-4-bb'

+cc') + cV.

In the seventh case the dyadic is again a tonic which may be


reduced in a doubly infinite number of ways to the form
<p

al

a (a a'

bb'

cc').

These seven are the only essentially different forms which a


dyadic may take. There are then only seven really different
three tonics in which the ratios of elongation are all different, two alike, or all equal, and the cyclo-

kinds of dyadics

tonic together with three limiting cases, the


the one complex shearer.

two simple and

VECTOR ANALYSIS

368

Summary

of Chapter

VI

The transformation due to a dyadic is a linear homogeneou


The dyadic itself gives the transformation of the
strain.
The second of the dyadic gives the transin
space.
points
of
formation
plane areas. The third of the dyadic gives the
ratio in

which volumes are changed.

s'=02

r'=0r,
The necessary and

v'=0 a

s,

sufficient condition that a dyadic repre-

sent a rotation about a definite axis


the form
4>

or that

d>

that

is

be reducible to

it

= i'i + j'j + k'k


=I
=+1
C
<P C = I
<P
3 >
<P

or that

The necessary and

v.

<P

(1)

(2)

sufficient condition that a dyadic repre-

sent a rotation combined with a transformation of reflection

by which each

figure is replaced

by one symmetrical

to

it is

that

= -(i'i +
<p.<p c = I,
<P
<P C = I,

or that
or that

dyadic of the form (1)

is

k'k)

(1)'

= 1
3
<P < 0.
3

(3)

j' j

d>

called a versor

one of the form

(1)', a perversor.

If the axis of rotation of a versor

be chosen or the i-axis

the versor reduces to


== ii

or
If

cos q (j
i i

any unit vector a

+ kk) +

cos q (I
is

= aa +
The

i i)

- j k)

sin q (kj

sin q I

(4)

i.

(5)

directed along the axis of rotation

cos q (I

a a)

sin q I

x a

axis of the versor coincides in direction with

(6)
<P V .

ROTATIONS AND STRAINS


If a vector be
'ie

drawn along the

and

axis

if

369

the magnitude of

vector be taken equal to the tangent of one-half the angle


rotation, the vector determines the rotation completely.
vector

f his

is

called the vector semi-tangent of version.

= tan 2 ^2

ft. ft

In terms of

ft

the versor

(9)

be expressed in a number of

may

ways.

&=-ftft

ft-ft

ftft \
-

r-

cos a (I

or

sin q

<J-<v

(I

x a)

(I

-I

=
Va-a

ft)-

(10)

(10)'

(ioy

ftft

-.
1

ft-ft

If a is a unit vector a dyadic of the

=2

aa

form

(11)

a biquadrantal versor.
Any versor may be resolved into
the product of two biquadrantal versors and by means of
is

such

resolutions

another.
of version

The law

any two versors may be combined into


of composition for the vector semi-tangents

is
ft

ft

ft

ft

dyadic reducible to the form

= aa' + cos
is

(bV +

called a cyclic dyadic.

simple rotation

an

cc')
It

sin q (cb'

be')

(17)

produces a generalization of

elliptic rotation, so to speak.


24

The

pro-

VECTOR ANALYSIS

370

duct of two cyclic dyadics which have the same antecedents


a, b, c and consequents a' b' c' is obtained by adding their

angles q.
idemfactor.

cyclic dyadic

may

= aii +
where

be regarded as a root of the

dyadic reducible to the form


&jj

+ ckk

(18)

a right tensor. It
a
the
principal axis i, j, k in the
stretching along
represents
1
are
a
to
c
to
which
called the principal ratios
to 1, b
ratio
1,
This transformation is a pure strain.
of elongation.
a, &, c are positive scalars is called

Any

dyadic

a right tensor,

(ai'i'

0=

or

may

be expressed as the product of a versor,

and a positive

(i'i

Jj'j'

or.

negative sign.
k'k') (i'i

+ j'j + k'k).(aii+Jjj +

j'j

k'k)

ckk).

(19)

Consequently any linear homogeneous strain may be regarded


combination of a rotation and a pure strain accompanied

as a

or unaccompanied by a perversion.

The immediate

generalizations of the right tensor

and the

cyclic dyadic is to the tonic

(21)

and cyclotonic
cc')

or

<P

= aaa' + .p cos

where

Any
(21),

=+

# (bb'

*/ 52

C2

+ c(cb'-bc)

(23)

+ cc')+^sin # (eb'-bc')
and tan I

=-

p +

(25)

(24)'

dyadic in general may be reduced either to the form


is therefore a tonic, or to the form
(25), and is

and

therefore a cyclotonic.
The condition that a dyadic be a
tonic is that the cubic equation

(29)'

ROTATIONS AND STRAINS


have three

shall

real

371

Special cases in which the

roots.

may be accomplished in more ways than one arise


the equation has equal roots. The condition that a

reduction

when

dyadic be a cyclotonic is that this cubic equation shall have


only one real root. There occur two limiting cases in which
the dyadic cannot be reduced to cyclotonic form.
cases it may be written as

=aaa' +p
and

is

a simple shearer, or

it

(bb'

is

a complex shearer.

cc')

cb'

(37)

takes the form

= al +
and

In these

ab'

Dyadics

bc'

may

(38)

be classified accord-

ing to their characteristic equations

=0
tonic
a I) (<P 2
2 p cos q <P + p 2 1) =
cyclotonic
2 =
a !)($
& I)
simple shearer
(<P
(0 al). (<P b I) =
special tonic
3
(0 a I) =
complex shearer
(0 a I) =
special simple shearer
a I) =
special tonic.
(<P

(<P

(0

aI).(<P

61).

(<Z>

cl)

CHAPTER

YII

MISCELLANEOUS APPLICATIONS
Quadric Surfaces
If

136.]

be any constant dyadic the equation


r

0.i =

const.

(1)

The constant, in case it be not zero, may


quadratic in r.
and hence the equation takes
be divided into the dyadic
is

the form
r

or

The dyadic

= 1,
r = 0.

is

may

(2)

an anti-self-conjugate dyadic, the product


is left

dyadic

is

as an exercise.

The

r.

By

r is

Art. 116 any self-conjugate

reducible to the form

-1-

If

r .,.

Hence the equation

a2

<>

bz

"

(3)
\.
/

2)

= ici + yj + zk,

r=
r

#2
_

7/2

|i

zz

(4)

=1

represents a quadric surface real or imaginary.


The different cases which arise are four in number.
signs are all positive, the quadric
is

if

proof of this state-

*=-y"

sign

For

be assumed to be self-conjugate.

identically zero for all values of

ment

(1)'

negative

it is

is

a real ellipsoid.

an hyperboloid of one sheet;

if

If the
If

one

two are

QUADRIC SURFACES

373

If the three signs are


negative, a hyperboloid of two sheets.
In like manner the
all negative the quadric is imaginary.

equation

r.0.r =

seen to represent a cone which may be either real or


imaginary according as the signs are different or all alike.
is

Thus

the equation
r

= const.
The

represents a central quadric surface.


a cone in case the constant is zero.

quadric surface

may

Conversely any central


be represented by a suitably chosen self-

form

in the

conjugate dyadic

This

is

surfaces

surface reduces to

= const.

evident from the equations of the central quadric


when reduced to the normal form. They axe

x2

a2

zz

The corresponding dyadic

= const.
i i

is

j j

a2

kk
%c2

The most general

scalar expression which is quadratic in


the vector r and which consequently when set equal to a constant represents a quadric surface, contains terms like
r

where

a, b, c, d, e

a) (b

(r

r,

c,

are constant vectors.

are of the second order in r

the

)r,

The

the third, of the

c,

first
first

two terms
and

order

independent of r. Moreover, it is evident that these


four sorts of terms are the only ones which can occur in a
last,

scalar expression

which

But
and

r
(r

quadratic in

is

a) (b

=r I
=r
r)

r.

r,

ab

r.

VECTOR ANALYSIS

374

Hence the most general quadratic expression may be reduced


to

r.0.r + r.A +
where

a constant dyadic,

is

conjugate if desired.
To be rid of the linear term r

(r'

- t)

r'

Since
equal.

r'

(r'

- t) +

-t

now $

r'

is

+2

r'

(|

<P

r'

In case

is

A-

complete the vector

is

and third terms are

incomplete

is self -conjugate.

as the case

soluble for t

If

may

or

4>

t)

may

= l 0-

C'

= 0.

be chosen so that

1 -

A.

reducible to the central form


r'

sible.

- t) A + C =

Hence

Hence the quadric

is

(r'

self-conjugate the second

is

self-

a change of origin

+
-fr'.A-t.A+C'^O.
r'

^A=

of

make

A,

as

t.

r'

r'

If

a constant vector, and

The dyadic may be regarded

a constant scalar.

by replacing r by

(7:=0,

r'

it is

lies

= const.

wm'planar or wmlinear because


in the plane of

or in the line

be the equation

and the reduction to central form

But unless

is

so situated the reduction

is still

is

pos-

impossible.

The quadric surface is not a central surface.


The discussion and classification of the various non-central
quadrics
here.

is

an interesting exercise.

The present

object

is

It will not

to develop so

much

be taken up
of the theory

QUADRIC SURFACES

375

of quadric surfaces as will be useful in applications to mathe-

matical physics with especial reference to non-isotropic media.


Hereafter therefore the central quadrics and in particular the
ellipsoid will be discussed.
137.]

The tangent plane may be found by


r

dr
Since

is

= 1.
dr

self-conjugate these

= 0.

two terms are equal and

T=

dT'0

differentiation.

0.

(5)

r.
Hence
perpendicular to
normal to the surface at the extremity of the vector

The increment d r
this

is

normal be denoted by

and

N=
n

d>

let the unit

n.

(6)
.

($

Let

r.

normal be

r)

r is

<P

r)

Let p be the vector drawn from the origin perpendicular to


The perpendicular
the tangent plane, p is parallel to n.
distance from the origin to the tangent plane is the square
root of p p. It is also equal to the square root of r p.
r

Hence

= r cos
r

p.p

p)

=p

LH = r

Or

as

p and

0.

2
.

p.

._Z_
p.p

= l.

r.0-r = r.H =

But
Hence inasmuch

(r,

l.

are parallel, they are equal.

= N=

p.p

(7)

VECTOR ANALYSIS

376

On

page 108

it

was seen that the vector which has the direcand which is in magnitude equal

tion of the normal to a plane

from the origin to the plane


as
vector
coordinate
of that plane.
the
taken
be
Hence
may
r is not merely normal to
the above equation shows that
to the reciprocal of the distance

the tangent plane, but


That is, the length of

is

also the coordinate of the plane.


r is the reciprocal of the distance

from the origin to the plane tangent to the ellipsoid at


the extremity of the vector r.
The equation of the ellipsoid in plane coordinates
found by eliminating r from the two equations.
(

Hence

may be

= 1,

d>

Hence the desired equation

is

H.^- -H = I.
I

~ ^ jj
^- U
*

Tf

kk
--

4.
"I

Let

N = ui +

and
where

= a?i +

w, v,

are

the

yj

(8)

vj +

kk.

zk,

k,

reciprocals of the intercepts of the

plane N upon the axes i, j, k. Then the ellipsoid may be


written in either of the two forms familiar in Cartesian

geometry.
/y

or

N. Q-i. N = a 2

u*

77

bz vz

wz =

l.

(10)

QUADRIC SURFACES
The

377

middle points of a system of


This plane is
parallel chords in an ellipsoid is a plane.
called the diametral plane conjugate with the system of
138.]

locus of the

the plane drawn tangent to the


ellipsoid at the extremity of that one of the chords which
passes through the center.
chords.

It

is

to

parallel

any radius vector in the ellipsoid. Let s be the


vector drawn to the middle point of a chord parallel to a.
Let

r be

Let

=s+

x a.

If r is a radius vector of the ellipsoid

Hence

= (s +

a)

(s

+ 0.s-j-2S.0.a +

#2

x a)

= 1.

a.0.a =

l.

Inasmuch as the vector s bisects the chord parallel to a the


two solutions of x given by this equation are equal in magnitude and opposite in sign. Hence the coefficient of the
linear term x vanishes.

Consequently the vector

s is

A
= 0.

perpendicular to

a.

The

therefore a plane passed through


the center of the ellipsoid, perpendicular to
a, and parallel
to the tangent plane at the extremity of a.

locus of the terminus of

If

with

is

s is

any radius vector in the diametral plane conjugate

a,

= 0.

The symmetry of this equation shows that a is a radius


vector in the plane conjugate with b. Let c be a third radius
vector in the ellipsoid and let it be chosen as the line of
intersection

with a and

of the diametral
b.

Then
a

planes conjugate respectively

= 0,
c = 0,
a = 0.

(11)

VECTOR ANALYSIS

378

The

vectors

the dyadic

The

vectors

= <P

b'

a,

=a

= 1,

and

=c
ab/ = a0b = 0,

a'

The dyadic

may

= <P

c.

4>

= 1,
=b
a = 0.

a, b, c.

= 1,

c'

a'

=b

b'

=c

c'

= 0,

be therefore expressed in the forms

= a' a' + b'V + c'c',


0" = a a + bb + cc.

(12)

and
If for

c'

b,

form the system reciprocal to

a', b', c'

For

0a,0*b,<P*eby

Let

<P.

a'

are changed into

a, b, c

convenience the three directions

a, b, c,

be called a

system of three conjugate radii vectors, and if in a similar


manner the three tangent planes at their extremities be called
a system of three conjugate tangent planes, a number of
geometric theorems may be obtained from interpreting the

invariants of 0.

system of three conjugate radii vectors

be obtained in a doubly infinite number of ways.


The volume of a parallelepiped of which three concurrent

may

edges constitute a system of three conjugate radii vectors is


constant and equal in magnitude to the rectangular parallelepiped constructed upon the three semi-axes of the ellipsoid.
For let a, b, c be any system of three conjugate axes.

0The determinant
pendent of the

= aa +

or third of

bb +

0~ l

form in which

is

is

-i=[abc]

cc.

an invariant and indeexpressed.


2
.

QUADRIC SURFACES
But

0- 1

if

a 2 ii

2
Z>

jj

=a b c
= ale.
[a b c]
-i

Hence

^V"

3,

The sum

and

1
,

kk,

In like manner by inter-

This demonstrates the theorem.


<P

preting

379

possible to

it is

show that:
drawn to an

of the squares of the radii vectors

ellipsoid in a

system of three conjugate directions

is

constant

and equal to the sum of the squares of the semi-axes.


The volume of the parallelepiped, whose three concurrent
edges are in the directions of the perpendiculars upon a system
of three conjugate tangent planes and in magnitude equal to
the reciprocals of the distances of those planes from the
center of the ellipsoid, is constant and equal to the reciprocal

upon the semi-axes

of the parallelepiped constructed

of the

ellipsoid.

The sum

of the squares of the reciprocals of the three perpendiculars dropped from the origin upon a system of three
conjugate tangent planes is constant and equal to the sum of

the squares of the reciprocals of the semi-axes.


If i, j, k be three mutually perpendicular unit vectors
a
-i
4> a

Let

a,

b, c

= i.

<p

0-i

+j

+j

0-i

j
.

+k
j

k,

0-1

k.

be three radii vectors in the ellipsoid drawn

respectively parallel to

Hence

ffi

=b
.

i, j,

<P

But

k.

$
j
\- ,

b
.

=c
.

0-

j
-

= 1.
.

the three terms in this expression are the squares of the


reciprocals of the radii vectors drawn respectively in the i, j,

directions.

Hence

VECTOR ANALYSIS

380

The sum

of the squares of the reciprocals of three mutually


And
radii vectors in an ellipsoid is constant.

perpendicular
in a similar manner: the

sum

of the squares of the perpen-

diculars dropped from the origin upon three mutually perpendicular tangent planes is constant.

The equation

139.]

mined by the vector a

of the polar plane of the point deter-

is

= 1.

(13)

let s be the vector of a point in the polar plane.


The
vector of any point upon the line which joins the terminus of
s and the terminus of a is

For

+ sa
x + y

ys

If this point lies

upon the surface


i/

O + y)

g.0.

If the terminus of

_J-

/v

i/

xy

O+
s lies

a _I_

>>

=1

.0.i +

2/)

X
,

O+

a.0.a = l.

2/)

in the polar plane of a the

two values

y determined by this equation must be equal


in magnitude and opposite in
Hence the term inxy
sign.
of the ratio x

vanishes.

Hence

<P

=1

is

the desired equation of the polar plane of the terminus

of

a.

Let a be replaced by

or

z a.

It is evidently immaterial

The polar plane becomes

d>

whether the central quadric determined by * be

real or imaginary, ellipsoid or


hyperboloid.

QUADRIC SURFACES

When

381

the polar plane of the terminus of z a


the
In the limit when z becomes infinite
approaches
origin.
the polar plane becomes
z increases

0.

Hence the polar plane of the point at infinity in the


a is the same as the diametral plane conjugate with

direction
a.

This

frequently taken as the definition of the diameIn case the vector a is a radius
tral plane conjugate with a.
vector of the surface the polar plane becomes identical with

statement

is

the tangent plane at the terminus of


s

or

The equation

a.

N=1

therefore represents the tangent plane.

The
which

polar plane
is

important.

be obtained from another standpoint


Q and a plane are given,

may

P= r

and
the equation

If a quadric

(r

- C = 0,

1)

k (r

C)

represents a quadric surface which passes through the curve


and is tangent to Q along that
of intersection of Q and

curve.

In like manner

the equation

(r

if

two quadrics Q and

1)

(r

0'

Q' are given,

1)

represents a quadric surface which passes through the curves


of intersection of Q and Q' and which cuts Q and Q' at no

other points.

In case this equation

is

factorable into

two

equations which are linear in r, and which consequently represent two planes, the curves of intersection of Q and Q'
become plane and lie into those two planes.

VECTOR ANALYSIS

382

A is any point outside of the quadric and if all the tangent


are drawn, these planes envelop
planes which pass through A
If

This cone touches the quadric along a plane curve


the plane of the curve being the polar plane of the point A.
a cone.

For
of

let

any

a be the vector

drawn

to the point

A.

The equation

tangent plane to the quadric is


.

= 1.

If this plane contains A, its equation is satisfied by a. Hence


the conditions which must be satisfied by r if its tangent
are
plane passes through

= 1,
r = 1.
r

The points r therefore lie in a plane r (0 a) = 1 which


on comparison with (13) is seen to be the polar plane of A.
The quadric which passes through the curve of intersection
of this polar plane with the given quadric and which touches
the quadric along that curve
r

(r

Hence

By

d>

4>

(r

Jfe

(a

- I) 2 = 0.

through the point A,

If this passes

(a

- 1) +

is

- 1) +

- 1)

k (a
a

(a

ffi

1)

- I) 2 = 0.
- (a
r - I) 2 =

transforming the origin to the point


whose vertex is at that point.

0.

this is easily seen

to be a cone
140.]

Let

sible in the

be any self-conjugate dyadic.

It is expres-

form

=A
where A> B,
more let

ii

+ ^jj +

tfkk

are positive or negative scalars.

A <B
- Bl =

((7-

J5)

<

kk - (B - Ay

ii.

Further-

QUADRIC SURFACES

VC B

Let

Then
Let

=c

and

= cc-aa=^
c

=p

(c

The dyadic

= a.

+ a)(c-a)+(c-a)(c+a)

and

= Bl +

Then

VB

383

|-(p q

j.

= q.

qp).

(14)

has been expressed as the sum of a constant


and one half the sum

multiple of the idemfactor

pq +
The reduction has assumed
are different

qp.

tacitly that the constants -4, B,

from each other and from

This expression for

zero.

closely related to the circular

is

sections of the quadric surface


r

Substituting the value of

Br

Let

r
r

<P,

+
r

This

= 1 becomes
r = 1.
r

p q

be any plane perpendicular to

= 1.

=n

p.

n q

By

substitution

0.

a sphere because the terms of the second order all


have the same coefficient B. If the equation of this sphere
is

be subtracted from that of the given quadric, the resulting


equation is that of a quadric which passes through the inter-

and the given quadric.

section of the sphere

r (r

w)

Hence the sphere and the quadric


r

and

difference

= 0.
intersect in

curves lying in the planes

The

= n.

two plane

VECTOR ANALYSIS

384

Inasmuch

as these curves lie

upon a sphere they

are circles.

Hence planes perpendicular to p cut the quadric in circles.


it may be shown that planes perpendicular to

In like manner

q cut the quadric hi


follows

The proof may be conducted

circles.

.Z?

p q

= 1.

If r is a radius vector in the plane passed

of the quadric perpendicular to p or q, the

Hence the vector

ishes.

as

through the center


term r p q r van-

r in this plane satisfies the equation

B r-r = l
and

is

of constant length.
The section is therefore a circular
The radius of the section is equal hi length to the

section.

mean

semi-axis of the quadric.

For convenience
stants

may

A, B,

let the

The

con-

reciprocal dyadic

0~l

quadric be an ellipsoid.

The

are then positive.

be reduced hi a similar manner.

0-1

Let

Then

i i

k k

ti

0-1

= ff -

dd

Jo

= *\\

(f

d) (f

- d)

(.

+ (f-d)(f +
+

Let

Then

0-i

=u

and

=iI +

(u

d)j

- d = v.

v u).

(15)

QUADRIC SURFACES

385

The vectors u and v are connected intimately with the circular cylinders which envelop the ellipsoid
r

=1

or

-N N+N

For
If

N=

<P~l

N=

uv

1.

1.

now N

be perpendicular to u or v the second term, namely,


N, vanishes and hence the equation becomes

uv

N N=
.

That

is,

the vector

is

B.

of constant length.

But

the equation

the equation of a cylinder of which the elements and tangent planes are parallel to n. If then N BT is constant the
is

The
cylinder is a circular cylinder enveloping the ellipsoid.
radius of the cylinder is equal in length to the mean semi-axis
of the ellipsoid.

There are consequently two planes passing through the


origin and cutting out circles from the ellipsoid. The normals
to these planes are p

extremities of the

two

and

mean

q.

The

through the
There are also

circles pass

axis of the ellipsoid.

circular cylinders enveloping the ellipsoid.

of the axes of these cylinders are

u and

v.

The

Two

these cylinders pass through the extremities of the


of the ellipsoid.

These

direction

elements of

mean

axis

can be seen geometrically as follows. Pass


a plane through the mean axis and rotate it about that
The section is an
axis from the major to the minor axis.
ellipse.

ellipsoid.

results

One

axis of this

ellipse

is

the

mean

axis of

This remains constant during the rotation.

the

The

other axis of the ellipse varies in length from the major to the
minor axis of the ellipsoid and hence at some stage must pass
At this stage of
through a length equal to the mean axis.
25

VECTOR ANALYSIS

386

the rotation the section

is

In like manner consider

circle.

the projection or shadow of the ellipsoid cast upon a plane


a point at an infinite distance
parallel to the mean axis by

from that plane and in a direction perpendicular to it. As the


ellipsoid is rotated about its mean axis, from the position in
which the major axis is perpendicular to the plane of projection to the position hi
to that plane, the

which the minor axis

is

perpendicular

shadow and the projecting cylinder have the

axis of the ellipsoid as one axis. The other axis changes


from the minor axis of the ellipsoid to the major and hence at
some stage of the rotation it passes through a value equal to

mean

the

mean

axis.

At

the shadow and projecting

this stage

cylinder are circular.

The

necessary and sufficient condition that r be the major


1
ellipsoid r d> r

or minor semi-axis of the section of the

by a plane passing through the center and perpendicular to a


that a, r, and
r be coplanar.

is

Let

and
dr

dr

if

r is to be a

maximum

dr

Furthermore

= 0.
r = 0,
a = 0.
r = 0,

Differentiate

=1

major or minor axis of the section for r is a


mininum and hence is perpendicular to dr.
;

or a

These three equations show that a, r, and


r are all orthogonal to the same vector dr. Hence they are coplanar.
[a r

Conversely

if

dr may be chosen
Then

d>

r]

[a r

r]

= 0.
= 0,

perpendicular to their

0.

(16)

common

plane.

QUADRIC SURFACES
Hence

maximum

r is a

or a

mininum and

387
is

one of the prin-

cipal semi-axes of the section perpendicular to a.


141.]

form
r

instead of

This

may

an advantage to write the equation

It is frequently

of an ellipsoid in the

be done

because

= 1,
r = 1.
r

(17)

if

i
kk
j j
-h
H-2
2
2
i

'

a dyadic such that W* is equal to 0.


a square root of
and written as $*.
is

membered that

there

example,
i i

__

For

-Ha

'

this reason it is necessary to bear in

meant by
been denoted by
is

for

_ kk
__

j *
j
*
i

root which

be regarded as
But it must be re-

other square roots of

are

and

W may

mind that the square

that particular one which has

<P* is

The equation

of the ellipsoid
r

or

Let

(?T.r).

r'

may

be written in the form

= 1,
(?F.r) = l.
W

be the radius vector of a unit sphere.

the sphere

is
r'

r'

= 1.

The equation of

VECTOR ANALYSIS

388

evident that an ellipsoid may be


transformed into a unit sphere by applying the operator
If r''=

^'rit becomes

r, and vice versa, the unit sphere may


be transformed into an ellipsoid by applying the inverse operFurthermore if a, b, c are
ator W~ l to each radius vector r'.

to each radius vector

a system of three conjugate radii vectors in an ellipsoid


2

=b

2P 2

Wz

=b

a.
a
If for the

moment
a'
a'

Hence the three


which

5F2

denote respectively

a', b', c'

r.c,

=c
c = c

=V
b' = V

=
=

b'

a'

c'

radii vectors

a,

b,

= 1,
a' = 0.

c'

c'

c'

of the unit sphere into

c'
a', b',

three conjugate radii vectors in the ellipsoid are trans-

l
are mutually orthogonal.
formed by the operator
They
form a right-handed or left-handed system of three mutually

perpendicular unit vectors.

Theorem
ellipsoid

Any

by means

ellipsoid

of a

be transformed into any other

may

homogeneous

strain.

Let the equations of the ellipsoids be


r

and

By means
ellipsoid are

of the strain

= 1,
W r = 1.
r

0* the

changed into the

radii vectors r' of

r'T'

r'=0*T,
By means

of the strain

radii vectors r of the first

a unit sphere

= l.

W~* the radii vectors r' of this unit


manner into the radii vectors f
Hence by the product r is changed

sphere are transformed in like


of the second ellipsoid.
into r.
f

V-*

0i

r.

(19)

QUADRIC SURFACES

389

The transformation may be accomplished in more ways


The radii vectors r' of the unit sphere may be
transformed among themselves by means of a rotation with or
than one.

without a perversion.

three mutually orthogonal unit

Any

vectors in the sphere may be changed into any three others.


Hence the semi-axes of the first ellipsoid may be carried over

by a

suitable strain into the semi-axes of the second.

strain is then completely determined

The

and the transformation

can be performed in only one way.


142.] The equation of a family of confocal quadric surfaces

is

az
If r

2
b'

1 and r

c'

1 are two surfaces of the

family,
i i

i i

nr
*

^tt

Hence

j *
*
j

L_

>

a2

j:

T2/

kk

j j

&

1 1 ~f"

0- - y-i=

Cb

k k

62

7i

72.

&

1 1

c2

"^~

n*&

(C

- n^ (ii + j j +
= (% - wi) T

(w a

---

71

K.

kk)

The necessary and

sufficient condition that the

=1
r = 1

and
be confocal,

r
is

two quadrics

that the reciprocals of

and

differ

by a

multiple of the idemfactor

0-i

y-i

= a .i.

(21)

VECTOR ANALYSIS

390
If

two confocal quadrics

Let the quadrics be

and

Let

Then

the quadrics

1 .

r,

s'.

be written in terms of

may

and

= 1,
W r = 1.
= W
r and
s and r = W~
r

*'

= 0-

do so at right angles.

intersect, they

s'.

and

s'

as

= l,
?T-i.s' = l,

0- 1

.s

where by the confocal property,

0-i
If the
is

yr-i

=xi

quadrics intersect at r the condition for perpendicularity


r and W r be perpendicular.
That is,

that the normals

But

= 0= ~
-s
y-

-1

s'

s'

s'

s'

= 0.
y~ +
s =
l

1 .
s'

5P-1 .

I)

8,

s'= 1

-s

SF-1

s'.

In like manner
r

- CM
a?

Add

Hence

- (0- - I)
= W-*
= s 008
= s (0~
s
)
=
s
0,

a;

s'

s'

aj

JP"-

= 00s = 8
s
s =
1

s'

'

s'

a; s'.

1.

a;

s'.

s'

and the theorem


If the

s'

s'

is

proved.
parameter n be allowed to vary from

oo to

+ oo

the

resulting confocal quadrics will consist of three families of

which one
and the

another, hyperboloids of one sheet ;


third, hyperboloids of two sheets.
By the foregoing
is ellipsoids

QUADRIC SURFACES

391

theorem each surface of any one family cuts every surface


The surfaces form a triply
of the other two orthogonally.
orthogonal system. The lines of intersection of two families
(say the family of one-sheeted and the family of two-sheeted
hyperboloids) cut orthogonally the other family

the family

The points in which two ellipsoids are cut by


of ellipsoids.
these lines are called corresponding points upon the two ellipsoids.

It

may

be shown that the ratios of the components of

the radius vector of a point to the axes of the ellipsoid


through that point are the same for any two corresponding
points.

For

let

any

ellipsoid be given

The neighboring

by the dyadic

ellipsoid in the family is represented

dyadic

U
ndn

W=

and

,
*^

dn

bz

JF-i= 0-1

Inasmuch as

kk

JJ
*^

a?

dn

dn.

homologous (see Ex.

are

by the

8, p.

330)

dyadics they may be treated as ordinary scalars in algebra.


Therefore if terms of order higher than the first in dn be

V=* + Idn.

omitted,

The two neighboring

ellipsoids are then


r

and

By

(19)

d>

dn)

(0 +

dn)~*

(I

dn)

= 1.
0*

r,

4>

dn)-

(I

= 1,

(0 +

_i

* .

r,

= r - d-~n

r.

VECTOR ANALYSIS

392

vectors f and r differ by a multiple of

The

<P

which

is

(P.
Hence the termini of f and
perpendicular to the ellipsoid
for
r are corresponding points,
they lie upon one of the lines

which cut the family of ellipsoids orthogonally. The com= x and


ponents of r and f in the direction i are r i
r

= _a = r

dn x
'-x- --^

- dn i.
.

Of

The

ratio of these

The axes

(7

-=!
x

is

of the ellipsoids in the direction

Their ratio

a.

components

TL

--

2 a2

are

Va2

dn and

is

Va 2

dn

V& 2

,
In v
like manner

dn

s
2

-i

j
dn

by

2 a2

dn = y- j Vc 2
and

x
x

dn = z

-.

Hence the ratios of the components of the vectors r and r


drawn to corresponding points upon two neighboring ellipsoids only differ at most by terms of the second order in d n
from the

the axes of those ellipsoids. It follows


immediately that the ratios of the components of the vectors
drawn to corresponding points upon any two ellipsoids, separatios of

rated by a finite variation in the parameter n, only differ at


most by terms of the first order in dn from the ratios of the

axes of the ellipsoids and hence must be identical with them.


Thio completes the demonstration.

The Propagation of Light in Crystals 1


143.]

The

electromagnetic equations of the ether or of any

infinite isotropic

netic
1

To

medium which

is

transparent to electromag-

waves may be written in the form

The

following discussion must be regarded as mathematical not physical.


from the standpoint of physics would be out of place here.

treat the subject

THE PROPAGATION OF LIGHT IN CRYSTALS


/72-n

V.D = O

+ UD + vF-o,

Pot

393

(i)

ct t

where D

is

the electric displacement satisfying the hydro-dya constant of the dielectric meas0,

namic equation V D =
ured in electromagnetic
due

to the function V.

units,

and

VV

the electrostatic force

In case the medium

is

E becomes a linear vector function

constant

not isotropic the


This function

(P.

self-conjugate as is evident from physical considerations.


For convenience it will be taken as 4 IT 0. The equations

is

then become

V-D=0.

VF=0,
Operate by

x.

d2

x Pot -n

Ct/

The

last

(2)

+47rVxVx0-D =

r
/
6

0.

(3)

term disappears owing to the fact that the curl of

the derivative

VF" vanishes (page

167).

The equation may

also be written as

Pot

d2 D
-r-y

+47rVxVx0.D = 0.

VxVx=VV

But
Remembering
d D

that

D and

vanish and that Pot

(3)'

V.

consequently

and
Ct v

V V

is

equal to

TT

the

Ct t

equation reduces at once to

? = V.V<P.D-VV.0.D,

= 0.

(4)

Ct t

Suppose that the vibration D is harmonic. Let r be the


vector drawn from a fixed origin to any point of space.

VECTOR ANALYSIS

394

Then
where

and

= A cos

That

the direction A.

vance

the

is,

in the direction

the quotient of

is

and n a constant

are constant vectors

represents a train of waves.

wave advances

n )

(m

scalar

The vibrations take place in


wave is plane polarized. The
m. The velocity v of that ad-

n by m,

the magnitude of the vector

wave is an electromagnetic wave in the medium


considered it must satisfy the two equations of that medium.
m.

If this

Substitute the value of

The

value of

in those equations.

V V

D,

VV

and

D,

D may

be

obtained most easily by assuming the direction i to be coincident with m.


r then reduces to
i
r which is equal to

m x.

The

and

variables y

no longer occur in D.

= A cos

= i 5D =
-=

VV
V

m2

m2 i

m A

V V

(m x

dX

V V

Hence

t)

A m

sin

(m x

cos

(m x

n f)

sin

Hence

cos

(m

m m

(mx

<P

nf).

t~)

VV.<P.D = -mm.<P.D.
Moreover

Hence

if

t*

n z D.

the harmonic vibration

tions (4) of the

is

to satisfy the equa-

medium

w2 D
and

d2

= m m <P D m m
m A = 0.
.

<P

(5)
(6)

THE PROPAGATION OF LIGHT IN CRYSTALS


The

395

once that the vibrations must

latter equation states at

m of propagation of the waves.


be put in the form

be transverse to the direction

The former equation may

0.D.
Introduce

The vector s

is

in the direction of advance m.

the quotient of
velocity of the wave.
of

=-

s is

(5)'

by

This

n.

The

vector

The magnitude

the reciprocal of the


may therefore be called
is

the wave-slowness.

D=
This

may

D.

s s

also be written as

=-

(s

D)

=s

x (0
n

Dividing by the scalar factor cos (m x

A = sx(0.A)xs =
It

evident that the

is

ss

wave slowness

D) x

s.

),

88.0.
s

A.

(7)

depends not at

all

upon the phrase of the vibration but only upon its direction.
The motion of a wave not plane polarized may be discussed by
decomposing the wave into waves which are plane polarized.
Let a be a vector drawn in the direction A of the
144.]
displacement and

let

the magnitude of a be so determined

that

The equation
a

1.

(8)

(7) then becomes reduced to the form

= sx(0-a)xs =

s.s

0.a = ss.0a

a = 1.

(9)

(8)

These are the equations by which the discussion of the velocity


or rather the slowness of propagation of a wave in different
directions in a non-isotropic

=s

medium may be
a

=s

carried on.
s.

(10}

VECTOR ANALYSIS

396

Hence the wave slowness


direction a

equal in

is

drawn

radius vector

due to a displacement in the

magnitude (but not in direction)


ellipsoid a

in the

to the

1 in that

direction.

axa = = ss

a x

= s.s(ax0a)'0a = axs.<Pa
But the

first

a twice and vanishes.

term contains
a x

The wave-slowness

[a

a]

0.

a.

Hence
(11)

therefore lies in a plane with the

direction a of displacement

a drawn to the

and the normal

1 at the terminus of a. Since s is perpena


ellipsoid a
dicular to a and equal in magnitude to a it is evidently completely determined except as regards sign when the direction

is

known.

Given the direction

of displacement the line of

advance of the wave compatible with the displacement is completely determined, the velocity of the advance is likewise

known.

The wave however may advance

in either direction

along that line. By reference to page 386, equation (11) is seen


to be the condition that a shall be one of the principal axes of
the ellipsoid formed by passing a plane through the ellipsoid

perpendicular to
there are

two

s.

Hence

for

principal axes of the ellipse cut

by a plane passed through


line of advance.

minateness of

any given direction of advance


These are the

possible lines of displacement.

from the

ellipsoid a

=1

the center perpendicular to the

To these statements concerning the deterwhen a is given and of a when s is given just

such exceptions occur as are obvious geometrically. If a and


<P a are parallel s may have any direction perpendicular to a.

This happens when a


axes of the ellipsoid.

is

directed along one of the principal


one of the

If s is perpendicular to

circular sections of the ellipsoid a

plane of the section.

may have any direction in the

THE PROPAGATION OF LIGHT IN CRYSTALS

When
slowness

397

the direction of displacement is allowed to vary the


To obtain the locus of the terminus of s, a

a varies.

must be eliminated from the equation

=s
-s
(I
a

or

The dyadic

&

ss

s s

<P)

<P

a
a

= 0.

(12)

in the parenthesis is planar because

vectors parallel to

The

a.

it

annihilates

This

third or determinant is zero.

gives immediately

s.s<P

(I

or

(tf-i-s.

This

is

= 0,
+ ss) 3 = 0.

ss.<P) 3
I

a scalar equation in the vector

(13)

It is the locus of

s.

the extremity of s when a is given all possible directions.


number of transformations may be made. By Ex. 19, p. 331,
(

e f) 3

0^

f <P3 .

Hence

common

Dividing out the

factor

and remembering that

self-conjugate.

s.

1.8
8-8

(0-

-&

si)-

S.

Hence

\-

Let

1-8.

-=

= 0.

= 0.

(14)

is

VECTOR ANALYSIS

398

i-.

?/
Let

Then

ic

yj

& 2/

and

\-c

= x2 +

z 2.

the equation of the surface in Cartesian coordinates

T
,

T
,

"

~~cf

(14)'

l-fl
The equation
directly

(0-1

The determinant
a2

By means

_8

obtained

ss) 3 :=0.

of this dyadic is

xz

x y
2

x y

may be

in Cartesian coordinates

from

is

Z>

of the relation s 2

= xz +

0.

(13)'

+
z 2 this

assumes the

forms

x2

a2
95
s'

ic

2
Z>

a*

y
*

2
c2 z

9 T
o'
_i

I
I

9
z

95

s'

==

**i

c^

This equation appears to be of the sixth degree. It is however of only the fourth. The terms of the sixth order cancel
out.

The

vector

represents the wave-slowness.

Suppose that a

plane wave polarized in the direction a passes the origin at a

THE PROPAGATION OF LIGHT IN CRYSTALS


At

certain instant of time with this slowness.

unit of time

it

will

have travelled in the direction

equal to the reciprocal of the

magnitude

be in this position represented


If

of

= ui +

s.

by the vector
vj

399

the end of a
s,

a distance

The plane

will

(page 108).

wls.

the plane at the expiration of the unit tune cuts off intercepts
upon the axes equal to the reciprocals of u, v, w. These
quantities are therefore the plane coordinates of the plane.

They

are connected with the coordinates of the points in the

plane by the relation

If different plane waves polarized in all possible different


directions a be supposed to pass through the origin at the

same instant they will envelop a surface at the end of a unit


This surface is known as the wave-surface. The
of time.
perpendicular upon a tangent plane of the wave-surface is the
reciprocal of the slowness and gives the velocity with which
the wave travels in that direction.

The equation

of the wave-

surface in plane coordinates u, v, w is identical with the equation for the locus of the terminus of the slowness vector B.

The equation

is

uz

v2

w^

^o^o^o =
P

9
a4

where

= u2 +

vz

~T""*>

o*

w*.

forms given previously.

This

The

may

\/

72

(15)

be written in any of the


is known as FresneVs

surface

Wave-Surface. The equations in vector form are given on


page 397 if the variable vector s be regarded as determining a
plane instead of a point.
145.]

In an isotropic medium the direction of a ray of

light is perpendicular to the wave-front.

the direction of the wave's advance.

The

It is the

same as

velocity of the ray

VECTOR ANALYSIS

400

In a non-isotropic
equal to the velocity of the wave.
The ray does not travel perthis is no longer true.
that is, in the direction of the
wave-front
pendicular to the
wave's advance. And the velocity with which the ray travels
is

medium

of the wave. In fact, whereas the


greater than the velocity
off
wave-front travels
always tangent to the wave-surface, the
is

ray travels along the radius vector drawn to the point of tangency of the wave-plane. The wave-planes envelop the
wave-surface; the termini of the rays are situated upon it.

Thus in the wave-surface the radius vector represents in magnitude and direction the velocity of a ray and the perpendicular

direction the velocity of the wave.

surface the surface which

wave

magnitude and
instead of the wavein

upon the tangent plane represents


If

the locus of the extremity of the


slowness be considered it is seen that the radius vector
is

wave; and the perpendicular


of the ray.
the
slowness
the
upon
tangent plane,
Let v' be the velocity of the ray. Then s v'
1 because
represents the slowness of the

the extremity of v' lies in the plane denoted by s. Moreover


the condition that v' be the point of tangency gives d v' perpendicular to

In like manner

s.

if s'

ray and v the velocity of the wave,


of tangency gives

and
v'

d v' = 0,

may be

v'

=1
s'

and

2a

=g

ds

<p

a_ s

da

v
d

v'

0,

= 1 and the condition

= 1,

= 0,

dv

d>

da

</>

da..

Multiply by a and take account of the relations a


.
. d a =
and a a = s s. Then

= 0,

a,

ads +
s s

s'

as follows.

_ ss
a

(16)

and

a, s,

<P

Hence

to v.

s'

expressed in terms of
a

d&

d &' perpendicular

be the slowness of the

s'

and

THE PROPAGATION OF LIGHT IN CRYSTALS


ds

or

But

ds

since v'

ds

a.

(s

0, v'

v'

= x (s

and

a)

= 0,

<P

401

= 0.
a have the same

direction.

v' = # (s

Hence

v'

=s

a)
.

a;

s.

a
(17)

v'

a),
CP

a
s

$as0a =

0.

Hence the ray velocity

v' is perpendicular to
a, that is, the
the
to
the
tangent plane
ray velocity
ellipsoid at the
extremity of the radius vector a drawn in the direction of the
displacement. Equation (17) shows that v' is coplanar with

lies in

a and
plane.

The

The

s.

vectors

In that plane

angle from

Making use
(16) (17),

it is

a,

s is

to v'

is

a,

s,

and

v'

therefore He in one

perpendicular to a ; and v', to


equal to the angle from a to

of the relations already found (8) (9) (11)

easy to show that the two systems of vectors


a

and

(0

a)

a be replaced
If
are reciprocal systems.
tions take on the symmetrical form

by

a,

v',

v'

a,

s,

X
a'

the equa-

=a a
a a' = 1,
s
v' = 1,
v'-v' = a'.a'
v'.a' =
a = s x a' x s
a' = v' x a x v'
s = a x v' x a
v' = a' x s x a'
-1
a
a = 1
a' = 1.
a'
s

a.
a.

(18)

Thus

a dual relation exists between the direction of displace-

ment, the ray- velocity, and the ellipsoid


26

on the one hand

VECTOR ANALYSIS

402

and the normal to the


~l

on the

ellipsoid
It

146.]

ellipsoid, the

wave-slowness, and the

other.

was seen that

if s

was normal

to

one of the

cir-

the displacement a could take place in any


For all directions in
direction in the plane of that section.
cular sections of

this plane the

wave-slowness had the same direction and the

Hence the wave-surface has a singular

same magnitude.

plane perpendicular to s. This plane is tangent to the surface


along a curve instead of at a single point. Hence if a wave
travels in the direction s the ray travels along the elements of

the cone drawn from the center of the wave-surface to this

curve in which the singular plane touches the surface. The


two directions s which are normal to the circular sections of
are called the

wave

primary

velocities

In like manner
[<P

The

These are the axes of equal

optic axes.

but unequal ray

velocities.

v'

being coplanar with a and

v'

a]

[a' v'

0~l

a']

a
0.

equation states that if a plane be passed through


the center of the ellipsoid @~l perpendicular to v', then a'
last

which

a will be directed along one of the prinis equal to


Hence if a ray is to take a definite
axes
of
the
section.
cipal
It is more condirection a' may have one of two directions.

venient however to regard

The

first

v'

as a vector determining a plane.

equation

[0

v' a] ==

states that a is the radius vector

drawn

in the ellipsoid

to

the point of tangency of one of the principal elements of the


cylinder circumscribed about
parallel to v' if by a principal
element is meant an element passing through the extremities
:

of the

major or minor axes of orthogonal plane

of that cylinder.

Hence given the

direction

v'

sections

of the ray, the

two possible directions of displacement are those

radii vectors

VARIABLE DYADICS
of the ellipsoid

which

lie in

403

the principal planes of the cylin-

der circumscribed about the ellipsoid parallel to v'.


If the cylinder is one of the two circular cylinders which
the direction of displacement
may be circumscribed about

may

be any direction in the plane passed through the center


and containing the common curve of tangency

of the ellipsoid

The ray-velocity for all


of the cylinder with the ellipsoid.
these directions of displacement has the same direction and
It is therefore a line drawn to one
the same magnitude.
At this singukr
of the singular points of the wave-surface.

point there are an infinite number of tangent planes enveloping a cone. The wave-velocity may be equal in magnitude
and direction to the perpendicular drawn from the origin to

any of these planes.

The

directions of the axes of the

circular cylinders circumscriptible about the ellipsoid

two
are

the directions of equal ray-velocity but unequal wave-velocity.


They are the radii drawn to the singular points of the wavesurface

and are

called

the secondary optic axes.

travels along one of the secondary optic axes the

If a ray

wave planes

travel along the elements of a cone.

Variable Dyadics.
147.]

The

The Differential and Integral Calculus

Hitherto the dyadics considered have been constant.


make up and the scalar

vectors which entered into their

which occurred in the expansion in nonion form


have been constants. For the elements of the theory and for
coefficients

elementary applications these constant dyadics suffice. The


introduction of variable dyadics, however, leads to a simplification

and unification of the

differential

and integral calculus of

vectors, and furthermore variable dyadics become a necessity


for instance, in the theory
in the more advanced applications
of the curvature of surfaces and in the dynamics of a rigid
body one point of which 4s fixed.

VECTOR ANALYSIS

404
Let

W be

Let

a vector function of position in space.

be

the vector drawn from a fixed origin to any point in space.

dr = dxi + dy j + dz
d

9W
W = d x 9W
+ a y ---h
^
v
ax
,

k,

9W

d z -7

Hence

W=dr

\i

9W
--dx

The expression enclosed

i-

spect of #,

and

9W
---h k 9W)
->
dz

dy

The antecedents

first partial

The expression

2/, z.

analogous to del

in the braces is a dyadic.


It thus
is a linear function of d r, the

appears that the diif erential


differential change of position.

and the consequents the

<y

is

derivatives of

found

in a

will in fact be denoted

are

k,

i, j,

W with

re-

manner precisely

V W.

by

(1)

Then

<2W

This equation

is

= dr.VW.

(2)

like the one for the differential of a scalar

function F.

dV=dr VF.
It

may be regarded

nonion form

as defining

VW.

expanded into

If

VW becomes
dx

Qx

Qx

9Y
3y

+
if

9Y

9Z

Toz-+kj^-+kk^
arz
dz
W
i

-,

VARIABLE DYADICS

405

and
The operators
x which were applied to a vector
function now become superfluous from a purely analytic
For they are nothing more nor less than the
standpoint.

V W.

and the vector of the dyadic

scalar

W = (VW)^
curl W = V x W = (V W)
divW = V.

The

(4)
(5)

analytic advantages of the introduction of the variable

VW

In the first place the operapplied to a vector function just as to a scalar


In the second place the two operators
and
function.
x
are reduced to positions as functions of the dyadic
W. On

dyadic
ator

are therefore these.

V may be

the other hand from the standpoint of physics nothing is to


be gained and indeed much may be lost if the important inas the divergence and curl
and
x
terpretations of

be forgotten and their places taken by the analytic idea


of the scalar and vector of VW.

of

If

the vector function

be the derivative of a scalar

function V,

=dT' VVT,
Qzy
Qzy

y
+ ij = ^- +
VVF= ii 32
^r
dx
d x*

where

ik 7:=-,
d x dz

dy

result of applying

v
^J

V twice

This dyadic

be a dyadic.
is zero its scalar

dy

fcj o

dz 3x

The

dy*

~f~

J J

dy 3x

n
dz

n
o
z*

'

to a scalar function is seen to

is self -conjugate.

V V V is evidently
= 52 Y

V.V V= (VV V) a

V
*

'

+
-=-j
z
5x

92

Its vector

.-s-j
d 2

52

+ -^-Tz

VX VV

VECTOR ANALYSIS

406
If

an attempt were made to apply the operator

V three

cally to a scalar function

sum

symbolitimes, the result would be a

of twenty-seven terms like

9s

1 1 1

93
i j

r,

d x*

k*

T
etc.

^- ,
a x d y dz

Three vectors are placed in juxtaposition


without any sign of multiplication. Such expressions will
not be discussed here. In a similar manner if the operator V
This

is

a triadic.

be applied twice to a vector function, or once to a dyadic function of position in space, the result will be a triadic and hence
outside the limits set to the discussion here.

x and

may however be applied


respectively a dyadic and a vector.

The

to a dyadic

operators
to yield

90
90
- + kx-90
+
Vx0 = lx-5r
ax jx 3
dy
3

-,

and

Or

90

+
^3x

90

-r-

dy

(7)

90
.

-r-.
3z

(8)

V = ui + vj + wk,

If

where

n, v,

w are vector

functions of position in space,

Vx

=V

x u

<P

=V

if

Vxvj

+V
u

v
v

Vx

+V w

+k

wk,

(7)'

k.

(8)'

w,

9y
and

V.0 = ^ + ^
+ ^.
d
dx

dz

(8)"

In a similar manner the scalar operators (a V) and (V V)


may be applied to 0. The result is in each case a dyadic,

IE DYADIC S

VECTOR ANALYSIS

408

The

bolically as a vector.

must be
to which

it is

x (vw)

(v

X w)

[V
(v

Hence
It

which the

implies

factor of a product

[V x (vw)] v + [V x (vw)]^

(v w)] v

Hence

[V

it

=V

w)] w

(v

x v w,

V w] v = v x V w.
Vx (vw) = Vxvw vxVw.

[V x

which

Thus

applied.

[V x

148.]

upon each

carried out in turn

Again

differentiations

[v

[V

(v

x w)] v + [V

(v

x w)] w

=Vv

(v

x w)]^

x w,

v (w x v)] = V w x
V(vxw) = Vvxw v^xv.
x w)] v

was seen

(Art. 79) that

and the

if

v.

denote a curve of

point
point is r the line integral of the derivative of a scalar function taken along the
curve is equal to the difference between the values of that
is r

initial

function at r and r

final

fZr.VF=F(r)-F(r ).

*> c

In like manner

and

idr.VW

Jc

id r

W (r

W(r)

VW =

),

0.

/o
It

may

be well to note that the integrals

fdr.VW

and

fvw

VW

dr

is a dyadic.
The
by no means the same thing.
vector d r cannot be placed arbitrarily upon either side of it.

are

VARIABLE DYADICS
Owing

409

to the fundamental equation (2) the differential

dr

VW.

The differentials must be written


necessarily precedes
before the integrands in most cases. For the sake of uniformity they always will be so placed.
Passing to surface integrals, the following formulae, some
of which have been given before and some of which are new,
may be mentioned.

f f dax

VF=

t dr

Cfd&x VW=Cdi W

ffdaVxW =
if

d&

<P

idi'W
/

dr

0.

The

line integrals are taken over the complete bounding curve


of the surface over which the surface integrals are taken.
In

like

manner the following

relations exist

between volume and

surface integrals.

fffd, V.W
f f fdv V X

///

W=

Cda, X

*-//

f f f^V VX 0=:CCdSLX

$.

VECTOR ANALYSIS

410

surface integrals are taken over the complete bounding


surface of the region throughout which the volume integrals

The

are taken.

Numerous formulae

of integration

by parts

like those

upon

The reader will find no difficulty in


himself.
The integrating operators may

page 250 might be added.

obtaining them for


also be extended to other cases.

To

the potentials of scalar


and vector functions the potential, Pot <P, of a dyadic may be
added. The Newtonian of a vector function and the Laplacian and Maxwellian of dyadics

Pot

fff

'"

w <>"

fff
*

The

be denned.

'-///

New W =
Lap *

may

12

analytic theory of these integrals may be developed as


The most natural way in which the demonstrations

before.

may
sum

be given

is

by considering the vector function

W as

the

of its components,

Yj + Zk
and the dyadic
as expressed with the constant consequents
k and variable antecedents n, v, w, or vice versa,

i, j,

These matters will be


tering

upon them

at all

left at this point.

was

The

object of en-

to indicate the natural extensions

which occur when variable dyadics are considered. These extensions differ so slightly from the simple cases which have

THE CURVATURE OF SURFACES

411

gone before that it is far better to leave the details to be worked


out or assumed from analogy whenever they may be needed
rather than to attempt to develop them in advance. It is sufficient merely to

may

mention what the extensions are and

how they

be treated.

The Curvature of Surfaces 1

There are two different methods of treating the curvature of surfaces. In one the surface is expressed in para149. ]

metic form by three equations

=fi <X

*0

or

y =/a O' *0
r

=f

(u,

=/8 ( w

>

v)

t>).

analogous to the method followed (Art. 57) in dealing


with curvature and torsion of curves and it is the method

This

is

employed by Fehr in the book to which reference was made.


In the second method the surface is expressed by a single
equation connecting the variables x,y,z
,

The

latter

method

z)

thus

0.

of treatments affords a simple application of

the differential calculus of variable dyadics.


Moreover, the
results
to
the
most
lead
connected
naturally
important
dyadics

with the elementary theory of surfaces.


Let r be a radius vector drawn from an arbitrary fixed
The increment d r
origin to a variable point of the surface.
the surface or in the tangent plane
at the terminus of r.

lies in

drawn

to the surface

dF=dr* VF=Q.
Hence the
surface.

derivative

V.Fis

collinear with the

Moreover, inasmuch

as

normal to the

F and the negative of F when

1 Much of what follows is


practically free from the use of dyadics.
especially true of the treatment of geodetics, Arts. 155-157.

This

is

VECTOR ANALYSIS

412

equated to zero give the same geometric surface,


considered as the normal

V F may be

either side of the surface.

upon

case the surface belongs to the family defined

In

by

F (#, y, z) = const.

V F lies upon that side upon which the constant


Let V F be represented by N the magnitude of

the normal
increases.

which may be denoted by N, and


in the direction of N.

let

n be a unit normal drawn

Then

(1)

If & is the vector

the terminus of

r, s

drawn to any point in the tangent plane at


r and n are perpendicular. Consequently

the equation of the tangent plane is

(s-r).
and in

like

V^=0.

manner the equation

of the normal line is

V^=0,
s = r + kV F

(s-r) X
or

where k

a variable parameter. These equations may be


translated into Cartesian form and give the familiar results.
The variation dn of the unit normal to a surface
150.]
is

plays an important part in the theory of curvature,


perpendicular to n because n is a unit vector.

VF

dn. is

THE CURVATURE OF SURFACES

413

-/v

The dyadic
dicular to

nn

an idemf actor for

is

all

vectors perpen-

n and an annihilator for vectors

parallel

to

n.

Hence

dn
and

= d n,

V^.(I-nn)=0,

dn =

Hence

N d r VV F* (I
= di

dr

But
Hence

n n)

(I

tfi

nn).

(I

VV.F. (I-nn)-

dn = dr<-(I-nn)

Let

nn).

= (I-nn).'.-

Then

= dr

dn

0.

In the vicinity of any point upon a surface the variation d n of


the unit normal is a linear function of the variation of the
radius vector

r.

The dyadic

N
Evidently (I

is

self-conjugate.

= (I - n n),

- n n) c =

is self-conjugate.

a vector parallel to

(I

Hence

For

VV F) c

- n n)

(I

and by (6) Art. 147

C is equal to 0.

n, the dyadic

- n n)

produces

When applied
zero.

lie

to

It is there-

fore planar and in fact uniplanar because self-conjugate.

antecedents and the consequents

VVF
The

in the tangent plane to

VECTOR ANALYSIS

414

(Art. 116) to reduce

It is possible

the surface.

form

=a

i'i'

to the

b j'j'

(5)

and j' are two perpendicular unit vectors lying in the


tangent plane and a and b are positive or negative scalars.

where

i'

dn = dr
The

vectors

and the

i', j'

151.]

The

conic r

is

=1

a and

is, if

the point

but

if

j'j').

have the same


the conic

is

from point to point

variable.
is

surface at the point in question.

that

scalars a, b vary

The dyadic

of the surface.

i'i'

(a

called the indicatrix of the


If this conic is

an

sign, the surface is

an hyperbola, that

ellipse,

convex at

is, if

a and b

have opposite signs the surface is concavo-convex. The curve


. r
1 may be regarded as approximately equal to the
r .
intersection of the surface with a plane drawn parallel to the
tangent plane and near to it. If r
the result is a pair of straight lines.
totes of the conic.
if

imaginary, an

r be set equal to zero

These are the asympan hyperbola

If they are real the conic is

ellipse.

Two

directions

on the surface which

are parallel to conjugate diameters of the conic are called con-

jugate directions. The directions on the surface which coincide with the directions of the principal axes i', j' of the
indicatrix are

known

as the principal directions.


They are a
The directions upon the

special case of conjugate directions.

surface which coincide with the directions of the asymptotes


of the indicatrix are known as asymptotic directions.
In case

the surface

is

convex, the indicatrix

is

an

ellipse

and the

asymptotic directions are imaginary.

In special cases the dyadic


cients a and b are equal.

may be such that the coeffimay then be reduced to the

form

a(i'i'+j'j')

(5)'

THE CURVATURE OF SURFACES


in an infinite

number

The

of ways.

any two perpendicular

directions.

415

and j' may be


The indicatrix becomes a

directions

i'

perpendicular diameters of this circle


Such a point is
give principal directions upon the surface.
The surface in the neighborhood of an
called an umbilic.
circle.

Any

pair of

umbilic is convex. The asymptotic directions are imaginary.


becomes linear and reduIn another special case the dyadic
cible to the

The
to

i'.

The

form

&=

(5)"

i'i'.

indicatrix consists of a pair of parallel lines perpendicular


Such a point is called a parabolic point of the surface.

further discussion of these

and other

special cases will be

omitted.

The quadric surfaces afford examples of the various kinds


The ellipsoid and the hyperboloid of two sheets
The indicatrix of points upon them is an ellipse.
are convex.
The hyperboloid of one sheet is concavo-convex. The inThe indicatrix
dicatrix of points upon it is an hyperbola.
The points are all
of any point upon a sphere is a circle.
umbilics. The indicatrix of any point upon a cone or cylinder
The points are parabolic. A suris a pair of parallel lines.
of points.

face in general may have upon it points of all types


elliptic,
hyperbolic, parabolic, and umbilical.
line of principal curvature upon a surface is a
152.]

curve which has

at each point the direction of

cipal axes of the indicatrix.

The

one of the prin-

direction of the curve at a

point is always one of the principal directions on the surface at


that point. Through any given point upon a surface two perpendicular lines of principal curvature pass. Thus the lines
of curvature divide the surface into a system of infinitesi-

mal rectangles. An asymptotic line upon a surface is a curve


which has at each point the direction of the asymptotes of the
indicatrix.

The

direction of the curve at a point is always

one of the asymptotic directions upon the surface.

Through

VECTOR ANALYSIS

416

any given point of a surface two asymptotic


lines are

the surface

if

is

lines pass.

These

Even when real


The angle
angles.

convex.

imaginary
they do not in general intersect at right

between the two asymptotic lines at any point is bisected by


the lines of curvature which pass through that point.

The necessary and

curve upon a
that as one advances

sufficient condition that a

surface be a line of principal curvature

is

along that curve, the increment of d n, the unit normal to the


surface

is parallel to

dn=

the line of advance.

= (a
+I
dr = x
+ yj'.

& dr

i'i'

For

j'j')

dr

i'

Then evidently d n and d r are parallel when and only when


dt is parallel to i' or j 7 The statement is therefore proved.
.

It is frequently

The

taken as the definition of lines of curvature.

differential equation of a line of curvature is

dnxdr = Q.
Another method of statement

is

(6)

that the normal to the surface,

the increment d n of the normal, and the element d r of the


surface
is

lie

in one plane

an element of a

when and only when

line of principal curvature.

the element d r

The

differential

equation then becomes

dn dr]=0.

[n

(7)

The necessary and sufficient condition that a curve upon a


surface be an asymptotic line, is that as one advances along
that curve the increment of the unit normal to the surface

perpendicular to the line of advance.

= dr
dn dt = dr

is

For

dn

If then

dn

r is zero

asymptotic direction.

dr.

r is zero.

The statement

is

Hence d r

is

therefore proved.

an
It

THE CURVATURE OF SURFACES


is

417

The

frequently taken as the definition of asymptotic lines.

differential equation of

an asymptotic

line is

dvdr = Q.
153.]

Let

(8)

be a given point upon a surface and n the


Pass a plane p through n. This

normal to the surface at P.

plane p is normal to the surface and cuts out a plane section.


Consider the curvature of this plane section at the point P.
Let n' be normal to the plane section in the plane of the
section,

But

with n at the point P.

n' coincides

unless the

p cuts the surface everywhere orthogonally, the normal


the plane section and the normal n to the surface will not
d n and d n' will also be different. The curvature
coincide,
plane

n' to

of the plane section lying in

is

ds

As

(Art. 57).

far as numerical value is concerned the increment of the

and the increment of the unit normal n' are


Moreover, the quotient of d r by d s is a unit vector

unit tangent

equal.
in the direction of

Consequently the scalar value of C

d n'.

dr _dn' dr
ds 2
ds

dn'

ds

By

hypothesis

= n' at P

and

dr

= n'

dr

d(ndr~)=dn'dr + n'd z r =
d (V d r) = d n' d r + n' d 2 r = 0.
(),

Hence

dn

Since n and

n'

dr

d2

= d n'

n'

d 2 r.

are equal at P,

dn dr =
Hence

dr

dn
- dr

dr

4>

dn'

dr

-^527

dr.

dr

<P

dr

rfr .rf r

0.;

is

VECTOR ANALYSIS

418

(J

=. n
d

dr

-f-

dr

dr

dr

Hence

C=

a cos 2

(i',

d r)

& cos 2

(j',

or

C=

a cos 2

(i',

dr)

J sin 2

(i',

d r),
df).

(10)

The

interpretation of this formula for the curvature of a


normal section is as follows : When the plane p turns about

the normal to the surface from

plane section varies

i'

to

j',

the curvature

of the

from the value a when the plane passes

through the principal direction

i',

to

the value &

when

it

The values
passes through the other principal direction j '.
of the curvature have algebraically a maximum and
njinimum
in the directions of the principal lines of curvature.
6

have unlike

signs, that

at P, there exist

two

directions for

normal section vanishes.


154.]

The sum

the surface

is, if

is

If a

and

concavo-convex

which the curvature of a

These are the asymptotic directions.


two normal sections

of the curvatures in

at right angles to one another

is

constant and independent of


For the curvature in

the actual position of those sections.

one section

is

and

=a

cos 2

(i',

c?r)

5 sin 2

(i',

dr),

in the section at right angles to this

Cz =
Hence

a sin 2

C1

dr)

(i',

& cos 2

(i',

+C = a + b=0 s

dr).

(11)

which proves the statement.


It is easy to

show

that the invariant

duct of the curvatures a and

2S

is

equal to the pro-

b of the lines of principal curv-

ature.
<P

Hence the equation

xz

zs
4> s

=ab
x

+ &zs =

(12)

THE CURVATURE OF SURFACES

419

the quadratic equation which determines the principal curvatures a and b at any point of the surface.
By means of this
is

equation the scalar quantities a and b

may

be found in terms

oiF(x,y,z).

(I-nn) VVJF. (I-nn)

VV^-2_
(nn

VV F

Hence

nn) 5
a

N
nn

(nn

VV^ ) 5 =
T

(nn

VV^) 5 = nn: VV^=n. VVJ?

(nn

Hence

= V.V.P

^-

V..
-

n.

(13)

These expressions may be written out in Cartesian coordinates,


but they are extremely long. The Cartesian expressions for
ZS are even longer.

<P

The

vector expression

may

be obtained

as follows:

(I-nn) 2 .(VVJQ 2 .(I-nn) 2


(I

nn) 2

= nn.

Hence

Given any curve upon a surface. Let t be a unit


tangent to the curve, n a unit normal to the surface and m a
155.]

VECTOR ANALYSIS

420

X t. The three vectors n, t, m constitute


The vector t is parallel to the element d r.

vector defined as n

i, j, k system.
Hence the condition

an

for a line of curvature

Hence
d (m

n)

x d n

becomes

= 0.

(15)

m d n=
= = m dn +

Hence

dm

= 0.

Moreover

dm

= 0.

Hence

x dm

= 0,

dm.

(16)

dmxdn = Q.

or

The increments

of

m and of n

(16)'

and of r are

a line of principal curvature.


geodetic line upon a surface

is

all parallel in

case of

a curve whose osculating


That the

plane at each point is perpendicular to the surface.

is the shortest line which can be drawn between


two points upon a surface may be seen from the following
Let the surface be smooth and
considerations of mechanics.

geodetic line

a smooth elastic string which is constrained to lie in the


surface be stretched between any two points of it. The string

let

acting under

its

own

tensions will take a position of equili-

brium along the shortest curve which can be drawn upon the
surface between the two given points.
Inasmuch as the
string is at rest

surface

must

lie

upon the surface the normal

reactions of the

in the osculating plane of the curve.

Hence

that plane is normal to the surface at every point of the curve


and the curve itself is a geodetic line.

The

vectors t and

mine that plane.

dt

lie

in the osculating plane

In case the curve

is

normal

n.

Hence

deter-

a geodetic, the normal

to the osculating plane lies in the surface

perpendicular to the

and

and consequently

is

THE CURVATURE OF SURFACES


n

n X

or

The

= 0,
d t=
t = 0.

X d

421

(17)

differential equation of a geodetic line is therefore

[n d r d* r]

= 0.

(18)

Unlike the differential equations of the lines of curvature


line, this equation is of the second order.

and the asymptotic

The

surface

is

therefore covered over with a doubly infinite

Through any two

system of geodetics.

points of the surface

one geodetic may be drawn.


As one advances along any curve upon a surface there is
necessarily some turning up and down, that is, around the
axis m, due to the fact that the surface is curved.

There may

to the right or left. If one advances


a
such
is no turning to the right or
curve
that
there
along
the
unavoidable
but
left,
only
turning up and down, it is to be

or

may not be any turning

expected that the advance

is along the shortest possible route


a
Such is in fact the case. The
along
geodetic.
Since n,
total amount of deviation from a straight line is d t.

that

t,

is,

m form

an

i, j,

k system
I

dt

Hence

= tt + nn + mm.

= ttdt + nndt + mmdt.

first term vanishes.


The second
and
of
amount
term represents the
down; the
turning up
Hence m d t is
third term, the amount to the right or left.

Since

t is

a unit vector the

the proper measure of this part of the deviation from a


In case the curve is a geodetic this term
straightest line.

vanishes as was expected.


curve or surface
156.]

may

be

mapped upon a unit

fixed origin is
sphere by the method of parallel normals.
of a
assumed, from which the unit normal n at the point

VECTOR ANALYSIS

422
given surface

upon

The terminus P'

is laid off.

If the

the surface of a sphere.

of this

normal

lies

normals to a surface at all

of a curve are thus constructed from the same origin,


points
the points P' will trace a curve upon the surface of a unit
This curve is called the spherical image of the given
sphere.

curve.

In like manner a whole region

T of

the surface

may

mapped upon a region T' the sphere. The region T' upon
the sphere has been called the hodogram of the region T upon
be

the surface.

If

d r be an element

of arc

upon the surface the

corresponding element upon the unit sphere

dn

is

dr.

be an element of area upon the surface, the corresponding element upon the sphere is da! where (Art. 124).
If da.

da!

=a
2

= ab

Hence

The
its

ratio of

hodogram

curvature at

i'

da!

(Pi* da..

i'i'

j'

= ab

J j'j'

i'xj'

nn

= ab

nn.

da..

an element of surface at a point


is

(19)

P to

the area of

equal to the product of the principal radii of


the reciprocal of the product of the prin-

P or to

cipal curvatures at P.
It
is

total

was seen that the measure

of turning to the right or left

d t. If then C is any curve drawn upon a surface the


amount of turning in advancing along the curve is the

integral.

(20)

For any closed curve this integral may be evaluated in a


manner analogous to that employed (page 190) in the proof
of Stokes's theorem.

Consider two curves

and

C' near

THE CURVATURE OF SURFACES


The

together.

which the integral undergoes when


integration is changed from C to C' is

the curve of

variation

Cm>dt= Cs

d (m
B

d t.

dt+ Cm.Bdt

(m. dt~)= C Sm
St)

= dm*

St

+m

Cm- dt= C Bm*dt- C dm

d Bt

Bt

+ C d (m

The integral of the perfect differential c (m


when taken around a closed curve. Hence
B

m^t=

The idemfactor

d t and

for t

423

is

Sm

dt

5t).

8 t) vanishes

dm

Sk

I^tt + nn

m m vanish. A similar transformation may


m St. Then

be effected upon the term d


8

By

m dt= /(Smn

differentiating the relations

dm

dt

n8t).

and n

it is

seen that

Smn=:
dm n =
Hence

m Bn

Bt

8nt

m dn

dt

dn

\m*dt=\ (mSn t>dn

Cm dt=

|(mxtSnxdn)=

t.

m dn tSn)
I

Bnx

dn.

VECTOR ANALYSIS

424

The
the

differential

8n x dn

hodogram upon
/

represents the element of area in


the unit sphere. The integral

n x dn

da!

the total area of the hodogram of the strip of


and C'. Let the
surface which lies between the curves
represents

curve

C start at

a point

surface and spread out to

upon the

The total amount of turning which is reany desired size.


quired in making an infinitesimal circuit about the point is
2 TT. The total variation hi the integral is
I

But

if

m dt=

dt

TT.

H denote the total area of the hodogram.

Hence

dt

H= 2

TT

or

?r

- Cm

H+ Cm

or

dt

dt,

(22)

= 2ir.

The area of the hodogram of the region enclosed by any


closed curve plus the total amount of turning along that curve
is equal to 2 ?r.
If the surface in question is convex the area
will appear positive when the curve upon the
so described that the enclosed area appears positive.

upon the sphere


surface
If,

is

however, the surface

is

concavo-convex the area upon the


This matter of the sign of the

sphere will appear negative.

hodogram must be taken into account in the statement made


above.

THE CURVATURE OF SURFACES


157.]

If the

closed curve

geodetic lines the

The

amount

is

425

a polygon whose sides are

of turning along each side is zero.

total turning is therefore equal to the

sum

of the exterior

The statement becomes : the sum of


angles of the polygon.
the exterior angles of a geodetic polygon and of the area of
the hodogram of that polygon (taking account of sign) is
equal to 2 TT. Suppose that the polygon reduces to a triangle.
If the surface is convex the area of the hodogram is positive

and the sum of the exterior angles of the triangle


2 TT.
TT.

is less

than

The sum of the interior angles is therefore greater than


The sphere or ellipsoid is an example of such a surface.

If the surface is

negative.

concavo-convex the area of the hodogram

The sum

this case less

than

of the ulterior angles of a triangle

TT.

Such a surface

sheet or the pseudosphere.

There

is

is

is

is

in

the hyperboloid of one

an intermediate case in

which the hodogram of any geodetic triangle is traced twice in


opposite directions and hence the total area is zero. The sum
of the interior angles of a triangle upon such a surface is equal
to TT.
Examples of this surface are afforded by the cylinder,

cone, and plane.


surface is said to be developed

lines

upon the surface

when

it is

retain their length.

so deformed that

Geodetics remain

One surface is said to be developable or applicable


geodetics.
when it can be so deformed as to coincide with
another
upon
Geodetics
the other without altering the lengths of lines.
the
other.
into
are
surface
one
upon
geodetics
changed
upon

The sum

of the angles of

any geodetic triangle remain un-

changed by the process of developing.


that the total
of the

amount

hodogram

of

of turning along

any

From

this it follows

any curve or the -area

portion of a surface are also invariant

of the process of developing.

VECTOR ANALYSIS

426

Harmonic Vibrations and

The

158.]

motion

Bivectors

equation of rectilinear harmonic

differential

is

dt*

The

may be reduced by

integral of this equation

a suitable

choice of the constants to the form

=A

sin

t.

This represents a vibration back and forth along the JT-axis


about the point x = 0. Let the displacement be denoted by
D in place of x. The equation may be written

D
Consider

=i A

=i A

sin

sin

cos

t.

m x.

a displacement not merely near the point x


2 k TT
but along the entire axis of x. At points x
where
,

This

is

is a positive or negative integer, the displacement is at all


times equal to zero. The equation represents a stationary
Te

wave with nodes at these points. At points midway between


If the
these the wave has points of maximum vibration.
equation be regarded as in three variables
sents a plane

wave the plane

of

which

is

#, y, z it

repre-

perpendicular to

the axis of the variable x.

The displacement given by


Dj

=i A

the equation

cos

(m x

f)

(1)

likewise a plane wave perpendicular to the axis of x but


not stationary. The vibration is harmonic and advances
along the direction i with a velocity equal to the quotient of
is

HARMONIC VIBRATIONS AND BIVECTORS


n by m.

If v

be the velocity;

the period; and

length,

n
= -,
m

2?r

= 2 TT
m

the

427

wave

= -.
p

(2)

The displacement

A z cos(mx

D 2 =j
differs

nt)

from DJ in the particular that the displacement takes

place in the direction j, not in the direction i. The wave as


before proceeds in the direction of x with the same velocity.
This vibration is transverse instead of longitudinal. By a
simple extension it is seen that

= A cos

(m x

f)

a displacement in the direction A. The wave advances


along the direction of x. Hence the vibration is oblique to
is

the wave-front.

A still

by substituting m

r for

more general form may be obtained

m x.

= A cos

Then
r

(m

(3)

t).

This is a displacement in the direction A. The maximum


amount of that displacement is the magnitude of A. The
wave advances in the direction m oblique to the displacement; the velocity, period, and wave-length are as before.

So much for

rectilinear

harmonic motion.

monic motion may be defined by the equation


o

Elliptic har(p. 117).

wr.

dt*

The general

integral
r

The

is

obtained as

= A cos n + B sin n
t

t.

discussion of waves

viously.

may be carried through as preThe general wave of elliptic harmonic motion

advancing in the direction

is

seen to be

VECTOR ANALYSIS

428

= A cos

dV =

7i

dt

(m
sin

(m

f)

t)

sin

(m

+B

sin

(m

(4)

t).

f)

(5)

the velocity of the displaced point at any moment in the


This is of course entirely differellipse in which it vibrates.
is

ent from the velocity of the wave.


An interesting result is obtained by subtracting the displacement from the velocity multiplied by the imaginary
unit

and divided by

n.

D+V

The

A sin (m

t)

+B

expression here obtained, as far as

its

sin

(m

form

is

t~)

}.

concerned,

an imaginary vector. It is the sum of two real vectors of


1which one has been multiplied by the imaginary scalar
Such a vector is called a bivector or imaginary vector. The
is

ordinary imaginary scalars may be called liscalars. The use


of bivectors is found very convenient in the discussion of

harmonic motion. Indeed any undamped elliptic harmonic plane wave may be represented as above by the product of a bivector and an exponential factor.
The real part
of the product gives the displacement of any point and the
elliptic

imaginary part gives the velocity of


reversed in sign and divided by n.
pure

displacement

analytic theory of bivectors differs from that of


very much as the analytic theory of biscalars
differs from that of real scalars.
It is unnecessary to have
159.]

The

real vectors

any distinguishing character

for bivectors just as it is need-

HARMONIC VIBRATIONS AND BIVECTORS


have a distinguishing notation for

less to

biscalars.

429

The

bi-

be regarded as a natural and inevitable extension


It is the formal sum of two real vectors
of the real vector.
vector

of

may

which one has been multiplied by the imaginary unit

The usual symbol i will be maintained for V 1. There is


not much likelihood of confusion with the vector i for the
reason that the two could hardly be used in the same place
and for the further reason that the Italic i and the Clarendon
i

Whenever

differ considerably in appearance.

it

becomes

especially convenient to have a separate alphabet for bivectors the small

bivector

Greek or German

may

letters

may be

be expressed in terms of

called upon.

i, j,

k with com-

plex coefficients.
If

and

= Oi +

rl

=x

r2

= # 2 * + 2/2 J + *

2)

or

Two

= TJ + * r a

i -f

(Vi

yl

* y-i)

= #i + yj +

bivectors are equal

when

z^
2

>

+ Oi +

*a)

>

2k.

their real

and

their imaginary

is the
Two bivectors are parallel
parts are equal.
product of the other by a scalar (real or imaginary). If
a bivector is parallel to a real vector it is said to have a real

when one

direction.

In other cases

it

has a complex or imaginary

The value of the sum, difference, direct, skew,


and indeterminate products of two bivectors is obvious without special definition. These statements may be put into
analytic form as follows.

direction.

Let

Then
if

r
if

= TJ +
r =
s,

r||s

= s2 + i sa
r =
and r 2 = s 2
r = a;s = (x + i x^)
and

r2

B,

VECTOR ANALYSIS

430
r

(r x

(T!

= (r

r
r

rs

Two

B!

(r 2

(rj BJ

r 2 Ba)

s )
2

s ),
a

a)

r2

(TJ

+i

<>!

(T I

s2

r2

gj),

r2

and

when

their pure imaginary parts

The conjugate

differ only in sign.

equal to the scalar or vector

of a real scalar or vector

The conjugate

itself.

and

sort of product of bivectors

Sj)

r 2 BJ).

bivectors or biscalars are said to be conjugate

their real parts are equal

is

- ra

BJ)

biscalars

is

of

any

equal to the pro-

duct of the conjugates taken in the same order.


similar
statement may be made concerning sums and differences.

Oi +
(TJ

Oi +

i r )
a

r2> ( r i

- i ra ) = TJ

<>!

i ra )

(r x

~*r)=

or cyclic factor as

= TJ +

ra

- i r ) = 2 i r2

rx

(ri ri

If the bivector r

i r
a

ra r2 )

rx ,

ri

-r

r2>-

be multiplied by a root of unity

frequently called, that

it is

ra

is,

by an imagi-

nary scalar of the form


cos q

where

the conjugate

i
2

= a + i J,
+ & 2 = 1,
sin q

multiplied by a

is

i J,

(7)

and hence the four

products
ri

'

ri

*2

are unaltered

Thus

r2>

*2

X *H

T l r2

r2 r l

T2 r2

by multiplying the bivector

r i'

= r/ +
'

'i'

i r
a

rz

'

'

'a'

(a

= TJ

5) (r x

rj

ra

by such a

if
r'

~T

i r a ),

ra , etc.

f2
factor.

HARMONIC VIBRATIONS AND BIVECTORS

160.]

431

closer examination of the effect of


multiplying a

bivector by a cyclic factor yields interesting and important

Let

geometric results.

r/

i r
2

Then

'

r/
ra'

By

real

=r

cos 1

it

sin q) (TJ

r x cos q

reference to Art. 129

duced in the

(cos q

will be seen that the

is

b and

by
same as

cyclic dyadic

sin q (c

c c')

change pro-

of a bivector

precisely the

would be produced upon those vectors by a


cos q (bb'

(8)

r i si*1 ?

and imaginary vector parts

used as a prefactor.

r2 ).

r2 sin #,

multiplication with a cyclic factor,

= a a' +

be')

supposed to be two vectors


and r2 a is any vector not in

c are

collinear respectively with r x

Consider the ellipse of which r x and r2 are a


of
pair
conjugate semi-diameters. It then appears that r^
and r2 are also a pair of conjugate semi-diameters of that

their plane.

'

ellipse.

are rotated in the ellipse from T I toward r2 by


is to the area of the whole ellipse

They

a sector of which the area

TT.
Such a change of position has been called an
rotation
elliptic
through the sector q.
The ellipse of which T I and r2 are a pair of conjugate semi-

as q

is

to 2

diameters

is

called the

directional

ellipse

of the bivector

r.

When

the bivector has a real direction the directional ellipse


reduces to a right line in that direction. When the bivector

has a complex direction the ellipse

is

a true ellipse.

The

angular direction from the real part T I to the complex part r2


is considered as the
positive direction in the directional
ellipse,

and must always be known.

If the real

and imagi-

nary parts of a bivector turn in the positive direction in the


ellipse they are said to be advanced ; if in the negative direction they are said to be retarded.

Hence

multiplication of a

VECTOR ANALYSIS

432
bivector by

cyclic

factor retards

sector equal to the angle

directional ellipse by

its

always possible to multiply a bivector by such a cyclic

It is

factor that the real

with the axes of the


r

To

in

it

of the cyclic factor.

and imaginary parts become coincident


ellipse and are perpendicular.

= (cos q + i sin

q) (a

b) where a

accomplish the reduction proceed as follows


r

If a

(cos 2 q

Form

(a

b)

i b).

= 0,
r

(cos 2 q

Let

and

With

sin 2 q) (a

= 0.

sin 2 q) (a

=a+

tan 2 q

b).

i b,

-.

this value of q the axes of the directional ellipse are

given by the equation


a

(cos q

sin q)

r.

In case the real and imaginary parts a and b of a bivector


magnitude and perpendicular in direction both a

are equal in

and b in the expression for r


q

is

The

indeterminate.

r vanish.

bivector whose directional ellipse


lar bivector.

The necessary and

is

If
r

The condition
is

= x2 +

0,

is

is

a circle.

called a circu-

is

r circular.

0,
i

a circle

sufficient condition that

non-vanishing bivector r be circular


r

Hence the angle

directional ellipse

2/j

+
+

which for

2k,
z2

= 0.

real vectors implies r

not sufficient to ensure the vanishing of a bivector.

= 0,
The

HARMONIC VIBRATIONS AND BIVECTORS


bivector

a bivector vanish

The

not necessarily zero.

is circular,

condition that

that the direct product of

is

433

it

its

by

con-

jugate vanishes.

Ol +

r2>

'

then

0*1

TI

- * r2> = r

= r2

rl

'

and

conjugate and
161.]

their product

The condition

f2

>

becomes equal

it

becomes equal

two

that

'

= 0.

In case the bivector has a real direction


its

r2

to r

to

r.

bivectors be parallel

is

that

the product of the other by a scalar factor. Any biscalar factor may be expressed as the product of a cyclic
factor and a positive scalar, the modulus of the biscalar.
If

one

is

two bivectors

by only a cyclic factor their directional


Hence two parallel vectors have their
similar and similarly placed
the ratio of

differ

ellipses are the same.

directional ellipse

similitude being the modulus of the biscalar.


It is evident
that any two circular bivectors whose planes coincide are
parallel.

circular vector

and a non-circular vector cannot

be parallel.
The condition that two bivectors be perpendicular
r

is

or

Consider
coincide.

first

s2

=r

0,

s2

r2

BJ

= 0.

the case in which the planes of the bivectors

Let
r

The

r2

sx

TJ

scalars a

and

dicular to r2 , and

condition r

(T!

i r2 ),

=b

(s 1

i s ).
2

r 1 may be chosen perpenbe taken in the direction of ra


The

I are biscalars.

BI

s =

r2

may

then gives
s2

and

TJ

28

s2

+. r2

BJ

= 0.

VECTOR ANALYSIS

434

The

equation shows that r2 and

first

are perpendicular and

s2

B I and s 2 are perpendicular.


Moreover, the second
shows that the angular directions from i^ to r2 and from s l to
s 2 are the same, and that the axes of the directional ellipses

hence

and s are proportional.


Hence the conditions for perpendicularity of two bivectors
whose planes coincide are that their directional ellipses are
similar, the angular direction in both is the same, and the
of r

1
If both vectors
major axes of the ellipses are perpendicular.
have real directions the conditions degenerate into the per-

pendicularity of those directions. The conditions therefore


hold for real as well as for imaginary vectors.

be two perpendicular bivectors the planes of


which do not coincide. Resolve r x and r2 each into two com-

Let

and

ponents respectively parallel and perpendicular to the plane

The components perpendicular to that plane contribute


nothing to the value of r s. Hence the components of r x
of

s.

and

r2 parallel to the plane of s

perpendicular to

s.

form a bivector

To this bivector and


The directional ellipse

r'

which

is

the conditions

of the bivector r'


stated above apply.
is evidently the projection of the directional ellipse of r upon

the plane of

two bivectors are perpendicular the directional


of either bivector and the directional ellipse of the

Hence,
ellipse

s.

if

other projected upon the plane of that one are similar, have
the same angular direction, and have their major axes perpendicular.
162.]

Consider a bivector of the type

D
where

A and m

= Ae, i(m

are bivectors

'

- n()

and n

position vector of a point in space.

'

(9)
is

a biscalar.

r is the

It is therefore to be con-

1 It should be noted
that the condition of perpendicularity of major axes is not
the same as the condition of perpendicularity of real parts and imaginary parts.

HARMONIC VIBRATIONS AND BIVECTORS


sidered as real,

be considered as

is

real.

the scalar variable time and

Let

is

435
also to

VECTOR ANALYSIS

430

same period and wave-length

as the individual waves.

vibrations, however, take place in a different ellipse.

The
If the

waves are

the resultant

is

A+

B)

(m

'

~"

trains of waves which advance in opposite


which
are in other respects equal a system of
directions but
stationary waves is obtained.

By combining two

Ae- m*- r

e-

<n '

<mi

(e

The theory of

+ e -' mi-

bivectors

The

carried further.

and

= 2Acos

mt

(n^

the reader

how

nt)

r)

__

e~^' T e~ int

their applications will not be

object in entering at all

short and condensed discussion of bivectors

to the

upon this very


was first to show

the simple idea of a direction has to give way


less useful idea of a directional

more complicated but no

when

the generalization from real to imaginary vectors


made, and second to set forth the manner in which a single

ellipse
is

bivector

D may

waves of

elliptic

vectors

be employed to represent a train of plane


harmonic vibrations. This application of bi-

may be used to give

the

Theory

of Light a wonderfully

1
simple and elegant treatment.

"

1
Such use of bivectors is made by Professor Gibbs in his course of lectures on
The Electromagnetic Theory of Light," delivered biannually at Yale University.

Bivectors were not used in the second part of this chapter, hecause in the opinion
of the present author they possess no essential advantage over real vectors until
the

more advanced parts

magnets and

of the theory, rotation of the plane of polarization

crystals, total

and metallic

reflection, etc., are reached,

by

University of California

SOUTHERN REGIONAL LIBRARY FACILITY


405 Hilgard Avenue, Los Angeles, CA 90024-1388
Return this material to the library
from which it was borrowed.

LII7

inr

9?

^**

OA
A

000 361 200

72

BKAKY6#.

R)

L1C