Professional Documents
Culture Documents
DO
164413
73
pall
No.
Author
Title
last
marked below.
PLANETARY THEORY
LONDON
Cambridge University Press
FETTER LANE
TOKYO
Maruzen Company Ltd
A II
rights
resewd
PLANETARY THEORY
BY
ERNEST
W.
I'/D/esxor Kniciitns of
Yale Unit
Mathematics
in
ersitif
AND
CLARENCE
Ati^i^tant I'tofensot'
Lcli i</h
A. **
SHOOK
~._
of Mathematics
in
I r itit'<T.'iiltf
CAMBRIDGE
AT THE UNIVERSITY PRESS
1933
IN
GREAT BR11AIN
CONTENTS
.........
Preface
Chapter
I.
EQUATIONS OF MOTION
page
General introduction
Frames of reference
ix
11
15
22
plane
28
30
31
planet
Motion referred
Chapter
II.
to
...
....
....
.......
34
and cosine
36
37
37
40
45
45
powers of the
.....
...
......
....
47
49
53
Bessel's functions
series
Hypergeometric
Numerical calculation of
series
series
54
55
56
59
CONTENTS
vi
.....
....
page 62
62
65
66
68
71
72
79
80
...
Chapter IV.
81
TION
82
cities
Expansion
........
Expansion
Development
....
.
Development
in
The
Chapter V. CANONICAL
94
95
97
99
100
102
112
114
117
function
118
Contact transformations
........
and
its solu-
The
93
91
87
89
121
1
27
131
136
CONTENTS
vii
page 138
138
143
146
152
156
First approximation
161
Summary and
169
special cases
....
Chapter VII.
174
174
17,8
The
first
approximation
185
194
198
207
able
RESONANCE
Chapter VIII.
216
Elementary theory
Solution of a resonance equation
General case of resonance in the perturbation
.
problem
A general
The
The
The
method
case e'=
1
T=
2 case
cases
e'^0
for treating
210
213
resonance cases
219
222
226
234
238
241
246
CONTENTS
viii
Chapter IX.
The
.....
.......
triangular solutions
General theory
Elimination of the short period terms
Higher approximations
Perturbations by Saturn
Indirect perturbations
Direct action of Saturn
.....
.....
.
page 250
.........
250
256
259
261
269
273
277
280
283
286
289
801
PREFACE
THE
is
If an
for
is
to
be
be
PREFACE
x
is
still
if
complete justification
The use
of formal
results,
justified
processes
otherwise obtainable, are useful for the prediction of physical
is
riot
We
they had not been given here, they would have been printed
In particular is,
contains
PREFACE
xi
shown how
this theory
may be
tion of the orbit of a planet the basis of the method is the use
of the transformation to eliminate the short-period terms as
;
first step,
with separately.
In Chapter vii, the direct calculation of the coordinates with
the use of the true orbital longitude as the independent variable
developed with sufficient detail for the formation of an approxi-
is
mate or of an accurate
method of approach
been the
case.
will
be found to contain
formulae for
its
be analysed is great.
We are indebted to Dr Dirk Brou wer and
of such functions
errata given on p. xii and discovered after the sheets had been
printed.
r
1933
ERNEST W. BROWN
CLARENCE A. SHOOK
Errata
p. 26, line 5 from bottom, insert factor
*
r2
before
11
p. 28, line 6
p. 28,
Equation
Equation
/or
(5),
v
p. 28,
Equation
p. 30,
omit lines
p. 63,
Equation
2tf 2
-
read
-T
..
?r
for
(7),
du
ou
for $q read
(4),
2
p. 28,
read uu'2 _-
^-oT
read
l\
12, 13.
(10),
read andt.
/or rtd
for
2
47r'
fc
read
/x,
/x/47r-.
p. 76,
Equation
for # = exp
p. 87, line 7,
p. 117,
p.
/or
Equation
a-.
^ read x-~exp
for
mid
(5)
,._ r
p. 157, line
from
(1),
-_-
/or
(8).
for
$S
mid,
--^-
read
(3)
is".
- 3S read -32/j.
Dv = l
read
Dv = l.
p. 230, line 5,
1.
(6).
p. 145, line 2,
for
differing
*J
read m.
\f/
^
(2),
p. 129, line 6
-j
Equation
read
{-)
read
a~
p. 87,
for
(J),
(10).
aB _
db
3 >/3 m' - 8a 3 /t 2
I
read
r*
should read
(y. m)*
_
"^
>/3~
12
,I'
S, slightly
CHAPTER
EQUATIONS OF MOTION
A.
GENERAL INTRODUCTION
in this volume for the investigamutual actions of two or more bodies are based wholly
on Newton's three laws of motion and on his law of gravitation.
It is assumed that there exist fundamental frames of reference
with respect to which the laws are exact and that the space in
which the bodies move is Euclidean. The modern theory of reI'l.
tion of the
corrections,
1*2.
assumed
EQUATIONS OF MOTION
[OH.
satellite or stellar
problem.
third division
exist.
the latter.
tire
inclined to
one another at small angles generally less than 20. The distances of the bodies from one another may have any values whatever provided they do not fall below a certain limit. In general,
the methods of this volume are developed for this case alone.
of the bodies
another and their centre of mass circulates round the third body.
The maintenance
relation
are also
In the
and other
cases.
give expressions for the coordinates in terms of the time which serve to give the positions
of the bodies over long intervals of time: the results are usually
the general perturbations. Practically all other cases have
at present to be treated by the method of special perturbations
named
INTRODUCTION
2, 3]
the
first
As
bodies.
sums
of sines
and cosines, thus play a large part in the work. The possibility
of obtaining these expansions in such forms that numerical
results
may
much
labour,
EQUATIONS OF MOTION
[CH. i
usually
For the majority of cases, this last ratio lies between *4 and
*8, and the feet that we are compelled to expand in powers of so
large a parameter
is
responsible for
many
of the difficulties of
the problems.
1'4.
difficulty,
however,
is
caused by
changed.
In either event, the terms which cause the chief trouble are
those with periods which are long in comparison with the period
of revolution of the body round the central mass. Such periodic
terms
may have
affected be carried to a
much
ASTRONOMICAL MEASUREMENTS
3-5]
work required
first
approximation
may be
approximation.
celestial sphere.
is
needed,
and
this
is
when a plane
Owing
to the rapid
its
axis the
in
on
it
The motions
EQUATIONS OF MOTION
[CH. i
ecliptic.
many cases two or more interpretations are possible and these can only
be separated by the use of more observations. An outstanding difference,
for example, between the observed and calculated values of the motion of
In
1 *6. Observations of bodies in the solar system are usually of two classes.
Those made with the transit telescope give the instant of passage of the
body across the meridian of the observer and the angle between its observed
direction and that of the earth's axis, the time being given by a clock which
is
often
such a standard
should give the position with a probable error less than
at least is aimed at in the gravitational theories of the principal bodies in
the solar system.
While
is
necessary because
LAW OF GRAVITATION
6-7]
them from the equations of motion without introducing complications greater than those which the equations already possess. They are
the
also needed because the planetary theories use the sun as an origin
eliminate
known
to us
with the effect of one planet on another, we need to know the ratio of the
mass of the disturbing planet to that of the sun. If the planet has a
satellite whose motion can be calculated, the mass of the planet can be
found with sufficient accuracy to calculate its effect on any other planet.
When it has no satellite, its mass can be found only by comparing its calculated disturbing effects with those furnished by observation. There is of
course a correlation between the degree of accuracy required to calculate
the perturbations and that with which the perturbations can be observed.
Thus the masses of all the major planets except those of Venus and Mercury
are fairly well determined. The masses of the minor planets can be obtained only from observation of the light they send with estimates as to
their albedo and density, since they are too small to exert
attractive effect on any other body.
any observable
distance r apart
a mass with these units has the apparent dimensions of the cube
of a length and the inverse square of a time. The adoption of
called the 'astronomical unit of mass'
is closely
a slight
with
This
law.
associated with Kepler's third
law,
law
of
modification (3*6) to bring it into accord with the
gravitation, states that if ZTT/H be the period of revolution of two
this unit
if
= C x sum
of masses.
EQUATIONS OF MOTION
[CH. i
will consist of
units of time,
B,
components of the
forces.
For two
m,
in
and at a distance r
+1
apart,
bodies with
ij
may be
used.
It
may
or
may
not depend on
not the same as those for the other particles of the system.
PLANETARY FORM
7-9]
tion of
d^i
m i~Ta^>
m
at*
d*m
J72~
at*
dgi
= SV
T~
'
dm
d2
BF
at 2
o%i
m i~J^^ = ^^'
Only the
defini-
in 1*1,
-(*)
needed and,
in
1*9.
is
mQ
are to be determined.
j/fc
These
!, 2,
...w,
^)
definitions, together
dt 2
__
/p
_
""8.'-*U
where J
of
=(=
^-
>,J
v
_ '^fc _ ~
>*3
The
last
;
;
m k jr k with respect
of mo/r k
v-
'
to the derivative
in
which form
Hence,
if
it
we put
for
mk
relative to
yk zk
,
mQ
become
EQUATIONS OF MOTION
12
so that the
new
coordinates
[OH. I
still
Now when 7 is
we
&
But
since
V contains
the
dV
dV=
(rfft
j|
<7i
But when
V is;to
37
dV
- rf>) +
^ (d& -
we have
d$>).
O^2
be expressed in terms of
TT-
ox
If
&,
only,
Whence, combining
TH,
;,
7
dx
+
=-,
dx
#,
,7;',
we have
,
.
ox
we
d^z
expressions for dV,
we obtain
ar
'
3 A''
...... (4)
x become, with
#,
^r,
dV
'
ftV
'
.
'
~i
dV\
^~~
'
(5)
SATELLITE FORM
12, 13]
n>l
2
with
?*i 2
m^mi
y7
we
calculation,
13
= .1? +
7*02
?/
2
/
(,'i/H
nil
,] +...
itiz
...,
giving
d2 x
?)
(?
77i
/M
//?
2)
~
///h
I
\r
wa
H---
r02
in
tQi~-
Kl
= ^yU,
1/r
K,
/ox
...... (8)
^)
of
which
motion
The equations of motion in the form 1*12 (6), (7) will not be needed
developments of this volume since they are, in general, useful only
when the distance between mi m^ is small compared with those between
1'13.
in the
wi
mi and
wt
?n 2 ,
that
is
when
r/r' is
small.
The
initial
development of
Put
xx' +yy'
EQUATIONS OF MOTION
14
so that
is
r()1
'2
i
-f
?/i
if
=r
'
V*f ------2 - \
cos
rr'
>S -i-
-fw 2
2m
r a2
Vm^wo/
/
Wi Y
----
?r eos*S-H
m! + 7?l 2
V> 1 1+ W1 2/
/^ be the zonal harmonic of degree i with argument
9
Hence,
Then
?, r'.
2W *2
[CH. i
02
'<;
,,
-i
/ l+ ;
V
JL.UH.
%2
'
'01
'*
(_
,=1
we have
ft,
'*_ TP)
T+'S 'V '/'
and thence
y^'/^wiy
Wp(wi-f^a)
Wf>
*
t=2
L.
F being
'^vi
The function
useless.
!_\
1*12 (7)
becomes
>
the
first
term
in this case
In each case,
motion (Chap,
if
being useless.
the function be confined to
in), the
By
putting
= l,
we
wo have by
(3-6),
In the
satellite
problem,
is
much
greater than
and
m-!
than
and
wi 2
a/ a', n'/w are small, so that both these factors are small. If
Q1 m-j,
refer to the sun, earth and moon, respectively, these ratios have the mag-
magnitude, but
a/a', n'/n
are small
shown
The disturbing
in the
6, pt. 3.
COORDINATE SETS
13-15]
C.
15
FRAMES OF REFERENCE
Choice
may be much
abbreviated by
may
But
since the
mean
A
due
effect
known and
to that
EQUATIONS OP MOTION
16
[CH. i
some function
of this distance
is
used
is
the usual
of
second
is
of these methods
it is
convenient to introduce an
is
the angular
consists of the
momentum
use of certain
continued
its
the
size,
'
COORDINATE SETS
16-18]
17
the major axis, the eccentricity, the longitude of the axis and
the time of passage through the nearer apse or position when
the distance from the focus occupied by the central mass is least
;
it
remains constant,
the relation between the variations of the major axis and the
period is always the same. Various combinations of these elements
and of the two elements which define the position of the osculating
plane are also used as elements as we go from point to point of
the actual orbit these elements will change. According to the
:
definition, the
changes
will
other than that of the central mass, and the term Variation of
the Elements' refers to these changes. It will be seen below
'
named
'
them
unfortunately
the theory of differential equations.
'
in
involved.
new
set
new
variables
j,
and
if
them be
t=l, 2, 3,
Xi=fi(ai,a 2 .. .,<,,),
then the three additional relations are almost invariably chosen
to be so defined that they satisfy the equations
,
B&SPT
EQUATIONS OF MOTION
18
As a
we have
by fit dfi/dt
replaced by
d 2 %i/dt2
is
[CH. i
in the equations
last process gives three equations and these with (1) furnish
the six equations necessary to find the a,-. The forms of the
functions /i, and consequently those of their partial derivatives,
This
is
in the general
The
which
by the new
first
to be developed in Chap,
chosen that the equations have the canonical form, is largely
responsible for the use that has been made of them in theoretical
in this volume, as well as that given in Chap, vn, possess to some extent
the characteristics of both categories. No sharp division is possible or
necessary, the sole test being that of convenience for the problem under
consideration. Whenever a new variable is introduced, it can generally
be related to some property of the ellipse, but this relation is not usually
helpful except in so far as it may have led to the choice of the variable.
COORDINATE SETS
18-21]
19
1*20. The frames of reference should also be regarded as fourdimensional in the sense that the time as well as the space coordinates should enter into consideration in making choices of
new
made
The
is
of a variable which
some
infinities are
apt to be introduced.
1'21. The choice of a method for the solution of any particular problem
depends on a number of factors which should receive consideration.
As between the two principal categories described above (1*16 and 1*17),
the elliptic frame requires the calculation of the expressions for six
variables as against the three coordinates which are alone needed by the
On the other
much more simple for
observer.
For this reason, certain sets of equations belonging to the first category
have been so developed that their solution is as simple as those for the
elliptic elements.
more
easily than
EQUATIONS OF MOTION
20
[OH. i
rather than to those which are mathematically possible, and to work which
has been done in the past rather than to what may be accomplished in the
future.
The question
is
to
be made, that
is,
the degree of accuracy with which the initial conditions, that is, the
arbitrary constants, are known. In cases where the deviations from
motion are large, the literal method may involve such extensive
computations that it becomes practically impossible, even if the infinite
series used were sufficiently convergent to give the quantitative results
needed. Sometimes a partly literal and partly numerical method can be
adopted with but little extra labour. In all numerical methods, provision
must be made for changes in the arbitrary constants which future obser-
elliptic
vations
may furnish.
Some details with reference to the problems of the solar system will
make these statements more concrete. For the eight major planets the
elements are known with considerable accuracy so that corrections to them
need scarcely be considered at the present time as a factor in the choice
of a method. It is impracticable to use the literal values of the ratios of
their mean distances from one another owing to the numerical magnitudes
of these quantities: numerical values must be adopted for these from the
outset and these involve numerical values for the periods of revolution
round the sun. Little is gained by the use of literal values ior their
along which convergence is least rapid and there is little doubt that its
numerical value should be used from the outset. Literal values for all the
other elements can be used with but little additional work.
For the minor planets, numerical values of the ratios of the mean
distances are again a necessity, but since there are groups of them in
which this ratio is nearly the same it is useful to devise methods in which
this ratio has a given numerical value while the other elements are left
COORDINATE SETS
21,22]
arbitrary.
21
satellites
by using the
demanded.
2.
available.
is
Possibilities for
arise.
6.
The degree
7.
The
literal
developments and
at.
made
of existing
numerical or
tion.
8.
The extent
to
may dominate
EQUATIONS OF MOTION
22
[CH. i
D.
The
angle
is
which
it
makes with a
fixed plane
its line
makes
90, 6
is
refers to that
be denoted by
6.
Fig. 1.
The angular
velocity of
S3
d0
d0
P perpendicular to
The change
- 9)
POLAR COORDINATES
23]
to the motion of
23
The
-r-sin(v
0)
(it
The
velocity of
is
_j_
It
is
.
-7-
dt
dt
sinicos(u
compounded of its
^d0 cos
d,( _ m
^v
'
dt
-j-
0)
.......... (1)
velocity relative to
therefore
d0
= dv
-T-
1r -r:
dt
dt
cos
^.
is
a term depending on
its
F be
the force-
dt
--dt~
dt
-r
di)
~^'
Jt\
~dt
dt\di
*
The
acted on
relation (1) expresses the fact that the material osculating plane
by any forces which do work.
is
not
EQUATIONS OF MOTION
24
[OH. i
force- function
as that of #,
= r cos (v
y~r
2
?*
0) cos
r sin (v
6) sin
cos
i,\
r sin (v
0) cos
cos
i, >
cos (v
0) sin
sin
= r sin Z,
#) sin i
(t;
-f
0,
---CO
definition of
plane.
1*24.
so that
97 ,_/rfY
~\di)
we can
<?
<fr_aT
r>'
dt~
dr'
.(3)
or,
more compactly,
dr Sr
.
- Sr
dr
+ d(?
Sy
- 8 G dv
.
of the
CANONICAL FORM
23,24]
If,
we
in the latter,
replace dv by dv
25
+ dO
Sv
- SG
S,
dv
We
six variables
equations
result
r, r,
G,
HI,
v,
6.
By making
we obtain
(4),
which
is
+
dt)
The equation
Eliminating
between
integrate,
we obtain
"
,*-**
may be
1'23 (3)
and
T,
written
<Pr_GP_
~ r SF
2
r
this
and
dr
dt*
(6)
we have
When
is
dF
-rr
dt
dF
dF^ dF * dF
= dF
^-r + ^-i;+r-f^-0+-^-.
.
dr
dv
ol
dv
dt
dF
3F
dF
6,
dF
/QN
................... (8)
jT^a-r+a-v+or
dt
dr
dv
dt
.
'
[OH. i
EQUATIONS OF MOTION
This equation when R replaces F, where F = p/r + J?, is evidently
26
still
true.
1*25.
d'R dR dR dR
dR
r +
-jr=-jT--57=-;r
-5- v
at
or
at
dv
at
.
jP^-fJ?,
r
'
first in
The coordinate v
is
t.
The equation
obtained from
$-*-/*
<
3>
?o
7*0
2
J
............. ^(4)'
dv
Further,
^ = ^Y + r
*""A
dR -^Y
- +
d""(ft" "dvar
di
to the
same
H?^
dwar
>
dt
order.
T $ VS--.
d^
20" *
r
Sr
--1
r
[dR ^
j4 +
,
dv
FdR
VQ
9F
.......... v/5)^
NEWCOMB'S SOLUTION
24-26]
27
Newcomb
notices that
He
two
R~Q
e,
07.
He
differential
Cy =y 2
is
Qy l dt - ^
y 2 -~ -y l
where
2
(7,
-J^
Qy 2 dt,
a constant.
expressed as a function of n,
e, e>
dX ~~
__
de
Thus
give
cos
e,
sin
w,
(15),
X can be taken
t,
X
- cos X
sin
__
~~
a sin
we obtain
They
motion.
The
difference
may
EQUATIONS OF MOTION
28
[OH. i
inclinations of the orbits of these planets to the ecliptic are small. The
constant and secular parts of this term are absorbed in the constants of
mean longitude, so that the only doubt which remains is whether the
term gives rise to any sensible periodic terms, and if it does, whether these
have been included in his final results.
the
1*27.
independent variable.
With
these definitions
we have
~ du
dr _ dr dv _
m'~
~~~
dz r
If then
d*u
'
~~~
du dG _
dG
d,
dv
dt\
d*u
dO du
d
*K_^
O
dv^
dtj
we put
==/*(>
"
4-
so that
TS-
MW2
dr
+ 72),
ur -^-
du
.................. (3)
and
d*u
-1-5
2
dv
ft
?
*
dR
?
* o
du
+ 1* ?* dqdu
;r :j,
dv dv
X
TT-T
av
\/A'
2
do
-T-dv
=1
^r
av
"^W'
dR
--22o~'-u 2 dv
- r ;r>
av
=
dv
72 3T
//f
/cx
5)
(&)>
^ " (
\ '
(
\ 6 />
>'(\ 7 )/
(8)>(9)
THE LONGITUDE
IN TERMS OF
26-28]
29
new
mass
to the
When
72
=0 we
have
const.,
for
is
is
(3-2)
e
cos (v
or)},
\jq-a
(1
-e 2 ).
Since dq/dv has the disturbing mass as a factor, we can change the
variable u to MI where n = Uif(q) without losing the easy integrability of
the equation for HI
The special cases
.
It
at the outset.
The
variables
the constants w
qi for
l/<?,
qa ot
equations and the variables are
1'28.
Latitude equation.
From
easily
d
,
all ratios.
namely,
(1), (2),
= d6 sin i cos (v
6)
di sin (v
we
/LI,
dv = dv
6),
cos
(1
i)
d0,
deduce
{sin i sin (v
.
a sin i cos
6)}
mi
(v
0)1
/j
sin
cos (v
i
T-a2
dv
-,
\
)
sin
sin (v
v
0) dv,
x
sin
sin (y
(9),
/ix
a)7
av +
sin i cos i
----rr
sm(v 9)
-;
7/,
d^.
sin i cos i d0
- ^') = -77
-^ -T~
sm (v - ^)
dv
tf
_
"
sin
cos
dR _
=
" >(
EQUATIONS OF MOTION
30
It will
be seen, by differentiation of
dR/dT
dR
^. _
If
/\\
Z when
to,
is
shown
is
to be
vR
/\\
(-(?) an (-<?)g
sin
4*1 (1)
contains sin
no discontinuity in
[CH. i
nZ =
so that sin
that for
is
u.
When Z
we obtain
dv
-7-
=1
Z sm i cos i sm
r,
-.
dv
1*29.
.................. (2)
Anomaly as independent
by integrating
/K
(v
u).
/
variable.
r dX
pdt,
dr
dt~ P dX'
p = const.
d / dr\ _ p a d 2 r
dt\'dt)'~~rdX*'
F^p/r + R
therefore gives
dX*
dr
dt .......... (1)
^ '
dt
^ -2^+2^
=
dt\a)
dt\a
(6)
may be
dt*
dt
written
$'+"(*)'-*:
the disturbed
is
................... v(2)
'
major
axis.
RATIAL COORDINATES
28-30]
From
(1), (2)
31
we deduce
7TT92
dX
a
2
p \a
V^V
'
v
9
2
dr
may be
11
= -- ^2- +
E>
f)
Jtt
3 Z?
f>/t
<
written
/^
1Tr
(o)
Ir^dX,
d2 r
LL
71*
"'
(t
r2
dR
/*"'*
'*'*'
'*''*
ft
'*
/
(6)
^
The transformation
If
?i
2
/i/?
=a
2
/i=n a
the definition
is
will not
is
thus adaptable
be developed further
It
that, since
exception
1*30.
is
an exception common to
disturbing planet.
i
T
r = rp,
Put
ndv'
j.
coordinates of the
,
/1X
(At
A. direct transformation
dt*
dv
, t
methods.
so that
d?r
to the
all
~r
/dv\
\dt)
_
~
cttV
2
/t'
fflp
2
left/
~p
/dv\
2
)
\dv') }
r, t
1 dh'
~
to p,
dp
v'
gives
d?r'
~
EQUATIONS OF MOTION
32
[OH. i
dF
F,
W,
c77*~
With the
w/
dh<,
hp
dv = ajp
dt
dv
7^
by
/\/r',
rtt
o?
Replacing
Now
moving
The
by
r'*dv'/h'
and
dt/
/^
d7""
'
we obtain
*
/.
/2
definitions (1) of
~di
Sw
A'
dt
K\
let r',v'
elimination ofd?r'/dt
2
,
by means of (6),
(7) gives
dtp
dv'*
(dv\*
3i
_/
\dv') ~h'* dp
'2
r^ dF'\ _r^d/^dp
k' 2
2 V
dp
__
dv'~h'*dv
h'
dr')
A'
dr'
dt dv'
h'
h'
a/
............................
F, 9 is easily made.
in
an
ellipse,
(Chap, in)
aP'
dtp
become
dv
'
...... (8)
dt dv'
_ SF2
~
dp'
af
we have
30, 31]
RATIAL COORDINATES
u
where
Fo = /i\+ JpVcosfv'-w')
%
-',
^
1 +e cos(i/
w')
33
,
, x
(11)'
v
no longer
is
when
elliptic,
we put
The
^=
y,
Vj
= v',
0,
= 0, ^ = r,
F =/V = ^ + m' \
For then
where
cos
8= (1
-------- 2
H?
i,
=
--
/i,
m k =m'.
p cos
l(H-// -2pcostf)*
S\
of certain problems in which it is necessary to take into account the perturbations of the disturbing planet by another planet < the indirect
effect' of the latter, that is, the effect transmitted through the
disturbing
A case
Another case
is
equations of motion.
The
elliptic
form
B&SPT
it
for
EQUATIONS OF MOTION
34
[CH. i
celestial
Now
the symbols
let
reference plane
and
let v',
&
',
refer to
9,
v,
i'
any
fixed
arbitrary
We
have rr cos S
Put
=V
xx
4-
Define T,
F=1
We
xx'
4-
yy'
+ zz'
1 so that
yy'
k, I",
cos
i,
is
2k =
ty) (x
iy').
by
1 4- cos
i,
F'
cos
i',
2k'
= 1 4- cos
i'.
and similarly
4-
cos (v
31, 32]
To
find cos
S we must
add
to this.
zz' jrr
The
35
latter
can be
written
sin (0
0')
v') cos (6
6') + sin (v
- cos (v + v') cos (0 + 6') - sin (v + v') sin (0 + 0')}-
We
fl')
in the
form
)>
KQ
replace
Here
-Or
1-^r,
1ST
W--OT',
|T,
/V3
20,
A cos a -h B sin a
where
(7,
(7
(7
sin
cos (a
)>
(7cosoo
(7
by
= J.,
<)=-#>
3-2
CHAPTER
II
sums of
way.
The majority
much
labour.
may be carried
It
monic analysis usually give higher accuracy and are less laborious.
The coefficient of a periodic term in the expansions of most of
the functions considered here takes the form
a '(a
l
where a
2
i
a parameter and a
H-a 2
-f ...),
(1)'
...
It is this latter
much wasted
labour
x = exp. 6 V^T,
2 cos
iff
=x
i
-t-
or*,
1 sin iff
=x
or*.
(2)
LAGRANGE'S THEOREM
1,2]
When
work
is
37
made
p=a
2a* v ^Tsin iff = z -p*z-*
2
1,
-\
We
where
o,
i,
x,
so that
The gain
where a
is
a parameter and
and
its
<>
38
Fx = F(x)
where
(f>
[CH.
x =<f> (x).
theorem
to several variables.
it
may
Regarded as a function
in the form
F=F
A=~
...... (3)
denotes that a
suffix
is
D = d/dx.
put equal
Evidently
to zero after
F = FX
Q
Put
The use
of the operators A,
implies that the
functions on which they operate are expressed as functions of #,
.
successively,
4> + a^ Aij,
AyJ = ^
Dy=I+a dy
fDy,
J
dy
Ay =
so that
and therefore,
From
for
x,
Dy,
that (5)
?/,
we can show, by
A
Assume
<f)
any function y of
this result
is
F=D
we have
n~1
induction, that
n
((f)
n
<f>
D (AF)},
the change in the first term being made by the use of (4) and
in the second by the commutation of A, D. But, by putting
we have D(AF) = (QDF),
in (4) and operating with
g=
"1
n
is
so that the portion under the operator
(<f>DF <f> ).
11
Hence
and since the theorem
is
true for n
= 1,
it
holds universally.
2, 3]
F become
<f>
Fx when a = 0,
39
the coefficient
proved.
Particular case.
When F(y) = y, we
yy2
have
,72
2*3.
If
,/
yi
where the
where
y3 ) = a;i + aai<l>, i = l, 2, 3,
are constants and a is a parameter, and if
= Xi + aai<l>(yi
a$
F and
tives of yi,
<f>
yz
...(1)
2/2> y?.>
and
further, if
Fx = F(XI,
o;2j ar8 ),
then
2/2,
yd^
x)
The proof
follows the
as a function of a
regard
Next, by differentiation of
where the
as
=j
for j
or
= 1,
But from
....... (2)
first
i 4= j.
2, 3.
(1)
(1),
The
result
1 or
a,-
according
and summed
is
we have
2JL. 2*,.
(5)
40
These two
may be
sets of equations
[CH.
set for the determination of J9yt-, and the second for that
of dyi/da. They are the same except that the absolute terms
first
</>
and therefore
The remainder
The most
2*4.
of the proof
<j>DF
Hence
is
2*2.
= -n +
Q/i
y* #0,
,
= 1,2,
3,
It is
not
difficult,
<tx
</>t
(#1
*2
D == 2
# 3)1
t <j> lx
a^
/^x
we obtain
in terms
or any function of
Lagrange's theorem can be used to find
of g from Kepler's equation X=*g e&\i\X (3'2 (16)), and was probably
suggested by this problem. The extension may be applied to the Jacobian
a single periodic term or to a Fourier set of terms. The more general case
mentioned above is that of the Jacobian solution where the disturbing
function contains any periodic terms.
2'5.
of an implicit equation*.
F(y)
and
^(c Gosiy-{-disiniy)
let
where a
*
E.
y
is
a parameter and
W. Brown,
>
x -f a cf> (y\
</>
........................ (2)
same form
IMPLICIT FUNCTIONS
3-5]
F (y).
as
41
a if
b it
when F(y)
of the notation,
</>
(x),
an
F(y)=F+'S,
Dn-
(^
DF)
= l,
2, ....... (4)
Let
On
).
The
...(5)
identity,
n
2
(<f>
shows
when
DF),
all
D^r D".
lire
the same.
On
hand member of
by
tl
(4>
DF),
applying
(5)
(-
l)
I)
we obtain
a series which
<f>
DF
the expansion,
Taylor's theorem, of
-DF.^(x-a<f))
..................... (6)
in powers ofa<p.
Hence, the constant term in the Fourier expansion of
(7)
when F(y}
is
term in
Fourier expansion of
the
expressed in terms of x,
is the
same as
-~F(x). + {x-a$(x)},
or in that of
F (a). ax
-]-$
[x
a$
the constant
............... (8)
42
[OH.
The use
different form. If we
this definition,
same as
the constant
According to
x, is the
But
is
a constant term.
we
Since
= x -f a<f>(y)-m
constant term in
Similarly,
bi
by taking Dty =
= constant
sin {ix
za</>
(x)}
jctx
F (x).
...... (12)
sin ix,
term in - cos
i
{ix
we
obtain
ia
<
(x)} -y
dx
F (x\
. .
.(13)
. .
.(14)
in
\
I
-7-
ax
<f>
(x)
>
)
F (x).
IMPLICIT FUNCTIONS
5, 6]
43
When
may
tegral.
Extension
2'6.
to
we have a second
to that used
before, namely,
F= F (x, x'),
A
J}
= d/dx,
= a/a*/,
D'
etc.
m=
x, x'
It evidently leads to
F (y
when
'
-a^(
ai
),
x'
same as
-')},
the
...... (2)
or in that of
,^{x-a^(x),
And
the constant
F(y,y') X (x,x')
..................... (4)
same as
is the
F(x,
x')
[OH.
that in
% (z,
z') -r-
=x
z'
a<f)(x),
x'
a'<f>
(#').
...... (5)
by
(5)
--F(x,x'}~^{x-a<}>(x\
or
in
->' OO},
*'
-(6)
interchanged.
In order to apply the theorem to the coefficients in
We
(2).
"
const,
term
When i = Q, we
"
in
'
+ i'x'
sin {ix
i, i'
(1),
we
successively in
ia </> (x)
(x'}\
i'a'<$>
= const,
term in
"*"
C S
+ sin {t
(
-a 1
Jl
or n
+ i'x'),
sin i'x',
and obtain
successively,
0t
sin (ix
obtain
-f i'x'),
+ COS
V-
-a'
^*'(^)}
</>(^)| |l
V-
\
/j
i'cftf
r (x')\
\
i'a'tf
r (x')}
/)
\l-a~<f>
dx^
^(*X),
^
(x)\
)
? ,a
l
tix
...... (8)
^ (^, ^).
^
'
...... (9)
The formulae
for
b iQ are similar.
When i = i' = 0, we
aoo
= const,
term in 41
^
-a
ox
<j>
(x,
(x)\ \1
J
-a'
1,
,
c?<^/
and obtain
<f>'
(x')l
J
F (x,
x').
SYMBOLIC OPERATORS
6-8]
27.
If p
is
If
45
D
=r
p jf(x\
\
then
x,
D = x ^-j-x = ^r---x
g
where
/>
p =a
if
Hence,
we have
(
n m^,n
4-
a2 D2
D oc n
= Sn
/(a?)
If then
aiD
4-
(px)
b n x n ),
<
4- ...
n.
= 0,
. . .
D
n
n
D
f(px) = 2 n b n (px) =p ^ n b n x =p f(x).
we have
Thus
/(Piffi,j02ff2.
ss
pi Dl .p*
D*
f(xi,X2,
where
-..)
A = ^3/9^.
The
stopping at
some
D
p = (1 + eyj* -
We
can also
make
z = D log p,
we put
Let the
(i)
4-
factor
series
in
powers of z
e.
series.
be
_
\l
then
for
Put
kD(kD- 1) + ....
2*8.
A=
+ ey kD +
1,
.B
= 6 + SS^a*
2
p= a
cos id,
2a* cos 16
5 =
&<)
4-
26*
z* 4-
p^"
1
',
2* .............
CO
if
46
AB = (A + '2a p
Then
z-*)(Bo
2,b i p i z-
[CH.
) .......... (2)
z*, i
>
0,
We
and then
to replace z i
thus reject
all
and
in (2)
it
is
A Q BQ + AoZbip'z-* +
The
On
and
. .
z*,
that
is,
+ a^) +
of 2a* cos
(2&<+2
2
p by a we find
,
iff
a+2&2> 4-
. . .
...... (3)
................... (4)
magnitude of
sum
in (3)
product by the
of the suffixes of a,
b.
is
made
. . . .
(ii)
If
A'
= 22a,a* sin
is
B'
substitution for
The product
iff,
= 226<a< sin
0,
iff,
a and we have
an even function of
in terms of cosines of
iff.
We
ff
and
is
therefore expressed
of z i in
z-* -f
8, 9]
which do
in
= &o
Hence we
0.
47
......... (3)
If
A'
the product
an odd function of
is
coefficient of z l
and we
-r-
0.
in
= 0,
(ai6 t-_i
^-2
a^o)
+
-f
(- i^+i
-H e&f+i&i)
...(7)
cosines or sines.
(i)
'
Let
G ^Sa.a'cos'fl =
Since this is
an even function of 6
of cosines of multiples of
2a cos 6 z -f a2/z, we have
6.
it
= 0,
1, 2,
....... (1)
With 2=aexp.
0V
1,
so
that
2i
+ (* +
2*
"2
-^
1.2.3
iff.
48
in
is
obtained by putting
[OH.
in this
formula.
The
result
by the use of
____
<+
___^
(ii)
Let
S=
The terms
'
1
Sa^a* sin
in
= 26, (2
sin 0)<,
0, 1, 2,
will
By
a similar procedure,
we
2 sin (2i
2i+ 3 H
1) 0,
(2^4-5) (2i
J
-----
+ 4)
2i+5
- 6
2t+7 +...!>
---T2T3---......
4.3,
6.5.4,
(5)
9, 10]
2*10.
49
series.
cos 20 +...),
......... (1)
in the form
+ 26 2 a 2 cos 2(9 4-
2&!
cos 6
made
60
powers of x.
As
and
/(a
. .
2a* cos id
=z
we put
1,
i
-f
expansible in
is
-l-#)
an even function of
Evidently f
a exp.
z
is
before,
that
0.
a 2 = p, so that
p*z~*,
A = mz + a 2 z 2 +
Put
iff
for
...,
> 0.
-B = ai^ +
and
a2
^+
>
so that (1) becomes /(oo -h ^4. 4- i?). This function will be first
obtain
expanded in powers of p by Taylor's theorem.
We
Put
is
put equal to
/i
zero.
is
to a Q
by
accents.
Then
\dP
Hence, replacing
2
p by a we
,
obtain
a2
"*
5 </'
50
second, third,
...
power of z required
[OH.
in the
We
plan,
from z
coef. of o?jz,
coef. of
a4/2 2
coef. of
a 6 /^3
to
from
z 2 to z 5
from
z 3 to
4
;
and these are all that are needed. The work of expansion is thus
reduced to operations with positive power series.
We next expand fj and its derivatives in powers of A, which
contains a as a factor. As far as a 7 the result is, with/ set for/(a ),
,
'
a2
,
^2
(A f"
7\
(2a 2 /
/'"
2V
+
,
'"
*
"
+ 0!%'") +
+ Oaxa./o'" + ^yo
(6a s /o"
vl
(2a/o
5
lv
4-
The
powers of
formed.
We
have, in fact,
A 1 = ajV, A 9 = ai6 * 6 +
The lower powers
8
7
6ai a a *
Thus
= (aiz 4-
+j (az + a^z y~
For j = 2, we have
aiz*)'
^ 3.
^2 = aj2^2 ^
2a 1 a 2 23
+ (a22 +
+ 2a!a 5
4-
-f-
(a3
....
A*
2a!a
2a 2 a4 ) ^ 6
+ (2aia 6 -h
10]
51
An
only
when
series,
expanded
(ii)
The expansion of
+ 2a a sin 2(9 +
f(2ai sin
.).
/(-Vin"4+V^~I/0,
so that the results of
(i)
the resulting series will contain both sines and cosines, but as
the important applications are confined to those cases in which
only, or
0,
in
which we
shall
di
(/3)
be replaced by
be replaced by
V 1 where
- A V- 1
-
it
occurs explicitly
/odd.
We
(a)
A2
"
When f is
n
44
**
an even function of
A6
0,
the expansion
is
^2a / "+^<
2
3!
+ ^(63/." -</-) + ^
A
^^1 ^"Z./ U
...(**)
4-2
52
When /
(b)
we
of z y
is
as in
"la
by
an odd function of
4/o'-~/o"' +
i
replace z
(i)
1
cos
the
(6
i2o
V^l
!o
3o
we
2a J sin
.(5)
{
replace z by
id.
a 2 ,a 8
._
which
is
2'11.
id.
the expansion
6,
is
|\/o*-/v
- A 6a 1 a2/ '" +
When
[CH.
0).
1 f,
a;
may be
conveniently
adopted.
Taylor's theorem
may be
/(tt
written
+ x ) "= ex P-
./(o),
^ = 3/3a
Hence
/(oo -f
ttia? -f
a-!^"
= exp. a^xD
exp.
a^x~ l D ./(a
).
. .
.(2)
if
we sum
for k.
It is
10-12]
The
by interchanging
i -f k, i,
that
is,
53
evidently found
is
by interchanging
cij
Hence
a_j.
...(3)
tf-i
The expansion
~ ex p
>
(3)
&'*
>
&'~~
= 2 cos id.
then gives
(9)
cos i0,
..(4)
we get
x~ l
=2V
1,
1 sin id,
and sines
for
odd
values,
and
+ 22i (- I)''
...... (5)
with i=0,
1, 2, ...,
first series
being omitted
when i = 0.
It
may
0'Qi(-^) =
/i(2*),
we make
use of
2ai
210
cos d
(3) with
4-
2&2a cos 20
series.
. .
,)>'
For
54
[OH.
It
integer.
For
+ 2a2 a2 sin 20 +
*an
(2a!
we apply
we put
when j
a positive integer.
is
is
.)'
2'13.
The expansion
(i)
of
/ = -/</' =yi) =
lv
is
A2
A*
sine series.
a2
^l
...
1.
A*\
/J4
The expansion
sin (2a!
is
we
obtained from
210
as in
replace
z*
210
by 2
(i)
l
cos
A*
3!
(5)
. .
.)
v
by putting /</ = -/o'" =/o =
"*"
A^_ ai a? (A + A*
4!
7!
"*"
A^\
z V21
6!/
,M
J. 4 \
+
In the
1:
A*
5!
id.
of
by V
2a sin
l
(2)
i^.
1.
BESSEL'S FUNCTIONS
12-14]
55
where ^
is
suffix,
first
kind,
we define
it
by
i
comparison of coefficients of x shows that
y _i (a;)
J, + i
(*;)
= 2?~Jj 00,
=2
*)
/M 0*) - ^+1 (*)
rf
...... (3)
differential equation
is satisfied.
2^(*)^
1, .......... (4)
j-0,
This result
Put 0= exp. g V
1 in (4), so that
21/2 = 2 V
We
1 sin g.
obtain
exp. (x sin g
= 0,
1,
. . .
...... (5)
In
this equation,
The
real
(g-e
sin g)
(ig V
V-
1).
1
j
g,
put x
= ie
and multiply
obtain
= S, Jy (fc)
-e sin g)^^ J
We
exp.
{(t
- j) gr %/- 1
}.
(ie)GOs(i-j)g
+1 ""
'~
'
56
The same
the results
process
may
sin(tf sin<7)
may be
applied to
With the
(5).
[CH.
aid of (2)
be written
also
22,Jw+1 (ar)
sin (2j
L> *>
l)g,\
...... (7)
also be defined
i r*
from
is
cos
(j<
sin
by
<) d<,
......... (8)
(6).
2*15.
The Hypergeometric
series.
This
namely,
series,
...... (1)
F=0,
...(2)
F(A,
B,
a~
C,X)=(I-X)-*F(A, C-B,
r ),
C, x) = (1 -^-*- F(C-A, G - B, C,
F(A, B,
The differential equation may be used
of
made
powers of y where x
and if we put
t
=a
-h y.
).
...(4)
in (2)
in
of
in
...... (3)
(a
- a)
(M
2) a n+2
{n
(2a
1)
+ a(4 + fi +
1)
0} a n+1
HYPERGEOMETRIC SERIES
14-16]
when two
of
them
are known.
By
all
57
the coefficients
fti
can
be obtained from
formula
may produce.
following formulae are immediately proved
The
~
~d~c~'
IT
>
>
'
............... ^ ^
'
G -f
by
1,
1,
tive
B unchanged, or
A 4-1,^.4-2 with
(7+2 with A, B unchanged or with A, B each increased
2 with
2*16.
we can obtain
(7,
etc.
unchanged,
Expansion of (1
powers of
ax)~ (1
/#)"* in positive
and nega-
x.
/V ~_ n(n + 1)
\r)
...
(n
rl
+ r--l)
/n\'_
""
'
'
\OJ
...... (
the accent being used to avoid confusion with the usual notation
Expand each
product
is
KO'-'H'G'M'
''
j --
'
The
58
The
coefficient of xn
putting
of j.
The
= n+j
[CH,
coefficient is
+jj
\JJ
(&
n.\
-4-
(& 4-
n.
///
-L 1 \
_i_
1 \
'
1.2
t,
s.
co-
+ n,
14-
n,
a 2 ),
When
coefficient of
s,
we obtain
in this
way the
x n or ar n namely,
,
where
(3)
1+n'll-a2
(l-)(2-)
+w.)'
a*
s(s+l)
1.2
(l
l-s l+n-s
4
n)(2+w)'
When
1.2
+-..|.
(5)
is 1.
16,17]
If x
exp. 6
1,
69
+ aa )~ 5
2a cos 6
in the expansion of (1
(Cf. 4*2.)
is
may
be-
aidJ
+ Gtefl2
If
x
= --- ^oH-AaorhA.2 a
1 41 -f x
...
-r'
.,
IK
\\-\-x:
+-.-K
)
......
where
Aa = a -ai, A 2 </o=o
-"i
H-
A 3a = "o
^2,
3(/i+ 3a 2
s>
...... (2)
is
easily proved by expansion of the right-hand
of x. If OQ, i,
are positive, the coefficients
. . .
member in powers
Aa A a ao,
may
,
. . .
...... (3)
where Aa
A2
<7
...
/=l-M? + &
a
a
...................... (4)
For
This
is
where X
when
60
The
+ l)th
term of the
[CH.
may
be written
+ - ~^l
K( Yfl
Ml-"V L
The formula
!ll+
by
(4)
which
(3) is
+ x)
rf*F
y are positive
the value of
The
0.
dF
^r
+^-0.......... (6)
(7), (6)
final integral is
given by
member can
be integrated.
F for
from
+{(2+7) .r+j + n + l}
/*,
When
>
...
F=sl when a
Since
first
since
-J
(afi
___^4-1
.......... (5)
+
j(j+l)lj + n + ll-S
applicable if we put
I
In the applications
chosen
-J-
differentiating
(6).
^l, n =
2*18.
sum
/S
is
known
coefficients
for a
up
to
+a # + a 2
series.
We
a?
+ ........................... (1)
x of x, and that we
then know the value of
particular value
a_i.
of the series
also
know
the
17-19]
We
61
have
dS
dS
approximation because
ful
it
i ,
a n from
(2)
mentioned.
2*19.
Note on
the
series with
different arguments.
If
we
2aV
where
2 cos j'0',
j,
/ = 0,
1, 2,
it is
. . .
is
omitted,
22aX,'. 2
where
it is
0,
/= 0, j
j'
are
2 a'/
that
is,
2 cos/0',
no attention
either j or
is zero,
is
to
and
be paid
to the
the factor 2
is
o',
sin j0
2aj2
j'
/0').
6 '\
CHAPTER
III
ELLIPTIC MOTION
3*1.
The
relative
hyperbolic
method
that one
of distinction
F
where
time;
The
is
V is
/JL
is
case
is
and
it
is
is
which
3*2.
#0
will
1, 2]
__
63
dt*
dt
r*'
\dt
integrals,
,dv
where
a,
The transformation
between
in 1*27
d?u
/j,
>
T-
2
'*
Vat;/
/i
1\
/X/,
2w - - )
a/
/2 (
/r\
...... (5),
The
u=
where
0, tn-
T*{ 1
I/?*,
fdu\
+ u - T22 =
j~22
o?w
(6)
is
+ ecos(t;-cr)},
............... (7)
<
A2
1
Hence
so that the
1 -he cos (v
= ----7T-- 2T
a(l
cr)
.................. ( y )
of r are a (I
e).
Equation
rdX =
by
The
ndt,
= n*a3.......... (10),
/i
(11)
of motion with
_
d2 r
--,
dv
aVr~e2
...... (12),
(13)
show that
A
ELLIPTIC MOTION
64
Since the
of (12)
is
is,
when v=
Thence,
The
if e
the solution
e),
.................. (15)
when
a (1
e),
that
we
fsm-
,
1
Vl
- (Ps'mX
v
e cos
-,
\
We
nr
-&.
y = nt
Finally
of r are a (1
= a(l-ecosZ),
where we define
[OH.
cosX
,
cos" 1 - ------- -.,
e cos
rcos/= a (cos
e),
sin/= a Vl
e sin
A,
...... (18),
With the
</,
(19)
df
a2 V
e2
dr
'
fig**
r*~
dg
__
~
dX _ a
ae sin/'
v'T^?
'
W/~~r'
...... (21),
(22), (23)
The constants 2a, e are the major axis and eccentricity of the
of the nearer apse from the initial
ellipse w is the longitude
the
line, and e, 'the epoch,' is
longitude of the body when it is
;
mean
as the
mean
longitude,
2-4]
Frequent use
3.
<jE>
= exp./V
be made of <, ^, ^,
will
=*
1,
exp.
65
tj
defined by
e(l+r,*)-2i,,
..................... (4)
so that
2 cos
t/=
<
!/</>*,
V-~7 sin t/
^ - !/<*, etc.
...... (5)
= $e + \eif, and
JHJ
....
(6)
The
section give
...... (7),
(8)
(9),
(10)
3'4.
An
is
of the form
whether 6 be
/,
or g.
It
it
was
were subjected.
B&SPT
appeared in the
last
chapter that
we
if
ELLIPTIC MOTION
66
in
[CH.
from equations
a,
(15), (16), (20) of 3*2 that r
in
terms
of
have
this
X,
expressed
property.
The operations to which they are subjected are those dealt with
in the previous chapter, partly expansions in powers and partly
It is evident
g - X, f X, when
rp
smq(fg),
reappears in rPcosq(f-g),
and the latter can always replace the former in
it
the applications.
Series having this property will be
named d'Alembert
The
of x>
in
<
in terms
?;
terms of <,
The
3*5.
series.
II.
77.
terms of
form
/ and
r in
terms of
X satisfy linear
furnish the reason for transforming the equations of motion to the forms
given in 1*27-1 '29. It will be noticed from 3'3 (7) that r cos/, rsin/' in
planets about the sun in equal times, i.e., that the rate of description of
areas is constant, moans that the right-hand member of 3*2 (2) is zero it
follows that the resultant force is along the radius.
;
Law
If
we
substitute 3-2
(9), (4) in
= n2 a 3
= 2T/>i,
give
is
KEPLER'S LAWS
4-6]
67
is the sum of the masses of the sun and planet. The largest
has a mass less than 1/1000 that of the sun, so that the
Jupiter,
planet,
difference between Kepler's third law and the exact statement is small.
where p
fj.
equations in which the terms are obvious ratios of times and lengths.
example, the equation,
acceleration = mass x [length]
which is the
formed to
initial
which
The
obviously correct in
constant a so defined
is
~2
,
its
is
~J
For
[length]
is
trans-
~2
,
dimensions.
usually called the 'mean distance.'
is
incorrect,
This
even when
if
true longitude or the time is so used; in the latter case, however, 3-11 (2)
is the mean value of I jr.
In the actual integration of the equations, it will be seen that the first
arbitrary constant to appear is that on which the distance depends: the
mean angular velocity is seen later to be a function of this and of other
arbitrary constants which have arisen in the integrations. But since, in
general, we can deduce the mean angular velocity from observation with
much higher relative accuracy than is possible for the mean distance, it is
Confusion
the letter
is
a.
ELLIPTIC MOTION
68
certain periodic
'
term
There
is
involved
3*7.
<t>>
'true
much
usually not
;
when
The logarithm
of
[CH.
X>
f V-
1/x
....
j=l,2,
(7),
we have
\p
(a
iT?)
The binomial theorem
binomial factors, with
.......... (1)
xa (i-w)"- a
7i\^^ Q
-
(i-^)
.-(2)
the
coefficients,
= 0, which
provided the term for j
not repeated. If we write
is
is
is
and define S^ Sj by
6, 7]
q.
.
1)
-j -~
"
"
Vixj + 27"
'
"
p +""
?)
""1.2
...... (3)
where J
= 0,
1, 2, ...
By putting
for
the term
its
for^'
value cos
being
qf+ ^l sin g/
in terms of/*,
rp cos qf,
as Fourier series with
The formulae
qf
argument X.
for Sj, Sj
r,
Hence,
if
= i+
we put Sf =
.
(1
2
T/
)?^*
T^
We
have,
so that
^_
-j-^2
4-
(j
+ !)(/+ 2)
(-5)
ELLIPTIC MOTION
70
Tf
s=
value of
T when
the sign of q
is
[CH.
ra
that
,.. .(6)
where j
0, 1, 2, ...
In the applications,
the values 0, 1, 2, ...
number
from 3-3
from which
The
?*^cos
pcf),
p = 0,
1,
gives
= 0, we
have T/
=T
The following
1,
when
= 0, and
is (
iy.
p=
2,
(8)
TJ = 1 and
For
</
p=
J.
It gives
particular case
sin/'
of terms.
(7).
we have
...).
When
Hence
...(9)
7,8]
71
Hence
= (l-*)-*Srf(l+j</l-*).2coBJX,
and
...(10)
3'8.
The formula
q = 0,
by
-e)-*a-
(1
We
a*
obtain
in
p=
2,
which j=
The
we
and the
factor 2 is
obtain, as in 3*7,
it
ty = J**-^df a **Jl-e2
An
0,
1, 2,
series for
by combining
/,
with q
= 0.
omitted when j
For
3*7 (5)
'
g vanish
It should
together, gives
is
ELLIPTIC MOTION
72
The
[CH.
in
functions of
in terms of
?;,
according to a
remark
in 3'4.
They
are,
X=f+2zl(-r,y S injf,
and from 37
(4),
with p
from 37(1),
= 1,2,...,
.7
...... (4)
= 0,
(5)
where
$j,
=
j
3*7 (3)
From
with p
these
= 0, and
we obtain
cos
qX,
sin
qX.
(7),
IN TERMS OF THE
MEAN ANOMALY
FOURIER DEVELOPMENTS
of
d
............ (1)
in the
expansion
is
'
cos/
sin^, replace the first sin in (1) by
The form of the first factor of (1) shows that developments
by means of Bessel functions (2*14) will be needed. In these
coefficient of
may
appear.
8-10]
73
rcosf, rsin/, in
A", r,
terms of g.
When/(A) = cosyLY,
k
7 cos
{(j
According to 314
(6),
of this expression
cos {(- j
is
k
............... (1)
.J^. k (-je\
forj=l, 2,.... When j = 0, equation
= 1.
zero if k ^ 1 and
\ e if k
3'9 (2)
shows that
it
is
we
have, for
cos
.
sm
kl,
k ,
cos
-J~- k (ie)
9(7,
v=2
^
kX
J9
sm^
When &=1, it
214 (3), although
convenient to
is
-f-l,
make use
/.>\
-f-2,
....... (2)
of the formulae
if
we add
We
cos
X = - \e + 2S
sin
-'-
(*
2S - Jj (je) sinjg,
The expansion
X=g
gives
X = # -f 22 J
-.
4- e
sin
X,
(je) sinjg,
= l,
2,
....... (4)
a(l
ecos X),
rcosf =acos
r sin/= a (1 -
e*fi sin
A.
ae,
3*11.
Expansions for
2
2
a/r, r /a ,
factor, it is
some-
by*
since r
The
result is
(2)
The
term of
a/r,
expressed in terms of
an important property.
unity
For the expansion of r2/a2 we have
is
is
jzl
\
by 3-10
(2),
^
*
The
310
(3).
j._ i(je}
The use
.?>,
'
./
of
214
=
l+f^-2JiJ,-(>)cosj7,
j-1,2,
....... (3)
10, 11]
The
relation
dt2
with
/L6
2
?^
a3
is
given by
aj
\r
This equation
(2), (3) is
78
easily
deduced from
3*2 (1)
and
3-2 (3).
2 cos (jg
The expansion
Hence
e cos
je sin g) (I
.
of 1/(1
6 cos
X)
(4)
g)~
in terms of
X,
e is
given by
3*7 (9).
e )"^ (\
(1
or
2(1- e
)~i
cos ig),
2
e )"^ 2^1*1 cos
(1
+ 22?7
*'
17!
cos
igr,
may be
{(i
77,
1, 2,
= 0,
is
. . .
1,2,
written
This
is
The
= 0,
1,
2,
. . .
expansion of
ti
/r
is (1
- e 2)"*, and
i=
when
The change
it is unity.
except for
the previous expression for the coefficient of cosjg therefore
gives
,,2
Jr
The expansion
for
is
>
A>
= 0.
df_a*(l-e^
r*
dg~
'
ELLIPTIC MOTION
76
arid
[OH. in
together.
,The result
is
sin;//,
where
While
= 0,
2, ...;
1,
............ (6)
l, 2, ....
is,
The term
of lowest
Hence the
is
The
will
5*
(9)
4.^ +
2.
5j 2 + ... + 0'~
.7-
11-13]
3*12.
77
recurrence.
The expansions
d*x
" "~
'd?
d^y = "
\*.x~
73
'
rf?
fjti/
'
yJ
of
.r,
_?/,
sin//?-
when those
of
2
?-,
I//-
are known.
which
is
deducible from
?)
rence, the series for r for all values of p,
when
by recur-
p
p
Expansions of r cos qf, r sin qf.
of
g.
The
series suffer
1 terms of the
-f
same order
in the
7
however, we do not need such expansions beyond e the
*
extensive tables given by Cayley for various values of p, q and
If,
literal
and
Mem. E.A.S.
e,
when they
vol. 3.
ELLIPTIC MOTION
78
3*14.
[OH.
The constant term in the expansion ofr? cos qfin terms ofg.
By means of the
is
immediately
same
Hence, by 3'2
,
qf -r a
(1
-e
(21),
we need the
)*,
-^~2
V cos j/cos g/
is
by j = #, that
is, it is
(1)
where
r-i
''~
rJ^
'
-P^
q+l
(_p_2)(- ? _
1.2
!.(?+!)
p=
values of p ;
in this expansion
is
obtained by putting
found to be,
(8),
LITERAL DEVELOPMENTS
14-16]
+2
and that of
The
(e/2)>
79
cosjg,
3*16.
The
may
reference.
is
e,
used.
= y _ 4 e s n / + (3 C 2 +
2 e*
4-
4-
(I
+ 3e 6 ) sin 2/~ f
A sin 4/- ^2
4-
2e
+^
^4e
/26
/1097
\W
B
C
95
2"
sin
_\
"36"
1223
15"
8e
4-
sin
7/
2
e
+ --
sin
902
3/
T5"
..(1)
8
c
?^
\ 6"
S1 "
^ /103
ff
4c 5
4-
6/-
2
g+ ^5e -
sm
5/
1
6
e
1
o
~ 5957
sin
sin
,\
8e 7
4-
c')
43
IT
+
sin 2#
S1 "
/;
= 1 + 2c 2 -
2e
oc
\j
4
e
4e
cos
)
/16
\T
128
~5"
2r;
cos
)
(,3e
C S
^t 9
108
e
6
c
""
/125
\ 12
.
cos6^r-
B ""
'^
4375
"72"
16807
--e
6
4-
C S
)
K
cos 87
.
/0 ,
cos75r ....... (3)
ELLIPTIC MOTION
80
- ~6
4
c
+^
6
c
(4e
/64
4
c
1024
TT
6
c
cos ty
cos
(9e
/625
"
5
c
[CH.
5
c
cos
~~
15025
/n
(4)
It is evident that the expressions for/, r are d'Alembert series
with respect to the association of powers of e with multiples of
or
tsr
(3;4).
is
that of numerical
=X
will
be found con-
venient.
When
= a(l -ecos
r,
Z),
See, for example, those of Boquet, Obs. d'Abbadia, Hendaye, and of J. Bauschinger, Tafeln zur I'heor. Astr. Leipzig.
16, 17]
81
small.
will
it
show how
x-e sin
Hence
sin
x=x
i^+iJotf 5
(g
+ x)
(1)
--
= esin(^+^)~Jtf 3 sm 3 ( + ^) + T i^e 5 sm 6
(7
Calculate
(<7-f.r)-
(2)
# from
(7,
C cos .'0
cos
sin
</,
= e sin
(3)
</
These give
C 2 =l 4- e 2
2<?
cos #,
C sin
+gr)
(.
= sin #,
<?sin (#
+#) = 8in #
(4)
With the
aid of
(3),
equation
(7sin(o;~^
may
(2)
be written
(5)
#=#
term
is
found to be of order
(7 sin (a;
gives
By
e 6 /46.
)
= -^sin3 (# + #o)
(6)
x with an
sin (x
- ff ) = -
Q(J
sin3 #o
=-
e3
gg*
sin3 #
these giving the same results as (6). Should still higher accuracy be needed,
can be obtained by substituting the value of x thus obtained in the right-
it
hand member
the error of
*
of (5),
An example
B&SPT
but this
will
is less
CHAPTER
IV
The disturbing
m'r cos $
.-.
where
_ m'
is
is
radii
We
r, r'
and
r, r',
is
Hence
S = cos
(v
6) cos (v
cos 2 ^ I cos (v
v')
6) sin (v'
6) + cos / sin (v
+ sin 2 J / cos (v + v' 20).
(3)
In this formula, /
the longitude of the disturbing planet from the same fixed line,
v is that of the disturbed planet reckoned to the node and
and
then along
If
cr',
ty
its orbital
The
of r
r, r' 9
v' 9 v,
respectively,
and
if/',
/be
the true
we have
ff
9
may
GENERAL PLAN
1,2]
or
mean
There
anomalies.
is
83
of them.
when
is
arbitrary
and the
e, e'
in-
and
respectively equal
tion becomes
with
The
/S
r, r'
reduce to
4- TV
a, a'.
g'
The disturbing
func-
tsr'.
R = ^ (1 - 2
cos
$ + a 2 )"* -
ft
~ a cos
(/<
/
Co
{l-Ja+t(2acu8S-a)+...},
The
the general
are due to the need for using values of a which are frequently
as large as '7 and to the fact that the coefficients may be needed
to five or
more
were used,
problem
is
84
[CH. iv
The disappearance of the term a cos S from the expansion has important
consequences in satellite theory where a is very small. In the planetary
theory it simply has the effect of diminishing to some extent the terms
with argument S, so that those with arguments S, 2/S' have coefficients of
about the same order of magnitude in the coordinates.
4*3.
When
than with
for
A much
more far-reaching
effect is
of the time, these multiples take the form jg j'g', where j, j'
are positive integers, and the coefficient of the term which has
of the eccenthis angle as argument contains the power j
tricities or inclination. The coefficient of t in the angle is jn j'ri >
and when an integration is performed this quantity will appear
|
The
as a divisor.
upper sign
will
tend to
may
in-
it.
Since
which
n, n'
integrals involve discontinuities which may require special treatment. It has been pointed out, however, that a term with argur
VARIOUS METHODS
2-4]
f f,
...,
and
in
85
(1)
small.
It
is
this
stitutes the central difficulty in the development of the disturbing function: a few coefficients of a given order with respect to
the eccentricities and inclination are needed to a higher degree
The problem
same
order.
is
used
several
is
to express
'
it first
in terms
-t-
cos/
'
1 -h e'
cos/'
second method
is
= a(l
ecos X),}
ro,o$f=a(cosX e), f
2
r sin/= a Vl
e sin X, )
............... (2)
#
Still
another method
anomalies from
is
/, /, /' by
the later sections of Chap. in.
?,
86
is
[CH. iv
used,
t is
g'
may be
= n't + e'
able chosen.
4*5.
When
the time
is
e,
~r Aj,f cos
is
+ jy ) + ~ Bu sin
(jg
-f
//),
where j
A,
(jff
On
much unnecessary
case should be
fit all
calculation
examined
is
Each
some
detail, especially the calculations needed for the long-period terms, that is, those for which
jnj'n' is small, and that plan adopted which would seem to give
the results needed most efficiently for the case in hand. Famili-
method
in
is
to
available
and advantage
is
IN
4-6]
87
4'6.
tricities.
=
proved in
2*7,
p=
d
__
exp.
dx
E V- 1,
Id
_=
so that
............ (1)
by putting
.
*,
x = exp.
un
p =
^
\f-ld^
ty>
d
exp.
r h,
cfyr
and
F(exp.
V-l).
#-7-7 -f\exp. ^r
...... (2)
care of
all
Put
r
=a
function of
when e = 0,
r</
e,
e' = 0,
e,
e'
and
= a'
function of
e',
r /r
'
= a,
r p,
r'
= r y,
..................... (3)
is
Again, regarding
and putting
............... (5)
88
[CH. iv
G xp.
^ V^T)
Then
and
for
two
variables,
Again the
(^f).
'
<ty
...(6)
< 1,
The
1-
ri/
v-
pi-*-'
1)
21
+ ...,
...(7)
where
The
latter
form
is
valuable chiefly
when
e is used, since
1
expansion of p'"^" which should be
made
in the form
(9)
The harmonic
analysis
is
l/p',
logp'/p',...
functions
IN
6-8]
89
statement
inclination
true of r',/'.
is
is
2
factor cos J7 in the functions of
(cf.
is a homogeneous function of r, /.
proper use of the fact that
With the methods given in this chapter, the development takes
Expansion of
I/
along powers of
e, e'
and multiples of
'.
r, r'
1+ccos/'
As
in 3-3,
put
= exp./V^l,
(I-.?
of the expressions,
+ e'cos/'
2
e(l
+<rj )
2 2
)
--
2<rj,
and
let
-fr
with similar expressions for r',
Thence, according to 4*6 (4),
?</, //.
where
Ai
=1
2a cos
-I-
The expansions of p D p
coefficients
~D"1
=-^,
r
u-=a~.
...(4)
</a
arguments
on
depending
positive integral powers of JD
,
f,f and
>,
90
[CH. iv
s, t,
x, respectively.
is
<f>
Hence
...(5)
with
= 0,
1, 2,
..,
when^ =
Similarly,
where
tively.
This
is
The
It
is
a
"
a'
-D-2,j + l,
and multiply the result by (1 ?;2 )/( 1
work to this formula, w-3 have put
7/
/1X
if),
...... (2)
In adapting tho
).
-V ?
s
8-13]
If
is
to
91
be used, we put
(4)
with
F-F(l-D,D,j l,-*),
F' = F2 + D,-D-l,j+l,
f_
/2
. .
.(5)
The operator
4*10.
render
its
^1
= D 4- ^
powers of A, that of
.,
A and
of the operator
F has been
made
substituting
77'
for
TJ.
namely
a/a', is
arises
when we have
to log a.
Now
.................. (1)
92
where, by 4*1
cos
[CH. nr
(3),
.(2)
v',
+ v'
20,
and
this
might
be achieved by first expanding into a Fourier series with argument S and then expanding cos iS into sums of cosines of
multiples of these two angles. More rapid convergence with less
computation can be obtained by making the development depend
With the
definition (3) of
sin
which
A
2
we have
1/ cos
(?;
/- 2(9)) ~*
(v
+ v'
20) are to be
all
practical
needs.
Define RI,
R9
...,
by
AI
We
3, 5, 7, ....
then obtain
= RI + RS tan 4 II + J R 9 tan 8 \I
+ v - 20)
+ v' - 2(9)
f
(i;
\,
(t;
+ / - 2(9).
...... (5)
of
coefficient proceeds
4
by powers of tan J/.
13, 14]
= /+tar,
93
The development
angles
f-.f+n->v',
/+/++' -20.
w -w', v+v'-20,
/',
and
4'14.
The development
of the RZS
to the
third order.
is
form
= W>COsi(/-/ + *r-O,
/8f
->
=&<>, *=0,1,2,....
...... (1)
The
coefficients.
^=
( F!
2/3 4
(t)
where
v'-O
(D +
1)
+ ip/ D(D +
I)
cos/
'
'
94
Fz =
[CH. iv
cos (j
+
= w+w _
'
20.
The double sign means that there are two terms each having
the coefficient set down.
The
final step,
in FI,
cosi(ff' +
FZ by
= /38
r,
and
-f VT
.
r')
receives
This is equivalent
each of the angles in
out.
to
The term
(l)
requires no treatment;
integral values
be carried
to
is
(ff
independent of
positive and negative
in FI
all
zero.
(6),
-f
C)
-f
E'
= g'
-f
2//
sin tf,
= exp.
(#
1 + E'
|>)
cos (jg
+ $<f +
C).
...... (1)
The exponential
expanded
powers and products of Ed/dg,
the
and
this
requires
expression of powers of E, E' as
E'd/dg',
Fourier series with arguments g, g
the operators d/dg, d'/dg
act only on the explicit functions of g, g' and not on E, E'
in
is
The same
cos
result
is
reached by writing
and expressing
*
*
sin*
j,
j'
r,
7ih.
COS
.
sm
arguments
needed.
The
..
Ji
&~,
14, 15]
95
monic
analysis.
is independent of the
Properties of the expansion. Since
directions of the axes of the frame of reference, it is independent
of the origin from which the angles used in the expansion are
measured. Hence the algebraic sum of the multiples of such
angles present in any term is zero.
Thus if w w be the mean longitudes, and
f
CT, CT',
we have
The
F, 20.
+ i' +j -h ;'
if
6 the longi-
any argument
+ 2hV,
=
+ 2h 0.
the
in
............... (2)
terms of g or v
-GJ
is
-CD-'.
has the same properties. Further, the expansion 413 (5) shows
is a d'Alembert series with
that
respect to F, 26. It follows
But
'
of R
is
to
coefficient
equal
of w, w' present in that term, and the
multiples of g, g' when
we put
sum of the
same
&, w'
is
g' -f
multiples
true
the
for
r'.
may be put
Jt
or
2 coBJf(CJtf cos//'
4-
j,/ =
0,
1,...,
8Jtf sin//')
'jj,
j,
/ = 0,
M,
96
[OH. iv
f
The
to multiples of
is
j,
= 0,
carried out
1, 2,
by using the
f=g + E. By
additional
work
The value
4*16.
of r in terms
r=a
off is the
series
aecos/-f
Actually, this gives series along powers of %e, because the long
period terms always arise from the expression of the product of
two cosines as the sum of two cosines; only one of the latter is
needed more accurately. The same is true of r'. But when we
substitute for/ in terms of g by means of the series
we
of magnitude for the series giving r in terms of/, and /in terms
of g. Further, many of the actual problems are those of asteroids
full
degree of accuracy,
much
The
4'17.
For
where
this calculation
it is
15-18]
97
to sine in (1)
___ ___
An
...... (4)
alternative to
p'
2 in
411
(1)
by
D -f 2
must
and in
of course be
made
is
4*18.
This is sometimes needed to a high degree of accuracy. According to the theorem of 2*6 the constant term in the expansion
of I/A in multiples of g, g' is the same as the constant term in
the expansion of
1
dgdg'
A dfdf
B&SPT
rV2
A aV VT^T Vl - e'
a
98
in multiples
of/,/'.
p = 2 is indicated.
The use
of the formula
41 1
..
(1) with
[OH. iv
p = 2,
/,/'.
vol. 40,
pp. 35-38.
of g
equation
X*=g + emi\ X,
From the
in 3'17.
tan J/
relation
T~~
the special values of /and thence those of any function of /are then obtained. As / g take the values 0, 180 together, there are only 3, 5 or 7
special values of / to be computed for each planet. Methods for analysis
into Fourier series are given at the end of this volume.
of
The functions of /needed are cosjf/j shy/ for a number of integral values
It is more convenient to calculate cos^'
(fg\ sin,;' (/ g\ and after-
,;'.
wards to deduce the expansions of cos.;/, ainjf by the use of the factors
sin^.
Most of the asteroid problems require the calculation of the perturbations
by Jupiter and Saturn only. The series for cosji/', sin j/', once computed
for these two planets will serve for all cases small changes in the values
coajg,
E.
any
18-20]
99
4*20.
The
6, /.
See App. A.
XV
% = exp.
1,
where
(9),
is
<
= exp. / V^HT,
it
has been
found that
r',
</>'.
If
we put
variables.
Now
Hence,
by
I/A
if
'
}
is
equal to a function of
A^. ?V be what
%, %', the
r/r',
<f>,
<f>'
divided by
r'.
</>'
theorem gives
xV-v'x'r*-
1-*'
....... (i)
It gives
'
...... (2)
7-z
100
The
result is
The product
-D-
/, /
putting
it has the same properties.
Since /, 6 are present in A^ in the same
1,
*;',
for
',
D,
is
[OH. iv
1, for
then
real Fourier
obtained by
B, x> j respectively in
rj,
(2),
and
way
as they were
present in AI, the expansion of A^ along powers of the inclination follows the same plan as that of
In fact, if we
x in 4*13.
X+
X +
=
=
*' in 413 (3), 4-13 (5) (taking note of
er, v'
put v
the different significations of r r ', a), the results can be used
1
Newcomb has given (Astr. Eph. Papers, vol. 3) & detailed expansion of
the disturbing function in terms of the eccentric anomalies, certain portions
of which are taken to the seventh order with respect to the eccentricities.
He
uses an operator but did not obtain the general formula which permits
any coefficient to be written down at once. The latter was given by one of
us (E.
W. Brown,
D and
certain integers
improved form
4'21.
to
mean
anomalies.
and
cosjX = A
sin
or, if
where
i\
H-
jX =
= exp. g V
-4
or
S< 1,
*=1,
2, ...(1)
tf-A + Z.ij^ice).?,
K
\e according as j 1 or j = 1
=}=
(2)
20-22]
101
by adding
- \e
to that in terms of
coef.
of cos X - \e'
ee
f
,
the terms
coef. of cos
X'
-T
X,
sum
of coef. of cos
(X
particular periodic term and that in 4*18 for the constant term can
evidently be applied to the transformation from eccentric to mean anomalies.
But the
A'',
with
coefficients in the
terms of
same.
It might be thought that the change from true to mean anomalies is
more complicated than that from eccentric to mean, because we cannot use
general formulae like the Bessel functions to make the change. As a matter
of fact, the actual labour of making the expansions differs very little in the
two
cases,
whether
The developed
literal or
series in
A point
tion.
102
[OH. iv
series for
of g,
cf.
'
converging
series.
in
anomalies in the form given in the text, consists in the fact that it is the
only method known by which any coefficient in the development of the
disturbing function can be written down from a general formula it contains the operators /), /?, B', and the Bessel functions. The highest power
;
The order of a Bessel function is known from its suffix. But the
formula suffers from the defect pointed out in the paragraph following
3*11 (6) for the case of the general expansion of/ in terms of g, namely,
former.
that
^'-f-1
part of order
this defect
The functions of
4*23.
the
major
is large.
axes.
The development
where
and
0,
for 25
definitions of
provided
|a|
1,
2, ...;
2*=
the numerator
1,
1,3, o, ...;
- (1)
()
&<->-&<>;
of the fraction is
unity.
The
T/T,
<
practical
This definition of the coefficients without the factor J is so much more conwho have followed him that we have
it
22-24]
103
deduced from these two by the use of finite formulae. The two
coefficients to be first found will be those for s =
i= 10, 11,
,
for reasons
which
will appear.
The more
calculate
/%" =
&
^i>
(1)
| (*i
where FI, E\ are the elliptic integrals of the first and the second
2
7=1. The tables of
kind, from the tables of Legendre, with cos
Runkle (Smithsonian Contributions, 1855) give certain of these
coefficients for different values of a; those of Brown and Brouwer
(Camb. Univ. Press, 1932) have higher accuracy.
424. The
series for
Define
i,
&>.
by*
aeosH/
1+ai2=
+ rf
= cosi/,
............ (1)
so that
(1
-f
-2
cos
^)~
s
.
.(2)
'
i)'
= at
F is
where
and
The
finding
(a*)-*
illl.Lr_
rtl
/'
summation
0, 1, 2, ...,
2 cos t>,
...... (3)
that for
in (3)
A from
l
sign of
*
^- 2 cos
sin
2A t
K sin
2A
being F(s,
= tan A\
aj
2
s, 1,
a = tan A,
).
and
104
[CH. iv
so that
in
which
This
last
s-1
,( + !)
(s-l)(s-2)
i,
evidently useful
since in this case the earlier coefficients of
<
\p\
that
1,
is
when
is,
is
that the
<
1,
that
is,
of p
PQ where p Q =(=
4*25.
0.
For brevity,
(1 -f
with
let
2
us put
- 2a cos 2
(i)
fi
an.
a -, so that
t
/ cos ^r)~*
= a + 2Sa
cos
fy
...(I)
The
values of
F(pQ ) F
t
(p
pQ
series:
4, 26]
105
hypergeometric series*.
2
On putting (1
for the cases
j9
)^
=0,
J, 1.
/a~v
46,189
-A/
V tea
/ca
-02272 727
-00213 068 p 2
- -00034 15
p3
-00007 47
- -000020
-000006
*\
factors 46,189
.
17
of the series,
multiplied
by
r
J one
\4/
+ 0-98913047
+ 1-00000000
-
13
+0'97912 120
-01926520 (/>-!)
'00141681 (V-l) 2
- -02082065 (p-i)
+ -0017101 (p-V?
- -0002304
(p-l)*
jo*
-000041
(/>-4)
p*
- -000009
(jo-i)
;>
(/J-i)
-000002
-00016 594
(p-
1)
- -00000451 (p'-l) 6
+ -00000091 (^-l) 6
- -00000020 (p-l) 1
1)
-00000005 (^-l) 8
00000001 (p-1) 9
10
au
176,358
+ 1-00000000
-
-02083 333
-00180 288
- -0002683
+ 0-99000
-
p
p'
-00005 48
p
p
-00001 4
'p
00000
The
<
series give
-82.
The
a!
=
=
/<*ip(<*i
-01920 796
(p-i)
+ -00146840 (p-b) 2
-0001854
-0000311
-000007
(p-)
-000002
(p-$f
aio,
an
to
(p-l?
Q0-I)
4
5
+0*98074 527
- -01786420
+ -00123022
- -00013597
+ -00001956
- -00000 335
+ "00000 065
- -00000014
+ -00000003
- -00000001
(jo-1)
(p-1)
(p- 1)
(p-1)
(p-
(^-1)
()
(jo-1)
(/>-!)*
(p-l)
when
table
-45
-61
'66
'71
'78
-82
'88
'25
-50
75
1-00
T50
2'00
3'00
i.
For
The numerical
series
ax =
*
1)
459-465.
given in the text for2? = |, 1 are taken from this paper. Extensions to the cases
p = 2, 3, 4 are to be found in Mon. Not. vol. 92, pp. 224-7.
106
form for a 10 a n
Alternative
The
[OH. iv
tend to the limit unity, but for large values of i the approach
is very slow, and after the first two or three terms the ratio changes slowly.
We can make use of this fact by expressing the series a -\-a l p + a 2
these series
in the
all
form
been obtained
In this
A.
way the
a 10 = '35239,4104
a u = '38183,0736
Sa 10 = (2131
where
These give
figures for
a,
10 , r
p<\
=
l
H to
+ 245/7 - 19/> 2 )p 2
4'26.
Eight significant
+ '00007,469^
829^ 3 + '00005,477/J 4
coefficients
when
275^),
by the expressions,
"00034,1 46/>
'
+ (1 +
Q
.
are furnished
~p p*
(&)
\ K* /
the /3
(<)
are known.
found, there
a,
is
no further need of
i;
K in the form
The
ai
ea
m- a
+1
= /3|
-
( *)
(ca
when two
m - 2a
of
them are
RECURRENCE FORMULAE
25, 26]
higher values of
for
107
i.
1+
~*
I + S
&<>,
......... (4)
to s the value J.
values of two consecutive coefficients for other values of
by giving
The
s are
found by putting
=
-
~~
""
4( 6 _2)
2s)
The remaining
which
^+
'
coefficients are
(2
+ 2-
+1)
"
2s) /9^
may be
effect, tests
When
(11)
/3
(0)
(6) with
(1)
y8^
#f
...,
= 0. A
cti
=\
for the
In general, there is a
in running down from
loss of less
(10)
y88
to /3
accuracy in the use of the formula (5), but this loss is balanced
by the fact that the higher values of s are present only with
the higher powers of the inclination
eccentricities.
108
4*27.
The
first
derivative
with s
easily
= f,
f, ...,
the derivatives
is
of the
obtained from
with s
^,
and
for
[CH. iv
(i)
.
from
&=
11
(D--*"
(3)
= (2*-l)JD'+ l & +
(D + lXA+i
/c
>,,
+ 4,(l--z
To
The
derivatives for s
&&? = <&&?+&/*?.
(5)
The
by
where
cf
as an identity.
a2 -
i.f ...(*-!),
= (a*)*-* 2&
i = 0,
1,
(i)
cos ty,
2, ...,
. .
.(1)
27, 28]
The
a2
sc 8 (1 4-
2/ea sin ty
109
f gives
(*)*"* 2/3
(?)
fi
sin if.
...... (2)
and
s 4- 1 in (1)
Replace
member
of (2). Since cs +i
by
2 sin
The
sum
sin
left-hand
i/r
= sc we obtain
2/3^ cos ty = 2/3^ i sin ty.
member
of this equation
of sines of multiples of f.
we obtain
i\|r,
which
expressed as a
coefficients of
'
is
may be
Equating the
(6).
(2),
4-27 (5).
2 2*a cos i/r and insert (1) in the
Again, multiply (2) by 1 + a
left-hand member of the result. After some reduction, we obtain
Ka
S (i + s) $M sin (i +
1)
^ + *a2 (i - s) /3
(i)
The
sin (i
= (1 +a a )SiA
1)
(t)
sini>.
takes
all
of them, gives
*a
(i
+ s) &* +ica(i+2- s)
which
is
(1
+ a2 -
2*a cos ^)
+2)
+ a2 ) (i + 1) #' +1)
when we put e = (1 + a 2 )/ tea.
S ^^ cos ty = 6-a/t 2 ^
obtained from
s,
yields
(1
when
cos if,
+
)
If we successively eliminate
and
(3),
Change
we obtain
into
1 in the
it
$+i\ff*+^ between
1 is
put
for
....... (4)
this equation
(5).
110
4'29.
The
[CH. iv
by
a,
2s (a 2
1
-f-
^^
o<?n
^ JL
2
^)
a2 \
1*
'
cos
/ca
a2 - 2*a cos
cos ty
2/tra
+ a2
01^ =
Hence
From
2/ea cos
^^+
Ktt.
a2
^2*^2
KCd
this equation,
in Fourier series
obtain
which
is
The
general theorem,
D'{oflf(a)}=<#(D
+ qy.f(a),
............ (1)
The proof
we put
p
=1
+ a2
2a/c cos
yfr,
is
more
difficult.
.................. (1)
we have
(Dp)
(2a
2a/ccos
2
>|r)
4-
dy^
Also
J5V' =
If
29-31]
111
= * (s
(s
-f-
......
Now, the
definition (1)
P~'
and 4'23
-(a*)*- RI. +
C. ................... (3)
(2)
(1) give
eliminate
placing s by s
1, s
re-
of the relations
we
by suitable
factors,
2
(D + * -s)
.
The
final
step
is
which
ity.
is
when
we putj=
The
4*27
The statement in
(4),
may be
to this equation
(4).
justified as follows.
on
Dftg, j3s +i
/8
. .
only; the
112
2
\I (1 + sec
4 tan2
nr
[OH.
D2
/9, f 2 is
/).
for values of j
greater
computation caused by
The
4'32.
of
the
literal
for
its
presence
expansion of I/A
very small
is
to the
*.
mean anomalies.
tf^a^a-t,
will
is
= sum
i,
including
zero, of
(1
T?
77
= 1,
were given by
31, 32]
The transformation
the theorem
415
The expression
77
- exp.
for Z? is
E=
4*77
8??
With the
8??
cos 2(/)
f'=g' +
of
+ j'g' +
0),
E'.
Expressed in terms of
given in 3*16.
-2
made by means
cos (jg
(EJ-+E' g2>)
f-g + E,
where
is
namely,
(1),
(tf+j'f + 0)
cos
mean anomalies
to
113
it is
sin
for
-f 5??
E'
sin
2jr,
From
EE'=8w'
these
{cos (g
we deduce
- #') -
cos (g
+ #')}.
exp.
tf
3
/
we
=
*
{1 -f
x sum
for i
1,
0,
2, ...
of
7?
2
(D + 4D -
H- 2^7 (
D+
4i a )
2t')
iw
4-
g)
+ g')
+ 27?' (D + - 2i) % cos (iw 2
2
2 4i
3D
4iD
+ 5i) a< cos (iw - iw' -f 2#)
-f
+
(D
+ (D2 - 4iD + 5Z> + 4i a - 9i -f 4) a, cos (iw - iw' 4- 20')
1
I'M;'
?7
/2
7/
w, w'
= mean
longitudes
=g+
r,
^' -f
w'
are used.
B&8PT
114
[OH. iv
{1 H-
a2
2a cos 2 / cos (w
{a cos /
(1
+ a2
tt/)}"~^
= 2a,- cos f (w
2a cos2 |/cos (w
w')
.(3)
= 26* cos i (w -
w')}~
w')
...... (4)
where
= 0,
2,
1,
These definitions
77
= ^, 77'=^'.
differences
between
Tisserand* in which r
'
rQ
a'
= a.
D*
replaced by
r\oa/
.-,
cos2 A/= 1 in
rf
will
the mass factor; the only condition actually used in the developis that accented letters shall refer to the outer planet.
ment
factor,
r cos
r cos
~/r~~ or
is
>
+ yy + zz*
>s
The equations
xx'
4-
yy'
4- zz'
>
^3
1, p.
309.
32, 33]
If
Since
a;',
latter,
y', z'
115
n'dt
we
do not contain
$r
and
/-t
= n 2 a8
x, y,
may
and
be written
is
v,
the formulae
is to
310
(5).
easy
mean anomalies by
way
S = A'
rr' cos
A'
ra ,
The
An
and these
r' sin/',
B' r sin/
cos/
aiT^'cos. 9
T r cos ,
2
= cosH7
+v
v
(r v ') + sin A/
2
^
r sm
r sm v
.
,.
B'
-20).
'
analysis, or
in series from
A' = Cx
r'
cos/'
+ Oa
r' sin/',
JB'
- ~ C3
r'
sin/'
+ C4
r'
cos/',
where
C 4 = cos2 J/sin (r
f
<72 ,
r')
sin2
/ COS (tsr +
2
w') T sin \I sin
r'
- 20),
+ *? - 20).
8-2
116
[OH. iv
The expansion
of
i=l,
value
- %a/c,
ai
we
find
so that
from
413
(4),
414
^i
Finally
a/c
cos (v
if this
(5)
that
(1),
becomes
a* tan 2 \I cos (v -f
i/)
......... (1)
v'
20)
D=
=
,
a cos
we have
we
of I/A
by
rcosS -Ai/
-= acos
-7a -{(l~27;
~7a2
(/ .
rt
-f
?;
COS (2g
/ax
2
,.
/
iL
^' -f cr
isr')
3?;
COS
(</'
-or
-f tir')
-f
tar
flf'
tzr
of
r'
-cr'
'2
tar
cos(5r~3/-f ^ --cj')
- 2(9)}.
by
inter-
CHAPTER V
A.
5*1.
be
d2 ^_atr
-.190
order.
We
shall
now
six
>
The
kinetic energy
Differentiating (2)
is
and substituting
%-%
From
(2)
Since
CT
(5)
may
and
(3)
in (1)
'-'''
we have
HTU,
dt
we have
<
respectively, (4)
and
118
[OH.
function.
On
= (ft.5ff,
2(<fo,.8y<-dy.&,)
(7)
SH
where the
is
d, S,
made more
precise in
5*3 below.
5'2.
Let a i9
differential equations,
dxt_m
dt'dyt'
where
is
derivatives.
variables
dyi _
dH
dt~
a*,'
xit y i
to
2n new variables,
p it
JIT
dt
is
m
W
ic it
y,, t only: it does not contain any
contact transformation is a change from the 2n
dp,
is
>
a function of
equations of the
where H'
*- 1 >4-> w
related to
dq,'
dq i=:
_dHr
v ~'
'
dt
dp,
expressed as a function of p iy q it t.
According to a theorem of Jacobi, relations
fulfil
= dS
dS
S = rfunct.a?
Pt^faS
function 8 must thus be
ct
j.
/o\
,9 < ,t
(3)
j.
g^>
<
The determining
expressed as a
function of one set, either the x t or the y t of the old variables,
and one set of the new. It must be so chosen that it is possible,
,
No
development
S and
that in the
CONTACT TRANSFORMATION
1-3]
119
.(4)
formations
is
information. It
may be
to have first given the canonical forms of the equations of motion (Brit.
Ass. Report, 1834, p. 513), Lagrange had given the equations for the
variations of the elliptic elements in 1809 (M4m. de VInst. de Paris, p. 343)
The theorem
in this form.
Rendus for
by Hamil-
we
shall give only those which are necessary for the later developments in
this volume. In particular, the proof of Jacobi 's theorem, given in 5*3,
docs not indicate the process of discovery, but it has the advantage of
$,
for
new forms of
canonical variables.
5*3.
desirable to define in
form
i
and the
=x
i (*>>
%>
<%>),
Vi
Hi
(*>
a l>
'
' '
>
a 2)>
the a ry
it is t
alone which
is
in terms of
and
symbol 8 implies
120
that any or
all
of the
t is
[OH.
not changed.
expressed as a function of x iy q iy
is
In these expressions
t,
commutative
for all
values of
*/ j
j
=*
2
2
(yidxt+pidqt)
x
fiS j
Ac,
dS
+ g-
dfif
......... (2)
- a/S
d^J
-^
d$,
......... (3)
S may
contain
since
X (dy
Set -
dc, 8 y< )
We
obtain
/dS\
JACOBFS THEOREM
3, 4]
which
The
is
121
the same as
*(t"--$*)-'(*+D
Finally,
if
we
define
in terms of p it q it
t,
by
^
dt
...................
'
~dq t
dPi
dt
that dS/dt
H'
is
B.
to
be expressed in terms of p^
qi,
t.
such choice
is
the following.
Suppose that
make the
identically zero.)
become
zero,
The equations
5'2 (2)
= const.,
qi
= const.
122
when S
is
[OH.
so determined as to satisfy
Notation.
where
1, 2,
be present, thus
t)
means /Oi,
#2,
-,#> 0>
o
5*5.
its solutions.
But
yi
H was
= dS/dtti.
by
5'2 (3),
x iy
as independent variables
and
When
all
that
is
necessary
#,-,
y if the constants
JACOBIAN SOLUTION
4, 5]
Theorem.
123
is
dS
where
p it
#,
is
an
integral, contain-
The constant
additive to
differential equations
its
presence
is
and
through
appears in the
but
its derivatives,
may be pointed out that the ordinary method for the solution of
order partial differential equations simply leads back to the canonical
equations, so that nothing is gained by attempting to use it. In the
It
first
form
S, containing half of
fore,
may be chosen
arbitrarily.
The
124
5 '6.
[OH.
explicitly.
start
S^Si+Ct,
........................ (1)
is
reduced to
<-w +t
equations.
This
may be proved
by multiplying them by
dyi/dt,
canonical
dxjdt, and
adding.
similar procedure may be adopted when any one of the
coordinates x t is absent from H. It is evident that each absent
down
of
an integral of the
equations.
5*7.
Application
to the
perturbation problem.
where UQ
equations completely
when
R = 0, and
it
+ J?,
is
replaced by
UQ
R, or
VARIATION OF ARBITRARIES
6-8]
R if HQ = T -
by HQ
126
written
-=
dxi
dy t
rr
-^(H,-R),
p\
(if
-^-(Ht-R)
/i\
(1)
the relations
as
ds
where S
is
,.
ds
5*2
satisfied
dt
for in this case
we
dqt'
dt
dp^
have, by (3),
/*-/*
,
da__x
^
^+g-
The
p iy
when 2? 4= 0,
numbered (4); hence the
variables
those
Constants.
and error
if it is
we know
5 '8.
in
R=
new
variables,
Osculating orbits.
126
[CH.
the solution of 5*7 (4) has been solved, so that the variables
p it
and
tersect at
=s tQ
intersection;
when
this
R=
is
tv,
to
In the great majority of cases arising in the solar system, the forces due
R are small compared with those present when 72=0, so that the oscu-
one revolution of the planet round the sun. Thus the osculating ellipse
can be used to predict approximately the place of the body for some time
before
and
=t
(}
Ordinarily two coordinates, which give the angular position as seen from
the earth, are observed neither the distance nor the velocities are directly
observed. From three such observations an osculating ellipse can in
;
general be deduced.
position predicted for some other time in order to
limit the area of search also needs only the two coordinates. There are six
constants present in the osculating ellipse, and according to the mathematical theory of the approximate representation of a curve, considerable
variations may be made in the six constants without altering the two
orbit
its
determination.
Since the curvatures are not in general the same, the word 'osculating'
not used in the same sense as in the theory of curves.
is
S, 9]
C.
127
5*9.
method.
The
dxi
dt
__
dHo
dyt
dt
'
-
9y t
9#o
'
dxt
where
The
division
by
is
m=
According to
MIS)
5*5,
+l
w-.,
Transform to the
is
+*,+*,)
tri-polar coordinates
TT ~
r,
^
i, X, so that
is
dS
,.
,~.
convenient to put
is
independent of
by
2r^
X,
in (2)
t,
we
--
and
.................. (3)
i,
aa are constants.
and rearrangement,
....... <>
The form
solution
128
[CH.
2i
2
2
2a
4-
=-
/9$3\
.
=-=
= a* 2
\dL )
2
3
^-,
UU&-JL/
where
TI is
2ai-f
*
and
let
may be
when
vanish
positive,
it
is
//
2jjL
-^
/M
= 0.
a22
a
/V1<fi
= 0,
/f .
,,
(5)
^
'
'
we
...... (6)
The next
step
is
= 9S/3a$, y i 'dSldxiy
equations by means of the relations
which would seem to demand a return to rectangular coordinates.
However, we do not need the latter set since the former gives the
necessary three relations between r, L, X, t, and the relations
between r, L \ and #1, #2 #3, are independent of aiy 2 3
The derivatives of 8 with respect to the a f are obtained
without carrying out the quadratures in (6), by means of the
&
formula
The
ft
gives
9, 10]
Similarly,
129
L
'
The
and the
and a or n
fit
a,
/3<
e, i, e,
con-
w,
6.
equation,
According to
5*9 (7),
/2i =
nt
a*
(1
the value of t
shows that
is
/33 is
2
),
giving c&i
e).
r.
when L = 0,
reference. Hence
the value of \
that
y33
5*9 (9)
/3j is
2a,
is,
= 0,
maximum
we put a3 = a 2 cosi
in the last
term of 5*9
Hence
(3) it
becomes
L
cos
Jo (cos
But
LdL
Z - cos2 z')*
__
~
CL
Jo (sin 2 !
(sin
L)
2
sin i)*
~ &m ^
__
L\
am?/'
/sin
\
130
[CH.
is
$2
=&
5*9 (8)
of canonical
elements
i=(e-
2
or,
a,-,
& when R
- d& 8a ) =
(da, 8&
f
4= 0,
eft
forth
above,
the
are
812,
(2)
written in extenso,
d&
<ft
5*11.
The
4&
9a t
eft
manner.
aecos A" in (7), and insert the
If we put r=a
found above, we obtain, after integration,
in the following
n
which, with the aid of the relation a?n?=p,
necting the mean and eccentric anomalies.
Equation 5*9
(8),
is
[X
since
6,
r,
becomes
dX
l~e cos
v=v
'
Z = sin i sin
sst X - tan*"
since X
(cos i tan
i/)
is
is
elliptic to
nature of the motion had been found by the easier method of Chap. in.
(Cf. 3-2.)
DELAUNAY'S ELEMENTS
10-12]
D.
131
5*12.
elements.
S = 2a,A,
........................ (1)
where the
set,
I,
g, h,
defined by
so that
Hence
= /*(-
2ax)~^
t.
new
variables be
5'2 (3),
/
- (?==
^ 9$
d$
^
i
ir = _ =/t (_2a
1 ri=V M a
/-
L 0, H, we have, by
# = ^S
^=
rr
3.
...... (3)
The remaining
H'-H +
Thus
this set,
which
is
-R-*-R +
............. (4)
equation
where
L = V'pa,
= nt + e
...... (5)
tsr,!
...... (6)
cos
9-2
132
The
H are angular
L, 0,
been divided
common
factor
"Ja/j,
becomes na
5*13.
If
[OH.
we put
If
/JL
= n 2 a8
the
2
.
set.
Ll + Gg
+ Hh,
in the
form
HG
by
c i?
d=
=
c3 =
c2
it
and
them
so that
= nt -f e = mean longitude,]
w 2 ~ TZ = n g- of apse,
v^a,
Vyu,a
ivi
(Vl
(?
e )
\/y^a (1
1),
(cos
long, of node.
w^
1),
...... (1)
The Hamiltonian
When
divisible
expansion
by
is
unaltered and
is
made
in powers of
e,
function
and
is
c$
make
properties which
i*
by
-h yu, /2ci
equal to
i, the element c 2 is
this set
This set
set
given by Poincare'*.
p2 =
jt) 3
\/
w,
q%, qs>
2c2 sin w,
V- 2c 3 sin 6,
where
02, CB
tested
by showing that
defined by
q
</
=V
=V
3
?i
dpiSqz
That
dq^pz = dczBw 2
2c 2 cos
sr,
2c 3 cos
^,
it is
canonical can be
div z Sc2t
f C. A. Shook,
I.e.
in 5-5.
1, p. 30.
tfON-CANONICAL SETS
12-15]
that
C2
pz
two variables p2
if
w2
in such a
qz are also
is
is
function
is
E -f
/*
/2ci
as before.
V/t
when
fa are related to
canonical variables.
The Hamiltonian
This set
manner
133
e siri
-GT,
V/xa
2
.
sin 0,
V/xa
e cos
f/ 3
t*j,
are
2
( 1
cos 0,
when i is neglected, so that the disturbing function is developable in powers of p 2) p$, 72 </3> such a development replacing
powers of e, i, and cosines and sines of multiples of -or, 6. The
,
of
possibility of such a development depends on the association
i
with
of
and
of
of
with
of
e
r,
multiples
powers
multiples
powers
of
5*15.
ex
The non-canonical
in terms of
set a, e it PI, w,
T!
r,
0,
tsr,
and the
r',
?0A0re
M;
=H
= (1 - cos i) Vl -
4- e,
2
.
...... (1)
The
by
=V/ua,
variables w,
wi,
u'2,
c2
w^.
or,
c3
by the equations
c2
=- iei
are the
V/xa,
same
c3
134
[CH.
The transformation
The
process
is
= p/ncP.
member
After rearrangement,
we
S0
2
multiplied by p/no
Since /-c2/2ci2 = /x/2a,
.
in terms of the
new
we
have,
if
be supposed expressed
elements,
-^
-2tt
3o.
o
2
,
-^-~
mu
SlV
oiv
which
da _
~
2 1 dn
IUL
_
dt
nadR
/M
dw
^6
2?ut 3JB
dt
'
naTidR
Jiae^dR
da
%IJL
dei
p,
3Fi
nae^dR
9-or
d9
naT^R
dt
we obtain
2na dR
dw __
=
~~
3 n dt
a dt
...,
90
p~~
dw
na dR
'
The objective in this transformation is the isolation of derivatives with respect to a, so that the operator D, which plays so
large a part in the development of the disturbing function,
may
NON-CANONICAL SETS
15, 16]
135
Chap. iv.
Since R has the dimensions, mass divided by distance, and n
has the dimension of the inverse of a time, the factor p/no?
the equations to relations between ratios.
The right-hand members of all the equations except the second
reduces
all
The
is
integrated.
relations
*
= *l(l-l*M
2,;^ 1 (l-iV)-*,
......... (5)
substitute for dR/dei in the equations (4), the value derived from
dRdT
/dR
^
-f-
f=i
01
\ce
^~
oe
is formed from F
For dR/dTi we have (dR/dT)
when dT/de
5*16.
This set
4-
de
DLL lda\ f a ?e/ de\
= I\ (1 e2 )~^ and
2
(1 -e )-*.
F
set a, e y
is
(4)
or
i,
or
by means
e,
w,
6.
of the relations
__
a dt
dw ^
3n
.
_ dR
_
fl
dt
da
,-
'
e*Jl
___
____
______
l-fVl
na
dR
_
A6
de
- _ _ __
IJL
._.
dw
/A
2na
r>
dt
de
_
^_
dt~~
_ V1
na dR
e A/1
--"
&
na
__
_j_
F
Vl -
dR
na dR
__
P ^F'
136
dfff
__
i~,
~dt~
na dR
vfT:^ 9F
[CH.
nadR
^e'de
'
30
This set
which
The
is
slightly
61, i
replace w,
definition of 61
respectively.
is
usually given by
nt
-f-
n dt -f
means
ei
of the equation
and
in the co-
dR/Sw = dR/de.
The
substitution of
first
F=1
power of
cos
for
i is
advantageous because
i occurs in the
5' 17.
a fixed
An
relations,
di
-r,
dR
.,
dR
to the
distance
from
origin.
16, 17]
137
ellipsoid.
The
mR
Let
$fj,mr
denote the force- function for the remaining forces which act on
m. If #1, #2 #s he the rectangular coordinates of
referred to
force-function for these ideal cases
is
>
motion
be
will
=x +a
When R = 0,
with
>*
with period
centre
is
2?r/n,
at the origin.
set,
is
1Q
If
=e
y=w
-or,
be replaced by
the replacement of
The proof
of 5*13
may
is left
also
by
nt
nL
to the reader.
be made.
0,
r,
/A is
a
replaced by n
ff.
is
CHAPTEK
VI
may be
change of variables such that the new variables are more nearly
constant than the old ones, is used throughout. But the application of the idea is different from that which Delaunay made to
the solution of the satellite problem, where the changes of
variables were very numerous*. Here it is shown that one
change of variables
is,
new
variables requires.
Particular sets of
vol. 3,
new
Chaps,
1-3]
139
latter
them
is
a linear function
oft.
of
given in Chap. IV
may be
used.
63.
In Chap.
IV,
replacement
of the time and the elliptic elements, the relations between them
is unchanged if
being those referred to in 6'2, the development
t is
n't
+ e'
ing planet
is
or <f
not
explicitly present
elliptic,
in
it
is
140
[OH. vi
canonical elements
d = V//,a,
Also,
and
it is
by 515
evident that
(2),
Ci
e,
difficulty.
515
TI in the developments,
powers of
in
we have
by 515
since,
F by
C;,
(5),
it
c 8/ci)^.
Derivatives with respect to c 2 C3 will
(
implicitly involve the presence of negative powers of e, F, and
it will be
necessary to show that these negative powers disappear
(-2c 2 /Ci)^,
when the
R and
variables
w1}
q 2j
difficulty does
powers of p 2) p 9) q 2
73 (514).
e sin sr, q t = e cos BT;
have, approximately, p 2
that ?o, ^o ftre the undisturbed constant values of e, *&.
,
We
get
suppose
then
We
tzr
of e sin
tzr,
e cos
of the work
is,
D'ALEMBERT SERIES
3, 4]
those in which
141
CQ is
because
6*4.
D* Alembert
The
series.
-or,
of powers
series*.
It is evident that if
g,
we have a d'Alembert
a coefficient
to
in powers of
exp.
the following theorem.
are d'Alembert series with respect
If f,
and the angle x, then the Jacobian,
d
is also
(A
a d'Alembert
Let
=A
exp.
xV
x)
2A
1,
\dA dx
to the coefficient
=A
A,
x.
__
exp.
xV
1,
yz
= A2
It appears that d'Alembert was the first to notice this property of the disturbing function with respect to the eccentricities and longitudes of perihelia, in
his memoir, "Becherches sur diffe'rens Points importans du Systeme du Monde,"
Sc. 1754.
142
The transformation
the variables
[OH. vi
A, x
to
y, z gives
2A \ABy
Adz)
8T_
_
a/'N
dy
dz )
d
-,
'
\dz dy
'by
dz
2
The definitions of /, /' make them expansible
since yz =
in powers of y, z and therefore their derivatives have the same
property. The Jacobian is therefore a d'Alembert series.
expressed in terms of
really require to know.
bian
we
when
6 '5.
it is
Other properties of
It.
- 1 with
homogeneous and of degree
respect to
length. The variables GI are homogeneous and of dimensions
(mass)* (length)* or, with the unit of mass actually used, of
(a)
It is
It
is
therefore
expressed by
tzr
txr',
tar
-f
&
26.
It
is
of the velocities.
V?
dR
PROPERTIES OF R
4-6]
which
143
is
which
result
is
dR
dR
dR
When
for ci.
move
in fixed ellipses,
the latter
is
is still
planet.
(d)
ra'/a' in
is fji/r, it
cube of the ratio of the distances of the two planets from the
sun.
consider
R as
first
We
term of
can then
6'6.
Elimination of a portion
In Chap,
iv,
sum
of terms having
the form
R = 2KcosN,
where
N = ji wi +j
w\
= n't +
2 iu 2
e'
g'
(1 )
144
[CH. vr
Of. 4-15.
R=R +R
Put
where
........................ (2)
may
"=jiw+jiV,
= ^/c^
............... (3)
The
elimination of
will
w#,
i=l,
(5'3).
2, 3,
and the suffix zero will in all cases denote that the old variables
have been replaced by the new. Thus
S=
y
so that
is
:E;C t
^o-S,
S=S~smJV
......... (5)
d and
the three
The equations
dS
dS
ELIMINATION OF S
6]
145
and new
01
01
by the
definition (3) of
But when
v.
variables.
Also
becomes
CIQ
whence by
(3)
dS = - ^
^
~ 7
dc
K COS NQ
-rr
~\T
-f
M
~-
~ CIQ).
\
3 (Ci
Ci
function
is
factor.
Ci
CIQ
...f,
The
last
3( Cl
P*
-c 10 )2
2c 104
2cio
2
^ T -'
,
(10)
first power of Ci
CIQ disappearing. Finally, since
c
R^
has the factor m' 2 the right-hand member of (9) may be written
the
^co-H
If,
therefore,
satisfied
9^~~2
terms
factor
factor
m' 2
m'
(11)
2
,
the equations
by the
^2
These have the same form as the original equations, but the
terms present in R have disappeared.
10
B&SPT
t
146
[OH. vi
At this point it is convenient to indicate the general plan of the remaining portions of this chapter. Two choices of the terms to be included in R t
will be made. In one of them, R t contains all the terms for which *>4=0 ;
Rc then contains only those terms which produce the so-called secular
motions of the elements. In the other, R t contains the short period terms
R
developed.
It may be pointed out that the work as far as equation (9) is quite
general in character, no approximation being involved. It in only after this
point that we assume the possibility of development in powers of m' and
first
Let
contain
all
The
z>=i=0;
then
is
explicitly.
satisfy
Since
6*6 (12).
cQ
obtained from
is
merely by
RcQ
0,
CIQ
= const. = ki
for WIQ is
_
dt
When m'
is
Cio
9cio
ki*
"
9cio
and then
M2
^io=r3^-fi,
KI
where
a^,
ki
substituted in
R<$
w = aa,
= a3
c,-
= *i,
...... (1)
are
all
W3o
it
FIRST APPROXIMATION
6,7]
of
jRco will
become
The
constants.
= ki +
c<o
w =
~t,
<
dWi-o
in
which the
The
coefficients of
new
values of the
obtained,
the
be
-< 2 >
T^
dCio
2, 8,
variables in terms of
variables
old
147
are
to
equations 6*6 (6) with the value 6*6 (5) for S. But since
c iQ have the factor m' we can replace c t by c t0 in S
Ci
and in its derivatives. We therefore obtain
-
S,
Finally, since the values (1) differ from (2) by terms having
we can replace WM, c^ by their values (1) in SQ
and
in its derivatives.
=R
where
-f
c
^K cos JV,
contains
the variables c,
all
jV =ji Wi +J2 w 2 + JB w2
terms
for
which
0,
-f ji
= ji
ji
Wi
4- jz
w*
>
the values of
by
cos N,
ij?
i -f
y~3
^i
OWi
+ 2ji
- -5dR c \^
U^ - vS Aa (K\
3ci/
dcAv)
si
sin
... (4)
cos
V
..
S - -5- sm Jv,
w =a - -^C4^ - ^
dCi
oCi
.
//
148
Equations (4)
may be written in
^ = *fl + S
c
d - ki = Sci,
MI
we put
- sinjy,
2
- M 1=
-3
#1
Swi,
*-, *~g
[CH. vi
.................. (6)
Wi-ai = Swi,
. .
.(7)
......................... (8)
is
If
we
define
to order
But
oo
m'
since
21 9R
2
Ci
/x-a,
Cl
-*
A-!
or
-f
/JL
/JL
ki
^oo
Ci gives,
A7
Sjj
J*
2 1 aft c
J*
V
.
.7
of a
is
not
oo
...(3)
1^
A?i
obtain,
,
/,
\
+ r^i-cos^)
"
=oo(l~
-gr
\
OWoodCi
/cj-^
we
/*,
e\
2~.K cos N \
= -M/,l
oo
depending on
2
,
tf
...
in Wi
is
the observed
mean motion,
is
FIRST APPROXIMATION
7-9]
The
terms in
149
w iy w2
w$ depend on
is
o~
K\ldK
(K\
I
oCi \ v /
since n
= jjf/Ci3
The presence
the theory of the long period terms which have small values
The simple manner in which this divisor arises with the
of vjn.
is
noticeable.
elements
If
a,
e,
/be any
From
5*15 (2)
r,
shown by
5'13
function of the
<S'
t-,
(1).
variations of them,
we deduce
(2)
On
a/_
The
l^df
d__
ct,
e1
TX
k$.
sa--
na
a Cl
we
/i ^
\_ty'
the differences of
dc% =*Ci
l_
fic t
ae=
e,
r,
we
_(lz^ /i
\e
find
160
[OH. vi
and
"
df
~~
no?
cci
net? \
d#2
a/;
e(l-erf ?/_
i -j_ (j _ e 2)i 96
df
da
c!e
r
(i
e 2 )^ c*rj
d/1
*
e vfi crj
'
a/
the factor
in
terms of the
when we
confine ourselves to
the
first
power of
ra',
6*10.
The perturbations of
the coordinates.
The
power of m'
"~*
be given by
"'
d'Wi
\dci
The
will
latter
~*\dC{,dtVi
dividci/'
form again introduces a function of the type conThe periodic part S of i/r is a d'Alembert series
sidered in 6'4.
vanish with
e.
functions of
of the solar system. The results in this form usually have sufficient accuracy
during the limited intervals over which observations are available. It is
evident that a continuation of the process to powers of m' beyond the first
SECULAR PERTURBATIONS
9-11]
will lead to
161
of
may
w2y
c2 ,
let
>
dR
dp 2
dq>2
dR
dt
dp%
,-jx
The
it
and i=
The
ct
where P, Q are functions of a/a'. Just as before we show that e and therefore a is constant. To the second powers of the eccentricities, we have, by
5-14,
e sin 37
Since a
put
is
constant
sin v^
e'
=p
f
,
e'
=p
we can use
2 /Ci
e cos
>
w = q. /ci
2
/z,
a, w,
may
are
unity. If
we
a.
{i
With
all
be written
leading to
where
{7,
=-
^2
aP<l
and m'<'001
is
As a contains
powers
e 2 , e' 2 ,
functions of a
of these parameters.
must be expanded
as far as
e 2 , e' 2
in
152
terms in powers of
and
still
[CH. vr
calculation.
arguments would be present. Even with a single disturbing planet and the
calculation confined to the first power of ra', the work becomes complicated
as soon as we proceed beyond the second powers of the eccentricities and
inclination and the labour becomes almost prohibitive in the case, for
example, of the mutual perturbations of Jupiter and Saturn.
In
of
afforded is lost
by the
fact that
many powers
of the disturbing
mass must
be retained.
The general theory of the secular motions of the elements, to which the
solution just given constitutes an introduction, will not be given in this
volume. The reader is referred to other treatises, particularly to that of
Chap, xxvi, and to later work referred to in Ency. Math.
concerning frequently quoted results, which give
limits to the eccentricities and inclinations, should be made. These investiTisserand, vol.
Wiss. Bd. 5.
1,
A warning
gations, in general, take into account the first powers of the disturbing
masses and the earlier powers of the eccentricities and inclinations only.
When
to be considered
6'12.
We
Long period
terms.
first
approximation we
may
sub-
any angle
N in the form
i
K-
wi')
jg
j'g'
k(wi +
<-
2(9),
11, 12]
When
elliptic
argument
values are
used,
= i (n
jn
It follows that if
t
is
jin
coefficient
of
in the
is
of
the
153
ji'n',
n')
k (n +
j'n
the order of
in
n').
which the
be \ji
coefficient
ji
as far
vjn
is
ji
(jiii
= j/ =
we
j\n')
-r
this
Since
it
is
Since
n, n'
continued fraction.
term.
may
If p/q be
= p,
for
any
firsfc,
ji
j\
?&'=
5n -
'483?i,
= '0334^,
=
72n 29n' - -0162*1,
The
third of these
is
2n'
order
order 3;
order 43.
obviously insensible.
154
[CH. vi
make use
whose denominator lies between 9, q' gives so close an approximation as p/q. Thus if q'
q is large, p/q is usually a close
and higher convergents are
is
small
approximation, v/n
very
unlikely to give sensible coefficients. An apparent exception to
the argument is the case of multiples of
when the term with
has a sensible coefficient. This case is dealt with
argument
(cf.
618).
i>
is
inserted
in
Now
e,
ra
occur in the
elliptic term
and the long period term, therefore, gives two
terms with motions n
As v is small, these have nearly the
v.
of
as
revolution
periods. One of the two coefficients is
period
true
longitude
2esin(w1
principally
cr)
The
fact that
would seem
to suggest
when the
OTHER SOLUTIONS
12, 13]
determination
Thus the
is
spread over
many
155
that such terms, having periods very long compared with the interval of
observation, can be expanded in powers of t. The constant parts of the
motion; the remaining portions are usually quite insensible during the
interval.
There
is,
coefficient of the
difficulty.
comparable with the periods of the so-called secular terms (6'8) and that
is not then possible to treat them independently even in a first
it
approximation.
6*13.
written in
0*6,
T*
t>
Here the
rdles of c t-,
= ^c tQ w
-f
--- sin
N.
little
con-
sideration will
this
form of
R = ZK
where
y8
cosN'
+ 2 #. sin -ZV',
= ji^i 4 ji'wi
-f /8,
this form
K K
156
p*>
</2>
pa, #3
and are
function to be used
[OH. vi
then
- sn
cos
#',
where the
suffix zero
rdles played
by
p, q.
6*14.
not taken of every feature which may help to shorten it. The
first step consists of an examination to discover what classes of
methods
will
computation may be
all
the
first
approximation
parts which
Ca,
SECOND APPROXIMATION
13-15]
It will
secular.
157
and
pendently of the
But the
latter.
and
secular terms on the short period terms are usually sensible, and
it is these effects which become most evident in
comparisons
possible are not necessarily the most convenient for the actual
calculation of the sensible terms, so that more than one of
The equations
WIQ.
Hamiltonian
new
^CIQ
CIQ,
(d -
cio)
+R +
S^Tcos
found
N - Stfcos N
CIQ
...... (1)
in which powers of
last
omission
is
ci
CIQ
ra' = 0.
W&
ra'
2
.
We
to this order.
c t-=c to,
have
^=
^,-0
power of
when
ra',
the
and
that, if
f(Ci\ Wi\
to the
same
/be
any function of
)=/(c t0 w#\
;
order.
The
enables us to write it
c if
+2
iy t,
?\f
(c,
- Cfl) +
7\f
(w<
158
On
%KcosN=R
applyiug this to
and
to
[CH. vi
%K cos N
we
0}
obtain
V
V cos
^
2<K
\T
iv
7?
V /^tO ~3$0
- 3$o\
---- SRtQ
= JK
~
tQ + i
i
9w t0
\3c i0
CiO
Since
order
...... (4)
dwtfdcio/
(5)
be made. The second term of (1) evidently has the factor 'in' 2
and may be replaced by its value in terms of the new variables
F +F
2c 102
................... (6)
where
/IT\
2
The application
expressions (7) contain the factor m'
of the theorem of 6*4 shows that they are d'Alembert series and
The
from FI, Ft R cQ
The Hamiltonian function (6)
= jV =
15, 16]
'
159
Ci,
is
it
easily seen to
Ci
" Cio 4.
+
be given by
4.
w*tofl
2
9 $o
9$o
6*16.
namely
(6).
Influence of the short period terms in the first approxithe second approximation.
mation on
F F
,
all
Q cos
Q is a linear function of
Q> where
they have the form
the WM, w\, Wz with integral coefficients, and
Q is a function of
the CM and of the elements of the disturbing planet.
F F F
0,
2
having the factor m'
If the short period terms were again eliminated by a Jacobian
transformation, the new variables would differ from CM, w^o
but
all
2
by terms having the factor m' and with no small divisors
the
value
of
As
m' in the problems of the
largest
present.
solar system
a'),
160
cos (-42+
),
mean motion,
in
which a
is
gives rise to
small and
[CH. vi
is
(Aa)t + A'a'
have the factor
to those in
cQ
with the
factor m'. Thus these terms will merely change the coefficients
of the long period terms by amounts of the order of *001 of their
observation.
ji=ji'-0.
Exception to these statements
fact
may
arise
on account of the
when we
that
2
divisor of order e
can be altogether neglected in proceeding to a second approximation, or at most, only a very few, and those with the largest
coefficients, need be retained. It follows that the long period
and secular terms can be obtained to the order m'2 with sufficient
As
all
short period
of the canonical
difficulty in the discussion
may
be of value.
product
(9/2 w /<to>2o)
'
By
free
acgo 8w>2o
to be substituted
respect to
is
As
a d'Alembert
for
^"SSiUfcii
and
is
therefore
w%) , c^ are
series,
we note
16, 17]
161
functions.
As a matter of fact, even if the method were used for calculation, the
terms would cause very little trouble. For we are actually interested only
in the long period terms present in
and the operator djdt introduces
the small factor v in such terms. The numerical effect of this fact would
be to cancel to a large extent that produced by the divisor e* or r.
,
6*17.
and secular
The
exercises little or no sensible influence on the second approximation to the secular and long period terms, enables us to calculate
the latter as though the former did not exist. Thus the equations
w o become the same as the original equations for cit wt
would have been if we had omitted all short period terms. For
for CM,
The equations
are
all
dci
__
dR
di^dwi'
If
we
dw% _
dR
of equations (2)
The
first
d/f^_dR\_
~ _ V9^_rf/W*\
.......
dcJ
d4 dwl dt \dcj
dt
B&SPT
first
we obtain
__
~~
dt*
dws __ dR
....... ^
dT""""3c3
Wj and make use of the
~dt~~fo*'
i*
+ 2jBcos(^ + "o),
Ci,
Wi in the form
............... (4)
II
162
if
we
i/
[OH. vi
= 0.
fti
Finally, B contains
power of v as a divisor in
which contains an additional part
the
all
first
the symbol 8 (in the case of Wi, the symbol Sw{ denotes the
difference between Wi and its undisturbed value n$ t + const.),
and if the additional part due to the second approximation be
denoted by
$2 > Taylor's
(2), (3),
gives
rs
-7;
820*
at
RR
v
toj-f ^
^^U
\dWidWj
dWidCj
/
d2
-a
5,
1=
ocj ),
1, 2, 3,
/e \
...(5)
\
/
o
= o2,3,
...(6)
rt
>
--
/PX
dCidWj
--r
\dwidWj
Ci
PR
......
1, 2, 3,
Since
(7)
all
a
right-hand members of these equations have the factor ra' ,
be
substituted
for
values
in
constant
the elements
the
may
may
*
SC<=
where
dS
tf-SsinJV,
*S
8c,-,
*",
g^,
/QX
............... (8)
R = 2KcosN,
t
............ (9)
17, 18]
with
N =ji Wi +J2 w
if
'I-*
ji -(- o*
163
f\
\jy
2/3
== u/
c\
/I A\
........ {L\J )
Hence
1
'
v ^C2
v uC$
(11)
Swi
= -35^- sin^-S-^sin^.
(12)
UxLfl
C/jtV/
uC% ==
OCi ==
OCa
Swi
= 0,
&W2=
'^
t -^
8w$
~^.
^
t
OCz
where
The
we are
(13)
,
OC$
~-
2*
dwi dwj
dcj
member
of (5) gives
y
"
2
presence of the divisor e or F, lowering the orders of certain
terms in the first approximation, does not affect the equations
for
This
6*18.
coefficient
of a second order
term.
164
[OH. vi
Swl
3K
~
Whence, integrating,
9
then,
we
S^n
N COS N'
K n*
- sm2^.
2
^i=^i
If,
~T ~2
9jl
j?
write
term in 6*17
&wi =
B sin N,
(12),
we
obtain
................... (1)
StWi^ffsmZN
o
This result
is
B may
coefficient
the
first
P cos
where P,
Q are
expression may
where
j\Wi
/3,
jB
+jiWi
is
all
wj +ji'wi)
functions
and
',
this
N.
-f /8 for
It
is
In
this case
we put
when
two or more long period terms are present. For two such terms
R denoted by cos N, cos N, we have
in
The
result, obtained
W. Brown,
18, 19]
If
Swi
= BcoaN+5 cos N
we have
Whence, on
165
integration,
(2)
Effect of
all
the disturbing
planet.
The
effect of
= _ jx
'
K sin N. a<.
If
we
are given
Swt'
f
where v /n is small,
followed in 6'18,
we
obtain,
*><>'),
by a procedure similar
to that
(1)
When we
to substitute in
166
c$,
That we cannot do
Wi.
[OH. vi
so is seen
by
solving the equations of 6'15. He will find, for example, that the solution
6*15 (8) of the equations 6'6 (6) to the second order will contribute
2
ji B sin 2N to the value of S 2 ^i f r the case considered in 6*18, while
I^Z^sin
2^V,
the
6*20.
The
is
large
terms.
2^
small divisor
difficult.
Now
the derivatives of
with respect to
i}
Ci
will
have been
To
calculate S 2 Wi,
f r
of
we need
the
value of
full
>
In 617
we can
Sivi
first line
Siv$.
19-21]
6*21.
167
effects.
Terms
of the form
= 0.
ji=ji'
(6)
4= 0.
to
= 0,
J!
constant
true of the terms arising in the first line of 6'17 (7).
the
terms
of
from
the
second
line
which
arises
gives a term
part
factored
by
in Sz wi.
an integral of which
x
rise to differential
equations of the
= tk cos (vt
is
tk
sin (vt
+ i/o) +
k
-a
cs (vt 4-
z>
).
We
have seen
Terms of this character arise in S%cit S%W2, B^w^.
that terms of the second order with the small divisor v 2 can
usually be neglected, and the terms with the factor t/v will
rarely be sensible except for large values of t.
168
[OH. vi
,
= ^& cos (vt
4- VQ),
,
The second
of these terms
is
-f VQ)
is
+ 2 -3 sin
of the
(i/
-f
i/
same order
)-
as those con-
furnished as follows.
By
definition,
,<$
is
independent of
ww
t',
The
Ci
= Cio + 9$/9wio,
to the
factor
form
j3Q
+ fat,
where
>
Pi are constants.
Hence, 3$ /9wio
will
GENERAL SUMMARY
21, 22]
169
B is
a constant. Since
CIQ
it
const.,
it
contain terms of
NQ
is
not zero.
Finally, the result is true for any function of Ci. For such
terms can arise only from products of terms of the form t cos
with terms of the form cos N^ the former have the factor m' 2
while the latter have the factor m', so that the product will have
In particular it is true for a = CIZ//UL and for any
dQ
const. 4-Scio,
Wio
= const. -hSw
t-o,
......... (1)
Wi =
T?OO t -f
The symbols
8c#), Sw; include all long period and secular perturbations as far as the order m' 2
.
$=2
In putting
smN
..................... (3)
we have included
of
Cio,
The
is,
in general, given
values of
Since
of
WM
c,-o,
c^o,
WM
by
given by
(1), (2)
terms of
(4).
Thus the short period terms to the second order are found
with sufficient accuracy by substituting in them the constant
170
[CH. vi
by
their secular
is
portions.
development of
literal
first
Integration by parts.
smAr
dt\jf
where
...... (1)
dt\N)
variables be
any functions of
a,
The equation
explicitly.
for
e, CT,
F,
(2)
for
variables, including
~ = 2PcosAT+Q,
..................... (3)
where P,
ra'
in both cases.
Hence P, Q may
= n't +
e'.
It follows that
and that
cos
(N - 90) = sin N,
SPsinJV'+Q.
this
4-
Q.
Since
SPECIAL CASES
22-25]
171
we have
...(4)
where #
is
a constant. Since
P has
the factor
in'
derivatives of
all
ra'.
Hence, the
for
first
approximation to the
is
.r
= a- + 2~"sinN, +
#o
For a second approximation, the values (5) are substituted in
the second and fourth terms of (4): in the third term constant
values of the elements can be used.
The
integrations
may then
The
first
approximation to Wi
substitution for
second approximation is made in a manner similar to those outlined for the other elements.
6*24.
The case of a
Whenever
is
172
[OH. vi
and particularly for those which are concerned with the phenomena
of resonance. But it is doubtful whether it lends itself most conveniently
tions
is
r of the
convergence
e'
series
in 3'17.
reader
and to the articles in the Ency. Math. Wiss. for the earlier literature. For
the later work, references and abstracts will be found in the mathematical
and astronomical publications which summarise the literature annually.
6*26. Throughout this chapter it has been supposed that the mutual
perturbations of two planets can be separated, so that in determining the
motion of one planet that of the other can be supposed to be known. As
factor.
SPECIAL CASES
26,26]
173
No
made
the small mass being relatively small. Thus the integrals are not useful
But when the two planets have masses of the same order
in such cases.
due to the two planets have the same order of magnitude, and it
would seem that this fact should be utilised to abbreviate the calculations.
It generally appears, however, that the lack of symmetry which their use
introduces, causes additional difficulties in the calculations. The more
integrals
useful procedure is that of following the usual method for each of the
planets and later making the integrals serve as tests of the numerical work.
These tests are particularly valuable for the coefficients of any terms of
may be present.
For theoretical work in the general problem of three bodies, these integrals have been much discussed. Since there are four of them, the system
of variables, namely six for each planet, can be reduced from the twelfth
order to the eighth.
shall see in a later chapter that it is not always possible to proceed
by following the process described at the beginning of this article. It
We
breaks down in certain cases of resonance and notably in the case of the
Trojan group. If, for example, we attempt to determine the action of
Saturn on a member of this group without taking into account at the
CHAPTER
VII
vector
is
The
force-function
fj./r
is
+ /j,R,
(1)
""""
__
dv"
(c)
\q/
-,
r
Q == Q
u 2 dv
= r2
r=l
dt
"~"
""
"^
9.
'
_
dv~\p
~^\
~T
==
cost,
(2)'
x
"
~*"\
t-i
~7
~7
\ <5 )
u2
'
dv
dv'
......
W,(
The
latitude
defined by
sin
may be found
directly
2
^-j-f
2
dv
(d
L = sin i sin (v
.-. x
dR
v
^-^^rr ....... (!0)
'
sin i cos i a
llsini^-^-T
sin
)
2
(v -0)u dr
EQUATIONS OP MOTION
1-3]
variables u,
v, t,
It
L (or F,
175
6) are to be deduced
v.
in 1'27 that
M-stt^fal)
=/(?!)
when the right-hand members have been expressed in terms of v. A transformation which renders the equations useful for the treatment of the
satellite
problem
is
q=q
M=z
1 gr 1
4.
so that
=-^2
Wj
""
ix8y
.
'
""
3 "^"
^* r
Dt = ~,^
if
=^ 1
/>
-..
^ ?^
'
Z)y=l-fr/)0,
'
"~^7
^F
'R
V and each
neglected we have R V
This portion constitutes the chief part of the dis-
independent of u\
For
for u*R,
"r
When
/?
+ Wl - q = - --3 /^ - wi
</!
is
straightforward.
is
No
by
this
elimination.
7*3.
solution.
disturbing planet to the sum of the masses of the sun and the
disturbed planet. This factor being always small (its maximum
value
is less
than
equa-
176
tions with
[OH.
solution
gives elliptic
vn
elliptic
For the
first
members
left-hand
The value
of q
is first
variables.
is
similarly obtained
by substitut-
ing the results from the first approximation in the terms which
have m'/fji as a factor and proceeding as before. It is rarely
necessary to go beyond this stage in planetary problems and, in
fact, a second approximation is necessary in general only for
those terms which, on account of their long periods, have received large factors during the integration of the equations
giving the first approximation.
The system
of differential equations
it
is
requiring six arbitrary constants. The additional arbitrary constant necessary in the new system owing to the differential
definition of
v
when
v,
will
be defined as
follows.
The
final
expression
v in terms of v
of
is
when
ra'
the relation v
= v is
to hold
when
suppressed.
7*4.
0, i
When R =
we have
q,
F,
ELLIPTIC APPROXIMATION
3, 4]
177
where
e,
ty
= u = q + qe cos (v
-or),
As we have seen
in
Chap,
ill,
this
is
TV
=X
e sin
tan J
X,
X=
-f"
tan
)
by the
vr),
6/
...... (1)
or by
nt
[cf.
3'8 (3)],
+ =v
The
Ef,
Ef = 20
sin (v
57)
2
Je sin 2(v
....
i gives
Z = sin i sin (v 6).
The
is,
r)+
...... (2)
it
is
The adopted
1 in length.
define n
that for
definitions of u, q
If
we put u/a Qy
2
by the equation no
Dt is altered and the
give
defined by
<?/a
0,
latter
approximation
we put
= a, n^=n
to the perturbations.
178
We
shall
that
u=
with a
=1
[OH. vit
(1
+ e cos/) -r a (1 - e2 )
is
B.
PERTURBATIONS
7' 6.
is,
approxima-
elliptic
is
pR we
here denoted by
have,
from 1-10,
rcosS\
n = m' (\
7?
cos
S=
cos (v
-730) cos (v
AA22 =
r2
6)
cos
,2
~
,
2rr cos S,
/ sin (v -
0) sin (v'
6)
= (1
The disturbing
r', v'
of the
by means of
1
r
v'
'-
iWcosfo'-tQ
= n 't +e' +
as found in 3*11,
a'(i- e '*)
2e' sin (n't
+'-*/) +
is
>
DEVELOPMENT OF R
4, 5]
is
179
when the
in the literal
But
as these powers of n jn
occur only in rapidly converging forms they cause little additional
trouble*.
directly.
The expression for R used above assumes that the plane of motion of the
disturbing planet is fixed and adopted as the plane of reference. It should
be pointed out that, as far as perturbations of the first order with respect
to the masses are concerned, it makes no difference whether this plane is
fixed or moving. For since its motions are produced solely by other dis-
turbing bodies, they contain the disturbing masses as factors. But the
effects of the disturbing body enter the equations of motion only through
plane of reference is fixed, the solution to the first order still holds when
we transform to another plane of reference which is actually fixed, the
words,
effects are, in
most
may
may
be neglected in making
be mentioned.
We
are sub-
stituting constant values for the various elements in the expressions for
the coordinates in R. To the first order of the disturbing forces it makes
*
A method, similar to that just outlined, for developing the disturbing function in terms of the true longitude, is given by C. A. Shook, Mon. Not. R.A.S.
vol. 91 (1981), p. 553. In this paper will be found the literal development to the
180
[CH.
vn
some manner, for all these sets differ from one another only by magnitudes of the order of the disturbing forces. But when we compute to the
second order actual definitions are necessary. In general, we get better
in
mean elements
if
are best found after the theory has been completed and their insertion
usually involves small corrections to those which have been used in forming
first
elliptic
Numerical developments of
following method
The
is
^ A'M
sin
*r
- w'),
(jf+ j/V/i),
(1)
.(2)
analysis for
The principal reasons for the adoption of this plan are first, the slow
convergence of the coefficients A*, K{ with increasing values of i, and
second, the comparatively rapid convergence of the series for AJJ A JtJ
'
so that only a few special values of /, /i are needed for the harmonic
w'
analysis. No additional calculation is involved by the retention of
in a literal
form as
t3
cos
~_
du
A3
dr
1
~~"2A
r/2 "~ r2
2A8
"/a
r2 cos
7' a
^
'
............
^-2tf)
('
...... (5)
NUMERICAL DEVELOPMENTS
fr-7]
(iR) = -
When 71
(5)
is
(10)
r'
used instead of
is
needed only
sin (v
- 0).
- 6) sin ( - 0),
181
. .
.(6)
the calculation of
Of. 1'28.
the factor F, these facts require that (3) shall be carried to the
highest accuracy of all the functions. The most extensive part of
is
.A
cos(v
expression for
tf = r2
From
(1)
(t;
......... ^
'
0).J
we deduce
............... (3)
)
sm2 0;-<9).j
.........
Next, define
?*i,
C by
+ r' 2 ) C2 = r2 + r
r^W^rA, ......... (5)
A = C (n2 + r' 2 - 2rx r' cos (v - v' - B)} ....... (6)
2
/2
(r,
so that
The
special values of
r, r'
tan\=
B&SPT
T
,
sin2\i
=A
sin2X,
............ (7)
IZ
182
and then
ri,
vu
from
/
cos \i
6'-,
n-rtan^,
The
[OH.
A2
/ox
..............
(8)
(5).
gives
...(10)
Methods
for the
with
+ a2 -
ri/r',
2a cos ^)-
tyv
The
==
namely,
a,
/P
(i)
"
v'
= /8/> + S 2/3/>
i=l,
B,
/3 8
(i >
are given
/v*
(t)
cos
ii/r,
s = J,
2, ...,
by 4*24
(6)
. .
.(11)
f.
with
K= 1,
^v
(i)
_____
"
..... _
,7
(t)
where
2.4...2t
in which
j9
= a 2/(l
->'
'-">
a 2).
The
tables of
I.e.
p. 103, give
Iog2^() for
t=0
to 11
and
for
s=,
NUMERICAL DEVELOPMENTS
7, la]
183
In this way, each of the functions 7*6 (3), (4), (5), (6)
is
expanded
22C cosi(v-v'-B\
(i)
............... (12)
We
next put
V
- v - B =/> *r
r
/i
-or'
BI,
Bi=f'-fi + B,
where
The
special values of
Bt
for
2t
il
L 1>il cos(jf+j1 f1 +
is
i il
,
L' itil
sm(jf+j1 fi
).
...(13)
The
results give series having the form 7*6 (1), (2). After
the derivative of 7 '6 (3) with respect to -cr' has been formed,
is inserted in all of them, the
the numerical value of
&'&'
same multiples of f f\
and each expression is put
ments are
(13)
collected
or, if desired,
in their argu-
it
-N
step
it fc )
is
............. (14)
the expression of
in terms of/i,/.
7'7a.
When
harmonic analysis
is
to
We
rapidly.
g=f-Ef
flr'=/i-^,=/i-/i,
/'
= g' + E, =/i -
8/i
+ E (/i -
and
184
where
Eg = E (fi
>
=/!
S/i.
obtain, if
vn
planet with
disturbing
g'
[CH.
Since f\
which
4-
E(fi)
The
/ in
=/
e' is
n'
- df
special values of
similarly obtained.
d*fn'
/ with
special values of
special values of f\ are
2
^(1 + e'cos/)
dfi~
(I-**)*
= (1 - e' 2 )~* (1 +
Hence
e'
cos/).
W
which
The
will
~T6
'
be sufficient
calculation
ei
cos/
of/,/
iv
to be sufficiently accurate in
7a,8]
185
order of magnitude of the term with the largest coefficient in the longitude,
pr, in the present case, in t expressed in terms of the longitude. Usually,
the term is one having a long period. The order of magnitude with respect
The accuracy
unlimited, but
coefficients
is
g^\ Those
C>
of accuracy with
with
enced computer
ways
in
which
may
be used
It is
decimals required, but it is not easy to say how many special values should
be used. The work can be started with the minimum number and others
The methods
of
in the
form
JZ-SJCcoB^+jj/i + i),
When a literal development
of
a/a',
r',
e, e',
w + ta'
F.
20,
j=
is
0, 1, 2, ...;
made
and the
^ = 0,
1,
2, ....
coefficients
186
same values of
ji are
j>
vn
[OH.
and
its
0o + 2
or in the form
4-
2/3' sin
jf + ^/i),
+2
cos
(j/+ji /i-lto,
where
>
The constant
y8o
/3
independent of v and
is
-GT',
OT -f TB'
when expressed
20.
is
implicitly a
when
D (Jf + ji/i) = J + h n
so that s
'l
=s
coefficient
>
when we
integrate
is
7*9.
J5 s
and
its
9 -f^cos(j/+j1/1 ) + SyS'sin(j/+jJ/] ),
integral
<jo
is
= ?o + /9ov + 2
q
where
q.
is
- S cos (jf+jtf),
f
o
later.
. .
.(2)
The term
magnitude as
shorter.
2-7T/71,
8-10]
q,
187
jy u + u = qQ -f
<?o
/3i v
ov -f
sin
'
/3x
is
v cos
/ 4- e
cos
/ 4- e
sin/ 4-
t*
...... (1)
where
Now
ec
as can be seen
D2 -f 1;
M) + 2 -~r sin
ea are
values of u, q are
__
with
e, -BT
1 -f e cos
as arbitrary constants.
As
/____
qQ
e c , es are at
our disposal
we can put
e
=___._
These values of ec
we
shall
e8 will
=o
put
that
u,
may be
188
- 1+
where
'-
-* +
**>.
[OH. vii
(3)
Here
is the value of e which has been used in the
development of R\ TI, ei are small constants whose squares may be
neglected.
The expansion
__
l+o cos/
Comparison of
of UQ in powers of
gives
e\, -CTI,
l-h#o
2eo 0_i
shows that
lO-V)
if
we put
il-eo'*
;,,
The argument
it
in 7*24
is satisfied,
so that
time
in
replacing
/ by / s^v
procedure which
7*11.
is
we
{1 4- e cos(/'
10-12]
189
<?i
= 0,
^=
0,
Hence
in
713
that if
term
term
is
3
cos/)~ into a Fourier
(1 4-e
+ ie
(1
j(l
The
we expand
- eoa)~*- The
)(l
therefore
constant
).
^of
1 "T #0
argument/.
When
+e
cos/)
The
& Bq
+ e=
v-E
fy
all
per-
..................... (1)
190
vn
[OH.
We
obtain, if squares of
The
first
e\, TZ\
are neglected,
in the third
2
<?
and fourth
)*
first
term of 7*11
+ = v - Ef -
ffiji
eij
is\ v
= 0. Hence
(1) gives
nt
-~^df+
r)
J?
TI
-j^
d/,
where
and /
is
the
integral of the third term is given by
formula 7*13 (4) below; that of the fourth term is &iEf.
The remaining terms in the expression 7*11 (1) for nDt, the
the centre.
The
form of which
mediately. We
nt + 6 = v
is
+S
sin
o
(jf + j,/,)
-2
cos
where s=j+jin'/n,
f is the equation of the centre with
and
true anomaly/ -B^V and, by 713 (4)>
e
+
eiV
eccentricity Q
= 2 (- 1)* eo " 1
1
+e2H
cos
if,
eQ
=
1
713. The Fourier expansions which are needed above are obtained from
fl-*)
[)(l+e COS/)
Differentiation with respect to e gives
1
2
CD
12-14]
and
The value
integrating
191
/ provides
(2).
identities,
d
a/" 1
the
sum
sin/ =
__
+ e cos/
2
sin/
_ 2<?4-cos/ 0cos /
=
3
2
(1+ecos/)
a/* (1+0 cos/)
cos/4-0
(1
+e cos/)
of the right-hand
'
members being
2-M2
~"
(1+ecos/)
Hence
sin
The value
l-i2
'
e (1
+ ecos/)
sin
\l+ecoa/
it
'
"~~e
(\+e cos/)
/
2
(1+ecos/) /
1
(dEf/de) is obtained immediately from this result, and
also completes the expansion 7*13 (3).
The calculation of these expansions by harmonic analysis is at least as
ofFfD'
numerical value
(?
coefficients divided
100).
The
difference
by k gives the
714. The equation for F. The equation 7*1 (7) for F gives,
on integration,
.................. (1)
+ FiV +
F^
Tp
is
F^i
2
the constant term in the expansion of
ql dR/u dO,
a sum of periodic terms, and k is an arbitrary constant.
where FI
is
The value
first
+ qp ) (1 -
order,
,)*,
whence
r
The
(1
first
- eoT* + 4
- ^o2 )* -f (*cio (1 - eoT*
(1
+ F! (1 - e 2 )-*} v + i* (1 - ^ 2 )* gp + (1 - ^o')"* ^P-
two terms in
part of F; denote
it
by F
192
mass as a
of
is
&= F
eQ2 )^.
(l
vn
given by
r=
When
which
we can put
factor
[OH.
is
To
[el e,
- eoV + Fx (1 - * T*} v
turbations.
The
secular motion of
Since
Fo (1
F=1
is
the coefficient of
v.
All
as a factor.
cos 7,
SF =
which
715.
0.
The
0=0 + 0iV4-0p,
where the
value of
4
Here
is the
and 0iv is its
secular motion/
is
to
make
it
the
mean
value of
14-18]
193
7*17. The equation for sin L. When T0P can be neglected, we can save
some labour by integrating the equation for sin L to replace those for r, 6.
If we treat the right-hand member like that of the equation for u, isolating
the terms with argument / and also the constant term, the equation takes
the form
............... (1)
The
sin
Z=sin 7
The constants
sin
- 4^i' v cos
(/+ w -^o)+y
shown
are
may
bo written
in the above
ellipti
motion
sin
if
L = sin 7
factor v
we put 71 v = S7,
#^ =
5 {sin
7sin
(sr
0,
sin (v
may
where
- 0)}=
#o)
57,
-^i'v,
7cos
q,
and when
(ETO
s is small, that
- ^Hi&v.
...( 4 )
is,
is
present.
The divisor 1 s 1 is
when s is nearly equal
term
is
e as
194
vn
[OH.
C.
new
These
u, q.
We
introduce
new
operators defined
by
DQ =~ ....... (1)'
De%-+D*r-,
de
dur
9v
+
Dt=Da!lda
we
are replacing
between the
variables
new
a, e,
tsr,
we have
two variables by
is at our disposal.
The
three relations
to be adopted are
~1
__
-f
0cos (v
CT)
a (l-e27~~
?==
'
1
*
a(l-6)'
-f
e cos (v
~ __
-or)
'
a(\-#)
...... (2)
Du = DQ u,
so that
when u
Since
is
.................. (3)
Dq =
{1 -f e cos (v
This,
1)^ = 0,
combined with
{e
cos (v
)}
0.
T\
Next, the
Dq + qD\
cr)}
\1
C/./V
I>!{C08(V-W)}--^J=
equation Du DQU gives
Di^
qe sin (v
since
But
=
=
=
D {gesin(v
DI
D^
{qe sin (v
.
e sin
(v
DO (^ sin (v
Dq.e sin (v
&)} -\-u-q
-or)}
tsr)
r)}
-or)
Hence
D*u + w
>
................ (4)
^Z>i {e sin (v
=u~q.
= (Dq/q)
DM.
-or)},
19, 20]
Whence, from 71
DX
The equations
y
(3),
/
{*
1
sin (v
M
n
r)}
195
Du'dR
,-.
--= - d-=R
-5- .......... (6)
=-
u* dv
du
are those which give the variations of
(4), (6)
"
may be
written
Substituting
+ *_
2
for u,
ag +
dv
Da, q from
(2), (5),
dR
which
is
^
on
we obtain
dR
a.
The disturbing
of
u, v, P, 6,
and of
through
its
presence in
Hence
r' y v'.
oR nT1 dR ^^ dR *.
dR ^
DRn = dR
Dv + ^- Dw
+ ^ DF
+ ^^ DQ + -^~ Dt.
dT
ou
dv
cQ
dt
T.
-p.
-5-
of the expressions
71
Dv,
DT, DO,
this reduces to
dR ^
dR
-~- + *- Du + -zrDt ................ (1)
DR = dR
du
dv
dt
T-.
r,.
,->
Now t enters
into
^n'.
new
v only in the
form
n't
+ e',
and
this angle
manner
that
Hence
a result which
old or
r',
is
true whether
R be
the forms
\W
since
Dt = qft~ku~ z and
...... (3)
196
7*21.
of the
[OH.
vn
new
2
<s7)}
where Ef
is
3v
7i
'
Hence
Dodv +
f^dvThe values
= (/4/a8 )^,
of n
Equation (2)
may
be written, since
D =D
D\,
f=f!dv-^+
fl) (^dv + const .......... (3)
n
\nj
Jn
}
1
'
As in
tion,/= v
w, equations
719
(4), (6)
can be written
.sinf=2u
............ (1)
...... (2)
the computation of r2 9JR/9r being carried out in the manner explained in 7*6. From these equations we deduce those for De,
The equation
for
The equations
The
latter
Equation
may
a becomes
for 6,
T remain
be replaced by
7 '21 (3) is
used to 6nd
t.
21-24]
7*23.
R is neglected a, e
-or,
F,
We
proceed as before.
197
When
is
The amount
mean
motion.
much
very
work
is
where q\ e are numerical constants, in order to find I/a, and the multiplications by cos/, sin/ to find e, w.
In either case the work is less laborious than when t is used as the
independent variable, chiefly because, in the latter case, the single terms
cos/, sin/ have to be replaced by Fourier series containing a number of
terms corresponding to the highest power of e we need to retain.
The chief saving of labour in the use of the methods of this chapter over
is due to the avoidance of the formation of derivatives
with respect to e. There is great advantage in using the numerical value
of e from the outset and if this be done we cannot find the derivative with
respect to e. Further, the development, to secure a given degree of accuracy,
requires the presence of one power of e higher in the latter case than in
those of Chap, iv
the former.
2
I/a and none of the form tcv in t, we use equations 7*20 (3). To
the first order, dRi/dfi has the same value whether RI be
B&SPT
v,
so that
13
198
The
secular terms in
&
e,
[OH. vii
These, substituted in
Ef, produce terms of the forms v cos if, v sin if, but no term of
the form tcv to the first order. Since there is no term factored
by v in
a
Smce
.
/ev
in
t.
w=
e=e + ^i v i Q
secular term in u.
part.
When
This
differs
We
depends on Da, De, Dcr, has no divisors and its integral will give
rise to terms of the same kind. The integral giving a gives rise
2
p, to divisors s
I/n, through n a*
This latter result which was stated in 7*18 is of some assistance in the
to divisors s
and that of
D.
present.
The
We
where
= l-f ecos(f
v
- Ef
/= v -
isr
SECOND APPROXIMATION
24-28]
in which e=*eQ
eccentricity
a Q e0y
,
from
found
e>
<*o>
R;
#o
e iy
up qp
199
TI,
,
Tp
p are
is
now included
in Up, qp
If we had solved
.
but
for u,
to be substituted in
u= {1 -f e cos (v
7*27.
tzr)}
-f-a(l
2
),
l/q
= a(l
2
).
still
and
,
1
u
37 = 57T~
9M
x
8/1
8/
~
du
8fi
200
the
first
/, fi their derivatives with respect to /i are immediately obtained. If the first approximation has been obtained
t
by finding
we
*-+-
M<i-<rt
UQ =
where
{1
replace Sqk,
Su by
+ e Q cos (v -
<sr
)}
4-
(1
-e
.......
2
).
Ft
(5)
sin/D (8,)
f
(
(oji \
d
r*
St.
or J
...... (6)
The second
tsr
eiv,
From
these
T!3-
1 v,
we get
,j
8u
5--i-
ceo
dtjj
to be substituted in the
right-hand
(5), (6).
fcv -f
members
SECOND APPROXIMATION
28, 29]
First, let
201
(1).
-D
{A
(/feu")
S|
tev
from Su,
is
= - (D +
since all the other portions are products of series one member
of which contains /i and the other is independent of fi. For
=
terms of this form we therefore have, since
are to be chosen.
/cv
This
is
The two
/3JZi,
latter
substitute for
/3JZi\
&t
3#i
3#i
3(8jR)/3/i.
the portions u p
Hence when we
have no
l n '_2fi d^l
*
90*3/1'
which in general will not be zero. Thus the theorem, that I/a
has no secular term of the second order when t is the independent
variable, is not true when v is the independent variable. It may
be noted however that it is true for a second order perturbation
arising from two different disturbing planets.
the right-hand
secular terms of the
to
we
will arise
shall
202
[OH. vii
quite low.
is
The practical importance of the theorem that there are no secular terms
of the first and second orders with respect to the masses in I/a when t is the
independent variable, has been much over-estimated. It is a result which
eliminates certain secular terms from
1/r,
or,
in
to be averaged. The fact is that the separation of the deviations of the coordinates into deviations of a, 0, or, etc. is an artificial one, convenient for
and
calculation
significance.
7'30.
Terms
by
the formulae
[
v cos sv dv
v sin
svav =
,
v
s
82 it,
will
Calculation of the
considered.
effects
due
to
to
be
a derivative of
2
Su, the divisor s is present; in &t, the divisors s, s are present.
The products just referred to will produce products
COS 5V
COS
s'
= ^ COS (S
$')
4-
i COS (S
-f $') V,
29-32J
The former
will
squares in S2
(s
203
s')
in Stu
and their
sensible.
s
or
5', s'
in the right-hand
members
the
the
the
the
is
We
in S 2
7'32. Calculation of the portion of a coefficient in &%t depending on the fourth power of the small divisor*.
We
& 2 t=
substituted in
S2
where
Bt arises
\n/
dv,
..................... (2)
from
substituted in
<$(-)dv,
..................... (4)
2 8
by the use of the relation n a = /A.
Suppose that the term in RI is
$ cos sv + a
/S'
sin sv
= aQ A
cos (sv
^),
sv
=
...... (5)
E.
W. Brown, Mon.
204
and the
[CH. vii
result integrated,
we
obtain
-}
CL/
= 2n' (&]
\/^/
The equation n2 a3 = /x
gives
3a,
g!
...................... (7)
2i?
ft
Whence, from
a
S
(4),
*
From
(1),
we have
(sv
from
The
(8), (9).
+ si)
!)
S<
g<
+ s1 ),
............... (10)
integral gives
v .......... (11)
^v-M,)
\
\
/
And
/1\
3 do.
Finally,
*.
In'.
(12)
32-35]
If
7'33.
we have two
Hi = a Q A
and
is
if
we wish
205
cos (sv
4- si)
to obtain the
-f
$/),
it
!- B
cos (sv
+ Sl )
'
'
-
CoS
V)
(^
(5
-f
s in K*'
the upper sign giving one term and the lower the other.
The
V = \ J Bff -gr^
In the case
= s,
s'
sin
[(,'
= jV- The
'
s,}.
_ frBB'
1 ^'
=-
+ sl
s)v
sin (2sv
+ sl + Si).
??Q
If
Sit;'
= B\
is
particular, gives
an additive part to
S2 t
t:
'
illustration.
206
[CH. vii
1/s*
Suppose that we have calculated the terms with argument (5n 2?i) t to
first order in the motions of both planets, and that we need the principal
second order portions, the latter can be obtained immediately from the
formulae in 7 '32. We shall perform the calculation and compare the
f
the
results with those given by Hill*. Since the latter uses Hansen's
be necessary to compare it with that of this chapter.
method
it will
Denote
Hill's notation
of this chapter
by
(v).
In
elliptic
motion we have
=
(T), v=z + n+E(z)', (v), nt+c v-JS(f),
where E(z) is the equation of the centre expressed in terms of the mean
anomaly z and E(f) is the same expressed in terms of the true anomaly/.
In disturbed motion we have
(H\ ws*g+te + vr+ff(2+bi);
(v),
nt+c=v-E(f) + nto.
first order,
neglected since
it
period perturbation
the portion due to the elliptic periodic terms can be
only produces portions with the divisor s. Hence, very
nearly,
n dt = - n fa,
ri dt'
=-ri &',
R. For the latter we have in the motions of Jupiter and Saturn, respectively,
*/,
v in
= noz,
dv^ridz*,
= 1196" sin N,
= - 2908" sin N,
n'8z'
two portions
are, in
our notation,
n 8t = - 1 1 96" sin
and
in
tf,
8v'
= - 2908" sin N,
- w**g and
fi for n't
e'
-cr'
=g\
for
the
1 1 96/>,
WoM=|
1196p
(f
1196/3
* American
Ephemeris Papers, vol. 4
t Coll. Works, vol. 3, pp. 560, 568.
Coll.
Works,
vol. 3.
TRANSFORMATION FROM
35, 36]
(first
+ second
term
approx.)
in
n 82,
TO
207
similarly,
Jupiter
n' 8z'
are *
- (first approx.)= -
Saturn
1"'4=
12"'4,
26"'84-3"-4 =
30"'2.
11"*0
After the change of sign necessary for accordance with our notation, the
comparison shows a close agreement between the two sets of results.
It may be pointed out that the percentage accuracy of the method is
greater the smaller the value of s. As far as the present demands for
accuracy in the various problems of the planetary theory are concerned,
the method will give all second order coefficients with sufficient accuracy.
It has been shown, however, that the terms depending on the divisor s 3
E.
be facilitated
too
+ e = v-Ef + P,
= v + Sv,
(1), (2)
/-v-*r, /1=s
The expression
of v in terms of
-) +
'-t*'
(3), (4)
Pub
n$t
-f e
tzr
=g
g-P=f-E}
Coll.
Works,
(5)
208
[OH.
vn
fore
be used. It gives
............ (6)
will
be very
terms of
The
first
them
f=g+E
g,
fi
= n't -f e'
g'
where
= g' + n
^E
g,
sr'.
+ B cos
The second
argument
............... (7)
this
term becomes
E^
B
(jg +j g' + l sin
~^J
j(j
g, either
formulae
l-t(j+Ji^)X'
+ ...,
(j+fi%)E,-...,
...(8)
+ ji
ft'/
71
is
coefficients
small.
/ in
terms of
of
P is substituted in
(6),
which then
TRANSFORMATION FROM
36, 37]
TO
209
If this first
in n^t
becomes a perturbation
in v,
where the last term contains the perturbations due to the transformation from an origin in the osculating plane to one in the
fixed plane. These, having the square of the inclination as
well as the disturbing mass as factors, are very small, so that
= 2esmg will serve.
the values (5), (6) for/,/i with
g
1/r, 9,
are transformed in a
similar manner, that of 1/r being found with the aid of 2*2 (2).
in terms of t is used in the
The value just found for
expression for 1/r, while the values (5), (6) will be sufficiently
accurate for substitution in the expression for 6, T.
The method of this chapter is closely allied, as far as its final form
concerned, with that of Hansen*. Substantially, his method requires
the expression of the true longitude in the form
7'37.
is
2
e'
it is
that
*
it
The
1838.
Fundamenta Nova,
etc.,
Gotha,
210
known
The two
[OH.
vn
may be
urged are,
first,
it is
only in the development of the disturbing forces, requires more labour than
that in terms of t. As pointed out in Chap, iv, each requires substantially
four operations, two of which are more simple in the present method than
in that of Hansen, while one of the remaining operations is more
complicated.
There
is,
exterior planet
those of
F.
7*38.
calculation
2
),
f*>
we put
=
q (1 + Sq) q0j u = (l+e cos/+ Su) q |
nQ t + = (!-#)% (1 -f e cos/)~ 2 + n St, J
,
(1)
APPROXIMATE FORMULAE
17-39]
may be
,hey
211
written
.............................. (2)
(3)
d *,_
-">*_
f ift
SM
~o
+ e cos/y/
= JSg -28w + (68^-8^)6 cos/ + ....... (4)
Consider a term with argument
md
<r in the
disturbing function
rise
to
terms
that
this
term
gives
suppose
si no-
in
^-(r*Rqo),
Bcoscr in rz
-^...... (5),
[n general,
A>
will
--
2^1
*
gg,
*
we obtain
da-
COS
CT,
T-
dv
................ (7)
^
+ i) Su = - f
(&
\
\rtw
/
cos
(<r
\eA
}he integral of
l-(*
The
+/)
cos(ov
/),
which furnishes
2A
jince
(6)
\coso-
l)
= + 2s-s2
...... (B)
.
jive n Q St.
7*39.
tihe
cosr
S
(4)
/2A
D \ cosott--f
-f
a
o
\ o +JB)=
/
-I-
which have
..-(1),
212
The transformation
/=#,/! = /
in
[OH. vir
we have Sv=
0,
n^St,
and
<r.
for all
e or
it
suffix zero,
by the
To obtain the
of
e, it
suffix unity.
2SiU
j-Sit-
To the same
power of
power
is sufficient
Hence
nQ
first
-h
(68Q u
- n Q Sit,
...... (6)
with/=<7,/i
The terms
s
or s
1 is
by SQ v =
argument a
7'40.
/ must
Solution
be considered.
to the first
development of a'/ A
Si {c^ cos i^r
(D
powers of
the eccentricities.
The
2mi) a f cos
e'
(i^r
+/)
= a/a',
-^
The development
=/ 4- w ~/i
to the
*',
D=
d/a a,
m = w'/w,
DETERMINATION OF CONSTANTS
39-41]
213
and
t
CN-
bince
a-^
r-^-
da
or
/-
/.
a (I
2,....
1,
0,
ecos/),
we have
r2
,r
=a
2i }Da $ cos
e (Z) 4- 1
i\Jr
4- e'
4-
a2
7?2/
-{
cos
-v|r
+e
4- e
-a
(1 -f
(1
(D +
2rm)
cos
(ti/r
4-/)
- 20 Dcti cos (
m) cos (^
/)
- m) cos (^ -f-/) -
/)
e(D + 2
2e' cos (
a,
sn
{- sin
yfr -f
ie(3
beyond
G.
7*41.
give
first
when
all
214
The constants
principal elliptic
[OH.
vn
e$, OTO
sented by
2
cos (v
cj ).
inclination to
slight difference
when
t is
be sufficiently accounted
for in
from observation.
7*42.
process of approximation is used in the determination of
the values of the constants from observation. The perturbations
examined
to see
7*43.
The
detailed
constants, as well as their correct definitions, has to be carried out whatever method be used to calculate the perturbations. The difficulty of
DETERMINATION OF CONSTANTS
41-43]
215
it
its
its
own
14-2
CHAPTEE
VIII
RESONANCE
8*1. Resonance is usually defined as a case of motion in which
a particle or body, moving or capable of moving with periodic
motion, is acted on by an external force whose period is the
same
it
-r
4-
n*x =
sin
n,,t.
at*
When
=f
we have the
But when n
f
,
solution
the solution
x=
c sin (nt
in
-
-75a
&m n IA
t.
is
+ a)
\ mn't cos
n't,
left-hand
motion,
is
member
1-3]
217
interval
is
due to the
it is
The
and a following
We
shall
made from
in the ratio of
RESONANCE
218
[OH.
vin
following chapter.
8*4.
We
Since the final expressions for the comultiples of the frequencies, each pair of these can be
regarded as a possibility for resonance conditions. But these frequencies
are observed quantities, namely, those of the mean periods of revolution,
ordinates contain
all
phenomena
is finite.
between two possible routes. And this process appears to continue as the
approximations follow one another. From his point of view, there can be
no general solution of the problem of three bodies, that is, there cannot
exist one. set of formulae giving the coordinates in terms of the time and
the initial conditions which will serve for more than one set of such initial
conditions, which will be valid for all time, and which can be used for
calculation of the position. This conclusion may be a result of the mathematical devices which he adopts, but is more probably due to an inherent
difficulty, namely, that of finding expressions which shall be continuous
MOTION OF A PENDULUM
3-5]
219
Any
proper solution of the problem requires also the consideration of the limitations placed on the observer it is not solely a mathematical problem.
;
8*5.
K*smx =
This
is
(1)
of length
t.
2
changes the sign attached to # the equation
2
2
with
/c
replacing /c gives the same motion as (1).
The equation has the integral
tion,
x -f- TT
for x,
Y-C'-f 2* 2 cos#,
where G
(2)
necessary that
(i)
(7>2/e2
As dxfdt never
it
is
We
dx
as the integral.
const.
we put
If
_
~ J^
2"
dx
dx^
2^ Jo (C + 2* 2 c
K?
nt
4- e H
/e
on
-f e)
(3)
periodic portion of this series can be regarded as an oscillation about the mean state of motion which is revolution with a
The
period 27T/W.
The half-amplitude
of this oscillation
is
evidently
RESONANCE
220
less
than
and
TT
[OH.
decreases as n increases.
it
It
vra
convenient to
is
consider
C<
(ii)
The
n, e
2/e
when x =
Here dx/dt =
2
.
a,
where
2
|a -sin |#),
a,
x = c sm
(pt
where
-f /?)
C
T^T sin
2
-^c
(1
3 (pt
ft) -f
. .
(4)
...).
(iii)
where
x -f
a<> is
gives
TT
4 tan" 1 cxp.
(fct 4- c/o),
(5)
value.
When t=
and
it
follows
of x vanish.
limits
<x>
TT, t
x=
TT:
by
Near
is
is of a quite different
(7,
type
2
a
C
or from
from
2*
2** < 0. This
as
>
<7-^2*
according
result is of course characteristic of unstable equilibrium, but the
arbitrary constant
C-
is
MOTION OF A PENDULUM
5-7]
(a)
of dx/dt in
(i) is
221
n and in
(ii) it is zero.
In case
2
(i)
or a which
is
related to
it,
and
TT.
2
depends on V/c There is no analytical
continuity between the two types of solution, and they cannot
be represented by one and the same analytic function of t.
powers of *
in case
(ii) it
(d) In case
with
A more general
type of motion
is
(~\ = C-
its integral
where/(#)
is
assumed
to
We
get the
same three
C<
the value given to (7. In the first case also dx\dt never vanishes and it has
a mean value different from zero in the second case x is a periodic function
;
of
x-72/yi
-y-2
(it
4-
/c
sin
x = mic2 sin (x
n't
e'),
......... (1)
RESONANCE
222
[OH.
vra
it is
'
advisable to use
The method
is
given
cases
is
periodic,
and
also the
8*8.
when
is
that of
"^'^"
d?
is
x~x(l,c\
where x
nt+e,
= func. c,
(3)
The
(I,
c)
is
Let
let
I,
and
(4)
dx
dx dl
variable.
ex dc
We
have
CHANGE OP VARIABLES
7, 8]
We
223
are related to
'~_
dx dc
dx dl
dx
then
di
Whence, since
is
dx
w
.................. (,~^}
'
dx
d*x
d*x dl
9 /
I,
c,
n,
dx\ dc
Substituting this in (1) and making use of the fact that (4)
zero,
is
we have
n
The equations
to find dl/dt
( dl
n,
9 /
dcc
dx
n x\
,,
m(t>
dc( dl)'dt-
/(7N
.......... (7 >
and
K=
Tjr
dc
where
n\
W'(dr
dx
9 /
dl
dx\
dx
(n -^ }. ~r
dc\ dl/ dl
- n (Px
^
2
dx
dx
.
dl
/r
^ /
5- ................ (9)
dc
and
dK/dl
0,
showing that
function of c only.
If <' = d<f>/dx, where
in the form
<f>
is
is
a function of
m
dc
_
_ d6
__
dt'Kdr
dl _
_
ij
dt~
where
its
independent of
x,
m
_
t,
is
therefore a
we can express
9<f>
r
Kdc'
/,
and
c, t
(8)
/irkX
/
............ v 1 1 1 '\
by inserting
for
RESONANCE
224
Since n
vra
[CH.
= Kdc, dB =
dci
n dc\ =
nKdc,
~ = -(B + m
/r>,
/D,+
/-i\
-(B
Kf)....... (11)
i\
<t>),
is small, the
89, Solution of the equations for I, c. When
usual method of approximation is the substitution of constant
which
values of j?
Q, Co for n, e, c in the terms factored by
=/
-f /!
= n Q t -f
/ l5
Ci
+ hj
beyond the
= Co -f Ci,
first,
we have
dn\
from which
Ci
and then
li
We
A
^ -~dc\
(
dt
the
obtain
ty\ *?
t
~K^c)
~'"~_
_m
fdn
d^
?l
-^ ^ - ^.
Kdldc'Tt
dt
i
_8^> ""
2
j>
-?
Kdcdi
m2 it
may, in general, be
given constants.
When
cty^jV
cty
dt
dl
'
8,9]
and the
first
d* 1
,
-T72 -f
ar
v/5
(in
<
where a it
m
- Jw /\a
)
is
9 /~
9<
=->
5iKoc
\in-jn
that in which
c^ cos ^-f&,
li
in a first approximation,
If,
225
M-
ai
oil
<
ilj(n't+e),
The equation
we put
....... /Q
w)
=c
......... (4)
for
n = n Q)
lt
is
then
K= K
(5) takes the form of the equation for the pendulum. [If the coefficient of sin l t be negative, we put
all
constants
Z.
= i7-j(?i' + e') + 7r
There are therefore the types of
(i) is that in which dli/dt is
Type
o)-j(n't
cti
(ii) is
+ e') =
l i0
that in
in the
/a
as a first approximation.
With type
we choose
(ii),
to be such that
and
li
is
an
-=*
jn'
iiiQ
i Q
0,
oscillating function.
liy
n^> c
= Co,
je
K = !Q
X,
X<>
(7)
TT,
or
<>
we
ma
"If"
fi n \
13")
/o\
>
()
With
dc
Whence
=c
f
t
mt\
- cos
(^ + Xo),
(9)
RESONANCE
226
where
of
is
[OH.
T? O
is
(9)
is
known
vni
function
CQ.
The
term in
..................... (10)
v
'
X,
first
put
Co
The
is
difficult cases
I in (5) is
justified.
in celestial mechanics are those
which
resonance
is
this
possible.
A case
l,j
=2
is
treated below.
8*10.
We
where
mR
. .
We
where
E = RQ + 2Aco*jiN,
was a function of the ct and
.................. (2)
JI'MI
'
GENERAL CASE
9, 10]
In a
first
integrals for
j\n*+ji!n'.
vanished.
227
It
is
one term
for
which this
is
+ j\n
is so
small
We
We
new
Wi =jiwi
ji
+ji'wi
+J2 W,
............... (5)
so that*
ji8Wi=
It
is
easily seen
/iSwi,
that
the left-hand
hand member by
ci
Next, replace
Ci
ji
= by the relation
Previously, n$ was defined for the case
-3 = ft
0) so that the second relation (8) is the same as
^2 Clo
= 0. This definition does not demand that rc shall be
+
o
?i
ji'n'
8 have the
same
signification as in 5 '3.
RESONANCE
228
mean
the final
value of
n,
since there
[CH.
may be
a constant portion
vm
at this
We
j^
The
first
(9)
do not
= /^io~ 3 =-ji
flo
we obtain
?*
...............
/ji>
cos
......
The
(10)
last expression is
(11)
It is to
be remembered
CIQ are, by definition, functions of n' only and are therefore independent of c,-, Wi, w {
that no,
8*11.
-^
The equations
for c n
= --mjiABUijl N',
dQ
-^
--
-7dt
ecu
Cu
Wi become
....................................... (1)
CIQ
~n """
3^0
~-
d
5
oc n
5~~
den
Ci<f
...... (2)
by
Ci, (72
= mi(7i,
7
all
Gz where
cu
and
02=020
+ ^^2,
c3
Cso
+ m^Cs,
...(3)
GENERAL CASE
nj
iv,
where TF2
disposal.
>
The
form.
w2
quantities Cao,
be noticed that the factor
turbed
It will
w$ to preserve symmetry of
constants which are at our
now
CM are
d = (/ia)*
c2
mean
variables
We
elliptic orbit.
have
= ci {(1 - e2 )* - 1},
The replacement
of
has disappeared
by
Ci
-f
CB
ci
(1
- ea ) } (cos i - 1).
The
factor
m = 0.
wi would seem
to
this is correct,
Mathematically
in shall not be zero, we can at present be content
demands that
The replacement
value
is
CQ
of c%
such that
necessary
if
(e
Cw+ndCt
by
e
-f-
is
is
of order '03/xi.
1,
implies
we have
that there
we contemplate expansions
mi
m%
e<
m//j, is
in
is
But
care
powers of m^C^Cm.
Thus expansions
in
powers of
^20)^
(<? 2
made
ment
B&SPT
15
RESONANCE
230
[OH.
vra
8*12.
mi
neglect
The
coefficient
function of
w2 w3 only
,
-3?!
The
first
we can
first
are
GI
-\
/i\
/n\
............ (1),
(2)
...... (3),
(4)
-,-^T
U-JL
710
*r
- 3w o a
smji tf = 0,
CIQ
rx
'
or, since
wa
This
is
dN/dT=dWi/dT,
jiN
-f TT,
we
we may
replace sin^i^V" by
tf=Xsin(p* + Xo),
if
dT by its
j^N
or
by
value m^dt,
= a>ji*\A* ~m,
...(6)
CIQ
where
X, Xo
The frequency p
is
proportional to the square root of the disturbing force, while the coefficient and phase are to be determined
from observation. In all cases except that of the Trojan group
GENERAL CASE
12]
of asteroids in which ji
and
231
/, some power of
e,
will
be
present in
it
is
therefore
>
in powers of ra*.
The value of Ci
Ci
is
given by
We
(2), (6).
find
).
. .
.(7)
^oji
The
is
small factor
made
small.
Since
R w 2 w3
Q)
~Ci +
C8 = w**!^+-i8
dw3 fr
Now RQ
multiples of
(72 ,
C9 shall
that
Since
wz w3 w2
w2 ^2 4- w2
contains
wa
'
',
'
ft
const.,
............ (8)
it is
necessary
all
thejiy ji
is zero,
however,
condition (10)
If,
is
where, as usual,
jl+jl'+J2 + J3 =
0.
We
15-2
RESONANCE
232
[CH.
vni
medium
of
the libration
wi, wi,
w3
ofjiN
are
all
is zero,
and
determinate since
range of values
for
When
manner that
N=
are possible.
The treatment
will
is
be found in Chap.
814. It
eliminates
is
t
ix.
from
same:
2
7i
2
1
4/1
is
= const.
an integral of the
GENERAL CASE
12-14]
ci to c
gives,
233
by
8*1
(9),
c 10
Cio
The
illustration afforded
from the
sufficient.
cos
2w x + w 2 ),
'
( \GI
In the ordinary planetary theory, the variation of w%' and especially its
secular part can be neglected in a first approximation and the result may
be later corrected sufficiently to satisfy the needs of observation. If, howformer angle is oscillating about a mean value, it is necessary to
consider the nature of the motion of the latter according as it oscillates or
ever, the
makes complete
revolutions.
that
of such cases
is
given below.
RESONANCE
234
[OH.
vm
Certain
features
of
have been
resonance problems
be given.
The
integers j,j
whose
for
than
5.
j^
which
jiV is very small or zero,
r
n
n
are
observed mean motions
Here Q
,
numbers
is
so close
eccentricities
and inclination
is
l^i
^i'l (cf.
it is
desired to
The terms for which jin Q j\n' is not very small or zero can be
eliminated by the method of 6 '6 and the resulting function therefore contains Wi, Wi only in the combinations p(jiWiji'wi),
where
we can
take as the
Hamiltonian function
...... (1)
RQ
j\w\) where
Any
canonical
still
remain
GENERAL METHOD
15, 16]
235
= (^a)*, n2 a3 = p,
Ci
it is
convenient
to put
Ci
=C
+ z)~l,
= (^a
wo
)*,
3
(to
/*,
NO
= ji
n'/ji ,
......... (2)
so that
and
is
ft
=w
or,
+ *)*
Co {i (1
on expansion in powers of
As
^ is
4- (1
*)-*}
01 JR,
#,
we have
Ci
to a constant
*2
The symbol
-*
z corresponds to
it
m^Ci used
follows that
in 8*11
............ (5)
and therefore
the factor m.
z,
-6mJfZ/7?
we have
i
Co) ................... (6)
we can go a
equation, which
8'16.
it is
of
is
that for
t.
RESONANCE
236
Suppose that
[OH.
for its
mean
As
vin
cQ cos/! TV
does not
of z gives
in C.
value
^oo
./i^oo-/!^'
~ ~~
36mM a
j/ A
\J *
-r-r-
~~~77i>
Y2
16
ftitQ
1~
.......... & )
I
z is large
in R.
dW
~dt=
z,
n Q in 8'15, givea
dR
rrr
Thus the
long period
and there
) t
+ const., -
sin
,r
^ ^- ^
.
first.
\~~c~~)
by a term of
is
is also
coefficient
- feji
(coef. of
sinjiN)
2ji
N having a
2
,
dW
Differentiating with respect to
dR
t,
we obtain
dz
dt)'
FUNDAMENTAL EQUATION
16, 17]
237
where, in the last term, i has the values 2, 3. Since the derivatives
of Ci, <?a, c3 Wz, ws contain
as a factor, we obtain, on neglecting
But the
first
815
Now K
is
and expand
a function of
.R in
(2),
2
,
z,
If we put therein
we obtain
implies that c
on
, t
Ci.
powers of
or,
ma
CI
= CQ (1
c.
PR
On combining
this
(d*R/dcidW)o disappears,
we obtain
we can put
~
^^
n Q dt
~
c a , 03,
....... (
w z w3
,
are
still
RESONANCE
238
vm
[CH.
= 0.
e'
Since
dR
dR
T"
3-R
,
i
?\
dR
>
dwi
of
dw2
>
On
dR _ "A
............. (*)
T^
ji'wi, so that
/*_
|
ji/dW
dwj'
dwz
dwz
810
(1),
we
obbain by integration
/,
1
\
Jji\
-r
ji/
ci -h c 2 -f c 3
,
= const.
If,
(2),
ow
dt
(6)
we put
we obtain
When
f 3
~
>
J
2
e'
'
The
last
term o
integral.
THE CASE = r =
The variables c3 w3 disappear and
<?'
819.
Wi
fixed
R and
by the resonance condition. Hence a
w* are present in
is
The
single angle
The system admits of the two integrals 815 (5), 818 (4). The
latter enables us to eliminate the variable c. From the former,
18, 19]
239
tf,
known
We
to be small in
assume that
z, e
--z =
E
Next,
and
(1)
a constant.
is
if
(4),
const.
# = E-l\l- J
4Jz
or
where
0y
additional labour.
The
where
z\ the
same
coefficient
Ap
limitations as those
made
in the case of
p=
RQ permit
to
(2)
constant part
2*
= G-6mAe J co*jiN.
(3)
becomes
= 3? 1 mAe J sin ji N.
(4)
RESONANCE
240
The elimination
[CH.
vni
gives
d?
\2
...... (5)
of
its
For the cases of chief interest in the solar system, J is, in fact,
than 5, and the equation includes all these cases. The most
less
serious limitation
8*20.
is
Particular
These are
classified
0.
according to the
values of /.
= 1.
ratio ji/ji
the
mean
sufficiently convergent
calculations
It is doubtful
The
case 1/2
for
is
numerical
discussed in
The
J=
is
also arises
orbit.
when we take
J are
chiefly of interest in
(1)
CASE OF THE
19-22]
THE CASE
ji
RATIO
l, ji'
241
8*21.
a)z,
t,
where
a>
~}
e,
for
\o/
??o&>
= (12A 2 m2 )%,
z, t, e
respectively,
W
-~>
Cvv
dr
*
,
N= Wl - 2^' +
OT
W+
<*,
Clu
The
scale.
new
but
Conditions
(i)
mean
to be satisfied.
fore lie
shall choose
mum
RESONANCE
242
(ii)
The
[OH.
= 0, so
vm
that
-(z*-C)* =
(iii)
The
member
left-hand
of (1)
................... (1)
must be >
values of z.
(iv)
The convergence
of the developments
is
'3; this
doubtful
if
gives a limit
The equation
de
e
= E+2z
dz
dt~dt'
gives
d 2e
/de
dt 2
\dt
"""3?"
e y de/dt
Since z
=s
if
e is not negative,
de/dt vanishes simultaneously: hence e is
roots of (1),
it is
easily
deduced
that
CW +
and that 8*21
(1)
CP--JJ-,
may
E = 4*ffi +
-28,
...(2), (3)
be written
-4(^)
These
results give
............... (5)
so that the
of e are given
by
...................... (6)
ZZ-Z4J
The
8-23.
identities
(6)
z
show that
but that
if 5
d2 - n
ZiS
-I-
l/4s
jV"
1/4$*,
and with
= s2
> 0, d >
d<
if
'416
6 cos
N,
TT
N are
TT,
0,
at both extremes,
= l/4sa
8*24.
The
z l = s -f d,
By hypothesis 2
that
is
z% = s
to lie
or that
(a)
between
For
> 0, d2 >
The
>
>
is
For
1/5,
d,
^2-
d<s-f^-2.
fr<S
the roots
=
always real between z z,
2
>
are
> 0, d 2 <
d
#1
dz/dt
1/5,
d<2s,
(6)
z\,
when
z%.
= s + l/4s2
d =l/s,
#3,
Hence
when d <
when d>
2s,
25.
RESONANCE
244
(c)
If s
<
0, all
[OH.
of
is
2s
Since d
zZ)
zi,
>
0,
<
it
that s
+d<
0,
and as we have
d<
The
cated.
and
-1-0
<
0,
that
is,
if
z can
1-6
conditions that z
The boundary
d, s
may vanish are therefore quite compliof the region consists of the four curves,
<
that z does
-0-5
On
for s
easily seen
is
It also requires
0, it follows
-1-5
vm
and
> 0, where
are simply
is real.
RESTRICTED
24, 25]
CASE
245
two
latter regions
is,
d=
= s 4- l/4s2
for s
>
to the
boundary
l/4s for s < 0. Along these two curves and also along
two
of the values of z are equal and
d?=l/s,
jzdt is
oo in the same sense as in the case of
indeterminate near t=
s
The same
solution
is
available for s
<
The
solution
is
0.
is
then
-f l/4s.
provided
s is
TT
when the
From
8-22
is
libration is zero.
(2), (3),
we deduce
C2 - E=(s*-d
C2
changes sign at the boundary separating the region in which
also when
can be zero from that in which it is never zero. But C 2
d2 =82 +l/s, a relation which does not enter into the discussion given above.
L
is not the necessary and sufficient
It follows that the condition
so that
O =E
condition that
In
vol. 4,
z shall
Ce'leste,
case
those in which
e Q is
small.
The
f show
W. Brown, Mon. Not.
statistical discussions
B&SPT
S.
G. Barton,
16
RESONANCE
246
[OH.
mean motions
is
vnr
variable z
may be regarded
and the
is
given by
=d
which z can be
and the maximum value of z is then 2s.
= J. The
least
maximum
of e
is,
8'21,
On referring
find that this gives a least
therefore, 1'5.
we
maximum
So
far,
exactly twice that of Jupiter. But it has been shown that if such
asteroids can exist, the elements, in particular the eccentricity,
must
oscillate
small
that
if
such orbits
exist,
N
is
zero.
We
to that of 1
2.
THE CASES
e'4=0
it is possible,
in a first
first
EXTENDED
25, 26]
W+
JV's=
equations
vr'.
for
247
at
_ __
*r
*T/\
sm N
+ e AAt sm N)
-
n^ at
'
z*
02,
w2
let
defined by
w2
sin
where, as before,
for Ci,
CASE
Arguments
W, z
--
2> (72
W, p 2)
<?2
are
isi^i
still
p2
#2
According to
is
a function of
5'14, the
equations
canonical.
,
q* to
new
sn cr,
where X
= 0i Ci* cos w2
Ci
only.
variables defined
= ^2 + ^
^2
We
by
cos
have
contain them.
dp2
/dt
is
may
dt
therefore be written
.
$H=dci
8W-dW. Sci + dp
pz
- dq2 Sp 2
- dq 2 Sp 2
Sq 2
Sq 2
'
'
(e cos OT'
dp2
'
e'
sin
cr'
dq%)
Sci.
C/Ci
RESONANCE
248
[CH.
vm
e cos
-or
e sin ts
= e sin *r + e
-~r sin IB
so that
e
A sin N -f e'A
'
'
N' = 4
sin
N,
cos
N' = e^.
cos
N,
e,
__
sin
cos -&
0.
and the
b'/b
s=s
-36, e'
= '048
so that unless e
is
small the
oscillations of
N or JV,
it
appears
or TT and that
must take place about the values
a
in
must oscillate
similar manner. But the argument,
that these
&-
tar'
The methods
TS' is
constant,
is
not necessarily
ment
problems
developed avoid the very difficulties which the actual problems present.
The methods given above apply to cases of resonance in which both
periods of revolution are present. The perihelia and nodes are angles which
in general revolve and there are possibilities of resonance relations between
their periods of revolution. In the comparatively short interval of time
during which observations have been made, such relations are unimportant
because, with the very long periods involved, expansions in powers of the
APPLICATIONS TO COSMOGONY
26, 27]
249
time give the required degree of accuracy. Comparable with these are the
periods introduced by the librations, and there are, therefore, further
possibilities for resonance relations. So long as the past history of the
new
solar system
was supposed
deductions as to
to be confined within
from
its
present configuration appeared to have some degree of value the extension of this interval to 10*
years or longer makes these deductions quite doubtful. The doubt appears
not so much in the ranges of values possible for the mean distances as in
;
in the future.
On
the other
hand, statistical evidence appears to indicate that these elements will tend
discussion of these and other
to be confined within narrow limits.
E.
W. Brown,
CHAPTEB
IX
The problem
finite solution in
bodies.
is
is
fixed
ellipses
lies
axis major of
tt
is
= sum
ellipses,
the relation
of the masses
Sun, apparently satisfies the given conditions, have been discovered, the first in 1901, more than a century after Laplace
gave the solution, and the last in the year 1932. They have re-
names taken from the Iliad of Homer and from this circumstance constitute what is usually called the Trojan group.
ceived
We
and
1,2]
The problem
differs
in several respects.
251
In the
first place,
and Jupiter
amplitudes and require special methods if an accuracy comparable with that of observation is to be secured. Another point,
brought out in Chap, vm, is the development in powers of the
square root of the ratio of the mass of Jupiter to that of the Sun,
instead of in integral powers of this ratio as in the ordinary
planetary theory since the square root of a small fraction is
;
much
vergence
consequence.
Still
another
libration.
Since the motion takes place in a fixed plane, the latter may
be used as the plane of reference. With the use of the equations
of 1*23, those
numbered
= v. Let us take
disappear and v
as origin and let the coordinates and
(5), (6)
r, v,
mi and
*-W
Jt)
_
dt*
~
\dt
d(
5
5V
~dr'
9/
'
dt \
dt
r', v' t
m'.
It is
252
where, according to
^
H
'
1'9, 1*10,
+ mi* ~
mo
~
__
,
I
J.'
AV)
in
fl
,
'
r
,
jp
7/?
+ w'
j_
are
..
~"~
r cos (v
V
r
(A
^j
elliptic
motion
dr
(5)
7 _,
^ I
*-*
/
'
\
^
/r
10)
v)\v
j
t/
60, with
According to 3'2(1),
From
r2
(A
zero,
v'))
/I
7o
2
(1
[CH. ix
(2),
these conditions
--_
~
~
dr'
--
'
dv
that
"~
r2
demand
fo'
we have
dF
-' _
w Q + mt
(r
r'
cos (v-v')
cos (v
t;'
dr
^ ~= _
dv
= mQ -f mi 4- rti'.
/u
The elements n> a,
are evidently the same for the two
2 8 =
n
with
a
7W
For the remaining elements
4- w'.
-f
m\
ellipses
we have e 6 =cr t*r'=4 60.
relations provided
the perturbations
methods.
is
of this chapter
EQUATIONS OF VARIATIONS
2, 3]
253
asteroid) is so small
other two bodies (the Sun and Jupiter), that it can be neglected
in the equations of motion. We then have F' = (m Q 4- w')/r' and
the motion of
will
disturbed values by
r
=r +
Sr,
= VQ -f Sv.
Sr, Sv
satisfied
when
8r, Sv
cedure gives
dt dt
dt )
d (
r
-r.
dt\
d
-j-.
dt
*
Sv
dv
+ O2r
\
5^
~-~ Sr
dt
)
<>
^-^-
^~22
~J7 ji
-Ji
&F \
= f 5-=-
S>
Sr
\drdv Jo
92
+ / 5-02
,
are
m'
5-^-=
9r3v
The
r
~ /I
Xa3
\A
(
limitations
IN
/a
r' 3 /
3m'
sm
,
(v
v
v /x)
= a = r = A,
/
VQ
v'
60,
55
ov.
\9^ /
')
,
Sv,
x)2
They
254
The
[OH. nc
s
or
2
n*or
zarz ~rA ov
dt
-m
= m j-=
3
4a
.3
^ ^* r
for
A/3
+
~ "T~24a2
dt
the latter,
,,
o?* ""
-f
-=-ra ov,
2
4a
m ,,^ r + 9m'
T4a 8v.
coefficients, their
The elimination
of the ratio
gives
X*
+ X'
8n
*-
a3
The use
of the relation
+
/
?i
a8
= 0.
T-
4a 8
a3
= r??o + ^i and
/
the introduction
of m, where
reduce this to
X 4 + X2 /i2 + ^-w4 /^ (1 - m) =
If
or
m < '04
the motion
0.
27//i(l-wi)<l,
approximately, the roots are all pure imaginary and
is
< *001 in the case of the Trojan
oscillatory. Since
If powers of
V^w,
2w
-T-
(1
m beyond
PERIOD OF LIBRATION
3]
With
27T/n
255
= 11*86
oscillation will
The
ratio
B:A
for the
m
3m'-ani-m
aB ~~
-27a3
?i
~~
m-9ra'
is
given by either
From
(2)
we
find
t,
+ T*
1
\/3m'
with the aid of the relations a?n2 = mQ + m' = m' jm> Since the
second of the two terms is large compared with the first, the
is 1
V3?Ai or 18'7
1.
As A\
is
The
which follows*.
are present in
We
*
now proceed
The small
oscillations
were
first fully
Lond
Soc. vol. 6 (1875) ; see also his Dynamics of Rigid Bodies, Part n, Chap, in
special case of them is treated by Charlier, Himmels-Mech. vol. 2, Chap. ix.
Math.
Incidentally, it
all
256
[OH. DC
is
,- x
(1)
where
/S
form
+m
F= mo + mi
+
r
'
,/l
D
Brsm
where
U~r
............... (2)
r'.
F can
R^ +
be written in the
R,
............ (3)
x 7
A 2 1 v*\
27*~ 2?j'
............ (4)
...
since the term | m'/r', thus introduced, can play no part in the
equations of motion which depend only on the derivatives of
first
derivative of
R with
respect to
The
mean
i+R
.......... (1)
GENERAL THEORY
4, 6]
Define a
new
Wi = n'
and in
257
-f e'
+ r = Wi + T,
Wi by
............... (2)
+ dr,
Since dw-i^n'dt
r.
8^1==8r, the
becomes
(3)
Since
all
d have
the
the dimension
d = Ci
(1
4-
X)
= \Va' (1 + #),
Since JK
is
near
us put
unity, let
is
dr
C2
#,
= Ci'^2
# 2 #3
,
C3
>
T,
= Ci'#3
w2t w9
still
.(4)
remain
a homogeneous function of
a, a'
of degree
and
factor
m = m'//^.
yet
We
mx2
and
it
is
eccentricities
but as
will
it will
much
follow.
In any
258
case
[OH. ix
we
With these
expanded
powers of
in
n'dt S (I a*
.
\2
where
is
W+
+ 2x
p
when a = a'
fJb
~]
OCLJ
As
Form of the
in
Chap.
of (5) can be
or
to order
x=
and becomes
x,
the value of
member
#4 ma?
. .
.(6)
and similarly
= a! (1
-f
for
2#) has
expansion of R.
of the eccentricities
there given
is
(cf.
4'13),
with
(JL
and
to derivatives of these
WI
WI=T, A becomes
a function of
lating angle.
Wi + Wi - 20 or
of terms of the form
g, g'
r,
wZ) WB,
Wi,
-cr'.
short periods and those in the latter have long periods or are
constant. The latter class is also distinguished by being inde-
is
an
ellipse.
5-7]
259
In general,
it
for observational
26 used in Chap.
',
orders
Arg. 0*,
>}
9'7.
0, e
9~9*>
orders
ee'
2#-2<7',
order
eV
e'
2
,
2
2
e\ e e' e'\
,
T2
IV,
e e
ee'*\
2
;
2g-(wi + wi'-20),
~~
~
2#)>
9 + o' ( w* + w i
or der 00'
- (w l + wi 2g'
order
order (?T
2(9),
f*
T.
is that used in
Chap, vi, namely, a change
of variables which leaves the equations canonical. Owing, however, to the fact that the variable r is contained in both coefficients
and
N^j'n't
as shown in
-f
>
#3, T,
K,
(6),
and put
are independent of x,
and
multiples of
T,
w2> w9 + const.,
9*6.
*
The term
of order
is
included in
260
means
n:
[CH.
by
sn
In this
sin jy
jn
\jn
cos jy
last expression,
Ki LI
t
L when
NQ
is
the value of
N when T
>
TO is
put
for
O,
To to be determined;
#3>
T,
^2, ^3*
new and
_as
XQ
dx>
3
__
=
" X _d_s_
aV
Cf
Ci
^ '
...(2)
__^_
==
'
9 S
and the equations for the new variables will still be canonical
provided we add d$/dt to the Hamiltonian function. From (1),
and
SS/dt
).
. .
.(3)
factor
from
m*
while
(2),
^ 3 = *o3
)
,
where SQ
is
the value of
8 when #
^2o> ^30
...(4)
of
Since the
denoting
by
by terms having the
c
factor
~z
oa
independent of
new variables
we can put
cO
/JL
"O
differ
/it
da
explicitly,
CHANGE OF VARIABLE
7,8]
261
terms of order
could
retained in the
Hamil toman
function, these
still
and dS/dt by dS
E, t
by R^o,
/dt.
If then in (4)
new
the
variables,
f*
2*
/JL
da
),
. .
.(5)
where, in all cases, the suffix zero denotes that the old variables
are replaced
by the new
The
which
relations
zero
and
is
a^
Q =
jV9V ^ ~JV8^'
= ZKrfN
still
it will
be reduced to
m%
will
............ (6)
when
the suffix
The remaining
9*8.
We
n S and
its
e, e',
F.
have, as in
cos
8 = (1 - JP)
cos (v
- v') + |F cos (v +
v'~ 20).
If we put
sCt, -<;')>
B&SPT
17
262
the expansion of
term of cos S as
[CH. ix
'
16
where
r'
RI==
cos 2
"A^
^ + 2 ^*~
this purpose,
+ '- 20)!,
(2)
2?*
'r
...(1)
For
e, e' y
F*,
put
a(l+/3),
~
(4)
so that RQ,
vanish. Taylor's
a
a'Ri_a'lt
~
+
~7~
r'
Z?i,
u'p
7-
r"
?R
8
a'dlto
+ ??
+ a'p* a2
2~7~
IT
S2 .Ro
3
+ -'
...... (5)
which
is to
be continued
to the fourth
powers of
p,
f for the
we
.&
2 (e
EXPANSION OF B
8, 9]
263
is
seen by a
jwo others have the same property, and two more differ only in
;he fourth order parts.
9*9.
The
coefficients
of the derivatives of JK
1 4- u\
a'/r'
1 4-
MI,
so that
of
-j'
pr
r
where
i -f
4.
The
of
= Ui
= a---
The
7
'
Ui
calculations appear to be
most easily
',
tables*.
the forms
ee'
sin (g
*
2 2
g'), e e' sin 2 (g
Mem. Roy.
g', so
that terms of
264
8
3
3
p /r' with f, f (in the fourth order terms the
takes the value a'). The remaining terms have the
products of p/r
divisor r'
[OH. ix
9
e' e)
sin (g
#'), and these disappear on account of
the relation between the coefficients of cos g, sin g in a/r, E,
form
(eV
respectively.
The
for those
- (2ee' + f ea e' -
f ee'
- (Zee' - feV -
3 r
'
cos (g
cos (g
e'
==
r
(4e
(3e
= ( e4 _ e '4) _ ( 2e3 e
ee' )
e, e',
'
e' 4-
4ee'
8
)
cos (g
which
is
the develop-
ment
To the second
order,
we have
...(1)
EXPANSION OF R
9, 10]
The
265
= ecosg
a' f/r' = 2e sin g
afpjr'
e'
cos g'
+e
cos 2g
we obtain
cos 2#',
e'
sin 2#',
w{ are
Wi
to
and
oa
a==1
it
substitute for
R Q beyond
Since
we can neglect
so that
by
-q
x.
266
Whence, when a =
1
T>
Q2 = 2 (1 - q),
9a 4
1 1
dRo
___
g8
[CH. ix
'
da 3
~
2
1_1_161
9 1
31
9T 2
_
~U
9'11.
'
,3
depending on F.
a'Rj/ji
argument
EXPANSION OF
10, 11]
In order to expand
it,
write v
+v
267
It
20 in the form
It
is
'-
20)
r, #, #'.
- 2(9) = cos (2wi' - 2(9) cos 2" - sin (2w/ - 2(9) sin 2"
= cos (2 Wl - 2(9) (1 - 4e' 2 + 4e' 2 cos 20')
- sin (2wi' - 2(9) (4e' sin #' + f e sin 2^'),
'
/2
and the only portions of this which will give long period terms
as far as the fourth order including those with factor F are
/
cos(2^i'-2(9)- 2e cos(2M; l
...... (1)
For
sin (2v
The
changed to
sines.
',
For the short period part which is taken to the second order
= Wi,v' w\ r = r' = a', A x = A = Q.
only, we put v
The third term of a'Rjp, in 9*8 (1), having the fourth order
5
2
a
factor F gives the single term 3F /16Q
The portions due to the factor 1 JF have been retained
throughout on account of the large numerical multipliers which
accompany them. As in Chap. IV, their presence causes but little
additional calculation, and adds considerably to the numerical
,
convergence
But
it is
easy
if
desirable to
268
R given in
,,
r
(^--Hr*-...,
~
s\
so that
where Q
9*12.
9*10 in powers of F.
=2-
2 cos
COS T
is
-no
COS2 T
,~.
......... (2)
r.
Transformation
The element #2
For,
2
d2 Q __ I cos r
~
~ r
1 cos
3Q __
[OH. ix
to the
given
(cf.
canonical elements.
and 513
9*5 (4)
2
*2 =(l+tf)f(l-* )*-l},
(1))
by
............... (1)
x$
= 61 sin w2 (ci/c/)*,
j9 2
......
(2)
first,
these give
esin
tzr
(l
i^
#).
^ costxr
F, 0, the substitutions
- \# -f J0) sin 0,
=
ps (2F)i (1
g3
= (2F)* (1 - \#
-h |a;)
cos 0,
...... (3)
= ^22 + 9 22
for
11-13]
in
The
through
e,
F when we transform
coefficient of
269
and in
i
-2a 5---!e
*
da
this expression,
^ 9 a'R
--r=
oTJ
9
5
de
JJL
second order.
913.
It has
T.
period terms.
The
The Hamiltoniau
new
variables
function, namely,
Q
An&
/-><
*Iff&
t\Aj
ft-iX/
Tt
VI
~\
C\
^O/
f 'in*
.
V\> C
_
"7T
oa
IJL
be denoted by
2
2#3
,-E
JLVg
/&
will
suffix is
-f (7
its
+ 2# F,
to order m*.
reference to 9'5,
<7,
Fare independent
t.
The equations
for #,
dx
r are then
= dU- 4-2^ 3F
-^
r^-
OT
OT
71 Cl6
~ r =.
n eft
3#
+ 6^
/1X
to order wi*
2F,
to order
(1)
(2)7
v
derivatives of
we obtain
;7T,- 2
^ tt>
270
[OH.
ix
On
substituting for dx/dt, dr/dt from (1), (2), we see that the
portions depending on
disappear and that the equation becomes
- 80/< ,1 - 4a? )
.
ss
The disappearance of
to order
m* .......... (4)
of which
it
is
x=
Since
^ -,
-y-
o n at
to order
becomes a
it
factor 1
-f-
where
m^
d2 r
"2
2
?i
dt
Ti2
2
Finally, if
it is easily
2 d (I dr\*
=
-TIT -/
o
3 n r/^ \?i T^
di/
I
we put
^=^
-f-
-S-2A + 3 ?T
is
-f-
dr
^
,
to order m*.
const.),
/Yl'^x-/-^
which
^ (r
- 3 dU
^~
x-1
=.
'
to order m*.
'
(5)
^
'
libration.
We
to r.
= C- 6U,
13, 14]
so that
n' fc
J (0
- 6 U)~*dr
From a theoretical
but the process
const.
r,
-f
271
it is
better
60,
is
to find r as follows.
Since
let
we know
us put
60
+ ST,
and expand the second term of 9*13 (5) in powers of ST. If J7be the ith derivative of U with respect to r with r =
60
the
becomes
inserted,
equation
When
In this
e'
case,
F = 0,
...... (i)
when powers
of Sr
beyond the
are neglected,
first
(1) reduces to
where
The
is
<f>,
v*
= 2?-m,
......
(2)
In the further
approximations, we
is
near
*3.
The statement
As
z>
= *079, we
in 9*5 that
have
is less
than
thus justified.
In general, when constant values of
Ji/&
= *008
in this case.
'01 in all
known
(see
is
e, e',
are used,
we
have,
approximation,
= & cos
2
i>
</>,
272
[OH. ix
_7X
= -3/i-f t/3& 2
The
addition to 8r
-7
&2
cos2<.
...(3)
is
is
TT
7T
TT
7y
TT
TT-
O JL
A,
"1
'
in the
- 3C/3 b cos $ ( or
-J
b
yj?
Y2
7j
The
argument 30
integral
particular
is
A '9-l
*
3(/2
3
fr
8Vi/2
</>
This shows that instead of the value 3C/2 for v 2 previously used,
,
we must put
in order that ST
may
still
remain periodic.
FURTHER APPROXIMATION
14, 15]
273
Sr
if
remain periodic.
is to
solved also
by putting
The
arbitrary constants
which become
9*15.
bQ
6,
The
6,
and
62 ,
Z>
...
coefficient of cos
first,
is
an
The development
0,
= const.
Further, since
2
e )* d/d' = T (1 - e )* (1 + x\
degree of accuracy, F = 3 = const.
#3 = T (1 -
or #3
,r
i^
>
p 2 = esin'cr,
The
has
a?
ecos
g^
tzr
................ (1)
pt)
g 2 become
__
n'dt~
with the briefer notation
9c
'
s for
ri
dt~
8s
p% and c for
'
q%.
274
The development
order, is given
by
[CH. ix
of
9*9 (1).
this
development
may
be written
T {- se' cos (r +
The equations
1
-,
n
1
-,
?l
for
s,
r')}.
c are therefore
d*
The only
is T
variable present in T
T has been expressed as a function of t by
,
with argument
</>.
9*16.
Solution of
As
usual with linear equations, we first find the complefunction, which is the solution of
is
mentary
--
dt
P = P + PI cos
where
with
</>
= vn't -\-VQ\P
<j>
s, c.
P2 cos 2< +
-f
............ (1)
^ '
0,
n dt
where
where
eo>
o)>
^ cos (*TI -f
p
''
(2)
We
(PP^n't
-f
TQ
+S
COS
PfiU't
+ Wo + S
thus have
\
~^ cos *0 )
lv
~ COS
>
'
p.
!</>
15, 16]
mean motion
mechanics
celestial
of the perihelion.
275
n'
is
the
sin
(P^nt
p
2~
tv
-f-
{sin
WQ)
(PQ n't +
c.
^ +
i<f>)
of the perihelion
is
divisible
by
e'
= e 2 (Si cos
c = e' S (Ci cos
r
of multiples of
<j>
The
{</>
icf>
9'15(3).
,..
In a
first
'),
Since
dcj>/dt
-(5)
satisfied only if
sc , cc
s,
</>,
m,
m*.
276
[CH. ix
as a
If, then, Po be the constant term in the expansion of
Fourier series with argument <>, the equations (5) and (6), with
the notation
(4),
give
expansion of
r')
PO SQ = const, term
- Tsiu (r + w'),
in the expansion of
m.
Ps, in
c iy c/, Si,
(5), (6)
approximation to these
coefficients.
Hence
s*, c*
are given
by
Sj
and they
the
first
{_
4- tar')} dt,
will have the factor m*. When these have been found,
of (5), (6) give s c c c the variable parts of which have
,
the factor m.
c it st
the factor
2
,
constant.
The
= 6 sin(60 + w'),
/
particular integral
c
is
is
reduced to a
then
= e'cos(60+'cr
),
...(7)
to the
(8)
HIGHER APPROXIMATIONS
16, 17]
277
is so
in
this
The
=#
cos
WQ
factor
CT
may
r.
9'17.
results.
/I
d*
~/2* j^s
\n
B&SPT
air
.,.
*\OT = additional
+ v.A*
)
terms
dU-%
or
iS
278
[OH. ix
now
first
paragraph
The
is
effect of substituting
variable for constant values of the variables with suffix zero will
equations
all
Numerical developments.
theory in which the expansions are made in powers of b, e r
m* can be formed which will give a close approximation without an
9'18.
A literal
e' y
excessive
amount of
Even with
calculation.
so large a value of b as
*3,
the
argument
<f>
in the calculation
r.
r=
and
lot
60
+ 2^008 ifa
d(t>ldt
v,
(1)
cos
<f>
+ eg cos2
<
4-
. . . .
17-19]
Then a
in
which
279
On
equating to zero the coefficients of cos 1$. The process can be repeated if
2
2
2
the principal part of
necessary. Since the principal part of 33 (7/d r is
Sr is found at once.
i>
The nature
tions
this volume.
A procedure
from these
An approximation to the short period terms can be obtained by substituting for the elements with suffix zero in the terms arising from S, their
osculating values. The same procedure is followed with the terms due to
By comparison
r,
6,
qf>
which are
The process does not differ essentially from that which would be followed
in the ordinary planetary theory if the methods of Chap, vi be used to
determine the perturbations. In the
at several dates
280
[OH. ix
special perturbations. Since all the short period terms have periods approximating to that of revolution round the sun or sub-multiples of this
period, a mean value of an element with suffix zero can be found by analysing
values at the various dates into a Fourier series with argument 2rr/w-'
first approximation to its value at
its
mean
date.
PERTURBATIONS BY SATURN
The
we
mation.
three,
is
and
due
revolution of Saturn
is
It will
PERTURBATIONS BY SATURN
19-21]
It will be
281
we can put
These additional
portions can then be separated into long and short period terms.
The latter may be eliminated as before by a change of variables
9*21.
The equation
for r
its solution.
laid
down
in the last
paragraph of 9*20,
(2)
where
t/"has the
constants.
amounts
this
t',
simplify the exposition, we shall put ti
to the neglect of terms of order higher than those
To
retained.
= TO
Let
where in
U and
its
= TO
is
(2).
inserted.
If
282
[CH. ix
The
We
can
the constants b VQ disappear identically.
differentiate (1) with respect to b and VQ and obtain
y
9T
'
db
*..
+
'
\fa)
n'* dt*
therefore
_
'
drf fa
...... (4)
Hence
gT
Sr=|?
ob
.................. (5 )
^<>
di/o
when
A = 0.
follows
A =f 0,
is
(6)
vy
i/J'O J
Ul/
where
Equation
(6)
may be
(7)
<j>
n't(dv/db)(dT/dvo)',
is
= 27m/4. With
this value of TO
is
6cos(im'
+ j>o)> where
we obtain
~p
will
21, 22]
283
Indirect perturbations.
9*22.
These
arise
for
the elements of
We
shall
Jupiter their disturbed instead of their elliptic values.
of
the
of
the
that
orbit
plane
Jupiter's
perturbations
suppose
reference.
of
Wi
a', e',
= w',
= w'.
wz
As
turbing function.
f
If we denote by 8a
equations
dd
-rr
at
dR
d*R
-T7=
ffjf
\Ci
-tto
Se' y SOT'
the perturbations of
a', e',
-or',
the
T become
2
d
~T*
dt
a /?
+ n ^T/S
=^drda
dr
dr
dt
for Ci,
+;Ta-/Be
drde
,
A
}
dR
d*R
+3-5- or
-/
PR
-- a a'
"5dci
dcida
,
>
.................. /1X
(1)
PR
~^i"a
,
'
ocide
-"d*R
->
BciPtsr
...... (2)
We
first
approximation.
284
[OH. ix
To
Sw'
Suppose
is
w'.
= B sin (n'pt + pQ
this in (3).
term
),
given by
8r =
But Sw = Sr -f
(5).
it is
Bp2
2
-%-_
sin (n'pt
4-
p Q).
Hence
J
ft
(4)
The
effect of
relative
magnitudes of v, p.
compared with p,
If v is large
(4) gives
approximately
Sw = $w'
general proposition:
of
one of the second order with respect to the masses while the
been treated
as short period effects and therefore do not enter into the discussion.
22, 23]
Terms
in
which p2
is
nearly equal to
v*
285
would give
rise to
much
is
9*23.
used.
When pjv
is
small, Sr is small
The
or, since
Hence
f 8n'.
= S&'.
The long period terms in the major axis of Jupiter are therefore
impressed on that of the asteroid.
For the perturbations of e, r, the equations 9'15 (3), which
are
still
true
if
ef',
BT'
Ps may be
Let
Then, as before,
80', SOT'.
it
e', or'
may
neglected.
receive
be shown
vanishes, the
reduce to
^8s-P S{e'cos(60
4V)}
^8c
=P
S {e'sin
60
+ w%
motion of Jupiter.
286
9*24.
[CH. ix
The disturbing
is
given by 1'10(1),
and it can be expanded in terms of the elements of the asteroid
and of Saturn by one of the methods used in the ordinary
planetary theory. The expansion, if made in a literal form, is
available whether the elements be constant or variable. The
disturbing function, denoted by R' adds a term a'-R'/A6 to the
Hamiltonian function in 9*5 (5).
y
is still
f a?
become
l^'-^J
~ 7 + ?E1
dr
n' dt
dr
n' dt
_'
then
(3) reduces to
^-_ql
n*d&~~
On
L^ T -__<}
I
'
(1), (2)
d_d_(dU
n'dt
dt\dx
JL
n'
rf2r
dt 2
JL + dU '\ _ 1 fiV'\
d
(
\ dr
~dr)
dt
\dx
For a term of long period, the last term of this equation is small
compared with dU /dr, and may be neglected in a first approximation.
to
J 8 ^j.Q^-.q^
8 +
'"
rfi
*ST~
dr
Saturn as a
24]
287
A sin(pn't-\-po).
The method of 9*21 is now available for the solution of the
equation. The principal part of the addition to r will be given by
The principal long period term due to the action of Saturn will
have nearly the same period as it produces in the motion of
Jupiter, namely, about 870 years, and therefore p is small compared with v. Hence the small divisor due to a long period term
much
is
and the
members of the group, that the value of the coefficient of the principal
long period term obtained in this way is doubtful. Another difficulty arises
from the fact that the mean motion of the perihelion of the asteroid is
comparable with that of the argument of this term so that
*
Paris Obt.
Mtm.
vol. x.
it
cannot be
288
[OH. nc
Group
"
9 '25. The literature connected with the triangular solutions of the pro-
W. Brown
(Mon. Not.
vol.
a linkage between the orbits of planets outside and inside that of Jupiter
and those of satellites of Jupiter, the passages between them going through
the collinear solutions. The linkage bears some resemblance to that which
joins the two sets of solutions of the equation for the motion of a pendulum.
A literal development sufficiently complete to give the position of an
asteroid of the group within a few seconds of arc has been made by
W. Brown ("Theory of the Trojan Group of Asteroids," Trans, of Yale
E.
The method
of Chap,
vn
is
with in
libration of
A.
Let
A'l.
is
= CQ + Cicos x + C2cos 2# +
(x)
. . .
is
+ c n cos nx +
2?r,
which
....... (1)
It is supposed that
c (x) conc^.
tains numerical constants only, and that after some term c n cos nx
the remainder of the series may be neglected.
Under the
latter
may be
condition (1)
If then
regarded as an
we choose n
-f 1
Ti
+1
relations
giving the n
The
unknowns
Ci, ...
cn
+1
linear equations,
effectiveness of the
tion of
(x)-Sismx + s%sm2x+
... -f
sn
smnx+
is
no con-
stant term.
If
if
we
of
(x) is
each case,
APPENDIX
290
[APP.
2?r
a),
Since
sin ia =
sin i (2zr
a),
we have
F(a) + F(%7r -a) = cosine
In future
will
it
series,
separately.
Next, since
(d)
+ Fc (7r-a),
F (a)-F
(7r-a)
F (a)-F
c
(7r-a),
(a)
+F
(ir-a)
two
Sj
sets, one containing the coefficients with odd suffixes
and the other those with even suffixes.
into
of Yale Obs.
vol. 6,
part
4,
pp. 61-65.
Since
Determination of the nth harmonic in
s
8
= 0, TT, these values are useless for the computation of the S{. Moreover, since sin nx vanishes when x is any
multiple of TT/W, it is the coefficient of the rath harmonic which
A'3.
vanishes for x
is
undetermined with
It is
sometimes
HARMONIC ANALYSIS
A-2-A-4]
291
value #
c 2 (see A*4).
For n
4,
we may proceed
as follows
calculation of
give the
the relation furnished by the other values of x inserted in 8t will
give the value of s n Only one of the two values is necessary,
.
the (n
l)th harmonic.
d + C2n -i + Ctn+i + C 4n -i +
if
we
include
< n,
Thus the
c t is
-
is c 3/l .
the value of
of the error of
we
all
c t-, i
principal part
When i = n,
Si
instead of
Si,
for i
<
/?,
4" fyn+i
&4ni
"^
is
$2n-i, and tne same rule with respect to the errors of the
coefficients of the lower harmonics holds.
When the harmonic with coefficient sn (which vanishes when
is a multiple of TT/U) is found for n odd by using the additional
value x = 90, it is easily seen that we actually determine
APPENDIX
292
for
without
find sn _ 2
its
sn
2sn+2
$ n is
use we find
+ sn + g
If
. . .
90
[APP.
sn_2 - sn +2 +
and with
cn _!
When n
# = 0,
TT
is
its
use
we
cn +a,
the errors
gives us
sn _i
+ s n+ s +
. . .
sn
+ 2sn + 2 +
. . .
s n +i
+ sn +3 4-
. . .
Thus
in the case of
A'6.
subtractions,
routine computation.
In the cases where the coefficients are
this reduction to
known
to diminish
little labour.
denoted by
we put
and
The
function to be analysed
x.
F = C + Ci COS X + 02 COS 2x +
. .
an odd function
for
special values of
HARMONIC ANALYSIS
A-4-A-6]
TO,
I7
^0
'
>
I7
I7
F will
37 '.
^4, ^4
^9;
be denoted respectively by
XT'
JP
^3> ^3
IT*
I?'.
^6, ^6
known
it
would be subtracted
and
all
in order
the examples
B&SPT
its
19
APPENDIX
294
A'7.
[APP.
SERIES.
A'8.
Values of x
= 60,
120, 90.
31053
'
22787
31085
30000
04772
53840
01085
31085
01085
08266
04772
HARMONIC ANALYSIS
A-7-A-10]
A'9.
295
Values of #
= 0,
.F
= (l -
-6cos#)*.
19-2
APPENDIX
= 0,
63246
F,
[APP.
1-26491
1-00000
- -53961
69311
- -93597
1-23272
-93463
83666
-1-87060
1-14018
- -00134
A =
B =
C =
1-89737
2c3
2-00000
-31199
- '00847
1-92583
97523
1-97684
3-89737
=(7+1)
4c
=^+2c4
-2c 4
3*90267 4c
-00530
-15364
- -05121
- -00020
- -02561
- -00423
-00089
- -00022
'00177
3-90090
- -10263
- -05101
-31177
-00005
2
- 4c e
^+5
A-11,A*12]
HARMONIC ANALYSIS
)-
298
APPENDIX
(a>,
y)
^,
where J,j'
efficients
= 2 (A jtj
When
a function of two
+ G*,f cos j
>aj
>
= 0,
>
[APP.
1, 2, ...,
harmonic analyses.
First, the choice of pairs of values a, 2?r
a separates the
Thus the
Consider the
first
analysis is
values for x, y
group given by
special values of
A XtV
x,
and each
The
Aj
. . .
for
y.
The
The process for finding the BJJ is the same with the exception
that the cosine analyses are replaced by sine analyses.
For Cjj>, the first block of analyses is that for cosines, and the
second that for sines.
A-13-A-16]
The
is
^299
A*15.
The
derivatives
cos jx terms in
disappear from dF/dx when x = 0, TT ;
those
of
the
cosj'y terms from dF/dy when 2/ = 0,
similarly
and
TT.
A*16. In the
and disturbing
vanishes for
0, TT
or for g'
= 0, TT,
is
replaced by a derivative
shown above.
as
When
An
sets.
INDEX
(The numbers refer
Anomaly,
true,
mean,
eccentric, 64
Approximations
in terms of time
first, 146
second, 156
in terms of true longitude
as independent variable, 30
elliptic expansions in terms of, 68
disturbing function in terms of,
99
270
Elements of
second, 277
Asteroids, Trojan, 250
resonance effect on, 245
Astronomical measurements, 5
Astronomical unit of mass, 7
Attraction
Newtonian law
of, 7
proportional to distance, 136
Bessel's functions, 55
66
discussed, 141
Declination, 5
(see
Development)
Earth-Moon system, 11
Eccentric anomaly, 64
176
178
second, 198
first,
d'Alembert
Disturbing function, 10
for double system, 11
for satellite problem, 13
for Trojan asteroid, 256, 258
(see
elliptic,
first,
to pages.)
Equations of motion)
by harmonic analysis, 80
Elliptic motion, 62
fundamental relations, 64, 65
Jacobi's method, 127
true longitude theory, 176
Elliptic variables, 16
Delaunay's, 131
Force function, 8
Fourier expansions
Chap. II)
(see
Contents,
INDEX
302!
Harmonic
analysis, ^289
Fertprbations (wntinued)
"'
Potential function, 8
Power
series,
121
Jacobi's transformation theorem, 119
Jupiter's effect
on Saturn, 205
on certain asteroids, 245
on Trojan asteroids, 284, 287
Kepler's equation
first used, equation (16), 64
numerical solution of, 81
Kepler's laws, 66
third law discussed, 7
Reference frames, 15
Bqsonance, defined, 4, 2J.6 (see Contents, Chap. VIII)
Bight ascension, 5
Satellite form of equations, 11
Satellite problem, 2
Saturn's effect
on Jupiter, 205
on Trojan asteroids, 284, 287
Latitude, 6
Latitude equation, 29, 193
200
Short period terms
Law
of gravitation, Newtonian, 7
66
Libration, 230, 255
Long period terms
denned, 153
second approximation to, 161, 202
of disturbing planet, 165
case of a single, 171
Laws, Kepler's,
7,
Longitude, 6
true, as independent variable, 28, 174
mean ,64
Mean distance, 66
Mean motion, anomaly,
longitude, 64
Measurements, astronomical, 5
Osculating ellipse, 16, 126
Osculating orbit, 125
Osculating plane, 15
elimination
effect of,
of,
143, 259
on second approximation
159
Small divisors
source
of,
84
discussed, 154
in true longitude theory, 193
Solution
of canonical equations, ,*138 (see
Contents, Chap. VI)
of true longitude equations, 185 (see
Contents, Chap. VII)
of equations of variation, 151, 197,
253
Perturbations
125
equations
of,
CAMBRIDGE PRINTED BY
WALTER LEWIS, M.A.
AT THE UNIVERSITY PRESS
: