Webster Theory of Electricity and Magnetism, Being Lectures On Mathematical Physics (1897)

THE THBOEY
OF
ELECTRICITY AND MAGNETISM.
THE THEOEY
OF
ELECTRICITY AND MAGNETISM

BEING
.ECTURES ON MATHEMATICAL PHYSICS
BY
ARTHUR GORDON WEBSTER,
A.B. (HARV.), PH.D. (BEROL.),
ASSISTANT-PROFESSOR OF PHYSICS. DIRECTOR OF THE PHYSICAL LABORATORY,

CLARK UNIVERSITY, WORCESTER, MASSACHUSETTS.
Honfcon
MACMILLAN AND
NEW
CO.,
LIMITED
YORK: THE MACMILLAN COMPANY

1897
[All rights reserved.]
(EamtmUge
PBINTED BY J. AND C.
:
F.
CLAY,
AT THE UNIVERSITY PRESS.
PBEFACE.
SOME
justification is perhaps necessary for the
appearance of
another treatise on Electricity and Magnetism, in view of the
numerous ones already existing

result of a
and
is
demand encountered
in
in
my own
based upon various courses of lectures that
of Maxwell, which
must always remain
To ask a student
the
The
have delivered
classical treatise
as a point of departure
the modern treatment of the subject,
purpose of a text-book.
is
experience in teaching,
at Clark University during the last six years.
for
This book
English.
is
ill
adapted to the
to attempt to assimilate
the contents of the two volumes of Maxwell in a year, or even

in
two
years, is only to expose
him
to the severest
pangs of mental
Again, Maxwell's own views are there presented by

him with not the greatest clearness, while severe demands are
made upon the student's mathematical attainments. The excel-
indigestion.
lent treatises of Mascart
follow
and Joubert and of Watson and Burbury
Maxwell with considerable
closeness.
Professor
Gray's
admirable treatise, though containing much recent matter, suffers

under the disadvantage of being in three volumes, while the very
convenient little book of Mr Emtage is somewhat restricted in
scope.
Professor J. J. Thomson's altogether delightful Elements
of the Mathematical Theory, which

appeared
w.
E.
when the present

b
PREFACE.
VI
book was nearly ready

as clear,
for
the press, while extremely modern as well
addressed to a somewhat different class of students
is
from that contemplated in writing the present book.
The
theoretical writings of Hertz, Heaviside,
Cohn and
others
have resulted in the systematization of Maxwell's theory and have

made possible improvements in the mode of its presentation and
nomenclature not contemplated by him. The extremely important
and original contributions of Mr Oliver Heaviside are unfortunately
adapted to the use of the student on account of their

character as a whole, as well as of an extreme
voluminous
very
The few brilliant
conciseness of expression in individual parts.
but
little
chapters on theoretical matters

of exposition, but rather of a
left
by Hertz are hardly by way

of the conclusions of
summing up
the theory.
It has
been
my
aim
in the
of this
preparation
volume to
present to the student the results of the theory as it stands to-
day
after the labors of Faraday, Maxwell, Helmholtz,
Here
Heaviside.
it
may be
Hertz and
convenient to state what I consider
be the essentials of Maxwell's theory as distinguished from

the old theories. To this question may very well be made the
to
answer of Hertz
But
tions."
"
:
Maxwell's theory
to specify
more
in the opinion of the writer
The
1.
is
Maxwell's system of equa-
fully the points of difference, they are
localization of the
energy in the medium.
The magnetic action of displacement currents.

While starting from the standpoint of Energy, I have not
2.
thought
many
it
advisable to abolish the usual terms repugnant to so
writers,
excellence,
who assuming the
deny the existence of
was not one of
of Maxwellians
attitude
Electricity.
par
Maxwell himself
Feeling that the consideration of the

Newtonian Potential Function is indispensable, not only for the
these.
old theory of action at a distance, but for the

in addition that
it
introduces the student to
modern
many
theory,
of the
and
methods
that he will need in various branches of mathematical physics, I
PREFACE.
vii
have prefixed to the treatment of Electricity a rather complete

treatment of the Potential by itself, including the properties of
been the custom of English writers

to include chapters on the Potential in works on Analytical Statics,
as in the cases of the admirable treatises of Routh and Minchin.
It has
polarized distributions.
It will probably be admitted, however, that the inclusion of this
subject in a treatise mainly devoted to the consideration of rods,

strings,
and
billiard balls is
devoted to Electricity and
no more appropriate than in one

likely to attract the student of
is less
the latter.
It is unfortunately the case that graduates of our
American
colleges are as a rule insufficiently prepared in the departments
of mathematics necessary in approaching the subject of mathe-
matical physics.
In
know
fact, I
of but three text-books on the
Calculus in English, those of Greenhill, Williamson and Byerly,

that give a treatment
considered
of Green's
Theorem.
have therefore
expedient to prefix a mathematical introduction
it
giving a short treatment of the important subjects of Definite

Integrals and of the Theory of Functions of a
Complex Variable,
For the same
indispensable to a study of the Potential Function.
reason, I have included a treatment of the fundamental principles
of Mechanics ab initio, including the deduction of the Principle of
Energy, Hamilton's Principle, and Lagraiige's Equations of Motion.
have followed the example of Boltzmann in making the deduction

of the equations of the Electromagnetic Field depend on Hamilton's
by means of the properties of Helmholtz's Cyclic Systems,

These chapters are exthe treatment of which is here added.
Principle
tracted from
my
lectures on Dynamics.
come about that the book

electricity
mentioned.
is
is
In this manner
it
has
nearly half finished before the word
This
may
be objectionable to some
persons, but I consider it of great importance that the student

should be well supplied with tools and practised in their use
before he
is
subject.
The
called
upon
to use
them on a new and unfamiliar
physical difficulties
connected with electricity are
PREFACE.
viii
mixed up with mathematical ones.

great enough without being
have
the
student get the idea that certain
to
a
It is also
pity
theorems pertain to electricity, when they really are simply
I have whenever possible atmatters of geometry or analysis.
tempted
processes
to bring out the geometrical or physical nature of the
involved before coming to the electrical application.
Thus these introductory chapters may serve

introduction to
as a sort of general
Mathematical Physics.
After a treatment of the problem of Electrostatics in a single
medium by means
of the Principle of Virtual
methods of attacking
electrostatic
chapters pertain to either the
Work
the usual
These
problems are treated.
new
or the old theory.
have
then inserted the chapter on Electrokinetics, somewhat out of
its
natural order, in order to bring out the geometrical ideas involved
Law
Ohm.
Of
made
in
the treatment of Dielectrics and Magnetizable Bodies, which
is
in the so-called
of
carried out in such a
between the two
manner
classes of
these application
as to
show the
phenomena there
is
close parallelism
treated, a point not
always insisted on by Maxwell, but clearly brought out by Hertz

On account of this the symmetrical notation of
and Heaviside.
I have
adopted in preference to that of Maxwell.
however kept the term induction used by Maxwell for magnetism
Hertz
is
alone, instead of the
term polarization used by Hertz, which I
have used in the more usual sense of moment of unit volume.

regret not having been able to respond to the appeal
Boltzmann
more important
it is
one,
and that
it
is
Roman and German
I feel
have a good notation than a familiar
to
first
should be symmetrical.
as
made by
to future writers to follow Maxwell's notation.
that
essential of a
The
letters
good notation that
it
indiscriminate mixture of Greek,
used by Maxwell
the dissymmetry with respect to
is
electrical
as unfortunate
and
magnetic
phenomena.
It is hardly necessary to say that vector
methods have been
used throughout, although the abbreviated notation of Hamilton
PREFACE.
and Heaviside has been
little
used.
IX
It is easy to lay so
much
on symbolism that the student loses sight of the real simIn order, however, to show its extreme
plicity of the method.
stress
connection with the operator V, the essen-
utility, particularly in
tials of
the quaternion notation have been explained in the
first
chapter.
As the aim
book has been to present an introduction

to the mathematical theory of electricity, little or no reference
has been
of the
made
to experimental
such subjects as standard
methods
in fact it
seems that
cells, dynamos, or galvanometers should
be treated in a separate work, and I have no desire to add to the

large
number
of such already existing.
At the same time
hoped that the principles involved in the various

surement are
The
of
all
of
it is
mea-
herein contained.
figures with
which the book
them are new, have
figures,
modes
but have been,
if
is
illustrated, while
but a few
no case been copied from existing

necessary, recalculated, and in every case
in
redrawn on a large scale and photographed down to the required

size.
For the amount of labor here involved I am under great
obligations
fellows
to
Messrs
of Clark
W.
P.
University,
Boynton and
T.
W. Edmondson,
who have undertaken the whole
As the proof has been read only by the author, it

probable that a certain number of errors have crept in, which
matter.
is
it
be excused.
hoped may
In conclusion
treatise
is
my
aim has been
to present a brief, connected
embodying the essential points of the theory and suitable

by the student in a period of time not exceeding
To this end I have considered only the usual methods of
for assimilation
a year.
treating the various subjects, and included enough examples to

illustrate their working,
and no more.
If
it
be considered that
unnecessary matter has been included it may be replied that this

may easily be omitted, and that it is safer to include too much
than to make unwarrantable assumptions regarding the knowIf the book shall succeed in
ledge possessed by the student.
PREFACE.
x:
clearing
up some
of the difficulties generally encountered
by the
student and in inducing him to read the classical writings of
Maxwell, Helmholtz, Hertz and Heaviside the object of

will
its
author
have been achieved.

A. G.
WORCESTER, MASS.,
Dec. 23, 1896.
WEBSTER.
LIST OF
WORKS CONSULTED BY THE AUTHOR
Maxwell, Treatise on Electricity and Magnetism,.

Green, Mathematical Papers.
Heaviside, Electrical Papers.
Thomson, Mathematical and Physical Papers and Reprint of Papers on

Electrostatics and Magnetism.
Thomson, J. J., Recent Researches in Electricity and Magnetism.
Thomson,
J. J.,
Elements of the Mathematical Theory of Electricity and
Magnetism.
Gray, Absolute Measurements in Electricity and Magnetism.
Faraday, Experimental Researches in Electricity.
Kirchhoff, Vorlesungen ilber Elektricitdt
und Magnetismus.
Helmholtz, Wissenschaftliche Abhandlungen.

Kirchhoff, Gesammelte Abhandlungen.
Hertz, Gesammelte Abhandlungen.
Gauss, Gesammelte Werke.
Elektricitdt und Magnetismus.
Vorlesungen iiber Maxwell's Theorie der Elektricitdt
Riemann, Schwere,
Boltzmann,
und
des Lichtes.
Drude, Physik des Aethers auf elektrbmagnetischer Grundlage.
Winkelmann, Handbuch der Physik.

Mascart et Joubert, Leqons sur V Electricite et le Magnetisme.
Mathieu, Theorie du Potential et ses Applications a I' filectrostatique
au Magnetisme.
Duhem, Lemons sur V filectricite
Poincare, $lectricite
et le
Magnetisme.
et
Optique.
Poincare, Les Oscillations electriques.
Betti, Teorica delle Forze
Newtoniane.
Ewing, Magnetic Induction in Iron and other Metals.

du Bois, Magnetische Kreise, der en Theorie und A nwendungen.
Thomson and
Tait, Treatise
Routh, Analytical
Statics.
on Natural Philosophy.
et
LIST OF
Xll
WORKS CONSULTED BY THE AUTHOR.
Minchin, Treatise on
Peirce, TJie
Statics.
Newtonian Potential Function.

Mechanik.
Kirchhoff, Vorlesungen uber Mathematische Physik

Somoff, Theoretische Mechanik.
F.
Neumann, Vorlesungen uber
die
und
Theorie des Potentials
der
Kugelfunctionen.
C.
Neumann,
Untersuchungen
Newton''sche Potential.
uber
das
und das
logarithmische
Harnack, Die Grundlagen der Theorie des logarith-ndschen
Potentiates.
Bacharach, Abriss der Geschichte der Potentialtheorie.

Jacobi, Vorlesungen iiber Dynamik.
Appell, Traite de Mecanique Rationnelle.
Kronecker, Vorlesungen
iiber die
Theorie der einfachen
und
vielfachen
Integrale.
Riemann, Partielle Differentialgleichungen.

Lame, Lemons sur les Coordonnees Curvilignes
et
leurs diverses Appli-
cations.
Darboux, Theorie des Surfaces.
d Analyse.
Cours d Analyse.
Picard, Traite
Jordan,
Laurent, Traite
$ Analyse.
ERRATUM.
Page 140, lines 12 and
13, for revolution
read translation.
MATHEMATICAL INTRODUCTION.
CHAPTER
I.
NUMBER.
Rationals and Irrationals. The primary objects of
Arithmetic are the natural numbers or integers, 1, 2, 3,...
an
unlimited sequence. Of these any two a and b may
forming
be added together, and we find the fundamental law that
1.
study in
a
This
known
is
as
+b=b+
a.
the commutative law.
For more than two
numbers, we find that

(a
This
is
known
+ b) + c = a + (b + c).
as the associative law.
multiplied together, and

the commutative law,
we
Any two numbers may be
find that multiplication
ab
is
subject to
= ba,
to the associative law,
(ab) c
(be),
and in addition to the distributive law,

a (b + c) = ab +
ac.
Defining the operation of subtraction as the inverse of addition,

so that c is defined as the result of subtracting b from a if b added
to c will give a, we find that the operation of subtraction is possible only if
definition of
must be added
less
than
W.
E.
b.
is
greater than
numbers
b.
in such a
We
way
are thus led to extend our

as to call that to
which b
in order to give a, a number, in the case where a is

are thus led to the conception of the negative
We
NUMBER.
[INT.
which we can add, multiply, and subtract according

same laws as the natural numbers.
integers,
Denning the operation of division as the inverse

a divided by b is c, if c multiplied by
tion, so that
I.
to the
of multiplicab will give a,
we
find that the operation can be performed only when b is a factor

of a.
are thus again led to extend our definition of numbers
We
so as to call that which, being multiplied by b, will give a, even

when b is not a factor of a, a number.
are thus led to the con-
We
ception of fractions, which
may be
operated upon like the positive

and negative integers. Every fraction is of the form m/n where
and n are positive or negative integers. This system of
numbers suffices for all the ordinary operations of arithmetic,
including the solution of equations of the
Any number may be
first
degree.
any power, the process being

define evolution, or the operation of
taking a root, as the inverse of involution, so that the 6th root of
a is c when c b = a, we find that the operation can be performed
known
when a
only
If
as involution.
is
raised to
we
one of the series of numbers,

c,
2
,
3
,
we
further extend our definition of number, so that that

which raised to the 6th power will give a, even in the contrary
If
case,
we
irrational
are led to the conception of irrational numbers.

No
number can be expressed as the quotient of two integers,
for any given irrational a we can always find two

integers
such that their quotient differs from a by an amount that is as
In symbols, if is any given positive number
small as we please.
though
as small as
we
please,
we can always
m
n
By
is
meant the
<
find
m and n so
that
e.
absolute, or arithmetical
value of
a, irre-
E.g., the square root of 2 is an irrational, but

spective of sign.
the rules for the extraction of square roots enable us to find a
it by as little as we
The ordinary
please.
of
enumeration
shows
that
we
can
express any rational
theory
number in terms of any integer, 6, called the base, as the sum of a
definite number of terms, each of which is the product of some
value that differs from
integer less than b by some power of b with positive or negative

exponent, or else as the limit of a sum of such terms, where as th
1,
NUMBER.
2]
exponents of the negative powers increase regularly in absolute

value, from a certain term on the coefficients are all obtained by
the repetition of a certain definite group
:
= anb n + On-
bn
~l
. . .
^=
e.g.
= TIO + 010 + MO- + 610- + 610- +
No
number can be so expressed, though by taking a

of terms we may obtain a number differing from
the given irrational by as little as we please. The coefficients in
irrational
sufficient
number
this case never repeat indefinitely.
expressed by means
of a finite
Since irrationals can not be
number
of terms each of which
is
they are defined by their properties, or as the limit

approached by an infinite sequence of rational numbers.
rational,
Limits.
If we have a sequence of rational numbers,
2.
a 1} Oa, a 3 ..., following each other according to a given law, and
can find a number A, possessing the property that, corresponding
,
to
any
find a
arbitrarily given positive
number
/A
number
however small, we can

than
for all values of n greater
such that
/-t,
'
an
then
is
called the limit of the sequence.
E.g. the
sequence
has the limit
2,
a rational number.
The necessary and

is that when e is
limit
such that
A\<e,
sufficient condition for the existence of a
we can find a number

n greater than
and for any
arbitrarily given,
for all integral values of
positive integral value of p,
an - an+p <
\
/u,
//,,
e.
even though the sequence

has no rational limit, the sequence has a limit, which defines an
If the latter condition is fulfilled,
lc
irrational
number.
12
NUMBER.
[INT.
The sequence
0!
= !,
does not
a2
=l +
2,
a3
I.
111
11
=l+ +
^ = 1+^ + ^ + -,
2
g,
the above condition, for
fulfil
-i
Ti+1
ct
Ti+2
+p
7i
n'
n+p
P
which cannot be made as small as we
matter how great n be taken.
On
please, for all values of p,
no
the other hand, the sequence
a1 -I,a t
-l + _L
l ^>
as
- 1 + _JL + _!_
i.2 1.2.3'
1
1.2
'
'
1.2.3
1.2.3.4'
does satisfy the condition, for
&n
~ an+p
n+1
1.2.3
1
n
:i+
1.2.3 ...... Ti
^+
*^. +
.+
..
1.2.3
"1.2.3 ...... 7i(n+l)2

which
is less
than
'
as soon as
is
taken greater than
1/e.
This sequence defines the irrational known as e, the natural

is not the root of
any algebraic equation
with rational coefficients. The class of irrationals is in fact much
logarithmic base, which
larger than the class of algebraic irrationals, which led to their

inclusion in the number-system.
3.
and
Complex Numbers. The system composed
irrational,
forming together the real numbers,
cient for the solution of algebraic equations.
simple equation
#2
= 0.
of the rational
is still
not
suffi-
For consider the
NUMBER.
24]
real
Since even powers of all real numbers are positive, there is no

number that has a square equal to 1. If we further extend
the idea of numbers, so as to call that a number whose square is

If we denote the
1, we have a means of satisfying the equation.
new number by i, defined by the equation i 2 = 1, we may multiply
number, positive or negative, integral, fractional, or

and thus get a class of new numbers, known as pure
imaginary numbers. Evidently no imaginary number is equal to
a real number, for the quotient of two real numbers is always real.
it
real
by any
irrational,
If
we
consider the
sum
and an imaginary number, we

complex number (in the narrow
of a real
arrive at the conception of a
Two complex numbers are equal when their real parts are
and
their imaginary parts also.
Any equation containing
equal
is accordingly equivalent to two equations connumbers
complex
In particular the equation
taining only real numbers.
sense).
where a and
+ bi = 0,
b are real, is equivalent to the two,
and
= 0.
complex number vanishes only when

parts both vanish.
4.
Complex Numbers
in the
real
its
and imaginary
Extended Sense.
As we
have formed numbers by multiplying the real and imaginary units

1 and i by all real numbers and forming sums therefrom, so we
may still further extend the notion of numbers to include sums of

terms each formed by multiplying any number n of different units
by real numbers. Such numbers are complex numbers in the
extended sense, a number involVing n units es being an n-fold
number. The units may have any properties by which we wish to
define them.
If they are all independent of one another, it is
obvious that two complex numbers are equal only when composed
of the same number of each unit es so that any equation contain,
ing
all
the units
numbers.
is
n equations containing only

a complex number
equivalent to
In particular,
a = a^j
vanishes only
when the
+ae
2
coefficient a8 of
a. n e
real
each unit
es is zero.
Two complex quantities satisfy the associative and commutative laws with respect to addition, and accordingly the sum of
i
4-
0^2
OnCn
and
ft-fii
+ y3
+ @n#n
NUMBER.
6
is
[INT.
I.
defined as
With
respect to multiplication, the units are

to
real numbers
respect
commutative with
aber ces
The
associative
(aer ) es
= aer bce
= er e abc,
abcer e s
and distributive principles
= a (e re )
and either a
or
We may accordingly
t)
also hold, so that
(er -f e s )
+ e = e r e + er e
er (es
etc.
= aer + ae8
any two com-
define the multiplication of
plex numbers
ab
= (ctA +
2e2
......
a n en )
(& + &e.
......
+ /3n en )
It will be convenient to consider a system of units of such a

nature that instead of the commutative property with respect to
= eser where r and s are different, and
multiplication we have er es
for
any
r, er
1.
we
consider a set of three units, each possessing the above

properties, and in addition the property that the product of any
If
we have the system

and
denoted
him
proposed by Hamilton,
by
by the letters i, j, k.
definition
Accordingly by
two taken
in cyclic order is equal to the third,
ij
ji
= k,
jk
kj
i,
ki
ik =j.
Multiplying each equation by the first unit appearing in

observing the associative law, we have
it,
and
= i*j = _ iji = - (ij) i = -ki = -j,

= fk = -jkj = - (jk)j = -ij=- k,
jjk
iij
kki = k2i
kik
necessitating
i'2
= js =
(ki)
k=
jk
i,
=_L
Even powers of the three units are real, and equal even powers
are equal, while odd powers of any unit are equal to real multiples
of itself, and equal odd powers of different units are not equal.
The product
and
6,
of
two threefold complex numbers of
this system,
NUMBER.
4, 5]
+ 0j + yk) (Y+ 0'j + y'k) = aa'i + /5fff + yy'fc + affij + /Sa'ji

= - (aa' +
+ 77)
ya'ki + ay'ik
8
(ai
'
is
accordingly equal to a real number plus a complex number of
the same system, and this may be considered as a fourfold complex
number compounded of the units 1, i j k. Such a fourfold number
t
was
We
by Hamilton a Quaternion.
seldom need the fourfold number, but
called
shall in this
book
shall frequently use the
threefold one.
5.
natural
The
Geometrical Representation of Numbers.
numbers may be represented by an unlimited series of
If we take
points laid off at equal distances along a straight line.
a certain point to represent zero, the positive integers will lie on
one side of
and the negative on the
it
other.
Points between the
integer points will represent fractions and irrationals, and to every

number will correspond a point. For any rational number we
real
find others lying as near it as

already stated, for any irrational we
may
we
please, and as
may find rational
we have
numbers
It may be shown, however, that

lying as near it as we please.
between any two rational numbers, however close together, there
can always be found an irrational, consequently the rational
numbers do not form a continuous series. It may be shown that
every point on the line corresponds to either a rational or an
irrational
number,
so that the
whole
series of real
numbers
is
continuous.
Quantities which, like the real numbers, require for

their specification but a single given quantity, which may take
any of an unlimited series of values, are said to have one degree
of freedom.
It is also said that there is a single infinity of
such
quantities.
Complex
quantities
in
the
narrow
sense,
involving
two
and
i, cannot be represented by points on a line.

If however we lay off the real numbers on a straight line, we may
lay off the pure imaginary numbers on a line at right angles with
different units, 1
it
through the point representing zero.

same distance from zero on this
at the
the other
line.
The two
The point
i is
to be taken
line that the point 1 is
on
lines are called respectively the axes of
and of pure imaginaries, or the axes of

and Y. Any
=
number
a
a
now
be
+
/3i
complex
may
represented by a point in
the plane whose rectangular x and y coordinates are respectively a
reals
NUMBER.
[INT.
I.
Whatever the values of a and (3 we may always find a

corresponding point, and to every point in the plane there corresponds a single complex number, including the real and pure
imaginary numbers as particular cases. As each of the real
numbers a and
may independently assume the value of any of
and
/3.
/5>
the single infinity of real numbers, there is said to be a double

infinity of complex numbers, or a complex number has two degrees
of freedom.
The distance of the representative point from the
origin
= + Va + #
2
by
modulus of the complex quantity and denoted
called the
is
since
it
includes
absolute value of a real number.
as
a particular case the
The angle that the
radius vector
from the origin makes with the X-axis is called the argument of
the number. This representation of complex numbers in the plane
was proposed by Argand and Gauss*.

threefold complex quantity a = ai + /5j + yk, not being
of
representation in a plane, may in a similar manner be
capable
represented in space. If we take three mutually perpendicular
The
axes, points at equal distances from their intersection will represent the three units i, j, k.
Multiples of these by real numbers
be represented by points on the axes of X, Y and Z, and any

complex number cd + jBj + yk may be represented by a point
will
y and z coordinates a, /3, 7. For every

find a point, and to every point there
As each of the real coefficients
a
number.
complex
corresponds
assume
a, /3, 7 may independently
any of a single infinity of values,
having the rectangular
x,
complex number we may
number has
three degrees of freedom, or there is a

such complex numbers. The distance of the
representative point from the origin was called by Hamilton the
We may apply the term modulus
tensor of the complex number.
the complex
triple infinity of
to the tensor,
and use the symbol
In this book the arrangement of the axes of X, Y, Z will

always be such that the motion of a right-handed screw along the
This will be
will turn the Y axis toward the Z axis.
axis of
called right-handed cyclic order, Fig.

*
les
Argand, Essai sur une maniere de repr&enter
constructions geometriques, Paris, 1806.

" Theoria residuorum
Gauss,
Bd.
1.
ii.,
p. 169.
les
quantites imaginaires dans
biquadraticorum, commentatio secunda."
Werke,
NUMBER.
5,6]
Geometric Addition. Instead of the point representing

number ai + fij + yk, we may fix our attention upon
6.
the complex
FIG.
1.
the line drawn from the origin to the representative point.

line is a geometrical
magnitude, which
This
is
completely specified only

Such quantities
direction as well as its length is given.
are called vectors, the name arising from the significance of the
when
its
operation of carrying a point from one end of the line to the other.
Quantities which do not involve the idea of direction, and are
completely specified by a single number, are distinguished by the

name scalars, the name arising from the possibility of their repre-
upon a linear scale*. To specify the direction of a vector

we must give two angular coordinates, which together with its
We may otherwise specify the vector
length make three data.
sentation
symmetrically by giving
perpendicular axes.
By
projections on three given mutually

projection on, resolved part or component
its
along a line, we mean the product of the length of the vector by

the cosine of the angle included between the direction of the
vector and the positive direction of the line. If the angle is acute,
the projection is positive, if obtuse, negative. In particular the
projections of the vector on the axes of i j, k, are the coefficients
t
of
i,
j,
number.
the representation of the vector by the complex

It follows from the definition of addition of complex
k,
in
numbers that to add two vectors means to find a vector whose

components are the sums of the corresponding components of the
two given vectors. This vector may be described geometrically as
*
If real.
Complex
scalars
may
also be used.
NUMBER.
10
[INT.
I.
the diagonal of the parallelogram formed from the two given

vectors as sides, or as given by applying the initial point of the
second vector to the terminal point of the first, and constructing a

new vector joining the initial point of the first to the terminal
Either of these geometrical processes shows
point of the second.
that the addition of vectors is commutative. The addition of
vectors in this
manner
is
known
as geometrical addition.
When
such geometrical addition of vectors is meant, as distinguished

from arithmetical addition of their tensors, we shall denote the
quantities to be considered as vectors
quantity otherwise used for the tensor
R = Xi +
Yj
by placing a bar over the

e.g.
the equations
+ Zk,
R= R =
are examples of vector and scalar equations respectively.
7.
by
As we have seen
Geometric Multiplication.
direct multiplication, the product of the
R = X i + Y^j + ZJc and R

1
is
in
4,
two vectors
- (X,X, + Y,Y + Z,Z ) + ( Y,Z - Z, F )

2
Of this the
scalar part
has an important geometrical meaning. The direction cosines of

the vector
being denoted by cos (Rx), cos (Ry\ cos (Rz), we have
the
of the projections of R,
definition
by
X = Rcos(Rx), F = jRcos(%), Z = Rcos(Rz).

Consequently,
cos
&
cos
+ cos (R$) cos (R$) + cos (R^) cos (R^z)}.

The
factor in the brackets is equal to the cosine of the angle

and R^. Consequently the scalar
l
between the directions of
part of the product of the vectors
and
of the two vectors, which will be denoted
or the scalar product
by the notation
SR^
is
equal to minus the product of the tensor of either multiplied by

the projection of the other on its own direction
SR,R 2 = - R,R
cos
NUMBER.
6, 7]
11
In order to avoid the inconvenience of the negative sign in

this, Hamilton's notation, we shall call the negative of the scalar
product the geometric product and denote
= X,X, +
RJl,
( i)
7,
4-
Z,Z2
it
= R,R
by R^R*,
cos
so that
(RA).
If two vectors are perpendicular, their scalar product
The
which
2)
vector part
will
R^, or the vector product,

VR^, has the components
- X.Z,, Z = X Y,- Y,X,.
3 of the product
be denoted by the symbol
F^ - ^F F = Z,X
2
<
It is to be noticed that the suffixes 1, 2,
as do the letters in the terms on the left

right.
If
is zero.
we multiply
ponents of either R^ or
RJt,
(3)
3,
appear in cyclic order,

first terms on the
and the
these equations by the corresponding com-
2,
we get
identically,
= X,X + Y,Y + Z,Z = 0,

3
= 0,
showing that the vector product is perpendicular to each of the
vectors involved.
Squaring and adding the equations (2), we get
F,
+ Zf) (X* + F + ^ ) -
(X,X, +
YF
1
- cos (RA)) = R*R* sin

2
(1
so that
(4)
The
= VEjE, = R R
i
vector product of two vectors
sin
is
accordingly perpendicular
their plane and its tensor is equal to the product of their

tensors and the sine of their included angle, or geometrically, to
the area of the parallelogram having them as sides.
to
The equations
(i)
and
(2)
show that
NUMBER.
12
The
[INT.
I.
is drawn with respect to

their plane in such a manner that
the rotation of a right-handed screw
advancing in the direction of the
vector product of two vectors
vector product would turn the first

vector towards the second, as is seen
by making the two vectors coincide
with two of the unit vectors

If
two vectors are
parallel, their vector
product
is
i,
zero.
j,
k.
The
vector and scalar products of two vectors cannot vanish simultaneously unless the tensor of one of the vectors vanishes.
NOTE.
Although the above consideration of vectors has been
inserted for the sake of logical connection, we shall seldom

use of the conception of a vector as a complex number, and
make
when
the term complex number is made use of we shall mean a complex

number in the narrow sense. We shall frequently use the terms
vector, scalar product and vector product, the latter being defined
by the equations
(i)
and
(2) above.
CHAPTER
II.
VARIABLES AND FUNCTIONS.
Functions. A real quantity is said to vary continuously

between two values a and b if it assumes successively all real
values, rational and irrational, comprised in the interval between
O
-El.
8.
and including the values a and b. The notion of continuity was

arrived at by considering the motion of a point which at successive
instants of time occupies the positions of all the points between
those representing a and b, and by the nature of motion cannot
omit any intermediate value.
quantity y is said to be a function of a variable x, in an

interval from a to b, if for every value that x may take in the
function
interval ab, there is assigned a definite value of y.
defined in this
one-valued.
somewhat
We may
restricted
manner
is
called uniform, or
extend the definition so that
for
each value
may have several values, in which case it is said to be a

multiform, or many-valued function of x. This definition, due to
Dirichlet, is independent of the question whether we can find an
of x, y
analytic expression for the value of
in terms of
example the analytic expressions

( T
-j-
ci
IT
--(2)
(3)
J*
(4)
e*
a raX*^
*"!-
n w^
x or
not.
For
14
a?
x
= tf-_
+ --
[INT.
II.
s intf
(6)
sm
are
all
functions of
--
111
in
......
......
any interval from a to
b,
where a and
6 are
both positive or negative finite numbers different from zero.

Of
these the first three are examples of the class called algebraic
functions, or such as are defined by an algebraic equation between
An algebraic equation is one in which only a finite

of powers with finite integral exponents and products of
such powers of the variables appear. All other functions are called
x and
y.
number
All
transcendental, and the last four above are examples of such.
the above are uniform functions, except (3), which has two values,
function such as (i)
one of which is the negative of the other.
A
called a polynomial, or a rational entire or integral function.
function such as (2), or the quotient of any two polynomials, is
called a rational fractional function.
(3) is an example of algebraic
is
irrational functions.
(4)
and
being defined by convergent
(6),
powers of x, are called integral transand the quotient of two such is called a
infinite series of positive
cendental functions,
fractional transcendental
The
between
rational and transcendental functions is similar to that between
rational and irrational real numbers, depending on the matter
function.
of finiteness or infinity in the
method
of specification.
rational
uniform, continuous, integral,

function is called holomorphic.
A function
to x
distinction
or
transcendental
taking the value 1 from the value x =
0, inclusive,
= 1/2, inclusive, to x 3/4,

inclusive, to x = 7/8, exclusive,
1/2, exclusive, the value 2 from
x = 3/4,
a
defined
as taking the value 1 for all rational
and
function
etc.,
all irrational points, would be, the first difficult,
2
for
and
points,
exclusive, the value 3 from
the second probably impossible to define by analytic expressions.

The former, being perfectly defined for every real value of x from
zero to
1,
excluding the
latter, satisfies
the definition of a function
in that interval, while the latter satisfies

9.
Limit of a Function.
If
it
in
y =/(#)
any
is
interval.
a function of the
continuous variable x in a certain interval including the value x=a,

and if there exists a number A having the property that to any
8,
9]
number e, however
sponding number B such that
small, there
positive
than
less
may be found
for all values of
a corre-
h of absolute value
8,
+ h)-A\<e,
\h\<6,
said to approach or converge to the limit

= a. This will be denoted
neighbourhood of the point x
then the function

in the
15
is
A
by
the equation
The necessary and

at
is
sufficient condition for the existence of
\f(a
+ h)-f(a + h')\<e,
\h\<S,
where
a limit
that
8 have the
e,
same
\h'\<S,
significations as before,
and h and
ti
are
any values whose absolute values are less than S. If the above
condition is satisfied only when h and h' are positive, the function
is said to approach the limit on the right of a, if when h and h' are
both negative, on the left. A function may approach different
limits on the two sides of a point. It is not necessary that a function
should be varying always in the same sense in order to approach a
limit, e.g.
the function
= x sin x
which alternately increases and decreases, approaches the value

The function
zero as a limit in the neighbourhood of x = 0.
_i_
y=e
approaches the limit zero on the right of x

the left.
The
0,
but not on
function
1
y = sin.
does not approach any limit whatever in the neighbourhood of
= 0,
for in
any
interval,
however small, from
X
where n
is
any
values from 1 to
integer,
1.
however great, the function takes
all
16
If a
[INT.
II.
does not approach any limiting value for a

it must be otherwise denned for such
function
certain value of the variable,
a point
e.g.
except x
we may
by the
for the point
assign to the function defined at
analytical expression sin
# = 0.
The
CO
all
points
any arbitrary value
function will then be completely de-
fined.
quantity that approaches the limit zero
an
called
is
in-
finitesimal.
If
is
a function of x defined in an interval a to
is
b,
we please, a number possessing the property

a given number as large as we please,
x greater than M,
b is
when
said to be the limit of
y as x
x>M,
\f(x)-A\<e,
for all values of
where
that,
as large as
is
increases indefinitely or, briefly, as

denoted as follows
x approaches
This
infinity.
is
\ime x =l.
e.g.
we change if to a negative number

whose absolute value is as great as we please, and consider all
values of x less than M, we say that A is the limit as x approaches
minus infinity.
If in the above definition,
If in the neighbourhood of a point x a, when If is any number

we please, we can find a corresponding number S such
as great as
that for all values of

|
h,
whose absolute value
/(a + h)
> M,
h
|
is less
<
than
B,
8,
then y is said to become infinite for x = a. If, as above, we change

the definition so that y is less than any negative number, y is said
to
become negatively
infinite, or
lim f(x)
The
oo
function
y
y
fails to
approach any
of the point x
0,
sin
limit, finite or infinite, in the
by reason of
its
neighbourhood
continued oscillation between
911]
17
greater and greater positive and negative values in any interval,
however small, including
zero.
Continuity of Functions.
continuous at a point x = a, if for any
10.
function
said to be
is
positive e there
is
a 8 such
that
+ h)-f(a)\<e,
\f(a
\h\<8
for all values of h whose absolute value
is less
than
8.
If the condition holds only for positive values of h, the function

said to be continuous on the right, if for negative, on the left of
function may be discontinuous at a point by reason of jumpa.
is
ing abruptly from one finite value to another, becoming

oscillating
through a
interval.
The
tinuous,
finite or infinite
function
last
and the next to the
infinite, or
range in an infinitesimal
defined in
is
nowhere con-
discontinuous at the points
last is
i^
x
1/2, 3/4, 7/8, etc., for the first reason, the function e is discon-
tinuous at x
for the
second reason, and the functions
111
sm
sm -
the third reason, e x is

discontinuous at the right of the point #=0.
are discontinuous at the
continuous at the
left,
same point
for
discontinuity arising from a finite jump, or an infinite increase or decrease, is called an ordinary discontinuity, while one
arising from an oscillation is called a discontinuity of the second
kind,
and the value of x at which
it
occurs
is
called
an essentially
singular point for the function.

11.
Derivative.
In the neighbourhood of any value of the
variable x, the difference -quotient
(x
+ h)
a function of the increment h of the variable.
If this quotient
a
the
value
of the limit is
limit as h approaches 0,
approaches
called the derivative of the function f(x) at the point x, and is
is
denoted by/' (x) or by

<fr
dx
w. E.
=lim /(*
h-Q
+ *)-/(*).
h
18
[iNT.
II.
If a limit exists on one side but not on the other, the function is
said to have a derivative on one side.
If no limit exists at the
point x, the function has no derivative at that point;

function sin - has no derivative at the point
derivative at any other point however near
--1- cos -1
of
x=
e.g.
the
although the
0,
is
which does not approach a limit as x approaches 0.

8 has no derivative anywhere.
function defined in
The
last
We may
also find transcendental functions defined
by analytic expressions,
The function proposed by
which nowhere possess a derivative.

Weierstrass,
n = 00
f(x)
=2
bn cos (an 7rx),
<
where
<
is
an odd integer,
tt=0
may be shown
to
have nowhere a derivative*.
Functions of two or more Variables.
12.
variables
If two real
x and y vary continuously in the respective intervals
if to every possible pair of values of x and y is assigned a

value of a quantity u, u is said to be a function of x and y. For
any particular value of a?, u is a function of y, and for any particular
and
Suppose that for a certain value

y, u is a function of x.
a
of
x
as
function
u
considered
approaches a limit as x approaches
y,
This limit will in general depend upon the value of y, let us
a.
value of
call it
lim u

x=a
(y).
again approach a limit as y approaches a value b. If we

consider the limit approached by u considered as a function of y,
It
we
may
have in general a function of
shall
#,
lim u = ^ (x).
y=b
If
x then approaches
a,
we may have
a limit, which
is
not neces-
sarily the same as before,
lim
y=b
j
(
lim u
x=a
j
j
= lim&(y)=A,
y=b
lim
x=a
lim u
j
(
y=b
I
>
= lim
x=a
(x)
*
Weierstrass, Abhandlungen aus der Functionenlehre, p. 97;
Morley, Theory of Functions, p. 58.
=B
Harkness and
11, 12]
e.g.
the function u
19
z/
has
y
lim u
lim
j
y=Q
since the limit for
as
X=Q
= lim (0) = 0,
2/=0
does not contain y, while
lim u
lim
\
i
a;=0
j
(
7/
oo
function of two or more variables
= a,
find Si
y = 6,
is
any positive value of

and S2 so that
if for
h,
for all values of
A function
+ k)-f(x,
of
y)\ <e,
h and k which
e,
is
\<$
satisfy the
two variables
continuous at a point
however small, we can
19
\k\<8,
above inequalities.
not necessarily continuous
if it
a continuous function of either variable; e.g. the function

2
xy/(x* + 7/ ) is a continuous function of x for any value of y, even
is
y=
and of y for any value of x, even x 0. It is not a continuous

a? and
y at x = 0, y = 0, since M = for # = 0, irrespective
of the value of y, and u =
for y = 0, irrespective of the value of
if
we
select
of
values
of x and y, such that y = mx, we
but
#,
pairs
0,
function of
have u
=
1
#=0,
2/
+ra 2
which
is
discontinuous with the value u
at
= 0.
Derivative.
If
w considered as a function of
x, for
any par-
ticular value of y, has a derivative as before defined, this derivative

is called the partial derivative of u with respect to x, and is
'
denoted by fx or by
5
CM?
If
^-
z,
&=o
/*
considered as a function of
2/,
say
<
(T/),
has a derivative,
(jCO
this is called the partial derivative with regard to
y of
and
V = lim
ju
denoted
by 5-^8
=o
22
is
20
[iNT.
II.
It is evident that
32
82
dy'dx
if/ and
its first
The
number
definitions here given

of variables.
derivatives are continuous functions of
may
x and
y.
be extended to functions of any
Point-Function. If a quantity has for every position

13.
of a point in a region of space r one or more definite values
assigned, it is said to be a function of the point, or point- function.
This term was introduced by Lame. If at every point it has a
Functions of the two or
single value, it is a uniform function.
three rectangular coordinates of the point are point-functions.

is continuous at a
if we can find correpoint
point-function
sponding to any positive
when
is
any
e,
however small, a value
point inside a sphere of radius
<
S such that
8,
\f(B)-f(A)\ <e.
We may
have vector as well as scalar point-functions, the
length and direction of the vector being given for every point. A
vector point-function is continuous if its components along the coordinate axes are continuous point-functions.
Level Surface of Scalar Point-Function.
14.
/
/M'
M/
If
is
uniform function of the point M, continuous

and without maximum or minimum in a
__
in
portion of space T, through any point
the region T we may construct a surface
having the property that

it V has the same value.
FIG. 3.
for
every point on
be c. Then since c is neither a

at
the value of
nor minimum, we can find in the neighbourhood of
two points A and B, such that at A, V is less, and at B, greater
than c, and that in moving along a line
through m, V con-
For
let
maximum
AB
If the line
tinually increases.
A'M'B, so that
V(A)-V(A')
and
AMB
is
displaced to the position
<c-V(A)
V(B)-V(B') \<V(B)-c,
then V(A')< c< V(B'), therefore there

A'B' for which V=c.
is
a point
M on the
line
1216]
As
its
AB moves
continuously
M describes a
and
this line in
every point of which V=c.

called a level surface of the function V.
motion describes a surface,
Such a surface
line,
21
for
is
level surface divides space into two parts, for one of which
than in the surface.
greater, and for the other less,
As examples
of point-functions
drawn from the point
of a line
The
(2)
level surfaces are spheres
the radius vector
The angle that

The level suraxis.
(4) The dihedral
plane through OX. The
with centers at 0.
(3)
OM makes with a fixed line
faces are right circular cones with
OX
as
angle made by the plane MOX with a fixed

level surfaces are planes through OX.
Coordinates.
15.
(1) the length

line until it
M parallel to a given
Its level surfaces are planes parallel to the

distance of
from a fixed point 0. The
cuts a given plane.
given plane.
we may take
V is
If a point
is
OX.
restricted to lie
on a given
surface S, the intersection of that surface with the level surfaces of

a function
are the level lines of the function on the surface S
examples (3) and (4) above, if S is a sphere with

center, the level lines are parallels and meridians respectively.
e.g. in
A function /( V F
is itself a point2
.) of several point-functions
its
level
surfaces are
a function of one
only,
,
If it is
function.
as
the same as those of V, for
when
V
V is
constant,
f(V)
is
also
constant.
Let q lt q 2) q 3 be three uniform point-functions.

through the point M.
surfaces do not coincide or intersect in a
level surface passing
Each has a
If these three level
common
curve, they
determine the point M, and we may regard the point-functions
The level surfaces of
as the coordinates of the point M.
<?i> <?2> #3
and the intersections of pairs

The tangents to the
lines.
are
coordinate
tne
(^1^2). (MS)' (Mi)*
If at
are called the coordinate axes at M.
coordinate lines at
#i> #2> ?s
are the coordinate surfaces,
M
coordinate
axes are mutually perpendicular, the
M
the
every point
system
is
said to be
an orthogonal system.
Differential Parameter.
The
consideration of pointfunctions leads to the introduction of a particular sort of derivative.

16.
a uniform point-function, continuous at a point M, and

possessing there the value V, and at a point M' the value V, in
If
is
22
of
virtue
continuity,
V- V= AV
is also.
when the
The ratio
distance
[INT.
MM'
is
II.
infinitesimal,
F'-F_AF
"
MM'
and as
is finite,
MM' = As
approaches
being given, the limit
-r
line
through
9s
Vin
the direction
M in the direction
we
as
that
Let
mal
M
-
MQ
and
MQ = OS
---
all
the
find
and
possible
surface
MN be the direction of the nor-
to the level surface at
represent the
Let
'
M, and
derivative
and
in
N be the
level surface, for
which
let
that
inter-
F= F',
Then
AF_~ &VMN
MN MM'
MM'
As
lay off
the locus of Q.
same neighbouring
MP.
of s a length
we may
direction.
sections of the
with
MP
"~~v
FlG 4
is
We may
s.
give s successively
directions,
MX
MM'
the direction of
-5-
As
defined as the derivative of
on a
0,
AF = 9F
r
lim
As=0
is
As
MM' approaches zero,

lim
s =o
we have
AF = 8F
-
-,
8s
As
=^
^9s
9n
Hence
'
lim
MN = cos
PMQ.
MM
cos
PMQ,
>--
the derivative in any direction at any point is equal to the

projection on that direction of the derivative in the direction of
that
is,
the normal to the level surface at that point.

points Q lie on a sphere whose diameter is MP.
The
surface
Accordingly
all
derivative in the direction of the normal to the level
was
called
by Lame* the/rs
differential
parameter of the
function F, and since it has not only magnitude but direction, we

shall call it the vector differential parameter, or where no ambiguity
*
G. Lam6.
Paris, 1859, p. 6.
Legons sur
les
coordonnees curvilignes
et leurs diverses applications.
16]
23
will result, simply the parameter, denoted by P or P r

The
above ^heorem may then be stated by saying that the derivative
in any direction is the projection of the vector parameter in that
.
The theorem shows that the parameter
direction.
gives
the
direction of the fastest increase of the function V.
If
is
a function of a point-function
surfaces are those of
___
q,
~dn~dq fa- f
and
Vf(q),
q,
its level
and
(q)
dn'
if
where the sign

same,
if in
-f
to be
is
taken
if
and q increase
in the
opposite directions.
Suppose now that
V=f(q
l}
qz q3 ...... )
,
8F = dVdft
dVdq,
dVdq,
ds
8g2 3s
dqs ds
dq
ds
and if h^, h2 ...... denote the parameters of q

theorem gives
,
q2
......
the above
P cos (Ps) = |Â cos (hjs) + ^-h, cos

oq 1
Now
qt,
;r
hi is
and we may call
have the same sign
cq2
the parameter of V, considered as a function of

it
h partial parameter
the
>
if ~
0,
opposite signs
and since Pi and
t-,
if
<
0,
we have
hi
in
either case
hi cos
(^s)
PI cos
(Pis),
vQi
and
P cos (Ps) = P
cos (Pjs)
+ P2 cos (P
2 s)
This formula holds for any direction s and shows that the
parameter P is the geometrical sum, or resultant, of the partial
parameters,
Hence we have the
rule
for finding the
function of several point-functions.
If
parameter of any
we know the parameters
24
^1,^2
x
......
of the functions q 1} q z
......
we
and the
II.
partial derivatives
lay off the partial parameters
h2 ...... or their opposites, according as
in the directions hl}
or the opposite,
......
[INT.
and
find the resultant of
If the functions q lf q2)
P P
lt
dV
>0,
.......
are three in number, and form an
......
orthogonal system, the equation
?=?,+?,+?
gives for the
modulus, or numerical value of the parameter
Examples. (1) in
from the plane be
Let the distance of
14.
AF= Aw =
u.
M in the given direction
where a
is
cos a'
the angle between
the given direction and that of a perpendicular to the given plane.

A*.
Aw
cos a
'
p= cos a
'
is perpendicular to the given plane P=l.

Accordingly for q 1 = x, q 2 = y, <?3 = z the rectangular coordinates of a
= Pz =\, and for any f unction f (x, y, z)
point, we have Px = Py
If the given direction
p=
The
projections of
[v_] +
L\3av
P on the
^_
+ [v.
\8,
w/
coordinate axes are the partial parameters
Consequently, if cos (sx\ cos (sy\ cos (sz) are the direction cosines
s, the derivative in that direction
of a direction
=P
dV
ox
cos (sx)
cos (ax)
P may be written in terms
+ P3 cos
cos (ay)
dV
cos (sy)
oy
97
cos
oz
of the unit vectors
J
v^y+jt+A.
dx
dy
dz
(sx)
i,
j,
(sz).
k as
16]
f)
7\
25
rt
which gives the vector differential

+
The operator *5~ +
./jr
^g~
parameter of a function, was denoted by Hamilton by V. (read Nabla).
If f(x, y, z) is a
homogeneous function of degree
n,
by Euler's
Theorem
&
+ y~+
nf= x~
J
Bx
,
z^-.
ty
dz'
nf= P {x cos (Px) + y cos (Py) + z cos (Pz)}.
or
Now
the
is
parenthesis
the distance from the origin of the
tangent plane to the level surface at
Calling this
x, y, z.
8,
or the parameter of a homogeneous function is inversely proportional

to the perpendicular from the origin to the tangent plane to the level
For example,
surface.
if
n= 1,
V = ax + by + cz,
P cos (Pa) = a,
The
and the parameter
level surfaces are parallel planes,
is
con-
stant
V is
proportional to the distance of the level surface from the origin.
If
7i
2,
Pcos(Px)=
Pcos(Py) =
a^
af
2F
I*
A/
V
For the
surface,
Pcos(Pz) =
^,
&i
"^
f_
2
!_
"^
Ct 2
V= 1,
S
*-y==^,
a familiar result of analytic geometry.
2
CT
-,
26
Polar Coordinates.
we
If
Examples
ii.
the point-functions of
call
and
3,
2,
[INT.
of
4,
14,
d,
r,
<f>,
we
obtain the system of spherical, or polar

and <j> may be called the
coordinates.
and longitude. The level surfaces of r being spheres, the normal coincides with r.
Accordingly
co-latitude
dr
dr
_
~
_
'
The
FIG. 5.
level surface of 6 is a circular cone
of angular opening
dn
The
W
=-
,,
= rd0,
level
0,
d6
j^
rdd
dn
surfaces
of
(Fig. 5),
and
=r
h6 = -.
are meridian planes
<
through the axis of the above cones, (Fig.
dn
FIG.
= r sin
dn
6.
r sin
6d<$>
r sin 6
6),
and
'
'
r sin
For any function /(r,

Jr f
0,
</>),
the partial parameters are
(}T
~T
~~ ?T~
dr
fly
==
(JT
or
p.-j{V-Jj
^
The
ii
^<f>
'
r sin
a</>
total parameter, the resultant of these, is given
p2
=
vaf )
by
+
^
Cylindrical, or Semi-polar Coordinates. If we take

the rectangular coordinate z, the perpendicular distance from the
18.
^-axis, p,
and
ing the point
o>
the longitude, or angle made by the plane includ^-axis and a fixed plane through that
M and the
we have the system of semi-polar, cylindrical,

coordinates, for which we have immediately,
axis,
1
IP
^.
or columnar
17
19]
The parameter
of a function f(z,
p, CD) is
27
the resultant of the
partial parameters
The equation
Ellipsoidal Coordinates.
19.
of a central
quadric surface referred to its principal axes is
where a l5 a2 a3 may be positive or negative.
If they are
negative, the surface
1.
Suppose one
is
imaginary.
is
negative, say
while
=a
c&!
Let
The equation
is
3?
-5
ft
The
section
2
.
> c.
= 1i
C
The
surface
is
cut by the
XF-
2
"Z/
Cb
foci are at
>
=6
semi-axes are a and
+|^=1, whose
plane in the ellipse
whose
a2
Z*
7/
+ ?r
#2
all
6,
and
a distance from the center
by the ZZ-plane
with semi-axes a and
c,
and
is
the hyperbola
foci at distance
Va2 + c2 = Va - a3 on
x
the X-axis.
The
section
by the F^-plane
is
the hyperbola
y^- z*-^
62
with semi-axes 6 and
and
foci at
'
a distance V& 2
the F-axis.
The
surface
is
an hyperboloid of one sheet.
+ c = Va
a
a 3 on
28
[iNT.
II.
Let two of the constants a it a 2 a 3 be negative, say
2.
The equation
is
a1 "*^
The
sections
'
d"""
by the coordinate planes and their
focal distances
are
/v,2
XT
ni2
-^ =
1.
= - 1.
+c
F^
The
surface
3.
If
is
Va +
62
= Va - a
Hyperbola
Va2 + c2
= Va - a
Imaginary Ellipse
V- (6 - c ) = Va -a
an hyperboloid of two
a 2> o are
j,
Hyperbola
all positive,
3.
sheets.
the sections are
ellipses,
and the
is an ellipsoid.
In all three cases, the squares of the focal
distances of the principal sections are differences of the three
constants a 1? a 2 a 3
Accordingly if we add to the three the same
surface
number, we get a surface whose principal sections have the same

foci as before, or a surface
confocal with the original.
Accordingly
7/
Zz
_
-
represents a quadric confocal with the ellipsoid

a?
for
any
If
b2
is
real value of p.
a>b>c
>p >
>p >
a?
p >
the surface
and
the surface is an ellipsoid. If

an hyperboloid of one sheet, and if
an hyperboloid of two sheets. If a2 > p, the surface
c
is
imaginary.
Suppose we attempt to pass through a given point

quadric confocal with the ellipsoid
a?
v2
z*
-^+r^-f-^=l,
or
cr
<r
a>b>c.
x, y,
z a
19]
29
Its equation is
X2
where p
is
to be determined.
2
(2)
yz
(a
Clearing of fractions, the above
is
+ p) (6 + p) (c + p) - x (V + p) (c + p)
- f (c + p) (a + p) - * (a + p) (6 + p) =f(p) = 0,
2
p.
a cubic in
22
Putting successively p equal to
2
,
62
a2
and observing signs off(p),
The changes of sign of f(p) show that there are three real roots.
\ lies in the interval
Call these \, p, v in order of magnitude.
X > c2 necessary in order that the surface may be an ellipsoid, ^
> /z >
may be an hyperboloid of one

2
a 2 that it may be an hyperand
in
interval
6
>
v
>
v
the
sheet,
boloid of two sheets.
There pass therefore through every point in
in the interval
6 2 that it
space one surface of each of the three kinds.
If
we
call
(3)
^=0
the equation
defines X as a function of x, y, z, and therefore as a point-function.
The normal to the surface X = const, has
direction cosines proportional
ax
to'
ax
ax
dy'
dz
'
'
da?
Now
since identically
F = 0,
and we have
3X
&((&)#-*-.
ax
30
[INT.
II.
Therefore
*2
2
+X
2
(a
(a
+ X)
8X
+ X)
F' (\)
__
+
2
2
-*
X)
(6
(c
+ X)
]
2
j
'
~
8^
The parameter
(c
+ X) .P
'
(X)
of the point-function
'(\)Y \(a?
+ X)
is
accordingly given by
\n2
+ X)
(6
4
that
is
Now
X=
the direction cosines
of
the
normal to the surface
const, are
1
8X
(7)
cos (n K z)
Similarly for the normals to the surface
= const.,
-()-
cos
The angle between the normals
to
X and n
is
given by
19]
cos
(8)
XB
*.
(c
Now by
31
I
*_
I
+ x)(c + ,*)f VFTx)
2
subtracting from the equation
the equation
we
get
_i
_J_|
or
(x
~ M)
x)
X=
cos (n^n^)
and the two normals are
ytt,
Similarly for the other pairs of surfaces. Accordingly the three surfaces of the confocal system passing through
any point cut each other at right angles.
Accordingly, unless
at right angles.
If
we
ellipsoid
a?,
y,
give the values of X,
/*, v,
we determine completely
the
and two hyperboloids, and hence the point of intersection
z (and its seven symmetrical points in the other quadrants).
Hence we may take X,

v for the coordinates of the point,
v
and the family of surfaces forms an orthogonal system. X,
//,,
yu,,
coordinates of the point.

shall proceed to find their parameters in a form not conmust find the rate of
taining any coordinates but X, //., v.
are called
the ellipsoidal or
elliptic
We
We
change of
Since
as
we go along
we have
the normal to the ellipsoid
identically
__
*
1
X=
const.
32
[INT.
II.
differentiating totally
zdz
Now
if
S x is the perpendicular distance of the tangent plane from

we have by the last formula of 16,
the origin,
Â
so that
Now
as
we may
-*-
a2 +X) 2
/7
rt
-v
\O
+ X)
.
(6
(c
+X)
'
write for the cosines,
we move along the normal, we have

^
dx = dn cos (n^x)
+ A, dn,
^,
dy = dn
cos (w x y)
^
dz
= C^TI cos (n K z) = - '-^- dn.

C + A/
Inserting these values in (10),
+ X)
a?
a?
(6
+X)
(c
c^X
((a*
+ A,;
ô^
+ A</
^c-
-i-
= 0,
A^rj
so that
3)
dX
2S
^ = ^- = os
i,
(J
2
A
~
,
,
i
/7
-v
\n
In order to express this result in terms of the

we may express x, y, z, in terms of X, /a,
ordinates alone
that the function
elliptic cov.
Observe
19]
has as roots X,
ft,
and being reduced
/',
to the
has a numerator of the third degree in
= X,
p
it
//,
common denominator
As
p.
33
this vanishes for
=v
can only be
Hence we have the
Multiplying this by p
identity
+a
and then putting p
a 2 we get
^_(
and in
like
manner
(62
(c
c2 ) (6
-a
'
2
)
+ X)(c + ^)(c +*/)

(c -a )(c -6 )
2
If X, p, v are contained in the intervals above specified, these will

be positive, so that the point will be real.
all
If
we
we
insert these values in 8 A ,
terms of
X,
This
is
JJL,
shall
have A A expressed in
v.
more
easily accomplished as follows.
Differentiating the above identity (14) according to p,

~
^+
-V>) (p
1
\p-\
If
we put p = X,
W.
E.
""
y*
1)
_1
/3
first,
and we have
z>
+ c ]'
2
p-p p+b* p-v
the terms on the right except the
X, vanish,
multiplied by p
a?
all
p+a
(X-yu,)(X-i/)
being
The
expression on the left
is -K
[INT.
ii.
Hence
o\"
(18)
-28
-2Â //(^
ZOx =
(X-/*)(X-
In a similar manner we find
,_
/(a
"V
+ K)(fe'+i/)(c + y)
(.-XX^-M)
2
and the parameter of any function F(X,
20.
Infinitesimal Arc,
/*, i/)
is
Area and Volume.
If
we have
any three point- functions q lt q^ q3 forming an orthogonal system

of coordinates, since their parameters are
8
?
k
* - 50?
1
7,
**
8 ^2
h - dq
^
an,
*
to,'
n,
FIG.
7.
the normal distance between two consecutive level surfaces q and

ql
_f-
dq 1
is
dn^
= -r^
q lt q l
consequently
q,
if
we take
six surfaces
q.2)
the edges of the infinitesimal curvilinear rectangular parallelepiped

whose edges are the intersections of the surfaces are
dql
dq 2
IT' "^'
dqs
~fh'
and since the edges are mutually perpendicular, the diagonal,
or
19, 20]
element of arc
35
is
*-***
the elements of area of the surfaces q l} q 2) q 3 are
respectively
,~ _dq2 dq3
dq3 dql
2
dq l
^"fcG?
~~I7
and the element of volume
^"X'X
is
Rectangular coordinates
Examples.
dq
x, y,
= h y = hz =l,
dSy = dz dx, dSz = dx dy,
hx
dSx = dy dz,
Polar coordinates
r, 6,
dr = dx dy dz.
= r^ sin 0d@d<f)
dS = r sin 0drd<j>
dS$ = rdrdO
element of area of sphere,

element of area of cone,
c?>Sfr
Cylindrical coordinates,
element of area of plane,
z, p, a>,
p-
-p,
dSz = pdpdco
dSp = pdndz
element of area of plane,

element of area of cylinder,
dSu
element of area of meridian plane,
dr
dpdz
= pdpdcodz.
Elliptic coordinates, X,
4 V(a2 + p) (6 2
4 V(a2
2
1/)
(6
+ /*)
+
(c
z/)
v.
/A,
(c
+ ^) (a +
__
2
z/)
(6
+ v) (c + v)
2
+ 1;) (a + \) (6 + X) (c + X)
hyperboloid
+ X) (a2 + /A) (6 2 + A*)
hyperboloid,
_
2
(a
7
_
~
+ X) (6 +
s
X) (c
d\d/j,dv(\
fi)(jj,
(c
v)(v
+ ^)
X)
32
CHAPTER
III.
DEFINITE INTEGRALS.
21.
Definite Integral of a Function of one Variable.
we
consider a continuous function of one real variable, the

notion of its definite integral may be illustrated by means of a
If
geometrical representation. If the function y =f(x) be represented

as the ordinate of a curve of which x is the abscissa, and if between
two points x = a, x = b, we place any number n 1 of points

#!, #2 ...... act, ...... #n-i> an d in the intervals between them erect
,
ordinates to the curve at points | 1? | 2
the
......
so that
sum
represents the area of the rectangles constructed on the bases
with the altitudes /(ft). The value of this sum depends on the
form of the curve or of the function f(cc\ on the choice of the
the
points of division, ^ ...... xn and of the points %k within
,
can be shown, however, that if all the differences &

are less than a certain value 8, all the values that S can take are
confined between certain limits, and if the number of intervals
intervals.
It
increases so that 8 decreases without limit while
remains always equal to b
a,
......
+ 8n
that these extreme values of
This limit will represent the area of

approach a common
the space bounded by the axis of X, the ordinates erected at the
x = a and x = 6, and the curve representing the function
limit.
points
DEFINITE INTEGRALS.
21, 22]
37
This conception may be extended to any function whether

continuous or not, and the limit, if there be any, approached by the
sum
n = oo
n = oo
as the
number
of intervals
A:
made
is
to increase without limit,
is
the definition of the definite integral of the function f(x) from a

to
It is
b.
denoted by
i,
f(x)
is
called the integrand,
/O) dx,
a and
b the limits,
and ab the
field of
Evidently the letter x in the symbol may be replaced

by any other without affecting the integral. If the sum has a
limit the function / (x) is said to be integrable in the region from
integration.
a to
b.
Condition of
22.
tion in a given interval
the least value that
it
I ntegr ability.
from the definition of continuity that

small as
we
The
oscillation of a func-
the difference between the greatest and

assumes in that interval. It is evident
is
please we may always

less than 8 and lying
if e is
find a
a positive number as
8 such that in
number
in the region ab in
function is continuous, the oscillation is less than e.
any interval
Let fx
division #j
......
which the
fn be a system of ordinates for a system of suband let / ...... fn be a different set of ordinates
'
...... scn
contained in the same intervals 81} 82
...... &n-
Then
Then we may
8,/(
.')
find 8 so that
Ss
when
(/(.) -/(?/))
all
S/s
become
less
than
8,
every
and consequently
2 Se (/(&) -/<&')) < 2 8 e = (6 - a) e.

s
As
?i
increases indefinitely, e decreases indefinitely, and
so that the selection of the ordinates in the intervals does not

affect the limit, if
one
exists.
38
DEFINITE INTEGRALS.
If
is
[INT.
the greatest value of f(x) in the interval Ss
III.
the
18
the
least, so that the oscillation in that interval is .DS =
S -1 8
n
n
two sums
S Ss s and S2 =2< 8S ls must approach limits as n
&=
increases indefinitely, for 8l is always greater than $2 and as we

can never increase and $2 can never decrease. Now
increase n,
,
&
the
sum 2 Sg /(fs ) always
lies
between
S and 82t
1
therefore if their
difference
Ss
sum 2
approaches the limit zero the
must approach a
Consider
limit.
now two
different
modes of subdivision of the
interval a&,
and
a?/,
a?/ ...... a?V_i,
and the corresponding sums
8 = 2S8fs and S'=iU//s

Let the points
form the system r
a^ ......
......
rp
n-\
and
Pl
In the interval a?^,

general
if ,%_!
= r^,
scs
a;g
will
and
'.
taken together
a?/ ...... #'n'-i
let
= Tl - a
there
be r^+t
may or may not

= 1), so that
fall
an
r.
In
(t
Then
+ ph+tf"h+t
+
PA+2 (/ -f"h+z) ...... + f>A+ (/
/"A+I)
V* == Ph+if"h+i + ph+zf'h+z
+
p A+1 (/g
where the
/A;//>S
......
are arbitrarily chosen values
p k and
,
where
P =1" [PA+1 (/. -/ V.) +
/>
of/
f"h+t),
in the intervals
DEFINITE INTEGRALS.
22, 23]
But
sum P,
in this
f f"n +k
is
39
for every s the greatest possible difference

absolute value not greater than the oscillation
8
in.
Therefore
In like manner
p
n'
V
Z
'
<>s
V
^
'
-f
fs
f"
phjh
where
P'
\
^ I S/ Ds
'.
Then
and
is
if
n'
2S./.-2
i
i
2
lim
Ss
the same for
D =
g
all
systems of division, the limit
for all
systems of division.
It
is
2Bf
s
easy to show that
if
the condition
lirn
S S s Ds =
satisfied for
is
one mode of division,
This is then the necessary and sufficient

is satisfied for all.
condition that the function f(x) shall be integrable in the interval
it
db.
Properties of Definite Integrals.

from
the definition
diately
It results
23.
imme-
= lim 2
f(x) dx
Ss
S)
if we
interchange the limits a, 6, since every S s changes sign,
the sign of the integral is changed. More generally
that
rb
ra
re
/O) dx + Jbf(x) dx + J /O) dx = 0.

J a
(i)
The arithmetical mean of a number

sum divided by their number.
their
in
tegrable
an interval
and
ab,
of quantities is defined as
If f(x) is finite and inSn
S,
Bn
8/
1111
divisions of the interval, from the last equation of
.'
/.'
22,
- H/, a i 8. D. + 2 ?.' A'.

|
Consider n constant and
let n' increase
without
Then
!
f
J
r/
11
*/() dx-2
Ss
f ÎS
s
JD.,
so that
I
1
limit.
are two
40
is
DEFINITE INTEGRALS.
[iNT.
an approximate value of the integral whose error
We may
put the S s 's
all
equal, so that 8S
=-
is less
III.
than
Then
If/
If
^
b-aj /(a
6
(2)
That
the definite integral of a function in a given interval

divided by the magnitude of the interval represents the arithmetical mean of all the values of the function taken at equidistant
is,
values of the variable throughout the interval,

of values taken is increased indefinitely.
From
the definition
when the number
evident that if/ (x) has the same sign
it is
rb
throughout the interval
ab,
and
there
if
is
f(x)dx has
the sign of (6
in ab a finite interval cd in
which f(x)
is
a)/(V),
not zero,
rd
then
J c
f(x) dx
is
not zero.
In particular, if the function is continuous in a whole interval

and
the integral between every two values of x in the interval
ab,
is zero, the function must be zero everywhere within the interval.
If therefore two continuous functions give in every interval ab the
same value of the integral, they must be equal everywhere in the
interval.
Suppose that the continuous function f(x) has in the interval

and a least value m, the integral will have a
ab a greatest value
value lying between
M(b
a)
and m(b
a)
and we may
write,
M > A > m.
where
Since f(x) is continuous, it

one value f of x between a and
(3)
will take the value

b,
so that
we may
(/(*)**=/(*)(& -a),
for at least
write
<<&.
J a
The above formula may be

(4)
generalized.
We
have always
DEFINITE INTEGRALS.
23, 24]
41
and <j>(x) are finite and

and m
always between
If in the interval ab, (a<b),f(x)
grable functions, and \f(x)
lies
rb
If
the interval
in
</>
m are
Mf(x) and/(#)
(x)
J a
Jo,
inte-
rb
f(x) <f>(x)dx
(5)
(\<t>(&)\) dx.
always has the same sign, since
positive,
rb
rb
(M
Ja
fx)
(x)
<j)
dx and
Ja
(f)
(as)
dx
f(x)
(x) dx,
dx
rb
rb
f(x) $ (x) dx
(x)
<j>
(f>
J a
J a
and
m)
(f(x)
(x) dx,
(f>
Ja
rb
have the same
sign,
rb
J a
(f>(x)dx
f(x)
(x)
<f>
<f>
(so)
lies
between
is
equal to
rb
<f>(x)dx so that
J a
f(x)<p(x)dx
J a
rb
I
dx
J a
rb
and
and therefore
dx multiplied by a
lying between
M and m.
some point f
in ab for
factor
If f(x)
continuous, there
is
= A, and
<#>
(^
J a
^ =/(f
This important theorem

theorem of the mean.
is
The
b.
zero for
a;
= a, and
known
du Bois-Reymond's
as
Let f(x) be integrable between
for
every value of x between a and b it has a

a function of its upper limit x.
It is therefore
it
by F(x).
/*#+/*
F(x + h)=
and
b.
integral
definite value.
Let us denote
< f<
J a
'
a and
[V () <fo,
is
Indefinite Integrals.
24.
which
accordingly
/"/(*)
(6)
is
â
.P (x
If x
+^
be another value of x in
/#
f(x) dx =\ f(x) dx
Ja
+ A) - ^(a?) =
fx+h
fje+h
/(a?)
J
a;
da;,
/(a?)
c?a7,
at,
42
DEFINITE INTEGRALS.
Hence
[INT.
F (x) is a continuous function of x.

b,
is
any number
rx
Cx
between a and
If a
III.
f(x)dx
differs
from
J a
J a
f(x)dx only by a conra
stant,
namely the value of the integral
J a
function
F (x) -f G is
f(x) dx =-C.
The
called the indefinite integral of /(#).
Suppose that h approaches zero either positively or negatively,

and let/(#) either be continuous at x, or have an ordinary discontinuity, i.e., by making a finite jump.
Then for any positive number e however small we can find a

number h^ of the same sign as h, such that for every x in the
interval x x + h^ (at most excepting x\ the value of f (x) for any
}
point differs by less than
is
from f(x
+ 0)
or
f(x
- 0),
according as
positive or negative.
Therefore the value /(f) in the expression
differs
from f(x
0)
by
less
than
and we have
lim
rx
That
is,
the integral
J a
=
f(x) dx
F (x)
is
not only a finite and
continuous function of x in the interval ab, but it has at all points

where f(x) is continuous a finite and determined derivative f(x)
and where/ (x) has an ordinary discontinuity, though not having
a determined derivative, F (x) has one on the right and left
If however f(x) has
0).
respectively equal to f(x + 0) and f(x
a discontinuity of the second kind, at
x,
the value of
F(x+h)-F(x)
h
as h decreases does not approach a limit
tive at x.
and
F (x)
has no deriva-
The
principle here proved enables us to calculate the definite

integral whenever we can find a function F(x) whose derivative is
f(x), for then
L
The
definition usually given of the definite integral, as
from the indefinite integral by the above formula,
is
deduced
unsatisfactory,
DEFINITE INTEGRALS.
24, 25]
43
the true nature of the definite integral being that of the limit of a
sum.
Integrand or Limit.
Infinite
25.
The
definition of the
definite integral presupposed that the integrand was finite in the

If there should exist points in the
field of integration ab.
region
ab at which f(x) became infinite, the integral would in general
have no meaning. In case however there is a single point c for
which f(x) becomes infinite, if h^ and h 2 are positive numbers
rb
/c-7t!
a,
finite
now
sum
If
meaning.
of each other the
as /^
f( x ) dx have a dec+h t
A2 approach zero independently
f(x) dx,
and
rb
re- hi
/<*)<**+ J
J a
c+hf
f(*)dx,
approaches a definite finite limit, the value of that limit

meant by the definite integral,
For example,
then
x=
for
f
J
c,
dx
= hm
,
what
is
let
f(x) becomes
a \^
is
,.
(-
~k
dx
>
7tj=0 J
C)
1
infinite.
c~h
(a
(&
l
c}
h 2 =Q J c+7i 2
G)
~k
+
~ (b
c)
~k
(h)
G)
~k
There
is
a limit as
/^
and h 2 approach zero only
if 1
>
0.
rx
In
limit
like
manner
if
the integral
J a
f(x) dx approaches a
when the
defines the
limit x increases indefinitely, then

meaning of the definite integral
Let, as before,
(x-c)
JaO-c)*
k)
~I-k
this
finite
value
44
DEFINITE INTEGRALS.
>
As x increases
1, when
[iNT.
III.
indefinitely, this approaches a finite limit only if
dx
(a
- c) ~k
l
k-l
Differentiation of a Definite Integral. Suppose that

26.
the integrand is a function of a parameter u as well as of x. Then
in the case of a function of x that is capable of representation by a
curve, if we change the parameter u we change the curve, and if
a continuous function of
is
f(x, u)
u corresponds an
an infinitesimal change in
u, to
infinitesimal change in the curve.
The area
rb
represented by the definite integral

J
f(x, u) dx changes by the
area of the narrow strip added to or included between the two

curves, and we may find the ratio of this change to the given
u.
We thus get a geometrical notion of the meaning of
the derivative of the integral with respect to u. Now by the
definition of the derivative
change in
+ h) dx -
O, u
,6
%dttj/
u)'
/(#,
v
dx
lim -^~
-^h
h=0
h=Q
f(x, u) dx
J a
now becomes
a question whether we may change the order

of taking the limits involved in the integration and in
making h
approach zero. If f(x u) is a continuous function of x and u we
It
may do
this*,
and since
lim
^-2
^=0
J^
'
U'
^' u '
du
we have
We
of its
have already considered the definite integral as a function

upper limit, and have found, 24,
dF (v)
*
So Kronecker, Theorie der einfachen und der vielfachen Integrate, p. 26, (the
word gleichmassig being superfluous, vid. Harkness and Morley, Theory of Functions,
For a more careful statement, see Tannery, Theorie des Fonctions d'une
64).
Variable,
166.
25
DEFINITE INTEGRALS.
27]
In
manner
like
fj
rbb
<& =
v
dw] w j/(*)
now
If
45
w, are
u, v,
functions of a variable
all
derivative of the definite integral according to
dF_d[
~
dt
dt] w
j(x> u} **
we have
for the
du c>Fdv dF dw
_dF
~
+
+
du dt
dv dt
dw
du
= -T7
cfa
df(x,u)
-^
dw
w
[
J
dt
dv
'dt
Double and Multiple Integrals.
27.
t,
t,
.dw
'
dt'
Suppose we consider
a continuous function of two variables, x varying from a to 6, and

y varying from g to h. We may represent f(x, y) geometrically as
the third coordinate z of a surface, erected perpendicular to the
plane of xy.
If
now we subdivide the

a<
<
x-i
x.2
<
interval ab
xs < xn ^ <
by points
b,
and the interval gh by points
and draw through these points
lines parallel to the axes of
x and y,
dividing the plane into rectangles, and at a point in each rectangle

erect perpendiculars meeting the surface, the sum
s=n r-m
(X8
=l r=l
-I
< f < X*
6-
represents the volume of the rectangular prisms constructed on

%S --L> 2/r "~ 2/r-i a s bases, and altitudes
the rectangles with sides xs
If as
we make the number of points of subdivision increase

sum approaches a limit, this limit defines the
without limit, the

definite integral
rh
J'b
a
/O,
y)
s=n
r=m
n=*m = oc=l
r=I
dx dy = lim lim 2
(xs
a?
(yr
- yr-i)f
We shall find by reasoning similar to that used in

the condition for the existence of a limit is that the sum
s=nr=m
2 2
s=\ r=l
(xs
-x _ )(y -y _ )D
s
sr
r-
22 that
46
DEFINITE INTEGRALS.
where
sr
[INT.
III.
the oscillation in the interval xs _ l} xS) yr-\, yr
is
approaches the limit zero.
In forming the doable sum we may proceed with the summax first, in which case
tion according to
(yr
- yr^) lim 2
?i=oo
- #-i)/(?
(x%
is
rh / rb
J
Or we may sum
(xs
- ^_j)
lim
M=o>
(\J
a f(*,y)dx)dy.
with respect to
first
m
2 (yr - yr-i)f(^s
(yr
- yr-i)
(n
[h
J
(J a
/(
y)
xs ^ and yr
\
*My=
)
rb
(y r
;?/
- 2/r
_1
%f ^= J a J g /(,
)
/(#, y)
(J g
Accordingly,
/& rh
rh
J a
(,]2
U
fm
a?_i)
rb
y) dy,
f( x >y}dy)dx.
(J
/(&,
is
But we have always

m
SCB
rh
J g
however small
which case
= 0* - x -i)
7)r)
y> in
and the double limit
f(%, Vr) dx,
Ja
and the double limit
rb
= (y r - yr-i)
s , rj r )
(In writing a double or multiple integral we shall write the

integral signs with their limits in the same order as the differentials.)
We
might now deduce theorems for the double integral similar

we have already deduced for the single integral. In
the
particular,
independence of the limit on the mode of suband
the
theorem of the mean may be demonstrated, and
division,
to those that
the extension of the definition
The
made when the integrand
or the
of an integral may be
in
an obvious manner.
and
extended to triple
multiple integrals
limits
28.
become
infinite.
definition
General Definition of Definite Integral.
We
have
in the preceding definition of a double integral assumed that the

limits of integration with respect to x and y were independent.
DEFINITE INTEGRALS.
27, 28]
If instead of a rectangle in the JTF-plane
curve, given by an equation
(f>
47
we should take any closed

could in like manner
= 0, we
y)
(x,
divide its area into rectangles, erecting perpendiculars in each,

and define the definite integral as the limit of a similar sum for
the
new
field of integration.
More generally, let f(M) be a function of a point M, moving

If we divide any area S in the
either in a plane or in space.
in
volume
r
or
any
space up into a number of parts, take the
plane
value of f(M) at any point within each of those parts, multiply
each value by the area or volume of the part in which it is taken,
and add together for all the parts into which the area or volume is
divided, the limit approached by this sum as the number of parts
increases without limit in such a way that each dimension of
every part approaches zero,

definite integral of
may
if
such a limit
exists,
is
called the
through the region in question.
f(M)
We
write the integrals
jff(M)dS
respectively.
In each
or
fjjf(M)dr,
case, the field of integration
It may be easily
pressly specified.
to
the
equivalent
preceding.
shown that
must be ex-
this definition is
particular mode of subdivision is by drawing level lines

or surfaces for two or three orthogonal coordinates q lt q2 q s
.
have then,
20),
dq.dq,
a/o
ar
-.
/lj /1 2
Then when we
(
its
fh
"<2
s
.
sets of coordinates
and p^p^pz with parameters, h^h^h^ and g^g^g*,
when taken through any equal

and
_ dq, dq, dq
Suppose that in two different

qi,q 2 ,q s
finite portions of
a volume
T.
meaning of the definite integral

manner of subdivision, we see as in
consider the
independence of the
23) that the above integrals, being respectively equal to
h l h 2 h s f(q l q 2
,
JJj
We
q.3 )
dr and
IjJ
g^g^ $ (p lt p, p3 ) dr,
DEFINITE INTEGRALS.
48
can be equal only
if
[iNT.
the point functions
h1 h h 3 f(q lt q2 q s )
,
it
= g,g g
2
3 </>
(p lt
p,, p.),
everywhere in the volume T for the same point

or p 1} p.2) p 3 in the respective coordinates.
<?i> <?2> #3
Calculus of Variations.
29.
follows
since
is
we
have
to
III.
M, denoted by
We
shall frequently in what

use of the calculus of variations, which,
always in connection with definite integrals,
make
shall use it
introduced here.
In the differential calculus, we have to consider questions of

maxima and minima of functions. A function of one variable has
a maximum or minimum value at a certain value of the variable if
the change in the function is of the same sign for any change in
the variable, provided the latter change is small enough. Since if
f(x) is continuous at x,
f(x+ h) =/(*) + hf (*) + ~ /"(*) + ......

,
If
is
small enough, the expression on the right will have the
sign of the first term, which will change sign with

the condition for a maximum or minimum is
h.
Accordingly
/-a
we change the form of the function
For
to take place gradually.
made
be
change may
a
in
have
curve
we
given
any way, e.g.
suppose
Suppose that
^1(0, V
= F,(t\
such a
instance
= F*(t\
where the F's are any uniform and continuous functions of an

independent variable t. If we change the form of the F's we shall
change the curve suppose we change to
x
To every
=G
value of
sequently to each point
on the other.
(t),
y=G,(t\ z=G>(t).
corresponds one point on each curve, conon one curve corresponds a definite point
Such a change from one curve
a transformation of the curve. The change
to the other is called
may be made
gradually,
DEFINITE INTEGRALS.
28, 29]
we
49
For every value of e we shall have a particular curve for e=0

shall have the original curve, for e = 1 the final curve, and
for intervening values of e other curves.

cause a small change in the curve, and if
the transformation an
call
changes in the values of
infinitesimal
x, y, z, or of
small change in
e will
we
e is infinitesimal
shall
The
transformation.
any functions
thereof, for
an
are called the variations of the functions, and
infinitesimal
change
are denoted
by the sign
e,
8.
Suppose we denote
dx
d*x
d kx
dt'
d?'
dtk
Ai~r*
by the
letters
of,
and by
<f>
any function
f
<
(t,
*,
x",
x, y, z, x',
y',z
......
*,
y,
*<*>, ......
a?<>,
y(), *<),
and consider the change in $ made by an infinitesimal

mation, where we replace x y, z"by
transfor-
where f
mi_
f are arbitrary continuous functions of
77,
Then
da
or
Tt
dx
j ^
replaced by
i
is
+ e di; and
rf (fc)
t.
fc
a?,
-^
c? a?
by
^+
+ep>.
ie.,by
Hence
<>
becomes
which developed by Taylor's theorem for any number of

terms in equal powers of e
gives on collecting
(t,
where
W.
E.
x, y, z,
...... )
<#>!
2
<
2]
......
e*
Jj#*
variables,
+ ......
DEFINITE INTEGRALS.
50
The terms e0 1}
eft
If for
we put
are called the
<f>k
80,
2
2
S 2 0, 6*01-
first,
second, &th varia-
thus see that the variations of

t,
= 8*0.
we
z, x', y', z'
successively x, y,
arbitrary functions of
last
III.
and denoted by
tions of
We
[INT.
get
x, y, z are infinitesimal
the independent variable, and from the
equation
'
dk
d*
d*
d*a>\
the operations of differentiation and variation are commutative, for the variables x, y, z.
that
is,
It is evident that 0^ is the &th derivative of

0.
e for the value e
with respect to
Since we may always change the order of differentiation, it

evident that the commutative property holds for any function.
is
Let us now find the variation of the integral

f*i
/=
Changing x
to
0(,
x, y
x, y, z,
z,
...... )dt.
to
+ 8a,
1 + 81 + ^1+
to
......
y+
By,
x to
x'
+ &c',
+ S0 + JS 0+
2
(0
[
J
etc.,
...... )dt,
and the variations are
JU
to
that
is,
tative.
the operations of variation and integration are commu(The limits have been supposed given, that is unvaried).
These two principles of commutativity of 8 with d and
form
the basis of the subject.

30.
Line and Surface Integrals.
in space joining a point
between
and
B we
If
to a point B,
place n
points,
we
consider any curve
and
p1} p
if
z
on the curve
-p n-i, whose
DEFINITE INTEGRALS.
29, 30]
^-coordinates are
51
xn-i, multiply the length
a? 1} x.2
of each chord
- *w)2 + (y8 - y-i)2 + (*. - ^-i)
P+-I PS,
a
,
by the value of the point function f(p) at some point TTS in the
arc bet ween p8 -i,p s and take the sum for all the arcs into which
,
the curve has been subdivided, then if this sum approaches a

finite limit as the number of subdivisions increases
indefinitely,
this limit is called the
along the curve
AB, and
line-integral of the point-function

is
/ (p) ds = lim 2/(7r,) V (x
If
f(p) =
1,
f(p)
denoted by
x^-tf
+ (y8
y8_^f
+ (z
^_ ) 2
the integral represents the length of the curve
AB
If in forming the line-integral we had multiplied the values of
/(TT S ) by the ^-projection of the chord, instead of by the chord
itself,
we should have
arrived at the integral already defined,

n
= lim S/(TTS ) (x8 - x8^\
except thatf(p) is a function not of x alone, but of the point on

the given curve corresponding to x. It will in general happen
that as we go continuously along the curve from A, x will not
increase continuously but will increase to a certain value
then decrease. As x decreases,
however, reassuming
values,
we
are
(7,
and
previous
still
continuing
along the curve and reaching
new
points and corresponding

values of
(p) which are to be
"
used in the integral.
The
fane-
FlG 8>
which would otherwise
in
not be uniform
x, becomes uniform when defined in this manner, so
*
tion /,
that
if
we
we must
interpret in the ordinary
separate
it
manner the
\f(p) dx,
each of which x
integral
into several integrals, in
varies in one direction throughout, taking in each the values of

belonging to the corresponding part of the curve. If however
we
write
42
DEFINITE INTEGRALS.
52
where
is
- ^-i) + (y - y -i) +
2
V(*.
and take
becomes
for
(z8
= dx
= cos (as,
-j,
x),
above integral
the independent variable, the
jfdx
in
Xg^
=
Xs
lim -
[INT. in.
|(/^)
ds
=
J*/cos
(ds, x)
ds,
which there can be no ambiguity.

In like
manner
if
we divide the area of any
surface
into
of a point-function
parts, multiply the area of each by the value
for
all the parts, the
and
sum
at
some
on
that
point
part,
f(p)
FIG. 9.
limit approached
by the sum,
if
any, as both the dimensions of the

f(p) over the
zero, is called the surface-integral of
parts approach
given portion of surface and denoted
by
jjf(p)dS.
Here
if
we multiply f(p) not by the
surface 8, but
by
surface-integral
its
to
area of the part of the

we reduce the
projection on the ^TF-plane,
the
double
integral
/ (p)
dSxy
already
treated, with the exception that the point-function depends not

only upon x and y but upon the surface 8. If, as is generally the
upon the same part of

the
the ^TF-plane,
integral must be interpreted in an analogous
manner to that used in the case of the line-integral. If n is the
case, several regions of the surface project
30, 31]
DEFINITE INTEGRALS.
normal to the surface
S drawn
surface, it
element
53
always toward the same side of the
easily seen that the area of the projection of the

on the
F-plane is dS^ or dSz dS cos (nz).
may
is
dS
We
accordingly write the surface integral,
f (p) cos (nz) dS =
fjf (p)
dSz
=jjf(p) dxdy,
with the understanding that in the last form the integral is to be

taken over the projection of the surface S on the
F-plane, in
such a manner that the projection is to be divided into regions for
S in the corresponding portions of S

the
towards
either
F-plane or always away from it,
always
points
and that those parts of the integral for which the normal points
It will be seen
in opposite directions are given opposite signs.
each of which the normal to
that this corresponds exactly to the interpretation of the lineintegral in terms of x, when x changes its direction of variation.
The
first
form of the integral above, with
integration,
is
preferable, its
as the variable of
meaning being unambiguous.
Dependence of Line Integral on Path. Stokes's

Theorem. Curl. The line integrals with which we shall have
31.
do are integrals of a vector point-function. If R is a

vector function of the point, whose projections are X, Y, Z,
functions of x, y, z, the component of
along the tangent to the
curve AB at any point is, since the direction cosines of the
ost to
dx
tangent are
-3-
ds
dy
,
-/-,
ds
dz
-j-
ds
R C os(R,ds) = X^+Y^
+ Z^.
ds
ds
ds
The
may be
line integral of this resolved
component
written
with the understanding of the previous section.
X, Y, Z, being given for every point x, y, z, the

curve AB.
/
will
in
integral
general depend on the form of the
If we make an infinitesimal transformation of the curve, the
The
functions
DEFINITE INTEGRALS.
54
integral will change,
have
variation.
and we
[INT.
now seek an
shall
III.
expression for the
We
.
r A- u
-7
a5
r u j.
~^
as
-7- -r
SX=^SX+
Now
and
We may
-f
as
r oz/
-j
rfs
jjSy
-;
z/
-r-
rfs
rfs
^Sz,
= dp
^
as
as
perform upon the term

o"s
an integration by parts
B
[
I
JA
where
dX
J
I
Bw~-
ds
ds,
IB
Xx
the point
X d(Sa;)
ds=XSa)
\
ds
signifies that
B we
subtract
its
from the value of the function
*L
j_
ds
dx ds
dy ds
at
Now
value at A.
dX = dXdx
_ __ dXdy
__
XSx
dX dz
___
dz ds
_j
<
'
Performing similar operations on the other terms we have
ds
dx
'
dz
J ds
y \dx ds
- Sz fdZ
N
dx
-j-
\dx ds
Now
fixed,
&,
if
dz ds
dy ds
dx ds
dy ds
dz ds)
dZ dy dZ dz\~\ ,
+ ^- -r)\ds.
-f
ds
dz ds/
+^
dy
in the variation the ends of the curve
By,
Sz vanish for
and
are
and B, and the integrated part
DEFINITE INTEGRALS.
31]
X8a)
+ YSy + Z&z
vanishes.
Collecting those terms under the
sign of integration that do not cancel,
Now
in
the
55
we
have,
the determinant Sydz

Szdy is the area of a parallelogram
the
projections of whose
F-plane
on the F- and j-axes are dy, dz, Sy,

That is, if we consider the infinitesimal
parallelogram whose vertices are the points
s, s + ds and their transformed positions, the
sides
Sz.
above determinant
is
the area of
its
pro-
If the area of the

jection on the F^-plane.
and
n
is the direction of
is
dS
parallelogram
as in
its
normal,
we have
30
dy = dS cos
Bydz
&z
Sz dec
= dS cos (ny),
Sydx = dS cos (nz),
Sxdy
(nx),
Sxdz
and
which
is
in the form of a surface integral over the strip of in-
finitesimal width.
If we again make an infinitesimal transformation, and so

continue until the path has swept out any finite portion of a
FIG. 11.
DEFINITE INTEGRALS.
56
surface S, and
sum
all
[iNT.
III.
the variations of I, we get for the final

/ for the two extreme paths 1 and 2
result that the difference in

is
the surface integral
cos
cos
..ds
taken over the portion of the surface bounded by the paths 1 and
2 from A to B.
Now /x may be considered the integral from B
1: is the integral around the

along the path 1, so that /2
closed path which forms the contour of the portion of surface S.
accordingly get the following, known as
to
We
STOKES'S THEOREM*.
The
line
integral,
around any closed
contour, of the tangential component of a vector R, whose components are X, Y, Z, is equal to the surface integral over any
portion of surface bounded by the contour, of the normal component of a vector co, whose components f, 77, f are related to
X, Y, Z by the relations
_
dx
'
dy
The normal must be drawn toward that side of the surface that
shall make the rotation of a right-handed screw advancing along
the normal agree with the direction of traversing the closed
contour of integration.
(lids
=
I
Xdoc + Ydy + Zdz
co
cos (con)
cos (nx)
The
77
dS
cos (ny)
+ fcos (nz))
dS.
related to the vector point-function

by the differential equations above is called the rotation, spin (Clifford), or
Such vectors are of frequent
curl (Maxwell and Heaviside) of R.
*
vector
co
The proof here given
von Helmholtz.
p. 73.
is
from the author's notes on the lectures
similar treatment
is
given by Picard, Traite
of Professor
$ Analyse,
Tom.
i,
DEFINITE INTEGRALS.
31, 32]
57
occurrence in mathematical physics. The curl may be derived

from the primitive vector by the application of Hamilton's vector
differential operator
dx
16) to a vector point-function R,
++dzj
dy
\dy
dz
dx
/BY
+J.fix_d_z\
'
So that the vector part resulting from the application of the

The scalar
operation V to a vector point-function gives its curl.
part
dY
_ fdX_
3,
has an important interpretation to be given shortly.
the
[The significance of the geometrical term curl can be seen from

physical example in which the
represents the velocity of a

point instantaneously occupying the
position x, y, z in a rigid body turnvector
ing about the J-axis with an angular

= cop
o>.
Then the vector
velocity
is
perpendicular to the radius p
its
components are
and
FIG. 12.
X = R cos (Rx) = R sin (px) = R - =
yco,
R-=
xco,
qj
F= R cos (Ry) = R cos (px)

where
o> is
constant,
and
dY
dx
dX_
~
dy
So that the ^-component of the
curl of the linear velocity
is
twice
the angular velocity about the Z-i

32.
Lamellar Vectors. In
/ in the previous section,
integral
are perfectly arbitrary functions of
finding the variation of the

since the variations Sx, By, 82
5, if
the integral
is
to
be inde-
DEFINITE INTEGRALS.
58
[INT.
pendent of the path, 87 must vanish, which can only happen
III.
for all
possible choices of &c, By, 82 if
_
that
is,
dition
the curl of
if
=
dz
dy
=
dx
dz
dx
==
dy
vanishes everywhere. In case this cononly on the positions of the limiting
/ depends
is satisfied,
is given,
and B, and not on the path of integration. If
points
/ is a point-function of its upper limit B, let us say </>. If
is
in
the
a
distance
in
a
direction
to
the
s
B',
displaced
change
given
function
<f>
is
rSf
<!>>-
<!>B
=
J
and the
8= o
we take
change to the distance
+-**JA
= x^+Y^> + Z^
ds
ds
ds
ds
s
the derivative of
If
(Xdx+Ydy + Zdz),
limit of the ratio of the
lim
is
(/>
in the direction
s.
s successively in the directions of the axes of co-
ordinates,
30
7T
dx
that
is,
function
= A,
Y
=v
d<f>_ 2
-^r
dy
= /Jy
7
3<t>_
TT
dz
the vector differential parameter of the scalar
is
<f>.
Accordingly the three equations of condition equivalent to

= are simply the conditions that X, F, Z may be reprecurl
sented as the derivatives of a point-function.
In this case the
expression
Xdx+
is
Ydy + Zdz
dx
dx
+
dy
dy + dz dz
= d<t>
called a perfect differential.
From
we
the definition of the parameter of a scalar point-function,

magnitude of the parameter is inversely propor-
see that the
tional to the
normal distance between two
infinitely
near level
Such a pair of surfaces will be called a

lamina.
For this reason a vector point-function
surfaces of the function.
thin level sheet or

that
may be
represented everywhere in a certain region as the
DEFINITE INTEGRALS.
32, 33]
vector parameter of a scalar point-function will be called a laminar,

or lamellar vector (Maxwell).
The
scalar function
(/>
(or its negative) will
sometimes be termed
the potential of the vector R.
Connectivity of Space.
33,
have
supposed
31
in
that
it
Green's Theorem.
We
was
possible to change the path 1 from

into the path 2 by continuous deto
formation, without
considered.
space
passing out of the
portion of space
which any path between two points

may be thus changed into any other
between the same two points is said to
be singly-connected. For instance, in
in
FIG. 13.
the case of a two-dimensional space, any

area bounded by a single closed contour will have this property.
If, however, we consider an area bounded externally by a closed
contour 0, and internally by one or more closed contours /, Fig. 13,

such as the surface of a lake containing islands, it will be possible
go from any point A to any other point B by two routes which

cannot be continuously changed into each other without passing
out of the space considered, that is traversing the shaded part.
to
The space in Fig. 13 between the contour C and the island / is

said to be doubly-connected.
may make it singly-connected
a
barrier
the
island with the contour (7,
by drawing
connecting
We
If no path is allowed which

represented by the dotted line.
crosses the barrier the space is Singly-connected.
A
closed
three-dimensional space bounded externally by a single

surface is not made doubly-connected by
For instance,
containing an inner closed boundary.
the space lying between two concentric spheres
allows all paths between two given points to be
deformed into each other, avoiding the inner sphere.
But the space bounded by an endless tubular

Fig. 14, is doubly-connected, because we
to
in either direction of the tube,
surface,
may go
from
and the two
paths cannot be deformed into each other. We may

make the space singly-connected by the insertion of
FIG. 14.
DEFINITE INTEGRALS.
60
[INT.
III.
a barrier in the shape of a diaphragm, closing the tube so that

one of the paths is inadmissible.
The connectivity
is
defined as one
number
of
of a portion of space
more than the
barriers
or
least
diaphragms
necessary to make it singly-connected.

Thus the space in a closed vase with
three hollow handles, Fig. 15, is quadruply-connected.
FIG. 15.
We
shall
always
suppose the spaces with which we deal

in this book to be singly-connected, or
to be
made
so by the insertion of dia-
phragms, unless the contrary is expressly stated.
Suppose that IT is a point-function which, together with its

derivative in any direction, is uniform and continuous in a certain
Then its deriportion of space r bounded by a closed surface 8.
dW/dx is finite in the whole region, and if we multiply it

of volume dr and integrate throughout the volume
the
element
by
r, the integral is finite, being less than the maximum value
vative
attained by
We
dW/dx
in the space T multiplied
by the volume
r.
have at once
keeping y and z constant, we perform the integration with

respect to x, the volume is divided into elementary prisms whose
If,
FIG. 16.
sides are parallel to the JT-axis,

sides dy, dz.
and whose bases are rectangles with
DEFINITE INTEGRALS.
33]
The portion
of the integral
due to one such prism
fdW,
ox
the integral
is
is
l-x- dx.
dydz
Now
61
to be taken
between the values of x where
the edge of the elementary prism cuts into the surface S and
where it cuts out from the surface. If it cuts in more than once,
it
since the surface
will,
Let the values of
times.
^~dx=W
then
being the value of
cut out the same
is closed,
number
of
x, at the successive points of cutting, be
- W.+ Wt-Ws
W for
and
#j.,
dW J j
-^ jdxdydz
,
Now
let
the surface
these
all
- W, +
d81} dS2
...
dSm denote
...
Fm - Tf
27l
_t ]
%^.
the areas of the elements of
cut out by the prism in question at xlt x2 ... x.m

have the same projection on the F^-plane, namely dydz.
,
Now if all these elements are considered positive, and if n be the

normal always drawn inward from the surface S toward the space
T, at each point of cutting into the surface S, n makes an acute
angle with the positive direction of the axis of X, and the projection
of
dS
is
dydz = dS cos (nx),

but where the edge cuts out n makes an obtuse angle, with
negative cosine, and therefore
dydz=
We may
dScos
(nx).
accordingly write
W
= W
F
dydzWi =
QOB(nl x)dSi,
dydzW?,
cos(n 2 x)dS2
cos(n 3 x)dS3
cos
62
DEFINITE INTEGRALS.
[iNT.
III.
and
in integrating with respect to y and z we cover the whole of

the projection of the surface S on the F^-plane. On the other
hand we cover the whole of the surface

integral
taken
is
all
S, so that the
volume
transformed into a surface integral,
over the surface S.
In like manner we
transform the two similar integrals
may
cos
Applying this lemma
(m)
dS.
to the function
Avhere both U, V and their derivatives in any direction are uniform

and continuous point-functions in the space T, we have
W= U
Similarly for
and
W=U,
for
oz
Adding these
three,
and performing the
dUdV
=-
cos (nx)
differentiations,
+
+ âF^^SZ
+ g- cos (ny) + =- cos (iw))
DEFINITE INTEGRALS.
33]
or,
transposing,
and denoting the symmetrical integral by
-|JV jôos (nx)
This result
is
known
as
+
|^cos
(ny)
|Jcos (**)}
J",
dS
GREEN'S THEOREM*.
the definition of differentiation in any direction the parenthesis in the surface integral on the right is
By
if
P v is
the parameter of V.
Since the integral on the left is symmetrical in

may interchange them on the right, so that
Writ
riting
which
this equal to the former value,
will
and F, we
and transposing, we obtain
be referred to as Green's theorem in
its
second form.
is
It will be noticed that the

integrand on the left in the first form
the geometric product of the parameters of the functions 7 and F,
We
shall,
unless the contrary
is
stated, always
mean by n
the
internal normal to a closed surface, but if necessary shall distinguish the normals drawn internally and externally as iii and n e
.
we do not
care to distinguish the inside from the outside

shall denote the normals toward opposite sides by n v and n^.
If
An Essay
Electricity
p. 25.
we
on the Application of Mathematical Analysis to the theories of

Nottingham, 1828. Geo. Green, Reprint of papers,
and Magnetism.
DEFINITE INTEGRALS.
64
[iNT.
Second Differential Parameter.

If for the function U we take a constant, say
III.
34.
1,
%-%-%-* *and we have simply
The function
which, following the usage of the majority of writers, we shall

denote by AF, was termed by Lame* the second differential parameter of F. As it is a scalar quantity it will be sufficiently
distinguished from the first parameter if we call it the scalar
We have accordingly the theorem giving the relation
parameter.
between the two
:
The volume
integral of the scalar differential parameter of a

uniform continuous point-function throughout any volume is equal
to the surface integral of the vector parameter resolved- along the
outward normal to the surface
the volume.
obtain a geometrical notion of the significance of

of ways.
Applying the above theorem to the volume
number
in a
8 bounding
AF
We may
enclosed by a small sphere of radius R, we have, since

direction of the radius, but drawn inwards,
is
in the
on
where
at the centre of the sphere,

s on the
of
a
definite
the
integral
significance
remembering
the value of
mean, we have
as a
Iim
is
Now
surface.
^ (Mean of F on surface
at center! x
Area of surface
R=o-tt
=
*
tions.
1 1 1
G. Lame.
A Vdr = (Mean
Lemons sur
Paris, 1859, p. 6.
les
of
A F in
sphere) x
Coordonnees cwvilignes
et
Volume
of sphere.
lews diverses Applica-
DEFINITE INTEGRALS.
34]
Now
surface
since the
65
volume of a sphere is J the product of the

we have, on making jR approach zero,
by the radius,
V on surface
(Mean
Tr
T
AF=3Lim!.
&po
.8=0
The negative
scalar
V at
center]
J
A V was
accordingly called by
Maxwell the concentration of V, being proportional to the excess of
the value of
at any point over the mean of the surrounding
parameter
values.
It is evident from this interpretation of
AF that
if
the
concentration of a function vanishes throughout a certain region,

then about any point in the region the values at neighbouring
points are partly greater and partly less than at the point itself,
so that the function cannot have at any point in the region either
maximum or minimum with respect to surrounding points.

function that in a certain region is uniform, continuous, and has no
The study of
concentration is said to be harmonic in that region.
such functions constitutes one of the most important parts, not

only of the theory of functions, -but also of mathematical physics.
By means
of the
same theorem we may obtain another repre-
A V.
Let us apply the theorem to the space included

between two small concentric spheres of radii R^ and R2 = R^ + h.
sentation of
Then
at the outer sphere
and the surface integral being taken over the surface of both
spheres, with the normal pointing in each case into the space
between them,
As we make h approach
destroys the
first,
zero,
the
first
term of the second integral
and
= Lim
so that
(f
H=Q JJ B,
Now hdS
is
the element of volume dr, so that
AF
92
defined as the
W.
E.
mean
value of the second derivative
may
be
for all
DEFINITE INTEGRALS.
66
directions as
we
[INT.
This interpretation
leave the point.
due
is
III.
to
Boussinesq*.
We may
operator
derive the parameter
by applying Hamilton's
Solenoidal Vectors.
Divergence.
35.
AF
twice to V,
If the components
of the vector parameter are
we have
= ax
ay
a*
dx
dy
dz
and the above theorem becomes

cos (Pn)
If
P is
integral
is
dS = -
lj(X
cos (nx)
+ Fcos (ny) + Z cos (m)) dS
everywhere directed outward from the surface

positive,
S, the
and
8F dZ\
mean ftX
^- + -5- +-5- >0.
\dx
Accordingly
+-^~"
+ ^~
oy
is
dzj
called the divergence of the vec-
whose components are X, F, Z, and will be

31 we find that the
denoted by div. R. Comparing with
divergence of a vector is minus the scalar part of the V of the
tor point -function
vector,
div.
R = - SVR.
The theorem as just given may be stated as follows, and will

be referred to as the DIVERGENCE THEOREM The mean value of
:
normal component of any
the
*
vector point-function
outward from
Boussinesq, Application des Potentiels a Vetude de Vequilibre
ment des
solides elastiqtLes, p. 45.
et
du mouve-
DEFINITE INTEGRALS.
34, 35]
any
S within which
closed surface
67
uniform and
the function is
continuous, multiplied by the area of the surface, is equal to the

mean value of the divergence of the vector in the space within S
The theorem was proved for a vector

multiplied by its volume.
which was the parameter of a scalar point-function V, but it is
evident that it may be proved directly by partial integration
whether this is the case or not.
Let us consider the geometrical nature of a vector pointwhose divergence vanishes in a certain region. In the
function
neighbourhood of any point, the vector will at some points be

directed toward the point and at others away.
may then
draw curves of such a nature that at every point of any curve the
at that
tangent is in the direction of the vector point-function
We
point.
Such
curves will
Suppose that such
lines
be called lines of the vector function.

be drawn through all points of a closed
curve, they will generate a tubular surface, which will be called

a tube of the vector function. Let us now construct any two
surfaces
8 and $
cutting across the vector
tube and apply the divergence theorem to the

portion of space inclosed by the tube and the
Since at every
two surfaces or caps 8l and Sz
the
surface
of
tube, R is tangent
point on the
to the tube, the normal component vanishes.
.
The only
parts contributing anything to the

surface integral are accordingly the caps, and since the divergence
everywhere vanishes in r, we have
If Rcos
JJsi
(Rn^
dSj.
11
JJs
R cos (Rn
2)
dS,
0.
If we draw the normal to S2 in the other direction, so that

we move the cap along the tube the direction of the normal
as
is
continuous, the above formula becomes
Rcos(Rn )dS l
R cos (RnJ dS. = 0,

2
or the surface integral of the normal component of

cutting the same vector tube is constant.
Such a vector
r)
condition
ôx
(Maxwell).
will
r)V
JT
|-
R over any
cap
be termed solenoidal, or tubular, and the
7^7
dy
We may
~- =
will
be termed the solenoidal condition
dz
abbreviate
it,
div.
R = 0.
If a vector point-
52
DEFINITE INTEGRALS.
68
function
of which
it is
is
lamellar as well as solenoidal, the scalar function
the vector parameter
dx
x-
ox
[INT. III.
solenoidal
dY dz =
+ 3- + ~oz
i
is
div.
harmonic, for
oy
vector
may be
represented by its tubes, its

the
by
tangent to an infinitesimal tube,
direction being given
and its magnitude being inversely proportional to its cross-section.

As an example of a solenoidal vector we may take the velocity of
particles of a
moving fluid. If the velocity is R, with components

the
amount of liquid flowing through an element of
X, Y, Z,
surface dS in unit time is that contained in a prism of slant
height R, and base dS, whose volume
is
R cos (Rn) dS.

The
total
surface S,
jJR
is
cos
quantity flowing in unit time through a
flux, or
(Rn) dS = [j(X cos (nx)
+ Y cos (ny) + Z cos (nz)) dS.
Such a surface integral may accordingly be

the vector
A tube
R through
of vector
called the flux of
8.
R is a tube such
that no fluid flows across
its
such as a material tube through which liquid flows, and the

divergence theorem shows that as much liquid flows in through
sides,
one cross-section as out through another,

holds.
If the liquid
is
As a second example
which
is
if
incompressible, this
the solenoidal condition
must
of solenoidal vectors
of course be true.
we have any
vector
the curl of another vector, for
_a_fa^_3F)
dz)
dx\dy
_a_
>az_a^| + ^_(8F_aj)
dy\dz
dx)
dz\dx
dy]
identically.
The equation
is
called Laplace's equation,
Laplace's operator.
and the operator
A=^ +^ +
2
DEFINITE INTEGRALS.
35, 36]
69
Representation of Solenoidal Vector. Multiplier.

We have obtained in 32 a means of representing a lamellar
vector-function by means of the level surfaces of its potential
function.
By means of Jacobi's multiplier we may find a somewhat similar representation for a solenoidal vector. If we suppose
the curves drawn whose tangent at every point has the direction
of the vector function E whose components are X, Y, Z, since the
36.
direction cosines of the tangent are
dx
dy dz
ds' ds' 'ds'
le
curve
is
defined by the differential equations
dx:dy:dz = X:Y:Z.
(i)
integrals of these equations will each contain an arbitrary

Let us suppose that an integral is of the form
constant.
The
\ (x,
y, z)
const.
Then we must have
ax
+ _ rfy + _ (b =
7 .
ax
dfl
ax
_
)
and since dx, dy, dz are proportional to X, Y, Z,

(2)
This partial differential equation may serve as a definition of an

I
Geometrically
integral of the system of differential equations ( ).
is perpendicular to the normal to the
it shows that the vector
surface
X = const.,
is, is
tangent to the surface.
If
/JL
= const,
a second integral, then
is
(3)
ox
and since
IJL
that
= const.,
oz
=
tangent to a surface of each family X const.,
are the intersections of the
the lines of the vector
is
X with the surfaces p.

X, Y, Z, we may determine
From
surfaces
in
-5-
oy
^r
d\
(2)
their ratios.
ax
and
(3), linear
We
obtain
equations
DEFINITE INTEGRALS.
70
If
[INT.
III.
M be a factor to be determined, we may put

MX = A, MY=B, MZ=C,
(5)
where A, B,
are the above determinants.
But the determinants A, B, C, if differentiated by x, y, z,

respectively and added, are found to satisfy identically the solenoidal
condition
SB
SA
-=
(6)
I-
dec
so that
Consequently
dz
ty
we have the equation
(7)
SC
+ -5- = 0,
-3for
M,
S(MX)
d(MY) .S(MZ)
~to~
~fcT
for
any continuous vector function R
to find a scalar multiplier
M that
components are
MX, MY, MZ,
itself solenoidal,
the equation for
say
1,
~^~
so that in this case
shall
solenoidal.
is
dz dy
_
dx dy
X,
/JL
If the vector
satisfied
x dz
if
possible
by any
is
constant,
dp _ 9X dp
dy dz
see
it is
the vector whose
we have
~ _ dX
But
we
make
'
'
'
dy dx
M denote the vector parameters of the functions

the
definition of the vector product,
by
PA
E
If we consider two infinitely near surfaces of the first family
which X has the values X and X + d\ respectively, the normal
distance between which is dn^, we have by
16 and 20
for
d\
and
Considering two infinitely near surfaces of the other family

/A + dp, we have in like manner for their normal distance
//,
DEFINITE INTEGRALS.
36, 37]
The area
a right section of
of
formed (Fig. 18)
,_
71
the four-sided tube
thus
is
-sin^x^-sinCPxP,)'
and multiplying this by the value
of*,
/tx
RdS = PA PM dii), dn^ = d\ dp,
FIG. 18.
which is constant for the whole tube. Consequently we obtain

a new proof of the fundamental property of a solenoidal vector,
for any tube may be divided up into infinitesimal tubes defined by
surfaces of the two families.
Principle of the Last Multiplier. If we have two
and N, each of which is a multiplier for the equations
functions
37.
they must each
(i),
satisfy the partial differential equation (7) so
that
dx
dy
dz
[dx
dy
dx
dy
dz
(ox
dy
oz
Multiplying the second of these by M, the
first
by N, and
subtracting,
dz
and dividing by
2
,
..
N\
2L
ox
That
is,
dz
+ Zi
oy
N
i
= 0.
dz
the quotient of the two multipliers is an integral of the

This result is of particular importance
differential equations (i).
when we have found one integral \ = const, and any multiplier, for
we may then find a last multiplier, which shall give us at once
the remaining integral.
By means of the integral equation
\ (x, y, z) = const, let us, by solving for one of the variables, say z,
express z as a function of
x, y, X,
= z (x,
y, \).
DEFINITE INTEGRALS.
72
If
IJL
= const,
is
new
of z just found, express
integral, let us
/u,
We shall
by introducing the value
= /i(#, y, X).
distinguish the partial derivatives of JJL thus expressed

derivatives when expressed in x, y, z, by brackets,
its partial
so that
we have
Accordingly we obtain for the values of A, B,

~~
Now
/ju
x,
X,
we have
^=[]^ + [|]^ +
= const,
is
an
integral,
But from the values of A and
d\
= 0,
B
_
8X'
8X'
[_ty
dz
dz
so that
da =
Bdx
,>
Adyr
OA,
But
since
this
becomes
^L
dyu-
= J/X, B = MY,
M Ydx
= r(
Accordingly although the expression
Ydx - Xdy
is
ax
being expressed in terms of
and since X
III.
in terms of x, y, X,
ji
from
[INT.
not a perfect differential, the factor
M
8X
Xdy).
DEFINITE INTEGRALS.
37, 38]
makes
it
the differential of a function

/-
fju
Ydx
//,,
Xdy)
73
and
const.
\Tz
is
a second integral of the equations
(i).
X,
and
^-
must
of
course be expressed in terms of #, y, X.
Consequently
if
we have the system
of differential equations
das:dy:ds = X: Y Z,
and we have found one integral X = const, together with a mul:
the partial differential equation
tiplier satisfying
d(MX)
(MY)
dx
then the expression
d(MZ)_
dz
dy
M
dx
dz
is
an integrating
When X,
Y,
factor, or last multiplier* for
Z satisfy
the equation
the solenoidal condition, the last multi-
plier is
1
8X'
dz
This result will be used in
103.
i
In illustrating
Variation of a Multiple Integral.
the method of the Calculus of Variations we have found the variation of a single integral, and in the example taken the functions
38.
varied were the coordinates #, y z, of points of a curve, the variable

of integration being t.
may in a similar manner vary a
t
We
volume integral, by causing the functions entering into

the integrand to change their forms by an infinitesimal transFor
formation, while the variables of integration are unchanged.
instance let
surface or
Vdxdydz
*
Jacobi, Vorlesungen ilber
Dynamik,
p. 78.
DEFINITE INTEGRALS.
74
be a volume integral, we
may
[INT. III.
define its variation
8V
by the equation
is any arbitrary function of x y, z multiplied by an infinitesimal constant e.

may also vary an integral in another
manner. Suppose we consider the volume in question to be occu-
where
We
pied by material substance, and that to each material point

Now let every material
belongs a value of the function V.
point be displaced in any manner by an infinitesimal amount
defined by the projections Sx, By, Bz.
The material point which
arrives at %, y, z brings with it a different value of F, and the
value of the integral through the same portion of space, since the
It is to
latter is filled with different material points, is different.
be noticed that this

in
is
the exact converse of the process exemplified
29, 31 for there the functions
X, F, Z were associated with

was over a field which was
fixed points in space, while the integral
varied, whereas here the function
while the field of integration

consider the integral
m=
goes with the varied point,

As an example, let us
is fixed.
pdxdydz
representing the mass of a body r whose density at any point is

limit of the ratio of the mass of
p, the density being defined as the
a portion of the body to its volume both being decreased inLet us consider the mass in an infinitesimal rectdefinitely.
}
angular parallelepiped, whose sides are dx, dy, dz, and whose
mass is dm = pdxdydz. When all points are displaced by the
amounts
Bx, Sy, Bz, particles in
the face normal to the X-axis and
nearest the origin move to the right a distance Bx, and the volume
of new matter that enters the parallelepiped through that face
is
dydzBx, whose mass
is
pdydzSx, p and Bx having the values
belonging to the face in question. At the opposite parallel face,

farthest from the origin, pBx has the value
d(pBx)
ox
and the amount of matter that moves out of the parallelepiped to

the right
is
j i
dydz \px
+ -%
v
'-
dx >
)
DEFINITE INTEGRALS.
38, 39]
The
total
through
gain
75
two sides
these
is,
accordingly,
the
difference
Similarly through the sides normal to the F-axis the gain
and through the
sides
normal to the
is
d(pSz)
~ dxdydz.
-,
The
total increase of the
and
this being
mass in the parallelepiped
therefore
taken for an element of our integral, the total
increase of mass, or variation of the integral,
We may
is
is
obtain this result in a more rigid manner by the use of

Through each element of surface dS of the
Green's Theorem.
boundary of the space in question there moves inwards an

finitesimal prism of matter whose volume is
dS [Bx cos (nx) + y cos (ny) + Sz cos

The mass of
+ py cos (ny) + p&z cos (nz)} dS,
so that the total gain of

1
(nz)}.
this is
{pSx cos (nx)
Sm =
in-
mass in the space r
{p$x cos (nx)
But by Green's Theorem
is
+ p Sy cos (ny) + pBz cos (nz)} dS.
this is equal to
This result will be of frequent use.

39.
Gauss's Theorem.
Reciprocal Distance.
the scalar point-function,
V=-
where r
is
Consider
the distance from a
DEFINITE INTEGRALS.
76
Then the
fixed point or pole 0.
parameter
[INT.
level surfaces are spheres,
III.
and the
is
and since hr
drawn toward
R=
1,
0.
16.)
Consider the surface integral of the normal component of

into the volume bounded by a closed surface S not
directed
containing 0, or as
(
we
shall call
R cos (Rn) dS = -
I!
the flux of
it,
JJi
R into S,
cos (rn) dS.
FIG. 19.
The
latter geometrical integral was

point in the boundary of an element
reduced by Gauss.
dS we draw
If to each
a radius and thus
get an infinitesimal cone with vertex 0, and call the part of the
surface of a sphere of radius r cut by this cone d%, d% is the projection of dS on the sphere, and as the normal to the sphere
in the direction of r, we have
is
d% = + dS cos
(m),
the upper sign, for r cutting in, the lower for r cutting out. If
a sphere of radius 1, whose area is 4?r, and
now we draw about
call the portion of its area cut by the above-mentioned cone day,
we have from the
similarity of the right sections of the cone
d% _ r2
da)
cZ2 = r do).
2
The
ratio d(o is called the solid angle
cone.
subtended by the infinitesimal
DEFINITE INTEGRALS.
39]
77
Accordingly
dS cos (rn) =
(2)
d^ =
do>>
77
-741
and
ff
ffdScos(rn)
JJ-T^-J/*^
(3)
Now for every element <&, where r cuts into S, there is

another equal one,
dm, where r cuts out, and the two annul
outside 8,
each other. Hence for
cos (rn)
is
If on the contrary,
~_
lies inside
<8i,
the integral
is
to be taken over
the whole of the unit sphere with the same sign, and consequently
within 8,
Hence for
gives the area 4?r.
(5)
These two results are known as Gauss's theorem, and the integral
be called Gauss's integral*.
(3)
will
These results could have been obtained as direct results of the

divergence theorem. For the tubes of the vector function R are
cones with vertex 0.
continuous in every
point within S, and since the area of any two spheres cut out by a
cone are proportional to the squares of the radii of the spheres, we
If
is
outside 8,
is
have the normal flux of
flux
is solenoidal, and the

spherical caps. Consequently
is
If
is within 8,
closed
surface
is
zero.
through any
equal for
all
between S and any sphere with center

lying entirely within S, and the flux through S is the same as the
flux through the
4?r.
sphere, which is evidently
solenoidal in the space
The fact that R is solenoidal and V harmonic may be

shown by differentiation. If the coordinates of are a, b,
(6)
= (x- a) + (y- b)' + (z- c)

2
directly
c,
2
,
Gauss, Theoria Attractions Corporum Sphaeroidicorum Ellipticorum

geneorum Methodo nova tractata. Werke, Bd. v., p. 9.
lionio-
DEFINITE INTEGRALS.
78
dr
(7)
dx
dx \rj
_3V1\
(10)
IN
/M
_x
dr
_y
'
K3
V
dz
r3
r2 dx
__!
_z
~
'
dy
dr
[INT.
3 (x
- a)
dr
82
'
v '
'
3 (x
- of - r
-S
- c) - 8r
2
and -
is
harmonic, except where r
0.
="
III.
CHAPTER
IV.
FUNCTIONS OF A COMPLEX VARIABLE.

Multiplication of Complex Numbers.
how the two-dimensional complex
number a + ib may be represented in
the plane by Argand's diagram. From
40.
We
have seen
in (5)
the definition of addition of complex

numbers it follows that two complex
numbers are added by the parallelogram construction, that is the sum of

the two complex numbers p = a : + ib-i
and q = a.2 + ib 2 is represented by the
diagonal of the parallelogram constructed on lines whose lengths
are equal to the moduli of p and q,
= V0i + &!
2
\P\
2
,
and which make angles with the X-axis equal to the arguments of
p and q.
Hence
If
we introduce the
JP
~M
polar coordinates
r=\p\
^tan-
-,
we have
a
= r cos
b = rsm
<f>,
<f>,
p = a + ib =
r (cos
</>
+ i sin
(/>).
80
Now
since
d>
it
[INT. IV.
2
<f>
<
follows that
cos
i*
It is easy to
complex numbers
the argument
show that the modulus of the product of two
equal to the product of their moduli, and that

equal to the sum of their arguments. For if
is
is
p = &! + $1 = n (cos r + i sin ^) = r^**

+ t&2 = r2 (cos $2 + * sin ) = rê***,
g=
= rjr& <*i+*) = r^ [cos (0! + ) + 1 sin (^ +
1
</>
o< 2
</> 2
then
><7
In
</> 2 )].
manner
for the quotient, substituting the words

and difference for sum. A complex number
vanishes only when its modulus vanishes, and is considered infinite
when its modulus is infinite, whatever its argument.
like
quotient for product,
Function of Complex Variable.
41.
containing
z,
function of the
+ iy,
given as an analytic expression
will be a certain function of the two real variables
complex variable z
if
x and y and will contain a real part, which we shall denote by

u (x, y) and an imaginary part, which we shall denote by iv (x, y).
}
Hence the study

to
of functions of a complex variable may be made

on
the
depend
study of functions of two real variables. Let
w f(z) = u + iv.
The representation of variable and function by means of abscissa

and ordinate of a curve is not here applicable, for both variable
and function have two degrees of freedom. The function may be
otherwise represented by means of another plane in which we
mark off lengths u and v as the rectangular coordinates of another
point representing w on another Argand's diagram. To every
point x, y in the first plane will then correspond a point u, v in the
second plane. As the point x, y moves, so will the point u, v. As
the point
x,
y representing the variable z describes any curve, u

t
v,
40
FUNCTIONS OF
42]
COMPLEX VARIABLE.
A.
81
representing w =f(z) describes another curve, if f(z) is continuous,

otherwise the point u, v may jump from one point to another.
The definition of continuity is that two points on the function
be made to approach each other as nearly as we please

the
corresponding points on the curve of the variable
by taking
near.
Or, a function is continuous in a region of the
sufficiently
s-plane continuing ZQ if to every real positive quantity e as small
curve
as
we
may
please,
we can
find a corresponding quantity 8 such that
<6
l/(*)-/(*o)|
if
|
z-z,\ <8.
In considering the representation by means of curves, it is of

if the curve of z
starting from an
importance to inquire whether,
arbitrary point ZQ returns to it after describing a closed curve, the

curve representing w =f(z) also returns to its point of departure.
If this is the case, the function f(z) within the region in which
,
be uniform, or single-valued, for to

z
one
of
value of w.
value
corresponds
every
this property holds, is said to
Analytic Function. Let us examine

an
infinitesimal change in z and the correbetween
the relation
The change dz = dx + idy has the
sponding change in f(z).
Derivative.
42.
modulus
ciz
|
= vda? + dy
and the argument
2
,
The change dw = du + idv has the modulus
and the argument
tan" 1
-jdu
w = tan"
dw
|
du
A = 9w
a*> + a-/2"
,
<
cfo
= dv
^
9w
+^
aa?
dw = du +
The
ay,
'
9a;
9y
.
idv
= du
rdx
dv j
j
^ + 9w
aV + ^ aa + =dy
dy
,
(dv
^-
*}
(dx
ratio
du
dw = du
Um
dx
\dz\=o dz
du
du
c&c
.dv
/du
'dx
W. E.
f9v
3w
dx^^dy^^ '\dx^dy
+
+ idy
idv
+ idy
.dv\dy
rfy
'
= VoV + dv*
Also
,
// ?/
1 -
82
is
dependent on
in general
-J-
that
[INT. IV.
on the direction in which
is
leave the point z. The value of the derivative will not then be
for the point z irrespective of the direction of leaving
unless the numerator is a multiple of the denominator and the
we
determined
it
dy
,.
expression containing -^ divides out.
may be
In order that this
'
'
<s
ft
du
.dv\
h i vr~
= du
dxj
\dx
-^ v
^
\dy
*\
ox]
is
^v \
^
I
dx
that
we must have
true
^
dy
dv
dy
dy'
Putting real and imaginary parts on both sides equal,
du
dw
~r
dz
and
dv
du
^~
dx
du
dv
.dv
5~
+l
= 5dv
ox
dy
du
5~
dy
dv\*
~
dz
dx
>
dx
dy
dy
In this case the function w has a definite derivative, and it is

only when the functions u and v satisfy these conditions that u+iv
is said to be an analytic function of z. This is Riemann's definition
of a function of a complex variable*.
(Cauchy says monogenic
instead of analytic.) The real functions u and v are said to be
conjugate functions of the real variables

It
is
obvious that
involving z
w =f(z)
dw
if
then
w
w
is
dw
satisfies this condition.
-~~
dw _df(z)dz _
= df(z)dz_,
Accordingly
given as an analytic expression
always
-~-
/du
dw
.dv\
s
5~^
*(5~+*5~)
dx
dx;
\dx
du
dv
dx
dy
x, y.
'
ass
o~
!=s
dy
dv
du
dx
dy'
du
5r
dy
Eiemann, Mathematische Werke,
.dv
+ *5rj
p. 5.
dy
For
42, 43]
43.
83
Orthogonal Coordinates. Conformal Representation.
We may
15 and 20 to the case of

apply the considerations of
in
a
If
a
coordinates
set
of point-functions are
plane.
orthogonal
independent of one rectangular coordinate, the geometry of all
planes perpendicular to the axis of that coordinate
is
the same,
and we have the uniplanar, or two-dimensional case involving only

two variables which we will take as x, y. If we take u and v as
any two point-functions, whose parameters are h u h v
,
idu\ z
/duV*
==
tt"
-\
\dxj
their level lines
coordinate
<5
r\
==
/dv
1
\dxj
\dyj
= constant
and v
2
1
fdv
r\
\dyj
= constant may
be taken
for
lines.
Their normals have the direction cosines

1
cos (n v x)
du
1 dv
= j-
h v dx
and the condition that u and
v shall form
du dv
du dv
dx
dy dy
fix
cos (n v y)
du
1 dv
= -j-^
h v dy
an orthogonal system
is
The lengths of infinitesimal arcs of curves, forming the sides

of a rectangle whose opposite vertices have coordinates u, v, u + du,
v + dv. are as in
20
du
hu
dv
'
hv
'
and the length of the diagonal ds, or element of length of a curve

whose ends have the above coordinates, is given by
du 2
If
now we take
functions u
and
dv 2
a?, y plane two

an analytic function of x + iy,
42 we have
for curvilinear coordinates in the
v such that
in virtue of the equations
+ iv
(A) of
du dv
dx dx
is
du dv
dy dy
Now in any orthogonal

for equal small increconstruct
a
set
of
level
curves
system
ments of u and v, they will divide the plane up into small
and u and v form an orthogonal system.

if
we
62
84
curvilinear rectangles the ratios of whose sides at

given by the ratio of the parameters h u and hv
.
equations (A),
[INT. IV.
any point are

But from the
we have
=
dw
Let
so that in this case the plane is divided into small squares.
us now construct in the second plane, in which u and v are
FIG. 21.
rectangular coordinates, the curves corresponding to u = constant

and v = constant. These are of course straight lines dividing their
plane into small squares. Moreover the length of any arc

curve in their plane, is given by
dcr
But
= du + dv\
2
in virtue of the above relations, this gives

dcr
dcr of
is
-j
question,
= h ds
2
2
,
accordingly the ratio of magnification at the point in
and varies
for different points of the plane.
Let us now construct, (Fig.

infinitesimal triangle
and
21,) at a point in the x, y plane an

the intersection of any three curves,
sides be ds1} ds 2 ds 3
Construct the
made by
the lengths of its

corresponding curves in the
let
u, v plane, intersecting to form an

infinitesimal triangle with sides
Now we
have
43, 44]
85
and therefore
d^
d(T z
da 3
c^
and the infinitesimal triangles are
ds 2 dss
:
similar.
sponding curves intersect each other in the

corresponding planes. Such a relation as this
Consequently correin both
same angle
is called a
Conformal
is
of
fundamental
and
it
Relation*,
importance in the theory of
The two planes are said
functions and in mathematical physics.
The relation is
to be conformal representations of each other.
sometimes specified by saying that the conformal representations
are similar in their infinitely small parts.

It is easy to
show that
if
the functions u and v give a conformal

must satisfy the equations (A).
representation of the plane, they

44.
Conjugate Functions.
Laplace's Equation.
differentiate the equations (A), the first
3%
and add, since
= * &v5~ we
or
is
have
satisfies Laplace's
equation in two variables,
harmonic.
Differentiating the other

satisfies
of
we
oyox
cxcy
so that the function
If
by x and the second by y
way and adding we show that
v also
the same equation
Thus every conformal development or every analytic function

a complex variable gives us two harmonic functions. The
It obviously will
question arises whether the converse is true.
not do to take any two harmonic functions for u and v, for they
must be related
function
is
given,
But if one
so as to satisfy the equations (A).
we may find the conjugate, for we must have
dv =
,
which by the
first
dv j
-
dx
equation (A )
dv
dx
dv j
dy,
+ 5-
is
du
^-
7
dx
+
dy
*
'
dy
Professor Cayley has called
du j
^- dy.
dx
'
it
orthomorphosis.
86
Now
if
we
call this
Xdx 4- Ydy
[INT. IV.
the condition for a
it satisfies
perfect differential
aX _ 8 Y
dy~dx'
&u_&u
fy
to a variable point x,
a?
limit,
and represents
2/0
fdu dx du ,
+ dy
3J ox
dy
is
its
upper
given,
[dv
-,
from a given
a function only of
is
Similarly if v
v.
'
+ ^-dy
\^-dx
dx
oy
Consequently the line integral

point
~8^
dv
^-dx
dx
dy
dy.
'
Furthermore the
+ udy
that vdx
udx
vdy
is
first of the equations (A) is the condition

a perfect differential, and the second that
is
such.
Accordingly the line integrals
$=
new
+ udy,
= udx
vdy,
-Jr
give two
vdx
point-functions
</>,
which in virtue of the equations
^r
_ d(f> _ _ 9-v/r
3</>
_ dty
'
dx
dy
dx
dy
'
are conjugate to each other, and give a new analytic function of z

+ iXjr, or T|T + i<^. From these by new integrations we may obtain
<f>
t
any number.
Examples.
The function
z*-
= (x + iy)
u=x
gives
^2 4y
2
,
both harmonic functions.
The curves u =
o?
const, and v
y*
2xy const, give two
of
which
intersect
sets
everywhere at right
equilateral hyperbolas,
angles, Fig. 22.
The
X
~~
-x+
function
x*
ly
Rives
=
*
9
*
+y
--~y
x 2 + y ;2
-
44]
87
FIG. 22.
The curves
u=
const,
and
= --
ni
,
x*
x*
-f
f - CjX = 0,
a?
+f+
+ y'
= const.,
%=
0,
give two sets of circles, the first all tangent to the F-axis at
the origin, the second all tangent to the X-axis, Fig. 23.
The power
zn
= (x 4- iy)n = rn
{cos
n$
+ i sin n<j>},
gives the two functions
= rn cos
n<f),
= rn sin n<f>,
and a sum of any number of such terms each multiplied by a constant
2rn [A n cos n<f> + Bn sin n<f>},

therefore gives a harmonic function.
If a function can be developed
in such a trigonometric series it accordingly is harmonic.

such as these may be called circular harmonic functions.
Terms
88
[INT. iv.
FIG. 23.
Integral of a Function of a Complex Variable.

Since the complex variable has two degrees of freedom, its integral
45.
is
not of so simple a nature as that of a single real variable.
Suppose the variable z moves from a point A to a point B along

any continuous path. The definite integral of f(z) = u + iv along
this path will be denned as the line integral
rB
rB
f(g)
dz=\J A
+ iv) (das + idy)
(u
8
.
udx
vdy
r-
+ udy.
4- i
J
Now in virtue of the equations (A) both the integrals above

It is
is a function of z.
are independent of the path, so that
is on the supposition
This
section.
of
the
last
+
evidently -v/r
i<f)
that the functions u, v are uniform and continuous in the whole

If this is the case the function w = u + iv is
region considered.
called holomorphic.
If
becomes discontinuous
between which
lies
in the region considered it ceases
the same over two paths AB

a point of discontinuity of the function w.
to be true that the integral
is
45]
For example the function

2
= 0.
w=-
Accordingly the integral
is
89
discontinuous at the point
around a closed contour con-
taining the origin within it is not zero, for it may be taken as the
between the integrals between two points
on the
contour along two paths between which lies the point of dis-
AB
difference
continuity of the function w.
The
integral around any closed
contour embracing the origin is however the same as around a

with center at the origin, for between the two
circle of radius
curves there
since z
is
Now
no point of discontinuity of the function.

if r is constant = R,
= x + iy = re,
and the integral from z
de
which taken around the

/*
The
integral
circle is 2iri.
is
defined as the logarithm of
sesses the property that as z describes
z,
and
it
pos-
any closed path enclosing the
function instead of returning to its original value

increases by a constant 2iri.
The function is then not uniform,
origin, the
but has at any point an unlimited number of values, depending

upon the path by which we arrive at the point. These values all
differ
by
We
integral multiples of the constant 27n.
see that this accords with the ordinary definition of the
logarithm,
log z
log (x
+ iy) = log (rei(* +2nn] ) = log r + ty + 2mri,
for if we increase the

argument </> of a complex number z by any
multiple of 2?r, the number is unchanged. A point such that a
function f(z) assumes a new value when the variable traverses a
closed circuit about the point is called a critical, or branch point

In this case the conformal representation given by the function/^)
is
multiple in character, for in the J7F-plane we are to take a point
for each of the values of the function
Each of these repref(z).
sentative points gives a conformal representation of the whole of
the JTF-plane on a part of the 7F-plane.
90
[iNT. IV.
For instance, in the case of the logarithm
= log r + i
as z takes all possible values in the JTF-plane, u

to + oo but v varies only in limits differing
log r varies from

2?r, so that the
by
whole .STF-plane is entirely represented on a strip of the UV- plane

in one direction but of the finite width 2?r in the
This strip is repeated an infinite number of times each
other.
infinite
giving the same conformal representation of the whole JTF-plane.

For instance the radii </> = const, and the circles r = const, cutting
them orthogonally
v
= const,
= const,
in
the XF-plane correspond to
in the
the
7F-plane.
FIG. 24.
Corresponding regions of the figures are similarly shaded.
lines
PART
I.
THEORY OF NEWTONIAN FORCES.
CHAPTER
PRINCIPLES OF MECHANICS.
46.
science
Matter and Energy.

of Matter
invariable
I.
UNITS AND DIMENSIONS.

Physics is the
laws are found to be
Dynamics.
and of Energy.
Its
and capable of exact statement, that
is
of presentation
in the language of Mathematics. The application of mathematical

analysis to the treatment of physical phenomena, enabling us to
deduce general laws from the results of experiment, and to infer

the consequences of general laws, forms the subject of Mathematical or Theoretical Physics.
Matter has the essential property of occupying space. It has

in addition
universally only the property of Inertia, to be denned
below. In order to define Energy, we must consider the motion of
matter in space. That portion of mathematical physics which
treats of the motion of matter is called Mechanics, or Dynamics.
It is the object of physicists
to reduce the explanation of all
and
descriptions of motion of matter,
indisis
accordingly the study of the principles of Dynamics
pensable to the study of any branch of theoretical physics. Before
physical
phenomena
to
considering the nature of electrical and magnetic

shall therefore devote a few
chapters to Dynamics.
47.
are of
Scalar and Vector Quantities.

two kinds. Quantities whose complete
phenomena we
Physical quantities
specification involves
92
[PT.
I.
CH.
I.
no idea of direction are called scalar quantities, for they may be

conceived as arranged on. a scale according to their magnitude.
Such are time, temperature,
size, density.
Quantities whose specification involves the idea of direction as

well as of magnitude are called vector quantities.
They may be
represented by geometrical directed lines, and all that has been

said of vector quantities and their addition, etc. applies to them.
48.
Degrees of Freedom. A set of magnitudes or parameters which completely specify a quantity are called its coordinates.
The number of coordinates required is called the
number
of degrees of freedom of the quantity.

For instance,
a point in a plane may be defined by two rectangular, or two polar
We may also say
coordinates, and has two degrees of freedom.
that there
a double infinity or oo 2 of points in a plane.

point
in space requires three coordinates of any sort, and has three
degrees of freedom.
Every independent relation that the coordiis
nates of a quantity are
made
to satisfy diminishes the
number
of
degrees of freedom by one. For instance, a relation between

the rectangular coordinates of a point restricts it to lie on a
certain surface,
it then has two
degrees of freedom instead of
its
three,
and requires but two coordinates
a point
satisfies
sphere of radius
the condition s?
a,
and may be
to specify
+ y* -f z
2
-
=a
fully specified
it.
2
.
For example,
It lies on the
by giving
its lati-
tude and longitude.
For the coordinates of a vector
R we
may
on the three coordinate axes. If we choose

and its three direction cosines,
a
= cos (Rx),
ft
= cos (Ry),
one of the four coordinates R,
a, ft,
is
its
take
its
projections
length, or modulus,
= cos (Rz),
redundant, for the latter
three satisfy the identical relation
This furnishes us an example of the general case where we

give n coordinates of a quantity satisfying k independent identical
The quantity then has only
relations, or equations of condition.
k degrees of freedom, and we

coordinates which completely specify
may
it.
find n
k independent
49] PRINCIPLES OF MECHANICS.
47
93
If a point change its position in

space, its
described
be
the
values
of
its
coordinates
by giving
may
for every instant of time, by means of equations such as
49.
Velocities.
motion
*=/i(0, y =/(),
The
functions
jump from one
= 0,
(x, y, z)
=/.()
must be continuous,
since the point cannot
position to another.
We may describe the
Ft (x, y, z)
motion otherwise by giving two equations

which denote the curve of intersection
= 0,
two surfaces along which the point moves. This curve is called
the path of the point.
must further give the distance s
measured along the curve, which the point has traversed, counting
of
We
from a fixed point on the curve. We must know s at all times t,

which is expressed by giving s as a continuous function of t,
s=
This, with the two equations of the path, gives as before
(t).
(j)
three equations to completely define the motion.
The
velocity of the point is defined as the limit of the ratio of

the distance As traversed in an interval of time A to the time
when both decrease without
limit,
ds
= TLim As = -j.
A f =0
AZ
dt
A point travelling with a given numerical velocity may however be moving in any of an indefinite number of directions,
accordingly a velocity is completely specified only when we give its
direction
and magnitude, or
velocities are vector quantities.
direction of the velocity is that of the tangent to its path.

direction cosines are accordingly
The
Its
'
a
Velocities
genera],
are
are
= dx
Ts'
n dy
^ = d^' t =
resolved
and
dz
ds-
compounded
in particular the projections of v
Vx
VOL
ds dx
= V dx
j- = -T -jds
dtds
dx
-r.
dt
Vy
ds dy
dy
= -T
= V/3 = V dy
-f--f = J ,
vz
=vy =v
ds
dz
-T-
ds
dt ds
dt
= ds dz
= dz
-j-j-.
dt ds
like
vectors
in
on the coordinate axes
-j-.
dt
94
We
[PT.
I.
CH.
I.
might have defined the vector velocity as the resultant of
the three vectors

_
_dx
50.
_dy
y
dt'
di
_dz
dt'
Accelerations.
with the time
we
If the velocity of a point is variable

define the acceleration of the point as the limit
of the ratio of the increment of velocity Ay to the increment of

time A, as both approach zero.
may consider either the
We
numerical change
w
or the geometrical change.
If
ds
2
dv
we draw
a vector
AB
to represent
AC
to represent the
the velocity at the time t and the vector
time
at
the
t + A, and draw the arc of a circle BD,
velocity
DC
will represent the
numerical change of
its direction,
velocity, Av, not considering
DX /
'B
BG represents
vector change, Av, for
while
its
geometrical, or
AB + BG = AC,
is
the vector acceleration
Lim -
Accordingly
FIG. 25.
= Lim BC
a.
Since the projections of the geometrical difference of two

vectors are the differences of the projections, the components of a
direction will be proportional to the changes of the correin
any
sponding components of the
velocities, that is
dv x
d>r.
_ dx
=
2
~di
~di?
dt2
a?
TT
'
51] PRINCIPLES
49
The vector
OF MECHANICS.
acceleration, a,
may be
defined as the resultant of
the components ax a y> az and accordingly

,
d~s
This
is
not in general equal to
the scalar velocity.
The
95
modulus
its
which
is
the acceleration of
is
-^-
direction of
is
given by
its
direction
cosines
d?x
51.
Physical Axioms.
with regard to motion are

of Motion or
Axioms
d'
The
d?
results of universal experience

his three Laws
summed up by Newton in
of Physics.
An axiom
is
defined by
Thomson
and Tait* as a proposition, the truth of which must be admitted

as soon as the terms in which it is expressed are clearly understood.
These physical axioms rest, not on intuitive perception,
but on convictions drawn from observation and experiment.
LEX I. Corpus omne perseverare in stata suo quiescendi vel

movendi uniformiter in directum, nisi qicatenus a viribus impressis
cogitur statum suum mutare.
Every body persists in
uniform motion in
be compelled by force to
its state of rest or of
a straight line, except in so far as

change that state.
it
may
The property of persistence thus defined
is
called Inertia.
This gives a criterion for finding whether a force is acting on

a body or not, or in other words a negative definition of force.
Force is acting on a body when its motion is not uniform. By
uniform we mean such motion that the vector velocity is constant.
If the body be a material point, that is a body so small that the
distances apart of its different parts may be neglected, the motion
is
uniform
if
dx
=c>Tt
that
dy
=c*>
it
dz
di
=c"
is
Thomson and
Tait, Natural Philosophy,
243.
96
[PT.
I.
CH.
I.
Accordingly we see that the force and acceleration vanish together.

Integrating the equations (i),
the path
is
a straight
line,
and since
traversed with constant velocity.

interpret the statement as giving us a
it is
We may on
the other hand
means
of measuring time.
Intervals of time are proportional to the corresponding distances
traversed by a point not acted on by forces.
The second law
LEX
II.
impressae,
et
gives the measure of a force.
Mutationem motus proportionalem esse vi motrici

ilia imfieri secundum lineam rectam qua vis
primitur.
of motion is proportional to force applied, and takes
in
the
direction of the straight line in which the force
place
Change
acts.
By change of motion is meant acceleration. If we have to

do with different bodies, however, the factor of proportionality
will be different for each.
LEX
Actioni contrariam semper et aequalem esse reactionem: sive corporum duorum actiones in se mutuo semper esse
aequales
III.
et
in paries contrarias dirigi.
To every
or,
action there
is
always an equal and contrary reaction
the mutual actions of any two bodies are always equal and
oppositely directed.
If
we have an
action between two bodies 1 and
were proportional only
to the accelerations,
This
we must introduce
is
so that
not the case, but
2, if
the forces
we should have
factors of proportionality,
51, 52]
The
factors m^, ra 2
are called the masses of the bodies 1 and
97
2.
This gives us a means of comparing masses. If we make two

bodies act upon each other in any manner, their masses are
inversely proportional to the accelerations they have at the same
The vector whose components
instant.
d?x
Xv = ro _,
are
z
Tr
= dz
Y=
m d*y
^, Z m ^
on the mass m. If the quantities

are .given functions, the above are called the differential
equations of motion of the material point m.
is
called the impressed force acting
X,
Y,
52.
Units.
The
involves two factors,
specification of
any quantity,
scalar or vector,
a numerical quantity (integer, fraction

or irrational) or numeric, and secondly a concrete quantity in terms
of which all quantities of that kind are numerically expressed,
first
a unit.
The simplest unit is that of the geometrical
shall adopt as the unit of length the centiquantity, length.
meter, defined as the one-hundredth part of the distance at temperature zero degrees Centigrade, and pressure 760 millimeters
called
We
of mercury, between two parallel lines engraved on a certain bar

of platinum-iridium alloy, deposited in a vault in the laboratory of
the " Comitd International des Poids et Mesures" at Sevres, near
This bar is known as the "Metre Prototype," and serves as

Paris.
the basis of length measurements for the civilized world (except
the British Empire and Russia*).
was proposed by Maxwell to use a natural unit of length,

namely the length of a wave of light corresponding to some welldefined line in the spectrum of some element, at a definite temperature and pressure, as it is extremely probable that such a
wave-length is extremely constant. Measurements were carried
out at Sevres by Michelson, with this end in view, which established
It
the ratio between the above meter and the wave-length in air of
a red cadmium ray as l,553,164f.
*
The United States yard is defined as 3600/3937 metres,

t Michelson, Journal de Physique, Jan. 1894.
W.
E.
98
The unit
of mass will be
assumed
to be the
1.
[PT.
CH.
I.
gram, defined as
the one-thousandth part of a piece of platinum-iridium, deposited

at the place above mentioned and known as the "Kilogramme
Prototype"
As the
unit of time
we
shall
take the
mean
solar second,
obtained from astronomical observations on the rotation of the

earth.
of time cannot be preserved and compared as in

of the units of length and mass, but is fortunately
The unit
the case
preserved for us by nature, in the nearly constant rotation of the

As the earth is gradually rotating more slowly, however,
earth.
this unit is not absolutely constant, and it has been proposed to
take for the unit of time the period of vibration of a molecule of

the substance giving off light of the standard wave-length. To
obtain such a unit would involve a measurement of the velocity
of light, which1 cannot at present be made with sufficient accuracy
to warrant the change.
53.
Dimensions.
Derived Units.
acquainted
may
It
can be shown that
physical quantities with which we are

be made in terms of three independent units.
the measurements of
all
and are most conveniently

taken as those of length, mass, and time. Other units, which
depend on these, are known as derived units. If the same quantity
is expressed in terms of two different units of the same kind, the
These are known as fundamental
units,
numerics are inversely proportional to the size of the units. Thus

six feet is otherwise expressed as two yards, the numerics 6 and 2
the
being in the ratio 3, that of a yard to a foot. If we change
in
ratio
units
r, the
any
magnitude of one of the fundamental
numeric of a quantity expressed in derived units will vary proof r, r^ the derived unit is then
portionately to a certain power
said to be of dimensions* n in the fundamental unit in question.
For instance, if we change the fundamental unit of length from the
;
foot to the yard, r
3,
an area of 27
sq.
ft.
becomes expressed as
27 = 1 3 2 = r~ 2
3 sq. yds., the numeric has changed
and the unit of area is of dimensions 2 in the unit of length.
in the ratio 3
We
may
express this by writing
[Area]
*
The
idea of dimensions of units originated with Fourier:
analytique de la Chaleur, Section ix.
vid.
TMorie,
52, 53]
The derived
99
same ratio that the numeric of

the quantity decreases. In our system the unit of area is the
2
In like manner the unit of volume
square centimeter, written 1 cm
3
is of the dimensions [Z ] and the unit is 1cm 3
The dimensions
unit increases in the
of velocity are
we
or as
velocity
write for convenience,
= length /time.
Two
quantities of different sorts do not have a ratio in the

arithmetical
sense, but such equations as the above are of
ordinary
great use in physics, and give rise to an extended meaning of the
terms ratio and product.
The above equation
to be interpreted as follows.
is
If
any
velocity be specified in terms of units of length and time the

numerical factor is greater in proportion directly as the unit of
length is smaller, and as the unit of time is greater. For instance
we may write the equation expressing the fact that a velocity

of 30 feet per second is the same as a velocity of 10
yards per
second or 1800 feet per minute
30^=10^ = 1800-^.
mm.
sec.
We may
sec.
operate on such equations precisely as
if
the units were
ordinary arithmetical quantities, for the ratio of two quantities of

the same kind is always a number.
For instance
30
10
The
ratio
vci
*~
= yd.
~
ft.
sec.
*
sec.
sf*r*
is
number
the
3,
while
= 1.
Also
sec.
ft.
1800_yd.min
10 "ftT^cT"
as -
Such an expression
The unit
of velocity
'-
is
is
read feet per second.
one centimeter-per-second,
cm.
= cm.
sec" 1
sec.
Since acceleration
is
defined as a ratio of increment of velocity to
increment of time, we have

,.
[Accelerate]
L_ZJ
L-JOJ = FZl
= [Velocity] = [Length]
|^J
72
100
[PT.
I.
CH.
1.
numeric of a certain acceleration varies inversely as the

magnitude of the unit of length, and directly as the square of the
For instance, an acceleration in which a velocity of
unit of time.
10 feet per second is gained in 2 seconds is equal to one in which
or the
a velocity of 9000 feet per minute
10
10
4
ft.
2
(2 sec.)
The
unit of acceleration
Since force
= mass
is
ft.
ft.
sec.
is
is
min. 2
one centimeter-per-second per second.
Pass]. [Length]
2
[(Time)
of force
90QO
x acceleration,
[Fence!
The unit
gained in a minute,
is
= p/
IT*
one gram-centimeter-per-second-per-second. It
called a dyne.
All physical equations must be homogeneous in the various

units, that is, the dimensions of every term must be the same.
This gives us a valuable check on the correctness of our equations.
The above system of units, which

fundamental units the centimeter, gram, and second, is
called the C. G. s. system, and was recommended by a committee of
54.
has for
Absolute Systems.
its
the British Association for the Advancement of Science in 1861.

It is sometimes incorrectly spoken of as the absolute system of
An
absolute system is any system, irrespective of the

magnitudes of the units, by which physical quantities can be
number of fundamental units, which
specified in terms of the least
units.
be independent of time or place, and reproducible by copying

from standards. A system based on the foot, pound, and minute
The idea
is just as much an absolute system as the C.G.S. system.
shall
of
an absolute system
is
due
to
Gauss *.
The ordinary method
of measuring force, used by non-scientific

(or including) engineers, does not belong to the absoThe unit of force is taken as the
lute system of measurements.
force
exerted
downward
by the earth upon, the
weight of, or
persons and
mass of a standard piece of metal, such as the standard pound or

kilogram. To measure the force in absolute units, we must know
*
Gauss.
Intensitas vis magneticae terrestris
Gottingen, 1832.
Ges. Werke,
v. p. 80.
ad mensuram absolutam revocata*
53, 54]
what acceleration the

if
receive,
allowed to
would cause this mass to

Experiment shows that in a given
earth's
fall.
101
pull
locality on the earth's surface all bodies fall in vacuo with the
same acceleration. The value of this acceleration is denoted by g,
and
its
value at the sea-level in latitude 45
is
f*TYI
980-606
sec.
Accordingly the force exerted by the earth on a mass of

is mg dynes, or the
weight of a kilogram in latitude 45
grams
980,606 dynes.
Now
as
we
the value of the acceleration g is not constant, but varies

go from place to place on the earth's surface, ascend moun-
tains or descend into mines.
Accordingly, the weight of a kilogram

not an invariable, or absolute standard of force. At the center
of the earth, a kilogram would weigh nothing.
Its mass is, how-
is
The value of g at points on the earth in laticentimeters above the sea-level, is given by the
ever, invariable.
tude
X and h
formula, originally given
g=
by Clairaut*,
- 2*508 cos 2\ -
980-6056
For further information with regard

consult Everett's Units
*
and Physical
Everett, Units
'000003/i.
to units, the reader
Constants.
and Physical Constants, Chap.
in.
may
CHAPTER
II.
WOKK AND ENERGY.

Work. If a point be displaced in a straight line, under
55.
the action of a force which is constant in magnitude and direction,
the product of the length of the displacement and the resolved
part of the force in the direction of the displacement, that is, the
geometrical product of the force and the displacement ( 7), is called

the work done by the force in producing the displacement. If the
components of the
ment
s are sx , sy
force
sz
F are
X,
the work
and those of the displace-
F, Z,
is
W = sFcos (Fs) = Fs = Xs
(i)
Since work
is
denned as
force
Ysy
+ Zsz
x distance, we have
for
its
dimensions,
[Work]
[L]
\^=
2
[I/Z ?
-2
].
The C.G.S. unit of work is the work done when a force of

one dyne produces a displacement of one centimeter in its own
This unit is called the erg = gm cm 2 sec~ 2
direction.
.
If the displacement be not in a straight line, and the force be

not constant, the work done in an infinitesimal displacement ds is
(2)
ds
and the work done
WAB
TIT
ds
in a displacement along
line integral
(3)
ds
any path
AB
is
the
WORK AND ENERGY.
55, 56]
The components
103
of the force are supposed to be given as
functions of s and the derivatives
-j-
ds
-^-
ds
.-
known
are
ds
as func-
tions of s from the equations of the path.
Understanding
this,
we may
WAB = (
(4)
write
Xdx + Ydy + Zdz.
Virtual Work.
56.
material points.
coordinates for
Suppose that we have a system of n

If they are entirely free to move, they require 3n
their specification.
They may be mechanically
manner that there must be
constrained, however, in such a

relations satisfied
by
certain
Let these equations of
their coordinates.
condition or constraint be
 2
Oi,
#2,
2/2>
*2,
#2,
2/2,
3s,
n, Un> 2n)
0,
0.
(5)
#n,
2/7i,
Zn)
Such constraints may be imposed by causing the
particles to
two particles 1 and 2 are

either
connected by a rigid rod of length I,
particle must move on
a sphere of radius I of which the other is the center, and we have
lie
on certain
surfaces.
For instance,
the equation of condition

* =
+
</>
x$
(x,
(2/1
if
- ytf + (*i -
^-
constraints defined by inequalities, e.g., if a

of
were
particle
obliged to stay on or within a spherical surface
radius I the constraint would be only from without, and we should
have
- a y* + (y- 6) 2 + (z- cf - V ^ 0.
(x
(We might have
We
is
assume that the constraint

defined by an equation.)
If
that
shall
any particle at xrt y r z r
it
is
is
toward both
displaced by a small
sides,
and
amount
so
has the coordinates
xr + xr yr
,
*
The sign = is to be read
variabiles, or is defined as.
is
+ 8yrt
identically
zr
i.e., is
Szr>
for all possible values of the
104
in order that the constraint

(6
may
[FT.
I.
CH.
hold we must have for each
II.
<
<
and
if
$ be a
continuous function, developing by Taylor's Theorem,
and accordingly, taking account only of the terms of the first order
in the small quantities 8xr $yr &zr and using equations (6), we
,
have
number
If a
of particles are displaced,
of expressions like the above for
T2 ]^f
(8)
r =i
a</>
*
Sx
r
a</>
+ -^
,
9y r
(d#r
we must take
all
the particles, or
+ 30
Sy r
^
d*r
* 1
$z
r\
the
sum
= 0,
must be satisfied by all the displacements

There must be one such equation for each function
Such displacements, which are purely arbitrary, except that
as the conditions which
&cr Sy r Szr
,
<f).
they satisfy the equations of condition, are called virtual, being

possible, as opposed to the displacements that actually take place
in a motion of the system.
The Principle of Virtual Work is an analytical statement of the

conditions for equilibrium of a system.
system is in equilibrium
when the
on
various particles, together with the

balance
each
other
in such a way that there is no
constraints,
forces acting
its
tendency toward motion of any part of the system. If the system

consists of a single free point, in order for it to be in
equilibrium,
the resultant of all the forces applied to it, whose components
are X, Y, Z, must vanish,
X=Y = Z=0.
(9)
we multiply these equations respectively by the arbitrary

small quantities Sx, By, $z and add, we get
If
(10)
XBx + YSy + ZSz = 0,
which expresses that the work done in an infinitesimal displacement of a point from its position of equilibrium vanishes. The
WORK AND ENERGY.
56]
is
equation (10)
equivalent to the equation
quantities Bx, By, Bz, are arbitrary, if

zero,
X,
we may take
Y, Z,
vanish.
X,
<f>
Y,
105
X, Y,
(9),
Bz respectively of the
Bx, By,
the
for since
Z, are different from
same sign
a>s
each product will then be positive, and the sum will not
If the sum is to vanish for all possible choices of Bx, By, Bz,
Z must
vanish.
If the particle is not free, but constrained to lie on a surface

Bx, By, Bz are not entirely arbitrary, but must satisfy
= 0,
(7)
dx
'
dz
dy
Let us multiply this by a quantity X and add
it
to (10),
obtaining
(II)
We may no
must vanish,
longer conclude that the coefficients of Bx, By, Bz
for Bx, By, Bz are not arbitrary,
being connected by
the equation (7).

Two of them are however arbitrary, say By and
Bz, X has not yet been fixed
suppose it determined so that
Then we have
in
(V+ X
By
+ (z + X
|)
which By and Bz are perfectly arbitrary,
it
Bz
= 0,
therefore follows of
necessity that the coefficients vanish.
dy
By the introduction of the multiplier X we are accordingly

enabled to draw the same conclusion as if Bx, By, Bz were arbitrary.
Eliminating
X from
the above equations we get
dx
Now
the direction cosines of the normal to the surface
are proportional to -^
ox
X,
Y,
dz
Z being
-^
oy
oz
=
<f>
consequently, the components
the resultant
proportional to these direction cosines,
106
is
in the direction of the
conditions the particle
is
In like manner we
may show
,
2) Xj&cj
Fjfyi
But under these
normal to the surface.
2
particle 1,
condition of equilibrium is
(1
2,
that
2
the forces
if
upon the
lt
l}
...... 4-
X T Z
lt
act
particle 2, etc., the
+ Z^z, + Xfe + F % + Zjbz*
where the displacements
II.
in equilibrium.
X Y Z
upon the
CH.
I.
[PT.
Zn &zn = 0,
satisfy
(13)
k- *
3^
O^P!
+
,
f
8y
fc
r.
fc
*
%
-~
OVi 4- -f- CZ 1 4- -f- bx.2 4.
'
dzi
dxz
%
OV
2
. . .
'
dy 2
Multiplying the equations (13) respectively by X 1}
adding to (12)
=*+ dz
= n0.
X2
A*,
and
we have
3^-A; are arbitrary,

multipliers X so that the coeffithen the coefficients of the 3n k
Of the 3w quantities 5^l5

we may however determine the k
S^w only
,
k other S's vanish,

must
$s
vanish, so that we get the 3n equations
arbitrary
cients of the
*
02rft
WORK AND ENERGY.
56, 57]
107
(15)
from these the k quantities X, we have 3n k

.
Eliminating
AnnaH
equations expressing the conditions of equilibrium, being as many
The equation (12), or as
as the system has degrees of freedom.
we may
write
it
3,(XSx+7Sy + ZSz) = 0,
(i 6)
expresses the fact that the work in a virtual displacement vanishes,

and is the condition for equilibrium. This is the Principle of
Virtual Work.
57.
D'Alembert's Principle.
point are
The equations
of motion of a
(51)
r= Y
(17)
(18)
the arbitrary
Multiplying the equations (18) respectively by
all values of
for
sum
the
quantities Sxr Syr Szr adding, and taking
,
the suffix
r,
108
[PT.
I.
CH.
II.
may be called the fundamental equation of

the analytical statement of what is known as
d'Alembert's Principle.
Lagrange made it the basis of the entire
This equation
dynamics, and
is
Interpreted by means of the principle of
subject of dynamics*.
virtual work, equation
(1
9) states
If, the motion of a system of particles being given, we find the

acceleration of every particle, and apply to each particle a force
whose components are
dx
then the system of forces X', Y, Z', together with the impressed
forces X, Y, Z, will form a system in equilibrium.
The
forces
reversed
X,
Y', Z'
are called the forces of inertia, or the
D'Alembert's principle
effective forces.
is
thus only
another form of stating Newton's third law of motion.
We
have now a measure of the inertia of a body, namely the

above defined^. We may now define matter as
force of inertia
whatever can exert forces of
inertia.
Conservative Systems. If in the equation of

58. Energy.
d'Alembert's principle, (19), we put for &e, By, Bz the displacements
which take place in the actual motion of the system in the time dt,
we
obtain
dXr
-
d*Z r dzr
+ dfyrdyr +
v
-Xr-Y-Zrd
dxr
__ =
d xr dxr
2
S lnc e
the
sum
mr
of the first three terms

!
(fdxr \
JS,m, ](-=-}
+
,
dijr
7 dzr \
_d
is
the derivative of the
(dyr \
-f-
r
(dz
-,- \\
sum
Lagrange, Mecanique Analytique. (Euvres, t. 11, p. 267.

t The inertia of a body is sometimes considered as the factor of the negative
acceleration in
the
expression
synonymous with mass.
for
the force of
inertia,
thus making inertia
WORK AND ENERGY.
57, 58]
109
and the equation may be written
-?
~M Y
a.
3*~
Integrating with respect to
Vr
t
7r
-,.
î
between the limits
tQ
and
t lt
<
The square brackets with the

of the expression in brackets for
value for
The
affixes
t
=t
is
to
^ denote that the value

be subtracted from the
^.
integral on the right of (22), which

l
r da;r
may be
+ Yr dyr + Zr dzr
written
denotes the work done by the forces of the system on the particle
r during the motion from tQ to 1} and the sum of such integrals
denotes the total work done by the forces acting on the system
during the motion.
The expression
the half-sum of the products of the mass of each particle by the

square of its velocity, is called the Kinetic Energy of the system.
If
we denote
it
T, the equation (22)
T - Tu = 2 r
(23)
This
by
tl
is
j\X
dx r +
becomes
Yr dyr + Zr dzT
).
called the equation of energy, and states that the gain of

is
equal to the work done by the forces during the
kinetic energy
motion.
The equation
of energy assumes an important form in the
particular case that the forces acting on the particles depend only
on the positions of the particles, and that the components may be
110
[PT.
I.
CH.
II
represented by the partial derivatives of a single function of the

coordinates
zlt
U(xl) ylt
dU
#2
2/ 2
z.^...zn )
dU
dU
In this case the expression
2r
is
{Xrdasr
+ Yr dyr + Zr dzr
dU
}
dU
the exact differential of the function U, and the integral
P2 (X
The equation
The
dxr
+ Yr dyr + Zr dzr ) = U - U
tQ
tl
of energy then
function
is
called
is
the force-function, and
its
W = U called the Potential Energy of the system.

W in (25) we have
T +W =T,,+ W
(26)
is
tl
tl
negative
Inserting
t,.
The sum
of the kinetic and potential energies of a system

a
force-function is the same at all instants of time.
possessing
This is the principle of Conservation of Energy.
Systems
for
which the conditions (24) are
satisfied are accord-
ingly called conservative systems.
The
an
potential energy, being defined by its derivatives, contains

Conservative systems possess the property,
arbitrary constant.
T+
is constant,
depends only on the coordinates, and
that T, the kinetic energy, depends only on the coordinates, or if
in the course of the motion all the points of the system pass
simultaneously through positions that they have before occupied,
since
the kinetic energy will be the same as at the previous instant,

in which the points may be moving.
irrespective of the directions
thrown
a
For instance, particle
vertically upwards, or a pendulum
swinging, have the same velocity when passing a given point
whether rising or falling.
56, may evidently be expressed
the
for
that
potential energy of the system
equilibrium
by saying
The
principle of virtual work,
WORK AND ENERGY.
58, 59]
is
maximum
or
Ill
minimum, and a little consideration shows that

a minimum.
for stable equilibrium it is
Examples of non-conservative systems are found whenever the

forces depend upon the velocities as well as upon the coordinates
for example, bodies moving through the air or other resisting
medium or bodies whose motion is opposed by friction of any sort,
form non-conservative systems. Even if the friction be constant

in magnitude, its direction will depend on the direction of the
velocities, being in such a direction as always to oppose the
motion, and to diminish the total energy of the system. The
dynamical theory of heat accounts for the energy that apparently
disappears in non-conservative systems.
Kinetic energy being denned as
-^-
the same as those of work.
The
as work.
unit of energy
is,
S^mv
is
of the dimensions
Potential energy
is
denned
therefore, the erg.
Newtonian
Particular case of Force-function.
59.
Forces. In the particular case in which the only forces acting
on the system are attractions or repulsions by the several particles
directed along the lines joining them and depending only on their
mutual distances, a force-function always
For
let
exists.
the force between two particles
and
at a distance
apart rrs be
be convenient to consider
positive -if the force is a repulsion.
It will
Consider
now
the force
(r)
acting
m and acting in the direction

Its direction cosines are
from m to m
on
Fm
26
'
s.
(r)
v ~
_Us~
those of the vector rr8)
112
Now
CH.
II.
the forces acting on
[PT.
I.
since
= (x - xr f + ( y. - yrj + (z - zr }\
rls
by xs
differentiating partially
*ig= 2 <"'-*>'
/
arrg
2 g)
=x
- xr
drrs
'
rn
fas
dy
= y -yr
drrs
rr8
^z -z r
8
rrs
dzs
and accordingly
X^__drr_
" s
If
we put
r8
Y^_drr,
M
g
'
F.
da.
Z.V ~
_ drra
FM
'
dy
dz8
'
such a function of rrs that
dU
drrs
rs
_ dUr
'
CXg
CtTfg
(jOi
'
drr8 dys
(r)
dy8
=
d
If
due
now we
find the resultant
to the repulsions
r
-s
by
all
dUlS
= ~oT"
o rr
of
^"~
mr
we
shall
have
dUns
dU~2S
f~
all
the particles
P
o7T~
o TT
^\ rr
dC/oo
o rr
vUg
^ = ^~ +^r+-- + OUns
^7 = ^r'
T-T
(29)
C/L/ifi
'
d^s
if
we
write
U = Uu +
8
^728
...+
CTn
Thus
dzg
the con-
satisfies
ditions for a force-function as far as concerns the point

summation s does not occur as the first index.
In the
WORK AND ENERGY.
59]
It is evident that the function
rs
113
serves the
same purpose
for
For the force F exerted on mr by me is equal and

to
that
exerted on ms by mr
But rrs is the same function
opposite
of ( xr ) that it is of x8) therefore
mr as for m
s.
drr8
XV
and
We may
IS)
(s}
add to
JL /
rrs
(r) 5
</>
drrg
terms independent of #s
Z7g
X Y
yg zs without
,
Zs If we make a symmetrical
s
s
affecting the values of
the
of
all
function
coordinates, containing sc8 yg zs as s does, then
U will serve as the force function for all the coordinates.
,
In particular, let the force of repulsion vary as the product of

the masses of the particles divided by the square of their distance
apart

and the symmetrical function
U will
'
2
\
12
be
8
,

The
factor
is
[PT.
I.
CH.
II.
introduced because in the above summation
But
in
no constant be added to
as defined above, both
every term appears twice.
each pair of particles
is
to
appear only once.

If
it
and
the potential energy
F = i2A^,
r
(33)
rs
will vanish
when every rrs
is infinite,
that
is
are within a finite distance of each other.
when no two
particles
This furnishes a con-
venient zero configuration for the potential energy, and is the one
may accordingly define the potential
generally adopted.
of
the
in
system
any given configuration as the work that
energy
must be done against the mutual repulsions of the particles in
We
order to bring
configuration.
them from a
state of infinite dispersion to the given
CHAPTER
III.
HAMILTON'S PRINCIPLE.
GENERALIZED EQUATIONS OF MOTION.

60.
we
Hamilton's Principle.
CYCLIC SYSTEMS.
If in d'Alembert's equation
consider Sx, Sy, Sz variations consistent with the equations of
condition,
we have
d2x
-_ d
(dx bX
(dx X
dx
~~~'
dx
MX
)~~di~dt
di\di
dx
Treating each term in this manner,
d ~
(dx
dy
dz
*
,
dz
If there is a force function
U we
member
of (1)
hence the right-hand
8T+
have
is
SU.
82
116
The
member being an
left-hand
grate with respect to
[PT.
exact derivative
I.
CH.
we may
III.
inte-
dz
f
fdx *
dy ~
Sac
% \m
+ -=I-,+-g8y
dt
dt
V<
["-,
(2)
If the positions are given for t and tl} that is if the variations
Sx, 8y, &z vanish for t and ti, then the integrated parts vanish,
and
or
(3)
This
is
known
as Hamilton s Principle*.
It may be stated by
the configuration of the system is given at two
and t1} then the value of the time-integral of
U is
saying that
instants
tQ
if
T+
the paths actually described in the natural

motion than in any other infinitely near motion.
less (or greater) for
Hamilton's principle is broader than the principle of energy,

inasmuch as U may contain the time as well as the coordinates.
It is true even for non-conservative systems (where a forcefunction
61.
does not exist),
we
if
write instead of
Lagrange's Generalized Equations.
Hamilton's Principle we
motion.
BU
By means
may
Suppose that by means of the equations of condition,

there are any,
m = 3n
we express
of
deduce the generalized equations of
all
k parameters qlt q 2
if
the coordinates as functions of

...
qm
which are known as the
generalized coordinates of the system,
Then W,
if
the parameters
* Hamilton.
=x
the system
is
(q l) q 2)
...
qm )
conservative, becomes a function of
q.
On a General Method
in Dynamics.
Phil. Trans. 1834.
60, 61]
Differentiating the above
dx
= dx
dt
dql
Since the
dx
is
every
dqi
by
t,
^x l dq%
dxl dqm
}
'dt^dq
~di' ~'"^dqm ~dt
are given as functions of qlt q2)
x, y, z's
given also as a function of the ^s.
velocity-component
is
a linear function of the
cients are certain functions of the

velocities
117
q's.
corresponding to the coordinates
Squaring, adding, and summing,
r = iSr
(2)
The
q's,
...
qm alone,
Hence every
whose
coeffi-
q's are called the
q.
we get
7T
a homogeneous quadratic function of the ^'s, whose coefficients

are certain functions of the q's, so that we may write
^=JQn?/
(3)
where
^^
Performing the operation of variation upon the integral occurring in Hamilton's Principle, we obtain
and since
we may
final
Since the initial and

integrate the second term by parts.
is supposed given, the Sq's
of
the
configuration
system
118
vanish at
=t
and
and
"i
Jt
Now
if
lr=i
all
=t
~>
is
Tf)
write
7
'
for the
d(*(T-W)\)'
dqr
Lagrangian function
d^
^-
if
equal to zero.
dt\
dqr
we
III.
the Sqs are arbitrary, the integral vanishes only
the coefficient of every &qr
or if
CH.
d ft(T^
dt\
dqr
~~j~*\
2qr
I.
so that the integrated part vanishes,
l)
-T - F
p FT
^ P
\S)
[PT.
f dL \
3Z
'
dt^qr')" dqr
**
Since the potential energy depends only on the coordinates,
dW =
->
0,
and we may write the equation
(6)
_-~
dt\dg)
dqr
dqr
Pr
is the generalized component of impressed force tending to

increase the coordinate qr
.
If the
of
system
not conservative, we must write, instead
is
-STF
ZrlXrSXr+Yrtyr +
and the integral
BfTdt + /2 [Xr Sxr +
(8)
ZrSZr],
is
Y Sy + Z
r
r $zr ]
dt
= $f(T + ^Xr xr + Yr yr 4 Zr zr) dt.
_
^ = dr
XT
Now
so that if
we
dxr
S?1
+ _S?2+
~
far ^
......
+^Sqm
write
we get the same equations
as before.
If there
is
a force-function
<W
fo,
and
'
J3F8^ + 8F8y, ^ aF3^
61, 62]
m of
There are
the equations
(7),
one
for
119
each coordinate
q.
These are Lagrange's equations of motion in generalized
co-
ordinates.
Proof independent of Hamilton's Principle.
62.
We
will verify these equations
by
direct transformation of the
equations in rectangular coordinates
do)
which are obtained from equation (15) of Chapter
by means of
II.
d'Alembert's principle.
Multiplying these respectively by
adding and summing

oomes
for all values of r,
Sr
\dx
dqs
dyr
If there are no relations
is
an
identity,
and
dq
!r =
&r dqj
between the
all its partial
Accordingly the terms in
We
"*"
'
\ lt \,
the coefficient of X a be-
<?'s,
derivatives
...
(n)
Now
m f^<^
T=
^^ +
f?r 95
'
the expression
-j~
disappear.
have then
_i
dq s
are equal to zero.
120
[PT.
CH.
I.
but by (i)
Differentiating
a?/
a?r
a~ = O
by gg
#r
9ft
O
oqityt
<?1+00
+
.
g2
. . .
-\fr/
7.
dt
dq^q*
Inserting these values in
dT
dT
=-in
.
,,
d /dxr \
dt
fej
(12),
+ yr,d
S
/dyr \
+ 2r
(dq J
r \]
fdz
dt \ftj\
>
j8^-87r
dx
r
_^
*rmr j^?/ r + dy
'
dyJ'
_l_
'
\.
'
'
'
\dt dqa8
dt
dq s
dt
dqs )
which, since
is
equal to
r
Hence we have proved by
direct transformation the expression
(^ _ ^
8^
d V^s
to be equal to
-p
III.
62, 63]
rlT
The
the q"s,
derivative
is
121
which
is
a homogeneous linear function of
generally denoted by
In the case of rectangular coordinates,

*/'
+ yr'
dT
momentum of one of the particles. In
be called the generalized component of momentum,
The equations of
belonging to the coordinate qs and velocity qs
motion may be written
the ^-component of the
general,
ps may
'.
dp 8 = dT
,-f-.
dT
if, as we shall in future do, we denote by

s simply that part
of the impressed force which is not derived from the potential
energy, under which are included all non-conservative forces,
or
Theorem on Reciprocal Functions. The ordinary

63.
notation for partial derivatives of functions of several variables
sometimes gives rise to a certain confusion, from the lack of
what
are to be considered as constant

For instance, suppose we have a
function F of any number of variables, which for convenience
we will divide into two classes, denoting them by the letters
indication
of
variables
during the differentiation.
#!, #2
%n,
and zlt z2
...
Suppose now we have n functions of these
zm
variables, given
equations
2/1
fi (#1, #2
%n,
Zi,
Z2--Zm)
by the
122
I.
[PT.
CH.
III.
Let us now consider the function
By means
of the equations (i)
we may
insert the values of
the y's in terms of the as's and ^s, so that G is explicitly given as
a function of the variables x^...xn and z l ...zm . On the other
hand
let
us solve the equations (i) for the #'s obtaining
(2)
2/2
and by means of the
latter let us insert in
and
the values of the
#'s
Let the function in this form, that

is, explicitly given as a function of y^...yn and zr ... zm be denoted by G. Then for all values of z's and of #'s and ys compatible with the equations (i) or (2), we have identically
in terms of the
G(x
T/'S
z's.
...xn> z^...zm )=G(y^...yn)
Differentiating both G
variables that occur, we have
5i
and
oxs
totally
s
i
...zm }.
by varying
all
the
5oz
but as these are identically equal, we get by transposing,
In this equation there appear 2n + m
differentials, only
of which are independent, in virtue of the equations
),
n+m
or their
we
equivalents (2). The equation (3) assumes importance when

define the functions y in a particular way, namely as the partial
with regard to the
derivatives of the original function
variables x,
dF
63]
123
Then the coefficients of every dx8 vanish, and since we may

take the dy's and dzs arbitrarily, in order for the sum to vanish
we must have
for
every dy8 and dzs
dG_
'*"
ty.~
The function
function
is
dF_dG
a*~a*'
called the
function to the
reciprocal
with respect to the variables x ...xnt

l
for
we have
the reciprocal relations
or:
Two reciprocal functions have the property that the partial

derivative of either with respect to any variable of reciprocation
contained in it is equal to the corresponding variable replacing the
original in the other function, whereas the partial derivative of one
function with respect to any variable not of reciprocation is the
negative of the derivative of the other function with respect to the
same
variable.
In case the function
is
homogeneous of degree K in the
variables of reciprocation
the theorem becomes more striking, for then, by Euler's theorem
and the reciprocal function
is
simply a multiple of the original
function.
is
If the original function is of degree two, the reciprocal function

have thus a
identically equal to the original function.
We
striking example of the remark made at the beginning of this

section, for here the derivative of the function when expressed
is exactly the negative of the derivative
the
same
variable
of
the
function expressed in the other form.
by
In this form the theorem will be frequently used hereafter. By
means of it the equations of motion may be transformed from
in
one form by a variable z
Lagrange's form to that given them by Hamilton.
124
We
Hamilton's Transformation.
64.
[PT.
I.
have seen,
CH.
III.
61
(3),
that the kinetic energy is a homogeneous quadratic function of

the variables q representing the velocities,
T = iQn?/ + ife?/
2
Qtfi
V + QirfiV
the coefficients being functions of the coordinates q. If we call

the reciprocal function with respect to the <?"s, T, by the last
section this is also the kinetic energy, expressed not in terms
of the velocities, but of the momenta p. Any p s is a homogeneous
linear function of the q"s, so that solving the equations
(I)
O/7T
Pn =
for
and
the
g"s,
is
momenta
$~
oq n
every q
= Qmqi + Qnfla
is
therefore a
By
p.
function we have
Qnn<ln
a homogeneous linear function of the p's,

homogeneous quadratic function of the
two properties of the reciprocal

q s (variable of reciprocation), and
virtue of the
'
for every
every qs (not of reciprocation),
8T__ar
,_ar
9
"dp.
dqs
62 (17), are transformed to
so that Lagrange's equations,
dps
dt
If
we put
~
+ d__
dqs
,_,_
T 4-
W,
'
~dt~dp
dq 8
Lagrangian function
'
dq s
this is the reciprocal function to the
L = T-W,
and the equations take the nearly symmetrical form,
s__,p
dt~
,
'
dt~dp
dq,
These are Hamilton's equations of motion.
we may immediately deduce the integral

of
By cross-multiplication of the above equaenergy.
equation
tions, after transposing and summing for all the coordinates, we get
From
these equations
,_
~
T^Wi
fy
dt
8
i
~dt'
64, 65]
But
a function only of the p's and qs, so that the left-hand
is
member
125
is
-^-
total energy.
and since
equal to
is
T+
it
represents the
P dq is the work done by the external impressed

P in the displacement dq so that the right-
Also
force
component
hand
side is the time-rate at
g)
which the external
the system, or the activity of the external
dt~,
forces
forces.
do work on
The equation
dt
is accordingly sometimes called the Equation of Activity, while if

there are no external forces, but only conservative ones, we have
the equation of Conservation' of Energy,
= 0,
H= T+W= const.
Cut
case of frequent occurrence is that where there are nonconservative forces proportional to the first powers of the velocities
f
= Ks qs '.
so that any
s
q
may then form a function
which
is
also
a homogeneous quadratic function of the velocities
F=^
(7)
We
Ks qs '^
P.
and since in
j>,
oqs
this case
>
dqs
F represents one-half the time-rate of loss, or dissipation of energy.

F is called the Dissipation Function. It was introduced by Lord
Rayleigh *, and, like the other function used above,
is
of use in the
theory of electric currents.

65.
Transformation of Routh and Helmholtz.
We
shall in general find

Lagrange's form of the equations of motion
more convenient than those of Hamilton.
intermediate form,
An
introduced by E-outhf, and afterwards by Helmholtz $,
is
of great
importance.
*
Proceedings London Mathematical Society, June, 1873.

t Routh. Stability of a given State of Motion, p. 61. Rigid Dynamics, i. p. 318.
Ueber die physikalische Bedeutung des Princips der kleinsten
J Helmholtz.
Wirkung. Borchardt's Jour, fur Math. Ed. 100, 1886. Wissensch. Abh. in. p. 203.
126
[PT
I.
CH.
III.
Suppose that instead of reciprocating with regard to all the

Hamilton's transformation, we do so with regard
a
to only
number r of them which we will choose so that they shall
have the indices from 1 to r, while the ^"s with indices r 4- 1, ... n,
remain in the reciprocal function, and with all the coordinates q
velocities q as in
play the part of the variables z in

of the reciprocal function
T = T-$qs'ps
(0
Then
63.
calling the negative
we have
dT = dT
,
9* 8 &
dT dT
(2)
for
dS-sg'**
s=l,
9
2, ...
?i,
-*^-"*
and
dT
Replacing
T in
dT
Lagrange's equations by T,
-i
we
c>
obtain
~
d(dT\_dT__SW
H
dt(dqs ')
dq s
8^
we may use for the suffixes corresponding to the un

eliminated velocities Lagrange's equations, using the function,
so that
tymT-W
instead of the Lagrangian function
l**T-W,
and obtaining
For the
suffixes corresponding to the eliminated velocities

form of the equations
Hamiltonian
the
use
dt
dqe
dqs
_d(T-W)
dt'
If r
= n, T
form,
64
we must
becomes
(4).
T,
~dj>r
and we have the complete Hamiltonian
65, 66]
The function

is
called
127
by Routh the modified Lagrangian
and on account of
its importance has received from

Helmholtz the special name of the Kinetic Potential, by which
we shall designate it. (Helmholtz calls <J> the kinetic potential*.)
function,
It is to be noticed that the equations for the elimination of the

velocities,
the equations (2) of
63 are now, instead of
Quqi
+ Qi
Qriqi
+ Qntqs '"+ Qrrqr = Pr ~ Qrr+i q'r+l
...
g2
Qi r qr
=PI-
Qir+iq' r +i
...
64
(i)
Qi n qn ',
~ Qro^n',
so that the q"s become linear functions of the

right-hand sides of
these equations and hence of
Pi,
thus
T becomes
p 3t
...
pr
is
q'r+i-" qn ',
homogeneous quadratic function
Pt-.pr and
but
tfr+i
of
qn,
not homogeneous in either the p's or the q"s alone, on account

pt q/ which are linear in either the p's or <?"s.
of terms such as
Concealed Motions. A system is said to contain conwhen the coordinates which become known to us by
66.
cealed masses,
observation do not suffice to define the positions of all the masses

of the system.
The motions of such bodies are called concealed
It is often possible to solve the problem of the motions
of the visible bodies of a system, even when there are concealed
motions.
motions going on. For it may be possible to form the kinetic

potential of the system for the visible motions, not containing the
concealed coordinates, and in this case we may use Lagrange's
equations, as in the preceding article, for all visible coordinates,
may be ignored.
are incomplete, inasmuch as they tell us nothing of
the concealed motions, but very often we are concerned only with
while the coordinates of the concealed masses
Such problems
Such concealed motions enable us to explain

the forces acting between visible systems by means of concealed
motions of systems connected with them.
the visible motions.
As an example
of a concealed motion let us take the case of a
closed box containing a gyrostat, or fly-wheel, pivoted on an axis

*
Helmholtz's notation
is
quite different from that here employed.
128
[PT
I.
CH.
III.
rigidly fastened to the box. If the box be at rest, there is nothing

on the outside to lead one to suspect the presence of the fly-wheel,
but if the box be moved about the reactions developed will be very
different if the concealed wheel is in rotation or not.
In the
former case the least experimenting will render us sure of the

To convince oneself of the truth
existence of a concealed motion.
of this statement
one's
it is
hand and turn
67.
it
necessary only to take a toy gyroscope in

about.
Cyclic Motions.
In certain cases some of the co-
ordinates do not appear in the expression for the kinetic energy,

although their velocities may. Such are termed by Helmholtz*
and we
shall distinguish them by a bar.

The
the
is
such
a
of
for
the
case,
gyrostat
angular
example just given
coordinate fixing the rotation of the fly-wheel does not appear, but
only its derivative, the angular velocity. Further examples are
cyclic coordinates,
furnished by the case of a heavy belt running over pulleys, or by

the case of a fluid circulating in an endless tube. The coordinate
expressing how far a point on the belt or in the fluid has travelled
does not enter, but its velocity does. The condition for a cyclic
coordinate being
motions
[61
we have
for the forces
maintaining cyclic
(7)],
d_/d
dt \3g/>
If the forces of the cyclic motions vanish
we have
'
dt \dqs
or, integrating,
8T
In this case we
may
with advantage apply Routh's transformation
in the case of the cyclic velocities.

tion are now
*
Jiir
The equations
for the elimina-
= cr - Q
Helmholtz, Studien zur Statik monocyklischer Systeme.

Wiss. Abli. in. p. 128.
Math., Bd. 97.
Borchardt's Journ.
CYCLIC SYSTEMS.
66, 67]
Let the solutions of these
129
the q"s be
for
The R's being the quotients of the various subdeterminants of the

determinant
ri
\rz
Qr
by the determinant itself, are functions of the coordinates only,

and since by hypothesis the function T did not contain the cyclic
coordinates, the R's are functions of only the non-cyclic coordinates.
The kinetic potential consequently is a function only of the non-
and
velocities, but on account of the presence of

not a homogeneous function of the velocities,
but contains a linear function of them, as was remarked in
65.
cyclic coordinates
the constants
cst it is
Cases in which the kinetic potential contains a linear function

may thus be considered as cases with concealed
of the velocities
motions.
case of this nature will
be found in considering
the mutual actions of magnets and electric currents.

the difference between the two cases is that while if
Physically
contains
<E>
only terms of the second degree in the velocities, if every velocity

reversed the kinetic potential is unchanged, and hence the
is
motion may be reversed without change of circumstances, but if

on the other hand there are terms of the first degree in the
velocities,
the motion cannot be reversed unless the concealed
motions are reversed as well.
As an example we
gimbals.
take the case of a gyrostat hung in

Let the outer ring of the gimbals A, Fig. 27, be
will
pivoted on a vertical axis, and let the angle made by the plane of
Let the inner ring B
the ring with a fixed vertical plane be -fy.
be pivoted on a horizontal axis, and let its plane make an angle 6

with the plane of the outer ring. The gyrostat is pivoted on an
axis at right angles with the last, and let a fixed radius of the
It is
gyrostat make an angle </> with the plane of the inner ring.
shown in the theory of the dynamics of a rigid body that the
energy of a body revolving about an axis

W. E.
is
one-half the product

9
130
of a constant called the
by the square of
its
I.
[PT.
CH.
III.
moment
of inertia of the body multiplied

angular velocity, and also that if we find the
FIG. 27.
angular velocities about three mutually perpendicular axes of

symmetry the energy may be found by adding the three parts
obtained for the energy of rotation about the three axes. We
motions of the gyrostat into three angular veloabout the axis of the top, the axis of the inner ring, and
an axis perpendicular to both. About the axis of the gyrostat
will resolve the
cities,
the angular velocity

velocity -~Jf
= ty'
is
-^-
but there
</>',
about the vertical
axis,
also the
is
which has the com-
The
ponent ^r cos 6 about the axis of the gyrostat.

about the second axis
is -=-
at
= 6',
and about the third
component of the velocity about the
vertical,
angular
-\J/
sin
6.
is
velocity
the other
If
A is
B
the
that
of inertia of the gyrostat about its own axis,
about either of the other two, we have for the kinetic energy
moment
T = \ [A (f + tf cos 0) + B (0' +
2
and
are cyclic coordinates.

increase ^, 0, (/>,
to
forces tending
so that
<jE>
^Jr
sin 2 6)1
For the components of the
CYCLIC SYSTEMS.
di (a
dT\
'
dt
+ +' cos *> cos
[A (0
no
+ ^' sin2 l
dt,
+ A(<t>' + ^' cos
is
dT = d
d0
If there
force tending to
- B^'
6) sin 6
sin 6 cos 0,
change the rotation of the gyrostat
in its ring
and eliminating
<'
by means of
(f>'
this equation,
= -T
-V/T'
cos
6,
jCi
3>
T-c<j>'
B
-^ I + \ (^ + ^'
sin 2
6>)
+ cf
the last term containing ^' in the first power.

 to determine the
forces, we obtain
P -1
^
The
if
(I?)
M = I (W
>
- ^'
influence of the cyclic motion
the vertical ring be held fixed.
sin
cos
'
cos
Using
0.
this
form of
sn
+c
may be most simply shown
Then
t]r
= const.,
and
ifr'
= 0,
Spinning the inner ring about the horizontal axis requires the
force whether the cyclic motion exists or not, whereas a force
developed tending to make the vertical ring revolve about its
same
is
7/1
axis,
which must be balanced by the force
force at once
shows that there
is
c sin
-y-
This
a concealed motion, even if the

unknown. This is exem-
disposition of the concealed parts be
the gyroscopic pendulum, which is simply a pendulum

with two degrees of freedom, containing a gyrostat whose axis is
plified in
92
THEORY OF NEWTONIAN FOECES.
132
[PT.
I.
CH.
III.
An ordinary pendulum
rigidly fixed in the axis of the pendulum.
set vibrating in a plane continues to vibrate in a plane, with a
The gyroscopic pendulum on the
periodic reversal of its motion.
other hand describes a curious looped surface, never remaining in
a plane nor returning on its course. This example is worked out
in
Thomson and
68.
is
Tait's
Natural Philosophy,
Cyclic Systems.
represented with sufficient
319,
Example
(D).
system in which the kinetic energy

approximation by a homogeneous
quadratic function of its cyclic velocities is called a Cyclic System.

Of course the rigid expression of the kinetic energy contains the
velocities of every coordinate of the system, cyclic or not, for no
mass can be moved without adding a certain amount of kinetic
Still if certain of the coordinates change so slowly that
energy.
their velocities may be neglected in comparison with the velocities
of the cyclic coordinates, the approximate condition will be fulfilled. These coordinates define the position of the cyclic systems,
and may be called the positional coordinates or parameters of the

the two coordinates of the
system. In the case of the gyrostat
for
the
taken
be
positional coordinates, while
gimbal rings may
the cyclic coordinate determines the rotation of the gyrostat. In
the case of a liquid circulating through an endless rubber tube,
the positional co-ordinates would specify the shape and position of
the tube. The positional coordinates will be distinguished from
the cyclic coordinates by not being marked with a bar. The
analytical conditions for a cyclic system will accordingly be, for all
coordinates, either
~\m
^m
or
(I)
alT
or if
we
Wr p
use the Hamiltonian form of
'
Q>
T obtained
by replacing the
by the momenta, which we shall denote by

non-cyclic momenta vanish
velocities
Tp
since the
^-0 ^-- Sq^ =

S
'
(2)
dp.
Sq,
We
forces tending to inaccordingly have for the external

62, (17)],
crease the positional coordinates [see
~~
W)
d(T-W)_d(T,+
~
CYCLIC SYSTEMS.
67, 68]
and
133
for the cyclic coordinates
*A motion
cyclic
which there are no forces tending to change the

coordinates is called an adiabatic motion, since in it no
in
energy enters or leaves the system through the cyclic coordinates.

may do so through the positional coordinates.) Accordingly
in such a motion the cyclic momenta remain constant.
The case
(It
of the gyrostat
worked out above was such a motion.
In adiabatic motions the cyclic velocities do not generally

In the above example, for instance, the cyclic
remain constant.
velocity
<f>'
was given by
p = -| - y cos
A
motion in which the
0.
cyclic velocities
remain constant
is
called isocyclic.
In such a motion the cyclic momenta do not generally remain

constant, but forces have to be applied.
is isocyclic, the only variables appearing in T are
the
The positional forces, (3), are
q's,
positional coordinates.
then derivable from a force-function
T *, so that even if the
If the motion
the
system possessed no potential energy, it would appear to possess

an amount of potential energy
If the motion on the other
T.
hand is adiabatic, the energy in the form Tp again contains only
the coordinates q s and the positional forces are now derivable
from the force-function Tp + W, so that in this case a system
,
without potential energy would appear to contain the amount of

In this manner we are enabled to explain
potential energy + Tp
.
potential energy as kinetic energy of concealed cyclic motions,

thus adding materially to our conceptions of the nature of force.
For
it is
to be noted that kinetic energy
only on the property of
inertia,
which
is
an entity depending
possessed by all bodies,
is
while potential energy is a term only employed to disguise our

ignorance of the nature of force. Accordingly when we are able
to proceed to an
explanation of a static force by means of kinetic
*
The reason
plained in
for the
appearance of
W with
62, end, P, denotes the external
equilibrium, are equal
and
the positive sign
is
that, as ex-
impressed forces, which in the case of
opposite to the internal forces given
by W.
134
[PT.
I.
CH.
III.
phenomena, we have made a distinct advance in our knowledge of

the subject. A striking example is furnished by the kinetic
theory of gases, by means of which we are enabled to pass from
the bare statement that
gases press against their confining
all
vessels to the statement that this pressure is due to the impact of

the molecules of the gas against the walls of the vessel.
Properties of Cyclic Systems.
69.
Reciprocal Rela-
Since by the properties of the kinetic energy we have

three different kinds of quantities represented by partial deriva-
tions.
tives of one or the other of
dT
two functions,
dT
^jP,
applying the principle that a derivative by two variables is independent of the order of the differentiations we obtain six reciprocal
theorems.
We
shall
throughout suppose that there
is
no potential
energy.
I a.
In an adiabatic motion
if
an increase
in
one positional
coordinate qr causes an increase in the impressed force

s belonging to another positional coordinate q s at a certain rate, then an
increase in the positional coordinate q s causes an increase in the
impressed force r at the same rate. For
o TD
d-*
(r\
dqr
I
b.
In an
isocyclic
o o/
" 7^
d 1P
^
3
dqr dqs
~D
'
dq 8
motion we have the same property as
For
above.
(6)
'
dqr
II
a.
dq rdqs
dqs
If in any motion an increase of
any
cyclic
momentum
pr the positional coordinates being unchanged, causes an increase

'
in a cyclic velocity qs at a certain rate, then an increase in the
,
momentum pS) the positional coordinates being unchanged, causes

an increase in the velocity q r at the same rate. For
'
(7)
Wr
II b.
If in any motion an increase in any cyclic velocity q r
the positional coordinates being unchanged, causes an increase in
'
a cyclic momentum ps then an increase in the velocity qs causes
',
(58
CYCLIC SYSTEMS.
70]
an increase in the
momentum pr
135
same
at the
rate.
For
3&_ _!*L.fr
:
a8&'~33."
a?/
Ill
If an increase in one of the cyclic
a.
momenta p r
the
positional coordinates being unchanged, causes an increase in the

impressed force s necessary to be applied to one of the positional
coordinates q s (in order to prevent its changing), then an

adiabatic increase of the positional coordinate qs will cause the
'
cyclic velocity q r to increase at the
same
For
rate.
^
dpr
III
If
b.
an increase in one of the
cyclic velocities
/,
the
positional coordinates being unchanged, causes an increase in the

impressed force s necessary to be applied to one of the positional
coordinates q g (in order to prevent its changing), then an isocyislic

increase of the positional coordinate qs will cause the cyclic
momentum pr to decrease at the same rate. For
'
dq r
dq r'dqs
Work done by
70.
dq.'
the cyclic and positional forces.
In an isocyclic motion, the work done by the cyclic forces

I.
double the work done by the system against the positional
In such motions the energy of the system accordingly
forces.
is
by one-half the work done by the
increases
cyclic forces, the other
half being given out against the Positional forces.

the energy in the form
we have
in
sr=i2. (?;
isocyclic change, every
<?/
vanishing,
8T=^ q;sPs
(2)
But
since
^=
p., Sp s
P,St,
and the above expression

(4)
if
any change
(1)
and in an
(3)
For
ST =
for
&
and since
q,
^|
q.'St
= 8?.,
the gain of energy becomes
q 8'P,to
= &P,8q,.
we use
1.36
But the work done by the
Hence the
last part of
I.
CH.
III.
cyclic forces is
8A=2 P
(5)
[PT.
Sq s
=2ST.
the theorem
is
Again, in any
proved.
motion
(6)
tf-xg-Vt^gifa
and in an
isocyclic
motion
(7)
M--*.|V
But
since the
work of the
positional forces
is
&A = 2 sPs Sq s = -2~Sqs =-
(8)
ST,
u$t
the
first
II.
part of the proposition
is
also proved.
In an adiabatic motion, the cyclic velocities
will in general
be changed.
Then they change
in such a
that the positional forces

caused by the change of cyclic velocities oppose the motion, that
amount of work. For since for any positional
is, do a positive
way
force
p _j>T
'
dqs
the change due to the motion
is
"r
Of
this the part
^'
due to the change
*r
'
have
^r
dqs dq r
and the work done by these
Now we
dqsdq r
dq 8
for
in the cyclic velocities is
^r
*r
dq s
forces is
any motion
and in an adiabatic motion
this is zero, so that
'
CYCLIC SYSTEMS.
70, 71]
sum with
Substituting this in the
respect to s in 8q>A
A = 2s Z r ^, Sq.' S?/ = 2a
S?
137
-S. r
G{ S
Qr,Sq.'Sqr
we get
'.
But this expression represents ( 61 (3)) twice the energy of

a possible motion in which the velocities would be &q8', and must
f
therefore be positive for all values of &q s
qr
'
^>A >0.
Accordingly
The interpretation of
known as Lenz's Law*.
this
theorem
for
electrodynamics
is
71.
Examples of Cyclic Systems. The expression for the
kinetic energy of the gyrostat worked out in
67 shows that
the system fulfils the conditions for a cyclic system if the velocity
0' is small
enough to be neglected in comparison with the other
velocities.
The forces acting have been already found, and we
can easily verify the theorems of the last two articles for this
case.
A very simple case of a cyclic system is that of a mass m sliding

on a horizontal rod, revolving about a vertical
axis.
Let us consider the mass concentrated at
a single point
at a distance r from the axis.
Let the angle made by the rod with a fixed hori-
zontal line be
to
the rod
</>,
then the velocity perpendicular

r
The
velocity along the rod

the
is
kinetic
being r,
energy of the body
is
r$
FlG 28<
-
If
may
we suppose the motion along the rod
neglect
r'
to be so slow that
we
T= Jmrty'
and the system
is
cyclic,
is
8
,
the positional,
<
the cyclic co-
ordinate.
A system containing a single cyclic coordinate is called by

Helmholtz a monocyclic system. We have for the momenta
dT
*~gp-0,
*
These Theorems are
all
dT
_.,
*-gy-Hf,
given by Hertz, Principien der Mechanik,
568-583.
138
and introducing these instead of the
We
have
for
[PT.
I.
CH.
III.
velocities
the positional force
This being negative denotes that a force Pr toward the axis

must be impressed on the mass
in order to maintain the cyclic
accomplished by means of a geometrical

The force or reaction
constraint, or by means of a spring.
r
which the mass
exerts in the direction from the axis in virtue
This
state.
may be
of the rotation
the motion
We
called the centrifugal force.

see that if
the positional force increases with r, while
is
is isocyclic,
when r increases. The

obvious.
The cyclic force
if it is adiabatic, it
decreases
of the theorems of
69
is
verification
when the rotation is uniform, and the radius constant.

the
motion
If,
being isocyclic, that is, one of uniform angular
velocity, the body moves farther from the axis, P^, the cyclic
vanishes
force
that
is positive,
is,
unless a positive force
P$
is
applied,
In moving out from r-i to r2

the angular velocity will diminish
work will be done against the positional force r of amount
-A = -
P dr = mf
rdr
Jr
(V2
- rf),
Jri
while the energy increases by the same amount.
Thus the
first
theorem of
adiabatic,
Pi
If the
70
is verified.
= mrty' =
If the motion is
c.
will accordingly
body move from the axis,
in Pr due to a displacement Br is
</>'
decrease.
The change
which, being of the same sign as Sr, does a positive amount of

in the displacement, illustrating the second theorem of
work
70.
CYCLIC SYSTEMS.
71]
139
The preceding example will suffice as a

illustrate
to
the phenomenon of self-induction
model
mechanical
current
of an electric
(Chapter XII). To illustrate mutual induction we must have at least two cyclic coordinates.
Such
models have been proposed by Maxwell, Lord
Rayleigh*, Boltzmann, J. J. Thomson f, and the
author J. In the model of J. J. Thomson, there
are two carriages of mass m^ and ra 2 sliding on
Dicyclic Systems.
parallel rails, Fig. 29, their distances from a

fixed line perpendicular to the rails being x^
Fl(J
29
Sliding in swivels on the carriages is

shall suppose that this
a bar, on which is a third mass ra 3
mass is movable along the bar, and is at a distance y from the
and
a? 2
line
We
rails, y being positive when

the distance between the rails,
midway between the

Then,
2.
if
is
and the kinetic energy

with #/, #2
is,
if
we may neglect y
is
nearer
in comparison
',
T=
-I )]The system
is
cyclic coordinates.
vanishes
if a?/
=#
2 '.
cyclic,
The
y beingf the
positional,
and
x.2
positional force
The
cyclic forces are
PWi. Mag., July 1890, p. 30.

t Elements of Mathematical Theory of Electricity and Magnetism,
J Science, Dec. 13, 1895.
p. 385.
the
140
[PT.
I.
CH.
III.
Suppose that the coordinate y and the velocity #/ are constant.

is increased, say by mt
a&i
starting from rest and moving
to the right by the application of a positive force PXi then PXi is
in other words,
positive if y < d/2 and m 3 is within the rails,
unless a force to the right is impressed on ra2 also, #2 will diminish,
and if #/ was also zero, ra2 will move to the left.
If
now
'
must be greater the smaller \y\. This is the

Xz
induction
of currents.
the
of
Similar effects may be
analogue
the
instead of applying a
rod,
3
produced by moving
along
The
force
force to
or ra 2
Fm.
30.
Maxwell's model, which undoubtedly suggested Thomson's,

from it only in having motion of rotation instead of revolution, so that there is no limit to the possible difference in the
differs
coordinates
the
as lf
moments
The independent masses
as 2 .
of inertia of masses
m m
1}
are represented by
by two shafts
carried
Fig. 30, each of which carries a bevel-gear wheel A,

at one end.
Engaging these is a pair of bevel-gears C running
$!,
loosely
upon a third perpendicular
mediate mass,
shaft,
carrying the inter-
3.
If all the bevel-gears are of the same diameter, and <f> lt <J> Z fa
are the angles made by the three horizontal rods with a fixed
horizontal line, then it is evident, since the velocity of the centre
of the wheel
is a mean between the velocities of its highest
,
and lowest
points,
rims of the wheels
which have respectively the

and B, that
velocities of the
CYCLIC SYSTEMS.
71]
141
Consequently the kinetic energy of the system consisting of

m l m 2 s at distances from the axis rlt ra r, is
the three masses
the velocities r can be neglected. The system is

being positional, the <f>s being cyclic coordinates.
if
make the model a more complete

currents,
Boltzmann modified
cyclic,
the
r's
In order to
representation of two electric

between the co-
so as to have
it
ordinates rlt r2 r 3 the relation

,
where y lt y z are two independent parameters. The two masses

l} Wa are chosen equal, being made one-fourth of m 3
The expression
for the
energy then becomes
T=m to ^'
2
and
+ \y^ + rfh'ti},
we may independently change
either
of
the three co-
efficients.
The Pythagorean theorem suggests a geometrical means

imposing the above constraints. To each of the masses m
of
is
attached a string, which runs along the rod to the axis of rotation, where, after passing round a pulley it is carried vertically
downward to be attached to the following device (Fig. 31). A
FIG. 31.
f
pair of rods are articulated at C the point of articulation being

made to slide in a vertical line CO. The string from ra3 is fastened
,
AB
and in slots
Sliding on a horizontal line
the strings
of
attachment
of
BC, are the points
and
outward
carried
which are then
upward
2
to the point C.
in the rods AC,
from m^ and
THEORY OF NEWTONIAN FOECES.
142
[PT.
I.
CH.
III.
over pulleys. The lengths of the strings being chosen so that

and 2 when
the rods are horizontal,
x
3 is at the axis when
the rods are vertical,
we must
2 _l_
<y
M. ~T /3
have,
I/ 2
y\
if
TZ
/v*
~T
m
AC = y BC = y
lf
T3
,v
/2
2/2
For the actual construction of the model, the reader is referred

Vorlesungen uber die MaxweWsche Theorie der
to Boltzmann,
Electricitdt
und des
Lichtes.
By means of these models all the properties of Cyclic Systems

may be illustrated, and all the phenomena of induction of currents
imitated, as will be described in Chapter XII.
Hamilton's Principle the most general dynamical

principle. We have seen in this chapter how by means of
Hamilton's Principle we may deduce the general equations of
motion, and from these the principle of Conservation of Energy.
As Hamilton's Principle holds whether the system is conservative
or not, it is more general than the principle of Conservation of
The principle of energy is not
Energy, which it includes.
If we know the
sufficient to deduce the equations of motion.
function
can
at
once
form
the
we
equations of motion,
Lagrangian
For we
find
the
and without forming them we may
energy.
72.
have
L = T-W,
Accordingly
so that the energy is given in terms of

If on the other hand the energy
vatives.
and
its partial deri-
given as a function
is
of the coordinates and velocities, the Lagrangian function
be found by integrating the above partial

involving an arbitrary function. In fact if
linear
function
satisfied
of the
not only by
homogeneous,
velocities,
must
differential equation,
be a homogeneous
the above equation will be
but also by
L+
F.
For,
being
71, 72]
143
Consequently a knowledge of the energy is not sufficient to

find the motion, while a knowledge of the Lagrangian function or
kinetic potential
is.
In case we wish to ignore some of the coordinates we may

modify the statement of Hamilton's Principle by the use of the
modified Lagrangian function and put
where we suppose only those coordinates which are not ignored

are varied.
CHAPTER
IV.
NEWTONIAN POTENTIAL FUNCTION.
We
have seen in
of Potenwe have any

according to the New-
59, (29), (31), that if
number
tonian
and fundamental properties
Definition
73.
tial.
of material particles
repelling
of the inverse square of the distance, the function
Law
mm
os
TT
U
s
r?
where rlt r2
rn are the distances from the repelling points,

the force-function for all the forces acting upon the particle
is
we put the mass
If
.
y _m
equal to unity the function

l
m^
mn
^2
Til
r\
called the potential function of the field of force due to the

n and its negative vector
lf
2
repulsions of the particles
parameter is the strength of the field, that is, the force experienced
is
m m
by unit of mass concentrated

term -
we have
possesses the
as
where r
for every term, for points

0,
at the point in question.
same properties
Since any
the function is
39,
not equal to zero,
and consequently
(2)
VTl/
74.
Potential of Continuous Distribution.
Suppose now
that the repelling masses, instead of being in discrete points,

form a continuously extended body K.
Let the limit of the ratio of the mass to the volume of any
infinitely small part
be p
= lim A T =O
AT
which
is
called the density.
73, 74]
145
Let the coordinates of a point in the repelling or

attracting* body
be
c.
a, b,
The
potential at any point P, x, y,
z,
due to the mass
dm
at Q,
a, b, c, is
dm
FIG. 32.
where r
the distance of the point
x, y, z from the repelling

at
x, y, z is the sum of that
potential
to all parts of the attracting body, or the volume integral
point at
due
is
a, b,
The whole
c.
(3)
Now we
have
dm
or in rectangular coordinates
pdr,
dr = da db dc,
dm = p da db dc.
If the
body is not homogeneous, p is different in different parts

body K, and is a function of a, b, c, continuous or discontinuous (e.g. a hole would cause a Discontinuity). Since
of the
pdadbdc
(4)
For every point x, y, z, V has a single, definite value.

accordingly a uniform function of the point P, x, y, z.
It
may be
differentiated in
any
direction,
we may
It is
find
its
is
level surfaces, its first differential
parameter, whose negative
the
and
to
the
unit
whole
of
action of
on a point
mass,
equal
lines
*
offorce, normal to the
level, or equipotential surfaces.
In order to save words, and to conform to ordinary usage, we shall say simply
attracting, for a negative repulsion is
W.
E.
an attraction.
10
146
[PT.
I.
CH.
IV.
If for any point x, y, z outside K, r^ is the shortest distance to

and r2 the greatest distance, we have for any point
any point of
in#
r2
>r>r
111
-<-<
r
r2
dm
r2
dm
TJ.
dm
<
lt
<
dm
;
r-j
dm
dm
(5)
Since
r^
and r2 are constant
Now
above
since
dm = M,
is
the whole mass of the body K, the
r2
r,
Accordingly for an external point
R is the distance of
distance from
K,
x, y, z
is finite.
from some point in or at a
finite
RM <RV< EM
n
r2
If
M
M-<V<.
Jr
(6)
If
dm
now we move
off
oc,
y, z to
r
lim
RV
infinite distance
we have
R = hm R =
1,
..
^ = 00 ^2
and accordingly since

same limit
an
lies
R=cc T!
between two quantities having the
lim(RV) = M.
(7)
22=oo
We
say that
vanishes to the first order as
becomes
infinite.
75.
Derivatives.
Consider the partial derivatives of
x, y, z.
The element dm
at a,
b, c,
produces the potential
dm
jr
dV
,
at x, y, z
by
74, 75]
Differentiating
by
(dm and
so,
being constant), we have
a, b, c
dm
/T\
111
VV
r\
ox
\SV9tV r\
'
147
dr
r2
ox \rj
-"
By
39,
( 7 ),
r
^
?Tr\
5-(dF)
9#
(2)
^W& #
r2
&
r
Now
a
a?
>
cos (rx),
where the
direction of r is taken from a, b, c to x, y, z.

This being
the derivative for that part of the potential due to dm, we have
to take the sum of such expressions for all corn's in K, that is,
perform a volume integration
9F
dadbdc
Ip
II
- cos
(rx) da db
2
R be cos A,
>r>r
Let the direction cosines of

r2
r2
cos (rx)
>
^
~3
cos B, cos G,
and since
rT
cos (r#)
dc.
n
>
'
-^ cos (ra;).
Multiplying and dividing the outside terms by cos
and
.4
integrating,
cosJ. /Tf
(5)
9F
cosfe) rfT>
7
>
cosjl f[f
cos(rx)
^ --^jjJ' ^3^rjJJ''^ZMultiplying by R and letting R increase without limit, since

R = lim R = lim cos(r#) =
lim
)
dT
R* ^
lim
\
.R=a>
,.
U'B
,.
\=
1,
M cos A,
_J
102
148
\im\ R^^] =
*=* L
fyJ
(6)
R=>
- M cos
[PT.
I.
CH. IV.
cz
derivatives of V, and hence the parameter,

vanish at infinity to the second order.
Therefore the
first
In like manner
for
the second derivatives
Every element in all the integrals discussed is finite, unless

hence all the integrals are finite. We might proceed in this
manner, and should thus find that
= 0,
At points not in the attracting masses, Fand all its derivatives

are finite and (since their derivatives are finite) continuous, as well
as uniform.
Also since
-//>
we have by
addition
9V 9V 9V
+
w
w w
that
is,
This
V satisfies
is also
each element
76.
Laplace's equation.
proved by applying Gauss's theorem
39
(4)) to
Points in the Attracting Mass.
Let us now examine
the potential and its derivatives at points in the substance of

the attracting mass.
75, 76]
If
P is
149
within the mass, the element - - at which the

point Q,
where dm is placed, coincides with P, becomes infinite. It does

not therefore follow that the integral becomes infinite ( 25).
FIG. 33.
Let us separate from the mass
a small sphere of radius
with the centre at P.
Call the part of the body within this

K'
and
the
rest
K".
Call the part of the integral due to
sphere
is not in the mass
and
due
to
K"
that
V. Now since
K\ V,
',
K" V" and

,
examine
its
V and
derivatives are finite at P, and
we have only
to
its derivatives.
Let us insert polar coordinates
"
so that,
by
23
(5),
if
pm
is
V, <
fe
4t7rp
Jo
rdr
the greatest value of p in K'.
As we make the
radius
diminish indefinitely, this vanishes,
hence the limit
is finite.
In
like
manner
for
the derivative
from K" is
Separate off K' from K". The part of the integral
coordinates,
finite.
In the other K' introduce polar
putting
150
[FT.
I.
CH. IV.
-|
(2)
37_'
dx
which
< pm
also vanishes
\'dr
Jo
with
*|
sin
cos
JoJo
Hence
e.
is
-^
like
manner
dz
we attempt
it fails
and in
finite,
.-=dy
If
everywhere
this process for the second derivatives

^
on account of
dr
which gives a logarithm becoming
oo
. .
in
the limit.
We
will give
We
have
another proof of the finiteness of
(3)
which by Green's theorem
This
is
everywhere
when
is
equal to
however only to be applied in case the function finite
and continuous.
is
This ceases to be the case
in the attracting mass, hence we must exclude

by
a
small
drawing
sphere about it. Applying Green's theorem to
the rest of the space K" we have to add to the surface -integral
is
',
the integral over the surface of the small sphere.

Since cos (nx)
which vanishes with
than p m
this is not greater
1,
e.
Hence the
infinite
47re/o
element contributes
nothing to the integral.
In the same way that
coo
was proved
r\
continuous.
Dividing
it
into
two parts
finite, it
V and
-^
r)
may be proved
--=
of which the
76, 77]
second
please
is
continuous,
and hence
r)Vl "
we
as small as
7\V '
.
ox
Then we can make -ôx
dV dV
we
small as
please, or
we can make
the increment of
please,
and
dV
- -
as
ox
continuous.
ox
Poisson's Equation.
77.
we
Hence by taking
is
-^
as small as
^-.
also the difference -=
near enough together,
ox
be -^- and
ox
dV
as shown, -^
by making the sphere at P small enough. At a neighbourP draw a small sphere, and let the corresponding parts
ing point
orr/
we may make,
151
By
Gauss's theorem
39
(5)),
we
have
cos (nr)
a.s
dS = -
when r
is drawn from 0, a
point within
mass concentrated at 0,
4-7T,
S.
Multiplying by
ra,
(i)
The
integral
the surface integral of the outward normal component of the

parameter P, or of the inward component of the force.
is
The
surface integral of the normal component of force in the

S is called the flux of force into S, and
inward direction through
we
4?r times the element of mass within

equal to
Masses without contribute nothing to the integral. Every
see that
8.
mass
dm
it is
situated within
point and
contributes
to the potential at
4>7rdm to the flux through the surface 8.
whole mass, when potential
is
V=\\\
flux
-4nrM=and
tariff
pdr,
any
Hence the
contributes to the
152
Now
The
surface
is,
[PT.
I.
CH. IV.
by the divergence theorem, equal
to
(3)
S may
we take
providing that
in the space T within S.
be drawn inside the attracting mass,

due to matter
for the potential only that
Accordingly for r we
attracting mass, and
may
take any part whatever of the
(4)
As the above theorem applies to any field

we must have everywhere (by 23)
of integration what-
ever,
AF+4-7r/3
(5)
This
at
Poisson's extension of Laplace's equation,

point the second differential parameter of
any
4-7T
= 0.
times the density at that point.
= 0,
where p
bodies,
and says that
is
this
is equal to
Outside the attracting
becomes Laplace's equation.
In our nomenclature, the concentration of the potential at any

point
is
proportional to the density at that point.
more elementary proof of the same theorem may be given
At a point
as follows.
x,
y,
z construct
a small rectangular
parallelepiped whose faces have the coordinates

x x + %, y, y + 77, z, z+%,
}
and find the flux of force through its six faces. At the face
normal to the X-axis whose x coordinate is x let the mean value of
-- =
the force be
ox
The area
integral
Px
of the face
1
JJ
is
77
P cos (Pn) dS the
At the opposite
so that this face contributes to the
amount
^
ox
77^
- - is
continuous,
face, since
we have
value
^
dx
S^r
\-zr-}
dx\dxj
terms of higher order in
f,
for its
77, 78]
and hence, the normal being directed the other way,

tributes to the integral the
153
this side con-
amount
7
<.
a /a
and the two together

fi??-^-; 4-
terms of higher order.

Of> T7"
Similarly the faces perpendicular to F-axis contribute f^f
and the others
Hence the
--F
3
surface integral
is
F
and by Gauss's theorem
is
the
mean
this is equal to
4-Trm
where p
92
density in the parallelepiped.
the parallelepiped infinitely small, and dividing by
Abbreviations
78.
for
Operators.
Now making
77 f,
we get
If
p is any point
to a distribudue
function, the potential function at any point
tion through all space of matter whose density at any point is p
has been denoted by Gibbs and Heaviside by the abbreviation
Pot p, standing for the definite integral
P
-dr.
Potp-JJj
(The
suffix oo
We may
denotes integration through
all
space.)
thus abbreviate Poisson's equation
- A Pot p = V 2 Pot p = 47F/3,

=
so that the operation Pot followed by the operation
performed on any scalar function, has the effect only of multiply-
ing
it
by
4?r,
each other.
or the operations
and
-.
are the inverses of
154
[PT.
Characteristics of Potential Function.
79.
now found
CH. IV.
We
have
the following properties of the potential function.
It is
1st.
I.
everywhere holomorphic, that
is,
uniform,
finite,
continuous.
2nd.
Its first partial derivatives are
3rd.
Its second derivatives are finite.
4th.
V vanishes
everywhere holomorphic.
at infinity to the first order,
R=ao
dV
=.... vanish to second

ox
order,
OX
R=a
V satisfies
5th.
everywhere Poisson's differential equation

92
32
and outside of attracting matter, Laplace's equation
d*V
ox*
dy*
function having
Any
&V
all
92
F~
dz"
these properties
is
a Newtonian
all
points outside
potential function.
The
force JT, F,
is
a solenoidal vector at
of the attracting bodies, and hence if we construct tubes of force,

the flux of force is constant through any cross-section of a given
A tube for which the flux is unity will be called a unit
tube.
The conception of lines of force and of the solenoidal
tube.
property is due to Faraday.
Since
a harmonic function outside of the attracting

has neither maximum nor minimum in free space,
is
bodies,
it
but
maximum and minimum must
its
lie
within the attracting
bodies or at infinity.
AF=4?r/) says that

or
the
at,
divergence of the
proportional to the density at that
In the attracting bodies the equation

the concentration of the potential
force
point.
from any point
is
79, 80]
155
80.
Examples. Potential of a homogeneous Sphere.
Let the radius of the sphere be R, h the distance of P fi-oi,
center,
Let us put
s instead of r,
using the latter symbol for the polar
coordinate,
Now
s2
=A +r 2
2hr cos
6.
FIG. 34.
Differentiating, keeping r constant,
sds
and introducing
if
= hr sin 0d&,
s as variable instead of 0,
We must integrate
P is external
first
r to h
with respect to s from h
+ r,
h-r
- =M
3h
Hence the
the same as
if
attraction of a sphere upon an external point

the whole mass were concentrated at the center.
body having the property that the
is
line of direction of its
resultant attraction on a point passes always through a fixed point

in the body is called centrobaric.
If instead of a whole sphere we consider a spherical shell of
internal radius R^ and outer
the limits for r being
1} R?,
2
V-
f*!
m
156
We
IV
have
_M
an
dh~
[PT.
I.
CH. IV.
'
h*
2M
'
dh
hs
on the other hand,
If,
the limits for s are r

.
h,
r+ h
= 27rp-[**
-r1
Hence
independent of
dV =
^r-
= 47773
is
1,
h,
that
is, is
constant in the whole cavity.
an d we get the theorem that a homogeneous
0,
spherical shell exercises no force on a body within.

of symmetry the force can be only radial.)
If
h<R
[**
JRi-Jr-h
>
is
rdrds
which
in the spherical cavity,
is
r+h
in the substance of the shell,
we
(On account
divide the shell into
two by a spherical surface passing through P, find the potential

due to the part within P, and add it to that without, getting
V=
\M r
(h*
T:
J-^\
-w
"
rf/i
dh~~!T\W~
'
-L
Tabulating these results
3
'
7T
/r
80, 81]
157
Plotting the above results (Fig. 35) shows the continuity of
FIG. 35.
and its first derivatives and the discontinuity of the second

derivatives at the surfaces of the attracting mass.
We
it is
see that the attraction of a solid sphere at a point within

= 0,
proportional to the distance from the center, for if
dV_
4<7rph
dh~
7T~'
and
is
independent of the radius of the sphere. Hence experiments on the decrease of the force of gravity in mines at known
depths might give us the dimensions of the earth.

81.
Disc, Cylinder, Cone.
Let us find the attraction of a

on a line normal
circular disc of infinitesimal thickness at a point
FIG. 36.
158
Let the radius be
to the disc at its center.
distance of
[PT.
1}
I.
CH. IV.
thickness
6,
from the center h

'
/*
2n27r
/*
F=|
/o
trdrd<f>
IJB
cTF
iv/ yz
rf/t
Attraction of circular cylinder on point in its axis. Let the

length be I and let the point be external, at a distance h from
the center.
By the above, a disc of thickness dx at a distance x from the

center produces a potential at
Hence the whole
is
rU2
J -Z/2
log
(x-
Circular cone on point in axis.
Let
^ be the radius
of base, a the altitude, h the height of
above the vertex.
disc at distance
x below vertex and radius r causes potential
at P,
dV= 2irpdx {V(A + xj + r - (h +

2
a?)}
81, 82]
and
= 27T/9
r
-
=E
159
R
a
x,
J(
If we have a conical mountain of uniform density on the earth,

and determine the force of gravity at its summit and at the sea
level, this gives us the ratio of the attraction of the sphere and
cone to that of the sphere alone, and from this we get the ratio of
the mass of the earth to the mass of the mountain.
Such a determination was carried out by Mendenhall, on Fujiyama, Japan, in
1880, giving 5'77 for the earth's density.
FIG. 37.
Circular disc on point not on axis.

Let the coordinates of
with respect to the center be a, b, 0. Then
_ r cos
2ff
CR r
v_
J o J o
an
elliptic integral.
given in
82.
erdrd<f>
Va +
2
(b
r cos
</>)
The development
in
+r
an
'
sin2
</>
infinite series will
be
102.
Surface Distributions. In the case of the circular

surface
e, ep is the amount of matter per unit of
disc of thickness
of the disc.
It is often convenient to consider distributions of
matter over surfaces, in such a manner that though e be considered
infinitesimal p increases so that the product ep remains finite.
The product
tion
is
We
cr is called the surface density,

ep
called a surface distribution.
have
<rdS
[[<rd
I
JJ
and the
distribu-
160
In the case of the
When
disc,
[FT.
I.
CH. IV.
we had
we have
27TCT.
The repulsion of a disc upon a particle in contact with it at its

center is independent of the radius of the disc, and is equal to 2?r
times the surface density.
FIG. 38.
If the force on a particle in contact
positive
if to
the right,
F.2
By symmetry,
with the disc
on the right be called
we have
=
+ 27T<7.
the force on a particle at the
left in
contact
is
F = - 27TCT,
1
Now
if
x denote the direction of the normal
to the right,
*--,
--,.
and we see that on passing through the surface there
continuity in the value of
dV
-^
Consider a thin spherical
V=
(Ef - Rfl =
of the magnitude
shell.
We have
(A,- JZ.) (Bf
is
a dis-
47r<j.
for
an external point
82]
and making
and on the outside
^ = lim R =
e,
lim
for
= R,
JJj
161
JK,
dV
dh=
Within we have everywhere
const.,
dV
-~- = 0.
dfi
Hence there
is
in like
manner a discontinuity in the
first
derivative
of the potential in the direction of the normal, on passing through

the attracting surface, of the amount 4?ro-.
FIG.
now any surface distribution of surface density a.

Gauss's
theorem to a small tube of force bounded by portions
Apply
of two equipotential surfaces d^ and d% 2 on opposite sides of the
Consider
element of the attracting surface dS (Fig. 39).

the tubes
The
flux out from
is
F,d^-F,d^
and
this
the tube,
But
and
if
must be equal to 4- 4?r times the matter contained

which is <rdS. Therefore
the length and diameter of the tube are infinitesimal
c?2 2 are the projections of dS,
d^ = d% = dS cos (Fri),
2
where n
is
the normal to the attracting surface.
F cos (Fn) dS - F
2
w.
E.
Accordingly
cos (Fn) dS =
11
in
THEOKY OF NEWTONIAN FORCES.
162
and since
The normal
to
is
cos (Fn)
[PT.
I.
CH. IV.
here drawn toward the side
2.
We
find
then that in general, on traversing a repelling surface distribution,

the normal force has a discontinuity equal to 4?ro-.
This
is
Poisson's equation for a surface distribution.
draw the normal away from the surface on each
side,
If we
we may
write
dv
=-+-dv
on
ont
or
F! cos
83.
two
(F^) +
jF2 cos
4-7TO-,
(F2n 2 ) =
FMl + F =
21la
4-Tro-.
Green's formulae.
functions, of
Let us apply Green's theorem to

which one, V, is the potential function due to
any distribution of matter, and the other,
where r
is
the
distance from a fixed point P, lying in the space r over which we

take the integral. Let the space T concerned be that bounded by
a closed surface $, a small sphere S of radius e about P, and, if
is
without $, a sphere of infinite radius with center P.
Now
in
33 (2) for the normal drawn

which means outward from $ and 2, and inward
the theorem was stated in
toward
from the
and since
r,
infinite sphere, as
82, 83]
in the
The
whole space
r,
so that (i)
163
becomes
surface integrals are to be taken over S, over the small sphere,

infinite sphere.
For a sphere with center at P,
and over the
a*
the upper or lower sign being taken according as the sphere

the inner or outer boundary of r
is
and
for
V vanishes,
Now
still
oo
hence this integral vanishes.
9F
at infinity,
vanishes.
1
is
of order
Also
and being multiplied by
r,
Accordingly the infinite sphere contributes nothing.
For the small sphere the case
becomes, as the radius
e of
is different.
The
first
integral
the sphere diminishes,
-V
(4)
The second part

e
however, since
-----
is finite
in the sphere, vanishes with
there remain on the left side of the equation only

obtain therefore
integral over S.
e.
4-TrFp
Hence
and the
We
.1
II
JJ
(s)
rrl^m^V
on
r on/
F
'=i/LI fe"-M ds
JJJ
-M/^ ^112
T
uteide *)'
164
the normal being drawn outward from S.

Green.
Hence we
[PT.
This formula
see that any function which
is
I.
CH. IV.
is
due to
uniform and con-
tinuous everywhere outside of a certain closed surface, vanishes

at infinity to the first order, and whose parameter vanishes at
is determined at
every point of space
at every point of that space the value of
the second differential parameter, and in addition the values on the
second order,
infinity to the
considered
surface
we know
if
of the function
and
its
vector parameter resolved in the
direction of the outer normal.
In particular,
have
if
harmonic in
is
all
the space considered,
we
(6)
and a harmonic function is determined everywhere by its values

and those of its normal component of parameter at all points of
the surface S.
Since
1
~" ~
r
1 or
r*
dn
If cos
=__
(M)
x
we may
_ + COS (ny)\3r + COS (,) 3H

^
dr
cos (nr)
write (6)
Consequently, we may produce at all points outside of a closed

surface S the same field of force as is produced by any distribution
of masses lying inside of S, whose potential is V if we distribute
t
over the surface
a surface distribution of surface-density,

1
--~
(Fcos(r) f
"S.\
In the general expression (5), the surface integral representing

the potential due to the masses ivithin S, the volume integral
JL
1
^LL dr
83, 84]
165
represents, since everywhere
pdr
that
is,
the potential due to
all
the masses in the region
T, viz.,
outside $.
84.
Equipotential Layers. As a still more particular case
of (7), if the surface S is taken as one of the equipotential
surfaces of the internal distribution, we have all over the surface
Vs = const., and
the constant
may be taken
integral,
Now
out from the
first
'
by Gauss's theorem
- -~
dS = 0;
accordingly,
*
so that
VP is represented as the potential
of a surface distribution
of surface-density
-
4t7r
The whole mass
dn
4?r
4?r
of the equivalent surface distribution
which, being the flux of force outward from S,

theorem, 77 (i), equal to M, the mass within S.
Accordingly we
Gauss*
may enunciate
is
is
by Gauss's
the theorem, due to Chasles and
We may produce
tribution
M the same
Chasles,
outside any equipotential surface of a diseffect as the distribution itself produces, by
" Sur 1'attraction d'une couche

ellipsoidale infiniment mince." Journ.
36.
Gauss, Allgemeine Lehrsatze,
p. 266, 1837
EC. Polytec., Cahier 25,
166
I.
[PT.
CH. IV.
distributing over that surface a layer of surface-density equal to
4f7T
times the outward force at every point of the surface. The mass
of the whole layer will be precisely that of the original internal
Such a layer
called an equipotential layer.

which
coincides with one of its
superficial layer
own equipotential surfaces.) Reversing the sign of this density
will give us a layer which will, outside, neutralize the effect of the
distribution.
(Definition
is
bodies within.
Let us now suppose the point P is within 8. In this case, we

apply Green's theorem to the space within S, and we do not have
the integrals over the infinite sphere. The normal in the above
formulae is now drawn inward from S, or if we still wish to use
the outward normal,
we change the
sign of the surface integral
in (5),
8
Tr ^
Vp = --r1 - M s f\V\ dn
4>7rJJ
(12)
3V
f/YAF
--dr,
r
,
r dne /ydtf--,47rJJj
(P
Note that both formulae (5) and (12) are
is drawn into the
space in which P lies.
inside S).
identical
if
the
normal
Hence within a
closed
surface
determined at every point solely by
a holomorphic function
its values and those of
is
its
normally resolved parameter at all points of the surface, and by

the values of its second parameter at all points in the space within
the surface.
harmonic function may be represented by a potential function of a surface distribution.
Now
if
the surface
is
equipotential, the function
cannot
be harmonic everywhere within unless it is constant. For since

two equipotential surfaces cannot cut each other, the potential
being a one-valued function, successive equipotential surfaces
must surround each other, and the innermost one, which is reduced
to a point, will
be a point of
maximum
or
minimum.
But we
have seen ( 34) that this is impossible for a harmonic function.

Accordingly a function constant on a closed surface and harmonic
within must be a constant.
and without S, the volume

denotes the potential due to the matter in the
If however there be matter within

integral, as before,
84, 85]
1G7
surface integral that due to the

space r (within S), and the
If the surface is equipotential, the surface
matter without.
integral
is
The
Vs
first
integral
is
now equal
r
to
4-Tr,
so that
being constant contributes nothing to the derivatives of V, so

may be replaced by a surface layer of
that the outside bodies
density
^ = T4-7T dn
04)
The mass
of the surface distribution
ne being the outward normal, is the inward flux of force through

8, which is equal to minus the mass of the interior matter, and is
not, as in the former case, equal to the
mass which
it
Potential completely determined by
85.
replaces.
its
charac-
We
have proved that the potential function

teristic properties.
due to any volume distribution has the following properties
:
It
1.
finite,
is,
together with
its first differential
parameter, uniform,
and continuous.
2.
It vanishes to the first order at oo
and
its
parameter to
the second order.

3.
It is
harmonic outside the attracting bodies, and in them
satisfies
The preceding
these properties
termined.
is
shows that a function having

a potential function, and is completely de-
investigation
168
[PT.
I.
CH. IV.
For we may apply the above formula (5) to all space, and then
the only surface integral being that due to the infinite sphere,
which vanishes, we have
*If however, the above conditions are fulfilled
except that at certain surfaces
its first
by a function V,
parameter
is
discon-
draw on each side of the surface S surfaces at

e from $, and exclude that portion of space
which we will call $j and $2
lying between these,
tinuous, let us
distances equal to
If the normals are
The
drawn
into
we have
surface integrals are to be taken over both surfaces 81
and
$ and
the volume integrals over all space except the thin layer
This is the only region where there is disconbetween Si and $2
theorem applies, and
r
the
in
hence
tinuity,
2
r ,_\
ff
>
s\
8F
vi "i
let
us
make
approach each other and
same on both
13F
f/7
-5^1-1JJs rdn
-^-dS- JJJ^
JJ Sl r GUI
Now
infinitesimal,
is
sides, hence, since
first
two terms destroy each
next two, for x

GUI
is
not equal to
dn
-dr.
then the surfaces
continuous at S, that
(-)
dni \rj
the
Oi
j j S%
rr
other.
Si,
is, is
the
^
(- ) in the limit
dn 2 \rj
This is not so for the
,
because of the discontinuity.
85, 86]
In the
169
then
limit,
The volume
integral, as before, denotes the potential
due to the volume
III- dr
distribution, while the surface integral denotes
the potential of a surface distribution
where
-+
1
(dV
Hence we get a new proof
- ,
||
dV
surface
of Poisson's
condition,
82.
86.
consider
Kelvin and Dirichlet's Principle.

a question known on the continent
We
of
shall
now
Europe as
Dirichlet's Problem.
Given the values of a function at
is it
on the
and
all
points of a closed surface

assuming these values
possible to find a function which,
surface,
is itself
is,
with
its
harmonic at
parameters, uniform,
points within S ?
finite,
continuous,
all
the external may be stated in

is the internal problem
manner, specifying the conditions as to vanishing at infinity.
This
like
Consider the integral
of a function
J
It
u throughout the space r within
must be
cannot
positive, for every
vanish,
= constant.
But
S, this will not
unless
since
be the
everywhere
is
is
- = 5-
dx
sum
of squares.
= ~ = 0,
dy
continuous, unless
that
is
dz
it is
constant on
case.
Accordingly
Now
element
8.
J (u) > 0.
infinite variety of functions u there must be,

less than for
to
Dirichlet, at least one which makes
according
difference
the
call
any of the others. Call this function v, and
of the
between
this
and any other
hs, so
u
h being constant.
that
=v+
hs,
170
The
condition for a
minimum
is
J(v) < J(v

for all
values of
[PT.
I.
CH. IV.
that
+ hs),
h.
du
XT
Now
^-
dv
ds
= ^+ô-,
.
etc.
+ n ^*+**+*fr.
02
00
dy 9y
of)
Integrating
Now
in order that J(v)

7
T- /
h 2 J(s)
(4)
-v7
fff
2/i
J JJ
may
be a minimum, we must have
ds
fdv
L
dv ds
dv ds\
+ =- 5- +
9^J
dz
~-
vto
limited,
term in h
c?r
dzj
But
for all values of h, positive or negative.
we may take h
dy dy
> 0,
as s is as yet un-
so small that the absolute value of the
greater than that of the term in

the sign of h opposite to that of the integral,
negative, the whole will be negative.
is
/t
2
,
and
if
making
we choose
the product
to leave the sum always positive is to have the

vanish.
(It will be observed that the above is exactly
integral
the process of the calculus of variations.
might put $v instead
The only way
We
of hs.)
The condition
for
minimum
[[fidvds
]}} (9# dx
But by Green's theorem,
Now
is
then
dv ds
dv ds\
dy dy
dz dz)
this is equal to
at the surface the function
have the same values, and
5 = 0.
is
given, hence u
and
must
86, 87]
17 1
Consequently the surface integral vanishes, and
///
But
since s is arbitrary, the integral can vanish only if every= 0, v is therefore the function which solves the
T, AV
where in
The proof
problem.
that there
is
such a function depends on the
assumption that there is a function which makes the integral J a

minimum. This assumption has been declared by Weierstrass,
Kronecker, and others, to be faulty. The principle of Lord Kelvin
and
which declares that there
Dirichlet,
a function
is
v,
has been
a number of special cases, but the above general

rigidly proved
is
no
proof
longer admitted. It is given here on account of its
for
histoiical interest*.
We
can however prove that if there
the conditions,
it is
For,
unique.
On
= v', u = 0.
0.
Accordingly J(u) = 0.
= const. But on
only be if u
and v = v'.
87.
Theorem
Green's
ordinates.
We
shall
now
is
In
r,
satisfying
Av and At;' are

we have shown,
since
But, as
S,
u=
0,
hence, throughout
Curvilinear Orthogonal Co-
in
consider Green's theorem in terms of
any orthogonal coordinates, limiting ourselves
U = const.,
v,
another, v, put
v.
the surface, since v
this can
}
=v
a function
is
there
A^t =
zero,
if
or the divergence theorem,
to the special case
35,
//-///"*
where n8
*
is
the outward normal to
Thomson, "Theorems with
8.
reference
to
the
solution
of
certain
Partial
Differential Equations," Cambridge and Dublin Math. Journ., Jan. 1878; Eeprint
of "Papers in Electrostatics and Magnetism," xm.
The name Dirichlet's Princip
was given by Eiemann

this
matter the student
theorie, as well as
Traite
(V Analyse,
Werke, p. 90).
For a
may consult Bacharach,
historical
and
critical discussion of
Abriss der Geschichte der Potential-
Harkness and Morley, Theory of Functions, Chap,
Tom.
u., p. 38.
rx.,
Picard,
172
Let the coordinates be ql}
The parameter
is
q^,
q3
[PT.
I.
CH. IV.
the resultant of the derivatives of
Fin
any three perpendicular directions. Let these be in the directions

of the normals to the level surfaces g 1? q 2 q3
,
Then, calling these
P P P
,
ft
fe
q3
~=Pcos(Pn )
(1)
Now P
dV
qi
i5
qi
cos (wjw,)
+ Pq
cos
(ntfig)
+ P q3 cos (n n
3
s ).
the partial parameter with respect to q lt
is
16)
9gi
Hence
(2)
8F =
^
ons
3F
/ii
8F
+ A ^ cos
,
cos
oqi
(fljn,)
3F
^s ) + /^3 ^
7
(?i 2
cos (n Bns ).
oqs
oq<>
If we divide the volume r up into elementary curved prisms

bounded by level surfaces of q 2 and q3 as in the case of rectangular
,
FIG. 42.
coordinates,
we
dScos (n^)
have, at each case of cutting into or out of S,

1} where dS^ is the area of the part cut by the
= dS
prism from the level surface q^
By
20,
accordingly
87, 88]
Transforming the other two integrals in
like
8F
cos (n#i 8 )
SV\
Md UK V
f h,
a&
Now
this is equal to
dV\
/ h,
+h
/ h,
WK a*' ^ U
1
173
manner,
8F
cos (n3n g )
\-
dS
dV
A Vdr.
But
Multiplying and dividing in the last integral of (4) by hjijt 3>
we find that, since the integrals are equal for any volume, the
integrands must be equal, or
3AA
9?/
8ft
*. a?,
3?3
3g,
This result was given by Lam^, by means of a laborious direct

The method here used is similar to one used by
transformation.
Jacobi,
and
is
given by Somoff*.
88.
Laplace's Equation in Spherical and Cylindrical
Coordinates. Applying this to spherical coordinates
hr
AF=
(6)
8 (
~-l-fr2 sinl9
9r/
_~
28F
We may apply
dr
~
+ r*~dffi
Lam6, Journal de
dd\
'
this equation to
nd
"\
(9
~a7sin
sm^-^
For external points

sphere.
independent of 6 and <,
les
T Sill D
rsin<9'
W)
_
^r (-^
d<f>\sm
r2 sin 2 6
tiellen Differentialgleichung
AF=0,"
Theoretische Mechanik, n. Theil,
51
'
and since by symmetry Vis
Lecons sur
"Ueber eine particulare Losuug der par-
Crelle's
2.
J)
determine the attraction of a
A 1^=0,
Jacobi,
d</>
9^>
I'Ecole Polyte.chnique, Cahier 23, p. 215, 1833
Coordonnees curvilignes, n.
dV
TJ-^T- }}
Journal, Bd. 36, p. 113.
Somoff,
TFÔRY OF NEWTONIAN FORCES.
174
(7)
dr2
I.
CH. IV.
TrTr
dr\ dr
rdr
"2
dr ~r*'
dr~''
r=-; +
But
[PT.
c>.
Lim (rV) = M,
since
r=ao
Lim
+ c'r] = M,
r=ao
we must have
c'
= 0,
M.
Apply the above transformation

/&2
=
If
dz2
= 1,
Ap
= l,
h (a
18F
8p
to cylindrical coordinates
=-
p dp
8ft>
we apply
homogeneous
this to calculate the potential due to a cylindrical

body with generators parallel to the axis of z and
of infinite length, the potential
independent of z and
is
satisfies at
external points,
_8*F
18F
p dp
8/o
8o)
is independent of
If the cylinder is circular,
the ordinary differential equation
dp
p dp
log
dV = -
-T-
dp
log p
const.
V=C\ogp+C'.
o>,
and we have
NEWTONIAN POTENTIAL FUNCÔN.
88, 89]
175
The
force in the direction of p is inverse y proportional to the

first power of p.
We may verify
this
direct calculation.
by
Let us consider the
FIG. 43.
cylinder as infinitely thin, with cross-section

component of force in the direction of p.
The
action of
dm
at z on
P at
The component
will find the
is
8"
We
distance p (Fig. 43)
dm
Vp +
dm
r.
parallel to p is
dm
= pcm
cos(ps)
^~.
Now
since, calling the density S,

total force in direction p
dm =
-&dz,
we have
for the
'
Svrpdz
.p=
= p tan ^,
dz = psee 0d0.
Put
=
P
89.
(7
as before.
'
Logarithmic Potential.
We may
state
the above
result in terms of the

Supfollowing two-dimensional problem.
that
on
a
there
be distributed a layer of mass in such
pose
plane
way
that a point of mass
repels a point of unit mass
in
176
the plane with a force -- where r

potential due to
m is V
m log r
[PT.
I.
CH. IV.
their distance apart.
is
and
it satisfies
The
the differential
equation
any mass distributed in the plane,

an element dm = /jidS produces the po-
Similarly, in the case of
with surface-density
dm log r,
tential
V=
where r
JJL,
and the whole the potential

\\drn log r
M//,
log rdS,
the distance from the repelling

repelled point a, y, so that
is
r2
= (x -
We may verify by
points, this V satisfies
direct
!?
a y2
=
r- 1 1* log rdadb
r
(y
- 6)
at
a, b
to the
2
.
differentiation
- Up
dm
that,
at
external
(log r) <*<
x
l
2(-a
9y
called the logarithmic potential and is of

of functions of a complex variable.
great importance in the theory
This potential
is
Green's Theorem for a Plane. In exactly the same

manner that we proved Green's Theorem for three dimensions,
we may prove it when the integral is the double integral in a
90.
plane
<>
89
91]
over an area
for
bounded by any closed contour
~dxdy=W*- Wl+
\
where n
is
Since we have
a continuous function
(2)
C.
177
...
W cos (nx)
Wm - Wm_
dy
ds,
the inward normal, ds the element of arc of the
contour.
Applying
this to
W= U
we obtain
0tH3Treating the other term
Interchanging
cV
dn
in like
U and V we
dn
manner, we obtain
obtain the second form
JJ A
where we write
Application to Logarithmic Potential.
91.
second form above
which
is
function
we put U=l, we
If in the
obtain
If the
the divergence theorem in two dimensions.
harmonic everywhere within the contour, we have
V is
,7
ds
A
= 0.
c on
FIG. 44.
W.
E.
12
178
*7
which r
r
3 log
CJ _
Jc
fa
is
7
/r O ^-
I.
CH. IV.
harmonic function logr, where P, the

measured, is outside the contour,
this to the
Applying
fixed pole from
[PT.
dr
cos (rn)
7
/y o
'
Jcrdn
J c
,
sy o
(1
is within the contour, we draw a circle

If the pole
of any
radius about the pole, and apply the theorem to the area outside
of this circle and within the contour, obtaining
(8)
3 log
& r
Jc
dn
These two
cts
=-
cos (rn)
r
JK
results are
c?5
=-
2w
/*
'
c?^
2?r.
Gauss's theorem for two dimensions.
We
They may of course be deduced geometrically.

may
deduce Poisson's equation for the logarithmic potential as in
Newtonian
for the
a mass
dm
being
77
The logarithmic potential due to

gives rise to the flux of force Zirdm
Potential.
dm log r
outward through any closed contour surrounding

mass m causes the flux
2-Trm
Put
now
= 2?r
it,
and a
total
pdxdy.
in terms of the potential this is
(9)
and since
this is true for
any area of the plane, we must have
AF=-2wv&.
(10)
This
is
92.
Green's formula for Logarithmic Potential. Apply-
Poisson's equation for the logarithmic potential.
logr and any harmonic

ing Green's Theorem (5) to the functions
the contour, and
the
of
to
be
within
function F, supposing
pole
extending the integral to the area within the contour and without
a circle
K of radius
dV
The
third term
is
about the pole,
91
93]
(since
V is
harmonic in
on
K
which, when we make e
K)
179
and the fourth
K r
decrease indefinitely, becomes
2-TrFp.
Accordingly we obtain the equation
Vp =
T^
which
In a similar way
83.
(6),
theorem on the Newtonian Potential
a corresponding theorem for the Logarithmic Potential. A comparison of the corresponding theorems will be found in C. Neumann's work, Untersuchungen uber das logarithmische und das
is
we may
the analogue of equation
find for nearly every
Newton sche Potential*.
The Kelvin-Dirichlet Problem and

and demonstrated
Principle
may be
stated
for the logarithmic potential precisely as in
86.
93.
Dirichlet's
Series.
We
Problem
for a Circle.
Trigonometric
a homogeneous harmonic function of

order n of the coordinates x, y of a point in a plane a Circular
n
Harmonic, since it is equal to p multiplied by a homogeneous
shall
function of cos
o>
call
and
sin
and consequently on the circum-
&>,
ference of a circle about the origin

function of the angular coordinate &>.
Vn of degree n satisfies the
so that a circular
an
simply a trigonometric
Any homogeneous
a solution of this and Laplace's
is
The homogeneous function
+ an^x
function
differential equation
harmonic
Equation simultaneously.
xn
is
n-^
......
câyf^
+a
of degree n
+1
terms, the sum of its second derivatives is a homofunction

of degree n
1 terms, and if this
2 containing n
geneous
is to vanish
1 coefficients must vanish,
identically each of its n
contains n
consequently there are n

efficients of
*
tiates
Vnj
or only
relations
two are
between the n +
arbitrary.
Accordingly
all
co-
har-
See also Harnack, Die Grundlagen der Theorie des logarithmischen PotenPicard, Traite d' Analyse, torn. n.
;
122
180
[FT.
I.
CH. IV.
monies of degree n can be expressed in terms of two independent

We have found in 44 that the real and imaginary parts
ones.
n are harmonic
functions of x, y, being
of the function (x + iy)
respectively equal to
n
n
p cos nco and p sin nco.
Accordingly the general harmonic of degree n
is
Vn = p n {A n cos no) + Bn sin nco} = pn Tn
(2)
We may call the trigonometric factor Tn which is the value of

the harmonic on the circumference of a circle of radius unity, the
,
peripheral harmonic of degree

If a function which
an
developed in
"P"
(3)
>
n.
harmonic in a circular area can be
is
infinite trigonometric series

W=<
% {A n cos nco
n sin nco}
y)
=
+B
oo
= ^Tn
on the circumference of the

Dirichlet's
Problem
for
circle of radius R, the solution of

the interior of the circle is given by the
series
F=2' +
(4)
|2'
+ ^r + ....
2
harmonic, and therefore the series, if conAt the circumference p = R, and the series
vergent, is harmonic.
takes the given values of V. The absolute value of every term
For every term
is less
is
than the absolute value of the corresponding term in the
n
n
series (3), in virtue of the factor p jR
converges, the series (4) does as well.

conditions,
by
therefore if the series (3)
Since the series
fulfils all
Dirichlet's principle it is the only function satis-
fying them.
We may
fulfil
negative degree.
the outer problem by means of harmonics

Taking n negative, the series
V=T +
(5)
^T1+ ^T
of"
+...
convergent, takes the required values on the circumference, and

vanishes at infinity except the constant term. For a ring-shaped
area between two concentric circles, we may satisfy the conditions
is
by a
series in
both positive and negative harmonics,
oo
(6)
V= 2/o n
{A n cos
nco
+ Bn sin
nco}
o
oo
+ S/o~ w
{A n cos
nco
Bn
sin nco}.
93, 94]
94.
181
We
Development in Circular Harmonics.
may
use
to obtain
the developformula (12),

92,
of a function in a trigonometric series on
the circumference of a circle. Let the polar cothe
ment
ordinates of a point on the circumference of the

within the circumcircle be R, &> and of a point
ference p, <j>.
Then we have for the distance be-
FIG. 45.
tween the two points

r
[R
+ p* - 2Rp cos (to - </>)]^.
z
Removing the factor R inserting for cos (&> p)
exponentials, and separating into factors we obtain
,
(7)
its
value in
Taking the logarithm we may develop
and
by
log
Taylor's
(8)
log r
l~r-* "
Theorem, obtaining
= log R - i 2
^J
(e
ni <"-*>
+ e~ni
= log R
This series
is
(w
-*>)
7i _Zt
convergent
if
<
71
cos
(o>
if
= R,
-j^n
R, and also
</>).
unless
co =
<f).
Inserting this value of logr in (12), differentiation with reR, we have

spect to the normal being according to
182
Expanding the
cosines,
I.
CH. IV.
we may take out from each term
of the
[PT.
n
n
integral, except the first, a factor p cos n(p> or p sin n<f>, so that VP
is developed as a function of its coordinates
p, $, in an infinite
which are definite

around
the
the
circumference, involving
integrals
peripheral values
This does not establish the convergence of the
of V and 3 V/dn.
series of circular harmonics, the coefficients of
series
on the circumference.
Admitting the
development, we may proceed

form.
In order to do this
function
Vm
which
the circumference
possibility of the
to find it in a
more convenient
us apply the last equation to a

a circular harmonic of degree m. Then at
is
let
we have
m_
v
~T~
mlftm-i
on
and
rZ
/Jo
Tm cosn(w-<l>)da>.
The expression on the right is an infinite series in powers of /o,

As this equality must hold for all
while
m (P) is simply p m Tm
values of p less than jR, the coefficient of every power of p except
the
mth must
vanish,
and we have the important equations
CZn
(11)
Jo
Tm cos n (co
1 C 2n
Tm (</>)=-
(12)
for all values of n,
constant,
we
and
$)
dco
= 0,
m=n,
Tm (w) cos m (co -
for all values of
</>)
m except 0.
dco,
Since
is
evidently have
m
*
1
x
C 2ir '
'
c~
These two important results can be very simply deduced by direct

integration, inserting the value of Tm (&>), but we have preferred
to deduce them as a consequence of Green's formula (12),
92, in
order to show the analogy with Spherical Harmonics. Let us
now suppose
that the function
F(o>) can be developed in the
convergent infinite trigonometric series

00
F(o>)
00
= S (A n cos nco + Bn sin nco) = 2!Fn (co).
94, 95]
Multiply both sides by cos
and integrate from
dco
<)
183
tO 27T.
r
oo
[Sir
(J
3)
Jo
F(o>) cos ra
(o>
</>)
do>
-ZTT
= 2)
Jo
Tn (w) cos m (co -
<f>)
dw.
Every term on the right vanishes except the mth which is equal to
7rTm (<t>). Accordingly we find for the circular harmonic Tm the
definite integral
rm (<) =
(14)
m = 0, we
For
Writing
7T
fV(a>)cosra(a>-6)da>.
Jo
must divide by
for
Tm
(<j))
its
2.
value
A m cos m<l> + Bm sin m(f>,

expanding the cosine in the integral, and writing the two terms
separately,
(15)
^-0
we obtain the
= ^1
-67T
coefficients
2jr
/*
Jo
V((o)dco,
Am =
2jr
/"
TJ
1
=:
f
I
w Jo
This form for the coefficients was given by Fourier*, who assuming
that the development was possible, was able to determine the
The question of proving that the development thus
coefficients.
found actually represents the function, and the determination of

the conditions that the development shall be possible, formed one
of the most important mathematical questions of this century,
which was first satisfactorily treated by Dirichletf. For the full

and rigid treatment of this important subject, the student should
consult Riemann, Partielle Differentialgleichungen; Picard,
d Analyse,
torn. 1,
TraiU
chap. ix.J
Spherical Harmonics.
Spherical Harmonic of degree
defined as a homogeneous harmonic function of the coordinates
x, y, z of a point in space, that is as a solution of the simultaneous
95.
is
equations
*
t
Bd.
Fourier, Theorie analytique de la Chaleur, Chap, ix., 1822.

"
" Sur la
Dirichlet,
Convergence des Series Trigonome"triques, Crelle's Journal,
4,
1829.
resume" of the literature
matiques, 1880.
is
given by Sachse, Bulletin des Sciences Mathe-
184
82
82
F=
^+8^+^
(I)
dV
x-^dx
(2)
+ 9y
j, o
y + an _
+ an _
+2+3
contains 1
+?z
of its second derivatives
is
CH. IV.
'
+z-^-=nV.
dz
The general homogeneous function

xn ~ l
I.
8F
8F
dy
[PT.
of degree n
#n
2>
+a
-y
0)0
2)1
+ l = (?i+l)(n +
a homogeneous
2)/2 terms. The sum

2
function of degree n
and accordingly contains (n
If the function is to
1) w/2 terms.
must all vanish, so
coefficients
vanish identically, these (n
1) n/2
the
relations
that there are (n
(n + 1) (w + 2)/2
among
1) w/2
coefficients
of
a harmonic of the nth degree, leaving 2n
+1
The general harmonic
arbitrary coefficients.
accordingly be expressed
pendent harmonics.
of degree n can
of 2n + 1 indefunction
as a linear
EXAMPLES. Differentiating the arbitrary homogeneous function,

and determining the coefficients, we find for n = 0, 1, 2, 3, the
following independent harmonics
n=
constant
n=l
x,
n=2
n=3
y,
2
x*
2/
3#
i/-2/
z* }
3^- z\
xy,
zx
yz,
3^ - x\
3^ 3z2x
If
we
insert spherical coordinates
x = r sin 6 cos
y = r sin sin
2 = r cos 6
the harmonic
Fn becomes
r, 0,
</>,
<,
<,
z*,
-X
s
,
3zzy
-y
s
,
xyz.
95]
185
Yn is
where
a homogeneous function of the trigonometric functions

and sin 6 sin <.
n being the value of
</>,
n on the
surface of a sphere of unit radius, is called a surface harmonic.
cos
6,
sin 6 cos
The equation n = represents a cone of order n, whose intersection with the sphere gives a geometrical representation of the
harmonic
Vn
If u and v be any two continuous functions of x, y,
/
N
A
A
= ukv
AA (uv)
+ v&u
+2
(3 )
Put
u = rm
du dv\
dz dzj
and since
9 (rm J}
2
we
du dv
/3w 3fl
;-+
\dx dx
dy dy
z,
= mrm~ 2 + m(m-
get
A (rm) = 3mrm- + m (m 2
(4)
m
2) r
~4
(a?
= w (ra + 1) rm ~
2
.
yn is a harmonic of degree
A (rm Fn = rm A Vn + m(m+ 1) rm- Fn
(5)
If
?i,
= [m (m + 1) + 2m^]
by virtue of equations
Consequently
if
ra =
(i)
and
-(2rc
rm
~2
Fn
(2).
+ l),
the product rm Fn is a harmonic.

of degree unity in the coordinates,
Fn is of degree n, and r is
yn s O f degree (n + I). Accordingly to any spherical
harmonic Vn = rn Yn of degree n there corresponds another,
Since
r -(2n+i)
(n+D
of degree - (n + 1). Compare this with the corresponding property

of circular harmonics, where the degrees of the two corresponding
harmonics are equal and opposite.
186
96.
Dirichlet's
Problem
for Sphere.
[PT.
By means
I.
CH. IV.
of these
harmonics we
may solve Dirichlet's problem for the sphere. If a

function harmonic within a sphere of radius
can be developed
at the surface in
an
infinite series of surface harmonics,
F=F, + F + F
(6)
the internal problem
is
solved
by the
F=F + r1+
(7)
For each term

is
is
harmonic.
vergent,
values of V.
series
,r +
a
harmonic, and therefore the series (7), if conAt the surface the series takes the given
Every term of the series (7) is less than the corren

n
sponding term of the series (6) in virtue of the factor r jR
therefore if the series (6) converges, the series (7) does as well.
Since the series fulfils all the conditions it is the only solution.
We may in
like
manner
fulfil
the outer problem by the series
of harmonics of negative degree, which vanish at infinity.
For the space bounded by two concentric spheres, we must use

the series in positive and negative degrees, as will be illustrated by
an example in 198.
Forms of Spherical Harmonics. Before considering

97.
the question of development in spherical harmonics, we will
Since if
briefly consider some convenient forms.
we have
and any derivative of a harmonic
is itself
a harmonic of degree n
(a + ft + 7).
=
the
harmonic
F_
x
c/r, we have
sponds
is
a?
fi\ " T , -
a harmonic, so that
Since to
F =c
corre-
96, 97]
If
/*j
187
be any constant direction whose direction cosines are

cos (&!#)
l
A
f
and
~r ( SIT
I
is
)
=l
=m
cos(A 1 2/)
lt
a harmonic of degree
cos
(h-^
and to
2,
it
corresponds the
harmonic,
(IO)
.
which
is
Since ^ 2 +
o
of the first degree.
mf + n?
1,
the harmonic
contains two arbitrary constants, and multiplying by a third, A,

we have the general harmonic of degree 1, in the form
If in like
cosines
Z2 ,
ra2
manner /i 2 /i 3 ......
n2 ...... ^n ^r., ftn,
,
/&
n denote vectors with direction
dhi dh z
is
dh n \r,
a spherical harmonic of degree
-V"n -r2n+1 31
d2\
L
^J
'
(n
or.
G/II o/i 2
+ 1)
3Z,
o/i n
and
to
it
corresponds
(-}/'
\r J
and since every h introduces two arbitrary

constants, multiplying by another, A, gives us 2n -f 1, and we
have the general harmonic of degree n in the form,
a harmonic of degree
fl3^
n,
=A
"
dfh dh2
The
directions h^, h 2
To
monic.
'"dhn
hn are called the axes of the har-
illustrate the
method
of deriving the harmonics
shall find the first two.
j)
an
= Ar5 [I
\
/i
+ m d^
^
dx
l
fH^-J ft 5dx
clz) \
+w 25d-
+i5-J
dzJ
(\r
we
188
[PT.
I.
CH. IV.
"5
"i-
\]
)]*
F =A 2
+ 3 (Ijltf? + mî^y 4- n^n^ 4- (^m +

+ (mjn + maWj) yz + (?^^ + r^y zx)}.
2
The
coefficients are of course subject to the relations
98.
coincide,
Zonal Harmonics.
we may conveniently
If all the axes of the harmonic
take the axis for one of the coor-
dinate axes, and write
It is evident that this will contain only powers of z and

that the surface harmonic will be simply a polynomial in
zjr
The equation
Yn (cos (rz)) =
r,
so
= cos (rz).
may be shown
to have
real roots
1, and hence represents n circular cones of

lying between 1 and
angles whose cosines are these roots, intersecting the surface of a
sphere in n parallels of latitude which divide the surface into
The harmonics are therefore called Zonal Harmonics.

The polynomial in cos (rz) which constitutes the zonal surface
harmonic, when the value of the constant A is determined in the
manner to be shortly given, is called a Legendre's Polynomial, and
zones.
denoted by
99.
Harmonics
transformed
88,
and
AV
Laplace's
in
Spherical
Operator
into
Coordinates.
spherical
becomes
1
We
have
coordinates
dV
in
97
100]
If
we put
Vn
in this
rn
Yn
we
189
obtain
sin^ B ( n + l)Fn+
(16)
as the
equation satisfied by a surface harmonic

the form of Laplace's equation originally given
n is the zonal harmonic
n which is independent
differential
This
Yn (^, <)
by Laplace*. If
of <, we have
is
(+'"
(i 7 )
or putting
cos
(18)
is known as
Legendre's Differential Equation. We shall
without
now,
considering more in detail the general surface harfind
the
monic,
general expression for the zonal harmonic. It
may be at once shown, by inserting for n (//,) a power-series in p
This
and determining the
coefficients, that for integral values of
n the
by a polynomial in /z. The form

shall find from one of their important
differential equation is satisfied
of these polynomials
we
properties.
We
Development of Reciprocal Distance.
100.
know
that l/r, the reciprocal of the distance of the point x, y, z from

any fixed point P, is a harmonic function of the coordinates
x, y, z,
and although
the fixed point

series of
is
it is
homogeneous
harmonics.
We
not a homogeneous function except when

it may always be developed in a
the origin,
shall
functions, that
now
is,
use the letter
any fixed point, reserving r
in a series of spherical
for the distance from
for the distance
from the
origin.
Let
us for convenience take the axis of z as passing through the fixed

point P, which lies at a distance r from the origin, and put
cos (rz)
/Lt.
Then we have
/
-|
-i
d
Considering this as a function of z
let
us develop by Taylor's
Theorem,
(2o)i=/(,-r')=/(,) + (-/)(|) r = o + ^(-ry(g') r/==(>+

,
Laplace, "Theorie des attractions des sph6roides et de la figure des plane" tes."
de Paris. Anne~e 1782 (pub. 1785).
Mem. de VAcad.
190
and since
for r
(2I)
Now
-,
0,
ay =
r-^-
dn
-
[PT.
I.
CH. IV.
(i
-
+ (-Or
'as
w+1 we find
multiplying and dividing each term by r
,
(22)
rn
r2
where
This
Legendre,
r'
the determination of the constant
is
for the reason that, since
<r, and /A=
I_1
makes
for
adopted by
1,
1
IT
d~r\
it
by the binomial theorem, for
_ ^l" - Mi + jL + ri
"r
every
n,
p.(i)=i.
(23)
The term
Pn /rn+l is a
and the
series (22) is
/ > r we
find
In order to find
spherical harmonic of degree
convergent
for r'
Pn as a polynomial in
and develop by the binomial theorem.
Developing the
=o r=o
<
last factor
r.
/A
(n+l),
In like manner
we may
write r/d as
if
100, 101]
Picking out
the terms for which s
all
-f
191
n we get
the
for
-J
(r'\
1.3.5
......
n!
1.2(271-1)^
n(n-l)(-2)(n-3)
2*4(2n-l)(2n-8)
The
first
"J"
polynomials have the values
P 00
P (/,) = ! (5^2
101.
We may
Development in Spherical Harmonics.
83 for an internal point, to obtain the

use the formula (6)
of
a
function
of 6, <, on the surface of a sphere in
development
the same
manner
94
as in
for
the case of a
circle.
Since the
polynomials in the development of the reciprocal distance involve

only the cosine of the angle between the radii to the fixed and
variable points,
we have
if
<
r,
(22)
and differentiating
this with respect to
r,
the internal normal,
Inserting these values in (6),
83,
namely
F =and applying
it
to the case that

*
we
7/1
F is a spherical
7/1
harmonic
>
obtain, since
(29)
o
-P
Ym
o
\f
^ sin
192
P be
If the coordinates of
r', 6',
we have an
while on the right
we
</>'
[PT.
CH. IV.
I.
have,
infinite series in
powers of
r',
with
Since the equality must hold for

must have, collecting in terms in r' 8
definite integrals as coefficients.

all
values of r less than
Ym (0,
Tm (ff,
4,')
</>)
= m+ S + l
"
so that
(3i)
s (fji)
we have
/*"
sin
d0d<f>
s 4=
0,
m,
(30)
we
r,
Vm
(0,
<)
M ) sin 0d0d<j>,
= m,
for the values of the integral
Ym (6,
Pm
</>)
M)
sin
Y
j m (ff, f).
6 d0d4>
In performing the integration, we must put
for
//,
the value
obtained by spherical trigonometry,
p = cos (rr )
f
& + sin 6 sin & cos
cos 6 cos
(</>
<').
By means of the above integral expressions (30) and (31) we

may find the development of a function of 0, <, assuming that the
development
Suppose we are to find the development
is possible.
(32)
+r +r +
=F
/(0,t/>)
......
n (iJ,)siv.0d0d<l>, and integrate over

Multiply both sides by
the surface of the sphere, and since every term vanishes except the
Tith we obtain
"
f 71
(33)
f 27r
Jo Jo
f(9,
Pn 0*) sin
<
Fn
(61',
(34)
6 d0d<f>
-jt- J o
J^
f(0,
<#,)
Avr
^Jy Y
n (6',
),
Pn (^) sin 0d0d<)>.
Accordingly to find the value of any term Tn at any point

P, (6', <') we find the zonal surface harmonic whose axis passes
through the point P, multiply its value at every point of the
for that point, and integrate the product

sphere by the value of
over the surface. It remains to show that the development is
possible, that is that the
^TT
S (2 + 1)
actually represents
sum
of the series
(' F"f(0, 0)
Pn 0*) sin 0d0d4>,
the function f(&,
demonstrated by Laplace, but without
</>').
This theorem was
sufficient rigor, afterwards
101, 102]
193
by Poisson, and finally in a rigorous manner by Dirichlet. A

proof due to Darboux is given by Jordan, Traite d' Analyse,
Tom. IT. p. 249 (2me ed.).
Potential of Circular Disc at points not on axis.
81 the potential of a disc of surface
have found in
a
at
radius
R,
point situated at a distance r from the
density <r,
102.
We
center on the axis to be
V = 2-770- {Vr + .#
2
(1)
r>
-r).
Developing by the binomial theorem

R,
for the
two cases r < R,
(2)
1 r2
1 r4
3 r6
V:
(3)
'I
E>2
2 r
1
.
4
E>4
4 r
Q
3
4 6
1
.
6
7?6
r>#.
FIG. 46.
If
now the
point be not on the axis, but on a line through
the center making an angle 6
r from the center,

(4)
<
?-
with the
axis,
and at a distance
we may put
F= 27r<r E - rP, (cos 6) +
(cos 0)
(5)
For both of these series are convergent under the assumptions

n
n+l are zonal harmonics,
made, both are harmonic, for r Pn and Pn /r
and both take the given values when 6 = 0.
W.
E.
13
194
Equations of Lines of Force.
103.
In
CH. IV.
[PT.
I.
36,
we have
considered the integration of the differential equations of the lines

of any solenoidal vector-function and have found that the lines
represented as the intersections of two families of surfaces.

shall now consider the same subject in terms of generalized
orthogonal curvilinear coordinates. Let us call the components of
may be
We
the force at any point q lt q2> q3 in the directions of the coordinate

16 are
axes at the point, R^,
2>
3) which by
R R
R,
(1)
If
R = -h
2
-h,^
= -A
~.
of a line of force, its projections on the
now ds be an element
three axes being
Â,
d, l= *2L,
fc,.&,
we have
ds :
(2)
or
dq t
dsz
dqz
dq 3
=R
R R
=hR hR hR
ds 3
3,
so that the differential equations of the line of force are
dfr
(3)
or,
dq,
dividing by hjiji 3
dqa =
hf^,
ft,'^-
h,
V |-
dV
dV
h,
hs
3V
*^*^-|Kiif. 3aa5 Bt5"* *
4)
while
we have by
Laplace's equation the relation,
fc
87 (5))
dq l hji3 dq! )
that
dqz
dq2
(hjl*
dq 3
(h^
dqs
is
SQ1 + 3Q2 aQ s =
+
oql
We may now
strated in
Qi, Qz, Qs-
use the principle of the last multiplier demon-
37, replacing or, y,

is to say, if we
That
Mtfi* &,
we may
(6)
dq s
dq 2
z, by q lt q2 q3 and X, Y,
have found an integral
,
<?s)
= const.,
obtain the other at once by a quadrature as

p,
^- (Q
dq,
- Q,dq ) =
2
const.
by
103]
195
and inserting the values of Q 2) Q lt

2
f-
f-J
-^d
hs
3q 3
where of course all the functions under the integral are to be

expressed in terms of q lt q 2) X. This principle will be made use of
in the treatment of the flow of electric currents in thin curved
surfaces.
two particular appliFirst let q 1} q^, q s be rectangular coordinates x, y, z, and

cations.
let V be independent of z that is, the problem is uniplanar, or
The theorem becomes very simple
in
the lines of force
= z = const,
From
this
lie
in planes all parallel to the
one integral and the other
is
we
-plane.
Then
is
obtain
,
d(j,
dv. dv,
dfji,
= dfji,
fdx + fdy = ^-dx-^dy,
J
dx
dx
'
dy
dy
(9)
dji^dV
dx
and the function
a/*__8F"
'
dx
dy
dy
'
the function conjugate to the potential

function V, as found in
42.
Since by
36 the flux of the
vector
across any cross-section of a vector tube defined by four
is
/JL
X + dX, /j,,fj, + d/j, is d\dp, the function /* represents

the flux through a tube bounded by two parallel planes z = Q,
= 0, and by the surface //, = const. If the
z
1, by the surface //,
surfaces X,
represent the velocity of a fluid motion, //, is called

Earnshaw's current function, and the amount of fluid crossing unit
vector
height perpendicular to the ^-plane of any cylindrical surface

projected into a curve on the ^-plane is given by the difference in
the values of p at the two ends of the curve.
We may call the
function
//,
for
any vector the flux-function.
In the second case
let q lf q 2 qs be cylindrical coordinates pyW.z,

be independent of o>, so that the lines of force are in
The figure is then symmetrical
planes intersecting in the Z-axis.
around this axis, and we have a problem of revolution. We then
and
let
have an integral
(IO)
X=
&>
= const,
and
*'-/*(*-
for the second,
dz
a,
= const.
132
196
The
function
the surfaces
[PT.
I.
CH. IV.
represents the flux for any tube bounded by

and two planes through the Z-axis
p = 0, p = const,
making a dihedral angle with each other equal
to unity,
and p
is
then called Stokes's current- or flux-function.
Functions of Complex Variable on Surface.
104.
Both
of the cases just considered are cases of a class of problems of

considerable generality.
If the vector lines lie in one of the
coordinate
\ = qs =
surfaces
we
itself,
have
and accordingly
h 2 dV,
const.,
Jh
P^ J\hnr
5
h
CO
the
particular
integral
dV
dq
or the differential
h 2 dV,
dp ,
dp
dp = f- dq + f- dq2 = Y-J- ^ dq ,
(2)
dql
From
this
dV
h,
j-j- ^
dq
Mi
9ft
we must have
=
~ JâF
8/
dq t
hJi-L
Differentiating the
,
hji^ dq 2
dq2
'
_fy ~
=
]h.
9F
9g2
hji s
9^
dq 2
first
'
of these equations
by qz the second by
,
and adding,
9ft
V in terms of those of p
Expressing the derivatives of
<nr
dq
=
~ h&dp
h z dql
Differentiating the
__
Now
if
A3
is
first
i
_8F = AA8/^
'
dq^~
Jh
'
dq2
by q l9 the second by q 2 and adding,

,
d_
independent of
q^
two consecutive surfaces q3 qs

the same distance apart, namely,
if
"
and q 2 which will be the case

dq s are parallel, or everywhere
,
then h 3 comes out as a factor of both differential equations, and

we find that V and p satisfy the same differential equation
l&afUi
as-.lA.agJ
as
103, 104]
197
In this case from the solution of any problem for the surface qs
we may by interchanging the functions V and /* obtain the
new problem,
as in the case of uniplanar problems.

extent
of the analogy.
We have for the
this is
20,
length of the arc of any curve on the surface q3) by
solution of a
But
not the
full
If
we can
find
any two functions u(q lt g 2 )>

du*
where
dqf
_dqi*
+ dv = Mds
2
&) suc h that
^(<?i>
2
,
is a function of the position of the

point, and does not
involve the differentials dqlt dqz we have
,
(8)
Now
each
member
of the last equation
complex
factors linear in
dq
dq
1}
(9)
(dti
+ idv) (du - idv) = M( 3l+i
Each of the four

and
factors
dq*, the first, for instance,
/du
1
Now
if
product of
If
<
in this
equation
is
linear in
/du
ft 4- ("5-
dv
+^5-
a product of two linear forms is identically equal to a

two others, each factor on one side of the equation
must be a multiple of one

case we must have either
where
factored into
being
dv \ j
5-
may be
and
^r are
of 'those on the other, so that in this
independent of the differentials dq lt
dq.2
we put
du+fo = (* +
(10)
\ hi
that
is
fin
(i i)
dqi
du
+ 3-
dq 2 + %
/dv j
dq^
U-
dv
j
+ 5- dq
-f-
+ i -^-
</>,
198
equating the coefficients of dq 1 dq2

,
I.
CH. IV.
obtain
*~_
.
du_
/
we
[PT.
dv
I2 \
du
.dv
+ iz
^
dq2
Now
eliminating
dqz
<j>,
du
~
(du
dv}
-=îl5~ + ^ -
.dv}
l
.,
(dq,
and equating the real parts on each

in like manner we obtain
,
(14)
"a2
du
o~
dv
">\
I
dq2
dv
du
"22 5T~
>
j
and the imaginary parts
side,
zr~
i
l
8g2
8gj
Bg,
Solving for the derivatives of v

.
dv
_h
~"
du
^ 3w
_
~~
dv
'
î
9^1
differentiating respectively
9^
A 2 Sî
8^2
'
by qz and qlf and adding
Solving for the derivatives of u
_ A! dv du _
= A dv
"a^/Hfy,' a^ ^8^'
du
differentiating
and adding
/^
7T~
9t;\
^
dq^ \h 2
dqj
T o~
dq2
/fe,
7~
^\_~
;:
>
">
dq 2 J
the functions u and v satisfy the same equation as the

and /u. Such a pair of functions,
potential and flux-functions
that
is,
forming a set of orthogonal coordinate lines on the surface q3 may

accordingly be taken for the potential and flux-function. If we
,
have a second pair of functions
we have
u', v'
as before
du'
idv'
du + idv
_&
'
cf>
such that
104]
199
r
which is the condition,

42, that the complex variable u + iv'
is an analytic function of u + iv.
Thus from the solution of
one problem for the surface q 3 may be deduced the solution of
any number of others
same
for the
surface.
If now the quantities u, v be taken as rectangular coordinates

in a plane, the arc of any curve is expressed in the form,
da 2 = du 2 +
To any point
same values of u,
u, v in
v,
dv*.
the plane corresponds a point with the

In virtue of the relation
on the surface qs
d<r
= Mds*
between corresponding arcs on the plane and on the
we
surface,
that corresponding infinitesimal triangles

see,
are similar, or the surface q3 is conformally represented upon the
If the Z7F-plane is conformally transformed to another
plane.
as
in
43,
plane XY, we have seen that we have u-\-iv an analytic function

of the complex variable x+4y and the real functions u, v are
potential and flux-functions in the JTF-plane.
As we have
the surface q S)
just proved that they retain this property on

see that the method of the functions of a
we
complex variable will give us the solution of any number of cases

upon a surface, and that the surface may be conformally represented on the plane in an infinite number of ways.
Such a
on a plane constitutes a map.
conformally represented on a plane
may be conformally represented on each other. The theory of such
transformations is the subject of an important memoir by Gauss*.
representation of a
Surfaces which may
surface
be
The method here given is due to Beltrami-f-, and may be applied

even when the coordinates q q 2 are not orthogonal. The method
l
particularly applicable to the case of electrical currents flowing

in thin conducting surfaces, and the conformal transformations
is
may be found by
is
thin space bounded by two

distributed a solenoidal vector which may
experiment.
surfaces qs in which
be represented by a potential or by a flux-function as here described, is termed a vector-sheet

*
"
Gauss,
Allgemeine Auflosung der Aufgabe die Theile einer gegebenen Flache
auf einer anderen gegebenen Flache so abzubilden dass die Abbildung dem
"
Werke, Bd. iv., p. 189.
Abgebildeten in den kleinsten Theilen ahnlich wird.
"
t Beltrami, " Delle variabili complesse sopra una superficie qualunque. Annali
di Matematica, ser. 2,
t.
i.,
p. 329.
200
Then by
17,
surface qs be a sphere of radius

and take
and the longitude </>.
q2 the co-latitude
^ and
,1,1
=R
(dd*
if
h3
1,
= (du + dv ) M,
sin 2 0d<p)
by u and
differential equation satisfied
du
n du)
and
u0+ 3(1
If
CH. IV.
we have
ds 2
and the
I.
Conformal Representation of Sphere
105.
Example.
on Plane. Let the
for the coordinates
[PT.
v is
we take
we
jMdu = Rd6,
choose J M = R sin 0,
71
= d<b,
av
J~Mdv =
R sin
then
7
du
= -s^0 -
sin
we
Integrating
obtain
v
=
(f>,
= log tan -
<a
If now we take u and v for rectangular coordinates in a plane,

the surface of the sphere is conformally represented upon the
This particular
plane by means of the above transformation.
representation
is
known
The meridians
as Mercator's Projection.
= const, correspond to the straight lines v = const.,

= const.*
parallels 0= const, correspond to the lines u
<j)
and the
Since the whole sphere is covered by a variation of c/> between the

2-7T, the projection on the plane has the finite width 27r,
= 0, = TT
but the length of the projection is infinite, the poles
limits 0,
= 00,^=00. If we make a conformal transcorresponding to u

formation of the 7F-plane by means of the function
u
we obtain the
+ iv = log (x + iy\
formulae,
log r
= log Jx + y
+f=r=
*
9
tan ^
-^
For an example
</>
= tan"
= tan~
see Fig. 71,
n
,
$/
177.
105, 106]
201
which give a new conformal representation of the sphere on the
= tan <,
plane, the meridians corresponding to radial lines yjx
This
is the stereographic
concentric
circles.
to
parallels
and the
projection, obtained by projecting points on the sphere upon a

plane tangent at one pole from the other pole as a center of
projection.
23 projected upon the sphere by

shown in perspective in Fig. 47.
Figure
transformation
is
this
FIG. 47.
106.
Diagrams.
If
we have a diagram representing a plane
section of a set of equipotential surfaces, corresponding to equal

increments of potential, and we superpose upon this a second
diagram representing a second set of equipotential

for
the same differences of potential,
we may
surfaces,
drawn
draw the curves
representing the equipotentials due to a distribution which is the

sum or difference of the other two by simply drawing lines connecting opposite corners of the curvilinear quadrilaterals into
which the diagram is divided by the two equipotential systems.

For as we go from vertex to vertex, the increase of potential due
to one system is just counterbalanced by the decrease due to
the other.
Fig.
49 represents a combination of Fig. 23 with a

manner.
straight field in this
202
[PT.
I.
CH. IV.
if we have diagrams of the flux-function of

any two systems superposed, we may draw
diagrams of the flux-function of the sum or
difference of the two systems, for if we consider two flux tubes bounded by the lines
AB, CD, and A'R, C'D', Fig. 48, the line PQ
In like manner
Q
D
/D,
has the flux A/A through it in opposite directions from the two systems, so that the total
through it is zero, or it is a flux-line.

In this manner the Figures 49*, 72, 74, 75, 76,
77, 78 have been drawn.
flux
C'
Fm.
48.
FIG. 49.
*
Fig. 49 is to be considered a diagram of lines of flow or of equipotentials
according as the directions of the component vectors at the origin are the same
The analogous cases of the rotational problem are represented in
or opposite.
Figs. 74, 75.
CHAPTER
ATTRACTION OF ELLIPSOIDS.
V.
ENERGY.
POLARIZED
DISTRIBUTIONS.
Newton's Theorem. If
Ellipsoidal Homceoids.
transform Laplace's equation to elliptic coordinates and
88 to the problem of finding
attempt to apply the methods of
107.
we
the potential of a homogeneous ellipsoid, we are at once conIt is not evident, nor is it true, that
fronted with a difficulty.
the potential is independent of two of the coordinates, and that
the equipotential surfaces are ellipsoids.
following theorem was proved geometrically by Newton.
shell of homogeneous matter bounded by two similar and
The
placed ellipsoids exerts no force on a point placed

Such a shell will be called an
anywhere within the cavity.
homoeoid.
ellipsoidal
similarly
FIG. 49 a.
be the attracted point inside. Since the

attraction of a cone of solid angle dco on a point of unit mass at
Let P, Fig. 49
a,
204
its
vertex
I.
CH. V.
is
r
we have
[PT.
for
dm
an element of the homoeoid the attraction
da)(BP-DP),
in one direction,
and
d(AP-GP)
in the other, or in the direction
PB,
da)(BD-AC).
Draw a
plane through
ABO, and
elliptical section conjugate to
let
ON
be the chord of the
AB. Since the
ellipsoids are similar
and similarly placed, the same diameter is conjugate to the chord

CD in both. But CD and AB being bisected in the same point,
AC = BD,
and the attraction of every part
is
counterbalanced by that of the
opposite part.
Condition for Infinite Family of Equipotentials.

Although the equipotentials of an ellipsoid are not in general
whether there is any distribution of
ellipsoids, we may inquire
mass that will have ellipsoids as equipotential surfaces.
108.
Let us examine, in general, whether any singly
infinite
system
of surfaces
F(x,y
z,
g)
can be equipotential surfaces. If so, for any particular value of

must be constant, in other words =/(#)
the parameter q,
= and from that
(x, y, z q)
x, y, z are given, q is found from
^
V
from the preceding equation.
Now
since
satisfies the equation

in free space,
function of q only,
F is a
dx
dq dx'
d^V_dVd^q + dqd_fdV\
~W~ dq da? dxdx\dq)
dq
'
\dx
dq
AF=0.
But,
107, 108]
205
In like manner
d2 V
~
dy*
dq dy*
&V ~ dV&q
dz*
2
dq 8*
dq
dy
'
+ (Wd*V_
df'
Accordingly
d?V
d
dV
dq
Now since F
is
a function of q only, the expression on the right

only, say $ (q).
Consequently, that
must be a function of q
may represent
a set of equipotential surfaces, the parameter q must

its second to the square of its first differ-
be such that the ratio of

ential
parameter
is
If this is satisfied,
a function only of
we have
q,
'
dV
(5)
There must be one value 9 such that the level surface

and for this V must vanish.
of infinite radius,
is
a sphere
206
These conditions are
satisfied
[PT.
I.
CH. V.
by the polar coordinate
r,
for
by
95, (4)
4-
For r
oo
we must have
We may get
A*? into
one.
0,
accordingly
a convenient expression for
we must put
B = 0.
by transforming
fi
terms of three orthogonal coordinates, of which it

it is independent of q 2 and q 3
is itself
Put q = q l} and since
Application to Elliptic Coordinates.
109.
to elliptic coordinates gives
AX
1.
g2
(a +X)(6
+X)(
V (X-^)(X-v)(a
4-
If
21
-v
~~I~
I
7 A
1
-N.
_o
Applying
this
108, 109]
which
is
independent of
and
fi
X can represent a family
v,
207
and hence the system of ellipsoids

We have
of equipotential surfaces.
(8)
= log V(a + X) (6 + X) (c + X),
V=A
The constant
integral
X being taken
V decrease
or
is
This
between X and
(10)
an
as
oo
for
oo
which gives the
F is
increases.
an
A may be
dV =
V(a
+ X) (6 + X) (c + X)
2
'
//AX2
a differential equation which
We may determine
by an
is satisfied
elliptic function.
the constant 'A by the property that
\im(rV) = M,
*=00
or that
lim
r2
-- = ]
faj
r =oo V
We
making
terms of X,
dX
positive,
elliptic integral in
For
elliptic function of V.
(11)
X=
obtained by taking the definite
is
the lower limit, so that
for
d^
'
must be such that
V = 0.
infinite sphere,
M cos (rx\
have
d\ dx
&T
2
(a
X)
(a
+ X) (6 + X) (c + X)
2
208
so
[PT.
I.
From
the geometrical definition of X,
Now
consider, for simplicity, a point on the X-axis,
The denominator becomes

Hence
does the numerator.
=#=r.
>
jr
-^
OCC
is,
where
r 5 and
,
= - 2A = - M,
lim
r=oo
infinite in X*, that
CH. V.
so that
r.?
()
Chasles's Theorem. We have now found the potential

of such nature that its equipotential surfaces are
mass
confocal ellipsoids, but it remains to determine the nature of the
110.
due
to a
number of ways we
attempt to find an equipotential surface layer. By Chasles's
84 (n), this will have the same mass as that of a
theorem,
This
mass.
varied in an infinite
may be
will
body within
If
we
it
which would have the same potentials outside.

on an equipotential surface
constant on S, it must be constant at
find the required layer
since the potential is

points within, or the layer does not affect internal bodies.
The
surface density
o-
must be given by
dV
1
-r- =
4?r 9?z A
where n^
is
84
S,
all
(10),
the outward normal to X,
and
dV
Now
hA
since
2S A
~2?r
Since
V is a
function of
a constant value of
is
SA
alone, the
constant.
~dX'
same
Hence
dV
is
cr
true of-rr* which for

varies on the ellipsoid
distribute on the given ellipsoid S a

surface layer with surface density proportional at every point to
the perpendicular from the origin on the tangent plane at the
as
SA.
Therefore
if
we
point, this layer is equipotential,
and
all its
equipotential surfaces
109
111]
209
are ellipsoids confocal with it.

Consequently if we distribute
on any one of a set of confocal ellipsoids a layer of given
mass whose surface density is proportional to 8, the attraction of

such various layers at given external points is the same, or if
the masses differ, is proportional simply to the masses of the layers.
For it depends only on X, which depends only on the position of

the point where we calculate the potential.
Since by the definition of a homceoid, the normal thickness of
an
homoeoid is proportional at any point to the

on
the
tangent plane, we may replace the words
perpendicular
above
etc.,
by the words homogeneous infinitely thin
surface layer,
infinitely thin
The theorem was given
homoeoid.
Maclaurin's Theorem.
111.
soids, 1,
in this form
by Chasles.*
Consider two confocal
Fig. 50, with semi-axes a lt ft,
and
71,
2,
ellip-
with semi-axes
FIG. 50.
a2> ft, 72-
The
condition of confocality
2
2
i
is
= ft - ft = 72 ~ 7i = s
2
>
say-
we now
If
1
and
2,
construct two ellipsoids 3 and 4 similar respectively to

and whose axes are in tKe same ratio 6 to those of 1 and
these two ellipsoids 3 and 4 are confocal (with each other,

though not with 1 and 2). For the semi-axes of 3 are 0a lt 0ft, 0y l}
3,
and of 4 are 02, 0ft, 07 2 and hence the condition of

,
is satisfied.
a2
0V =
ft
ft
72
confocality,
- 0V =&8
Now if on 3 we distribute one infinitely thin homoeoidal
layer between 3 and another ellipsoid for which
is
increased by
and on 4 a homoeoidal layer given by the same values of and

dd, and furthermore choose the densities such that these two
homoeoidal layers have the same mass, then (since these homoeoids
are confocal) their attractions at external points will be identical.
dd,
Chasles,
"Nouvelle
h6trogne sur un
W. E.
solution
point ext6rieur."
du probleme de
1'attraction
Journal de Llouville,
t.
v.
d'un ellipsoide
1840.
14
21
[PT.
Now
the volume of an ellipsoid with axes a, 6,

of the inner ellipsoid of the shell 3 is accordingly
and that of the
shell is the
c, is
I.
CH. V.
f irahc, that
increment of this on increasing 6 by
d0, or
(vol. 3)
= 47r0 d0a /3 y
2
(vol. 4)
Similarly
Consequently, if we suppose the ellipsoids 1 and 2 filled with

matter of uniform density p l and /> 2 the condition of equal masses
of the thin layers 3 and 4,
is
f trp&fi-ff!
simply
=f
And since for

is, equality of masses of the two ellipsoids.
such
homoeoids
as
3
and
4
and 6 + dd)
two
(6
corresponding
any
the attraction on outside points is the same, the attraction of the
entire ellipsoids on external points is the same.
that
This
Maclaurin's celebrated theorem
is
solid ellipsoids of equal
Confocal homogeneous
masses attract external points identically,
or the attractions of confocal
homogeneous
ellipsoids at external
points are proportional to their masses.*
Potential of Ellipsoid. The potential due to any

112.
homoeoidal layer of semi-axes a, /3, 7, is to be found from our
109 (12),
preceding expression for F,
jfr-
ds
where X
is
the greatest root of
Now
if
the semi-axes of the solid ellipsoid are
shell a
is
unity,
Oa,
= Ob, 7 = 0c, we
ft
have
a, 6, c,
those of the
if
the density
M = îrO^ddabc,
and
Maclaurin,
Treatise on Fluxions, 1742.
Ill, 112]
where X
is
defined by
"
'**'
To get the
and
VWX
If/o no
For every value of
potential of the whole ellipsoid,
for all the shells,
(2),
211
there
is
\ /7
rt
/Trt
one value of
we must
ft ,/lft
integrate
\ given by the cubic
we may say X =
Let us now change the variable
= 62 dt and put \ = 0%.
fit, ds
s to
t,
where, 6 being constant,
Then
(4)
^=
F=2*-ofcJo
where
is
defined
by
62
Since ^ 2
is
Differentiating (5)
will call
cr,
+w
0.
thus given as a uniform function of
change the variable from 6 to
When
+w
= 0,
= oo
by
u,
we
will
0,
and when
= 1, u
has a value which
defined by
Accordingly, changing the variable,
(8)
F=
The three double
integrals above are of the form
du
where
/() =
now
u.
142
we
212
This
may be
[FT.
I.
CH. V.
integrated by parts.
Call
f(t)dt
\"
u
= $(u\
r^(u}du_ r>wr
~~L^T^J,
"J, (CT^)
r*
Now
<
(oo
=
J
/() dt = 0,
r#(u)du
c+u
(since /(oo
= 0),
00
*() -("/(O*.
J
<r
f ()/(>
Inserting these values
or the variable of integration being indifferent,

t in the first integral.
we may put u
for
Applying this to our integral, by putting C successively equal

to a 2 b 2 c 2 multiplying by # 2 ;y2 2 2 and adding,
,
Now
the
first
three terms of the integrand are, by definition,
equal to 1, so that
'
(a
+ w ) (6 + u) (c
2
This form was given by Dirichlet*.

If the point x, y, z lies on the surface of the ellipsoid
then
cr
and
jr
f^fi
V=7rabc\
11
/
(14)
Jo
*
//.
Dirichlet,
--
^'
a*
--
+u
"Ueber eine neue Methode zur Bestimmung
der Berliner Akad., 1839.
vielfacher Integrate."
Translated in Journ. de Liouville,
t.
iv.,
1839.
112, 113]
We
du
ZTTCLOCX
a2
u) V(a + u) (6
2
cr
definition of
By
find for the derivatives of
213
o-,
<r
4-
u) (c
+ u)
the parenthesis in the last term vanishes,
-=-
(15)
3F =
^
du
ZTTOOCZ
Internal Point. In the case of an internal point, we

pass through it an ellipsoid similar to the given ellipsoid, then by
Newton's theorem it is unattracted by the homceoidal shell without, and we may use the above formulae for the attraction, putting
113.
for a, b,
c,
the values for the ellipsoid through a,

is on the surface of this, a = 0.
y, z,
say Oa,
Ob, Be.
Since the point
dV
du
+
Now
let
- -
fa
us insert a variable
divides out,
+ u')
and writing u
J(a? + u) (b
for
for
any
+ u') (c + uf)
2
the variable of integration

.
+ u)J(a + u)(b + u)(c* + u)

internal point, we put a- =
in the
2
(a?
x, y, z,
general
we have
a+
+u
+u
The constant term must be taken
= 0*u
Integrating with respect to
Joo
surface
du
f
I
Jo
So that
formula.
i
77 =
F
nrabc
zirabcx
u,
'
6* (a 2
00
BF =
^
proportional to
u'
6W
6s b x f"
Jo
The
2
2 2
2 2
u) 7(0 a + u) (0 6 + u) (# c + u)
may
be continuous.
(a
+ w) (6 + w) (c + w)
2
as above in order that at the
214
[PT.
I.
CH. V.
In the case of an internal point the above four integrals

be made to depend on the first. Calling
= Trcibc
<3>
9 (a
jraOC
/(
aa + u
2 (a 2
Jo
may
du
+ u) >/(a + u) (6 + u) (c + u
2
and accordingly,
The
and
integral 3> is an elliptic integral independent of #, y, z,

2
2
so are its derivatives with respect to a , 6 , c 2
Calling these
.
..
M
N
~r>
-r
respectively
>
we nave
a symmetrical function of the second order, and since L, M,

are of the same sign, the equipotential surfaces are ellipsoids,
is
how-
to the distance
from
Their relation to the given ellipsoid

ever transcendental, their semi-axes being
similar to each other.
V3 (a2 )
We
have
8(6
9(c
for the force
Hence, since
for
dx
dy
dz
two points on the same radius-vector,
^ = ^=^ = ^,w
#1
The
2/1
forces are parallel
and proportional
the center.
114.
point,
Verification
by
Differentiation.
For an outside
we have
C
c
+ u)
+ u } ^ + u} (c + u)
2
V(a2
113, 114]
du
J* (a
+ u)
+ u) (6 + u) (c + u)
2
\/(a
(a
(a
3<7
+ w) V(a + w) (b- + M) (c
2
+ <r) V(a + <r) (6

2
-f <r) (c
2#
#2
Forming ^-^ and
^-
+ <r)
^~
4- M)
+ a)
may
lS>
(6
-{-
|
2
(c
+ o-)
'
2
|
""
)
(
-\-
U)
V (o?
'
-f-
-ft*)
be at once effected.
d (uvw) = uvw
Since
2
'
and adding,
\af -\-ii
integration
+ a-)
rj ~5~Ii
J y
(a
\
\
-l
we have
1
V(a + 'M)(6 + w)(c +w)>

2
The
'
19, (5),
The
r
J
Now by
215
If
2 (a 2
+^
integral
J_
b2
+u
..1
j
z
c
u\
du
du
V(a
+^
becomes then
which cancels the second term, and
For an internal point
r
J
__
2
(a
AF= 0.
+ u) \7(a + w) (6 + M) (c + u)
2
216
At
oo
cr
infinity
and
and
[FT.
derivatives
its
I.
CH. V.
accordingly
vanish.
Hence the value

115.
Ivory's
V found satisfies all
of
Theorem.
If
x,
the conditions.
z
y,
a point
is
on
the
ellipsoid
the point
on the
lies
ellipsoid
We
shall now
will be called corresponding points.
smaller.
and
the
con
are
two
these
that
assume
focal,
(2)
ellipsoids
These
Then
action of (2) on the external point x, y, z
The
X, = -
27ra 2 6 2 c^
du
00
J^
(^ + M ) V(5?+
2
a/'
where
is
-r
v
rv;
tt
= 1,
z*
- + y* + =
(^ + 1*) (c,
a?
and
since
we must have
If
now we
<r
1,
= X.
substitute
iff
+ c^ - a 2 = w + cr,
7
00
Now
C&o
a?
the attraction of the ellipsoid (i) on the interior point
^2
,
y j-
^=
^2
is
a2
ZTraAcj a?
o
(d!
+ w) V(a + w) (6 + u) (d + w)
2
114116]
The
same
definite integrals being the
aA
in
217
both
cases,
we have
'
is Ivory's theorem:
Two confocal ellipsoids of equal
act
on
each
density
corresponding points on the other with forces
whose components are proportional to the areas of their principal
This
sections normal to the components.*
Ellipsoids of Revolution. For an ellipsoid of revoluthe elliptic integrals reduce to inverse circular functions.
116.
tion,
Put
c,
a being the axis of revolution,
V=7ra
(i)
du
(2)
where
a2
+ <r
^+
+^
= 1.
<r
Put
2
2
2(6 --a )
When
QO
when u = <r,
_ ^2
= \//^2
j-rV
+ O"
so that
- J (Xx +
Ivory,
"On
the attractions of
homogeneous Ellipsoids/'
Yy)
Phil. Trans., 1809.
218
[FT.
I.
CH. V.
/b*^tf
*+'
27ra& 2
fV
=
j]T==J
vo
a? J Q
(4)
ds
:/f=iVI s
<
Zv =
(6*
-a")
ds
so that
62
-a
(8)
Now
=i
-^=
VI -
[
J
1
[sin- s
-5
so that
2*y
-6 ^^
2
"
7^2
'
For sin
in" 1
V/
T2
S11
/(fr-rfXa'
/6^?
V6
V
+
_
- ^^2
- we may
+ cr
for if
sin 6
then
cos 6
62
<r
"
/^2
"j.
J-
write tan"1
V/
+ <r
_ ^2
*"
a- 4-
cr
/?L5
~V
+ o-'
/6 - a
tantf "
V a + o62
'
These formulae all serve for an oblate spheroid, where a <b.

For a prolate spheroid, b > a, they introduce imaginaries, from
which they may be cleared as follows.
116, 117]
sin- 1 (iu)
Call
e~ ie
0,
= sin 0, Vl + w = cos 0,
= cos - i sin 6 = Vl + u + w,
- i6 = log {Vl + ^ + M},
2
iu
then
therefore
sin" 1 (iu)
Put
=6 = i log {Vl -f u
2
-f-
M}.
tt
Va + a- + Va - 6
2
2
)
Hence
F = 7 ==log
va -6
,
Tr
(10)
4?rafe
Va2 -62 + Va2 +
^-(^^t ^- ^7PTT~
all
(a
a2
-(^T^j- HP+F"
these formulae,
is
er
-6
2
)
-V^TÎ'
- 6 ) (a + a)
2
/a2
o-
v
In
f.
^>
the larger root of the quadratic
r-2
for
an outside point, and a =
case,
we have
an inside point.
a*
functions only of the ratio
Energy of
117.
for
Distributions.
In the latter
j-
Gauss's Theorem.
If a
particle of unit mass be at p, (x, y, z) at a distance r from a particle

of mass
the work necessary to bring the unit particle from an
q,
infinite distance against the
will be
g
repulsion of the particle
W = ?*=V(x,y,,)-V,.
(0
If,
instead of a particle of unit mass,
the work necessary will be
Wpq =
(2)
we have one
of mass
mp times as great,
2-mp
= mp Vp = m q Vq
V = ^.
where
Thomson and
Tait.
Natural Philosophy, Part n.,
527.
mp
220
[PT.
I.
CH. V.
In other words, this is the amount of loss of the potential

energy of the system on being allowed to disperse to an infinite
distance from a distance apart r.
Similarly, for any two systems
of particles
mp mq
,
w
v mpm
w pq _ v
Zq&p
(?\
\3)
_v
zpmp v
vp _ v
vq
2, qm q v
'
Ipq
Vp being the potential at any point p due to all the particles q

and Vq being the potential at any point q due to all the particles
This sum is called the mutual potential energy of the systems
p.
p and q. If however we consider all the particles to belong to one
system, we must write
= iS,r,
'F-i22^
r
(4)
pq
where every particle appears both as p and

because every pair would thus appear twice.
been given in 59, (33).
the \ being put in

This expression has
q,
If the systems are continuously distributed over volumes
r,
r'
we have
Wn
(5)
The theorem expressed by the equality of the two integrals is

known as Gauss's theorem on mutual energy, where Vp represents
'
the potential at
the whole mass
The above
p due
to the whole
mass
Mp *
equality
may be
also
q,
proved as
Vq
that at q due to
follows.
Since
p^-
and
the triple integrals in (5) become respectively,
(7)
and
* Gauss.
"
Allgemeine Lehrsatze in Beziehung auf die im verkehrten Verhaltund Abstossungs-Krafte." Werke,
nisse der Entfernung wirkenden AnziehungsBd. v. p. 197.
ENERGY.
117, 118]
Now
since outside of
AF=
T,
and outside of T', AF' = the

But by Green's theorem,
space.
integrals may be extended to all

both these integrals are equal to
a*
221
'
ay
dy"*"&
since the surface integrals
nish at infinity. Gauss's theorem accordingly follows from Green's

theorem and Poisson's equation.
Energy
118.
in terms of Field.
distribution consisting of both volume

the sum (4) becomes the integrals
Now
If,
For the energy of any

and surface distributions,
at a surface distribution Poisson's equation
as
tributions,
J_f3F
8F)
4?r
dn2 }
\diii
is
'
85, we draw surfaces close to the surface disand exclude the space between them, we may, as above,
in
extend the integrals to
all
other space, so that
the normals being from the surfaces 8 toward the space

by Green's theorem, as before, this is equal to the integral
r.
But
r- S
U o)
Thus the energy
is
expressed in terms of the strength of the
field
at all
points in
space.
This integral
is
of fundamental
im-
portance.
It is at once seen that this is always positive.
We
may obtain the same expression as follows. Suppose

that the matter at a point x, y z is displaced to a point x + &r>
y
222
+ &y,
+ 82.
The amount
[PT.
I.
CH. V.
of matter in the fixed infinitesimal
dm = pdxdydz
is
thereby changed, and
same as that required to bring the mass
the point x, y z, where the potential is V,
parallelepiped dxdydz,
the work necessary is the
Sdm from infinity to

namely, STF= VSdm.
We
have found in
fr&p)-
38,
d(pSz)}
+ d(pSy)
-vp- + ~.
>
dx
dz
dy
Consequently the whole increase of energy

dx
j , ,
dxdydz.
is
dz
dy
Integrating by parts
dxdydz
the integral being over
= -fvpSxdydz +
all
space,
'dxdydz,
pBx
and the surface integrals vanishing
at infinity.
But
this
since
-r-
47T
A V,
becomes
_! |Y f 1 ^Z 8iK
~^7r]j] ao \dx dx
?? 88F ?? ^Zl
dy
dy
dz
dz
a^
00
so that
For a third deduction, since in moving a mass
whose components are Sx,

work done by the system
$y,
z the energy lost
- 8 W= dm {XSx + YSy + ZBz]

8F
3F,
(14)
dm
is
a distance
equal to the
ENERGY.
118,119]
The whole
variation of the energy
223
is
Applying Gauss's Theorem to the mutual energy of two

butions, one of which has density p, producing the potential
other the density
p,
distriT7 ,
the
producing potential SF, we have
jjj
W==
and
gives in agreement therewith
[[[
The
integrals
8p.Vdr=jjj pSVdr.
may be now
restricted to the space occupied
by
matter.
Maximum theorem
119.
for
Energy.
By making
use
two different expressions for the energy we can deduce

an important theorem relating to the energy of a distribution.
of the
We may
use the form,
(I)
118, (8),
<rVdS +
pVdr,
which is distinguished by the suffix d to denote that the densities

occur explicitly.
This form, by the definition of the potential,
holds for any law of force, whether the Newtonian or not.*
On
the other hand
we may use the
form,
118, (10),
which we give the suffix / in order to denote that it is expressed only in terms of the field at all points, and does not
to
By
this
we mean any conservative law in which the action is proportional

and to some function of their relative position.
the product of the masses,
to
224
explicitly contain the densities.
[PT.
I.
CH. V.
This expression holds good only
a distribution acting according to the Newtonian Law.

these two expressions must be equal for all distributions, we
for
As
may
write
(3)
If in this latter expression for

tion in the form of the function V,
of
W we make an arbitrary variawe
obtain for the varied value
an integral containing the variations of V,
we suppose
and
If
p.
distribution also acts according to the

virtue of Poisson's equation there will be
Newtonian law, in
between SV,
relations
We
a,
new
that the
So-, Bp.
however remove
shall
this restriction,
and consider V,
cr,
independent functions, which can be varied inde-
as perfectly
pendently.
We
and Bp as zero, in other words we shall

be varied from the values that it actually has for
shall choose Bo-
to
suppose
the original Newtonian distribution, the variation being entirely
Calling the variaarbitrary, while the densities are unchanged.
tion
(4)
under these circumstances Sp-W,
W + S r W=a-(V+SV)d8+
p(V+SV)dr
~~
+ /8(F+SF)y|
From
(5)
Sv
we obtain by
this
W=
subtraction of (3),
<rSVdS +
L fff
4 7rJJJ 00
pBVdr
8F d8V
d-T
(das
dx
dv 88F
a8F
<
dy
dy
dz
dz
dr
dSV^
Integrating the third integral by Green's theorem,

finally
we have
ENERGY.
119]
225
'
V a
V3*/ 1
a Newtonian one, by Poisson's

equation the factors multiplying SV in the first two integrands
is equal to minus the last
are zero. Consequently the variation of
Since the unvaried distribution
is
integral, which, as the integrand
Accordingly S V
positive.
W<
sum of squares, is necessarily

and we may state the theorem
is
If the potential due to any given distribution of matter acting

according to the Newtonian Law is known, the energy calculated
by the formula (3) is a maximum for the actual distribution of
potential as compared with arbitrary distributions differing

an infinitesimal amount from the actual.
We may
by
theorem in physical language, avoiding

is to be expressed, as follows.
specification of the form in which
We may consider V + S V as the potential due to a Newtonian
distribution whose densities differ at each point of space by an
state this
amount from the densities of the given distribution,

the differences being otherwise perfectly arbitrary. We will call
the supposed distribution 2, the original distribution being 1.
infinitesimal
Then the terms
jj<r(V+SV)dS
are the
mutual energy
jjjp(V+SV)dr,
TT12 of the distributions 1
and
2,
117, (5).
The
integral
*
is
the energy TF2 of the distribution
Accordingly equation (4)
2,
by
118, (10).
is
Wi+*rW-Wn-W*
so that
8V
W.
E.
W= W
12
- (W, +
W,)
< 0,
Tf12
< W, +
W,.
15
by
226
That
is
The mutual energy
of any
[FT.
I.
CH. V.
two Newtonian distributions
differing infinitesimally from each other, and supposed co-existing,

is less than the sum of their individual
This theorem
energies.
is
it
probably true for

in
all
120.
We
repulsive forces.
180 to deduce the laws of
dielectric
shall
make
and magnetic
use of
actions.
Potential of Polarized Bodies. Double Distribumass dm be placed at a point where the
If a particle of
tions.
field-strength is F, it experiences a force of

If any distribution
direction of the field.
amount
Fdm
in the
be placed in a
field
is uniform, that is, for which at all points

is constant in
value and direction, the force
experienced by the whole mass is
which
IIIFdm
=F
[ (dm
= Fm.
Certain bodies exist in nature which, on being placed in a uniform

field, experience no tendency to move in any direction, so that
must be zero, or their

Accordingly for such bodies
be
in
some
must
and
in others negative.
points
density
positive
Such bodies experience a couple when placed in a uniform field,
is
zero.
although the resultant force vanishes. Not only does the above
property hold for the whole body, but if it be broken into any
number of parts the resultant force on each part is zero. A
is the most familiar example of such a body, for placed in
a uniform field of magnetic force it experiences no resultant force,
no matter into how many pieces it may be broken. In such a
magnet
body then the mass of any

us consider
how such a
part,
however small, must be
condition
is
zero.
Let
possible.
Let us suppose that any portion of space r is occupied by a

body A of constant density p and that occupying identically the
same space is a second body B of constant density p. The two
bodies will then completely neutralize each other's action in every
way, and when placed in a uniform field would experience neither
force nor couple.
Now suppose that the first body A is displaced by
an infinitesimal amount, so that every point in it moves a distance

The effect will
h in the same direction.
be that while in the
common by
space occupied in
the two bodies the densities
is a space on
with positive matter, and on
neutralize each other, there

FIG. 51.
one side
filled
POLARIZED DISTRIBUTIONS.
119, 120]
227
the other a space filled with negative matter, (Fig. 51). The
volumes of these two spaces must be equal, since the bodies A
and
The
originally coincided.
a body covered on part of

positive, and on the remainder
of
surface
If n,
distribution.
effect of the
system
surface with a
its
with
Fig. 52,
is
is
now
that
negative
the normal
drawn inwards, and h represent the direction as

well as magnitude of the displacement, the amount
of matter contained in a right prism standing on the
element of surface dS will be
FIG. 52.
ph cos (hn) dS,

but
this is equal to crdS where o- is the surface

Accorddensity.
ingly the surface density of the equivalent distribution is
a=
(1)
If
we now decrease h and

and equal
finite
product
ph cos
(hn).
increase p without limit, keeping their

to /, we obtain a body charged with
surface density
&=
(2)
/ cos
(hn),
possessing the property of experiencing a couple, but no resultant

force when placed in a uniform field.
To
find the
magnitude of the couple
let
us divide the body up
into prisms with their generators parallel to h and standing on the

elements dS. Such a cylinder of length I carries upon one end
the charge adS which experiences the force Fa-dS, and upon the
other the charge - vdS which
the force Fa-dS in the
experiences
The
moment
of
the
direction.
opposite
couple thus produced is
I
sin
(hF) Fa-dS.
For the whole moment we must take the integral of
this over the
positively charged surface,

III sin
(3)
Now
cos (hn)
(hF) F<rdS
dS
is
= -F sin
(hF)jjlI
cos (hn) dS.
the area of a right section of the prism on dS,

rl
so that
cos (hn)
dS
is its
volume
dr,
and the
total
moment
Jo
becomes
(4)
F sin (hF)
fjj^r
= FI sin (hF)
r.
152
228
[PT.
I.
CH. V.
Such a distribution may be called a double or sliding disand a body possessing such a distribution is said to be
polarized in the direction h. The conception of the sliding
distribution is due to Poisson.
The moment of the couple
tribution,
experienced by the body when placed in a uniform field of

strength unity whose direction is perpendicular to the direction
of polarization, is called its moment of polarization, and J, the
moment of unit volume, the intensity of polarization. / is a
vector quantity, for the whole couple may evidently be supposed to arise from three bodies occupying the same space,
and polarized in three mutually perpendicular
directions, the in-
tensities of polarization being respectively
4= /cos (70),
For
if
J3
= /cos(/y),
the components of the
= /cos(/*).
field are
X=Fcos(Fx), Y=Fcos(Fy), Z = Fcos(Fz),

the polarization B produces the couple BZ about the X-axis, and
CY about the same. In like
the polarization G the couple
manner the couples about the other axes

(5)
L = BZ-CY,
are obtained,
M=CX-AZ, N = AY-BX.
We
accordingly find that the resultant couple is the vector product

of the intensity of polarization and the field-strength, whose
magnitude
is
FI sin (FI),
agreeing with
(4).
Suppose now that we have a body of such a nature that every

element of its volume has a double distribution, although the
direction and magnitude of the polarization / may vary from
element to element.
Such a body
is
polarized in the most general
manner, and the volume-density will not vanish throughout. Let

Conus seek its value in terms of the polarization at each point.
sider a rectangular element of volume dr, whose sides are das, dy, dz,
and in which the values of the component polarizations are A,B,C.

Then the face dydz on the side next to the origin has the charge
Adydz, while the opposite face has the equal and opposite
charge Adydz. In the next element of volume on the right,
whose center is at a distance dx from the center of the first, the
Jf-component of polarization
will
be
dA
A+^-dx,
ox
120, 121]
so that the face
dydz next the
origin will have the charge
superposed on the charge

on
that face is
charge
This
is
229
Adydz,
4-
so that the whole
jdxdydz.
Similarly the faces dzdx and
origin, have the charges
dxdy on the
dB
-
and
oz
sides farthest from the
dxdydz.
dxdydz.
If we consider these faces as belonging to the element dr, while

the opposite faces belong to adjacent elements, the whole charge
belonging to the element dr is
\dA
dx
\
\~o"
dB
+ ^r~ + dC]
"o~
dy
Thus the charge per unit volume
dz)
is
dC
Integrating this throughout the volume occupied by the body
(7)
II{A cos (nx)
=
[jl
so that the total
121.
cos (In)
+ B cos (ny) +
Ccos
dS
dS=IjffdS,
mass of the volume and surface charges
Induction.
(nz)}
Comparing the expression
is zero.
for the density
with the ordinary expression for the density as l/4?r times the
divergence of the force, we have
p
~~
230
g whose components
so that the vector
1
A
4?r
{Fi cos (Fiiii)
+F
cos
I.
CH. V.
are
has no divergence anywhere. Its normal component

at the surface of the body, for since by
82,
<r
[PT.
(Fe n e )\ =
is
continuous
/; cos (/i/ii),
(where the suffixes i and e denote values inside and outside, and
the opposite directions of the corresponding normals), we have
gi COS (giî)
( I O)
= Fi COS (Fify) + 47T/i COS (liUi)
cos
(Fe ne ) =
The
vector
any
closed surface vanishes, so that as
g being everywhere
ge
cos (g e ê )
=g
cos (g e ^).
solenoidal, its surface integral over
unit tubes enter as
many
Tubes leave the polarized body where a is

it where a is negative.
enter
and
They form closed
positive,
The vector g
tubes, every one of which passes through the body.
is called by Maxwell the induction, and is characterized by the
The line separating the region of positive a
solenoidal property.
leave the surface.
from those of negative
is
linked with
all
the tubes of induction
belonging to the body. The induction is not in the same direction

unless the polarization / is.
as the force
We
obtain another physical conception of the induction by

considering the force in a cavity in the conductor. By hollowing
out a space in the body we remove a portion of the volume distri-
We
shall
bution, but give rise to a new surface distribution.
suppose the cavity so small that the volume-density of the part
removed may be considered constant. Now if we consider the
forces at corresponding points of geometrically similar distributions

of constant densities, we have for the action of the volume-density,
V=
and if we increase the dimensions in the ratio n, the element of
B
volume and the potential at a corresponding point are dr = n dr,
"
and the
=p
[(f dr
'
hr=p
[[(n*dr
hr= n
force
_W _ _c
~
ds'
\(n*V)
d(ns)
dV
'
ds
121, 122]
231
while for surface distributions
dV
dS
ffdS' = o-n
[
nrJJ
8F
nr
JJ r
ds'
Accordingly as we decrease the linear dimensions indefinitely, the

force from the volume distribution decreases indefinitely, while the
force from the surface distribution remains finite.
Consequently
infinitely small cavity the force
in
an
all
points
does not depend on the

volume density of the part removed, but only on the surface
This will be at
densities formed on the surface of the cavity.
(7
= 1 cos (In),
the normal being directed into the cavity. Suppose the cavity is in
the form of a cylinder with generators in the direction of the
polarization.
Then the density on the
/ and
If
sides
is zero,
and on the ends
the radius of the cylinder, 26 its length, the

a point at the center of the cylinder is the
the
ends
on
action of
same as the action of two circular discs, of surface density / and
/,
/.
which, by
is
81, is
47T/
a function only of a/6, as we have just shown that the action

independent of the linear dimensions. If the radius is infinitely
This
is
is
small in comparison with the length the action vanishes. Accordingly in such a cavity the force is that due to the action of the
rest of the body, or
Y=
-"-
dv
^
dx
>
v=
-*
<3F
o
dy
"
dV
-s
oz
hand the length of the cylinder is infinitesimal in

with
the radius, the force is 47r/, so that the total force
comparison
in the cavity is
If on the other
or the induction
is
equal to the force in a thin crack perpendicular
to the lines of polarization.
122.
Potential due to Polarized Distribution.
If
we
introduce the expressions of the volume and surface densities in

terms of the polarization, we obtain for the potential due to a
polarized distribution
232
[PT.
I.
CH. V.
"
(0
r C{
A cos (nx) + B cos (ny) + C cos (rig)}

1
_ j{dA
r \dx
,~
dB
i
'by
dC
_
dz
Integrating the volume integral by parts by Green's method, the

surface integral cancels the surface integral in F, leaving
as the
volume integral
If as usual
we use
point, and a, b,
we must write
(3)
Now
V=
z to denote the coordinates of the attracted
x, y,
the coordinates of the point of

integration,
c for
ffff
J J J
(~)
P)l
(-)
+ B^l
+ C-gl\dr,
\A-^
oa
do
oc )
(.
dr^dadbdc.
since
r2
= (x
a)
6) 4- (z
(y
c)
(r)
da
__l*-a =
2
r*
H*
iy?l
r2
= n& cos (rz),
the integrand is the geometrical product of the intensity of polarization and r the vector distance from the polarized element to the
We might
attracted point, divided by the cube of the distance.
have obtained this result from the consideration of a doublet, or
pair of points of equal masses of opposite signs, placed at a distance
= mh. Then if
apart h, so that the moment of the doublet is
and r2 are the distances of the attracted point from the positive
and negative ends of the doublet, we have
T-L
Tr
m m m (r
r^
122]
But
if
is
infinitesimal,
we
233
have, neglecting infinitesimals of
^ = h cos (hr) and r^ = r

v _ mh cos (hr) _M
the second order, r2

(5)
Now
in a polarized
body
in general, the element of
volume
dadbdc contributes the potential
_ Tr
Idadbdccos(Ir)
r2
sp that
we obtain the form already
If the direction of the

is,
dr
r3
'
given.
axis of the doublet
is
that of the
we have
-
r3
to which,
by
103
(10),
we
'
find the conjugate function
FIQ. 53.
234
From
[PT.
I.
CH. V.
the lines of force due to the doublet, or the lines
this,
^ = const., are drawn by a simple geometrical construction, Fig. 53.

(Nature, Vol. XXL, p. 371.) The combination of this field with a
straight field in the direction of the axis of the doublet, drawn by
the method of
the
field.
The
shown
in Figs. 74 and 75, according as the

in the direction of, or opposite to that of
lines are drawn for equal increments of W.
106,
axis of the doublet
is
is
Potential due to uniform Polarization.
123.
We may
easily find a convenient expression for the potential at any point

constant.
Let
to uniform polarization, /
represent the
due
potential at a point
n
P whose coordinates are x, y, z of a body occupy-
ing the space r and filled with a single or ordinary distribution

Then after the body A of the
of the uniform density unity.
double distribution has been displaced the distance h in the direc-
is the same as if the body had remained

tion I, the potential at
had been displaced the distance h in the opposite
fixed while
direction, that
is,
_an
dh
The
potential at
of the negative body
and the potential of the double distribution
is
is
the
sum
of these
two, or
and inserting the value of the
polarization,
an
Consequently
if
we know the value of the
.an
potential due to a
obtain by differen-
single distribution of constant density, we may

tiation the potential for a body of the same form uniformly
polarized.
The
The expression
potential due
due to a point
is
holds both for inside and outside points.
to a doublet illustrates this, for the potential
122124]
while the potential due to the doublet
235
is
af-
a spherical harmonic of degree

124.
and
Solenoidal
2.
Lamellar
Polarizations.
The
120 (6),
volume-density of polarized matter has been found,
to be equal to the convergence of the polarization.
If the polarization
is
solenoidal, the
volume -density vanishes, and the
polariza-
equal to a surface distribution, as in the original assumption of

120.
We may then divide the body into tubes of
tion
is
Such a solenoid possesses the
polarization, or polarized solenoids.
property that
if it
be cut anywhere the two cut ends
will
bear
equal and opposite charges, their amounts being the same wherever
the cut be made. The potential due to a solenoid of infinitesimal
section depends only on the position of its ends, and a solenoid
may be considered as equivalent to a doublet of points at a finite
distance apart.
Again the polarization may be lamellar, that is it
may be the vector differential parameter of a function <f> which
We
be called the potential of polarization.
will
A
A.
(R\
( O
)
= ^<t>
=
da
of
d<!>
;ry-
db
ri
,
(j
then have
d</>
.
dc
Outside the polarized body, since / = 0,

at the surface.
$ must be
constant,
and accordingly discontinuous
Inserting this in the value of
the potential becomes
dc
dc
Applying Green's theorem we obtain
-///-
(10)
But
point
is
since 1/r is harmonic except for r = 0, if the attracted

is
outside of the polarized body,
given by the surface
integral,
(ii)
F=-
236
[PT
I.
CH. V.
Accordingly the potential at points outside of a lamellarly

polarized body depends only on its form and position, and on the
values of the potential of polarization at the surface.
If the attracted point
we may
body,
and
l/r,
is within the substance of the

polarized
integrate (9) in the other manner, interchanging
<
obtaining
which, by the theorem of
84 (12) applied to
83 (5) or
</>,
becomes
V"
d3)
Ufl
In the case of lamellar polarization the induction becomes
(14)
so
dV
dd>
dV
= -^+ 47r^-, $ = ---
that the induction, being
F+47n,
is also
function
this
is
lamellar.
equal,
**
**
air
the parameter of the function

For both inside and outside points,
except for a constant, to the surface
integral
(15)
as
<
|/W
we
is
see from (13)
and (n), together with the
fact that outside
constant.
125.
Polarized Shells.
The
characteristic of a
lamellar
we
construct two infinitely near equipolarization

= ^>a and = 2 the polarizapotential surfaces of polarization cf>
is
that
if
<
<
>
normal to them at all points, inversely proportional to the

distance between them, and in the direction from the smaller to
the larger value of <. The portion of matter included between
the two surfaces, which need not be closed, is called a simple
tion
is
we
consider the infinitesimal portion of the

shell, the surface integral (i i) is
taken over both sides of the shell, the portion over the edge
polarized shell.
potential due
If
to such
an unclosed
Consevanishing, since the width of the edge is infinitesimal.

n
and
2
away from
quently, replacing n, the internal normal, by
124, 125]
the
237
shell,
dS.
The geometrical
integral,
39 to be equal to the solid angle
has been found in
tended at
by the surface S,
toward the side of
S on which
if
n-^
o>
sub-
points
P lies.
Con-
due
sequently the potential at any point
to the shell is equal to the product of the
difference of potential of polarization on the
two sides of the shell by the solid angle
subtended by the shell at P, the potential

if
is on the positive side of
being positive
the shell, that
polarization
is
angle integral
and
is,
Now we have seen in
directed.
equal to
is
FIG. 54.
the side toward which the
39 (5) that the
solid
4?r for a point inside a closed surface,
an outside point, that is, it experiences a disWhen the surface is

of
as
crosses the surface.
4-7T
continuity
not closed the same thing takes place. For the integral
to zero for
is
as
a continuous function of
P does not lie
on the
P so
long as r
is
P lies
If
surface.
not zero, that is, so long

on the surface, the in-
We
remove this by cutting out a

be that part of the integral due
to the remainder of the surface, a is finite and continuous even
tegral has
an
infinite element.
small area around P.
when
If
now
o>'
As
approaches the surface
passes through the surface.
the solid angle subtended by the small area cut out, which may be
treated as plane, approaches 2w, so that at the surface on the side
= a)' 4-
2?r.
At an infinitely near point on the side 2, howthe
in
cosine
the numerator has changed sign, for the small
ever,
area, so that the solid angle subtended by the latter is to have the
1, &>!
negative sign.
cordingly,
(18)
Accordingly on the side

!
o> 2
4-7T,
2,
o> 2
CD'
2?r,
and
ac-
238
[PT.
I.
CH. V.
and the potential

4?r (fa
experiences a discontinuity of
fa)
in passing through the shell from the positive to the
negative
side.
The
by
discontinuity may be also explained

considering the solid angles subtended
at points 1
the surface
solid angles
and 2 approaching a point on

from opposite sides.
If the
have different signs on opposite
the points come together the sum

absolute values of the two angles
approaches 47r, so that at the surface
sides, as
of the
If the thickness of the shell
is e,
the polarization
is
(</>!
<
2 )/ e >
and the moment of the equal and opposite charges on the element
of surface dS on the opposite sides of the shell is, since the volume
of the element
is
edS, equal to
(fa
fa) dS.
Thus the surface density times the thickness, or the moment

of polarization per unit of surface of a simple polarized shell, is
The value of the constant <3> = fa fa is called the
constant.
strength of the shell, and it is this strength that is multiplied by
the solid angle in the expression for the potential.* Suppose now
that the intensity of polarization increases without limit, so that
the strength of the shell fa
fa is finite, instead of infinitesimal.
Then the
difference of potential on the two sides of the shell is

or
the potential is discontinuous in crossing the shell, by
finite,
the amount
The derivative, dV/dn, is however continuous. We may prove

the converse of this proposition. If a function satisfies Laplace's
equation, vanishes at infinity, and is continuous everywhere except at a certain surface, its first derivatives being everywhere
continuous, the function represents the potential of a double
If the function were
distribution on the surface of discontinuity.
uniform and continuous, it must, by Dirichlet's principle, vanish

everywhere. The demonstration will be given in 210.
*
v., p.
Gauss.
119.
"Allgemeine Theorie des Erdmagnetismus,"
38, 1839.
Werke,~Bd.
125, 126]
126.
Energy of Polarized
distribution
is
mutual energy
which, by
239
Distributions.
If a polarized
placed in a field of which the potential

is,
120,
by
is
is
F, their
117,
equal to
W= - JJV {A cos (nx) + B cos (ny) + C cos (nz)} dS
(i)
US
'
'
\te
Sy
S}*-
Integrating by Green's theorem, this becomes

(2)
The integrand
the negative of the geometric product of the

This result may be obpolarization and the force of the field.
tained directly for a doublet as we obtained the potential in
is
122.
If the polarization
tion
is
lamellar, the energy of the distribu-
is
For a polarized shell the

surface integral becomes
volume integral disappears, and the
Accordingly the energy of a polarized shell is equal to the

product of its strength by the flux of force through it in the
direction opposite to the polarization.
If
itself,
we wish to find the energy of the polarized distribution

we must put for V in the above formulae the potential
due to the distribution
itself,
and multiply by the
factor one-
It is important to notice that the energy of

is defined as the work that they are capable
distributions
polarized
of doing if every particle is allowed to retire to infinity carrying
half, as in
117.
240
[PT.
I.
CH. V.
own charge. But if the distribution should be cut up into

small parts, new surface densities would appear on each part.
To prevent this the distribution must be supposed split up into
its
on separating
would be formed,
so that the energy as calculated would be the work obtained
by letting these shreds be bodily removed to infinite distances
from each other.
Similarly polarized shreds side by side of
infinitely thin shreds along the lines of polarization
these from each other no
new
surface densities
If we
course repel each other, so that this energy is positive.
should further break up each shred into infinitely short lengths,
and separate these from each other, we should have to do positive
them apart, and if we should remove all the parts

from each other, it has been shown by Lord
distances
to infinite
Kelvin* that we should have to do exactly as much work as was
work
to pull
obtained by separating the shreds.

be defined by the first operation
must
energy
Consequently the
before
alone.
Development of Potential of Polarized Body in

We have seen in 123 (7) that
Spherical Harmonics.
127.
the potential due to a doublet placed at the origin is a spherical

harmonic. We may develop the potential due to any polarized
If we call r and r'
distribution in a series of spherical harmonics.
the distances from the origin of the attracted point x, y, z, and the
point of integration
we have
if
<
where
for
a, 6,
c,
so that
the distance between the two points, by
100
(22),
r,
p, the cosine of the angle
(ax
Inserting
this
by
between
+ cz)lrr
r,
r,
is
f
.
P P P
value and those of
lt
Q,
100,
we
have
1__1
d~~r
*
Thomson.
Magnets."
ax + by
3
"On
cz
3(aa?+ ly
+ c/) -rV
2
the Mechanical Values of Distributions of Matter, and of
Papers on Electrostatics and Magnetism,
p. 437.
126, 127]
241
Now inserting this value of "l/d in the expression for the

122 (3), in which 1/r is to be replaced by l/d,
potential in
or performing the differentiations, #, y,
z,
r being constant,
jjjc (1
and collecting in powers of x, y, z, which can be taken out from
under the integral signs, we get the development in spherical
harmonics
Ax + 5 +
where the
coefficients are the definite integrals
C=
A=jjJAdr, B=jjJBdr,
L=
jjJAadT,
P=
Cz
(Bc
jjJGdr,
M=jjJBbdr,
Cb) dr,
Q=
N=JljCcdT
(Ca + Ac)dr,
R=
(Ab +Ba)dr.
In like manner the coefficients in the harmonics of higher

orders are definite integrals throughout the polarized body of
the components of polarization multiplied by powers of the co-
ordinates of the point of integration.

By a change of the origin
the integrals L, M,
may be made to vanish. For putting
a
W.
E.
= a + a',
=b +
&',
= c + c',
16
242
[PT.
I.
CH. V.
we have
L = a, (llAdr + nlAadt =
aÂ
M= b
N=c
and
if
jUsdr
dr
jjJBb'
+L
f
,
S+M
',
jjjCdr
we choose
_M
_L
~Z'
_N
=
'
~B'
the integrals L', M',

vanish, and V_ 3 reduces to three terms.
The values of the integrals A, B, C, are not changed by this
change of origin, but those of all the others are.
The new
origin is called the center of the polarized distribuIf the polarization is uniform, it is the center of gravity
If we find a vector
of the body.
tion.
A, B,
T;r
V-+-
C,
we have
If (cos (Mx) cos (rx)
-f
cos
(My) cos (ry)
+ cos (Mz) cos (rz)}
^
_
M cos
Mr)
i*
equal to the potential due to a doublet of moment

is called the moment of the polarized
situated at the center.
But
this
is
body, and since at great distances the first terms are relatively
the most important, we see that at great distances the body
The line through the center
acts as if concentrated at its center.
having the direction of
is
called the axis of the distribution.
PART
II.
ELECTROSTATICS, ELECTROKINETICS
CHAPTER
ELECTRICAL PHENOMENA.
AND MAGNETISM.
VI.
SYSTEMS OF CONDUCTORS.
Fundamental Experiments. We shall begin the

128.
treatment of Electricity by the description of a number of simple
experiments, for the most part due to Faraday and described by
Maxwell, the explanation of which will devolve upon the theory,
when mathematically
established.
EXPERIMENT
Let a piece of glass and a piece of resin,

I.
which exhibits properties different from those of
ordinary bodies, be hung up near each other by silk threads.
They do not affect each other, and the threads hang vertically.
Let the glass and the resin be rubbed together, and left in contact.
They still exhibit no peculiar properties. Let them now be
separated.
They attract each other, and the strings take an
inclined position.
The system composed of the glass and resin
has now acquired energy, which has enabled it to do work against
neither of
the force of gravity in lifting the two bodies through a certain

distance.
Let a second piece of glass be rubbed with a second piece of

Then it may be observed that
resin, and be similarly suspended.
the two pieces of glass repel each other, and have therefore
acquired energy, which
lifting themselves.
is
evinced by their overcoming gravity in

1C
244
ELECTROSTATICS.
[PT.
II.
CH. VI.
The two
Each
pieces of resin in like manner repel each other.
attracts
each
of
of
resin.
All
of
these
piece
piece
glass
phenomena,
each of which indicates the acquisition of a positive amount of
potential energy, are known as Electrical phenomena, and the
bodies exhibiting them are said to be electrified, or charged with
Electricity.
The
properties of the two pieces of glass are similar, but

to
those of the resin.
What the glass attracts the resin
opposite
Bodies
and
vice
versa.
repelled by the glass and attracted
repels,
said
to
be
are
resin
vitreously, those attracted by the glass
by the
and repelled by the
we
convention
say
resin,
positive,
resinously electrified.
instead of vitreous,
By
general
negative
for
resinous.
EXPERIMENT
Let a hollow metal vessel be hung up by

II.
and let a lid completely closing it be also so hung, so
may be removed and replaced without touching it. Then
silk threads,
that
if
it,
it
the electrified glass be hung inside the vessel without touching

and the lid placed on, the outside of the vessel will be found
vitreously electrified, and the manner of the electrification will be

exactly the same in whatever part of the interior the glass may be.
That is to say, if we place successively at different points of the
external space the
same small
electrified
body,
it will
be acted
upon at each point by a certain force. The direction and magnitude of this force determine a vector called the strength of the
The field may be geometrically repreelectrical field of force.

sented by lines of force in the usual manner. The electric field is
the tangible evidence of the electrification, and the measurement
of a force
is
the means of
its
measurement.
We may therefore
describe the above experiment by saying that the field external to

the closed metal vessel is independent of the position of the charged
body within. If the glass be removed without touching the vessel,
the electrification of the glass will be unchanged, and that of the

If resin be substituted for glass the
vessel will have disappeared.
Such electrioutside of the vessel will be negatively electrified.

of
the
electrified
on
which
bodies, is
fication,
proximity
depends
called electrification
by
influence, or induction.
In this manner a
body may acquire energy without contact with other bodies, and
it is natural to suppose that the energy has passed through the
intervening medium from the electrified body. Such a medium,
128]
which allows
electrical influences to pass
245
through
it,
is
called a
was proposed by Faraday*.
dielectric, as
EXPERIMENT
Let the vessel be positively
III.
electrified
by
induction as before, let a second vessel be suspended by silk

threads, and let a metallic wire, similarly suspended, be made to
The second will be found to be

simultaneously.
and
the
positive electrification of the first is
positively electrified,
touch both
lessened.
EXPERIMENT IV.
If instead of a metal wire
we had used a
rod of glass, sealing wax, or hard rubber, no such effect would have
been produced. Bodies may accordingly be divided into two
classes,
fication
1, those which, like metals, allow a transference of electrifrom place to place. These are called conductors. The
second body above is said to be electrified by conduction 2, those

which do not allow such transfer. These are called non-conductors
:
The dividing
or insulators.
line
cannot be drawn with perfect
no bodies have been found to be absolutely nonAll insulators are dielectrics, but not all dielectrics
sharpness, since
conducting.
are necessarily insulators.
EXPERIMENT V.
In Experiment II
it
was shown that the
external electrification of the vessel due to the introduction of the

electrified glass
was independent of the position of the
latter in
we now
introduce the piece of glass together with

the piece of resin with which it was rubbed, without touching the
We therefore
vessel, the electrification of the latter disappears.
the vessel.
If
conclude that the electrification* of the glass and resin, which are
able to counteract each other's effects, are equal in amount.
By
number of bodies, and examining the external field,

we may show that the induced electrification is proportional to
their algebraic sum.
We thus have an experimental method of
putting in a
adding the
effects of several electrifications
without altering the
electrifications.
EXPERIMENT VI. Let there be two insulated metallic vessels,

and B, and let the glass be introduced into A and the resin
All electrification
into B, and let them be connected by a wire.
wire be removed,
the
let
as
was
to
be
Now
disappears,
expected.
and then let the glass and resin be taken out. It will be found
Exp. Res. ,1168.
246
ELECTROSTATICS.
is
ducing
that
now
negatively electrified, and
and the
[PT.
positively.
II.
CH. VI.
By
intro-
glass together into a larger metal vessel C,
its outside will be found to have no

charge, consequently the
induced charge on A is equal and opposite to that of the glass.
In like manner the charge of B may be shown to be equal and
The charge of A, which is not

opposite to that of the resin.
as
as
the
is
is said to be bound by the
within,
apparent
long
glass
on the inside of A.
of
the
and
resides
inducing charge
glass,
By the withdrawal
outside of A.
of the glass
We
becomes free, and appears on the
it
method of charging a vessel with

amount and opposite in kind to that of a
have thus a
an
electrification equal in
given electrified body without changing
EXPERIMENT VII.
its electrification.
Let the vessel B, charged with a quantity
of positive electricity, which we shall take for a provisional unit, be

introduced into the vessel C without touching it. C will be found
charged on the outside with a unit of positive electricity. Now let

B touch the inside of C. The external electrification is unchanged.
If
be
now removed from C without touching
it,
and taken to a
unchanged, that is, G is

with
a
of
B
unit
but
is completely discharged.
charged
electricity,
If B be now recharged with a unit of positive electricity, and
again introduced and made to touch (7, on removal it will again be
distance, the field external to
is still
found to be completely discharged, and the charge of C will be

This may be repeated indefinitely, and no
increased by one unit.
matter how highly
C may become
charged, it will be found that B is

is a cardinal point in the
This
always completely discharged.

theory of electricity. Since when in contact
virtually forms a
of
state
that
there
is no electrificathe
conductor
we
(7,
may
part
of
a
on
the
inside
conductor
left
to
now
tion
itself.
charged
have a means of charging a body with any number of units of
We
electricity.
on
machine
for
the purpose of generating electricity
this principle is Kelvin's Replenisher,
considered
The
whose theory
will
be
later.
last
experiment
may
be modified by examining the field of

This cannot be done by
force within a hollow charged conductor.
introducing anything through a hole, but was accomplished by

Faraday by building a closed conductor large enough for a person
to remain inside.
Even when the outside was so highly electrified
that large sparks were flying off from
at points within was absolutely zero.
it,
the strength of the
field
"
128]
247
EXPERIMENT VIII. Suppose
that while the pieces of electrified

in
as
are
Experiment I, we surround them with a
suspended
glass
dielectric fluid insulator, such as turpentine, kerosene, or melted
be found that, if the buoyancy of the liquid be

just counterbalanced by weights, the threads will now hang more
It will
paraffin.
The energy
nearly vertically, showing that the repulsion is less.
We see then that the energy
of the system is consequently less.
of a system of electrified bodies depends not only on their charges
and positions, but on the nature of the dielectric medium in which
they are placed. The consideration of the part played by the
is now one of the principal' parts of electrical theory.
medium
any of the other experiments be repeated with the closed

any dielectric fluid, the results will be unchanged,
values of charges induced on a closed conductor
that
the
showing
If
vessel filled with
by charges within are unaffected by the
We
will
now
dielectric.
briefly recapitulate the results of our experi-
ments.
We may
examine the nature and the magnitude of the charge
of any electrified body without altering it, by placing it within

an insulated hollow conductor without touching it, and examining
the charge induced on the outside of the latter.
it
The amount of electricity on a body remains unchanged, unless

be put in conducting communication with another body.
When
a body electrifies another by conduction, the quantity of

electricity on the two remains unchanged.
i
When
electricity is
produced by
friction (or otherwise, as
shall find) equal quantities of positive
and negative
we
electricity are
produced.
When
electrification
is
caused by induction
from
a body
surrounded by a conductor, the amount of electricity on the inside

of the conductor is equal in quantity and opposite in sign to the
charge of the inducing body.
There
is
no
electricity
on the inside surface of a closed hollow
conductor, charged but under the action of no internal bodies.
The
between charged bodies, and their electrical energy,

The
the
dielectric medium in which they are placed.
on
depend
do
not.
within
on
conductors
closed
by charges
charges induced
forces
248
ELECTROSTATICS.
electrification,
II.
CH. VI.
Law
of Force. We
electricity, and charge to
Mathematical Conclusions.
129.
have used the words
[PT.
denote a measurable quantity, which possesses the property of

For if
conservation, that is of remaining unchanged in amount.
by the disappearance of two equal

of
opposite sign, whose algebraic sum was zero.
quantities
need define these terms no further than by their properties, and
for the present, the single property of exerting force is sufficient.
electricity disappears, it is
We
We may
speak of electrification occupying definite portions of

space, for the field of force is such that lines of force issue from
positive electrifications and run into negative electrifications.
examined by examining their
Electrifications being always
of force,
fields
we may
consider the field of force as specifying the

Certain writers have gone farther, and insisted
electrification.
that electricity does not exist, but that lines of force and electrical
energy are the only real entities. Such a question is purely
metaphysical, and of no importance to the physicist. It is
obviously of no importance whether
which exists where

electricity exerts force
use the term
lines
of
force
upon other
"electrical fluid" or
we
define electricity as that

converge, or say that the
electrifications.
"matter" we
If
may do
we wish
so,
to
provided
"fluid" or "matter" simply as convenient terms, without

or
attributing to electricity any of the properties of ordinary fluids
we use
of matter.
It has, so far as
we know, no
inertia,
the fundamental
We
may then define

property of matter, nor is it incompressible.
a charge of electricity as a "something," "fluid," or "matter," which
other
possesses the unique property of repelling or attracting
of
the
two
to
the
charges.
signs
charges of electricity, according
definition the force is proportional to the charge, and it is
By
natural to suppose that the force between two electrified elements

will be in the line joining them, and proportional to some function
Experiment VIII shows that the force depends

on something beside the distance, but if we suppose all space to be
filled with the same dielectric medium, such as air, the assumption
This supposition will accordingly be
is justified by experiment.
made for the present. We shall also suppose all conductors to be
of the distance.
made
of a single material.
We
shall
now deduce
the law of the force from the result of
is no force within a hollow conductor.

Experiment VII,
Let the conductor be in the form of a sphere. On account of
that there
LAW OF
129]
symmetry the charge
FORCE.
249
so distributed that equal areas

possess
is
Let the charge per unit area be a; and let us find

equal charges.
the form of f(r), so that the resultant of all the forces
<rdSf(r)
due to
the elements
all
dS
at distances r from a
given point
when resolved in any direction.
within the sphere, shall be zero
On
on
account of symmetry, the force acting

in the direction of the radius
Q must be
OQ.
We
shall accordingly consider the radial
component R. Let
OQ =
and
6,
the radius
let
of the sphere be a.
Let the distance PQ,

where P is any point on the surface of the
sphere, be r, and let the polar coordinates of
P be 6, $, the co-latitude 6 being measured
from the radius OQ.
force at
OQ
R = fjffdS.f(r) cos 8 = ^ (
We may
is
be
Then the whole
S.
proportional to
a-
(1)
PQO
Let the angle
resolved along the radius
FIG. 56.
2
"f(r) cos 8 a sin Od6d<j>.
.
at once integrate with respect to <,
R = 2?%
(2)
cos B sin Od6.
P/(r)
Jo
Now OQ
is
the
sum
of the projections of
r cos 8
+ a cos =
--
and
PQ on the radius
b,
b-acosO
(3)
OP
cos 8
From
the relation between the sides of the triangle

r
(4)
we get on
=a +6
2
partial differentiation with respect to
dr
^=b
a cos
dr
a cos 6
b,
.,
do
(5)
_b
r~
db~
6,
= cos
B.
Substituting this value of cos 8 in the integral,
~ sin Odd,
R = 27mV C/(r) vb
J o
and
if
we
call
f(r)
= <&' (r),
POQ,
250
ELECTROSTATICS.
[PT.
II.
CH. VI.
so that
J(
_
'~
Sb~
db
'
we have
R = 27raV CO
J- ("<& (r) sin 6d6.
(6)
J o
We may now change
the variable from 6 to r by differentiating
the relation
r2
rdr
=a +6 2
2a& cos
0,
= a6 sin Odd,
rin^-^.
(7)
For ^
= 0,
=a
b,
and
for
= TT, r = a +
rO (r) =
Calling
>,
so that
(r),
we have
fa+b
(8)
Ja-&
(r)dr =
'*'
(a
+&)-
(a -6),
and
R = 27rao- ~
(9)
By
ji
(^(a + 6) -
^ (a -
6))|
the conditions of the problem this must vanish, so that
have the
differential equation for
we
W,
(10)
which being integrated gives
{^
(ii)
(a
+ &)-
(a -b)}
a functional equation to determine ^.

respect to
(12)
6,
0,
Differentiating twice with
LAW OF
129, 130]
This equation holding for
b are entirely
value for
all
FORCE.
all values of
independent variables,
arguments.
251
a and
-9"(r)=.A,
(14)
V(r) = Ar + B = O(r),
(15)
*(r)
= A+*,
(16)
tf(r)
=/(r) = -J.
Consequently the force /(r)

This proof
of the distance.
also
is
is
inversely proportional to the square

due to Laplace *. The law of force
deduced by Cavendish as a consequence of the
a conductor
is
by Maxwell f.
The experiment was repeated very careThe law of the force may also be deduced
from the result of Experiment
The law
fact that
completely discharged by contact with the interior
of a closed conductor.
fully
+ 6 and
have the same
since a
Accordingly, putting r for the argument,
(13)
was
b,
V must
II.
of the inverse square
was obtained by Coulomb by
direct experiment with the torsion balance, but such experiments

could not be exact enough to demonstrate the law with the same
accuracy as by reasoning from the results of the experiments of
Cavendish and Faraday.

130.
Dimensions of Electrical Quantities. Since charges

medium act on each other
of electricity in a uniform dielectric
according to the Newtonian Law, the whole mathematical investigation of Newtonian forces and potentials at once becomes
The volume density of electrification, or the charge

applicable.
per unit of the volume, will be denoted by />, and the surface
density, or the charge of unit area of a superficial distribution,
The charge of a body e is
cr.
(i)
e=jjjpdT+fj<rdS,
and the potential at a point,
Mtcanique
Celeste,
i. 2.
t Electricity and Magnetism,
i.
p. 79.
by
252
ELECTROSTATICS.
[PT.
II.
CH. VI.
the work that must be done against the electrical forces in bringing a unit of positive electricity from an infinite distance to the
is
move from
Positive electricity tends to
given point.
and negative
greater to places of less potential,
places of
electricity the
contrary.
The
unit
of electricity
which
must be defined
as the
amount
of
concentrated on a very small body, will repel

electricity,
a body similarly charged and placed at unit distance from it in
vacuo, with unit force. This unit is the basis of the electrostatic
if
system of electrical units. In the c.G.s. system, the unit of electricity repels a similar unit at a distance of one centimeter with a
force of one dyne.
It is necessary, as shown by Experiment VIII, to specify the
If air were adopted instead of
in defining the unit.
medium
vacuum, that
is,
ether, the difference
escape detection by
ments at present in
consider
all
use,
hence we
experiments to be
The dimensions
would be so slight* as to
but the most refined electrostatic measure-
all
made
may
with very slight error
in air.
of the unit of electricity are found from the
equation of force
ee>
- = Force,
and the
C.G.S. unit of electricity is

1
The dimensions
[p]
of p, a,
= Volume
W-
1
gm.^ cm.^ sec."
V are
density
Surface density
found from
=
= Electrification
Volume
^=
$
[e]
Less than one part in a thousand.
rur^r
[M*l
ELECTRICAL EQUILIBRIUM.
130, 131]
253
The strength of electric field, or the intensity of electric force

at a point, F, is defined as the force acting on unit of electricity
Its dimensions are obtained from
placed at the point.
[Fe]
This agrees with the definition
The energy
H8
of the system
Force
F=
may be
dV
^
which
dn
is
of dimensions
written in either of the forms,
(8),
the integrals having dimensions

[Surface-density x Potential x Surface]
and
or,
= [ML T~
2
],
[Volume-density x Potential x Volume],
118
(10),
the integral having the dimensions

2
[Field-strength
x [Volume]
= [ML T~ ],
2
giving in either case the proper dimensions for energy.

i
Electrical Equilibrium. Suppose we have an electric

field due to the
presence of a number of charged insulating bodies
with
a number of conductors K, insulated and originally
D, together
131.
either charged or not.

cannot move
The charges of the bodies
in the bodies, since
shall assume that the
they are insulators.
We
dielectric properties of the bodies
The electrification of the

manner in the conductors,
charge
es of
virtual work,
the same as those of
air.
s is constant.
By the principle of
find the condition for equilibrium.
any conductor
we can
D are
conductors, however, may move in any

subject to the condition that the total
Suppose that in any assumed distribution V is the potential

due to the total electrification of the conductors K, in which the
volume and surface densities are p and
cr.
Let
V be the potential
254
ELECTROSTATICS.
due
to the fixed electrification of
p.
The
total
D in
[PT.
CH. VI.
II.
which the volume density
is
energy of the system will be
or
Suppose now that
and
o-
to
o-
+ So-,
if
we change
the integral
p in the conductors to p
W becomes W + 8 W, while p
+ Sp
and V'
is unare unchanged at all points, since the electrification of

of
the
conductors
since
the
are
to
be
unAlso
charges
changed.
changed, we must have
&. =
(3)
for
any conductor
pdr +
ads = 0.
The condition for stable equilibrium is that for all possible

values of the functions Sp and <r subject to the conditions Ses = 0,
we must have
BW >0,
58).
Making the above changes
in the integral (i),
we have
dS
and subtracting W, we get

(5)
131]
Now F,
SV are potentials due respectively to distributions of
for F, p in the
densities p, cr in the space
for V, and
space
for
and
in
the space
BV,
8p, Sa
accordingly by Gauss's theorem
t
of
117
mutual potential energy,
(5),
<7$VdS =
(6)
VS<rdS,
In virtue of these equalities, the integral reduces to
(7)
(V+V')SpdT+jf (V+V')**dS
W=fjj
the integrals being taken throughout all the conductors only.

In order to take account of the conditions Se s = we must multi- cg and add the
ply each such equation by an arbitrary constant,
all
sum
to the above value of
SW,
STT-S,c.&.0,
(8)
that
is,
to the first order of small quantities, while the terms
of second order
must be
positive for a
Introducing the values of

Ses
(9)
pdr +
jjj^
II
Se,
minimum.
(3)
adS =
Bpdr
HI
jj
S<rdS,
we get
(10)
(V+ V'-c ) *pdr+
2S
The equations
treat Sp
(11)
and
So-
(V+ V'-c SrdS

s)
of condition having been introduced,
as arbitrary,
and
if
we may
we put in each conductor
F+F'-cÔ,
the above reduces to the terms of second order
+
SVSrdS,
|/// SVSpdr ljj
ELECTROSTATICS.
256
[PT.
which being the energy of the distribution
Sp, So-
II.
may be
CH. VI.
written
which is necessarily positive. Consequently the condition for

stable equilibrium is that in each conductor the total potential
is constant*.
F+ V
The
integral which takes a
minimum
in the above investiga-
the same one that appears in the demonstration of Kelvin

and Dirichlet's principle, 86. We saw that in general there was
tion
is
a doubt as to the existence of a function making the integral a

In the electrical case, experiment shows that there is
minimum.
always an equilibrium distribution, so that the only doubt which

may affect the mathematician does not trouble the physicist.
Reasoning depending upon such physical facts was frequently

made use of by Green, and while not legitimate for purposes of
mathematical
demonstration
is
frequently
of
service
to
the
physicist.
Since in any conductor
F+
c,
or there
is
no force in the substance of a conductor
further
A(F+F') = 0.
But
ductor,
lies
outside of the con-
AF'=
(13)
Consequently, in
superficial.
Now
AF=0=-47rp.
=
every conductor p
0,
or the distribution
at the surface distribution
dV
tinuity in the derivative
But
since the distribution causing

in the conductor, and
we have a
is
discon-
and
Si"
since within the conductor
a(F+F )_
<r
V+
V=
c S)
The above demonstration
p. 164.
aF__aF' = 3F;
rt
'
dm
dn}
is
dni
dne
'
given by Betti, Teorica delle Forze Newtoniane,
131]
the derivative of
for
257
continuous on crossing the surface, as

lies on the surface.
Accord-
is
none of the distribution causing

ingly the surface equation becomes
(16)
dne
The
surface density at any point of a conductor is the derivathe

total potential in the direction of the external normal
of
tive
to the conductor multiplied
by
that
is
rrTT
to the field-strength
away from the conductor divided by 4?r.

due to Coulomb, at least by implication*. The
This theorem
directed
is
a conductor
total charge of
is
The
following form of the investigation is shorter, and depends

on the variation of the second form of the integral
w
F = ^-i fff
(2)
(V^Fy + /9Fy + /aF

U- hr
-H
where we put, as we shall hereafter

heretofore denoted by F+ V,
*
do,
Pfwwwr
fffx [dx
^TrJJJ
._
\dz
\dyj
STTjJJ^ (\dxj
ft
^TrJjK
dx
vff +
dni
dy
dy
for
the total potential,
dz
VSpdr.
Now in Z), 8/3 = 0, and in external space p = 0, so that the

volume integral can be taken through
only, and introducing the
equations of condition $es
2S
(19)
Coulomb.
v-c
= 0,
s " ds
"Suite des recherches sur
la
v ~c*)
distribution
entre plusieurs corps conducteurs. " 1788.

Collection de
pub. par la Soc. franf. de Phys. Tom. i. p. 230.
W.
E.
/>
=0,
du
mem.
fluide 61ectriqne
rel.
la physique,
17
ELECTROSTATICS.
258
necessitating
V=c
dV
5
[PT.
A7 = 0,
p=Q,
dV =
47T0-,
=0,
dne
dni
CH. VI.
II.
aF
ao,
.9,
the same results as above, writing
We may easily show that

For
tion.
values of
for
V + F'
above.
but one equilibrium distributhere be another, "F, <r, for which the constant
if
there
is
are cs let us apply Green's theorem to the difference

,
u=V-V,
the volume integral on the right being extended to
But
outside of the conductors.

47T//,
ductor
and accordingly
Aw =
0.
AV
all
=
in that space
Also at the surface of-
or
any con-
SJ
The
space
AV =
integral
=C 8
Cs
J(u) therefore becomes
Now
the surface integral is equal to

l/4?r times the difBut the
ference of the charges of
s in the two distributions.
charge being originally given this
J (u)
vanishes,
du
5-
ss
osc
Since
distribution
is 0.
Accordingly the integral
and everywhere
da
5oy
du
A
= ^= 0,
oz
u=V -^
V= const.
T7
and
the same in either case.
is
vanish at infinity, the constant
is 0,
and the
Consequently we see
that the constant values of the potential on the surface and

throughout the substance of the conductors, or as we shall say, the
potentials of the conductors, are determined by the electrifications
of
D and the
total charges of the conductors.
ZERO POTENTIAL.
131, 132]
259
Zero Potential. If we have a single charged conduc132.

tor in the form of a sphere, uninfluenced by other bodies, since the
surface density will be constant, the potential at the center will
p
be - where r
is
the radius of the sphere, and since the potential

p
throughout the conductor

increases
earth
may
is
constant, its value will be -
As r
the potential decreases in absolute value. Now the

be considered as a conductor of a radius which is in-
comparison with the dimensions of our apparatus. Its

potential may be therefore regarded as zero, and any conductor may
be kept at zero potential by being connected with the earth.
finite in
If within a conductor there
is
electrified body, the function
a hollow space, not containing

V is harmonic in the cavity,
any
and being equal to a constant at the inner surface of the conductor, must by Kelvin and Dirichlet's principle be constant throughout the cavity, consequently there is no force at any point in the
cavity.
Or a
closed conductor screens the interior from the effect
of an external field of force.
This explains Faraday's Experi-
ment VII.
If the system is composed of a single hollow conductor in
communication with the earth, containing within several rigidly
electrified bodies D, then the total potential V being zero at the
outer surface of the conductor and at infinity is by Dirichlet's
principle zero everywhere in the external space, and there is no

That is, a closed conductor connected to earth shields
force there.
external space from the action of an electric field of force within.

This principle of electric screens is of great importance in practice
in connection with electrostatic instruments.
The
surface density at the outer surface of the conductor

<r
i
-r4-Tr
then vanishes, and the outside
is
dv
^~
dne
without charge.
On
the inside
surface, the charge is
the normal being, as usual, drawn
away from the
conductor.
172
But
260
ELECTROSTATICS.
AF
j
47T
[PT.
vanishing except in the bodies D, in which

p.
Consequently the charge
II.
it is
CH. VI.
equal to
on the inside of the conductor
equal and opposite to the whole charge of the bodies D.

explains Experiment VI.
is
This
not connected to earth, the same result as

There is, however, an outer
far as the inner charge follows.
If the conductor
is
charge, but as there is no force in the substance of the conductor,

the distribution of this charge is unaffected by what is within.
This explains Experiment
II.
Tubes of Force.
133.
If
we apply
the above reasoning
FIG. 56a.
to the space inclosed by any tube of force, which must end either
= 0,
at infinity or at conducting surfaces, we have, since A
any tube of force cover equal and opposite charges,

the flux of force through the tube or the number of unit tubes
contained in it being 4?r times the absolute value of the charge.
or the ends of
Theorems on Sign of
134.
Electrification.
By means
of the properties of tubes of force and of the Potential Function,

we may deduce a number of theorems on the sign of the electrification on the surfaces of conductors.
These theorems are taken
from the excellent Legons sur V Electricite

P.
We
density
in
et le
Magnetisme, by
Duhem.
shall call a distribution in
is
which the sign of the surface-
everywhere the same, monogenic.
we
If the density varies
shall call the distribution amphigenic.
We
sign
suppose
that all the conductors are external to each other, and that in
each case there are no conductors present except those mentioned.
132
SIGN OF ELECTRIFICATION.
134]
THEOREM
I.
201
If the system consists of a single electrified
For on the conductor

conductor, the distribution is monogenic.
the potential is constant, while at infinity it is zero. In free
it
space, being harmonic,

34), hence
has neither
maximum
nor
minimum
the equipotential surfaces are closed surfaces

(
surrounding the conductor, and the tubes of force proceed from
all
the conductor to infinity.
dV
Thence
is
of the
vfl e
the surface of the conductor, and the theorem
THEOREM
is
same sign
all
over
proved.
If the system is composed of two conductors,

the distribution of at least one of them is monogenic. For the
II.
greatest and least values of the potential are two of the three
values of the potential on the two conductors, and at infinity.
The potential on one of the conductors is accordingly an extreme
value, so that the derivative has the
THEOREM
III.
If
same
sign over its surface.
an insulated conductor with zero charge
is
placed in presence of a charged conductor, the charge of the

former is amphigenic, of the latter monogenic.
For since the
charge of the
derivative =
first
is
zero, the
must be
surface-density and hence the
positive
in
some
regions, negative
in
(jflf,
others, consequently its potential lies
between the extreme values,
which are accordingly on the second conductor and at
FIG. 57.
infinity.
ELECTROSTATICS.
262
[PT.
II.
CH. VI.
On the line on the first conductor separating positive and negative

values of &, the potential does not vary as we leave the surface in
the direction of the normal, or in other words the equipotential
surface of which the conductor forms a part has a sheet cutting
This sheet is closed, containing the
the conductor normally.
other conductor with the monogenic charge. This sheet is dotted

The direction of the lines of force is shown by the
in Fig. 57.
arrows.
THEOREM
IV. If the system consists of two bodies with

and
opposite charges, the distribution on each is monoequal
if not, it is evident on inspection of Fig. 57, that
For
genic.
a sufficiently large surface including both conif we draw
ductors,
same
the tubes
direction
of
force
will
cross
(outward or inward).
everywhere in the
But the total outward
it
flux of force is equal to 4?r times the total charge within the
surface, which is zero, accordingly all the tubes of force must
issue from one conductor
THEOREM
V.
and end on the
other.
charge concentrated at a point produces a
monogenic charge on a conductor whose charge is of equal amount

and opposite sign. For this is a particular case of the preceding
theorem.
THEOREM VI. A negative charge concentrated at a point

produces a monogenic distribution on a conductor with a positive
charge of greater absolute value. For the charged point may
GO
be considered as the limit of a conductor with potential
.
then the lowest value of the potential occurring. The

value at infinity, namely zero, is not the greatest value, for then
all the values occurring would be negative, but as we approach
74 (7),
is M/r,
infinity the value approached by the potential
This
is
is the total charge of all the distributions, which is

where
Since positive values occur, the highest value
here positive.
attained must be on the conductor, whose distribution is there-
fore monogenic.
The remaining theorems
are expressed in terms of

of
the conductors.
of
instead
charges,
potentials,
known
two conductors have potentials of the same

is monogenic on the one whose potential
sign, the distribution
has the greatest absolute value, and the density has the same
THEOREM
VII.
If
sign as the potential.
GENERAL PROBLEM.
134, 13")]
263
For the potential of the conductor having the greatest absolute

is, if positive, the highest, and if negative, the lowest value
value
occurring, so that in the former case
a- is
positive, in the latter,
negative.
THEOREM
On
VIII.
each of two conductors whose potentials

is
For the
monogenic.
are of opposite signs the distribution

potentials of the conductors are the
highest and lowest oc-
curring.
THEOREM
If one of two conductors has the potential zero,
IX.
the other a potential not zero, the distribution is monogenic on

both, and on the second the density has the sign of the potential,
on the
the contrary sign. For this is a limiting case of the

as
the potential of one of the conductors approaches
preceding,
first
zero.
THEOREM
concentrated
On
X.
at
For
opposite to its own.

theorem.
Theorem
THEOREM
a conductor connected to earth, a charge

point causes a monogenic charge of sign
may be
XI.
this is a particular case of the preceding
generalized as
In a system formed of any number of con-
For the
ductors, the distribution on at least one is monogenic.
on
of the
or
be
one
must
lowest
value
of
the
highest
potential
conductors.
General Problem of Electrostatics.
135.
If
we have a
number
of conductors in a state of equilibrium, of which some

are insulated and charged with quantities es others connected
,
to earth, or kept,
to
be hereafter described, at given
and influenced by certain bodies D rigidly

with density p, the problem to be solved consists in
constant potentials
electrified
by means
s,
which, 1, is constant in each confinding a potential function

which the
ductor, taking the values
s in those conductors for
constant
is
given, 2, in the bodies
and 3, in the
D satisfies
the equation
rest of space is harmonic,
AF=0.
We
can satisfy these conditions if we can solve n

pendent problems, n being the number of conductors.
+1
inde-
ELECTROSTATICS.
To
I.
all
is
II.
find for each value of s from 1 to n, a function vg ,
at the surface
where r
[PT.
takes the constant value
from
different
is
space external to
CH. VI.
which
at all the surfaces^.
1,
takes the constant value 0, and in

Each of these n problems
s,
harmonic.
is
a different problem of Dirichlet.
To determine a
II.
zero, in the bodies
and in the
function w, which in
D satisfies the
all
the conductors
is
equation
rest of space is harmonic.
+1
These n
functions being found, the required function
V is
given by the linear function
F= V v
(i)
where
V V
1
...
V is harmonic
On any
is 1,
+ Vnvn + w,
...
harmonic in
is
the vs 's are harmonic, and
sum
V.& 2
satisfies
all
Aw =
and
4?r/);
D, where
everywhere except in
conductor
For each of
space except D, where
the given constant values.
V.n are
the functions vs and
therefore the
it satisfies
the vs vanish except v8t which
all
hence
v=v
s.
From any
surface
of the functions v8
the integrals
Qs
-7
dvs
[[
I
and
00|
Vr
vn-JJxrOnt
let
us calculate for any
dw
[(
I
00.
4arJJx,ant
Since the finding of the function vs is a purely geometrical

8t
problem, depending on the form and position of the surfaces
all the n2
are
constants
for
the
rs
quantities q
geometrical
given
system of conductors.
ductor
We
have now
for the
charge of any con-
or inserting the above notation for the

integrals,
(3)
e,
=q
ls
V,
q2s
2 ...
+ qns Vn + Q
135, 136]
There
is
one such equation
265
each conductor.
for
These n equa-
tions determine the charges in terms of the potentials, and if

the potentials of some of the conductors are given, and the charges
of the rest, all the remaining charges and potentials are deter-
Q8
the conductors (including
all
is
consequently
V
v
v\ -V
Vn
r
Coefficients of Induction.
136.
We
the charge of the conductor

8 by induction from
are
to earth, and
connected
8)
mined.
when
v.
Reciprocal Relation.
now suppose the system of conductors to be under

= 0. Then we have
only of their own field, so that Q
shall
action
el
= quV +? F
1
2 ...
21
q nl
the
Vn)
(4)
The constants qrs
are called coefficients of induction, and any
defined as the charge induced on the conductor

8 when
q
it and all the others are earthed, except
r which is brought
to potential 1.
Any coefficient with double suffix qss is the
rs
is
at potential unity, and all the other

s when it is
charge of
It is called the capacity of the conconductors are earthed.
TT
[-].
We
shall
[0~l
now show that the order of the suffixes in qrs is immaterial.
Applying Green's theorem in the second form to the functions vg
and vr we have
(s)
IL (<
*) ds -
The volume
integral being taken throughout the space exthe conductors where vr and vs are both harmonic,
vanishes, and since vr vanishes on all conductors except
r where
all
it is constant and
conductors
equal to unity, and vs vanishes on
ternal
to
except
(6)
where
it is
equal to unity, (5) becomes
266
ELECTROSTATICS.
We may accordingly
The quantity
state the reciprocal
of electricity induced
system when another conductor
tential
and
all
brought to
is
theorem
CH. VI.
II.
upon a conductor
is
of a
brought to a given po-
are earthed,
when it and the others

the same potential V.
The energy
Energy of System.
137.
the others including
as the quantity induced on
except A, which
[PT.
is
the same
are earthed,
of the system
is
(7)
Vn en \
or introducing the values of es
and bearing
in
mind the
relation
W= i?u
(8)
That
the energy is a homogeneous quadratic function of the

potentials of the n conductors, the coefficients being the coefficients
is,
of capacity
We
The energy expressed
and induction.
be denoted by
Wv
in this form will
have
(9)
=#
1S
F,
+# F +
2
28
+ q ns Vn = e
or the charge of any conductor is obtained by differentiating the

energy- function expressed in terms of the potentials partially with
respect to the potential of the given conductor.
Coefficients of Potential.
138.
tions (4)
we
io >
where any
Solving the linear equa-
get n equations
coefficient
prs
is
the minor of qrs divided by
determinant
#11
#21
#12
#22
an
yin
^/27i
>
..
#ni
#712
nnn
*
in the
136
138]
The
coefficients
are
dimensions are
called
of potential.
coefficients
Since the determinant
-^
L*J
267
Their
is
un-
jAl
changed by interchanging columns and rows, the determinant

of the p's must have the same property, or pr8 = p sr
We may
.
we did
prove this directly as

the potential when
for the q's.
has the charge
Let V be the value of

and all other conductors
be the value of the potential when

Let
and the others charge 0. Then as in (5)
charge
0.
charge
and since on any conductor Ki,
spectively equal to
'
and
'
has
are constant and re-
TV,
<*>
Now
is
since the potential
charged,
due
is
to a distribution in
which only
the integrals
all
-^
en
vanish except for

the integrals
= 5,
for
JO = A
dk
which the integral
is
4>7re,
likewise
all
on
si,K |Z<*
.
vanish except for
= r, when
the value
is
4>7re.
Consequently
we have
(13)
Now
from the equations (10), putting
er
= e,
the other
e's zero,
V = p rs e.
s
Again putting
es
= e,
the others zero,
V '=p
r
3r e.
Whence
Pr,=P*r>
and we have the reciprocal theorem
receive a certain charge, e, all the other conductors of the system being uncharged, the potential of any other
If a conductor
the same as would be attained by A if B should

receive the charge e, all the other conductors being uncharged.
conductor
is
268
ELECTROSTATICS.
Making use
of the equations (10)
[PT.
II.
CH. VI.
and the condition prs =psr
>
W = ^-2 V
the energy
S es
becomes
W= %pu eS
(15)
or W is a homogeneous quadratic function of the charges of the

n conductors, the coefficients being the coefficients of potential p.
This form will be denoted by
If we differentiate partially by
e
e
we
any charge s
get
dW
or the potential of any conductor is the partial derivative of the
energy of the system as a quadratic function of the charges, by the
corresponding charge.
As the energy of an
system is intrinsically positive, the values of the coefficients q and p must be such that the functions
e
v and
shall be positive for all possible values of the Vs and e's. We may
Properties of the Coefficients.
139.
electrified,
deduce certain properties of the
from the elementary

Let
properties of the tubes of force and equipotential surfaces.
one conductor s receive a positive unit of charge, all the others
coefficients
being uncharged,
its
is
potential
then
pss> and
the energy
= 4#w,
TPe = Jp^
and since
this
must be
positive
of potential with double suffix
p ss
is
positive, or
A ny
coefficient
is positive.
the same
s has
completely enclosed by
=
conductor
r outpotential, so that for these two p rs
Any
p ss
side of
same sign but of less absolute
s has a potential of the
Any
conductor
value.
For the charge of a conductor
of unit tubes issuing from

conductor accordingly has as
all
tubes have one end on
and the potential of

that of
s and that at
All
coefficients
r,
is proportional to the excess

An uncharged
over that entering.
it
many leaving as entering. Accordingly
rs is
and the other at
infinity (Fig. 57),
consequently intermediate between
infinity,
of potential are
positive,
suffixes are not greater than those with single
and
those with double
suffixes.
138
140]
Secondly, let
which
is
all
269
the conductors be at potential
at potential unity.
The energy
0,
except
9t
is
so that the capacity of any conductor is positive.

The number of
unit tubes of force issuing from
is
s
proportional to the charge
Some of these extend to infinity, while others end on the

qss
other conductors. At the latter the charges will be negative, but
.
sum
of all such charges is not as great as qss

double suffixes is negative, and
with
every q
the
07)
If,
qss
however,
If a
> - (qis + $...+ qs-i, s + q
is
s +i, s
. . .
Accordingly,
+ &,).
completely surrounded by a conductor
new conductor be introduced
r,
into the field, the coefficient
of potential with the double suffix for any conductor is diminished.

For if any portion of the field be made suddenly conducting, elec-
move
tricity will
in
it
so as to
make
the energy less than before.
was the only charged body, the energy ^p ss e* must be

s
diminished, but as the charge e has not changed, p ss must be
If
diminished.
new body into the field increases the capacity of

and
diminishes the absolute value of every coany conductor,
efficient of induction. For if the new conductor and all the others
be at potential 0, while
is at potential unity, some of the tubes
s
of force which before ended on the other conductors, now end on
the new conductor, which receives a negative charge. This induces a positive charge on
its charge q ss and positive
s increasing
Introducing a
charges qrs
140.
charges on the other conductors
diminishing their negative
Work done
during displacement of conductors.
Suppose that we deform or displace the conductors of the system,

thus changing the geometrical coefficients p and q. Suppose the
configuration of the system is specified by
parameters
= 6n.
so that if the conductors are displaced as rigid bodies
Let the mechanical forces due to the electrification be denoted by
<E>
so that the force tending to
change the parameter fa
is
4>.
270
ELECTROSTATICS.
Then the work done
ty
<!> 1
(I)
and
if
no energy
in a displacement
+ <M<fc
S^,
. . .
S<j)
II.
CH
VI.
is
+ 3>,tym,
furnished to the system this work must be done
is
at the expense of the electrical energy
-8TF=2,*
(2)
In the
[FT.
differential S
W and
Sfc.
we may use according
to circumstances
either of the three forms
WeV = ^ V W
s
which are of course

If
variables.
identical,
we choose
es
s,
or
Wv
though expressed in terms of
Wev the
different
total differential
(3)
does not contain the S<'s explicitly. For neither the coefficients p
nor q appear in
However, the Se's and SF's are not indeeV
pendent, being connected by either set of equivalent linear relations

(4) or (10) above, which in the coefficients q or p involve the para-
meters
consequently
<p,
and replace them by
Now we
see
we may
eliminate either the
BVs
or Ses,
Sfis.
by the
relations
e and
v are reciprocal functions ( 63)
with respect to either set of independent variables e l} ...en or
Fj, ... Fn
containing also the independent variables </>, corre-
that the functions
sponding to the variables z of

equations
(5),
63.
Accordingly by the
last of
63,
(4)
If the conductors are insulated, so that all the charges are

so that any force <E>S has the value,
we use the form
e
constant,
from
(2),
The system tends
to
move
so as to diminish the energy.
hand the potentials are maintained constant we

In this case we must supply energy from
must use the form
v
If on the other
140]
271
outside and the equation (5) can no longer be used, but in place of
it
we
have,
by
(4),
*The system now tends
move
so as to increase the energy, and

exactly equal to the work done by the
to
the increase of energy
is
For
electrical forces.
(7)
We
system.
accordingly see that the system is analogous to a cyclic

The forces <E>S correspond to the negative values of the
positional forces
for the latter are defined as the forces that
must be applied from outside in order to equilibrate the reactions

of the system.
Comparing equations (5) and (6) with (i) and (3)
of 69, we obtain the analogous results
'
'
a*.
9ft
P-
*.'?!'.'
3ft'
The
electrical
energy
a*.''
W plays the role of the kinetic energy T
in the cyclic system. In order to determine which of the variables

are to be assimilated to velocities and which to momenta,
e or
we must
recall that in
an adiabatic motion work
is
done through
the positional coordinates at the expense of the energy T. This

corresponds to the case of constant charges (2). The charges are
accordingly the analogues of the momenta, and the potentials of
Accordingly to an isocyclic motion will correspond

a motion in which the potentials are maintained constant. We
have already seen that in this case electrical energy must be
the velocities.
supplied from without, and since this must not only do work but
also increase the energy of the system by an equal amount we
have the analogue of the Theorem I of
In any motion
70
:
of a system of conductors in which the potentials of the conductors

are maintained constant, an amount of electrical energy must
be supplied from without equal to twice the amount of work done
by the
electrical forces
during the motion.

s*
The equations
applicable.
for
the cyclic forces
3B
/Y\
-j?
are
here not
272
ELECTROSTATICS.
[PT.
II.
CH. VI.
In the case of a single conductor, the coefficients

reduce in each case to a single one, the capacity and its
Example.
q and
reciprocal respectively,
and
e
V=pe,
If the conductor
is
a sphere of radius
r,
we have
The only geometrical parameter <j> is here r, and since

tends to decrease,
v to increase, r tends to increase.
If the sphere is elastic, as in the case of a soap-bubble, the

forces tending to enlarge the sphere may be held in
electrical
equilibrium by a greater pressure of the air on the outside than on

the inside, or by the surface tension of the film. If
denote this
is, the force acting normally on a unit

by the whole surface S in increasing the
If T is the surface tension of the film, or
excess of pressure, that

of surface, the work done
radius by dr
is
PSdr.
the elastic force tangential to the surface exerted normally across

a curve on the surface per unit of length, in increasing the surface
by dS we must do work TdS.
Hence we have
= PSdr + TdS = -dW = dW v

e
W = ~,
e
= STrr3 {Pr + 2T},

F = Sjrr {Pr + 2T}.
e*
2
If the soap-bubble
be blown on a tube connected with a manoP may be observed. T may be
meter, the difference in pressure
determined by an observation when the bubble is unelectrified.

Calling r the radius under these circumstances, P the pressure,
T= -
being negative,
ELLIPSOID.
140, 141]
and using
273
this value of T,
F = 8?rr {Pr-P r
2
}.
may be measured
Accordingly a potential
manner by a measurement of
in
this
simple
and
and r having been

r,
on
which
tube
the
If
the
bubble
is
blown is open to
observed.
=
the air, P
0, and
Distribution on an Ellipsoid. We have found the

potential due to an equipotential layer of amount e distributed on
an ellipsoid of semi-axes a, b, c to be, 109 (12),
141.
V(a
where X
is
+ s) (6 + s) (c + s)
the greatest root of the cubic
^
.a
At the
'
+X
[-
62
.-
+X
surface of the ellipsoid
so that the capacity q
H
c
+X
1.
X=
V(<
is
the elliptic integral

2
Jo V(a2
The
+ s) (6 + 5) (c + s)
surface density of the charge
JL
<^_
47ra^~
which by
is
given by
JL f^Yh
47rV"dX
=o
110 gives
a4
64
+
c
Ellipsoids of Rotation.
If
a=
6,
the ellipsoid
is
one of rotation, arid the
elliptic integral
simplifies into
ds
sin" 1
W.
E.
Va2 - -c2
.,
,
if
>
c,
18
274
i.e.,
ELECTROSTATICS.
when the
ellipsoid is oblate,
Jo
when the
II.
CH. VI.
and
Vc 2 - a 2 +
[PT.
if
2
c>a,
(a
The capacity
ellipsoid is prolate.
i\
sin"
<$
1
is
02
_^
Va2 - c
oblate ellipsoid,
la
'..
For a very long prolate
so that as
more
a approaches
slowly,
viz.,
ellipsoid, neglecting (-)
w/
zero, the capacity
approaches zero, but
logarithmically*.
In the limiting case of an oblate

circular disc,
prolate ellipsoid.
whose capacity
ellipsoid, for c
"
2
~*
+ y*
6 4+
ll
I
'
7,2 /
f
,72
v&
-S
ly*
0,
we have a
is
If in the expression for the surface density

the equation of the ellipsoid,
^8
Z*
tf
1
7A "a*~
log
we eliminate
by
If
and
if
we now make
CONCENTRIC SPHERES.
ELLIPSOID.
141, 142]
0,
we get the
a-
275
density on an elliptical disc
b,
Va2 - r2
4?ra
'
At the edge
for a circular disc of radius a.
of the disc,
and the density is infinite, so that this case is not physically

It is however of considerable theoretical importance.
possible.
For the case of the circular disc the potential at any point
becomes
x
^s
rjr
-i
where X
is
-i
'
Vx
'
the greatest root of the quadratic
a2
+X
+X
Suppose we have a sphere of
Concentric Spheres.
142.
^H _ e
a
~~Zj\(a + s)^s~a\2
2
R^ surrounded by a concentric spherical shell of radii R

and R s
In the space between the conductors and outside of the
radius
outer,
V satisfies
the equation,
dr*
whose integral
88
(7),
r dr
is
dr ~r*'
V=--+R
T
If
r
l
is
the potential of the inner sphere,
of the shell,
182
ELECTROSTATICS.
276
which determines
The
surface density on the inner sphere
J_9F =
l
47T
1^3F" ~
=
(R 2 -
_!_
4?r r 2
'
RJ
is
In like manner, at the surface r

charge
=R
_ D
T>
'
l)
==
charge at the outer surface r

redetermine the constants A and B. Since F =
V
V
~~
T>
>
r2
dr
and the charge at the outer surface

e?
The whole charge
kirRfffs
VR
of the conductor 2
is
- F,)
4 + e " = JK^\(V.
2
We accordingly
r,
we
^1
the
e,
by
differentiating
:,
2 ',
=
find
CH. VI.
is
4vr dr
The charge on the sphere
get for the
II.
4>7rdn e
To
[PT.
have
for
It,
the coefficients q
T>
=R
we must
CONDENSERS.
142, 143]
143.
creases,
277
Condensers. The capacity q n decreases as R 2 inwhen R 2 = x

becoming equal to
Accordingly by the
presence of the envelope the capacity of the sphere
is
increased
JD
may be made very large. Such an

K!
arrangement of two conductors, by which the presence of the
in the ratio
-=- which
KZ
second largely increases the capacity of the first, is called a condenser, for by it a larger quantity of electricity is condensed on
the first by raising it to a given potential, the second being to
earth.
The coefficient q u or
which is here equal to it, is
<? 12
,
called the capacity of the condenser, and will be denoted by K.

It is not necessary that one conductor shall surround the other.
If
we
it
does not,
we
have q l2
shall not
q n but in any condenser

,
q l2 to be nearly equal.
suppose the coefficients q n q Z2
In that case we need not distinguish between the two conductors,
shall
or plates, of the condenser.
The energy
is
F - V^ + i (2* - 0n) F + ten + qu)

2
i 0u
FxF2
In virtue of the supposition made regarding the qs, the

two terms are small compared to the first, and we may write
last
or the energy of a charged condenser is proportional to the square

If one of the plates
of the difference of potentials of the plates.
is to earth this is accurately true, and this is generally the condition in practice.
Now e =
=
e*
l
F=
M F! + ^ F =
U F!
12
= -e = S (F
r
or
e1
- F ) + (5n 4- 0i ) F
- FO +(gf - U ) F + (0M +
U(F
U (Vi
ffl
U ) F,
-F
2 ),
to the same order of approximation, and the two plates of a condenser receive equal and opposite charges, proportional to the
difference of their potentials.
7
we get
Using
-F = -^
2
in TF,
ELECTROSTATICS.
278
[PT.
or the energy of a condenser
II.
CH. VI.
is proportional to the
square of the
Since the forces tend to cause
v to in-
charge of either plate.
the potentials remain constant, and

e to decrease, if
the charges remain constant, the capacity
tends in any case to
crease, if
increase.
In the case of the sphere,
if
be the area of the sphere whose radius

of RI, -R 2
144.
is
small,
is
be denoted by
is
r,
and
the geometrical
if
mean
R
If T
S
.
4-7TT
approximately the surface of either plate.
Concentric Cylinders.
If the internal conductor
is
the external a hollow cylinder of

radii R% and
both of very great length, at a sufficient distance
s
from the ends we have
dependent only on r, and in the space
circular cylinder of radius
l}
between the conductors
88
(9)),
d*V
dV
+ - ~j-=0,
AF=-j-r
r dr
dr
2
dr
The
potentials of the conductors being F,
and
V!=A log R, + B,
R,
At the inner
surface
JL d-Z=
^=--74-7T dn
e
The charge on a length
JL
^Z.
4vr
3r
logI
of the cylinder
is
143
PLANES.
145]
At the
surface r
we get
279
in like manner,
Accordingly the capacity of the condenser formed of

length of two conductors is
If
we put
R = r, and
l
J\.
is
=
4-7TT
47T7-'
the surface of one condenser plate.
Parallel Planes.
145.
units of
consider T small,
2r
where 8
If the conductors are two parallel
planes, of great extent, parallel to the plane of XY, at a sufficiently

is
great distance from the edges in the space between the plates,
independent of
x and
y,
and
dV
If z is
measured from the plate whose potential
the distance between the plates,
The
surface density on the plate 1
4-7T
dn
4-7T V c
is
is
lt
and T
is
280
ELECTROSTATICS.
and on the plate
[PT.
II.
CH. VI.
2,
4?T V 32
The charge on an area S
The
z=r
4-7TT
of either plate
capacity of the condenser
is
is
agreeing with the results in the two preceding cases.
In fact, for any condenser in which the two plates are separated
a
small distance, which is the same over the whole of their
by
opposed surfaces, we may use the above value for K.
Standard Condensers. For the purpose of furnisha

standard
of capacity or for measuring quantities of electriing
city when their potentials are known, condensers of one of the
146.
three forms just described,
plates, cylinders, or spheres, are
viz.,
The
spherical condenser is the only one for

which our formulas are exact, for in the other two cases some of
the dimensions have been supposed infinite, and we have dis-
nearly always used.
regarded the charges on the backs of the plane plates, or on the

outside of the outer cylinder.
This difficulty is surmounted in the
B
FIG. 58.
(Fig. 58) be separated

following manner. If a portion of the plate
from the surrounding portion C by a narrow cut, but be placed in
conducting connection with (7, the charge on A, if the edge of the
outer portion
is sufficiently
remote, will be that calculated, for
145
ELECTROMETERS.
147]
281
between A and B the field is uniform. A

being then disconnected from C, its charge may be used
itself.
The outer part
by
G is called the guard-ring, and its function is to render the field
uniform all over the working conductor A. In order that there
may be no
charge on the back of A, the guard-ring
is
made
part of a closed conducting box D, which has no charge in its

inner surface, hence none on the back of G and A.
The principle
of the guard-ring is due to Lord Kelvin*.
It may equally well be
applied to the cylindrical condenser, by separating a portion of the

outer cylinder from the ends, which are connected with an
enveloping annular box.
Absolute
147.
Electrometers. The potentials of the

V1 and F2 the energy is
plate of a plane condenser being
The
force tending to increase r is
'
'
dr
The negative
sign shows the force to be an attraction.
If the
working plate be hung from a balance, and counterbalanced by the

weight of a mass M,
We
thus have an electrometer, or instrument for the purpose of
measuring differences of potential. Lord Kelvin's f original instrument has the plate B carried by a micrometer screw, so that r can
be varied, while A is hung from a system of springs, whose
In this case Fj
tension, replacing Mg, is constant.
to
r.
In
the
balance
form, used by
proportional
others, r is constant,
and
have in this case a practical

*
loc. cit.
358.
is
directly
We
proportional to JM.
difficulty, in that if the upper plate
l
is
Electrometers and Electrostatic Instruments.

on Electrostatics and Magnetism, p. 263.
Rowland and
B. A, Report, 1855.
Papers
282
ELECTROSTATICS.
[PT.
II.
CH. VI.
approach too near, the force becomes greater and the plate is
attracted still nearer, and is accordingly in un.
stable equilibrium.
The electrometer
of Bichat and
Blondlot*
consists of a cylinder with rounded ends inserted

concentrically into a hollow cylinder and sup-
ported by a balance.
If the ends of the internal
cylinder are far enough from the end of the

external cylinder, the distributions upon them
will
FIG. 59.
be independent of the depth to which the
internal cylinder enters the other.

For a certain
distance the field between the cylinders (whose
equipotential surfaces are shown in Fig. 59)
will
of infinite length.
be the
same
as
the
if
cylinders
Let the length of this portion be I
were
Then
we have
-,
2 lo
3Z,
and the
force tending to increase
cylinder into the outer,
I,
that
is,
to
draw the inner
is
dl
The
the
difference of potential is proportional to the square root of

force, and independent of the position of the inner cylinder,
provided only that the cylinders be long enough. This electrometer is of coarse less entitled to the designation absolute than
Lord Kelvin's, on account of the assumptions made.
We
have now seen that in any absolute measurement of

potential, we must measure a force and certain geometrical
quantities of the nature of lengths.
*
Journal de Physique, 2e. S6rie,
t.
v.
ELECTROMETERS.
147, 148]
148.
283
The
Symmetrical Electrometers.
described
last
electrometer forms an intermediate type to the Kelvin absolute

lectrometer and the class of symmetrical electrometers, of which
Lord Kelvin's quadrant electrometer
the commonest example.
is
Suppose a conductor C, Fig. 60, in the form either of a thin

plate or a cylinder, to be surrounded by two conductors A and
/>'.
FIG. 60.
composed in part respectively of

together forming a box enclosing
parallel planes or cylinders,
We
C.
and
must then consider the
coefficients
#33,
#22,
#11,
#12,
#13,
#23>
where the suffixes 1, 2, 3, refer to A, B, C. As in the last

example, the distributions on the edges or ends of G will be unBesides this there
affected by a slight change in its position.
be charges on portions where the
will
field is
uniform, and pro-
Sf
portional to j-
the field
is
where
uniform.
S is
that part of the surface of
we
If
displace
position by changing a coordinate
amount proportional
to
#11
If
11
and C are
and
0,
at potential
Cl#,
cl are positive constants.
at potential 0,
#22= a 22
from the symmetrical

shall change S by an
are at potential
charge
If
0,
on which
0.
Accordingly, if B and
unity has the charge
where a n and
C
we
at potential unity has the
c 2 0.
at potential
is
not affected
are at potential 0, the charge of
is
unity
#33
33
+ (GI -
by
at potential unity,
the position of C at potential
on
If
is
12,
=
#13
13
at potential
whereas the negative charge
contains a part proportional to
=
#12
c 2 ) 6.
6.
C l&>
Accordingly
#23
2S
+ C.B.
284
ELECTROSTATICS.
If the apparatus
is
CH. vi.
II.
[PT.
symmetrical
13
Cf
(7o
Ci
47TT
AE
In the quadrant electrometer, the box
is
circular
flat
box divided
connected
into four quadrantal sectors,

alternately to form the two
conductors
and B.
The conductor C, called the needle,

of the shape shown in the figure, and
rotates about its center, the angle of
is
rotation
FIG. 61.
being the coordinate
"
The
0.
couple tending to produce rotation
is
where
Wv =
- c 0)F +
2
V,
F + (a 2
(c,
l3
V.
c 2 ) 0}
+ (a + c
23
0)
FF
2
giving
If the electrometer
is
correctly constructed
and adjusted
and
The needle
by a
bifilar
usually suspended either by a torsion fibre, or

suspension, so that the force of restitution <3> is
is
proportional to the deflexion, the factor of proportionality being
denoted by A.
is
In the usual method of use, the potential

made large in comparison with V1 and F2
of the needle
We may
then
neglect the second term in the brackets, and the deflection is

This is called the heterostatic method
Fj.
2
proportional to
of use, the needle being charged
by an extraneous source of
potential.
In the
idiostatic
method, the needle
is
put into connection
with one pair of quadrants, which are put at the potential to

be measured, the second pair of quadrants being to earth. Then
ELECTROMETERS.
148]
F =F = F,
F,-0,
28")
*-|r*
proportional to the square, instead of to the
This method does not show the sign
of
the
potential.
power
like the former.
the deflection
first
of
is
If the electrometer
is
not in perfect adjustment,
we
use the
more general form
In order to be able to adjust c and c 2 to equality, two of the

quadrants should be capable of motion toward or from the center,
:
one roughly, the other micrometrically, so that the amount of

In order
surface of the needle covered by them may be varied.
to
make the adjustment, we may
to earth
earth,
is
first put both pairs of quadrants

and observe the deflection when the needle, originally to
Calling this # we have
charged.
,
ô = i<A-c )I7,
2
which shows whether c or c 2 is greater. We may then adjust

until there is no deflection, however the needle is charged.
x
If a high potential
proceed as follows
is
not available for
we may conveniently
By means
and two commutators, we may
of a voltaic battery
either
charge either of the quadrant-pairs to a given potential
to earth.
other
the
or
being
quadrant-pair
positive
negative,
We
thus have four combinations, as follows:
reverse
commutator
reverse
The
deflections are given
A6,=
reverse
by
2
i^F^-f J(c -c 2 )F3 -c
1
FF
- c )F + c FF
286
ELECTROSTATICS.
[FT.
II.
CH. VI.
Taking differences we obtain
In this manner we can accurately bring the ratio
C2
to unity.
Whether the adjustment be made or not, and without the

necessity of making F3 large, if we can reverse the sign of V
we may by observing 6 and 2 get the correct value, since
l
so that
is
directly proportional to the difference of the two

mean of their absolute values.
deflections, or to the arithmetical
Induction
149.
Electrical
Machines.
As a
further
example of induction in a system of conductors, we shall consider the action of a class of electrical machines typified by Lord
Kelvin's Replenisher.
This consists essentially of two semi-cylindrical conductors
and B called the inductors, and two smaller conductors C
and D called the carriers, which may be rotated as a rigid system
about the axis of symmetry.
If
be the (positive) potential
any time, F2 that of J5, supposed

if C and D be put in conductthen
negative,
ing communication with the earth while in the
of
at
position shown,
will
have a negative, and
a positive charge induced upon it. Now on

insulating C and D, and turning them until C
is opposite B and
opposite A if C be put into
communication with B, being nearly surrounded by B,
it will give
of
the
the
absolute
value
negative
up
charge, thereby increasing
gives up
being put into communication with
potential of B.
its positive charge, and increases the positive potential of A.
its
The connections of C and D with each other and with A and

are made automatically by contact springs once in each half
revolution.
If
(7l)
and
revolutions,
(n)
KI and
are the potentials of

z the capacities of
A and B after n halfA and B and whatever
INDUCTION MACHINES.
148, 149]
287
conductors they are respectively connected with, their

charges are
If ql and q, be the absolute values of the coefficients of induction

of
and B respectively upon the carriers when they are connected
to earth, the charges induced on the carriers are
on
C and -
( ' l)
These are given up after a half-turn to
on D.
and
respectively,
so that
or in terms of the potentials,
ri
Ai
If
we
ft
A
write (f and p- for the positive constants -~
and
Multiplying these equations respectively by
p and
q,
ft
,
we have
adding and
subtracting, gives
i)
i)
+ q y(n+D = ( p y(n)
_ q JTCn+D = ( p yjn)
Consequently the linear functions'
pVi + qVz and

decrease and increase respectively in constant ratios for each halfturn,
and
= (pV^ + qVJ ) (1 -pq) n

p y(n) _ q yw> = ( p yM _ q y^ ( l + pq y^
pVf* + qV
(n)
o)
increases, no matter what the original signs or values of V

and V2) pVl + qVz tends toward zero, so that ultimately Vl and F2
become opposite in sign, and since pV
qV increases, the values
of the potentials may be made as large as we please, and increase
As n
If the replenisher be turned in the reverse direction

are rapidly reduced to equality.
very rapidly.
and
ELECTROSTATICS.
288
[PT.
II.
CH. VI.
In Lord Kelvin's form of the quadrant electrometer, a replenisher is used to bring F"8 the potential of the needle, to a
,
definite value,
which
is
controlled
by a small guard-ring attracted
disc electrometer, called the gauge. Descriptions of the complete

instrument are found in Lord Kelvin's Reprint of Papers on Electro-
and Magnetism, and in the usual treatises on Electrical

Measurements.
statics
CHAPTER
VII.
METHODS FOR THE SOLUTION OF PROBLEMS IN

ELECTROSTATICS.
Equipotential Layers as Screens.
150.
The theorems
of Green, given in
84, have an important electrostatic application.
By the first theorem we may produce at all points outside
of a closed surface
fication within S,
whose potential
surface
the same
field as is
is
produced by any electriby distributing over the

and if S is an equipotential
V,
a certain surface distribution,

must be
surface, the surface density
If now.
we
is
place on the surface S a distribution whose surface

cr instead of a, its effect on outside bodies will be the
density
negative of that of the internal electrifications.
closed
Accordingly
surface
if
electrified
equipotential
completely surrounding
bodies be made conducting, we may cover it with such a charge
as to completely screen external bodies from the electrical action
of the internal charges.

By the same theorem the magnitude of
the surface charge is equal and opposite to that of the internal
electrification.
By the theorem for an interior point, we see that
upon such an equipotential surface made conducting we may place

a charge that shall shield internal points from the action of the
external electrifications, the magnitude of the shielding surface
charge being
W. E.
now
equal to that of the internal charges.
19
ELECTROSTATICS.
290
[PT.
II.
CH. VII.
As a means of solving certain

Green's Function.
introduced a certain function*,
Green
in
electrostatics
problems
whose properties we shall next consider. Green's problem for a
151.
S may
portion of space r bounded by a closed surface

follows
be stated as
It is required to find a function

conditions
satisfying the following
1.
is
harmonic in the whole space considered with the ex-
ception of a single point P.
2.
function
is
P by the
infinite at
of
any function
V harmonic in
^, =
given at the
called Green's function
is
and pole P.
The problem
so that
is
4>
function satisfying these conditions
for the space r
two solutions
surface integral
(i)
P, but in such a manner that the
harmonic, r being the distance from the pole P.
The value
3.
pole
becomes
is
unique,
and
Gz by
,
if it
has a solution.
For
if
there are
3,
by subtraction
//
any harmonic function V.
for
But by 2,
G,--r and G,--r

are harmonic, so that their difference
6r2 is also
Applying the above result to the harmonic function
harmonic.
6r2 ,
(5)
*
//(ff,
Green, Essay,
applies
it,
5.
The name
Green's Function
however, as does Maxwell, to the function
is
due
G--
to C.
Neumann, who
GREEN'S FUNCTION.
151]
But by Green's theorem
this is equal to the
291
volume integral
'a^-G^y /a^-Gyy)
,
Gz = const. That is, with

which, as in 86 can vanish only if G
the exception of a constant, Green's function is unique. But as in
the employment of the function only its derivative is used, the
constant makes no difference.
Since the function
G --
is
harmonic, we have by
33 (2)
or transposing,
//('-)
o
by
83
(6).
If on the surface
G=
?z
we obtain
Consequently if we can solve Dirichlet's problem for the given space,

obtaining a harmonic function F which takes at the surface S the
values
G = F 4- -
then the function
Conversely if we can solve Green's problem

the space and for any pole P, the equation (1) enables us to
find any harmonic function
from its values at the surface,
solves Green's problem.
for
solving Dirichlet's problem.
The problems
of
Green
and Dirichlet
are
thus
exactly
equivalent.
In physical language, Green's function is the potential due to

a positive unit of electricity placed at the pole
together with
that of the charge which it induces on the surface S made con-
192
ELECTROSTATICS.
292
ducting and connected to earth.

induced charge,
If
dG
cr
[PT.
is
II.
CH. VII.
the density of the
~4^8V
and
(1) is
Suppose that G is Green's function for a certain space, with

the pole P, whose co-ordinates are a, 6, c, and that G' is Green's
function for the same space, but a different pole P whose cof
ordinates are
a', b',
c.
Then there
exists the reciprocal relation
that the values of either function at the pole of the other are
equal.
For
rp
r?
where the suffixes indicate from what point the distance is measured.
Now since F and F' are harmonic, by the property of the two
Green's functions G and G,
(8)
so that
The
last integral
but one vanishes because
F and
F' are harmonic
functions, while on account of the surface values of

last
F and
F',
the
becomes
l
s(}\
rp -^L)ds.
dn
Since both the functions l/rp and 1/ry are harmonic except at
their poles
and P', by constructing small spheres about the
and P' and proceeding as in
83, we find that the two

points
of
the
last
each
other
parts
integral destroy
(each being equal to
GREEN'S FUNCTION.
151, 152]
4-TT/rpp'),
so that
FP = F P
<
293
Accordingly we have for the two points
P and P',
G? = IV + r
(10)
= FP +
PP
= G'P
>
In order to show the dependence of the function

ordinates of
its
pole
P let us write
= g y, *,
= g (#, y, *,
y,z)
G(x,y,z}
(! i)
and
G'
(a?,
on the co-
it
(a?,
a, 6, c),
a', 6', c').
Then by the above theorem

G(a',b' >C ')=G(a,b,c),
(12)
(a', b' t c', a,
or Green's function
and a,
is
b,c)
= g (a,
b, c, a', b', c'),
a symmetric function of
its
variables a,
b, c
b', c'.
152.
Examples of Green's Function. Plane. Let us seek
Green's function for
Let
a given plane.
all
that portion of space lying on one side of
be the given pole, at a distance a from the
FIG. 62 a.
on the left, and let B be its geometrical image in the plane.

Let the distances of any point at the left of the plane from A and
B be r and r respectively. Now for every point at the left of the
plane,
plane the function
where r
= r,
it
-,
is
harmonic, and for points on the plane,
assumes the value
of the preceding article.
dG
_
dn~i~
We
cos (ni r)
r*
It is therefore the function
have then
cos
(rii
r'*
r')
_ 2 cos 6
r2
'
where
is the acute
angle included between the radius r and the
normal to the plane. Consequently, the equation
ELECTROSTATICS.
294
solves Dirichlet's
for
problem
[PT.
II.
OH. VII.
the left-hand side of the plane.
If
we suppose a charge of a positive unit placed at A, and a negative

unit at B, the plane of symmetry will be an equipotential surface
of zero potential, and we may apply the theorem of equipotential
If the plane of
layers.
symmetry
is
made
conducting, and the
1
removed, the conducting plane receives a charge
charge
which screens the space on the right from the action of A. The
cos #/27rr2 so
surface density on the plane is a# =
Gauss's
the
on
whole charge
theorem,
plane is, applying
,
that the
This is an example of the second theorem of

150, the space
on the right being considered internal.
1 at B is said to be the electrical image in the
The charge
plane of the charge
1 at
A.
B are said to be electrical images of

point-charges
each other in a certain closed surface separating them if either one,
say B, produces in the portion of space in which the other, A, lies,
Two
and
the same effect as would be produced there by the charge induced

on the surface made conducting and connected to earth, by the
point
alone, the
image B, being removed.
Planes intersecting in a sub-multiple of two right
153.
angles.
Let us seek Green's function for a portion of space lying in the

acute angle between two planes intersecting in an angle which is
equal to two right angles divided by an integer. Let the planes
be denoted by
and
2,
let
be the
the pole be P, and let
l
P
of
that
that
of
in
2 in 1,
2
3
Pj
2,
in 1,
geometrical image of
and so on alternately in the
two planes.
P in
that of
2,
that of
in
1,
Q2
in
Let
2,
Q be
the image of
and so on. Since the
x
a submultiple of TT it is easily seen that the series of

images will be finite, the Q's and P's finally coinciding. Let the
distance of any point from
be denoted by r, from any s by r8
angle
is
and from any Q8 by r8

Then the reciprocal of any distance rs or
rs is a harmonic function in the space between the planes since
none of the images lie in that space. Also for all points lying on
.
'
the plane
1,
152
and
GREEN'S FUNCTION.
155]
on the plane
for all points lying

1
i_
n
v/j
2,
l-l-o
I-I.o
r
r
r
i-iô
r
295
v/j
r2
\j
v/j
Consequently the function
vanishes for points on either plane, and being harmonic except

is Green's function.
at P,
Two
lie in the space between

parallel Planes. If
successive
the
images will all lie in a straight
parallel planes
in
number.
and
will
be
infinite
line,
Using the same notation as
154.
two
the last example, we have the same equations, and the same
form of Green's function, except that we shall have an infinite
in
series.
155.
Sphere. Let A be the given pole, at a distance a from
the center of the sphere of radius R. Take a point B lying on the
same radius as A, at a distance from the center b such that ab=R 2
.
Then
sphere.
and
If
OMB
triangles
angle at 0,
are said to be inverse points with respect to the

be any point on the surface of the sphere, the
and
are similar, for they have a common
MAO
and the sides including
it
are proportional, for
hypothesis,
--R-
(i)
a'
FIG. 63.
Accordingly, for points on the surface

(2)
/ = r-75
1
and ,
= -Rl-,
a r
by
ELECTROSTATICS.
296
r and
[FT.
II.
CH. VII.
and
being the distances of any point from
respecis
since
harmonic
in
the
Therefore,
1/r'
tively.
space containing
A, Green's function is for that space
1
*-r-a.f
*^L~
n
(3)
+ ~ r'
-rîi
dru
a,
?]-L
2
n
~ _ cos ( r )
i
~^~
dn
R cos faO
*"
'
r*
F<-
(5)
so that the density of an equipotential layer induced

on the sphere made conducting is
charge at
by a unit
(ny)
^-^{^
Now
R cos (n<r')\
^-~w~$'
OMB and MAO we have
JL_ [cos
(e
in the triangles
a-
62
= R + r - 2 J?r cos (X*r),

= E + r - 2.fir' cos (%r'),
2
/2
so that
cos (ntr)
__^___
O)
_ E cos (W') == __ a
a
- (R + r )
and
1 _i_
S^r
T>o
which by
::
r/2
jR (6 2
^
(2) gives
^ = ^-
(9)
jR2
"
4-7T
The whole induced charge

e
and
if J. is
--A
is
ff /cos far)
47rJJ V
-Rcos(n
an outside point by Gauss's theorem
so that
(12)
a~pS
6'=--.
- (R + r' //
)}
2r'
2
GREEN'S FUNCTION.
155]
297
we should
If
place a charge e at the outside point A, and a

at the inside B, the potential on the surface of the
charge
F= er- -f r
e'
sphere would be
The
zero.
action
we make
which,
if
of the charge
at
Re 'a becomes
I
in portions of space
may thus be exactly replaced by making the

and
removing the charge B. Accordingly the
sphere conducting
B
at
A
and
are
electrical images of each other in the
charges
outside of the sphere
sphere.
Suppose now that the sphere, instead of being connected to

and charged to a potential F, then beside the
induced charge it will have a uniformly distributed charge VR of
density F/4?rJ?, so that the whole charge of the sphere is now
earth, is insulated
E=VR- e^,
a
(13)
The
+
F=f
JK
a
surface density
<r
vanishes along the circle
two parts oppositely
J? 2
(F
-a
'
4-7T
into
and
Rr*
(R
=a
Vr
which divides the surface
If however
electrified.
v > a*-R
v < a?-R
(a+Rr
(a-R)*
the surface density is of the same sign all over the sphere.
Since
the action of the induced charge on external points is the same as
would be that of a charge e' at
and the action of the uniform
charge
repulsion of the
is
e at
whole
Ffle
e'e
-3- +71
(15)
a charge VR at the center, the

charge of the sphere on the charge
the same as that
is
L^=
(V
of'
ea
~__
This
if it is
point
making
F=
0, or
negative, so that there is an attraction, when
a
is small; that is if the
sphere is connected to earth,
is
0, or
insulated without charge, or in any case if the charged

is very near to the
On the other hand, by
sphere.
F or E of the same
sign as e and great enough in absolute
ELECTROSTATICS.
298
value,
we have a
[PT.
eR3 (2a*-R 2)
E\>
Electrical Images in a Sphere.
156.
CH. VII.
when
repulsion,
or
images of each
electrical
II.
besides
other,
Points which are
having the properties
described
connected with
above, possess
equipotential layers
peculiar reciprocal properties with respect to the portions of space
There thus arises a
in which they are respectively situated.
method of finding from the known solutions of electrostatic
problems a new class of problems whose solutions can be found.

This method of electrical images was discovered by Lord Kelvin in
1848*. Suppose as before that A and B are inverse points with
and
reference to the sphere of radius R,
being outside. Let
'
other
at
distances
inverse points situated
(M outside) be two
from the center, and at distances r and r respectively
I and I'
and OM'B are similar,
and B. Then the triangles
from
0AM
ab=
= R*.
Suppose a charge e placed at A, and a charge

due to
at
If we call
the potential at
B.
e =
eR/a placed
the potential at M' due to the charge, e',
the charge e, and
since
II'
we have
.
V~r'' e~
If then
and
r'~
we have any number
find their
at
'
images B, and
any external
point,
of electrified points such as A,

the potential of the system
M, then
I'
will
V be
if
V>-
(2)
b~
'
M' the inverse point to

the electrical image of the system A.
be the potential at
which
We
is
shall give
M,
of the system
an analytical proof of the same proposition,
based on the method of curvilinear coordinates.
If x, y, z are the
coordinates of the point M, x', y' /, those of the point

= 2
z'fz and since II'
y
xjx = y'fy =
R
,_R z
,
,_R>x
>
,_R*y
=
>
Papers on Electrostatics and Magnetism,
p. 144.
',
we have
ELECTRICAL IMAGES.
155, 156]
If x,
y z are given, we know

',
we may
299
and the position of M,
x, y, z
so
given by the above equations, as

curvilinear coordinates of the point M, disregarding for the present
that
consider
x', y',
their relation as rectangular coordinates of the inverse point

Forming their differential parameters,
It
is
easily
seen
the
that
x'
surfaces,
= const.,
'.
= const.,
/ = const.,
cut each other orthogonally, for example the cosine of

the angle between the normals of #' = const, and y' = const, is
proportional to
dx' dy'
8^
a^â^
We
(6)
dx dy'
ay
dy'
p,
"^
8^ 8^
have then by
[4 (afy
+ anf + xyz*)
/
87, (5)
AF=^^,(i ^
+ J,fo
'
"
(i
and performing the differentiations

a /!N
?_UV_8F
'
I'*
Now we
have
ra^
.
~~
'
'
4ay| "
ELECTROSTATICS.
300
Forming the
derivatives for
y and
and comparing with the expression
But
1/1' is
harmonic except at
z,
for
[PT.
II.
CH. VII.
writing
AV
above,
we get
and therefore
0,
(8)
If
we put
A'F' =
(9)
and we get the proposition that if V is a harmonic function of the

is a harmonic function of the corresponding
point M, then
point M'.
If the distribution causing
is
ously in three dimensions, the density is p

image the density is p = A' VJ^TT so that
'
If
cfo
and ds be corresponding infinitesimal

y z
we have
surfaces,
for
in the
arcs,
expressing ds
the ratios of corresponding infinitesimal arcs,
and volumes
dJ_&
ds~
The
A F/4?r and
in terms of the curvilinear coordinates x,
so that
distributed continu-
l?_
E2>
dS^^& =
dS
I*
l^
&'
dr^
=^=
dr
l^
R'
ratio of charges of corresponding infinitesimal
_ p'dr' ==
_R_l'
de~ pdr
~R"
de'
and of the surface densities
dS_ J^ = ^
<r~d8 de~ R*~ I'*'
<r__de^
volumes
is
ELECTRICAL IMAGES.
156]
301
It is to be noticed that if the original distribution is an equipotential one, which is the case if it is on the surface of a
conductor, the image will not be equipotential, on account of the

variable factor !/', but if we place at 0, the center of inversion, a
V, its potential
V/l' added to the potential due to
charge
the image gives zero, an equipotential distribution.
any problem of
electrical equilibrium
Consequently
whose solution is known
gives us the solution of the problem of induction by any pointcharge on a conductor whose surface is the inverse of the given
conductor with respect to the point at which the inducing charge
Conversely the solution of a problem of induction by

a point-charge gives us the solution of a problem of undisturbed
is
placed.
Thus the solutions of the problems treated above

equilibrium.
furnish us new solutions.
The image
of a sphere
is
a sphere (including a plane as a
special case) for the equation of a sphere
becomes, using the equations
(3),
x/2 +
y'
that
is,
+ z'
E (x
If
x'
+ y' + z'
x**
'
+ y' + z' ) + BR x +
2
is zero,
originally a plane,
sphere passing through the origin, while

through the origin inverts into a plane.
As an example
'
x' 2
'2
'
+ y' + z'
CRy + DR>z' + AR* = 0.
we have
'
+ y' + z'
if
which inverts into a
is zero,
the sphere
of the method, let us invert a sphere of radius
about a point on its surface, with

a/2 charged to potential
= a. The sphere inverts into a plane
radius of inversion,
tangent to the sphere at the point diametrically opposite the
center of inversion.
surface density
The charge
of the sphere being Fa/2, the
is
Consequently
,
a3
Fa3
535'
ELECTROSTATICS.
302
and
if
we put
at the center of inversion a charge e

>
II.
[PT.
CH. VII.
Fa,
ea
27Tl'*
the density of the charge induced on the plane, agreeing with

the result of 152. Inverting the distribution induced by a point-
is
charge on the two parallel planes gives us the equilibrium

distribution on two spheres tangent to each other, and inverting
the distribution on two intersecting planes gives the equilibrium
distribution on two spheres intersecting at an angle which is a
For the full treatment of these

sub-multiple of two right angles.
and other examples the reader may be referred to Lord Kelvin's
Reprint of Papers on Electrostatics and Magnetism, xiv, XV,
to Maxwell, Treatise, Vol. I, Chapter XL
and
Distribution on Spherical Bowl. As a final example

we shall work out the solution of the most remarkable problem
that has been treated by this method, namely the distribution of
electricity on an open spherical bowl, or segment of a sphere.
157.
the only case in which the distribution on a portion of a

geometrical surface has been solved, except in the case of the
This
is
distribution on a circular plate, the inversion of which gives the

shall not follow the method of Lord Kelvin, but
circular bowl.
We
that given by Lipschitz*, who solved the problem independently,

being unacquainted with the existence of a previous solution.
Let
R be the radius of the sphere of which the bowl is a segment,

Fig.
64,
and
let
the radius of
the
opening be a. Let the surface of the

bowl be denoted by $, and let the plane
surface which closes the bowl, of radius
a and distance c from the center of the

sphere, be denoted by 2.
Inverting
the figure with respect to the center
bowl and radius R, let the
of the
segment into which the

be denoted by
plane
Let the space enclosed between $ and
2 be denoted by T, that between 2
and 2' by T", and the remaining portion of space by T'. Let
spherical
inverts
Lipschitz, Borchardt's Journal, Bd. LVIII., p. 162, 1861.
'.
us
303
ELECTRICAL IMAGES.
156, 157]
now form a
function
potential function,
and
is
which behaves at
harmonic
infinity like a
in all space, except that it is
discontinuous everywhere on the spherical segment S. (If it were

not for the discontinuity, such a function would vanish everywhere.)
Let us also form a function

(i)
W'(x y,z) =
defined by
Rx
Multiplying by l/R
= R/t,
R
(2)
If the values of the functions for points internal and external

to the sphere 8 be distinguished
surface 8, since x
x\y y,z z
TF/=TF.,
(3)
TF' (x, y, z) vanishes for
(4)
(x,
y )Z )
oo
^ F (^,
y', z')
z=o
suffixes i
and
e,
on the
WY=TF.
and
lim Tf
by the
is finite for
since
and
= i Km
Tf (a/,
y' ,
/)
const.
-ftr=<
Let us now put

'
iv7
( e ^
\y)
and we
will
shall
f 'r- -/
\ > y>
z-N)
show that the function
may be
so defined that
be the potential of an equilibrium distribution on the spherical
segment.
We have seen in 141 that the potential at any point due to

an equilibrium distribution on a circular disc of radius a is
e (IT
a \^
(2
where X
is
tan"1.J\}k
a }
the greater root of the quadratic

L
^ + 1 =1.
+X
The derivative of this function according to the normal of

course has a discontinuity by changing sign on crossing the disc.
If we consider the disc
placed in the mouth of the bowl, on
304
ELECTROSTATICS.
account of the change of ^-coordinate

of the equation
ZJZ
~~
+ (*~<0
\
and if X' be the same function

must have
x">
We
will
now
+ y'
of
(af
we must take X
define our function
II.
CH. VII.
as the root
af, y',
- c)
[PT.
z that
is
of x, y, z
we
W by two
different analytic
expressions.
In the space
and
in the space T'
This makes
X=
T we
take
and T"
continuous at
as required, since on the disc
term makes the

normal derivative continuous in crossing the disc 5. By the
we have in T and T" (since the inverse of T is T',
definition of
and of T" is itself)
0,
and the change of sign
in the second
and in T'
W'--(Accordingly
,
T.
we have
-iJ^\
^^.
for the values of
F'=-tan-
V in
ELECTRICAL IMAGES.
157]
The
function
305
for
is
everywhere continuous,
continuous except at S and there, by (3),
(10)
so that
is
=W + W{ = W
'
W=V
e
and
are
continuous.
We
is continuous
have already seen that the derivative of
in crossing 2, and accordingly that of
is continuous in crossing
Now
2'.
the derivative of
continuous in crossing
is
2', since
denned by the same continuous analytic expression in T' and

is continuous in
T", and the derivative of
crossing 2, since
is denned by the same continuous analytic expression in T and
is
as well as
itself satisfy
T".
Accordingly the derivatives of
the required conditions of continuity, and on S, since
ai
we have
V = TT
= af,
y = y, z
= z',
= R,
\'
and V, the function assumed,
potential of an equilibrium distribution on

center of the bowl we have on .the one hand
while in order to employ the formula (9)
X=c
= \,
is
therefore
the bowl.
the
At the
we have x = y =
= Q,
2
.
But when I is infinitely

we see by making
order, as
small,
I
must be
infinite of the
second
infinitesimal in the equation
-+y + U-:
(8)
The terms
of the lowest order are
+*+*='.
Hence approximately
and
W (0) = Km y (f ~ tan~
'
Therefore
03)
we have
E.
T")
lim
T tan
~x
l&
=
lfi'
finally
w.
20
ELECTROSTATICS.
306
If
=
a/c
II.
CH. VII.
7 the angular half-opening of the bowl, a/R =

tan 7 and the charge of the bowl is
we
call
(14)
giving as
its
= R(ir
7)
sin 7,
+ _Rsin7,
capacity
To complete the problem we have
We
[PT.
to find the surface density.
find
dV __1_
8^~
8X
X'2aVx87i;
_L_^
V'
a2
a2
VV\
dl fir
-E
*r o
s - tan
a J
drii \2
.
I-
1
/
8^~
+ ^'2aV\8^
(7T
dne \2
on the surface
v _.:A"
a2
R dl
- T;
5T
Now
8X
dV'_
X1
a2
VV\
+ tan-1 -a
t
$
8X
8X'_
an~~8n'
and therefore
8F_
~
/TT
/V
"
8X
X)
Now the
direction cosines of the normal n e being a}/M,y/R,g/R,
we have
x ax
8x
The quadratic
(21)
(s}(-\\
X(a
for
is,
y ax
t/
z ax
I
cleared of fractions,
+ X)-( +
2
2
i/
)X-(^-c)
(a
+ X) = 0,
30
ELECTRICAL IMAGES.
157, 158]
which being partially differentiated gives

2
[a
[a
(22)
+ 2X - O + f) -(z- c) ] ^ = 2\x,
2
+ 2X - (* + f) -(z- c)
2
Multiplying these respectively by x/R, y/R, z/R and adding

gives
8X
dne
~~
2X [Q2 + f + z 2 ) - cz] + ^tfz(z - c)

R [a2 + 2X - (a?2 + y 2 ) -(z c)2]
Putting now x + y 4- 2
2
(21), the numerator 2 {X (R
the denominator,
2R (X +
=R =a +c
cz) + a ^
2
cz
),
and using the quadratic

a 2 cz}, becomes equal to
multiplied by (a
+ \)/R
so that
finally
8X
(24)
and
9X
dn e
for
dfii
X
=a +
R^
2
the density within and without

1
**"
G"i.
~"
~A
dV_
_1 ^
'^
fvr ~~
^r
~.
~\
we
obtain
tan -i^__?_
,
"
a]
f
(25)
"e
8F'
~
T
^^ = 4t7rRD
4>7r dn
1
~A
(TT
ir
|2
+ tan
.VS.,
- + .al
-7=1
a
VxJ
-:=
uo-u
Vx
-.=.
Vx.
The difference of the densities within and without is Vj^TrR.

The smaller the opening of the bowl, the smaller is the density
within.
At the edges of the bowl, where X becomes zero, the
density
is
infinite, as
in
the case of the circular
disc,
but the
capacity in either case remains finite.

158.
Application of the Conformal Representation to
two-dimensional problems. In cases where the densities of a
distribution are the same at all points situated on the same line
parallel to
a given direction, as for instance, in the case of the
202
308
ELECTROSTATICS.
[PT.
II.
CH. VII.
electrification of very long cylindrical conductors at a distance

from the ends, the potential is independent of the coordinate whose
axis is parallel to the given direction, and Laplace's equation
reduces to two terms.

9
92
Such problems are
called two-dimensional, or uniplanar, since

the same in all planes parallel to a single one.
For the solution of two-dimensional problems the method of the
the distribution
is
conformal representation by means of functions of a complex

Let
be a function of as, y,
variable gives a powerful method.
in
a
certain
of
the
Let us make a
region
plane.
holomorphic
of
the plane by means of two conjugate
conformal transformation
functions u,
V then
v.
That
formed plane.
becomes a function of
u,
is,
are
to
u, v in
the trans-
be taken as rectangular
Then to every point

in
coordinates in the transformed plane.
the original plane, having the coordinates x, y, there correspond
definite values of the functions u, v, and the point P' in the
transformed plane having the coordinates u, v corresponds to the
and P' belong the
To corresponding values points
point P.
V in the JTF-plane correin the ?7F-plane.

form
We shall
spond
first show that a certain function of the derivatives of F remains
unchanged if we replace in it one set of rectangular coordinates x, y
same values of
Thus
V.
level lines of
to level lines of different
by the other
u, v.
Considering
F as a
function of u, v
dVdv
^
du
dx
fix
dv dx
dVdv
dV^dVdu
'
dy
du dy
dv dy
~~
dx*
du*
du dx*
dvdu dx dx
dx
d*Vdudv_
dudv dx dx
~
=
d*v
f
dv*
(dx
dv dx2
'
'
'
\dyJ
dvdu dy dy
du dy 2
dv
'
CONFORMAL REPRESENTATION.
158]
Adding the second
derivatives,
and making use of the funda-
mental equations of conjugate functions,
we
42 (A),
44,
obtain
(3)
\3J
S^d
\Jw
we
If
call
'V
V "
we
309
4+
see that in a conformal transformation the second differential
parameter A'F of any function F in the transformed plane is equal

for the corresponding
to the second differential parameter
in
divided
the
the
original plane
by
square of the ratio of
point
AF
linear magnification h
a harmonic function of
of u,
-=-
at the point
(Hz
x,
is
43).
Consequently
transformed into a harmonic function
v.
In
like
manner squaring the
first
derivatives
and adding, we
obtain
a
Calling
we
see that the square of the first differential parameter h v possame property with regard to the transformation.
sesses the
Dividing equation (3) by (4) obtain
AF ~=
2
aF\ 2
/aF
y
or the ratio of the second differential parameter of any function to

the square of the first is unchanged by a conformal transformation.
We may
call
such a quantity an invariant of the transformation.
310
ELECTROSTATICS.
[PT.
II.
CH. VII.
=G
require the condition that an equation <f> (x, y)
In this case we shall
represents an equipotential family of curves.
have for the potential function V, V=f(<f>) and as in 108, (2)
We
now
(6)
so that
ifr
d>V
3-
The right-hand member depends on alone, consequently the

= C shall
left hand must also.
Consequently in order that $ (x, y)
c/>
represent an equipotential family the ratio of the second to the

must be a function
square of the first differential parameter of
<
of
and
alone.
let

Let now
(u, v)
y)=C represent an equipotential family,

C' be the transformed family.
Since by (5)
(x,
(f>
V v
'
2
and since A^/A^ 2 depends only on
A'^//^/
<>, for $ and are constant together.
c/>,
will
depend only on
Accordingly a conformal transformation leaves every equipoIt is upon this property that the
tential family equipotential.
other physical problems depends.
and
application to electrostatical
we
over a portion of the

does not vanish, using the element of area in
curvilinear coordinates
If
integrate the second parameter of
JTF-plane where
it
dSa =
7
dudv
.
-j-
dudv
<8>
and now considering the second integral to refer to the transformed plane, and e and e' to be charges of corresponding regions,
(9)
=
1
dudv
or corresponding regions in the

densities being different).
= e,
two planes have equal charges (the
158, 159]
we have
If *P be the conjugate function to V,
upon any conductor

B
e=
[
I
crds
--r-
JA
VG between
311
points
for the
charge
and B,
[*dVj
ds
4>7r]
A dn e
1
fiV C
f*
^r J A \te
or since
cos (n x) ds
--
(fa/,
cos (WT/) ds
dx,
CM?
/
C
-
4<7rJ
so that the flux of force of
dy
any tube of force
times the difference of the values at

function to the potential, as in
159.
by means
of the function
log r
103.
w = log z
= log V# + y
2
giving
2
transforms into a pair

r=r2 To the potential
.
&
-.
= ^80
= and
"^
<f>
/
v
(4)
x+
u +iv = - iy+a
^~
x + iy a
result.
Replacing
/*2
of the fractional linear function
new
that
2-Tr,
(*
-I
w = (z + o,)/(z
a),
in
i t as
a reference to the theory of the complex variable shows
always possible, gives
(5)
2)
gives us an important
by
is
&
T7"
The use
is
= tan"
F= u
and the capacity

\
<j>
we have the conjugate function

the charge of the cylinders we have between
function
Let us transform
45),
u=V ,u=V
l
pair of parallel planes
of concentric circular cylinders r = rl
for
l/4?r
two sides of the conjugate
its
Eccentric Cylinders.
Examples.
( i)
measured by
is
u-iv=
x - iy + a
-
x-iy-a
312
ELECTROSTATICS.
II.
[PT.
CH. VII.
and multiplying together,

u*
(6)
+v
(x
We
- of +
now make use of the results of the preceding example,

however,
denoting,
by our present u and v what were there denoted
will
by x and
y.
The
cylinders
U
transform into
~,2
(7)
'
rj,
>
which on clearing of fractions,

($}
^ + y^ ^ "- rf + 2ax ^ +
>
(tf + y
(1
r22 )
Tl
+ 2ax (1 + r
^ + a'( l ~ r^ =
2
-r )=
) + a? (1
'
0,
are seen to be eccentric circular cylinders.

Their trace on the new
is
in
shown
which
JTF-plane
Fig. 65,
represents the transformation
of the right-hand part of
Fig. 24
w = log
If
we denote
for either cylinder
of the center from the origin
by
by means
z
of the function
the radius by
d, since
FIG. 65.
R and
we may
the distance
write (8)
158]
x"-
(9)
313
+ y*l
=0.f(i^y_iU
'
r(i-
((I-T-)
we have
r + 1/r
r + \
= ar= a --TT:>
2
(10)
R=
from which
(ID
and r must be positive, we take the upper signs in
(i i) for r > 1, which makes d > 0, and gives the circles on
the right, the lower for r < 1, which makes d < 0, and gives the
circles on the left.
Since
.ft
(10) and
Now making
for the functions
use of the results of the last example,
V and ^
V log r = log
(w
-f-
we have
v2 ),
*-tA
w
so that in the transformation, for r
=r
2,
= log n = log
,
= r1}
log r2
= log
and the capacity of the pair of eccentric cylinders
is
K=
(H)
(d^
*(d2
In case r2 =l, d2 and
become
infinite,
and we have
for the
capacity of a single cylinder in presence of the infinite conducting
plane x =
0,
(15)
The formula given above

of which one
since
we
is
for the capacity of
internal to the other
are given not the distances
is
dlt
a pair of cylinders
not convenient in practice,

d2 but only their difference
,
314
ELECTROSTATICS.
= d.
(16)
CH. VII.
= a~-
-,
r-j,
r2
= 2a
'2
d2
(20)
so as to obtain
R,
(19)
'
2
/ 2
^ = 2a-^-,
(18)
d2 = a
(17)
j-
/i
d^
^-,
~~
-
= d,
R 1} R2 d. We need for
d
d
Eliminating 1} 2 from (20) by (16) and
in terms of
c? l5 c 2 ,
a,
use only the ratio r^r*.
(17) and (19),
and a by
(18), (19),
7'2_1
Taking the sum and
,2
difference of these
two equations we obtain
R r - R,r, = d,
(23)
/
II.
d 1} the distance apart of the lines of centers, together

R l} R 2 We must therefore solve the equations
with the radii
(1 8),
[PT.
-*-*'2
( 2 4)
-^^1
=d
7
Multiplying these equations together
r/y
2
'
or
,
r2
a quadratic
_
~
for r^r^ or rjr-^.
Solving we obtain
It is easily seen that taking the square root with one sign makes
the whole expression the reciprocal of its value with the other
Consequently we use the upper or lower sign according as
sign.
r2
is
greater or less than ra
The capacity
is
accordingly
log
which
for
becomes
1/2
log(^ 2 /^j) as
in
144.
CONFORM AL REPRESENTATION.
159, 1GO]
315
If the two cylinders are external to each other, we must insert

the minus sign on the right of (18), so that the equations are the
same as before, with

we obtain
its
replaced by
negative.
Accordingly
-Rp-tp}},
(29)
K=
-(B + B )^[<l'-(Ii + R,y][d -(Bi '
,,
21
The formulae
!f
~îTn~
(15), (27)
and (29) are important
in calculating the
capacities of telegraph wires.
and Hyperbolic Cylinders. In the prewere given a function of a complex variable,

we
ceding examples
and from that obtained a conformal representation. We will now
consider a case in which we are given a set of orthogonal curves,
and we shall seek a function of a complex variable which will
make them the conformal representation of orthogonal straight
The functions \ and defined by the equations
lines.
160.
Elliptic
//,
T 7o
f
,
*>
>
'
Solving for x and y we obtain
are a pair of orthogonal coordinates.
+ A,)(a
a -6
/(a
-V
y
+7*)
'
V/ffl+XXft'
&*- a
+ Ai)
and differentiating logarithmically

c?
d\
2 Va2
+X
a2
+ /V
d
'
\
At/'
ELECTROSTATICS.
316
[PT.
II.
CH. VII.
so that
(4)
b*
'
62
+X
For a conformal relation we must have
Consequently
if
we put
d\
du
du,
the functions u v will give us a conformal relation in which the

= const., v = const, in the 27F-plane correspond to
straight lines u
the ellipses and hyperbolas X, //-, in the XF-plane.
t
Integrating the differential equations (6)
u=
(7)
log (Va
=1
*=22
+ X + V& + X},
6
4- b
2/A + a +
(2yLt
2
cosS
C
a*-6
Taking the antilogarithm and

u
e
its reciprocal,
+X+
X,
_ M __V
Solving these for
Va2 + X and
Va2 + X =
From
the integral for v
\/6 2
M
{e
+ X,
+ (a - 6
2
we get
+ cos 2v
-
cos 2o
=c
_,
of the
first
equation,
160]
317
which give
Vo 2 +
= cos v Va
b 2,
/JL
V& 2
4-
//,
= sin v V&
a2
Inserting the values of the four square roots in the values of x
and y
in (2)
=i
=
y
a?
(12)
+ iy =
M
[e
{e
+ (a - b ) e~u
- (a - #) *-}
2
je
(cos v
cos
v,
sin
v,
+ i sin v) + (a - 6 ) e~w (cos v

= 1 [ 6U+iV + (^2 _ J2)
2
sin v)]
which gives the form of the function sought,

z
(13)
w
{e
(a -b*)e-
w
)
or
w = log
(14)
{z
\/z
(a
b%
We may
distance
now conveniently change our unit

Va2 b2 shall equal unity. Then the
so that the focal
function z becomes
table of comparison, of the principal

the hyperbolic cosine of w.
properties of the hyperbolic and circular functions is appended*.
*
Hyperbolic and Circular Functions.
160A.
cosh x = \
x
(e
+ e~ x
sinh x = \
cosh- x = log
(a;
*Jx
tanh x = sinh /cosh x
1
x
(e
-e~ x)
sinh" 1 x = log (a
1)
*Jx*
+ 1)
sech x = 1/cosh x.
(1)
sinh
- x) = - sinh x.
cosh or.
(2)
cosh(-z)=
(3)
cosh 2 x - sinh 2 x=l.
(4)
(5)
sinh (x
i')
sin
- x) = -
sin x.
cosx.
(2')
cos(-a;):=
(3')
cos 2 x + sin 2 x = 1.
(5')
sin
- tanh 2 x = sech 2 x.
y)
= sinh x cosh y
(xy) = siuxcosy
cosh x sinh y.
(6)
cosh (x
y)
cos x sin
sinh x sinh y.
^sino;
tan
(7)
(8)
sinh 2 x =
cosh 2x - 1
(8')
cosh 2x
sin 2
z=
tan y
tan x tan t/
- cos 2x
2
(90
-T-
sinh x = cosh #.
(10')
j-
sin
= cos
sin?/,
.r
+1
(9)
(10)
?/.
= cosh x cosh y
a:.
'
ELECTROSTATICS.
318
By means
of
it
we may show
cosh
II.
CH. VII.
that the four functions
sinh
z,
[PT.
cos z
z,
sin z
give conformal representations which are identical except for interchanges of the axes. By means of the equations 160 A, (15), (16),
(i5')> (16')
we obtain the
pairs of functions
v for the four cases,
u,
and by the use of equations (3'), (3) after division by one of the
factors on the right in the values of u, v we get
w = cosh z,
u = cosh x cos y,
I.
(a)
cosh 2 x
w = sinh
II.
cos y,
w = sin
IV.
w
(a)
-5-
Ax
1
(
4)
(15)
i/
sin
1,
sinh y
sin
cosh 2 ^
sinh 2
(6)
= i sin x.
=
cos x.
(ix)
sinh (# + iy) = sinh
a;
x sinh y
1.
-J
sin2
cos
1,
= cos x sinh y
(6)
?/
(ix)
(a;
= sin & cosh y,
cosh
cosh
1.
cos y
1.
sin 2
cos2 #
(n')
-T-
cos x = -sin #.
(12')
dx
-r-
tan
dx
a;
= i sinh x.
= cosh #.
sin (x + iy) = sin x
(13')
sin (ix)
1
(
cos (ix)
4')
(15')
= sec 2 x.
+ iy = cosh x cosy
+ i sinh a; sin T/.
)
(16')
sinh a; cosh g + 1 sin y cos y

2
cos 2 y cosh 2 x + sin 2 y sinh' x
eo8(*+iy)=co*#ooahy
-i
(.70
(17)
cosh y
+ i cos x sinh y.
+ icosli x sin?/.
(16)
=
sin 2
I/ 2
cosh # = sinh x.
sinh
V2
a?
^7)^
\w /
XJ
(12)
(13)
cos2 y
= cosh
W = COS Z,
u = cos # cosh y
r =
v?
cosh
(b)
_
1
(a)
= sinh x sin y
(n)
1>
sinh 2 #
w = sinh x
III.
tan (. + .,)=
sin
a;
cos
sin
# sinh y.
^
a:
+ i sinh ?/ cosh ?/
+ sin 2 x sinh2
cos 3 x cosh 2 y
?/
'
160]
319
see that the straight lines x = const, correspond to confocal ellipses, in (I) and (II), while the lines y = const, correspond
We
to
confocal
const,
while
hyperbolas,
in
y
ellipses
The geometrical character of all
the
therefore
(It
is
same,
we
to be noticed that
shall
and (IV), the
(III)
and x
correspond to
four
const,
lines
to
hyperbolas.
transformations being
consider only
we have interchanged
case
I,
z and
Fig. 66.
in (13).)
x = const, corresponds the ellipse whose semi-axes

cosh x and sinh x.
When x = the ellipse reduces to the
To any
"
line
FIG. 66.
and u = l,
straight line between the focal points u = l, v =
v = 0.
As x increases the ellipses become continually larger, until
x
infinite they become infinite circles.

For any negative x we
the
same
as
for
the
get
ellipse
corresponding positive. The lines
const,
to
the
y
correspond
hyperbolas whose semi-axes are cos y
for
and
sin y.
When
y=
the hyperbola reduces to those portions of
320
ELECTROSTATICS.
[PT.
II.
CH. VII.
the Z7-axis outside the focal points and as y increases the hyperbecome continually less sharp, until when y = Tr/2 the hyperAs y goes on increasing we obtain the
bola reduces to the F-axis.
bolas
same hyperbolas
Accordingly the whole of the

the
portion of the JTF-plane lying to
represented by
7F-plane
and y = 2?r.
the right of the F-axis and between the lines y =
Other regions of the JTF-plane of similar dimensions correspond
in inverse order.
is
repeatedly to the J7F-plane.

to the two focal points.
We
The point x
may
= 0,
y=
corresponds
get a good idea of the corre-
spondence of the two planes by describing any path composed of

portions of horizontal and vertical lines in the JTF-plane, and
noticing that when we turn to the right or left through a right
angle in the JTF-plane we turn in the same direction in the UV-
Such a path is represented in Figs. 66 a and b, where

the correspondence of the various regions is indicated by the
plane.
shading.
FIG. 66 a.
.XT-plane.
of the space between two ellipses corresponds to a

of the
rectangle of altitude 2?r in the right-hand upper quadrant
The whole
JTF-plane.
CONFORM AL REPRESENTATION.
160]
As we have interchanged z and w

interchange
x,
use, putting
a2
y with
since equation (13)
(Va
+ X -f
FIG. 666.
X-
<v/
1}.
UT-plane.
Now
and
choosing for the functions
obtain on the cylinders whose' semi-axes are a :
a2
= \/& + ^2
2
The
we must
we
u, v throughout, so that instead of (7)
= 1,
x = log
321
the whole charge
\,
=y, we
= Jo? + Xj
and
with the potentials
capacity of the pair of cylinders
is
accordingly
2 log
By
interchanging the functions
and
^ we
may
find the dis-
tribution on the hyperbolic cylinders, and in particular by putting

y = 0, y = vr/2 we may find the distribution on the edge of an in-
presence of a second infinite plate at right angles

beyond the edge.
finite plate in
with
it
w.
E.
21
322
ELECTROSTATICS.
[PT.
Logarithmic Transformation of
161.
II.
CH. VII.
last case.
If
we
transform the last figure by means of the function w' = log w, we

obtain the whole of the Z7F-plane represented on a strip of the
U' F' -plane of width 2?r, so that the transformation
w = log cosh z
1I)
transforms the right-hand half of a horizontal strip of width 2?r in

the XF-plane into the whole of the same strip in the 7' F'-plane.
Taking the antilogarithms we have
e
(2)
that
'
= cosh z = cosh x cos y + i sinh x sin y,
is,
(3)
( cos v'
+ * sin v') = cosh x cos y + i sinh x sin y
so that
= sinh x sin
'
u>
(4)
e u sin v'
cosh x cos y,
cos v
?/,
from which, in the same manner as above,

/
cos 2
sin 2
v'
cosh
a?
cos 2
v'
sin 2 v
v'
sinh
a?
2w ''
sin 2
cos y
_.
?/
2**''
and taking logarithms,

1
--
id
2
1
From
cos 2
-,
(cosh
,
and y =
log
D
cos 2
(cos
v'
2
f-
v'
-sin
v'
sinh 2 x
sin 2
sin2
v'
const,
these equations the curves corresponding to x
of
of
tables
aid
the
be
may
immediately plotted by
const,
logarithms and hyperbolic functions. They are shown in Fig. 67.

It is at once seen that u is a periodic function of v', the period being
TT.
The
figure is the
same
for negative
x and y as
for positive.
In
order to represent the whole of the ZJF-plane corresponding to the

to
half strip in the XF-plane, we must however let v' vary from
2?r.
The curves x
= 0, TT, 27T,
u' = log cosh x
v
const, are sinuous curves, u'
and minima for v = ?r/2,

and minima u' = log sinh x
. .
having maxima
3?r/2,
....
differ
but
for
The maxima
little for
large
sinh x = e?/2 so that

values of x, since then approximately cosh x
we may then take out this factor from u', obtaining u' = x log 2
for all values of v, so that the curves x = const, are nearly straight
lines.
As x
maxima and minima both diminish, but

maxima being always positive, while the
diminishes the
get farther apart, the
161]
323
The curves all cut the axis

negative.
of u' to the right of the origin, but stretch out farther and farther
minima eventually become
27T
3r
FIG. 67.
toward the
minus
left,
infinity,
so that for
x=
r/'F'-plane.
the curve reaches from
coinciding with the left-hand half of the
to
u' axis.
212
324
In
ELECTROSTATICS.
we see from the equation

when x is zero.
fact
zero
II.
[PT.
that for any finite
u' , v
CH. VII.
must be
The curves y =
const, are different in appearance, on account of

the minus sign, u has minima for v'=0, TT, %TT... having the
values u = log cos y which are all negative, and decrease more and
more rapidly
y increases to w/2. The maxima of u' are howIn fact while u increases continuously as v' varies
as soon as v > y the parenthesis becomes negative
as
ever infinite.
from
to 7T/2,
and u
is
The curves y =
imaginary.
horizontal
= y.
asymptotes v
These
const, accordingly approach

curves correspond to the
hyperbolas of the last figure, the sinuous curves corresponding to

the ellipses. Corresponding regions in the three figures are similarly shaded.
The
circle in Fig.
66 corresponds to the vertical
V-
axis in Fig. 67.
If
M^
= x,
have
we choose
for
the functions
and consider the
between
strip
the case of the electrification of
edge, lying between
two
and
v'
an
the values
"*&
= Tr/2
and
with
we
with a free
infinite plane
infinite planes parallel
V=y,
Tr/2,
at distances
it
and extending to infinity on all sides. Since at a

distance from the edge x is equal to u + log 2, the field is straight,
but the charge from the edge to the point u' is greater by
(log 2)/47r than if the plate extended to infinity instead of stopping at the edge. Thus the edge increases the capacity of the
7T/2
from
it,
upper or lower side of a portion of the plate of any width by the
amount
K = (log 2)/4?r F - FO = (log

(
2)/2-7r
This result
may
be used to find the capacity of a circular plate between two infinite

parallel plates at a distance from it d so small that the edge of the
The effect of the edge
circular disc may be considered straight *.
the same as that of increasing the radius by (log 2)
that the capacity would be, counting both sides,
is
S+
2-rr.fl
2d (log
2)/7r
B* 2R log 2
_
~
+
7T
Maxwell, Treatise, Vol.
i.
Art. 196.
/(7r/2), so
CHAPTER
ELECTROKINETICS.
VIII.
STEADY FLOW IN CONDUCTORS.
Ohm's Law.
The condition of equilibrium of electrihas been found to be that in each

conductors
homogeneous
conductor the potential has a constant value. If this condition is
162.
city in
not
any conductor, the electrification changes with the

the conductor be left to itself, or in ordinary terms elec-
fulfilled in
if
time,
tricity
moves from one place
to another in the conductor.
The
laws of this flow of electricity were enunciated in 1827 by Georg

Simon Ohm*, although the notion of the potential was unknown
If at any point in the conductor we construct an element
of surface dS, the quantity of electricity q crossing the surface in
the unit of time per unit of area will vary according to the direc-
to him.
dS at the point. That direction of normal

which the quantity per unit of time is greatest is called the
direction of the current at the point, and the quantity q is called
tion of the normal to

for
The current density
the current density.
and
its
components according
ft>
a vector quantity,
the axes will be denoted by
is
If the quantities u, v, w are independent of the time, we

call the state of the conductor a state of steady
shall
flow.
u, v, w.
We
now
consider the properties of the steady state.

If we consider
of
a
in
is
no
conductor
which
there
any portion
electricity created
nor destroyed, as much electricity must enter the space during any
interval as leaves
normal must be
(I )
it,
or the whole flow resolved along the inward
zero.
q cos
(qri)
Accordingly
dS
\\{u cos (nx)
+ v cos (ny) + w cos (nz)} dS
Mdu
ox
G.
S.
Ohm, Die
galvanische Kette mathematisch bearbeitet.
dv
h
ay
dw\ j
^- }dr,
oz J
Berlin, 1827.
326
ELECTROKINETICS.
[PT.
must be true for any portion

above conditions, we must have everywhere
and as
jfc
dx
This
is
+ f + |dz
CH. VIII.
of space fulfilling the
this
(2)
II.
in such regions
0.
dy
called the equation of continuity, and shows that the

is a solenoidal vector.
Current lines and tubes
current density
accordingly possess properties similar to those of tubes of force in

the case of equilibrium.
The law
Ohm
of
is
identical with that stated
by Fourier*
for
the conduction of heat, and connects the current density with the
potential, the corresponding quantity for heat being the temperaIf the conductor be isotropic, that is if its properties are at
ture.
each point the same for all directions, the direction of the current
the same as that of the electrostatic field, and their magnitudes
is
are proportional, the factor of proportionality depending on the

If n is the
physical properties of the conductor at each point.
normal to an equipotential surface at the point in question drawn

in the direction of the current, we have
j
(3)
XF=-x|?
Ohm's Law. The factor of procalled

the
conductivity, and its reciprocal the
portionality
If X is the same
or
resistivity of the conductor.
specific resistance,
as the mathematical statement of
is
at all points of the conductor, the conductor
geneous,
if
is
variable, the conductor
The above equation

(4)
is
3F
^=- X^
dx
is
is
said to be
homo-
heterogeneous.
equivalent to the three

v
3F
X^
iu
3F
X -5-
dz
dy
Inserting these in the equation of continuity,

(5)
5~
dx
W
5~
dx J
If the conductor
is
dy\
dy
dz \
dz
homogeneous, this becomes
A V= 0. Hence
the density p is zero, or there is no free electricity in any portion of

a homogeneous conductor in the state of steady flowf*.
*
Fourier, Theorie analytique de la chaleur, 1822.
t Kirchhoff. "Ueber eine Ableitung der Ohm'sehen Gesetze, welche sich an die
Theorie der Elektrostatik anschliesst." Pogg. Ann., Bd. 78, 1849. Ges. Abh., p. 49.
162164]
Flow.
of
Refraction
Boundary Condition.
163.
327
Lines
of
In passing from one homogeneous conductor to another,

discontinuous, and since the current must be continuous,
\ may be
we must have
at the surface
cos (q2n z )
ql cos
cos
(6)
or
v/.
dn2
dn l
The boundary condition
;r
= 0,
= 0,
(6) has a simple geometrical
meaning.
are discontinuous only on crossing the

have the derivatives in any direction t tangent to the
Since the derivatives of

surface,
we
surface,
dV/dt the same on both sides of the surface.
If
be the
acute angle made by the current line with the normal on one side
of the surface, 2 the acute angle on the other, resolving along the
normal,
\!
(7)
=\ F
cos 0!
cos
Resolving along the tangent plane, since this component
is
continuous,
(8)
6l
sin
=F
sin ft.
Dividing the second of these equations by the

,
tan0j
^9/
~~T
_ tan0
^
Aj
or the line of flow
is
first,
we obtain
2
>
A-2
refracted on passing the surface, so that the
tangents of the angles of incidence and refraction are in the ratio

The law of refraction is
Xi/Xa dependent only on the media.
,
from the optical law, in which we have the sine instead

of the tangent, and in the case of the
different
tangent law
nomenon
we do not have
the phe-
of total reflection, since the
tangent takes
all
values from zero to
^,
infinity.
164.
All the
are
Systems of Conductors.
heretofore made
statements
true for the
flow
of heat,
if
represent the temperature, but whereas

in the case of heat in
passing from one
conductor to another the temperature
a
FIG. 67 a.
328
ELECTROKINETICS.
[PT.
II.
CH. VIII.
I
continuous, in the case of electricity, in passing from the conductor 1 to the conductor 2, we have at the surface of separation
is
F -F = #,
(10)
2,
where
is a
1Z
quantity depending on the nature of the two
conducting substances.
In the theory of heat, if we have a chain of conductors in
contact with each other, surrounded by a non-conductor, we may
have equilibrium, but in the case of electricity this is the case
only
the
if
sum
of the discontinuities of potential
T \
(T
77*
Jjj
Conductors
a nature that
J71
"
JUSTt)
T
\
'
J71
JL>OQ
T
t
~r
'
JJ1
-tv-jn
' tl
is zero,
^ U.
f\
divided into two classes. Those of such

any number of them ah equation of this sort
may be
for
first class.
To it belong all metals (their
the
To
the second class, for which
same).
temperatures being
in general such equations do not hold, belong solutions of salts
holds constitute the
and dilute
If
ft
acids.
we have a
r+l
conductivity X as a pointshall show that the problem of flow is determined
being given, and also the
function,
as soon as
Let
set of conductors of either class, the constants
we
we
Fj.
are given any two equipotential surfaces.
be the potential at one of the surfaces A,
VB
that
at the other B.
Let <3? be a function holomorphic in the whole space occupied

the
conductors, satisfying the differential equation
by
a /,
(12)
dx \
a<\
dx
ao>\
dy\dyj
a
dz \
dz
and the boundary condition
at surfaces of .separation of two conductors, taking the value

unity for all points of the surface A, and the value zero for all
= for all points of surfaces

points of the surface B, while 3<E>/3w
from
the
the
conductors
surrounding insulators, or at
separating
infinity, if
the conductors reach so
far.
Then
if
vx be the potential
function in the conductor 1 (in which lies the surface A), v2

that in the conductor 2, ...v n that in the conductor n (in which
164]
the surface B),
lies
the potential
Vl
Vz
v,
vn
is
we may show
329
that in the different conductors
given by the functions
(H)
For since the function
<E>
the differential equation
satisfies
any v8 which is a linear function

Also at any surface
of <t>, must also satisfy the same equation.
r
r
and
the
conductors
+
1,
separating
that
is satisfied
by the
potential,
and
from the definition of the function <.

dary of
that
is,
At the
insulating boun-
any conductor
there
the surface
the value
is
no flow across the boundary. The function v l takes at
A the value VA and

VB But these are
,
potential function.
the function vn at the surface B

a,ll the conditions satisfied by the
It remains to
show that the function

is
uniquely determined by the conditions that have been imposed

upon it. The problem of finding the function is of the same
nature as Dirichlet's problem, differing from it in that while
the values of

are given over part of the bounding surface,

the values of d<&/dn are given.
over the remainder instead of
Suppose that there are two functions 3> both satisfying the
and <3> 2
Let them be denoted by
Then let us form the integral taken throughout the conductors
conditions of definition.
considered
ELECTROKINETICS.
330
By
Green's theorem this
is
[PT.
II.
CH. VIII.
equal to
-///<*.
The
and B, and
surface integral is taken over the surfaces
the surfaces bounding the composite conductor, the integrals over
the surfaces separating two conductors vanishing in virtue of (13).
But
at the surface
A,
and
4>j
 2
^-^ =
and at the surface B,
^ and O
are both equal to
1,
hence
0,
are equal to
2
0,
while on the remaining
surfaces dfy/dn = d<& 2/dn = 0.

Consequently the surface integrals
But the integrand in the volume integral vanishes in
vanish.
virtue of the differential equation satisfied by both functions.
Consequently the integral J vanishes, but as in Dirichlet's demonstration this can only be if <>!
But since
<l> 2 is constant.
and <3>2 are equal on the surfaces A and B, they must be everywhere equal. Consequently the solution is unique.
Properties of Vectors obeying Fourier-Ohm Law.

The vectors F, the electrostatic force, and q the electric current165.
density are typical of a class of pairs of vector-functions of frequent

occurrence in all parts of mathematical physics, distinguished by
the following properties. The first vector is lamellar, the second
In isotropic bodies the vectors have the same

and
their
ratio depends only on the physical nature
direction,
of the body at each point.
When two vector-functions have these
properties we shall say that they satisfy the law of Fourier- Ohm.
is
solenoidal.
The study
We
of the properties of such vectors
is
of great importance.
shall in general call the solenoidal vector the flux-density,
and the surface integral of its normal component over any surface
the flux through that surface.
It is remarkable that the characteristic properties of such
vector-functions are embodied in the single statement that if V,
the potential function of the lamellar vector, is uniform, finite, and
continuous, in a certain region
T,
its first
derivatives possessing
164, 165]
331
the same properties with the possible exception of certain surfaces 2 at which they may be discontinuous, then if the values
of
are given on parts of the surface S bounding the region r,
and the value of dV/dn

throughout the region
is
minimum*
solenoidal,
(2)
is
zero on the remainder, the integral
for that function
\,
if
we change the form
of the function
(8F8SF
=/(F) + 2 JJJ \J
f/33F\
-5
l\ Sa?
3a?
(8
The
q cos (qz)
by the arbitrary
8V
, /Y/;
(3)
which makes the vector q
= q cos(qy) = X-
w=
For
where
u = qcos(qx)
amount
r,
8F8SF dV
+ --8SF)
5-4
Sy
9^
9y
3^
+
V 3y /
integral with the coefficient 2 is equal,
by Green's theorem,
to
a ^ aF \
8
^ /^ 8F\
f
oF^5-(X^~
14-5- ^-5-1 + 58^
9o?/
\
13\
9y
8y/
f f f ^T^
JJJ
,
A 3F\)
T- f^ T
where n and w2 are the normals on opposite sides of a surface

of discontinuity of the derivatives.
bounding surfaces
for
*
which
is
On
given
>
8/J
those portions of the

F=0, and for the re-
Kirchhoff, Ges. Abh. p. 44.
332
ELECTROKINETICS.
[PT.
II.
CH. VIII.
maining parts X3F/3ft = 0. Consequently the integrals over the

bounding surfaces disappear, and we have
S/=-
(4)
+
In order
+ /8SFy
Mf/88Fy
-5) +
9#
3y
ffr
JJJ
I
for
-3
(V
J(V)
to be a
minimum,
this
must be
choices of the arbitrary function 8F.

be true only if we have everywhere in the region r
for all possible
positive
This can
a
d
dy J
and at every surface of discontinuity 2,
Consequently the statement that
J" is
minimum
is
equiva-
lent to stating that q is solenoidal.
166.
Integral form of Ohm's
Law.
We
have seen in
35
that the solenoidal condition signifies that the flux,
llqcos(qn)
dS
any surface bounded by the sides of a vector tube is the

same for all parts of the tube. In the case of electrical flow,
across
the flux
the tube.
not.
is
called the current (current-strength, or intensity) in
Although F has discontinuities, the function <3> has

have between any equipotejitial surfaces A
Since we
and B,
the ratio of the flux to

plus the
sum
of the
E+VA VB
sudden
the difference of potential

we go in the
rises of potential as
direction of flow, thus depends only on the function <, which

depends only on the configuration of the space r, and the values
of the function X.
That is, the ratio of the flux in any tube of the
165
167]
333
vector q to
plus the difference of potential between two equisurfaces
depends only on the physical properties of the
potential
substance in the tube.
This
is
the usual form of the statement of
Ohm's Law, and is the integral form, whereas our previous statement was the differential form. In the case of electrical flow,
- B is called the external

A
potential
to B, and it is
or electrostatic electromotive force from
of
electrostatic
force
line
the
along any line
integral
evidently
the
difference
of
from
to B.
is
electromotive force.
called the impressed, intrinsic, or internal

ratio C of current to total electromotive
The
force is called the conductance of the tube.
is
and
Its reciprocal
called the resistance of the tube.

If
we
consider a closed tube of flow, the two surfaces
will coincide,
and we
shall
have the ordinary expression of Ohm's
Law,
For any closed tube of flow, the current is equal to the

or:
impressed electromotive force divided by the resistance of the
tube.
Heat developed
167.
sider the physical
meaning
in Conductors.
of the integral
We
shall
now
con-
in the case of elec-
In passing from a point where the potential is A to

VB a unit of electricity does A VB units of work,
and that quantity of electrostatic energy thus disappears. Also at
trical flow.
one where
it is
units of work must be done

every surface of discontinuity,
r r+1
it.
But
if
we
consider
heat
as a form of energy, if
upon
mechanical energy disappears, an equivalent amount of heat must
make its appearance. If accordingly we find energy appearing in

no other form, the electrostatic energy
that disappears, together with the work done by the impressed electromotive forces,
must be converted into heat. In the case of steady flow we find
this to be the case.
In unit time the quantity
I
crosses
\\q cos
(qri)
dS
any section of a tube of flow, so that considering that part

between the equipotential surfaces A and B we
of the conductor
334
ELECTROKINETICS.
have /units entering at potential
The energy converted

will accordingly
(i)
[PT.
II.
CH. VIII.
A and emerging at potential

B
into heat in that portion of the conductor
.
be
But transforming the
165
(i), by Green's theorem,

and taking the normal at A, B, and the surfaces of discontinuity
integral,
always in the direction of the current,
dz\
The volume integral vanishes

in virtue of the surface conditions
dz
by the equation
164 (10) and
162
(5),
and
(13),
J=
(3)
The
which
integral
is
minimum
in the actual distribution of
current accordingly represents the heat generated in the conductor

in the unit of time.
The equation
(i),
written
(4)
the equation of activity for steady currents. It may serve for

a definition of the magnitude of an impressed electromotive force,
as the rate at which energy is taken into the system per unit of
is
current in
Ohm's Law,
(5)
we have
(6)
its
direction.
Combining with
(i) the equation of
167, 168]
This
is
335
the equation of Joule's Law*:
The heat developed
in
any portion of conductor in unit time
is
equal to the resistance of that portion of the conductor multiplied

by the square of the current traversing it. This law is universally
whether the flow
true,
is
steady or not.
Sources of Electromotive Force.
168.
Suppose we have
a closed circuit of a number of different conductors. We have

already seen that if all are of the first class there can be no current.
suffix
Suppose that one only is of the second

be 1. Then for all the others we have
^23
+ EM
......
and
let its
+ En-i, n + En2 = 0,
so that the total electromotive force
E=E
class,
around the
circuit is
E + E = E + E + En
depending only on the conductor of the second class and the two
Such an arrangement is called
of the first class in contact with it.
a galvanic or voltaic
cell.
In a conductor of the second class traversed by a current,

chemical actions go on, whose laws were discovered by Faraday
and Helmholtz. Such actions belong to the subject of electrochemistry, which is a branch of thermodynamics, and will be
treated
by the author elsewhere.
For the same reason the theory
of thermoelectromotive forces will not be treated here.
We
have so
far considered
impressed electromotive forces to
exist only at certain surfaces, where the potential is discontinuous.

If, starting at any equipotential surface in a closed conductor, we
plot the potential as an ordinate, on a

abscissa is the resistance from the initial to
surface, the curve will
lines,
whose slope
diagram in which the

any other equipotential
be composed of portions of parallel straight
is
proportional
to
the current.
The
total
impressed electromotive force will be equal to the sum of sudden

rises minus the sum of sudden falls as we pass in the direction of
the
downward
occur at as
braic
sum
slope.
It is evident that the discontinuities
is
evident, comparing the

*
Joule.
may
points as we please, and that provided the algethe same the current will be unchanged. It is
many
two figures in which
this
is
the case, and
" On the Heat evolved

by Metallic Conductors of Electricity, and in
Phil. Mag. 19, p. 260, 1841.
the Cells of a Battery during Electrolysis."
336
ELECTROKINETICS.
the slope of the broken lines
is
II.
[PT.
CH. VIII.
the same, that the more evenly the

is the maximum difference
discontinuities are distributed the less
of potential
between various parts of the
circuit.
By making
the discontinuities small enough, we may therefore, without changing the current, make the differences of potential in the circuit as
small as we please. In the limit the electromotive forces would
be continuously distributed, and there would be no difference of
In that case there would be no electrostatical electropotential.
motive force. Such a continuous distribution of electromotive
force
may be produced by
which
will
electromagnetic induction, the theory of

The existence of
be given at length in Chapter XII.
a current does not, therefore, imply differences of potential.
Conductors in Parallel and Series.
169.
By
the defini-
tion of conductivity of a current tube, it is evident that the

conductivity of any number of current tubes between the same
two equipotential surfaces
is
sum
the
of their individual conduc-
Now
those portions of the surface of any conductor

which are in contact with an insulator are portions of the sides of
a current tube, for there is no flux across them. If then two equitivities.
potential surfaces are given in such a conductor so that the

current flows in at one and out at the other, these surfaces are
known
as electrodes for the conductor,
and
if
the electrode sur-
conductors are brought into contact and kept

the
equipotential,
conductivity of the system is the sum of the
faces of several
individual conductivities.
The
essential in this proposition is that
the contact of the several conductors shall not change the form of
the equipotential surfaces that have been called electrodes. This
condition
is
if
is
sure to be fulfilled
if
the conductors are linear, that
each conductor forms a tube of flow whose cross-section
is
small enough to be neglected in comparison with its length. The

electrodes in this case reduce to surfaces of infinitesimal area, and
may be
regarded as points.
Conductors having two
common
electrodes are said to be connected in parallel, or in multiple arc,

and for such the resistance
of the system is given by the
equation,
!_.! 1
R~R,'R,"
The
* JL
Rn
'
resistance of the system is evidently less than any of the

If several conductors be placed in order so
separate resistances.
168
170]
that each
is
in contact only with the preceding
337
and succeeding,
the system forms a single current tube, and the current is the
same through any cross-section. The conductors are then said to
be connected in
tial
series.
If the surfaces of contact are equipotento each conductor.

The potential
we may apply Ohm's Law
at the entering electrode of the rth conductor being

r ', we have
issuing electrode
r,
and
at the
F.-F.'-ltf,
(2)
Adding these equations we have

(3)
Vi+E^ + E.
so that if
R be the
+En- ,n-Vn'=E+V -Vn
'
resistance of the system,
R=R +R
(4)
or
......
......
+ Rn,
the resistance of conductors in series
individual resistances.
is
the
sum
of their
If the conductors are linear the conditions
at the ends are sure to be fulfilled.
Networks of Conductors.
170.
We
have so
space.
Kirchhoff's
Laws.
filling a singly-connected
In order to treat a conductor filling a multiply-connected
far considered
conductors
space we have only to reduce it to a singly-connected region by

the insertion of cross-sections, and it is easily seen that if the
difference of potential on the
the
potential
is
two sides of a cross-section
determined.
These
cross-sections
is
given
are
most
naturally taken as the surfaces of impressed electromotive force.
FIG. 68.
Suppose now that a conductor has in a certain region a forked or

embranched form, as in Fig. 68. Then a portion of the tubes of
w.
E.
22
338
ELECTROKINETICS.
[PT.
II.
CH. VIII.
flow which
fill the
of the conductor continue in the
portion
while
the
remainder
leave them to traverse the portion
B,
portion
G.
Then if we consider successive equipotential cross-sections
beginning in the portion A we shall finally reach an equipotential

which is divided into two parts, one lying in B and the other in C.
The last equipotential which does not break up into two consists
of two parts touching each other and touching the surface of the
fork of the conductor in a common point.
This point, and this
equipotential surface may be taken to define the branching of the
conductors, and the surface will be taken for the common electrode
In a similar manner we may

have a conductor branching into any number of portions at a
common equipotential surface. Consider now any network of con-
for the three portions
A, B, and
C.
ductors forming a figure of any degree of connectivity. The

distribution of current and potential is determined when the im-
pressed electromotive forces are given. If the equipotentials of

embranchment are given, we may consider each conductor r between
two successive surfaces of embranchment as a separate conductor,

to which we may apply Ohm's Law,
Er +Vr -Vr'=RrIr
(1)
Ir
for
the total current in the branch,
At every surface
holds, so that if we
of
perfectly defined.
embranchment p the equation
call
positive if they all flow

(2)
is
of continuity
the currents in the s different branches
away from the embranchment,
Ipi
+ Ip*+
/jw
= 0.
For every conductor there is an equation of the form (i), and

embranchment one of the form (2). The equations are
linear in the currents in the different branches and the poten-
for every
all
tials of
the embranchments.
therefore suffice to determine
They
the currents and potentials, in terms of the resistances and

impressed electromotive forces, except that the potentials may
contain an arbitrary constant. This is determined if the potential
all
at
any one equipotential surface

In the above we
ment
is
given.
have assumed the equipotentials of embranchwill

easily seen however that these surfaces
It is
given.
the impressed electromotive forces vary. Suppose
as
form
in
vary
for instance that an electromotive force be impressed in the branch
of Fig. 68 so as to
make the
total current in that
branch
170, 171]
Then
zero.
all
the tubes of flow in
of
potential surface
embranchment
A
is
pass into
as it
339
and the equi-
were sucked up as
FIG. 68 a.
shown in Fig. 68 a. The conductor B is now longer than

and we accordingly see that the resistance of a branch
constant, but depends upon the electromotive
forces.
before,
is
This
not
diffi-
culty immediately disappears if the conductors are linear, when

the surfaces of embranchment reduce to points, where the several
conductors join. The resistance is then between definite points,

and the above linear equations determine the distribution of
currents.
Kirchhoff,
who
treated the general problem of a network of
first
linear conductors*, eliminates the potentials by adding the equations of the first kind above for any group of conductors of the
forming a closed circuit. The potentials thus disappear,

and for the circuit we have the equation
series
(3)
This and the equations (2) for the junctions are generally
referred to as the equations of Kirchhoff' s two Laws.
Maxwellf
treats the problem in the following more symmetrical form.
171.
Maxwell's treatment of Networks.
Consider n
is
points of junction, each of which, in the most general case,
connected with each of the others by a conductor. The number of
conductors in this case
is
1)/2.
are lacking this will be expressed

*
Kirchhoff.
"Ueber
die
If
some of the conductors
by putting the conductivities
Auflosung der Gleichungen, auf welche
man
bei
der linearen Vertheilung galvanischer Strome gefiihrt wird."

Bd.
Ges. Abh., p. 22.
1847.
72,
Ann.,
Pogg.
t Maxwell, Treatise, 280.
der Untersuchung
222
340
ELECTROKINETICS.
[PT.
between the corresponding points equal to

from the point p to the point q be Ipq and
Cpq
the conductor pq be
evidently
Ipq
lqp)
The equation
(i)
Epq
Let the current
the conductivity of
the impressed electromotive force
pq
,
Then
zero.
CH. VIII.
II.
let
=E
qp
may be
Cpq
Cqpt Ipp = Epp = Q.
written
(4)
Substituting the values of the currents in the equation of

continuity (2) for the point p
y
Cpl (Epl +Vp -
(5)
Vd+C^B^+Vt-Vl) ......
+ Cpn (Epn +Vp -Vn ) = 0.
Let us introduce a symbol
Cpp denned by
,
C--(O +C
rt ......
fl
+ Gpn)
or
the equation
Tc^ =
The equations
(5)
may then be
0.
written in the symmetrical
form
(6)
Cpn Vn = CplEpl
If
we add
these equations for
all
Cpn Epn
the points of junction, the
be an identity, so that the equations are not all

independent. The equations therefore suffice to determine the
result
will
differences
of
potential
potentials themselves.
between the junctions, but
not
the
Since in the equations (4) only the differences of the potentials

appear, it is evident that we may choose one of the potentials
Let us therefore put Vn equal to zero, and use the

which are independent, to determine the potentials Vlt F2 ...... Vn^. Calling A the determinant
arbitrarily.
first
1 of the equations (6),
of the coefficients of conductivity,
v 2l
j,
171, 172]
A rs the minor of Grs we

A rs = A^, since Cpq = Cqp
and
and
have
The
841
a symmetrical determinant,
solutions of the equations (6)
are of the form
A V=A
(7)
4-
1(
+G E +
(Cn En
A n_
lf t
12
((7n_j,
ClnEln )
4-
12
En - +
l} i
(7w-i,
......
-E'n-i, 2
+ Cn-i, ^n
?i
i,
.serting these values of the potentials in the equations (4),

>tain the currents in all the branches as linear functions of
the impressed electromotive forces in the branches. Picking out

the terms containing rs or its negative
w in the current Ipq we
obtain
In
like
manner the
3/rg
or
V9/
coefficient of
in
Irs
_ Crs Cpq (&pr
d&pq
But
since
is
or
A rs = A^,
etc.,
this is equal to ^-~
q
.
OJ^rs
Consequently the current produced in a branch pq as a result

in a branch rs is the
of introducing an electromotive force
same as the current produced in the branch rs on introducing an
equal electromotive force into the branch pq. This theorem is

analogous to the reciprocal property of electrified conductors given
in
(
136.
10)
If
Apr + A gs = A^ + A^
an electromotive force applied in one branch produces no current

in the other, and the conductors are said to be conjugate.
172.
Heat developed
in the System.
If
we denote the
coefficient
Cpq Crs (A^ + A 9S - A 9r - A^

we have
=^ S
r
(ii)
Ipq
r=l s=l
s)
by Cpqrs
342
ELECTROKINETICS.
[PT.
II.
CH. VIII.
Now
the activity of the electromotive force

pq is
pqLpq>
Forming the products for all the branches and summing, bearing
in mind that each branch appears twice, we obtain for the total
activity
p=nq=n
r
2 2 tf/ M -i22
r
p=lq=l
pqrs
Epq E
But since there is supposed to be no electrostatic energy, this

must be the heat developed in the system in unit time. The heat
is
accordingly a homogeneous quadratic function of the impressed
electromotive forces. If we should solve the equations (n) we
should obtain the electromotive forces as linear functions of the
Then forming the expression for the activity we should

obtain a homogeneous quadratic function of the currents, and by
our general theorem for the heating this must be equal to
currents.
<1
This might be obtained from the equations above by the aid of

certain properties of determinants.
173.
Wheatstone's Bridge.
As an example
of the above
us consider the case of
principles let
Wheatstone's Parallelogram or BridgeIt consists of four points connected by

six conductors, which may be represented
by the
sides
ogram,
or
Fig.
and diagonals of a parallelmore symmetrically as in
69.
Suppose
that
electromotive force
FIG. 69.
the
is
only
in the
impressed
branch 12,
and that we require the current

branch
84.
The equations
CnF, +
tl
I2
(6) are
F +C F +
2
13
F,
+ 0, F + CV,
V,
42
14
F = C,,EK
4
F + C F + 0F = 0,
2
43
in the
172
174]
from which, putting
F = 0, and
4
yi
C>2
<?42
343
using the last three equations,
A
From
this
we
obtain the current in 34,
The current vanishes and the conductors
34 are conjugate,
12,
if
CWk=CW7M>
This arrangement
is
used
that
for
conductor in terms of three
is
f-^fi
^41
^42
measuring the resistance of a
known
inserted in one of the conductors 12, 34,
resistances.
battery
is
and a galvanometer in the
which is called the bridge wire. The resistances in the

other branches being varied until the galvanometer shows no
In practice we
current, the condition of conjugacy is attained.
other,
how much current will pass when the condition of

from by a certain amount, in order to
deviated
conjugacy
determine the accuracy with which a resistance can be measured.
wish to know
is
We
therefore have to calculate the determinant A.
174.
Resistance
of
Linear Conductor of
Section.
If the cross-section df a conductor
equal to w,
we may
Variable
and
is infinitesimal,
write for the total current
dV
being the length of the conductor measured from a certain point.

Integrating with respect to s from Si to s 2)
and the resistance
is
given by
ds
344
ELECTROKINETICS.
This formula
[PT.
II.
CH. VIII.
important in the case of standards of resistance

with mercury, the varying diameter of the
determined
tube being
by a calibration. If the conductor is
formed of tubes
homogeneous, X
is
filled
is
constant,
and
r>
.#
if
the cross-section
is
constant,
S1
S.2
Xo>
or the resistance of a uniform wire
and inversely to
175.
of
(5),
proportional to its length
Non-linear Homogeneous Conductors.
homogeneous conductors,
162
is
its cross-section.
X being
In the case
constant, the equation of flow,
becomes
&v &y
dy
CPF
dz*
or the potential is harmonic.

Consequently every theorem on
harmonic functions applies to the potential in this case, and
every method of solving problems of electrostatic distribution may
be applied to the solution of problems of steady flow. We must

have the electrodes of the conductor given. Now by the equation
Ohm's Law
evident that the effect of increasing the

conductivity of any portion of a conductor is to make the potential
vary less rapidly there, the current being given. If then a portion
of
it
is
of the conductor be
made
infinitely
become constant throughout.
conducting
Accordingly
if
potential will
introduce a thin
its
we
plate of infinitely conducting material, this will form an equipotential surface and may be taken as an electrode for the conductor.
This supposition will be made in the following examples.

the electrostatic problem the capacity is given by
Since in
ff TT
dn
7TJJ
'2
and
we
in
~~
v
"l
'
the problem of flow the conductance by
find that the conductance of a portion of a homogeneous conis equal to 4?rX times the capacity
ductor between two electrodes
of a condenser whose plates have the geometrical form of the

electrodes of the conductor, and whose dielectric occupies the
174, 175]
345
space corresponding to that occupied by the conductor.

145, gives
The
case
of a straight field,
as in the case of
the uniform wire.
The
case of flow radially
between concentric cylindrical electrodes gives,
144,
K.
This formula might be used for calculating the resistance of
the plates are concentric
the liquid in galvanic cells where

cylinders.
The
case of radial flow in a sphere from a spherical

Q (
142) gives, if the outer electrode is at an
electrode of radius
infinite distance,
This formula
may be used
to find the resistance of the earth
between two telegraphic earth-plates.
If both earth-plates are
equal spheres buried deeply in the earth at a distance apart so

great that it may be considered infinite in comparison with their
diameters,
we may consider the
resistance from one to the other as
that of two conductors of the last case in series, so that
If, as would more nearly represent the practical case, the conductors are hemispheres, with diametral planes in the surface of
the earth, we may consider the space in the preceding problem
split
along the surface of flow formed by the plane through the
centers of the spheres, and take the lower half, whose conductivity
will be half of that just found, or
1
In like manner the problem of the ellipsoid and the circular

disk will give us the resistance between earth-plates in the form of
circular disks laid on the surface of the earth as Tr/2 times that for
a hemisphere of the same radius.
It is important to notice that
in any case of geometrically similar electrodes, the resistance is

inversely proportional to the linear dimensions of the earth-plate,
346
ELECTROKINETICS.
and not
[PT.
II.
CH. VIII.
This of course comes from the fact that
to its surface.
the lines of flow diverge in all directions from the electrode

It explains the necessity for largeinstead of remaining parallel.
sized plates for telegraphy or for the earth connection of a light-
In practice, the conductivity of the earth varying from

rod.
point to point, the conductivity of the portions near the electrode
plays the most important part, so that it is important that the
ning
The problem
earth-plate be buried in good-conducting material.
of the spherical bowl shows that if such a bowl should be made an
electrode
immersed
in
an
infinite conductor, the other electrode
being at a great distance, nearly all the current would flow from
the outside of the bowl, the current density being greatest at the
UP.
The method of the conformal representation furnishes a means

of solution for the case of two-dimensional problems, in particular
for the flow of current in a thin plane sheet.
Fig. 67 for instance
shows the
slit
lines of flow in the case of a long ribbon of conductor
along the axis of U'.

176.
Correction for
this subject with the
End
of Wire.
We
shall conclude
consideration of the practical problem of
finding the correction that must be made in the value of the

resistance of a uniform wire when it ends in a conductor so
This is of
large as to be capable of being considered infinite.
of
of
mercurial
standards
in
case
the
resistance, for
importance
We shall consider a
the tubes end in large cups of mercury.
right circular cylindrical conductor ending in a conductor of in-
FIG. 70.
175, 176]
347
extent and bounded on one side by a plane perpendicular to

We may obtain an upper and lower limit for
the cylinder, Fig. 70.
the desired correction by an artifice due to Lord Rayleigh *. It is
evident that if we introduce anywhere a portion of conductor of
finite
greater conductivity we increase the conductance of the whole. Let

us accordingly introduce in the mouth of the cylinder a plane sheet
of infinite conductivity, thus rendering that circular section equiThe flow in this case will resemble the actual flow in
potential.
that
V will
be continuous in crossing the plane, while it will differ

case in that dV/dn will be discontinuous, its
from the actual
integral over the section, or the total current being continuous.

may then use for the portion below the mouth the solution for
We
a straight
field,
so that the resistance of a length
of radius a
is
Above the mouth
of the cylinder we may use the formula for the

flow from a circular disk of radius a to infinity, so that the resist-
ance on the upper side
is
Consequently the lower limit of the resistance
is
In a similar manner the resistance of the system will be increased

we introduce non-conducting surfaces not coincident with the
walls of current tubes.
Let us below the mouth of the cylinder
suppose the cylinder split up into an infinite number of cylinders
of infinitesimal cross-section, by means of cylindrical non-conducting
if
surfaces introduced, and let the current density in these filaments

be maintained constant, in the whole of the cylinder.
Then
below the mouth the equipotential surfaces will be planes, but on
the upper side of the plane of the mouth the potential will not be
constant, as we shall show.
Consequently at the mouth of the
cylinder
V is discontinuous, while 3 Vfbz is in
Since below the
mouth
q
this case continuous.
= \ dF
xdz
Kayleigh, Tlieory of Sound, Vol.
i.
305.
348
is
ELECTROKINETICS.
dV/dz
if
dV/dz
is
CH. VIII.
side,
V must
we have
constant by hypothesis,
and
II.
[FT.
= const.,
same value on the upper
to have the
there be the same as the potential due to a fictitious (non-equipotential) distribution on a disk of radius a of constant density
JL
Z ~=
JL i
2?r X
2?r dz
The mass
'
of such a distribution would be
m = Tra
cr
a?q
ZA,
The
resistance of the upper side
may be
calculated
by
Joule's Law,
The
integral in the numerator being through one-half of infinite

space is STT\ times one-half the energy of the distribution on the
The integral in the denominator

mass of the disk. Consequently
disk.
is
4?rX times one-half the
W
where
energy
W
is
the whole energy of the distribution of the disk. This

very easily calculated. The potential at the edge of a
is
disk of radius p with constant surface density
a-
is
-is?
we introduce polar coordinates, the origin being the attracted

point on the edge, and
being the angle included between r, the
radius to the point of integration and the diameter through the
If
origin, this
becomes
2
= o-
rr
TrJ
r=
rdrdO
176, 177]
349
in increasing the radius of the disk by dp

increased by ^Trcrpdp,
The work done

the mass
is
is,
since
dW=V .27r(rpdp,
p
so that the whole energy of the distribution

f
Tr=27r<r|
Jo
TT
Vppdp
= 87TO-
Jo
Inserting this in the value of
p'
8
dp = -^
&
is
= 8~ m
7ra-*a
" ir&
we get
Consequently the
infinite
conducting mass necessitates a cor-
rection equivalent in value to an increase in the length of the

wire of between Tr/4 and 8/3-7F, that is '785 and *849, times the
Lord Rayleigh has succeeded in bringing the

nearer together, and the results have been confirmed by
radius of the wire.

limits
still
experiment.
177.
Current Sheets. The
vector, all
current-density being a solenoidal

that has been said about lines and tubes of such vectors
may be applied to current lines and tubes. The current tubes

may be defined by the intersection of two families of surfaces. A
current sheet will be defined as a portion of space bounded by two
infinitely near parallel surfaces, in which currents flow, convergIf the
ing to or diverging from certain' points called electrodes.
=
of
and
the
const,
and
are
two
coordisurface
is
equation
ql
q2
qs
nates forming an orthogonal system, the flow may be defined by

either the potential
or the current-function M*, which both
satisfy the equation,
104
(7),
Problems of plane current sheets
method
number
may be
conformal transformations.
practically for a surface
by the
them any
at once solved
of functions of a complex variable, and from

of problems for other surfaces may be solved
by finding
Such transformations may be found
by constructing
it
of thin metal, intro-
350
ELECTROKINETICS.
[PT.
II.
CH. VIII.
ducing current from a battery to it by various electrodes, and

finding by touching the surface with two sharp conducting points
connected with a galvanometer, the locus of points of equal
potential.
Doing the same for any other surface with the same
number
of electrodes of the
same sign
formation of the two surfaces*.
will give a
conformal trans-
If the whole of one surface
is
transformed upon only a part of the other, it is necessary to cut

out corresponding parts of the two surfaces, making the whole of
each equipotential edge an electrode. This can be done in practice

by soldering the edge of the sheet to a massive three-dimensional
Fig. 71 represents the flow in a
sheet
Mercator's
corresponding by
projection to the flow
spherical
in a plane sheet given in Fig. 22, and Fig. 47 represents Fig. 23
transformed to the sphere by stereographic projection.
conductor of great conductivity.
FIG. 71.
*
Kirchhoff.
krummten
Kelvin.
"Ueber
die stationaren elektrischen
leitenden Flache."
Stromungen
Monatsber. der Berl. Akad. 1875.
in
einer ge-
Ges. Abh. p. 56.
" Generalisation of Mercator's

Projection performed by aid of Electrical
Electrician, p. 553, 1892.
Instruments."
CHAPTER
IX.
PROPERTIES OF DIELECTRICS AND MAGNETIZABLE BODIES.
178.
Magnetic Phenomena.
we have supposed
of electrostatics
conductors to be
chapter
we
shall
filled
of space not occupied
with a single uniform
remove
tion of the forces
In the previous treatment

all
this restriction
when any number

Inasmuch
dielectric.
by
In this
and consider the distribumedia are present
of varying
as all the
phenomena here
we
treated have exact analogues in the phenomena
A magnet is
shall first briefly describe magnetic phenomena.
besides the conductors.
of magnetism,
a piece of loadstone, or of metal, generally iron or steel, possessing

the property of attracting iron, and of attracting or repelling
other magnets.
small
forces.
The
forces thus developed are called magnetic

magnet in the form of a filament or needle, under
the action of any other magnet, tends to set
itself in
direction at every point in space, and this direction

the direction of the magnetic force at the point.
is
a certain
said to be
portion of
space in which such forces are exerted on the magnetic needle
is called a field of
magnetic force, and may be represented by
drawing lines of force in every portion of it. If the lines of force

are straight and parallel, the field is said to be straight or uniform.
Different parts of a magnet possess opposite properties with
regard to attraction or repulsion, we may therefore consider them
charged with matter of different signs. We make use of the term
matter here precisely as in connection with electricity, not to

denote something which has inertia, but simply something which
attracts, and which is measured by its power of attraction. Experi-
ment shows that any magnet placed in a uniform field experiences

no resultant force, but only a couple.
We therefore conclude
ELECTROSTATICS AND MAGNETISM.
352
[PT.
II.
CH. IX.
is polarized ( 120).
The intensity of the polarizacalled the intensity of magnetization.
magnet may be
and
will
then
to
be
magnetized solenoidally,
appear
charged only
that a magnet
tion
is
superficially.
to two equal
long thin magnetic solenoid
and opposite magnetic
may be
assimilated
These are called the

a matter of indifference which one is
points.
It is
poles of the solenoid.
taken as positive in practice, as the earth
is surrounded
by a
run
north
and
the
end
south,
magnetic
roughly
of any magnetic solenoid which tends to move toward the north is
field
whose
lines
called positive.
It is to be noticed that a
but
is
when
magnetic point never exists alone,

always accompanied by an equal and opposite point, just as
and opposite amounts always

In
the
case
of electricity we often lose
appear simultaneously.
of
the
but
in magnetism we cannot
one
of
charges produced,
sight
electricity is generated, equal
though we may remove one of the charges as far as we

please by making the magnet long enough. It was in this manner
that Coulomb, by experiments with a torsion-balance, was able to
investigate the forces between magnetic poles, finding that they
do
so,
acted upon each other according to the Newtonian Law of the

The unit magnetic pole is then defined as the
inverse square.
pole which will repel with unit force a similar pole placed at unit
This definition is the basis of the magnetic
distance from it.
system of measurements, which stands in the same relation to
magnetic quantities that the electrostatic system does to electric

All the mathematical work that has been done for
quantities.
electricity (with
one exception)
is
then directly applicable to
magnetism. Magnetic potential, density, energy, and so on, are

defined in a similar way to the corresponding electrical quantities,
and their dimensions in the magnetic system are the same as those
The exception
of the electrical quantities in the electric system.
there are
is that phenomena of magnetic flow do not exist
noted
no magnetic conductors, and no dissipation of magnetic energy

It may accordingly seem that the principal
into heat by flow.
part of electrical phenomena, namely the distribution of charges
on conductors, forming the subject of electrostatics, has no place in
magnetism. While this is true, we shall find that a very important
part of electrostatics, namely the consideration of the field in
has exact analogues in magnetism, and these are yet to
dielectrics,
be treated.
DIELECTRICS AND MAGNETIZABLE BODIES.
178, 179]
353
we perform an experiment analogous
to Experiment VIII
two
namely, suspend
magnets so that two of their
in
the
each
other
and then surround them
air,
poles may repel
for
instance
a
another
of a salt of iron,
solution
medium,
by
by
we shall find that the magnets fall together, or the system seems
The experiment in this form would be difficult,
to lose energy.
but if we introduce into only a portion of the space a different
If
of Chapter
I,
for instance by introducing a piece of iron between the

magnets, the effect is unmistakeable. We are thus led to conclude that the energy of a magnetic or electric distribution depends
not only on the distributions themselves but on the media which
medium,
surround them.
Many
of the mathematical developments of the
preceding chapters must therefore be abandoned, and

examined in the light of this conclusion.
Parallel treatment of Electrostatics
179.
Inasmuch
statics.
as all the
phenomena
all
must be
and Magneto-
to be considered in this
chapter are exactly parallel, for electricity and for magnetism,

we shall in general not distinguish which they may be, but shall
consider in all cases the words magnetic or electric to be used
shall accordingly in this chapter not ininterchangeably.
We
troduce different symbols for the two sorts of quantities, the

necessity for so doing occurring only when both sorts of pheno-
mena
exist simultaneously.
Experiment shows that in the general case here considered the

forces experienced by a point-charge are conservative, consequently a potential exists. The law of the inverse square however ceases to hold in general, and the potential is not harmonic
in free space outside the acting distributions.
When the charges are given, since the forces are different from
those previously calculated, the relation between the density and

the potential must be different from that given by Poisson's equation.
Since however
we suppose the
force
due to any element
to be proportional to the
charge of the element, the differential
equation must be linear.
Let us examine what conclusions are
true irrespective of the law of force. By the definition of potential

as a quantity of work
necessary to bring unit charge from infinity
to any point, it follows, as was found in
any distribution is
117, that the energy of
W^
(,)
w.
E.
23
354
The theorem on mutual

117
[PT.
II.
CH. IX.
potential energy of two distributions,
(5), also holds,
(2)
jjV'trdS
=
\\\ypdr
jjVv'dS'
//]
Vp'dr.
use of these theorems, we find by the process of

All the
131, that the potential in conductors is constant.
unalso
remain
136
theorems on systems of conductors,
140,
By making
changed, except that the coefficients of capacity and potential

receive different values from those there given.
New Law
180.
of Force.
Action of Medium.
We
118 that the energy of any distribution acting

the
Newtonian Law was exactly accounted for by
to
according
of volume in all space to contain a
element
supposing each
have found in
quantity of energy equal, per unit volume, to l/8?r times the

square of the total force of the field. Since the phenomena now
of Newtonian distriwas long before any difference
was discovered, this proposition must be nearly true. We have
found however that the energy depends on the medium as well
to be considered resemble the
phenomena
bution to such an extent that
it
as on the distribution.
We
shall therefore, in order to explain
the phenomena, make an assumption deviating as little as possible

from the above proposition in regard to the energy, and containing it as a particular case, but allowing us to take account of
the medium.
be justified if its consequences accord

assume merely that each element
with experiment.
of volume contributes to the energy an amount per unit volume
the factor of
proportional to the square of the force of the field,
which may vary
proportionality being a property of the medium,
The assumption
will
We
from point to point.
shall
The whole energy
is
accordingly
a positive point-function, which is given as soon as

the substances occupying each region of space are known. For
where
JJL
is
a homogeneous medium //, is constant.

form of the energy in the expression
We
shall
W=2W
now
insert this
and apply
f,
179, 180]
the
maximum theorem
We
hold.*
119, which
of
we
shall also
355
assume
to
have now
WjjtrVd8+jjj pVd
(4)
Letting
V vary without
changing the charges
W + S r W*=<r(V + SV)dS+
(5)
cr,
p,
p(V+SV)dr
from which
Sv
(6)
W=
aSVdS +
pSVdr
far d_sv
+
rrr
dv dsv
ar
d
and by Green's theorem applied to the third integral
If
now the energy
to be a
tegrals
space
maximum
of the actual distribution of potential
for all possible values of
8F, the
must vanish, which can be the case only
first
if
two
throughout
we have
^""
fa
ar\
a /
dv\
See Helmholtz, TTm. ^6^. Bd.
a /
i.,
is
in-
aF
p. 805.
232
356
[PT.
II.
CH. IX.
and at every surface of discontinuity,

1
a=
(9)
T~
4-7T
These equations
fa
(;
3F
8^
f~
SV\f
^2 o~
'
8n 2 J
will henceforth
be known as the generalized
Poisson's equations. They give us the law of distribution of force

in the differential form, and contain the forms heretofore used as a
special case, obtained
by putting
//,
= 1.
The form of the integral expressing the energy is the same

as that of the integral
of
All the properties of
165 (i).
the integral
are accordingly possessed by the integral TF/.
In
particular it follows that if the potential is given at certain

surfaces the condition that the energy shall be a minimum re-
quires that in the space between
^?)+^)4(4F )=.
<
ox
ox
oy\
dy
oz \
of
and on surfaces of discontinuity

(ll)
h MO"
Uiv;
8??j
We may
= 0.
on 2
the problem of finding a function that shall

satisfy these differential equations, and take the required surfaceThe function
values, the generalized Dirichlet's Problem.
may
call
be called quasi-harmonic.
if
It is evident, as in
there be any, is unique.
86, that the solution of the problem,
We
have heretofore said nothing regarding the localization of

the energy of a distribution, which we have represented either by
an integral
d throughout the acting distribution, or by an integral
which
is
Wf,
expressed in terms of the field at all points of space.
Whereas both representations are equivalent mathematically, it is

a fundamental point in Maxwell's theory to regard the energy as
localized in the medium wherever a field exists.
181.
(12)
Induction.
g)
If
we
= gc
define a vector
g by the equations
DIELECTRICS
180, 181]
AND MAGNETIZABLE
the vector has, by (8), the property of being solenoidal in

That is
of space where there are no charges.
div
(13)
The
g = dx +
force, or field-strength F,
general, but does so in a

out as a constant factor.
is
+ =
p f
dz
dy
all
parts
0.
|3-
no longer enjoys
homogeneous medium,
The
357
BODIES.
vector
is
this property in
for which /x, comes
called the induction,
and
connected with the force by the equations

ae
(14)
The
induction
accordingly
everywhere the
satisfies
law of
Fourier -Ohm.
The
surface integral over any surface of the induction resolved

normally to the surface is called the total induction, or inductionflux,
The quantity
more usual name for it
through the surface.
of the
medium.
/JL
is
called the inductivity
is
the specific inductive
capacity or dielectric constant, in the electric case, magnetic permeability in the magnetic case. The latter name is due to Lord
Kelvin, to
whom
flow in electricity
the recognition of the analogy to the case of

is due.*
The name permeability comes
and heat
from the hydrokinetical analogy of water flowing through a porous
medium.
The
lines of induction suffer refraction in the
manner described
163 when passing from one medium to another.
in
In the
FIG. 71a.
*
Magnetic Permeability, and Analogues in Electro-static Induction, Conduction

and Fluid Motion. Papers on Electrostatics and Magnetism, p. 487.
of Heat,
358
[FT.
II.
CH. IX.
does not differ widely for different media, seldom

/JL
ten
times
the value for air or empty space, and never
reaching
than
for
less
empty space, while in the magnetic case, //,
being
electric case,
thousand times as great as for air, and in

some cases is slightly less than for air. Consequently lines of
force emerging from iron into air are generally nearly normal to
may
be, for iron, several
the surface in the air unless very nearly tangential in the iron.
This is exemplified in Fig. 7 la, showing the distribution of lines
of force
between the pole-pieces of the field magnet of a dynamo

In virtue of the analogy to electric
with the armature removed.
evident that the lines of force exhibit a tendency

to crowd together into parts of the field where //, is large.
conductivity,
182.
it is
Relation of Charge to Induction.
Since the force
no longer possesses the solenoidal property, except in homogeneous

media, while the induction does, it is more logical to speak of
tubes of induction than of tubes of force, although geometrically
The flux of force through various cross-sections
the two coincide.
of a tube, however, varies, while the flux of induction
for
is
constant
the tube.
The volume density
is
no longer determined by the divergence
of the force, but of the induction, being equal to 1/47T times the
divergence of the induction,
as
while the surface density
is
180
,
(8),
ag)
331
l/4?r times the discontinuity of its
normal component,
180
(16)
T-^{8, +.}.
(9),
Accordingly the charge of any portion of space
=-
j- (({$
cos (nx)
+ g)
cos (ny)
r,
+ 3 cos (nz}} dS,
equal to l/4?r times the excess of the number of unit tubes

shall call
issuing from the space over the number entering.
the densities thus defined, for a reason to be presently explained,
is
We
the densities of true electricity or magnetism.
181
DIELECTRICS
183]
183.
AND MAGNETIZABLE
BODIES.
359
Since on passing from one medium

is different the force is dis-
Apparent Charge.
where the inductivity
to another
continuous, the surface acts as a charged surface has previously

82.
Also since the force is not solenoidal
been found to act,
medium, there appear to be bodily charges.

The magnitude of these apparent charges, whose densities are p </,
are given by the usual equations
in a heterogeneous
''"
(18)
and comparing these with the equations
we
for the
true densities
find
i
(20)
r + 4?r
=-\P
A
^
(
\dx dx
dz
dy dy
homogeneous medium
or in a
(20')
p'
For the surface density
(21)
o-l
= -f-fr
The
potential
that
is
is
4>7T
/>!
- fr\
/A!
-]Jp 2n
a= ----
then determined by
\- -.
4,77
85 (18), as
distribution of charges acting according to the Newtonian Law,

of densities p and a', would produce everywhere exactly the same
field as that actually produced
by the true charges p and <r. The
Newtonian Law thus reappears, and may be used
to calculate the
only the true charges do not follow the law, but the apparent
charges, which are known as soon as the true charges and the
properties of the media are given.
forces,
360
184.
Reluctance.
Capacity.
for the capacity of a
In the
II.
electric case
condenser whose dielectric
is
'
that
[PT.
CH. IX.
we have
homogeneous, so
AF=0,
/-s
CvO
(23)
is, the capacity equal to that found in Chapter VI multiplied

This is Faraday's capital disthe
by
inductivity of the medium.
to
the
development of the theory according to
covery,* leading
that
which the energy resides
in the medium, so that electrical actions

of the medium, and not by action at a
means
are transmitted by
distance.
Faraday experimented with condensers in the form of
concentric spheres, the intervening space being filled by the
The material of the dielectric outside the
dielectric in question.
was
accordingly immaterial. Instead of capacity the
larger sphere
term permittance has been proposed by Heaviside.

In the magnetic case, the value of the quantity analogous to
the capacity has been called the permeance or inductance, while its
case of electric
reciprocal, corresponding to the resistance in the
called
resistance
was
flow,
by Bosanquet, a name which
magnetic
has given
way
to that of reluctance.
Induced Charge. The apparent charges defined above

185.
minus the true charges are called the induced charges due to the
If we examine the amount of the
action of the forces of the field.
induced charge in a body r surrounded by a homogeneous medium,
we shall obtain an important result. Let the constant inductivity
of the
external
medium be
and
toward the interior of the body r by

outside by
ne
Then
charge on the surface
HI
let us denote the normal

and the normal toward the
we use the formula
if
we
(21) for the apparent
find
and transforming the second integral on the right by Green's

theorem this becomes
*
Exp. Res.
1252 seq.
184- -186]
dV
But by the
definitions of true
361
9F
and apparent volume density
this
is
(26)
so that, transposing,
(2 7 )
dS +
JjaUS' +////>'*
[jja
jjjp dr^
or the total apparent charge of a body surrounded by a homomedium is equal to the true charge of the body divided
geneous
by the inductivity of the surrounding medium. In particular a

body which has no true charge has a total apparent charge equal
to zero, and since this remains true however the body may be
is polarized.
In the magnetic case, the body
found
to
be polarized, consequently we must conclude
always
subdivided, the body

is
that the true magnetic charge of all bodies is zero, or in other

words, true magnetism exists only as polarization. This is a second
apparent difference between electricity and magnetism, bat
remember that whenever

tities of
if
we
electrification is
produced equal quanboth signs appear, the difference disappears.

i
186.
Polarizations. Since experiments on electrification
and magnetization are almost always made on bodies surrounded by
a homogeneous medium, namely air, it has become customary to
regard their apparent charges as due to the polarizations of the
bodies themselves, although it is evident by 120 that the surface
charges are due only to differences of polarization on the two sides
of the surface.
The surrounding medium might be uniformly
polarized to any degree without producing any effect, consequently

absolute polarization cannot be determined, and is of no importance whatever.
The apparent polarization of the body must
its
produce the surface density, by

(28)
120
= -/cos(7w
(2),
i ).
ELECTKOSTATICS AND MAGNETISM.
362
But using the equation
for the
the true surface density
so that if
CH. IX.
apparent surface density (21) when
is zero,
we put
7-
(30)
in the direction of
density.
From
we obtain
for
0'
F and g,
the components of
we obtain the proper
surface
/,
the volume density due to the polarization,
120
(6)
fdA
+ _ + ._
_(_
dB
(32)
II.
''
(29)
giving
[PT.
dC
33e
33\
3g)
|
/3z
3F
and since by (18) we have

,
(33)
fiX + 37 + 3\ =
= n1 K^- ^47rV3
3y
3^/
1
div
4?r
.P,
we must have
is solenoidal.
The induction accordingly possesses
the property of the vector called the induction in 121, and by the
or the induction
equations (31)
is
equal to
it if ft is
equal to unity.
187.
Examples. Point-charge in Medium bounded by
Plane Face. Suppose we have a point-charge e placed at P at
a distance a from a plane face separating two media of inductivities

We may solve the problem of
ft, ft, their extent being infinite.
induction by the method of images as in
152.
Suppose that e
on the left of the dividing plane, and that at its geometrical
lies
186, 187]
363
Then we may determine the

image P' we place a charge e.
charge e' so that on the left the potential will be the same as
that due to charges e and e placed at P and P' in a uniform
medium, while on the right we shall have V the same as would be
produced by a charge
on the
V= re +-,r
left
+e
For
placed at P.
if
we put
dr
e
r'
on the right
0;
r2
9rii
V=
dV
But
at the surface these
8F
^ 8wj ^
,
Consequently
if
+e
dr
must
9F =
+- e
cos
u.
satisfy the equation

,
,.
~ ^2
tin's
we put
=e
'
,
4-
FIG. 72.
MO
/M
'
364
The
the condition will be satisfied.

the surface of separation
II.
CH. IX.
surface density induced on
is
so that the distribution
[PT.
is
proportional to the distribution on a

medium 2
conducting plane, as in 152. The lines of force in the

are straight, and the refraction on crossing the plane is
shown
in
the diagram, Fig. 72, in which pj^ = 4. The greater /^ 2 the less
is the force in the medium 2, and if we make yu 2 infinite, the force
in the
medium
2 vanishes, and the surface density becomes
as in the case of a conductor.
Slab in Uniform Field.
188.
tivity
fi 2
with parallel faces of
field parallel to
Suppose a slab of inducuniform
infinite extent is placed in a
the equipotential surfaces, and
of the surrounding
homogeneous medium
be p l
the inductivity
let
Then the
potential
equation in the slab as well as outside it. Accordingly the solution for either of the three parts of the field, in,
above or below the slab, is a linear function of the single cosatisfies Laplace's
ordinate perpendicular to the equipotential planes, and the force

has values which are constant, but different, in the three regions.
FIG. 73.
1 and
2 are any two equipotential planes outside the slab at
distances ch and d2 from its faces,
3 and
4 the potentials of the
faces of the slab respectively facing them, and d the thickness of
If
the slab,
that
(2)
we have the
conditions at the surfaces
is
fl
-F -*l
t
F F
3
Now by
so that
AND MAGNETIZABLE
DIELECTRICS
187, 188]
365
BODIES.
145 we have
we have the equations
Solving these for
and
Fi
=
from which we get
for
the force outside the slab
p - V>- V*In the
F F are the
V is
electrical case if
1?
the density on the upper plate

C7
Si
4 7r
2,
= M\
^'
and the capacity of the condenser of area

V,
By measuring
surfaces of conductors,
-F
the capacity with the slab and with it removed

dielectric constant of the slab in terms of
we may determine the

that
of the
capacity
which
is,
constant.
slab
surrounding medium.
If
and d 2 are
as
was stated in
The apparent
so that the intensity of polarization is

r
is
the
184, proportional to the dielectric

surface density on the upper face of the
is
which
zero,
is
in the direction of the force if
fi 2
> fa.
366
[PT.
II.
CH. IX.
Suppose we have a pointR of homogeneous

charge
surrounded
an
of
substance
infinite homogeneous
inductivity^!,
by
Point-Charge in Sphere.
189.
e at
the center of a sphere of radius
Then
medium
of inductivity
in either
medium, we may use the
either
medium
yu/2
Laplace's equation being satisfied

solution of
in
142, and put
equal to a linear function of 1/r
F =-.
r
2
The
condition at the surface r
gives, since
dF__^
<W\_A
dn ~7tT2
=R
'
8?T~
'
so that
e
Vsince
the integral
must be
of
fi
over any surface inclosing
kire.
The apparent
surface density
4-7T
The
%n
( dtii
and charge of the sphere are
dn2 )
4?r
real charge e at the center acts,
'
(jj^
by
/z 2 j
183 (2O
),
like
an
apparent charge e///,i,and the apparent charge of the sphere e acts

at outside points as if concentrated at the center.
Accordingly
the whole force in the medium 2 is
e*i
which
is
+ e'
the same as found from
F /9r.
a
Unit of Electricity or Magnetism.. If the charge

190.
be situated in a medium of inductivity extending to infinity, the
2
force of the field is, by the above, equal to e//j,r and the action of
//,
Now the
on a charge of e^ units is e times as large, or ee^r1
unit charge has been defined as the charge which repels an equal
charge placed at the unit of distance from itself with a unit of
We accordingly see that the magnitude of the unit will
force.
e
189
DIELECTRICS
191]
AND MAGNETIZABLE
depend on the medium, and
medium
of inductivity
the ratio V/A than
We
inductivity.
/^,
for
/j,
if
367
BODIES.
the experiment be
made
in a
the unit thus obtained will be larger in
had been determined in a medium of unit
if it
also see that the dimensions of the unit involve
we must have the dimensional equation
_\MU
I
so that
customary to choose the unit of inductivity so that the

empty space is unity, or as it is sometimes stated,
the inductivity of the ether is unity. This is, as we have seen,
purely arbitrary, as experiments enable us to determine only ratios
It is
inductivity of
The
of inductivities.
netic, differs
purposes
very
inductivity of air, both electric and magfrom that of a vacuum, so that for practical
little
we may
consider the size of the units determined by

must notice that even if fju is put equal
We
experiments in air.
to unity its dimensions
of
p we have no means
remain in the equation and the dimensions

As the matter of dimensions
of knowing.
we may make any supposition

Two
are led to contradictory results.
when
different suppositions are convenient.
may,
dealing
with electrical quantities, assume that the dimensions of the
is
always more or
that
we
less arbitrary,
please, until
we
We
electrical
inductivity are
system of units.
We may
zero.
This
gives
the
electrostatic
on the other hand, when dealing with
magnetic quantities, assume that the dimensions of the magnetic

This gives the magnetic system. Both
inductivity are zero.
these systems are due to Gauss, and when we use both systems for
their respective kinds of quantities, we shall say that the quantities
are measured in Gaussian units.
preceding chapters.
This has been the case in the
When we come
to deal with both electrical
and magnetic quantities at the same time, we must choose one or

the other of these assumptions, as we shall find in the next chapter
that both together are incompatible.
The equation giving the apparent

Susceptibility.
of
a
medium
of
polarization
inductivity //, 2 surrounded by a medium
191.
of inductivity
is, (
186 (30))
368
[PT.
II.
CH. IX.
so that the polarization is proportional to the total force of the

that is the sum of the external field and the field of force due
field,
to the polarization.
The
coefficient
4-7T
is
fr
called the magnetic susceptibility, in the magnetic case. In the

case the quantity K has never come into practical use.
electric
The equation
I= KF,
was the basis of Poisson's theory of magnetic induction, K being
supposed a quantity inherent in the body, and equal to zero for
air.
We see however that K depends on the medium by which the
is
body surrounded, as well as on the body itself: K2l may therefore
be called the relative susceptibility of the body of inductivity 2
//,
medium
in a
tion
to
is
If jc a is positive, the polarizaof susceptibility ^.

in the direction of the polarizing force, and the body is said
be paramagnetic, or simply magnetic.

is
polarization
said to be diamagnetic.
Accordingly
medium of greater inductivity than its
is
we
consider always the

=1
respect to a vacuum, so that /^
netic.
If
If
/c.2l
is
negative, the
and the body

in a
immersed
any body
in the opposite direction to the force,
fj,
=1
own
will
appear diamag-
polarization of a
body with
we may put
4- 4>7TfC.
Bodies are accordingly magnetic or diamagnetic as yu, is greater

or less than unity.
It is evident that the assumption that K is
zero for a vacuum is arbitrary, in the same degree as, but inde-
pendently of the assumption that the inductivity of vacuum is

unity, for we might assume all apparent polarizations to be the
differences of the polarizations of bodies from the polarizations of
vacuum.
Uniform Polarization by Induction. When a body

different inductivity from the rest of the medium is inserted
192.
of
into a field of force, the configuration of the field is disturbed

owing to induction, the polarization due to which produces new
forces FI
which must be added to the
forces of the undisturbed
DIELECTRICS
191, 192]
We
field FQ.
shall
AND MAGNETIZABLE
now examine
in
369
BODIES.
what cases the introduction
of a polarizable body into a uniform field will produce such a

resultant field that the polarization of the body will be uniform.
and the potential
Let the potential of the undisturbed field be
of the forces due to the induced polarization be

total potential of the field is
t-,
so that the
F=F,+ F
(1)
i.
If
X F Z
,
the undisturbed
denote the constant components of the force of

we have
field,
V =C-X x-Y,y-Z^.
(2)
Let
a, ft, 7,
be the constant direction cosines of the uniform
polarization, so that
4 = 1*. B-Ip, 0=77
(3)
Then
since
/=
icF
we must have
for the total potential
F= a -Xx- Yy- Zz = C' - -K (ax + fty + yz).
(4)
But we have seen
123
be the potential of a
single distribution of density unity occupying the space filled by
the polarized body we have for the potential due to the polarizain
(6),
that
if 11
tion
Jt =
T7
(5)
an
r
-^^zr=--;rdh
dx
^^-
_an
d
Consequently
if
we put
'
dz
for fl
a = C" - i [La? + Mf
(6)
so that
(7)
(8)
all
(9)
Vi =
/ (Lax + M/3y + Nyz) = LAx + MBy + NCz,
V=V+V = C+(LA-X)x+(MB-Y )y + (NC-Z)z,

i
our conditions will be
fulfilled
by taking
LA-X^-^, LB-Y = --, LC-Z = --.

fC
fC
fC
Now the only body for which fl has the form of a quadratic
function of the sort given is an ellipsoid.
The values of the
We
113.
in terms of the axes are given in
M,
have accordingly found that an ellipsoid introduced into a uniform
24
w. E.
constants Z,
370
field is
uniform.
We
[PT.
II.
CH. IX.
uniformly polarized, and the field inside the ellipsoid

The field outside will not be uniform.
is
obtain from equations (9) for the components of the polari-
zation
IT
'
ft
-D
4- /cL
and
for the field in
terms of the undisturbed
(II)
+ fcL
The
field
-TT"
-T7-
'
1 -f
'
tcM
1 4-
KN
'
the polarized body is not in the direction of the

= = (sphere), or the unfield, unless either L
field in
undisturbed
M N
disturbed field has the direction of one of the axes of the ellipsoid,
F ^ vanish. The force due to
when two of the components
the polarization,
has the components
it
icL
'
l+/cL
tcN
If K
is
positive, this force is in the opposite direction to the

If the body be magnetic, and hard, so that
force.
undisturbed
when the
field
force Fi alone acts,
is
removed, the polarization is retained, the

to produce a magnetization in the
and tends
The force Fi is
reverse direction, or to demagnetize the body.
accordingly called the self-demagnetizing force, and since we may
write
Xi =
(13)
L,
M,
N are
called the self-demagnetizing factors for the three
axes.
193.
The couple exthe components
Couple experienced by Ellipsoid.
perienced per unit of volume has, by
BZ,
-CT
Suppose that the
CX -AZ^
Q
120
(5),
AY -BX
force of the undisturbed field
is
in one of the
principal diametral planes of the ellipsoid, then the couple tends
192
194]
to turn
it
the force
is
371
about the axis perpendicular to this plane. Suppose

in the XY- plane, and makes an angle 6 with the Z-axis.
Then
and the couple about the
volume 4}7rabc/3,
The
4f7rabc(
values of
4<7rabcK
N are by
L M,
t
on the whole ellipsoid of
-Z-axis acting
(M - L)
jP
113,
du
Joo
(q* 4- u)
V(a
4 u) (b + u) (c
2
'
4-
u)
N are obtained by putting

c respectively for
in
the reverse order of magnitude from
Accordingly L, M, N are
where L, M,
a, b }
c.
a, b,
a > b and < ?r/2 the couple is positive,

the ellipsoid tends to turn its longer axis
whether K is positive or negative. This is in
Consequently
or from
X to
q.
Y, that
if
is,
parallel to the field,

contradiction to a statement frequently made, that diamagnetic
bodies tend to set their longest dimension across the field.
They
if the field is uniform.
If an ellipsoid be suspended
a
a
fibre
in
torsion
by
magnetic field, the field will cause it to
vibrate more rapidly when its long axis is parallel to the field, and
do not do so
more slowly when
it would do in the
however extremely difficult, if not
impossible, to obtain a magnetic field nearly enough uniform to
show these phenomena in diamagnetic bodies, on account of the
absence of the
extreme smallness of # 2
194.
across the field, than
it is
field.
It is
Polarization of Sphere.
we have by 80,
In the case of a sphere
for
inside points
n = 27r
(i)
so that
\
which
is
the self-demagnetizing factor.
Accordingly the force
is
in the direction of the original field,
242
372
and the polarization
4fir
The self-demagnetizing
force
(5)
be
CH. IX.
4?r
1T
is
2
fi
II.
is
(4)
If
[PT.
infinite this
field,
and
This
is
becomes equal to
force inside the sphere is zero.
sphere in an electric
a magnetic field.
Q,
so that the total
the case for a conducting

nearly the case for soft iron in
is
Outside the sphere we have a different form
for
H,
(6)
so that
cos (hr)
(7)
dh
The field due to the polarization is accordingly, by 123 (7),

the same as the field of a doublet of moment 4?r^ 3 //3 =JR s Fi} and
the total field outside the sphere is obtained by superposing this
If p = oo the moment of the sphere is
upon the uniform field
The
lines of force
due to a uniform
FIG. 74.
field
disturbed by a doublet
DIELECTRICS
194, 195]
pointing in
its
own
AND MAGNETIZABLE
BODIES.
373
shown
in Figs. 74
shown
for /x//^
direction or the opposite are
and 75 respectively.
FIG. 75.
The
the sphere in a uniform field

respectively in Figs. 76 and 77.
field of
is
=3
These figures were

//-//A!
of
Kelvin's
in
Lord
Papers on ElectroReprint
originally given
statics and Magnetism, (p. 492), where the equations of the curves
are discussed.
The figures have been re-drawn for this book, the
and
lines
or p.
oo
being drawn for equal increments of the flux-function W,

103 (10).
FIG. 76.
195.
Infinite Elliptic Cylinder.
If one of the axes of the
ellipsoid is infinite, we have the case of an infinite elliptic cylinder.

If c
oo ,
is zero, and L, M, reduce to
trigonometric forms. The
force parallel to the
"-axis is
the same as that of the undisturbed
374
[PT.
II.
CH. IX.
a consequence of the distributions on the ends

being infinitely distant. We may then measure K as the ratio of
field.
This
is
FIG. 77.
the longitudinal magnetization to force, a method often used in

practice, and accurate only when the cylinder is extremely long.
Ellipsoid of Revolution. In the case of an ellipsoid

of revolution the form of the integrals simplifies, and inserting in
196.
e^lo?
116 the eccentricity
the formulae of
b*/a
we
obtain for a
prolate ellipsoid
1-^(1,
-I
(i)
1-e
M=N=fr
(2)
and
for
an oblate
ellipsoid
L=4
(3)
Vl-e
2
.
sin" 1 e
b.
(4)
For
6=0
all these
expressions become indeterminate, but on
the
indeterminate
form they take the common value
evaluating
for
found
the
For e = 1 the expressions for the
already
sphere.
become
indeterminate, and on evaluation we find
prolate ellipsoid
195
197]
DIELECTKICS AND MAGNETIZABLE BODIES.
375
This gives the case of the infinitely long circular cylinder, for
which, as we have previously found, the longitudinal demagnetizing factor vanishes, while for transverse magnetization it is equal
to
2?r.
When e = 1 the expressions for the oblate ellipsoid give L = 47r,

M = N= 0. This gives us the case of a disk magnetized normally,
for
4-Tr,
which the demagnetizing factor

or parallel to the faces,
is
the largest possible, namely
when the demagnetizing
factor is zero.
For a long prolate ellipsoid, for which e is nearly unity, we

may conveniently use an approximate formula. Putting = a/b
for the ratio of the
length to the diameter, since
i-#-?=l
m
a
&
~~~
o
2
o
2
>
we have approximately
l/=^=
(6)
7YL
table of values of the demagnetizing factor
Ewing*, and a larger one by du Boisf

197.
is
given by
We
Magnetization of Hollow Cylinder.
shall
now
consider a few cases of induction in which the induced magnetization is not uniform. In the first case let us consider the uniplanar
problem of the transverse magnetization of an infinite homogeneous circular cylinder, place4 in a field such that the lines of
intersections of cylindrical surfaces with planes
to
the generators of the cylinder. If the cylinder is
perpendicular
circular the method of development in series of circular harmonics,
force
are the
94, gives the general solution of the problem.
Let the cylinder be hollow, the inner radius being b and the
and of the space
2
a, the inductivity of the cylinder being
outer
/Lt
within and without
Let the undisturbed
fa.
field,
with potential
while the
represented by
induced polarization is
with the potential
t
,
suppose that the bodies
producing
the
field
as before, be
field
F;.
lie
due to the
We
shall
outside the
Ewing, "Magnetic Induction in Iron and other Metals," p. 32.

t du Bois, "Magnetische Kreise, deren Theorie und Anwendung,"
p. 45.
376
[FT.
II.
CH. IX.
and its derivatives are finite

cylinder, so that the potential
and continuous at the surfaces of the cylinder.
Let
be developed at the outer surface in the
it
infinite series
of harmonics
r,
(i)
Then
= T.(4>)
at points for
which p
<a
it is
given by the
F.-r.+Jr. + gr.H-
(2)
series,
......
The
ments
potential Fi is represented by three different developin the three different regions, (1) p >a, (2)
b, and
We
< b.
a> p>
(3) p
distinguish these by an
vanishes at infinity, we have outside the cylinder
will
affix.
Since
F$
Fi
(3)
In the substance of the cylinder we must take
V^=^(Bnpn + Cnp-n )Tn
(4)
while in the cavity, since F$
is finite
at the center,
V^
(5)
o
(2)
w
continuous, at the surface p = a we have F$ = F^
all
values of <j> we must
and as this must be identically true for
Since F;
have
for
is
every term the coefficients of
An
(6)
In
like
or n
= Bn
manner, at the surface
DJ>
(7}
Tn
equal.
+ Cnarn
p = b, we have
an
for
every term,
= BJ>+CJr*.
Beside the conditions of continuity, we have at each surface of

the cylinder
F=F
the whole potential

+Fi. The potential of the external field being continuous, with its derivatives as well, we have
for
+-
197]
which being multiplied by
377
^ and subtracted from (8) gives

9F
At the
surface p
this gives, differentiating (2), (3), (4),
by p,
^(-A nn
(ii)
and consequently
for every
- frAncr" - ^ (Bn an - Cna~n ) =
(12)
and
=a
(/JL.,
- ^),
= b,
at p
and consequently
The
four linear equations (6), (7), (12), and (14) determine the
four constants
A n Bn Cn Dn
we
Solving,
obtain for their values, putting
= M,
Nn =
(15)
Since the absolute value of

b/a <
1,
negative.
Nn
is
is
greater than
and accordingly
1,
Dn
and since
is
always positive,
always
Accordingly the effect of the induced polarization
378
[PT.
II.
CH. IX.
within the hollow is to oppose the external field. We have for

the whole potential in the cavity, by (2), (5), and (15)
The
absolute value of the coefficient of any
Tn
is
less,
the
smaller the ratio b/a, that is, the thicker the walls of the cylinder.
By making the walls thick enough, we can make the coefficient
approach as nearly as we please the value 1

is smaller, the greater the ratio fi z /fa.
For
internal
would be reduced to
field
zero,
l/M
/^//î
Now
this
infinite the
as in the case of the
sphere.
This principle was used by Lord Kelvin in his marine galvanometer, in which a thick cylinder of iron shields the galvanometer
from the influence of external magnetic
ment
is
now
fields.
Such an arrange-
nearly always necessary to protect magnetic instrufield of electric railroads (in America).
ments from the
In case the external
We
then have,
if
is
field is uniform, the internal field

the strength of the external field,
is also.
008
In this case
= 1000,
and if the thickness of the cylinder

outer diameter, the field within is reduced
to two per cent, of the value outside.
The effectiveness of the
shielding is thus plainly shown.
is
if /^//î
only one-tenth of
The
total field is
its
shown
for this case in Fig. 78, for
which
This represents approximately the distribution of the lines of

force in a ring-armature of a
dynamo.
Magnetization of Hollow Sphere. The case of the

sphere may be treated in precisely the same manner as the case of
198.
197, 198]
DIELECTRICS
AND MAGNETIZABLE
BODIES.
the cylinder, substituting spherical for circular harmonics.
379
Let us
FIG. 78.
again suppose that the bodies producing the undisturbed

outside, so that at the outer surface of the sphere
(i)
Accordingly for r <
The
potential of the induced polarization

(1)
F,r i
(-2\
\j/
a,
5LA n'r~< n+1

**-'*
>
Fn>
-*
is
given by
r">a
-^ **>
'
= 2 (Bnrn + (V-w) Fn a>r>b,
(4)
=2Dnrn Yn)
(5)
The
r<b.
conditions of continuity of F< give as before
(6)
A n a-<n+ v = Bnan + (7na-<n+1
(7)
Ai&
= ^n&n +
>,
field are
380
The condition
[PT.
II.
CH. IX.
(10) of the preceding section gives the corre-
sponding equations,
(IT)
fr%{o
2 {Bnnan->- Cn (n + 1)
a-<
o
(12)
-frA n (n + l)
- ^ {Bnnan -Gn (n+

(13)
1)
-y
o
(14)
- frDn nb + K
{Bn nb
Solving the equations
-Cn (n
(6), (7), (12), (14),
we
get, putting
271+1
,^
*-The
results are similar to those in the last section.
Mn and M^ have the same sign, and

than unity, Nn
positive, and Dn
is
cavity
is
Since
are greater in absolute value

The field in the
is negative.
given by
(16)
M.
198, 199]
For a uniform external
the internal field
If b/a
is
= 9/10,
381
field
uniform, and given by
fr/fr
= 1000,
the internal
field is
1/67 of the ex-
ternal.
The sphere thus
shields
more
effectively
than the cylinder,
table of the relative strengths of the

as might be expected.
internal field for various ratios of b/a and for fj.Jfh. = 100 and
= 1000,
is
given by
Theory of Electricity
If b
Thomson, Elements of
and Magnetism, p. 264.
J. J.
the
Mathematical
the results of this section agree with those of
194.
For instance in the sphere
agreeing with
199.
194
(3).
Forces acting on the Polarized Body.
In virtue
of the polarization of a body whose inductivity differs from that

of the surrounding medium the body experiences certain forces.
These forces
may be
calculated
when the induced body

another, during which
is
by considering the work done
moved from one
motion
its
part of the field to
polarization
will
in
general
change.
Before considering the general problem, let us, to fix the ideas,
electrical condenser.
We have seen that
examine the case of an

the capacity
Accordingly
is
proportional to the inductivity of the dielectric.

given charge, the difference of potential of the
for a
is inversely as the
inductivity, consequently the force of
the field varies in the same ratio.
The energy being proportional
to the product of the charge by the difference of potential is
plates
accordingly inversely proportional to the inductivity. Now since,

the charge being given, the energy tends to decrease, if the dielectric is
movable, and
its
inductivity variable,
it
will
tend to
382
move
the
[PT.
II.
CH. IX.
so that portions of greater inductivity shall be drawn into

If on the other hand the potential of the condenser
field.
plates is maintained constant, the charge is directly proportional

to the inductivity, so that the energy is also
directly proportional.
We
have seen in
140 that in this case the energy tends to
increase, so that again the forces tend to bring substance of greater

These properties of the energy should
inductivity into the field.
maximum property mentioned in 119,

the variation was in F, tending to make it differ
from the values necessitated by the differential equation
not be confused with the
180, for there
dx
yu,
dv\
o~
/*
~o~
%oc )
d (
o~
A6
dy V
dv\
~o~ Hdy
Here the variation

depend on geometrical parameters
being unvaried.
to
polarizable
bodies,
precisely
as
geometrical parameters, and in
in
dv\
o~
dz V
A6
-^-
is
dz
in
//,,
and may be made
fixing the position of the

140 we had changes in
V must
this case the variations of
be such as are consistent with the above differential equation.
We may
of view.
look at the matter from a slightly different point

we found in
140 that the capacity tends to
Since
when the forces of the system produce motion, the system

move so as to increase
The capacity will be increased
when a body of greater inductivity moves into stronger parts of the
increase
will
//,.
consequently magnetic bodies are drawn into the strong parts

field, while diamagnetic bodies are repelled from the stronger
This property was correctly
portions to the weaker portions.
field,
of the
by Faraday, and was demonstrated by Lord Kelvin. It

tendency of diamagnetic bodies to move to the weaker
parts of the field that often makes them set themselves across
stated
is
this
the
field,
instead of along
We may
it
as they should do in a uniform
calculate the mechanical forces experienced
field.
by unit
of volume of a substance by the proposition that the work done

by the forces in a displacement is equal to the loss of energy
of the system.
Let us call the force per unit volume H, H, Z.
Then
if
a body
is
displaced so that a point so, y, z comes into

+ By, z + Sz, and the corresponding change
the position x + &e, y

in
(I)
W be 8 W, we have
BW = -
JJJ(S&
1% + ZSz) dr.
AND MAGNETIZABLE
DIELECTRICS
199]
BODIES.
During the displacement the distribution of the
383
field varies,
W given in 119, namely,

W=2W - Wf
we have the important simplification that S V W = 0*, for after the
but
if
we use the form
of
motion as well as before the potential satisfies the conditions

180 (8) and (9). Accordingly in considering the variation SW
we have
to consider only the variation of p and /JL produced by

in other ways being taken account of
the motion, the change of
in the condition B V
by matter
in
W=
differently charged
38, putting
like
manner
point x
Sx,
coming
in p at
any point is caused

and we find as
to the point,
dm = pdr,
dx
In
The change
0.
//,
dz
dy
has changed to the value it formerly had at the

Sz, which has moved to x, y, z, so that
By, z
Accordingly (considering surface distributions as a limiting case of
volume distributions) since we have
we
W as
obtain the change in
"
00
dz
dy
and integrating the
first
[\dat/
integral
by
parts, the surface integral
vanishing at infinity,
(To the
\dyj
first order.)
384
Since this
[FT.
II.
CH. IX.
to be equal to
is
Jff(SS0
1% + ZSz) dr
00
for arbitrary values of &x, By, Sz
we must have everywhere
-<>-%- tor dj\(te

37 i 9Mf/aF
P 37
The
first
term
is
at each point.
s^tv^
the force exerted by the field on the real charge
The
It is of course in the direction of the field.
to the polarization, and is proportional to the

of
field
the
square
strength, and in the direction of fastest decrease
second term
is
due
That is, any point of a polarized body tends to move toward

ft.
the side on which the inductivity is less, or to bring more inductive
matter into the field, as stated above. !.The form of deduction
of
here given
is
due to Helmholtz*.
Medium. The modern theory of

and
due
to Faraday and Maxwell, assumes
electricity
magnetism,
that bodies do not act directly on other bodies at a distance, but
by means of actions transmitted through the intervening medium
from particle to particle. The influence of the medium has been
200.
Stresses in the
in this chapter, as we in fact started from the

of
the
expression
energy as being distributed in all space. It
remains to find a system of stresses that shall account for the
made apparent
electrical or
magnetic forces which have been here investigated.
per unit volume act on all portions of a body,

for example gravity, these forces will throw the body into a state
of strain, and in order to produce equilibrium the applied forces
If forces B,
H,
E, H, Z must be balanced by a set of elastic stresses developed in

the body. These are forces acting from point to point in the body,
and may be specified as follows. Suppose at any point
the
body divided into two portions, 1 and 2, by a plane whose normal

is n.
If we consider a small area dS of this plane containing the
point
the portions of the body on the two sides of the plane

*
Wiss. Abh. Bd.
i.,
p. 811.
See note in Appendix.
AND MAGNETIZABLE
DIELECTRICS
199, 200]
385
BODIES.
exert forces on each other, whose combined action may for either
Let
part be represented by a single resultant force applied to dS.
the force acting through the area dS on the portion of the body 1
be denoted by n dS, n is not in general normal
but has a tangential component. This

tends to cause the two portions 1 and 2 to slide
to dS,
over each other, or to be sheared. The normal

component of n if directed toward the body
Fm
the two portions of the body
approach each other, and is called a traction or tension, as in
the case of a stretched rope. If the force
n on 1 is directed
2,
make
tends to
toward
1,
liquid
a pressure, as in the case of

traction will be considered positive, that is
the force
pressure.
called
is
the force acting on a portion of the body has a positive component along the normal drawn outward from that portion. We
shall denote
the components of
Fn
by
X n Yn Zn
,
the suffix n
denoting the direction of the normal to the element of surface to

which they are applied. If we consider three sides of an infinitesimal cube at any point, we may specify the stress at that point by
giving the components of the stresses on each side, those on the
X Y Z
side perpendicular to the JT-axis being

X)
X)
side perpendicular to the F-axis being
y ,Yy ,Zy
those on the
and those on the

being
face perpendicular to the ^-axis

If we consider the equiliZ)
z
infinitesimal tetrahedron formed by
X Y Z
z,
brium of an
cutting off one corner of this cube by a plane

whose normal is n (Fig. 80), the areas of its
four faces being
dSX) dSy dSz $Sn

,
(the suffixes
denoting their normals) and its volume being

dr, we have for the equations of equilibrium, resolving along the
three axes,
+ Xx dSx + XydSy + Xz dSz - X n dSn = 0,

Rdr + Yx dSx + Yy dSy + Yz dSz - Yn dSn = 0,
Zdr + Zx dSx + ZydSy + Zz dSz - Zn dSn = 0.
Etfr
(i)
Now
the faces dSx dSy dSz are the projections of the face
the coordinate planes, and accordingly
,
dSx = dSn cos
w.
E.
dSn on
(nx),
dSy = dSn
cos (ny\
dSz = dSn
cos (nz).
25
386
If
we now
let
[PT.
II.
CH. IX.
the dimensions of the tetrahedron diminish in-
volume dr is of a higher order than the surface of

and can accordingly be neglected, accordingly the equa-
definitely, the
any
face
tions of equilibrium become,
dSn
dividing out,
Xn = Xx cos (nx) + Xy cos (ny) X

-f
cos (nz),
Yn = Yx cos (nx) + Yy cos (ny) + Y
cos (nz),
(2)
Zn = Zx cos (nx) + Zy cos (ny) 4- Zz cos (nz)
which proves the statement that the stress at any point, involving
the action on a plane element in any direction at the point, may
be expressed in terms of the nine components at the point,
*,
**> %X>
Xy, Yy, Zy,
g,
fi
Let us now consider the condition of any finite portion of

Let the body-forces H, H, Z, per unit of volume be
r.
If now the forces
element.
to
each
n Yn Zn applied to
applied
matter
each unit of surface are to produce the same effect as the given
system of body forces, then the system of body forces with their
signs reversed, together with the surface forces, would produce
For equilibrium we must have, resolving in the
equilibrium.
JT-direction,
(3)
Let us now express

equations
(4)
Xn
in terms of the nine components
by the
(2),
[Xx cos
(n e x)
+ Xy cos (n
y)
+ X z cos (n
e z)}
dS
Transforming the surface integral into a volume integral we obtain
every portion of the body is to remain in equilibrium under

the stresses, in order that the integral shall vanish for every field
and
if
of integration
we must have everywhere
(o)
dXg
x
d-^v
w = -^-a
h -^--r
das
_i_
-75
dy
dz
AND MAGNETIZABLE
DIELECTRICS
200]
387
BODIES.
In like manner we find
y_
~^
'
d#
'
82
9y
as the equations of equilibrium.
In order to explain electrical and magnetic forces by means of

stresses we must therefore be able to transform the expressions
already found for H, H, Z, into forms involving partial derivatives
as above.
Introducing into the expression for

i
(8
8F\
a /
the value of p from
8F\
8 /
SF
and transforming the derivatives we obtain

i
fl=
(7)
area
dv\
8F\
JL
The expression now has the required form of a sum

If we perform similar transformations on
vatives.
of three deri-
shall find that the equations of equilibrium are satisfied
Z we
and
by putting
J>
/3\'
V te J
* dvdv
- M
3F8F
4?r 3a-
dy
=dJ
1
~*
1
252
388
[FT.
CH. IX.
II.
= Zy etc., it is easy to see that the couple tending to

Since
z
turn any element of volume about either of the axes vanishes, as
If the body is not
is the case with ordinary elastic stresses.
does
not
condition
hold.
this
isotropic
We shall now apply the expressions found to determine the

nature of the stress in two particular cases. First, let the element
dS be perpendicular
\
cos (nx)
**
Xn
cos (nz)
="
in the equation (2)
Xx
+ Xy
+ Xz
~^r^ + 8^
-
These components of
37*
the force
Fn are
direction cosines of F, which
That
cos (ny)
-p,
and using these values
T
Yn
Then we have
to a line of force.
is
equal to $F/87r multiplied by the

in the direction of the normal n.
n is perpendicular to its plane.

plane
this
property is called a principal plane of the stress.
possessing
The stress being positive represents a tension. Accordingly the
is
medium
in a state of tension along the lines of force, of an

per unit of surface equal to %Fj^'jr, which, it may be
is
amount
noticed, is the
amount
of energy of the
medium
per unit volume.
Consider secondly an element tangent to a line of
we have
X cos
-
(nx)
Y
-
cos (ny)
Z
cos (nz)
= 0.
Multiplying this equation by FX/^-rr and subtracting

the expression for
Then
force.
it
from
Xn gives
$7 cos (ny) + ^TT EZcos (nz)

Zn = -O7T (2fZ - $F] cos (me) + -47T
-
(10)
i
\X cos (nx) + $F cos (ny) + $Z cos (nz)} =

47T
CK ET
77- cos (nx).

O7T
DIELECTRICS
200, 201]
In
like
AND MAGNETIZABLE
389
BODIES.
manner
CE
Yn = -
Tjl
Zn =
COS ( n V\
^W
COS (m).
Here again the components of n are equal to

multiplied by the direction cosines of n, or the force is normal to
its plane.
Consequently any plane tangent to a line of force is a
principal plane of the stress, and the stress is symmetrical about
the line of force. The negative sign shows that the stress is a
The state of stress consisting of tension along the lines
pressure.
of force combined with an equal pressure at right angles to them

was described by Faraday*, who expressed the matter in words that
state in effect that the lines of force tend to contract
and
to repel
each other.
This
may be
illustrated
by supposing the medium
to be divided
into filaments along the lines of force, and

these again to be subdivided into short
Then each
filaments.
short filament
polarized body which acts
is
like a doublet,
and
ri
i
f
since unlike
poles of successive elements are
in juxtaposition, the filaments all attract each
,
other endwise.
For filaments lying side by
side,
__
M
i|P
?1P
ill*
il
however, since
like poles are together, there is a sidewise repulsion.
Permanent Magnets and
201.
The fundamental laws
Polarization.
summed up
may
and in similarly acting bodies* the
be
induction
the force
is
in the
solenoidal.
Or
Electrets.
of magnetic
Intrinsic
and
statement that in
force is lamellar,
and
electric
soft iron
//,
times
in brief
curl.F=0,
(i)
(i')
Iron for which this statement
is
true
is
said to be perfectly
When
the external field affecting such iron is removed, the

As a matter of fact, this is an ideal
polarization disappears.
condition not exactly realized by any sort of real iron, for when
soft.
removed, a part of the polarization persists.

called residual magnetization.
The harder the iron or
the external
field is
This
is
steel,
the greater
is
the fraction of the induced polarization which

*
Faraday, Exp. Res. (1297).
390
[PT.
II.
CH. IX.
A substance in which, when the external field is removed,

the whole induced polarization remains, is called perfectly hard,
and a body consisting of such substance is called a permanent
persists.
The
magnet.
same as of
inductivity of such a body is to be considered the

Such bodies do not exist any more than perfectly
air.
We may however treat actual bodies as if they were

formed by the superposition of perfectly hard and perfectly soft
matter.
The portion of the polarization which permanently
remains is called the intrinsic polarization*. In order to carry out
the analogy, Heaviside has proposed to call a dielectric persoft ones.
manently polarized body an
electret,
Certain natural crystals
tion.
and
its
polarization electrizathis condition.
when heated assume
or intrinsic polarization now forms a real

or
electric
charge, and if the intrinsic polarization be
magnetic
The permanent
denoted by / with components
C we
BO,
have
for the real
density
(<u
0+ a/jo.
\dx
with a similar expression for
(3*
1
S
S~
~,
dy
<r.
182
this with the expression for p in
Comparing
(
aa
3g
1
4-7T [da;
SB.
(dA.
33|.
T 7T~
"\~^(
dz)
dy
we
find
SC,}
1
dz
dy
[das
(15),
the
divergence of the induction is equal to 4?r times
the
of
the
intrinsic
expressions
polarization. Comparing
convergence
for the apparent density, that is the sum of the real and induced,
186 (33), and in terms of the total
in terms of the force F,
the
or
120
polarization /,
,_
we
(6),
(dX
find
dY
dZ)
(dA
dB
d(
Accordingly
(
5)
or
j-(X
more
briefly,
div(J+47r/)=0.
(5)
The
solenoidal vector-sum,
the induction.
denote
*
it
We
by % M since
Thomson.
F + 47r7,
shall call it the

it
has been called in 121,

Maxwellian induction, and
corresponds to the definition of the
Eeprint of Papers on Electrostatics and Magnetism, p. 578.
201, 202]
391
It is solenoidal in
induction given by Maxwell.
intrinsically
as
as
well
bodies
The induction, g, which
elsewhere.
magnetized
divergent in intrinsically magnetized bodies, and which is

defined as /j,F, we shall call the Hertzian induction, and denote by
g#. In magnetically soft bodies these two inductions are identical,
is
but in intrinsically polarized bodies they

If
we
write equation (3) as
div(g H
(6)
and from
differ.
+ 47rTo) = 0,
subtract
it
div
(5)
(F +
47T/)
= 0,
we have
div
(7)
Now
if
we
call
g H - F) = 4-7T div (/ - 7
the induced polarization,
ii = K F =
(8)
).
we have
as always
(H^lF /=/+/,,
i
Inserting these in (7)

div
(gfl.
- F) = 4-7T div Ii = div
{(p
1) F},
and transposing div F,
divg ff =divo*n
(9)
agreeing with the definition of g H
Heaviside's treatment of Intrinsic Polarization.

202.
The treatment given by Heaviside differs in several respects from
that just given. According to that author the induction
solenoidal, so that true magnetic charges do not exist.
is
always
The only
reason given for this assumption seems to the present writer

insufficient, being, as stated by Heaviside, "to exclude unipolar
magnets." It appears that the exclusion of unipolar magnets

merely means that for any magnet the integral charge is zero,
which simply means that the distribution is what we have called

and lays no restriction on the divergence of the
polarization,
polarization or induction.
induction was what

it
It
might be supposed that Heaviside's
here called the Maxwellian induction, were

"
not for the fact that he says that " we use always g = /j,F. In
is
392
order to
make
we
these two statements, which
exclusive where there
[PT.
CH. IX.
II.
hold to be mutually
is intrinsic
polarization, appear consistent,

Heaviside proceeds in the following, as it appears to us, artificial
manner. In our notation, Heaviside* considers the field F as made
up
of a part h
4rr/ /M) defined as the intrinsic force, together with

we shall denote by F^ (F in Heaviside's
a lamellar force which

notation),
F=h + Fh)
curl
Fh = Q.
In order to make the induction solenoidal, he then puts
/<=**,
instead of /;
= /cF.
Then the induction
defined, not as
is
g* = ^+471-7,
but as
Inserting for
tions, this
the
sum
of the intrinsic
and induced
polariza-
becomes
= fj,Fh +
fjuh
= /jiF.
This gives, in conjunction with the equation supposed to be

fundamental, namely
div
= 0,
the equation
-
div (fiFh )
"rTr
which we
may compare
Heaviside's
pFh
=47T
div fih =
- div 7
with our equation (6).

Accordingly
has the property of our Hertzian induction. The
difference in Heaviside's treatment may be summed up as
A
1.
:
different definition of the total field.
Induced polarization
2.
produced by only a part of the field. 3. The Hertzian induction

considered solenoidal, even in case there is intrinsic polarization.
We
have stated the
difficulties of Heaviside's treatment as they

appear to us, without wishing to dispute the dicta of so weighty
an authority. The theory as we have given it seems to be that of
Helmholtz and Hertz, both of whom explicitly state that real
magnetism exists in permanent magnets. Neither they, however,

nor any other author, so far as known to the present writer, have
*
Papers, Vol.
i.,
pp. 453
4.
202, 203]
DIELECTRICS
AND MAGNETIZABLE
BODIES.
393
worked the matter out in detail as has Heaviside, nor have any
problems been solved in which a difference becomes of importance.
In
treatment, the flux of induction issuing from a

the
same, which is the quantity with which we are
magnet
concerned in practice, the ambiguity existing only in the substance
either
is
The difference between intrinsic and

of the intrinsic magnet.
other magnets is that in the former two independent vectors are
necessary to characterize the state of the body, while in the latter
one
suffices.
These
may be taken
% H and % M)
it
as
$H
as
/,
or as
% M)
/.
203. Variability of /A. Hysteresis. Throughout this chapter

has been assumed that the value of //, at any point was constant
This assumption is not borne out by the facts,

but was necessary in order to make the subject amenable to
for that point.
mathematical treatment.
It is found that
//,
is
a function of the
and that for magnetic bodies, in which this

has
been
most carefully investigated, as the force
phenomenon
strength of the
field,
increases, //, diminishes, finally tending towards the limit unity, so

that the ratio of the induction to the force approaches unity. At
the same time the difference between the induction and the force
maximum value, which is equal to 4?r

times the greatest intensity of magnetization that the substance
can assume. This is known as the intensity of saturation. For
tends towards a constant
wrought-iron this intensity of saturation has been found to be about

1 700 C.G.S. units.
The variability of //, does not affect the validity
of
Ohm's Law, which determines the
distribution of the tubes
of induction, although it seriously complicates the mathematical

In fact no cases of magnetization have been worked out
theory.
taking account of the dependence of jj, upon F. But this is not

the only defect of our theory. It has been found that for a given
value of F there is not a single determinate value of ft, but that
the value depends not only on the actual value of F, but upon the
values which have acted at the point in question at previous
If we plot a curve having as abscissas the values of
times.
at
a given point at various times and as ordinates the values of g at

the corresponding times, we may express this phenomenon by saying that the value of JJL at any point of the diagram depends on
the path by which the substance has been brought to the point,
This
is, on the whole history of the field at the point.
that
394
[PT.
II.
CH. IX.
phenomenon, discovered by Warburg*, and thoroughly investigated

by Ewingf, was named by the latter Hysteresis, to denote the aftereffects of the fields to which the substance has been submitted.
Warburg and Ewing found that if the field was increased to a
certain value, then decreased, and then varied successively between
the same limiting values, the path of the representative point on
the F-% diagram was a closed curve, which was re-traversed after
the first periodic cycle. This is called the hysteresis-loop, and its
area has an important physical significance. Such a loop is shown
in Fig. 82.
If instead of continuing to repeat the same cycle
we vary F between different limits the point may take any
position between the two limiting curves of the loop, as
shown in Fig. 83, both these figures being copied from Ewing.
FIG. 82.
FIG. 83.
If the cycle be so chosen that at
some
point, F, while decreasing,
passes through the value zero, the value of / calculated as

the corresponding value of g/4?r is the residual magnetization.
If the force F is still further decreased, its value when /= 0, 8 = F,
is called, after Hopkinson, the coercive force, since it measures the
negative force necessary to destroy the residual magnetization.

Besides these
phenomena
of hysteresis, there
is
another more
complicated effect, which causes the magnetization to arrive at

its final value only gradually, taking a certain time to reach
This is denoted by the name of viscous
its permanent value.
hysteresis,
tinguish
it
magnetic lag, or after-effect (Nachwirkung), to

from the proper or static hysteresis just described.
*
Warburg, Wied. Ann. 13, p. HI, 1881.
t Ewing, Phil. Trans. CLXXVI., p. 523, 1885.
dis-
AND MAGNETIZABLE
DIELECTRICS
203, 204]
395
BODIES.
204.
Dissipation of energy in Static Hysteresis. Since
we have seen that fi is not uniquely determined by the value of F,
so it must be for the energy of the field.
Accordingly the forces
acting on polarized bodies cannot be derived from a single-valued

potential, but must be non-conservative. In taking a body through
a cycle of magnetization, accordingly, a certain portion of the work
done upon
it
be stored up as energy, and is therefore

easily find an expression for the
The potential energy of a
energy.
to
fails
We may
dissipated into heat.

value of this dissipated
polarized
equal to
body in a
field
whose potential
is
is,
126
by
(2),
or in terms of the field
W=-l
we
If
\(AX + BY+CZ)dr.
consider an element of volume dr, and suppose
a point where the
X + dX,
the work
it
moved
to
field is
Y+dY, Z+dZ,
dW
done upon the particle during the motion is accordto

the increase in the value of the energy,
ingly equal
dW=-dr(AdX + BdY+CdZ).
(i)
In the second position the values of A, B,
have changed to the
values
A+dA,
B+jlB,
C+
dC,
but the change made by using these values in the expression for the
work would be of the second order and may be neglected. If instead
we change the strength
of moving the particle
done
be the same.
of the field the
Inserting the values of A, B,

of the induction and force we obtain
(2)
If
will
work
in terms
dF = -^{(3e-X)dZ + (g)- 7)dY+$-Z)dZ}.
now we vary X,
Y,
the value from which
(3)
Z through
we
a cycle of values, coming back to
started, the integral
396
vanishes, since the value of
at both limits
is
[PT.
II.
CH. IX.
the same.
The
integral
may be
Of
integrated by parts, giving
this the integrated part vanishes, since, as
found by Warburg
and Ewing, after the cycle has been once traversed

the same value on traversing the complete cycle.
that in taking the particle
operations, and leaving
quantity of work, which
it
is
3E
We
returns to
thus find
through the whole cycle of magnetic

in its original state, we have done a
equal, not to zero, but to
the integral being taken around a closed loop. Each term of the
The
integral must of course be obtained from a separate loop.
whole energy dissipated in the body
is
7d + Zd3\dr.
Of course the general theory is so complicated that it is not even to
be assumed that when we have carried the magnetization through
a closed cycle in one point of the body we have done so at all
In practice we can calculate the dissipation only in the
case of a uniformly polarized body, where A, B, C are the same at
points.
all
is
points of the body and in the direction of the force. The cycle
for all points, and the energy dissipated is
then the same
equal to
vol. of
body x
-.
^\\Tr
The
is
integral
evidently the area of the hysteresis-loop.
This area
is
inde-
pendent of the time of description of the cycle. In the case of

viscous hysteresis there is an additional dissipation which depends
in a complicated manner on the rate of description of the cycle.
204, 205]
205.
Hysteresis couple.
In the examples of
397
192198,
evident that a sphere or cylinder turned about an axis of

symmetry in the field would experience no resisting couple, for no
work would be done against the forces of the field. In like manner
it
an
an
is
ellipsoid would require on the whole no work to rotate it about

axis, for the forces hindering the motion in one part of the
revolution would have corresponding forces helping the motion in

another part of the revolution. If hysteresis exists, however, the
Then the ellipsoid in a position in

a diminishing acute angle with the
axis
makes
long
direction of the field experiences a mechanical couple tending to
The magnetic force parallel to the long
accelerate its motion.
case
quite different.
is
which
its
is then increasing, so that when the force has reached the

same value in the symmetrical position in which the axis
makes the same angle with the direction of the field, but on
axis
being then on the decreasing branch of the

the value of the magnetization is greater, so
that the mechanical force, which now retards the motion, is
the other side,

hysteresis-loop,
greater.
Accordingly the motion
is
on the whole retarded*, and
easy to see that the mean retarding couple

the mean difference of the ordinates on the
it is
is
proportional to
upper and lower
branches of the loop, that is to the area of the loop. Upon this
principle is based Evving's Hysteresis indicator")*, in which a long
sample of iron is rapidly revolved between the poles of a magnet,
and the mean couple between them measured by the pull on the
magnet.
The couple
is,
as seen above, independent of the time of
revolution.
*
An effect of this sort was observed in diamagnetic and very slightly magnetic
bodies by Mr. A. P. Wills, in the physical laboratory of Clark University, in the fall
of 1895, and was discovered independently by Mr. Wm. Duane, in the physical
Wied. Ann. Bd. 58, p. 517, 1896.
laboratory of the University of Berlin.
t Ewing, Journ. Inst. Elec. Eng. 24, p. 398, 1895.
CHAPTER
X.
CONDUCTION IN DIELECTRICS.
206.
We have hitherto
be perfect insulators. This can hardly be
even for the best insulators. On the other
Variable Flow. Relaxation-Time.

dielectrics to
supposed
said to be the
case,
hand, although, as we have seen, the greater the inductivity of a

dielectric, the more nearly does it act, as far as concerns electrostatic distributions, like a conductor, it is by no means likely that
the inductivity of conductors is infinite. Still less is it likely that
it is zero.
We shall now consider the consequences of considering
a dielectric to possess, in addition to
We
electric conductivity X.
shall
inductivity /A, an
deal with currents which
its electrical
now
are not in the steady state, and shall require to assume that at
any instant Ohm's Law determines the distribution of the currents,
namely
q
This assumption
is
justified
\F.
by experiment.
Instead of the sole-
noidal condition for the current, however, we must obtain a new

equation. This is obtained by the consideration that, if we consider
a portion of substance r bounded by a closed surface S, the total

charge within that surface increases in any interval of time by the
amount of total current flowing

is, if n is the internal normal
(
Ipdr =
into T through the surface, that
l{u cos (nx)
+ v cos (ny) + w cos (nz)} dS
Since this equation must hold for any portion of space, we must
have everywhere
dp_
dt~
But
399
206]
in a dielectric,
Differentiating (3)
by
(dS
93
ag)
and eliminating
t,
from
(2),
we
obtain for
a conducting dielectric
i d&
d
a
+ i-^r + 5^-V
4?r
dx
dt)
dy
we put
substance
u, v,
is
w, 3,
homogeneous
r
a^
(5)
),
+
+
dgn
d3
+ 4?r
^r + 5~
;r--^7
dt
dz ^
+T
4?r dt)
If
(4)
terms of the
as regards both
assuming that the

p and X, this becomes
field,
dx
^lf
a
8y
dz)
+ T- r
^XF+ T4?r-u-Tr^+r-^X^
U,-rr\
^ dt
4nr
dt
9
(
=0,
or in terms of the density
Integrating this differential equation,

-
we have
(7)
Accordingly whatever charge the body has originally decreases in

geometrical ratio as the time increases in arithmetical progression.
The constant
at
any
/x/4?rX,
which
is
the time
it
takes for the density
point to fall to l/e of its original value, has
been called by
Cohn* the
relaxation-time, a term used by Maxwell in connection

with the Kinetic Theory of Gases. For ordinary metallic conductors this time is so short as to have hitherto defied observation.
The importance
of its discovery was recognized by the committee

setting subjects for an international prize competition in 1893, who
proposed this as one of the questions for investigation!. It appeared
that no experimenter ventured to attack the problem, it being

The finite relaxation-time was
evidently considered too difficult.
determined
for so
good a conductor as water in some remarkable
experiments by Cohn and AronsJ, who

discovering the finiteness of
T for
are entitled to the credit of
conductors.
Cohn, Wied. Ann. 40, p. 625, 1890.

t Elihu Thomson Prize, Electrician, 1892.
J Cohn
Ann.
u.
Arons.
28, p. 454, 1886.
"Leitungsvermogen und Dielektricitatsconstante."
Wied.
400
ELECTROKINETICS.
[PT.
II.
CH. X.
207.
Method of Cohn and Arons. Consider a condenser
which
A,
may or may not be connected in parallel with the condenser B and the resistance wire
Let the capacity of A be K,
the inductivity of the dielectric
Let the conductivity of the
dielectric in A be X and in B zero.
Then the charge of one of the
182 (16),
plates 1 of A is in terms of the induction,
K
//,.
(8)
e*
=
4^: JJ
(^ cos ( nx ^
cos
n2/)
3 cos (*)) dS.
s,
On the other hand the quantity flowing through the dielectric in

the condenser in unit time is
=
(9)
-fij
+ v cos (ny) + w cos (nz)}
II {u cos (nx)
dS,
Si
so that,
assuming X and n constant,

dei
-W
4-TrX
- -
= 0,
(10)
If
we assume
that an electromotive force
is
applied to the
plates in order to establish a steady difference of potential

until a steady state of flow is attained, we have everywhere in
the dielectric p
= 0.
If the electromotive force is

suddenly re-
moved, we have from that time on

p
= 0,
and accordingly the difference
el
= e e -î
*
of potential of the condenser plates
is
V=V,e
(ii)
-*i
*
If the difference of potential
meter at any time
t,
V can
be measured by an electro-
we have
('>
logF -logF'
If in the second place the condenser

is connected in parallel
and wire of resistance R, we have for the
with the condenser

charge
of B.
e^
of the plate 1 of B, e/
= K' V where K'
is
the capacity
207]
If after the steady state
is
established,
401
we remove the
electro-
motive force and leave the system to itself, we have flowing through
the wire R per unit of time the quantity
F
I'
Accordingly
we have
the decrease of the charges
for
which when combined with the equation

(8)
~i
gives the differential equation

(e
(14)
+
'dt
Substituting for the charges

difference of potential F,
elt
/,
their values in terms of the
we have
KV+
at
IJL
Jti
which being integrated gives
Putting
R = oo
K' =
Considering the condenser
we
obtain, putting
07)
we
obtain the solution (i i) just found.

alone discharging through the wire,
K= 0,
V=V e~&x.
Q
conducting condenser accordingly behaves, when left to itself,

exactly like a perfectly insulating condenser discharging through a
The relaxation-time
wire.
of such a condenser
is
KR,
but
for a
conducting condenser, although we may use the same formula,

the relaxation time is independent of the form or dimensions of
the condenser, since, as we have seen in
the
Q be
184, if
capacity of the condenser with air as a dielectric,
the
we have
The relaxation-time is accordingly a characteristic constant of

medium, and may be determined independently of other
w.
E.
26
402
ELECTROKINETICS.
[FT.
II.
CH. X.
media, whereas we may determine only the ratio of the inductivity

of a medium to that of a standard medium.
we make two experiments with the above combination
If
condensers, one with
and
if
B which
alone,
which gives
T,
and a second with
of
gives
we know K' and
R we may from these two results determine K,
is made in
any shape suitable for calculating
from geometrical data, we can then determine
In this
manner Cohn found for water //, = 73'6, the largest value of the
and
if
the condenser
//,.
electric inductivity yet
metals, all that
found
we know
is
for
that
In the case of
any substance.
is
extremely small.
of course due to the large value of

so that whether
small we have as yet no means of knowing.
Condenser with two Dielectrics.
208.
is
This
is
large or
Absorption.
In
the preceding section we have seen that a charge residing in any

part of a conducting dielectric will gradually disappear, and that
no electricity will accumulate at any part of such a dielectric.
We
have considered only the discharge or leakage of a condenser,

We shall now consider the
starting from a state of steady flow.
state
which precedes the attainment of the steady state when
an electromotive force is suddenly applied to produce a difference

of potential between the plates of the condenser.
We shall also
dielectrics
contains
two
of different
that
the
condenser
suppose
shall
consider
a
and
for
we
only
plane consimplicity
properties,
of
Let
the
the
two
be
and
F2 and
denser.
Fi
potentials
plates
let that of the plane separating the two dielectrics be F3
,
Let the thickness, inductivity and conductivity of the upper

The force in
be d1} fo, X 1} and of the lower c 2 fa, X2
dielectric
the upper dielectric will be the same at all points,

lf which
In
the
lower
dielectric
let the
however depends on the time.
force be
Z)
also a function of the time.
Let the currents in the two dielectrics be ql and qz respectively, and let F1} F2 q l} q 2 be considered positive when measured
,
from F! to
2.
Let the condenser
plates, of area S,
be connected
by a wire of no resistance, into which we can suddenly introduce

an electromotive force E, which can be suddenly removed. The
207, 208]
circuit
may
also
be broken.
If
403
the current in the wire from
is
the plate 2 to plate 1, we have, since whatever charge arrives by

the wire is uniformly distributed over the plates
<2>
w=
(3)
^-
The
densities are determined
by the equations
(6)
Beside these we have always, taking the line integral of the

force plate to plate, the equation
(7)
From
(i)
and
(4), (2)
and integrating from

Idt
and (5)
to
= r,
X.F.dt
If F! be the greatest value of
[F, (T)
in the interval
= 0, t = r,
have
T
F dt<F r,
l
Jo
262
we
404
ELECTROKINETICS.
and since J\
is finite,
as
we
II.
CH. X.
we have
in the
[PT.
decrease r indefinitely,
limit, since ^(0), F.2 (0) are zero,

.
do)
That
is,
the forces
jump suddenly from
e=
the total quantity of electricity

the other.
From
This
is
zero to
and
while
lldt passes from one plate to
called the instantaneous charge.
the equations (10),
(7),
we
find
(II)
(12)
=
4-7T
the same as
if
there were no conductivity, as in
188.
The
ratio
or the instantaneous capacity, is the same as the true capacity.

in the wire, electricity
If we now keep the electromotive force
continues to flow into the condenser,

the same difference of potential a
its
plates always maintaining
F = E.
The capacity appears

In order to examine what goes on, we
must integrate the differential equations. Eliminating / from (8)
2
to increase without limit.
and
(9),
By means
we may introduce
of the equation (7)
and E, and differentiating the equation
from which we
may
obtain
dF /dt
in terms of
(7),
in terms of dFJdt, giving finally
as the differential equation for
lt
This
is to
be integrated with
208]
the arbitrary constant so determined that for
405
2
= 0,
has the
value
F^=l
(16)
The
integral
and from
From
this
is
and
accordingly
(7)
(4), (5), (6),
putting for brevity

__
=6
we have
(19)
for
^2
'
^>i
=&
the densities,
o- 2
=-
(a2
+ (6 - a )
2
The plane 3 accordingly acquires a charge, which is not the

case if the dielectric is homogeneous, or if the relaxation-times of
We
the two dielectrics are equal.

shall distinguish the values of
attained
after
the
time
t
F, <r, etc.,
t
by an affix, F^. Suppose
that the circuit be
now
broken.
We
accordingly have
therefore
'
4>7r
dt
(20)
+ \J*=0,
**jf
dt
4-7T
1=
and
406
ELECTROKINETICS.
[PT.
and the charges of the three planes begin to die away at

rates.
At any subsequent time later by an interval 2
II.
CH. X.
different
47TA.
F^ = F^ e~
(21)
_4m\t
"
(22)
If the condenser be
If
entirely disappear.
now
left to itself,
the charges will finally
however the plates are short
have the same conditions as in
the
first
stage,
circuited,
with
to
2
Accordingly the forces change suddenly from F^\
and there passes through the wire the instantaneous
2
(s)
charge
c'
(23)
We
(idt
=^
W"
"
have,
}S =
by
(FZ(S)
F^,
dis-
~ F'*) S
.
4?T
4*7T
now
Fl
(2)
we
E=0.
(7)
and since
we
obtain
e'
Vyu-^2
M'A
(If #2 the time of leakage through the condenser, be zero, and if

the
time of charge, be either zero or infinity, we find that conti
duction is without effect, and the instantaneous discharge,
e, is
,
equal to the instantaneous charge.)
There now remain the charges

(3)
47T
4tTT
<7 3
(3)
=<7 3
<
2
>.
208, 209]
407
If the circuit be again broken, the circumstances are the same

2, so that if the condenser is subsequently again short-
as in stage
we obtain a new instantaneous discharge, called the

residual discharge, and this may be repeated as often as we please.
It will be seen that the residual discharge arises from the charge
circuited,
that has accumulated by conduction on the plane 3, and that

there will accordingly be no residual discharge in a condenser in
which the relaxation-time is the same in every part. This is a type
<r 3
what would occur in any non-homogeneous dielectric, and it is

manner that Maxwell gave a possible explanation of the
phenomena of electric absorption, and of residual charge (Ruckof
in this
Maxwell's explanation has found confirmation in experimental results of Rowland and Nichols, Hertz, Arons, and Muraoka*,
all of whom found that when the dielectric was
perfectly homostand).
geneous there was no residual charge.

209.
Total and Displacement current.
mental equation
206
(4),
we
q
In the funda-
see that the vector

1
tty
4*77
dt
-f-
=,
JLî
1_Î
4-7T
If
is solenoidal.
dt
we
'
4-7T
dt
'
w + L^3
4-7T
dt
a-
fa cos (nix)
=
4f7T
'
consider the condition at the surface of an
ordinary conductor, in which we consider

insulator (in which q = 0), we have
do-
dt
3=
0,
surrounded by an
+ v cos (nil/) + w cos (n^)}
{# cos (n^c)
+ g)
cos (ne y)
+ 3 cos (n^)},
so that here also the solenoidal condition is fulfilled.
4-
-r
is
called
by Maxwell the
total current
The
It is a funda-
mental principle of Maxwell's theory that the magnetic

*
p.
Rowland and Nichols,
279,
1883;
p. 328, 1890.
Phil.
Mag.
(5) 11, p.
414, 1881
Arons, Wied. Ann. 35, p. 291, 1888;
vector
effects of
Hertz, Wied. Ann. 20,
Muraoka, Wied. Ann. 40,
ELECTROKINETICS.
408
[PT.
II.
CH. X.
the current are due to the total current, and not to the conduction
In insulators the part dft/dt 4?r, which alone
current alone.
.
insulators
called
the
displacement current, since

the
electric
The corresponding
Maxwell calls g/4?r
displacement.
magnetic quantity, which, since there is no magnetic conduction,
exists
in
is
constitutes the magnetic current, has important physical properties,

which will be considered in Chapter XIII.
PAET
III.
THE ELECTROMAGNETIC
FIELD.
CHAPTER XL
ELECTROMAGNETISM.
%
Magnetic Force due to Linear Current. The diswas

made by Oersted*, in 1820, that if a linear circuit be
covery
traversed by an electric current, the space in its neighborhood
constitutes a field of magnetic force.
The nature of the forces of
the field was completely investigated by Amperef, who found that
they were of the same nature as if they proceeded from permanent
210.
magnets.
derivatives,
They accordingly have a potential, which, with its first

is continuous and vanishes at
infinity, and which
satisfies Laplace's
equation at
all
wire, supposing that a single
points outside of the conducting
homogeneous medium
is
present.
We
have however seen that a single-valued, or uniform function

having all these properties vanishes everywhere. Accordingly the
magnetic potential due to a current is not uniform.
*
Oersted, Experimenta circa effectum Coriflictus Electrici in Acum Magneticam,

Copenhagen, 1820.
t Ampere, " M&noire sur la theorie mathematique des ph6nomenes electrodynamiques, uniquement deduite de I'expSrience. Gilbert's Ann. 67, 1821 ; Mem.
de I'Acad.
t.
6,
Ann. 1823.
THE ELECTROMAGNETIC FIELD.
410
We
shall in future denote the
[PT.
III.
magnetic force by H,
CH. XI.
its
com-
ponents by L, M, N, the magnetic induction by 53, its components

s
)Jt, $1, and the magnetic potential by H, reserving the notaby
,
tion F,
X,
Y, Z, g, $,
For the
quantities.
g),
3>
I7
",
for
the corresponding electric

we shall use the letter e,
electric inductivity
leaving //, for the magnetic inductivity. These distinctions have

not before been necessary, since we have not at the same time
considered both electrical and magnetic quantities, as we must do
from now on.

a point
If
we form the line

we have
integral of magnetic force from
to a point B,
fB
(i)
J
Ldx + Mdy + Ndz =
1A
-1
which must be independent of the path AB, for otherwise, by

changing the path infinitely little, we should, starting with the
given value fl At cause 1 B to change by an infinitely small amount,
and could thus cause 1 B to take at the same point a series of continuously varying values. The integral is acc<:dingly the same
for all paths that can be changed into one another by continuous
If, however, the current separates two paths ACB,
the integral is not the same for both. In other words, while
the integral around any closed path not linked with the circuit is
deformation.
ADB,
zero,
the integral around a path linked with the circuit is not.

integral around any two closed paths each linked once
But the
with the circuit
the same, for they may be continuously

deformed into each other. Or in other words,
we may connect two such paths 1 and 2, Fig.
is
The
integral around the

is not linked
which
ABPQDCQPA,
with the current, is zero, but this is equal to
the sum of the integrals PABP around 1 in
84,
by a path PQ.
circuit
the positive direction, together with the integral QDCQ around 2 in the negative direction,
while the integrals over the coincident paths
PQ, QP in opposite directions destroy each
other.
We
Accordingly
shall say that
two geometrical
circuits are linked positively,
ELECTRO MAGNETISM.
210]
411
when, given a direction of circulation about each
circuit,
the
direction of circulation in one circuit agrees with

the forward motion of a right-handed screw,
whose rotation corresponds to the direction of

circulation in the other circuit.
sents two circuits
negatively below.
Fig. 85
reprelinked positively above and

By an extension of the above
+ Right
reasoning we see that the integral around any

circuit linked n times in the positive manner
with the current
around any
is
circuit
nJ, where J is the integral

linked once.
Accordingly
any point is an infinitely

valued function, whose values differ from each other by integral
We may however make the potential a uniform
multiples of J.
function, if we prevent passage from one point to another by
the
potential
at
paths not continuously deformable into each other,

that is, if we reduce the doubly connected space
about the current to a singly-connected one by

means of a diaphragm covering the current
Then no two paths can be separated

If we consider the potential
by the current.
circuit.
FlG 86
-
at two points infinitely near each other but lying on opposite

sides of the diaphragm, Fig. 86, to get from one to the other we
must perform a closed circuit about the current, so that their
potential differs by the amount J, accordingly in crossing the
diaphragm, the potential is discontinuous, the amount of the
discontinuity being
where
on the positive side of the diaphragm. There is, however, no discontinuity nor lack of uniformity in the derivatives
is
of n.
If we now consider all space, except a small sphere of radius
with center at the point P, and apply to it Green's theorem
and for U the

function 1/r, where r is the distance from P, the volume integrals
vanish, and the surface integrals are to be taken over the infinite
where
for
we put the magnetic
potential
fl,
412
sphere,
[PT. III. CH.
XL
where they vanish, over the small sphere about P, where

as in
83, on making R decrease indefinitely,
we have
and over the two
sides of the diaphragm,
where we have
"i
continuous, the first terms in the two
integrals cancel each other, the normals n^ and n 2 being in opposite
directions, and since
Since, however,
dl/dn
is
^
dWj
this
becomes
and
finally
a47rn P
(3)
|T(n s
Since
flj
H = J we
2
*? =
n,) -
o.
vUi
JJ
have
so that the action of the current
is
the same as that of a magnetic
double-layer or shell of strength
This result was given by Ampere, by different reasoning. Experiment shows that the magnetic forces are proportional to the
strength of the current, so that
tionality,
(6)
if
be a factor of propor-
ELECTROMAGNETISM.
210, 211]
413
the surface integral, being by Gauss's theorem,

30, equal to co
the solid angle subtended by the current circuit at the point P.
positive side of the shell and the one toward which the normal
to be drawn is the side toward which a right-handed screw
The
is
advancing in the direction of the normal to the diaphragm, would

move when rotating with the current. The line of force is
Since the potential everypositively linked with the current.
substance
of
in
the
the
where, except
conductor, satisfies Laplace's
equation, the force is everywhere solenoidal, and the tubes of
force are endless, and are all linked once with the current.
211.
factor
is
a natural constant,
extremely small, that
It is
The determination
Electromagnetic Units.
A, which
static units of electricity
is,
is
an enormous number of
must pass
of the
a matter of experiment.
electro-
in unit time in order that the
may produce magnetic forces of appreciable amount. If,

however, we choose a new 'unit for /, defined by the assumption
A = 1, so that
current
we get a new system of measuring currents known as the electromagnetic system. The unit magnetic potential is defined as the
potential at unit distance from the unit magnetic pole in vacuo,
accordingly the electromagnetic unit of current is referred at once
to a
magnetic pole, instead of to an
electrified point.
definition~oT^the
new unit
of current
whole system of
electrical
unite.
we may
We
From
this
at once obtain a
define the
new
unit of!
quantity of electricity as the quantity passing in unit of time

when a steady current of one electromagnetic unit flows. From
this definition of unit charge we obtain, as before, new units of
of electric potential, of resistance, capacity, and the rest.
Conversely if, measuring the current in electrostatic measure, we
field,
put A = 1 we shall get a new unit of magnetic potential, from

which we may obtain a complete set of units for magnetic quanreferred to the unit of electric charge, instead of to the
unit magnetic pole.
may thus measure electric quantities in
the electromagnetic system, or magnetic quantities in the electrostatic system, or as before, each kind of quantity in its own
tities, all
We
appropriate system, thus obtaining the Gaussian system.
414
[PT.
III. Ctf.
XI.
Dimensions of the Units. If we denote the numeric

of a quantity when measured in the electrostatic
system by the
suffix e and when measured in the
electromagnetic or magnetic
we
suffix
have
for
the
the
m,
system by
magnetic potential
212.
fl
(1)
m = AIe a) = Im co,
le
(2)
=I
(0.
(D=-Îm
Consequently the number A denotes the ratio of the numeric of a

when measured electromagnetically, to the numeric
certain current
same quantity measured
of the
number
I/
is
the
of electrostatic units of current in one electromagnetic
If
unit.
electrostatically, or
equation, by
denote a magnetic charge, we have the dimensional

190,
the quantities being measured in either system. Also since the

dimensions of solid angle are zero, the dimensions of O are the same
as of /,
and
(4)
Since the unit of electric charge in either system is obtained

from the unit of current multiplied by the unit of time,
(5)
and we accordingly have
for
the ratio of the two units of electricity
or of current, inserting the suffix
m in
(5)*
Now the fundamental assumption in denning the magnetic system

was that the dimensions of
were zero. Also the assumption
the
electrostatic
defining
system was that the dimensions of e
were zero. Accordingly the dimensions of ee and of mm both
belonging to the Gaussian system, and defined by precisely the
same considerations, namely
//,
Evidently any dimensional equation holds
on both
sides.
when
either suffix e or
m is
inserted
ELECTROMAGNETISM.
212]
415
Hence the dimensions of the quantity I /A are accordAll that has been said of
same
as those of a velocity.
the
ingly
course applies to any absolute system of units, and has no re-
are the same.
If the units of length, mass, and

striction to the c.G.S. system.
time are given, we can by definition immediately obtain the
unit of electricity in either the electrostatic or electromagnetic

system, and by experiment determine the number of electrostatic
units contained in
one electromagnetic. If the unit of mass is

define our electrical units as before, the size
now changed, and we
of both units of electricity has changed, but in the same ratio,

so that the number of one kind contained in one of the other
is
the same as before.
If,
on the other hand, we change the
unit either of length or time, the two electrical units change,

but in different ratios, so that the numeric expressing the number
of one kind in one of the other
is
changed from
its
former value.
however, changed in precisely the same way that the

numeric expressing any given velocity has changed, so that we
It has,
the
number
A represents
a certain definite velocity,

of
the
units
chosen. When the units
totally independent
and
of mass, length,
time have been settled upon, the numeric of
may say that

which
I/
is
may be given. This velocity will be denoted by v.

It is to be noticed that the determination of the quantity v
this velocity
depends upon the determination of a certain numeric, the units

being settled upon, and that there is nothing of the nature of an
actual velocity involved. We shall, therefore, not as yet be understood to speak of v as a velocity, but merely as a quantity whose
numerical expression changes like that of a velocity, with any
change of units. The quantity y is the most important electrical
Numerous determinations of its value have

by Wilhelm Weber* and Rudolf Kohlrausch,
The number now generally accepted is
natural constant.
been made, the

in 1856.
first
v=3
Electrical
x 10 10 cm./sec.
and magnetic potential are defined in terms of
work, so that
[eV]
(8)
[mtt]
= [ML*T-*l
which agrees with the other possible definition
Weber, Elektrodynamische Maassbestimmungen
iv.
1856; Werke, Bd. in. p. 609.
41 6
From
(8)
and
[PT.
III.
CH. XI.
we obtain
(6)
(9)
[ee
V = [em Vm
e]
],
and there are v electromagnetic units of potential in one electrostatic unit.

Capacity is denned as ratio of charge to potential,
so that
do)
from which
or there are
unit.
v2
electrostatic units of capacity in one electromagnetic

is defined as ratio of potential to current, so that
Resistance
or there are
213.
was due
electromagnetic units in one electrostatic unit.
The absolute system of units

Practical System.
and was introduced to practice by Weber. The
to Gauss,
first made practicable for general use by the exertions

of the British Association, which issued copies of the unit of
system was
resistance,
and decided on various multiples of the
magnetic units
for practical units.
Its action has
c.G.s. electro-
been seconded
congresses, at Paris in 1881, 1884 and 1889, and

at Chicago in 1893, which determined on the following multiples of
the electromagnetic units
by international
=
=
=
Ampere
Coulomb =
=
Farad
=
Joule
=
Watt
1 Volt
1
1
1
1
Ohm
10
C.G.s.
electromagnetic units of Potential.
10 9
Resistance.
10" 1
Current.
10" 1
Electric Charge.
10~
Capacity.
10 7
Work.
10 7
Activity.
are used before the preceding names

of the units to denote respectively multiplication and division by
The
prefixes
mega and micro
417
ELECTRO-MAGNETISM.
212214]
These units form a consistent system, so that electrical

measured in these units require no
a million.
relations involving quantities
For instance, a current of one ampere is

force of one volt is impressed in
an
electromotive
produced when
a circuit whose resistance is one ohm, and the activity of one watt
thereby exerted dissipates energy at the rate of one joule per
numerical
factors.
second.
the above definitions
we
compared with Practical Units. From

with the value of v given and equation (9')
Electrostatic
214.
find
1 c.G.S. electrostatic unit of Potential
From
1
The
= 300
Volts.
(10')
Farad contains 9*10 n
c.G.S. electrostatic units of Capacity.
electrostatic unit of capacity is the unit of length, accordingly

1
Microfarad
900,000 cm. of Capacity.
sphere of nine kilometers radius in free space would have a

From (6)
capacity of one microfarad.
1
Coulomb
3'10 9 C.G.S. electrostatic units of Electric Charge.
From ( u')
Megohms = 1 C.G.S. electrostatic unit of Resistance.

From equation (7) we may find the dimensions of e and m, when
900,000
those of
and
//,
are settled upon.
Any
convention that
may be
made
It must be noticed,
gives us a possible system of units.
however, that there is always a relation between the dimensions of
e
and
fju.
From
equations (4) a*nd (5)
m
Squaring this and dividing by
(7)
we obtain
m
2
r^ i _
~
w.
E.
m
w
'
27
418
[PT.
CH. XI.
III.
Consequently the dimensions of the product of the
electric
and magnetic inductivities must in any system be those of the
The
square of the reciprocal of a velocity.
absolute dimensions
of either factor are arbitrary.

Attempts have been made to settle
the absolute dimensions of e or //,, but they are evidently based
The two
upon misconceptions of the theory of dimensions.
common assumptions
are, that
= 1,
A*
= -.
Secondly we
This gives the electrostatic system.

/*
This
gives
the
= !,
may assume
= -*
electromagnetic
system.
We
shall,
when
dealing principally with the magnetic properties of currents, use

the electromagnetic system, but when dealing equally with electrical
and magnetic phenomena, to avoid ambiguity, we shall,

Helmholtz and Hertz, use the Gaussian system,
following
quantities in the electrostatic system,

in
the magnetic system, and introducing
magnetic quantities
the factor
with the numerical value 1/v.
complete table
all
measuring
electrical
all
dimensions
of
of
the
various
units
is
given at
the end of
Chapter XIII.
Potential due to Circular Current. The potential

to a current being fl = /o>, where a> is the solid angle
215.
at
P due
subtended at
by the current
circuit, if
is
situated at a dis-
tance x from the center of a circular current of radius R,

line through its center
perpendicular to its plane,
the
have
for the area of the
about
on
we
segment of the sphere of unit radius

cut off by the right cone whose vertex is P, and base
the current,
(i)
ft)
=
=
27T
~-
27T
This may also be obtained, according to

123, by differentiating the expression for the potential of a disc at a point on the
axis.
ELECTROMAGNETISM.
214, 215]
The
419
force in the direction of the axis is

7"
At the center
,^_
_^
. ..__
of the circle
L=
From this expression comes the definition often given of the

unit of current as that current which, flowing in a circle of unit
radius, produces the field 2?r at its center, or less correctly, the
current, which, flowing in
produces unit
circle,
The expression
an arc equal to the radius in a unit
field at
the center.
for the force is
an example of the proposition
that
traversed by equal currents,
geometrical
produce at corresponding points forces inversely proportional to
For at corresponding points the solid
their linear dimensions.
In the circuit of
and
therefore
the
potential is the same.
angle,
similar
circuits
n times the dimensions, the potential changes by equal amounts

for displacements of n times the length, hence for equal displacements the change is l/n as great, and the force is n times smaller.
When
the point is not on the axis of the circle, the cone,

an
oblique section circular, is elliptic, and we must calhaving
culate the area of the spherical ellipse cut out by it from the unit
sphere.
This involves an
We may
elliptic integral.
however develop the result in an
infinite series of
zonal spherical harmonics, as in 'the case of the potential of a disc,

in
102.
Developing the above expression for co at points on the
axis
by the binomial theorem, we have
co
272
420
Accordingly for points not on the

center of the circle,
24
axis, at
[PT. III. CH.
a distance r from the
+LMWp
2.4.6W
+
XL
r> R
1-4
0)=27T
In order to find the direction of the force we must differentiate

this in the directions parallel and perpendicular to the axis, and
take the resultant.
figure of the lines of force is given by
Maxwell, Plate 18.
Law of Biot and

Straight Current.
If we have a current flowing through a straight linear
Infinite
216.
Savart.
conductor of infinite length, we may consider the circuit completed by conductors lying at an infinite distance all in the same
The solid angle subtended by the circuit at a point

be that sector of the unit sphere with center at P included
between the plane through the straight conductor and P, and a
plane.
will
plane through
parallel to a given plane, which is assumed to
be the plane of the circuit. This angle being <, we have the
ratio of the solid angle co to the surface of the unit sphere equal
to the ratio of the plane angle to the circumference of the unit
circle,
But $ is equal to the angle made by a plane through P and

the conductor with a fixed plane through the conductor. Consequently the equipotential surfaces are planes through the conductor,
and the
lines of force are circles
whose planes are per-
pendicular to the conductor.
The
line integral of force
about a
circle of radius r is
the value
H, which is tangential to the circle, times the length

of the circumference, and this must be equal to 4?r/,
of the force
47T/
= ZirrH.
Accordingly the value of the force
is
H-*.
r
This
is
the law of Biot and Savart*.

*
Biot et Savart, Ann. Chim. Phys. 15, p. 222, 1820.
215
ELECTROMAGNETISM.
217]
421
Force due to any Linear Current.
217.
at a point
the distance
H and
PQ = &h,
at a neighboring point
is
and
if
is
If the potential
is
H + SH, where
the magnetic force at P,
we
have
Stt
(2)
This change in the potential is the same as the change that

would be made in the potential at P by moving the whole circuit
parallel to itself the same distance Bh in the opposite direction.
The change Bfl is proportional to the change So> made in the
solid angle
subtended at
P due
to the
motion of the
circuit,
which
is easily seen to be exactly the solid angle subtended at

the narrow ribbon of cylindrical surface whose edges are
and
by
the
positions of the circuit, and whose generating

and parallel to Sh. But any arc ds has described
in the motion an area dS of a parallelogram equal to
initial
final
lines are equal
dS = dsShsm(ds,
(3)
and
Sh),
n be the normal to this element of area, we have for the

element dSco of the solid angle subtended by it at P,
if
dSco
(4)
where r
is
dsSh sin
(ds, Bh) cos (nr),
the distance of the element from P.
Consequently
integrating around the ribbon

(5)
fr
(6)
m
If
we
consider that each element of the current of length ds

dl and the force dH, we
contributes to the field the potential

have, by (2),
- dHSh cos (dH,
(7)
The numerator
are
There
of
r,
is
-4K> - ?
r&Mrinfrfe M)eoB(nr)
_
the volume of the parallelepiped whose sides

It therefore vanishes if the direction of Sh coincides
r, ds, Sh.
with that of
to)
is
r.
accordingly no component of the force in the direction

is
perpendicular to r. In like manner if &h has
or the force
THE ELECTROMAGNETIC
422
FIELD.
[PT.
III.
CH. XI.
the direction of ds, the force vanishes, so that the force is perpendicular to ds. If &h is perpendicular to the plane of r and ds,
we have
cos (dH, Bh)
-
dH
,
to
sin (ds, Bh)
1,
rds cos (nr)
rds sin
(r,
1,
ds)
~^~
^~
Accordingly if we call da- the component of ds perpendicular

the magnetic force due to the whole conducting circuit will
r,
be obtained
if
we suppose each element
ds to contribute to the
the amount
field
dH = %,
(9)
which has the direction perpendicular to the element ds and the

radius
r.
The total field is the vector sum of all these infinitesimal parts.
The proper sign to be chosen may be found by considering the
way in which the lines of force are linked with the current, and
we find that the direction of the force is given by the rotation
a right-handed
of
direction of
cisely stated
3
da-.
screw advancing with
The complete
by saying that the
current
in
the
be most con-
specification may
force due to the element ds is
times the vector product of Ids and

to the point P,
1/r
the
r,
the vector r being
drawn from the element ds
dH = -Vds.r.
r
(10)
The
the
resolution of the field into elementary fields is artificial, for

Moreover
course due to the whole closed circuit.
field is of
may be performed in an infinite number of ways,

the integral of the above differential taken around the
whole circuit which gives the field.
the resolution
for it is
We may
ential of
consequently add to the differential above the differany function of the coordinates of the element ds, for in
integration around the circuit this function returns to

value so that the integral vanishes.
its original
If the coordinates of a point in the current circuit are
these of P,
x, y, z, since
#a ylt zl}
the direction cosines of r and
ds-^
are
ELECTROMAGNETISM.
217]
423
respectively
'
ds l
'
ds l
ds :
'
we have
field
for the components of the vector-product

representing the
due to an element dslj
dL =
z
~B [fyi (
dM = -{dz
(11)
dN== ~s dx
\
We may obtain the

31.
~ *0 ~ d*i (V ~ 2/i))>
(x-x^-dx,(z-zJ\,
(y
same
- 2/1) - fyi ( x -
result
i)}-
by the use of Stokes's Theorem,
Since
the component of the
field in
the direction h
is
Let the constant direction cosines of h be

n be
X, p,
v,
a.,
/3,
7,
and those of
variable over the siirface of the diaphragm.

a
Then
(13)
= X^-- p
9a?
8
_
5~
en
Now
\-
+ve
since
we have
_
dx
'
d%!
'
'
dy
dyi
dz
dzi
424
[PT. III. CH. XI.
so that
-^f-
= -(
-^- + -J^- + 7 -^
(r)
V-
-+
(14)
'dzj\r
32
Now
since 1/r satisfies the equation
M3-.,
we may
write
a^2
with similar substitutions
for 3 a
2
2
(l/r)% and a^l/r)^
1
these substitutions, and arranging the terms differently,
'
(15)
+P
Consequently
if
we put
Making
we obtain
ELECTROMAGNETISM.
217]
425
0-/\/8
FA =
(16)
7^y
W*=I\*
U
resultant of
h
Tf/^ equal to / times the
F"^,
vector product of the unit vector h and the vector parameter of
1/r, the force in the direction h is
making Qh the
7k_m,\
?i
=But by
(18)
(curl
Qh ) cos
Stokes's theorem this
(Uhdx^
around the current
is
(curl
Qk
9a?!
n) dS.
equal to the line integral
Vh dyi+ Wh dz )=1
Q cos(Qhds)ds
Jl
circuit.
Accordingly attributing to each element ds the amount of
field
dH,
(19)
dH cos (dH,
and since
this
coefficients
we
h)
= adL + /3dM +
must hold
for
every value of
obtain
=/l!irU-
which give the values obtained in
( 1 1 ).
a, j3, 7,
equating their
426
218.
Forces
[PT.
III.
CH.
on Conductor carrying Current.
magnetic energy of a pole
m at P in the field
due to a current
XL
The
is
(i)
dn
where
Om
is
the potential due to the pole. For any number of

manner ft m being the potential due to them all,
of poles, in like
W=
(2)
which
is
the flux of force through the current circuit in the
due to all magnets. The potential energy

negative
tends to decrease, consequently a current in a magnetic field
tends to move so as to make the surface integral a maximum, that
direction,
embrace the largest possible number of tubes of force linked

in the positive direction. This statement of the mechanical
action of magnetic forces on a current is due to Faraday.
is,
to
with
it
219.
Mechanical Force acting on Element of Circuit.
We may
consider the forces acting on the whole circuit as the

of
the forces acting on each element ds lt with the same
resultant
degree of arbitrariness as in the case of the field due to the current.

By the principle of reaction the force on c^ due to the presence of
must be equal and opposite to the force dL, dM, dN

a unit pole
on the unit pole, due to the current element dslf Consequently if
d"&, cZH, dZ, are the components of the mechanical force acting on
da,
d%=- {dy, (X - z) dR =- [dg
(3)
dZ =-
dz,.
(yl
(zl
- y)},
-
( Xl
-sc)- dx
[da^
(2/1
- y) - dy (^ - x)}.
*)},
But
A"4
A*
,y'-
'-
,'
y
*'
are the components of the field at ds l due to the unit pole at P.
ELECTROMAGNETISM.
218, 219]
427
Consequently
JB
dU
(4)
dZ = I (Mmdx l
and the whole
force
due to the presence of any number of
magnetic bodies producing a field L, M,

the individual actions
is
the resultant of
all
(5)
That
is
the mechanical force on the element
product of the current element Idsl

where it is situated*.
is
the vector
and of the magnetic
field
Suppose that the magnetic field is due to a second element ds2

72 at a distance r from dslt Then since by (i i) 217,
ds.
for x, y, z,
2 for ds l} x1} y l} z
putting
of strength
dL = -
{dy* (zl
dM = -J {dz
we have
for the
- z ) - dz
2
^2 )
(^
(yl
- y )},
(z^
z}},
dx%
mechanical force acting on dsly by
=~
Adding and subtracting the term
(5),
dx-^dx^x-^
3
x^/r this
maybe
written
(7)
I I2 ds ds 2
1
r2
cos (r#) cos (dSidSf)
r being drawn from c?^ to ds2

*
force
(cfoatf)
cos
would be hard to devise a simpler rule

than the one given on p. 12.
It
cos
for
remembering the direction of the
428
[PT.
III.
CH. XI.
In like manner
2
_2 cos
(ry) cos (ds^dsj)
1 12 ds
1
The
ds
cos rs cos
^s*
- cos (ds
cos
y) cos
cos
d zR has a component
resultant
COS
in the direction of
r,
and one of magnitude

cos
in the direction
cfo 2 .
(r, cfet)
This resolution into infinitesimal forces
is
unfortunate on account of the lack of symmetry with regard to
the two elements.

220.
Mutual Energy of two Currents.
acting on the
already found
(8)
H = /i/
The whole
force
found by integrating the expressions (7)

2
2
2
cZ
B,
H, d Z, around both circuits 1 and 2,
circuit 1 is
for
-^
J lJ
[da^
C'd/i
-|
J
<foj
1^2
with similar expressions for H and Z. If we now suppose the

circuit 1 displaced or deformed in any manner, so that a point
#i> 2/i î> i
displaced by the amount 8^, Sy1} 8^ the circuit 2
>
being
(9)
fixed,
the forces H, H,
do the work
429
ELECTROMAGNETISM.
219, 220]
The second
may be
factor in the second integral
written
and we may then perform the integration around the

integrating by parts, obtaining
circuit 1,
ds l
*''
/,
1
2
!+
1
-
[
J
-^
-
T-
dsl
dz.
dz,\
-
dsj.
ds l
The integrated part vanishes, for the factors Ba;lt 15 8^ are the
same for the beginning and end of the circuit. Accordingly the
expression for the work becomes
T r
ja(l)
[ [
af\r
(i)
Vr/g
ni
2^
r
l
~|
JiJ2^
Sj
denoting the change made by changing x^ y^zl} keeping

We have accordingly obtained the work as the change
constant.
due to the motion in the value of a
line integral
around both
Consequently the mechanical forces are derivable from a

force-function, and the integral represents the negative mutual
potential energy due to the magnetic forces acting between the
circuits.
two currents.
(1 1)
W= I 7
I
J 1J 2
- (dx^dx^ + dy-^dy^ + dz^ dz

^
z)
cos
This form of the integral was given by Franz Emil Neumann* in

1845 and is generally known in Germany by the name of the
Electrodynamic Potential of the currents

*
1845.
Neumann, "Allgemeine Gesetze
and
2.
der inducirten Strome."
Abh. Berl. Akad.
430
221.
[PT. III. CH.
XL
Various Resolutions into Elementary Forces. The

if we add to the
integrand
value of the integral will not be changed
any expression,
where
F is any function
of
r, for it
around either
when
will disappear
If
circuit.
we put
F=
integrated
r,
dtp,
ds
fi'Y*
\J I
uSi
OXi
fi
H
?*
C//
nf*
Ct/t//|
G/SI
y\
fiî
\JU U
r\w
U f ri *y
(JL&'\
^vî
u2!i
dSi
If, Fig. 87, we drop a perpendicular from ds2 on

the tangent at ds lt and call the length of the
tangent thus cut off p, we see by infinitesimal
r/T*
\J I
geometry that
FIG. 87.
ds l cos
dr
Accordingly
ds^
(r,
= dr
/
= -cos(r,
-
ds2 cos
(^5!,
ds2 )
= dp.
dj)i
C/oj
dp = cog ds
(12)
ds
ds 2
dr
= cos
ds 2 ).
(r,
o^
p=r
But
cos
(r,
Consequently
(13)
cos(cfei, d,
dp_
_8_
VO 2
UO 2
dr\
= _drdr_
2
d r
and
&r = -- {dr dr
-+ cos
r -{5-51
(14)
GS^SZ
\p8i
cos
ds 2
(r, (foj)
cos
(r, c?s2 )
~ cos (^5
>
^2)
"7~
and multiplying
this
to the integrand in
by an arbitrary constant
1 1
(Yl
+
The value k =
may
1 gives
&)/2,
and adding
)*
ni
_
(1
-+-
k}J
"
(1
-=
cos(ds1} ds 2 )
k)-
cos
(r,
Neumann's form
ds^ cos
(r,
of the integral, from which
be obtained the resolution into elementary forces already

*
Helmholtz, Wiss. Abh. Bd.
i.
p. 567.
ELECTROMAGNETISM.
221]
given.
For k =
Weber and
1 we get a resolution into forces proposed by

Neumann, and for k = one implicitly suggested by
C.
Let us examine the case k
Maxwell.
W = IJ,
(16)
COS(r
I
I
J IJ 2
f
From
(17)
431
we
this
STF
-^
1.
fl3r<dr
obtain
= Jj/a
Sr
^*
^*
^î ^^2
^^2
C"î
^*
,r
^
C/Sj
C/5 2
ds^ds^.
Integrating by parts, the second term around the circuit 1, and

the third around the circuit 2, the integrated parts vanishing in
both
cases,
2
J 2
9*i 9s2
9*i
\r 9sJ
8s 2 \r
9i
(18)
,
T
= /j/o
JiJz
--1- 5- ^dr + -2
2
8 r
Since the integrand contains the factor Sr, work is done only when
the distances apart of some of the pairs of elements are changed, and
we may resolve the action into attractions between dsi arid dsz of
the magnitude
2
-T2
= /i/
{2 cos (ds l) ds2 )
3 cos
(r, ds^)
cos (r,
This form for the elementary forces was given by Ampere*. According to this form, we see that parallel elements perpendicular to the
line joining
them
attract each other with a force
r2
,
Parallel elements having the direction of the line

joining
repel each other with a force
Ampere.
dynamiques."
"Memoire
Mem
sur la theorie mathematique des

vi., 1823.
de VAcad. T.
phenomenes
them
electro-
432
[PT.
III.
CH. XI.
while mutually perpendicular elements exert no action on each

other if either is perpendicular to the line joining them.
We
Currents distributed in three Dimensions.
222.
have seen in
210 that a current /
is
equivalent to a magnetic
shell of strength
=
4^'
where
J is
the line integral of the magnetic force around a circuit
once with the circuit, and in the electromagnetic

positively linked
system /
= 4>.
Accordingly
J=4>7rl.
now we
consider steady currents distributed in any manner in a

conducting body with current density q, the integral of magnetic
force around any closed curve depends only on the tubes of flow
If
with which
it is
linked, being equal to 4?r times the total current
through the curve.
Consequently
Ildx + Mdy + Ndz
(i)
4-7T
{u cos (nx)
dS,
the surface integral being taken over any surface bounded by the
curve.
But by
Stokes's theorem
Ldx + Mdy + Ndz

dN~
or
~ dM\
-57
cos
nx ) +
fdL
\JJJLM.
+ (-5
\dz)
The
if
we have everywhere
---dy
4>7TV
4f7TW
3dz
-- ^d
= -=dL
dx
dz
=
-- dL
-=
dx
-5-
dy
cos
OJu\
5- J cos
dyj
surface integrals can be equal for all surfaces
curve whatsoever only
(2)
dN\
Ur ~ ^r
(nz)}-
dS.
'}
bounded by any
221
ELECTROMAGNETISM.
223]
433
These are the fundamental equations of electromagnetism. In the

Gaussian system, we must introduce the factor A on the left. By
these equations the solenoidal vector q
magnetic
The magnetic
H.
force
expressed as the curl of the

be derived from a
is
forces cannot
potential except where there is no current, but must be found by

In order to
integration of the partial differential equations (2).
show how
may always be
this
accomplished,
we
shall
prove a
general theorem.
Helmholtz's Theorem.
Vector Potentials.
223.
Any
uniform, continuous, vector point-function vanishing at infinity

may be expressed as the sum of a lamellar and a solenoidal part,
and the solenoidal part may be expressed

if
as the curl of a vector
vector point-function is completely determined

its divergence and curl are everywhere given.
point- function.
Let
R be
it
the given vector, with components X, Y, Z.

Let us
it
to
as
the
sum
of
the
vector
express
possible
parameter
suppose
of a scalar function $ and the curl of a vector-function Q, whose
Then
components are V, V, W.
_84>
~x
doc
dW dV
r
~~
tiz
dy
dV
--r -^
-- dU
~ 9<
^
7r~
oz
Finding
first
dx
dy
the divergence of R,
*
,,
div
dY dz =
R = dx
+
Arf>,
^- +
dx
for the curl of
any vector
is
dz
dy
35.
solenoidal,
But by 85 (18) we know that if $ and its

are everywhere finite and continuous, we have
Since
is
continuous by hypothesis, div

,
/,x
<3)
<
w.
E.
= -Ti
frf
0* ar
.
R is
azi
first
finite, so
derivatives
that
*
28
434
[PT.
III.
CH. XI.
Consequently the lamellar part of R is determined by its divergence.

Secondly finding the curl of R, say o>, with components f, 77, f,
e=_
9y
__-
=-_
8#
9?/
8#
V 8-z
(4)
3^7
Since
dv
as yet undetermined except
is
equations (i)
we may impose on
dz
dy
[ex
it
-.
dz \dy
dy )
by the
partial differential
the condition of being sole-
noidal,
^ + ^-
(5)
dx
dz
dy
Hence
,.
and
But
in like
since
-- dY
= dz
o
^-=9y
dz
dx
dz
manner
R is continuous,
curl
must be
finite,
and therefore as
before
The
vector Q, whose components are l/4?r times the potentials of

the components of o>, is derived from
77, f,
the scalar functions
may
by the operation
G)/4?r
Pot, considering
as a vector, so that
we
Q = ^Potft>,
(8)
and
call
4?r$ the Vector Potential of
the curl of Q,
potential belonging to R.
of
G>
write
is
part of
is
we
o>.
shall also
We
Since the solenoidal part

say that Q is the vector
accordingly see that the solenoidal
determined by curl R, and accordingly the vector
ELECTROMAGNETISM.
223, 224]
is
divergence and curl. This theorem

by Helmholtz in his celebrated paper on Vortex
uniquely determined by
was given
Motion *.
435
its
Symbolic Formulae. These relations may be conmeans of Hamilton's and Gibbs's symbols V
cisely expressed by
and Pot ( 78). In words we may say that any solenoidal vector
224.
the curl of the vector potential belonging to

vector potential of 1/4-ir times its curl.
is
By
which
it,
virtue of the definition of Hamilton's operator
is
the
we have the
vector equation
R=V
(9)
so that
we may
call
the
sum
of the scalar
and the vector Q
<
the quaternion potential belonging to R, from which

is derived
the
V.
vector
the
values
of <f>
by
single
Inserting
operation
andQ,
(
R = V j -L (- Pot div R + Pot curl 12)1
o)
so that the operator (Pot curl

Pot div)/4?r
when applied to a vector-function.
is
the inverse of V,
For a lamellar vector we have

curl
i)
and
R = 0, R = -
i V Pot div
4?r
JR,
for a solenoidal vector
(12)
div
R = 0, R = -1 V Pot curl R = ~
Taking the
curl of
CD,
curl
(R being
we
o>
find in like
= curl
curl Pot curl R.
manner
R= - AE,
solenoidal) so that
1
- Pot curl
R = 4?r
(13)
/,
R.
In fact since the operations of definite integration and partial

are commutative, the operations Pot and curl
differentiation
must
be.
*
Helmholtz. "Ueber Integrale der hydrodynamischen Gleichungen, welche
den Wirbelbewegungen entsprechen," Crelle's Journal, Bd. 55, 1858, p. 25. Wiss.
Abh. Bd. i. p. 101.
282
THE ELECTROMAGNETIC
436
FIELD.
[FT. III. CH.
XL
Magnetic Force from Current. Applying Helmtheorem to the solenoidal vector H, the magnetic force,
and calling the components of the vector potential F, G, H* and
222 (2), we
using the fundamental equations (7), together with
225.
holtz's
obtain
(is)
-///-:*
oo
or the vector potential belonging to the magnetic force

vector potential of the current density.
Energy
226.
of
Magnetic Field of Currents.
netic energy of the field
Wm =
(i\
\
'
*"*
is
by
118
)W
([[{L(
the
The mag-
(10),
and introducing the vector potential

(
is
this
becomes
_^? \+M(* -
+ N(
}\dr
oo
Integrating by parts for any volume r bounded by a closed

surface S,
*
It is to be noticed that the letter
is here unfortunately used for both the
resultant magnetic force and one component of the vector-potential. This is
because we have followed Maxwell in using the letters F, G, H. The ambiguity
need cause no confusion.
ELECTROMAGNETISM.
225, 226]
jj{(MH
437
- NO) cos (wa?) + (NF - LH) cos (ny)

+ (LG-MF)cos(nz)}dS
This important theorem in integration

(4)
fjj{H
The
space,
curl
may be
Q-Q curl H] dr =jj"V.HQ cos (n, V. HQ) dS.
integral representing the energy
is
and the surface integral vanishes
the value of curl
abbreviated as
extended over
at infinity.
H in terms of the current density,
infinite
Inserting
222
(2),
we
obtain
Wm = lfjf( Fu + 0* + Hw) dr,
(5)
and since no portion of space contributes to the integral unless

it is traversed by currents, we may take the
integral simply
The components of the

through conductors carrying currents.
vector potential are however themselves triple integrals over the
same portions of space, so that if we distinguish a second point
of integration by an accent, we have the double volume integral
r2
= (x - xj + (y- yj + (z- zj,
where each point of integration traverses

occupied by currents.
the whole volume
This form of the energy corresponds to the form in terms of

density given in 117 (5), the integrals being there taken through
all
distributions of matter.
If
we perform the volume
integration
by dividing the space
up into current tubes, of infinitesimal cross-section 8, ds being

the length of the generating curve, and / = qS the total current in
the tube,
integral
(7)
we have
becomes
for the
element of volume dr
= Sds,
so that the
438
[PT.
III.
CH. XI.
The sextuple inds' traversing all current tubes.

as
a
line
around
here
interpreted
integral
every current
tegral
tube and then an integration for the double infinity of tubes for
both ds and
is
each variable s and
If the currents
s'.
consist
of two
linear
circuits, or closed tubes of infinitesimal cross-section and strengths

/! and 72 the sextuple integral .reduces to a double-line-integral,
>
and since both variables
and
may divide the integral

coincide with Sj or s2
up
are to traverse both circuits,
we
into four parts according as s or s
(8)
The second and
third integrals are equal, for
it
is
evidently
a matter of indifference which point of integration is associated

with either circuit, so that we may write for the sum of these two
terms
T T [
C S (dSldS.2 )
'JiJs
where each point of integration goes once around one of the

circuits.
This term
energy
of the
is
equal to the negative of the mutual potential

electromagnetic forces acting between the two
currents, as found in
220
In like manner the
(i i).
and
first
last terms,
where each point of
integration goes once around the same circuit, are the negatives
respectively of the potential energy of either current in its own
from which the electromagnetic forces acting between

If we call the integrals
different parts may be calculated.
field,
JiJi
we may
cos (dsds)
(<focfe')
^
7
_
LTr =
2
J2J
cos (dsds')
f
say then that the magnetic energy of the
both currents
(9)
rt
^dsds,
Wm = \LJ?
its
field
due to
ELECTROMAGNETISM.
226]
439
But the potential

the negative of the total potential energy.
if the current
are constant,
and
to
tends
decrease,
strengths
energy
is
while the circuits are moved or deformed, their position and form
being specified by a certain number of geometrical parameters q 8
the forces
(10)
according to these parameters are given
P _i7
^
r
A,H<IA
77^2 ,172^
*'
dq.'
dq.
Sjf
The magnetic energy
we
by
=
2,P.fy.
of the field then tends to increase,

in the
find the system
same manner
and
as a cyclic system
behaving
The energy which must be
TO.
during an isocyclic motion,
caused by the electroa
motion
to
the
furnished
system during
of work done by
amount
magnetic forces must be double the
the electromagnetic forces, which is equal to the loss of potential
energy, and must be furnished by the impressed electromotive
We have already seen that
forces that maintain the currents.
in the case of concealed motions
we cannot always
tell
whether
and that in
cyclic systems the

has
of
a
function for either
the
force
properties
energy
We are therefore led naturally
isocyclic or adiabatic motions.
energy
is
potential
or kinetic,
kinetic
to consider a
system of currents as a cyclic system, and, instead
of considering
kinetic energy.
energy, and denote
as potential energy, to consider

shall henceforth call it the
We
it
by
Wm = W as
electrokinetic
T.
These considerations, assimilating an electrical system to a

mechanical system, are due principally to Maxwell, and by means
of them we shall in the next chapter be able to deduce the laws of
induction of currents.
we
If in the integral (5)
manner just explained,
for
integrate over current-tubes in the
udr we must put
q cos (qx) Sds

so that
(12)
we
= Idx
etc.,
obtain for each current
T=
(Fdx
+ Gdy + Hdz\
are
is around its own circuit, but F, G,
the definite integrals over all currents, as previously used. Applying Stokes's theorem to the above line-integral, we obtain
where the integral
THE ELECTROMAGNETIC
440
FIELD.
[PT. III. CH.
XL
dy
over any surface bounded by the current.
But by the equations
T= -
(14)
225 (14),
(t{L cos (nx)
+ M cos
(ny)
+ N cos (nz)}
dS,
or the electrokinetic energy of a system of currents is equal to onehalf the sum of the strengths of each current multiplied by the total
magnetic force through its own circuit in the positive direcThe part of the flux due to the current itself constitutes the
flux of
tion.
term ^Z/ 2 while
for any two currents 1 and 2, the portions conof

one
-half
the strength of either times the flux through
sisting
its circuit due to the other current, being equal to the two
,
middle terms of
mutual
We
are equal.
may consequently express the
kinetic energy of two currents as the strength of either
(8),
multiplied by the flux through
its circuit
of the magnetic force due
to the other.
227.
Mechanical Forces.
We may deduce
the mechanical
forces acting on conductors carrying currents from the expressions

found in 219 (7). Calling the forces per unit of volume H, H, Z,
and writing
we have
for
Idx the value
in terms of the current density udr,
')
drdr'
-/f//4-"
The
other.
first
terms in the
first
and second integrals destroy each
The second terms may be written
respectively, since the
accented quantities are independent of the unaccented,
fff.itffl^fc'U-fff.g*,
a# IJJJ r
JJJ
JJJ dx
j
and
,
,
(/Y/V
,)
- dr
dr
5~\
[
r
fa
V
dy (JJJ
=-
r/Y dF , T
v oT ^
JJJ dy
'
226
ELECTROMAGNETISM.
228]
441
Consequently we get
dF
and we obtain
dH
mechanical forces on the conductor per unit
for the
volume,
3 = vN wM,
R=wL-uN,
Z = uM- vL,
(3)
The mechanical
force per unit volume is the vector product of the

current density and the magnetic field.
228. Effect of Heterogeneous Medium.

Let us consider
what changes are necessitated in our equations by the presence of
magnetizable bodies, so that the magnetic inductivity is not constant throughout space.
In the reasoning of
210 we supposed
the magnetic force to be both lamellar and solenoidal in all space
//,
not traversed by currents.

As soon as we have variations in the
the
force
in
is
inductivity,
general no longer solenoidal, but the induction is.
cannot, however, apply the reasoning unchanged to
the induction, for this, in general, is not lamellar. The reasoning
We
connecting the current strength with the work of carrying a pole

around a closed circuit is however unchanged, and if the circuit lie
in any other medium than air, the work is the same as if the circuit
lay in air, namely zero if the circuit is not linked with the current,
4s7rnl if linked
n times
positively.*
For consider a
circuit
composed
of two infinitely near circuits each embracing the current once,

corresponding points of the two lying infinitely near each other on
opposite sides of a surface separating air from another medium.

Then if we carry a pole around the circuit in air in one direction,
and back around the
circuit in the other
medium
in the opposite
direction, since the double circuit is not linked with the current
work has been done.

circuit in
much
as
For otherwise,
in going
no
around the double
one direction or the other, we might store up energy, as

But this would
pleased, by repeating the operation.
we
be in opposition to the principle of conservation of energy, which

says that the energy is definitely determined when the positions
and strengths of poles and currents are given.
Consequently our
442
[PT. III. CH. XI.
222 (2) remain unaltered. When we

electromagnetic equations
consider the energy and the mechanical forces, however, we have
changes. The potential due to a current is no longer proportional
to the solid angle subtended by it, and accordingly we can no
longer deduce the forces as simple line-integrals.

write for the energy of the field, by
180,
We
must now
so that if a given current is placed in an infinite homogeneous

medium, since the distribution of the force is independent of the
medium, as long as
it is
homogeneous, the induction, and therefore
the energy, are directly proportional to the inductivity.

Contrast
this behaviour of a current with that of a permanent magnet, which
homogeneous media always emits the same total flux of

and therefore the energy are inversely
The flux of force emitted by the
to
the
inductivity.
proportional
in different
induction, while the force
conductor carrying current
is
constant.
Since the magnetic force is no longer solenoidal, it can no

The inlonger be represented as the curl of a vector potential.
duction, on the contrary, can be so represented,
potential belongs to the magnetic induction.
and the vector
= a#_d
o =
dz'
J\-
rs
ox
_
*
dy
On
account of this change it is no longer possible to integrate the

222 (2) in the same simple manner as in
225, for
equations
while the current

induction that
is
is
the curl of the magnetic force,
the curl of the vector potential.
curl of the induction,
_
dy
or using
222
dz
dy
dz
(2),
JF^
dy
it
is
the
Taking the
ELECTROMAGNETISM.
228]
443
so that the vector potential is not related simply to the current,

but the magnetic forces still occur in the differential equation.
We have the same difficulty as occurs when we undertake to find
the potential of the field when the inductivity varies. As we

there made use of an apparent density, so here we may define an
apparent current as
dz
ty
so that the vector potentials are
-//* *-//*
'-//*
If each magnetizable
body
a/A
is
homogeneous,
= 8//:= a^ =
da)
dz
dy
and the apparent currents are the true currents multiplied by the
inductivity, except at the surface separating two media, where the
derivatives of p,
infinite.
We
and consequently the values of &F,
have the above form
and when they are
for F, G,
AGr, AZT, are
H only when A.F,
etc.
a surface we must proceed

finite,
as in the case of a surface distribution of matter, that is we must
Conconsider an apparent current-sheet between the two media.
are
infinite at
sidering two surfaces infinitely near each other and situated on

opposite sides of a surface of discontinuity of /u, at a distance dn
from each other, and integrating the equation (3) over the volume
between them we obtain*
of the thin sheet
(4)
jjj \vy
v*
j-^-
{(W
cos (ny)
$1) cos
(ny)
dndS
cos (nz)
j-
D^'
9JI)
cos (nz)}
dS
*
The second integrand in (4) is equal to the first since n is the direction of
most rapid (infinitely rapid) change in the functions 31, 581, in the infinitely thin
sheet.
THE ELECTROMAGNETIC
444
FIELD.
[PT.
III.
CH.
XL
S
)JT, 91', being the components of the induction on the side
toward which n is drawn,
$31, $1, on the side from which it is
',
We
drawn.
must now,
as
F the
integral already found for
which
is
(W -
R) cos (ny)
shown
85 (18), add to the volume
in
surface integral
the effect of an apparent current whose ^-component per

unit of surface is 1/4-7T times
- (9R' - 9R) cos (nz)

= 53' (cos (23 Y) cos (ny) - cos (93'y) cos (nz)}
23 {cos ($$z) cos
(ny)
cos (53y) cos (nz)}.
Now
the normal component of the induction is continuous, its

tangential component being discontinuous, while the tangential
component of the
force is continuous.
to the line of force is the
same
The normal plane tangent

and the amount of
in both media,
the discontinuity in the tangential component of the induction

33'
sin S3'w
-33
is
sin
Referring now to the definition of a vector product, we see that

the first parenthesis above is the ^-component of the vector product
of the induction and a unit vector in the direction of the normal,
which vector product has the magnitude
S3'
sin
(93'ra).
The apparent
accordingly in the surface, perpendicular to the normal

plane containing the line of force where it crosses the surface, and
its magnitude per unit of surface is 1/4-Tr times the discontinuity
current
is
in the tangential induction.
If the lines of force are normal to
the surface, the apparent surface current vanishes*. If, however,

there is a surface carrying a true current-sheet, by the same
222 (2), we find a discontinuity
reasoning, applied to equations
in the
component of the
dicular to the current of
force
tangent to the surface and perpen4?r times the current density.
amount
Mutual Energy of Magnets and Currents. If we

229.
have permanent magnets and currents situated in a homogeneous
medium of unit inductivity, we may represent their mutual energy
in
two ways.
We may
in the first place consider the
magnets to
*
This apparent current- sheet was overlooked by Maxwell, and it was not until
the appearance of the Third Edition of his Treatise that the correction was made
by
J. J.
Thomson.
ELECTROMAGNETISM.
228, 229]
445
be traversed by apparent currents and current-sheets, as in the
We there
preceding section in the case of temporary magnets.
introduced the discontinuity in the induction, but we might have
In the case of the
introduced the intensity of magnetization.
permanent magnet
be more convenient
this will
in either case
the form of the vector potential will be the same.
We
have
medium
for the potential
of unit inductivity,
where A',
B', C' are
due to a magnet
122 (3),
the values at
a, 6,
in a
homogeneous
c,
dr' = dadbdc,
r2
The
field at
= (x - of + (y
any point
x, y,
b)-
+ (z
c)
z has the component
Now the derivatives of I/r with respect to a, b, c are the negatives

of its derivatives with respect to x, y, z, so that we may write
L" =
(,)
(3)
But
JJJ
since 1/r
is
-D
harmonic we
dxdy
may put
and thus the integral becomes
L=
ox
(4)
dy /
dz
THE ELECTROMAGNETIC
446
But
since
/*
= 1 we
FIELD.
[PT. III. CH. XI.
have
dz'
dy
and accordingly the components of the vector potential may be

taken as
(5)
da
Every element of volume produces at
sc,
y,
z a portion of vector
2
potential equal to 1/r times the vector product of its magnetization by its vector distance from the point x, y, z.
The mutual
energy of currents and magnets is then obtained by the equation

226 (5), omitting the factor -J-. This method of treatment is
that of Maxwell*.
From the above form
for the vector potentials
express the solenoidal vector F, G,

vector potential.
derivatives
so that if
by
H as
itself
we may
easily
the curl of another
For again replacing derivatives of 1/r by
a, b, c
by
x, y, z,
we introduce the vector
potential of magnetization, with
components
>.
*-///>.
*
Treatise, Vol. n., Art. 405.
ELECTROMAGNETISM.
229]
447
the vector potential belonging to the magnetic force
is its curl.
aR_3Q
dz'
Sy
This leads us to a second manner of obtaining the mutual energy,

The expression for the energy of a permanent
due to Helmholtz.
magnet
in a magnetic field L,
TF=
(9)
M,
126
is,
(2),
-fjj(AL+BM+ CN) dr,
where the volume of integration is that occupied by magnets, or

it may be extended to infinity, since elsewhere
^==(7=0.
We may
transform the integral into one taken throughout the

If we introduce the vector potential
space occupied by currents.
of intensity of magnetization, we have from (7), if the magnetization
is
everywhere
finite,
A=-
^ AP,
4-7T
B = -~^Q
(10)
A.
C = -.-.
4-7T
Introducing these values of A, B,

(11)
W=
(^ AP +
into the integral (9),
M AQ + lYAE) dr,
00
and transforming each term by Green's theorem
in its second form,
the surface integrals vanishing at infinity,

(12)
F=
Let us now substitute
for L,
M,
N their values in terms of the
vector potential belonging to them, noticing that, since the differ-
448
T^HE
ential operator
we may
is
ELECTROMAGNETIC FIELD.
commutative with any
dG
_-_
\dy
if
partial differentiation,
write
= Ax ftH
But
[PT. III. CH. XI.
dzj
the vector potentials F, G,
dz
dy
H are those of the currents
=-
u, v,
w,
4>7TU,
so that finally
(14)
w
W=
which by the theorem of
226
(3), (4) is
equal to
Tf=
(15)
226 (5) we find a difference in sign,

Comparing with
being
mutual potential energy,
m electrokinetic energy, while the
factor
is omitted in mutual energy.
integral may now be restricted to the space occupied

The form involving the curl of P, Q,
currents.
is that used
The
Helmholtz*, who writes L, M,

P, Q,
R by
their values (7)
instead of P, Q, R.
dy
which differs from the result of substituting

the same way as (15), above.
have thus seen how we
may
(5) in
replace every
.apparent current
*
Replacing
we obtain the double volume-integral
+ (uC'-wA')
We
by
by
Helmholtz, Ges. Abh. Bd.
i.
p. 619.
226 (5) in
magnet by an
ELECTROMAGNETISM.
229, 230]
=-
i
/dR a
AJP=--r- A --- .
4t7r
4-7T
1
= --r1= --A(r
4?r
4?r
was due
Ampere was
dz
\dy
fdP
-- 3R
AU
y9s
9#
.
which would produce the same magnetic

reason that
449
effect.
It
was
led to the hypothesis that all
to currents of electricity circulating
for this
magnetism
about the molecules
of matter.
The above formulae all refer to currents and magnets placed in

a homogeneous medium, and as has been already seen, lose all
their simplicity when the inductivity varies.
For although we
still
the
to the induced
calculate
vector
due
may
potentials
magnetization, the process will be
complicated, and in
general
For this reason, and because both scalar and
impracticable.
vector potentials are quantities whose physical significance is
much less apparent than that of the strength of the field,
Heaviside and Hertz have been led to avoid the employment of

potentials, and to deal directly with the electrical and magnetic
We have however introduced the vector-potentials here

on account of their important mathematical relations, and the
fields.
fact that
they have been so
much used by
the highest authorities.
Magnetic Field due to Current- Sheet.
230.
We
have
found that in a current-sheet the amount of electricity that flows

in unit time across a curve connecting any two points in the sheet
equal to the difference of the' current-function M/" at those two
This quantity is the same whatever the curve connecting
points.
is
We
shall supis an electrode

lying between.
that
a
flows in
current
sheet
has
no
the
so
that
electrodes,
pose
closed circuits in the sheet.
may find the magnetic field of
them, unless there
We
such a sheet, at points not lying in the sheet, by the consideration

that the strip of the sheet bounded by the curves "*& = const, and
"fy
+ dW = const., dW
being a constant difference in the values of
the current-function for the two curves, is equivalent to a linear

current of strength dW.
Such a current, by 210, is equivalent
to a magnetic shell of strength dV.
The whole current sheet
therefore be replaced by an infinite series of magnetic shells,
whose edges only are given, the form of the shells being in-
may
w.
E.
29
450
CH. XI.
[PT. III.
them.
These
be considered to form a continuous body, which is,
being divided into shells, lamellarly magnetized, the potential of
different, so long as the attracted point lies outside
shells
may
magnetization
<f>
being equal to the current-function
The magnetic
potential
is
124).
accordingly at outside points,
by
124(11),
All}
rj ,~
But
since the form of the magnetic shells
as their edges are of the given shape,
is indifferent,
we may
as long
them
consider
all
deformed so as to coincide with

the current-sheet, as
in
Fig.
88.
The
is
illustrated
shells
overlap
each other continuously, so that

there are more shells laid on the
sheet the greater the values of M*.
cross the sheet, the poten-
FIG. 88.
tial
As
is
As we
discontinuous, as in the case of a single magnetic shell.

component of the magnetic force,
in that case also, the normal
being continuous for all the shells, is continuous on crossing

the sheet. The tangential component in the direction of the
lines of flow is also continuous, but, as we found at the end of 228,
the component" perpendicular to them experiences a discontinuity
equal to 4?r times the current-density, that is 4nr<y9/dn*. This may
also
be very simply obtained by taking the line-integral of magnetic

around any circuit composed of two infinitely near portions
force
lying on opposite sides of the current-sheet and coinciding with
an
electrical equipotential line, the integral being equal to 4?r

times the difference in the values of the current-function at the
two points where the

231.
Examples.
circuit cuts the sheet.
Coefficients of Induction.
Toroidal
straight coils. We shall now calculate the energy due to

currents in a few simple cases. The coefficients of the half-squares
and
and products of the current-strengths in the expression
for
the
electrokinetic energy, are called, for reasons to be explained in the

next chapter, coefficients of induction, or more briefly, inductances ,
distinguishing coefficients of half-squares by the name self-inductance, coefficients of products by the name mutual inductance.
ELECTROMAGNETISM.
230, 231]
Any
451
mutual inductance is the magnetic flux through a

due to unit current in its own or another circuit respec-
self or
circuit
tively.
We
shall first consider a solid
surface generated by revolving

axis in its plane not cutting it.
If the tore be uniformly
bounded by a
of revolution
any closed plane curve about an

Such a solid may be called a tore.
wound with wire carrying a
current, so that
every winding
very nearly in a
plane passing through the axis of
lies
we may very
approximately consider the layer
of wire as a current sheet, the
difference of value of the current
revolution, Fig. 89,
IIG. 89.
function between any two points

being ml, where / is the current in the wire, and
m is
the
number
of turns of wire between the points.

By reason of symmetry the
lines of magnetic force must be circles whose planes are perpendicular to the axis of revolution,
and whose centers
lie
on the
axis.
is a function
Consequently the strength of the field
only of
the distance p from the axis, and the line integral of the fieldstrength around any line of force is equal to the constant value
of
H on
the total
that line times the circumference of the
number
of turns of wire on the tore,
the substance of the tore
same
is
circle.
any
If
n be
circle lying in
linked with the current n times in the
direction, so that the value of the
above line-integral
is
4wrl
(i)
This gives as the value of the force for internal points
circle lying outside the tore, however, is not linked at all with
the current, so that the line integral is zero, and therefore the
force
must be zero. Such a closed coil or toroidal current-sheet
accordingly emits no tubes of force, but all its tubes lie within
The force accordingly
the doubly-connected space of the tore.
has a discontinuity at the sheet equal to 4?r times dV/dn, which
is the amount of current
crossing unit of length of a circle coIf the tore be filled
inciding with a line of force, or nl/27rp.
a homogeneous medium of magnetic inductivity /x the
= Zn/jul/p. This whole reasoning
induction at any point will be
with
pH
292
452
[PT. III. CH. XI.
has supposed that the external medium is homogeneous, but since

there is no field there, the value of the inductivity is immaterial.
\z
p be rectangular coordinates parallel and perpendicular to
the axis in the plane of any orthogonal cross-section of the tore,
the whole flux of induction through the section is
If
(3)
ffpHdS
JJ
= 2^7
= 2^7
ff*&
JJ
Jflog (&
\pi
P
P! and /3 2 being the least and greatest values of p on the contour

of the section for a given value of z, and being given as functions
of z by the equation of the contour.
If another circuit be
wound
in
any manner about the
tore,
the flux through
it is n' times that

embracing
just
and
the
mutual
electrokinetic
found,
energy of a current /2 in it
and a current /j in the former winding is, according to the last
n' times,
it
sentence of
226,
T-
(4)
The mutual inductance
of the circuits
is
accordingly
Mu
(5)
If the second coil coincide with the
first
the flux through itself
is
so that the self-inductance of the toroidal coil is
L=
(6)
The
electrokinetic energy
'//
For a
radius
coil of
is
is
square cross-section whose side
R,
ft
(8)
_I_
L.
Cb
is
2a and whose mean
ELECTROMAGNETISM.
231]
453
For a circular cross-section of radius a
L=
(9)
L-
og
to*n
dz
= 47m> (R - Jit - a
2
2
).
If in equation (2) we insert the number of turns of wire per unit

of length of the line of force, m, since n = 27r/?ra,
H=4}7rml,
(10)
or the force depends only on the amount of current per unit of

In case the radius of the tore is increased indefinitely,
length.
is the number of
we get an infinitely long straight coil,
turns per unit of length of the coil, and we have within a uniform
If any coil of n' turns be wound
field of the magnitude 4t7rmL
so that
on outside, the mutual inductance
will
be
It is noticeable in all these cases that it is of
no importance whether
contact with the inner or not, for in any case

If there were any field
it is threaded by the whole flux of force.
external to the tore, the case would be different. It is however
the outer
coil is in
necessary that the tore be entirely filled by the medium of inducThe formulae of this section are applicable to induction
tivity p.
and transformers, providing the coils are endless. The lineintegral of magnetic force 4?r/i/ is called the magnetomotive force,
and the problem of finding the magnetic induction in the tore is
the same as that of finding the current in a tore of conductivity
in which there is an impressed electromotive force of the
amount 47m/, the lines of flow being circles. In case the cross
coils
/JL
section of the tore
is
the curvature of the
T
1 1
we may
184),
neglect
by
174,
we have
so that
(
small compared to its radius,

and find the reluctance (
coil,
.-,,
Induction Flux
& _,
= Magnetomotive
.
force
-=
4>7rnl
7
Reluctance
pS
This formula
used in practice in finding the flux in the field

of
a
magnet
dynamo-electric machine, although it is accurate only
in the case that we have treated, where all the tubes of force are
is
the current turns, so that the numerator is the

Any tube being partly in iron and partly
the reluctance of any infinitesimal tube is found by the
encircled
by
same
every tube.
for
in air,
formula
for
all
the resistance of conductors in series, as
454
[PT.
III.
CH.
XL
In order to find the influence of the ends of a uniform straight

any cross-section, we may consider that each current turn
replaced by a plane shell, so that the whole current sheet is
coil of
is
replaced by a uniformly magnetized cylindrical magnet with

= ml. The free surface charges
intensity of magnetization d<f>/dz
of all the shells accordingly cancel each other except for the two
These ends are single distributions

plane ends of the magnet.
identical with each other except for the difference of sign.
If Vi
at
to
is the potential
a uniform single distribution
any point due
of unit density on the positive end 1, and

2 that
identical distribution on the negative end 2, then at
due to an
any point
outside the region bounded by the cylindrical current sheet and
its plane ends, the potential due to the sheet is
n = m/(F -F
(12)
2 ).
We may
find the potential at a point inside the space in question

that for an infinite cylindrical sheet the force is
the
result
by
if z is measured parallel to the generators of the
so
that
47rra7,
cylinder in the direction of the force,
O=
(13)
If IV
is
4>7rmlz (for the infinite cylinder).
the potential due to
portion which
we
of the infinite coil except the
all
are considering,
we have accordingly
H -f ft' = -
(14)
But the space in question is outside the two magnets replacing

the two infinite parts of the sheet, so that for a point between
the ends,
IT =
mI(V
-V
1 ),
giving
n=
(15)
Now
we
is
pass one of the ends of the coil the potential
its
but
a
of
the
continuous, being
single distribution,
potential
derivative has a discontinuity of amount 4?r by
82, accordingly
as
continuous, the
discontinuity in changing from the formula (12) to (15) just canIn the case of a circular
celling the discontinuity in dV/dz.
and
2 may be found by the
l
coil, the potentials
the potential fl
cylindrical
development
is
discontinuous, but the force
is
in spherical harmonics given in
102,
and the devia-
ELECTROMAGNETISM.
231, 232]
455
tion from uniformity of the field at any part of the solenoid

In a long solenoid the field is very nearly uniform
calculated.
for
a considerable distance from the middle of
its
length.
By
2 with respect to r,
differentiating the expressions for Vl and
the distance from the center of either, multiplying by the element
of the area of a sphere of radius r, and integrating, we may find
the flux due to either end through a circle perpendicular to and

with center in the axis, and hence the correction due to the end
to be
made
mutual inductance of the coil with another

and thence by another integration with
in the
circuit of a single turn,
respect to any concentric
coil.
Pair of Rectangular Circuits. In the case of two

we may use Neumann's formula for the mutual
232.
linear circuits,
inductance
M=
,
cos (dsi ds^) ds 1 ds2

/Y -
JJ
in those cases
which are simple enough

two circuits are
us to effect the
for
If the
integration.
equal rectangles
ABGD and
A'B'C'D',
Fig. 90, of length ^ and breadth 1 2

with corresponding sides parallel, and
the lines joining corresponding corners perpendicular to their planes and

of length a, then for pairs of sides
which are perpendicular the integral
2'
while for pairs of parallel

sides the cosine is either plus or minus unity, according as we
consider corresponding or opposite sides in the two rectangles.
vanishes,
For the
sides
AB, A'B' we have

dzdz'
TD
I
Ivj
Va +
2
(*i
A/~iTT
The integration of the logarithms in the second integrand may

be performed by taking as a new variable the quantity whose
logarithm is to be integrated, and then integrating by parts, the
result being
MAB A>v = 2
,
a-
V^TT? +
1,
log
456
[PT. III.
CH. XI.
For the pair of sides AB, C'D', substituting the square of their
distance apart, a 2 + If, for a2 and changing the sign, we have
MAB,
C'D'
=~2
Wa + - Va + +
2
Z2
lt
Z2
J,
log
The
v Co
t'o
BC, B'C', and 5(7, D'A', are obinto 1 2

We have then considered
portions for the pairs of sides
tained from these by changing
two
just half of the
lt
circuits, so that,
adding these four parts and

the
we
obtain
value
of the inductance
two,
multiplying by
M = 8 {a - Va
+ If - Va + If + Va + 1* +
2
log
attraction of the
by unit current
233.
of radii
is
+ v a + If
lt
2
}
v a2 + ^ 2 [
two
each other when traversed
circuits for
obtained by differentiating this expression by
Pair of Parallel Circles.
R R
/2
Va'
ft
The
their
a.
If the circuits are circles
planes being perpendicular to the line

joining their centers, of length a, we
may put
[p^
[^
cos
= RI cos fa,
a^
2/2
zl
z.2
=0,
2
Ho cos
<!>!
<
2)
+ (^tj sin fa
cos (fa
V ot
~H -Ri
2
~l~
Rf
7T,
d (fa
fa)
= 2cfyr,
COS (0!
the integral becomes

.
cos
(rf>!
^>o)
amounts merely
integration with respect to fa

If we put
cation by 2?r.
2\/r
K sin
<f>A
l2jLt 1 jfi 2
The
fa=
= -^2 sin fa)

= a,
cos (6,
M=
fa
cos fa,
&2= R%
== -î sift
y\
fa)
p(
/
V1
-K
fa)
to multipli-
COS
2\/r,
ELECTROMAGNETISM.
232234]
457
and writing
1-2 sin
we have
2
<\|r
= -2 (1 -
2
/c
sin 2 ^) 4- 1
finally
M=
2
47T
-^ ()+(-?
K
[K
where
- -2
^ and F are the elliptic integrals

o
Vl-K
sin 2 A/r
These definite integrals are functions only of the parameter K,

and their values have been tabulated by Legendre for various
values of
9'j
If
K.
we put
and r2 are the
maximum and minimum
on the circumferences of the two
circles
distances
of points
from each other.
The
^R^
expression M/4>ir
being a function only of K and therefore
of 7, has been tabulated by Maxwell as a function of 7. (Treatise,
Vol. 2, Art. 701.)
We may
also find the value of
M in a series of zonal spherical
harmonics by means of the series of 215 by differentiation with

respect to r and integration over a spherical segment bounded by
the second
circle.
circular coils
For a
the reader
full
is
treatment of the properties of

to Maxwell's Treatise, to
referred
Mascart and Joubert, Lessons on Electricity and Magnetism, and

to Gray, Absolute
Measurements
where a great variety of formulae

234.
in Electricity
will
and Magnetism,
be found.
Non-linear Currents in Parallel Cylinders.
If
the expressions in the two preceding sections be used to find

the self-inductance of a linear circuit we find a difficulty, for on
= in 232, the expression becomes logarithmically

putting a
while
on putting a = 0, l = J?2 in 233, K becomes unity,
infinite,
ibe elliptic integrals reduce to trigonometric, and
(K) becomes
logarithmically infinite (log tan
TT/ 2).
This
is
easily seen to be
the case for any linear circuit, for if dsi and dsz traverse the
same circuit there is an infinite element in the integrand, and,
458
considering the element in which
it
[PT.
III.
CH. XI.
occurs as straight the integral
fds
J
becomes logarithmically infinite. We may see the reason for the

self-inductance becoming infinite in another way by considering
Biot and Savart's Law, for as we approach a linear conductor
the force
is
conductor.
this
inversely proportional to the distance from the

flux therefore increases like the logarithm of
The
distance,
and
when we approach
not finite
is
conductor indefinitely.
sidering conductors of
the linear
We
may avoid this difficulty by confinite cross-section, for in that case the
corresponding element of the integral, in which the integrand
becomes
infinite,
potential,
- is
not
infinite, as
was proved
for
an ordinary
76.
We shall now consider currents flowing in three-dimensional

conductors in the form of cylinders of infinite length whose
generators are all parallel. We might treat the problem by the
application of the law of Biot and Savart to each infinitesimal
tube of flow, but we shall prefer to make use of the general
228 (2). It is evident that the lines
222 (2), and
equations
of force are in planes perpendicular to the conducting cylinders,

and the field
which we shall take for the JTF-plane, so that
N=Q
is
independent of the coordinate
z.
The problem
is
accordingly
a two-dimensional problem, and all the quantities concerned are

= = we have F = G
so that our
independent of z. Since u v
equations are
dM
-4f7TW=^dx
(l)
dL
5,
dy
(3)
from which
results, if
fju
is
constant,
&H
(4)
But
-**i=w
this is Poisson's equation for the logarithmic potential,
91 (10),
459
ELECTROMAGNETISM.
234]
so that its integral is
H=C-Z
(5)
r2
where
= (x - a) + (y 2
6)
a constant, which, though infinite, does not affect the

force.
the
value of
and
is
If the conductors are concentric circular cylindrical tubes and

the current-density is uniform, we may find the magnetic force
without finding the vector-potential, in the same way as in 231,

for it is evident that the lines of force are all circles in planes
perpendicular to the conductors. At points outside the outer
tube, at a distance p from the axis, the line integral of magnetic
force (which
we
will
denote by
instead of
H,
fusion with the vector-potential) around a circle

27T/3P
(6,
to prevent con-
is
= 47T/,
..P-f,
where
(7)
is
the total current through
all
the conductors.
Accordingly at
external points the field is the same as if the current were concentrated in the axis of the conductor.
If different tubes are
made
part of the same circuit, so that all the current flowing

in one direction is returned in the other direction by concentric
conductors, the total current
is
equal to zero, and the force
is
zero at
Such a
doujble tubular conductor accordingly,

like a toroidal coil, emits no tubes of
For
magnetic induction.
all
external points.
when it is wished to protect delicate magnetic instruments from the action of strong currents, the circuit should be
formed of concentric conductors. The mutual inductance of any
this reason,
external circuit with such a concentric conductor
is accordingly
next chapter, no currents
would be induced in the concentric conductors by external cur-
zero, so that, as
rents.
we
shall see in the
Such a conductor would thus be
suitable for telephone
circuits.
In the space outside the conductors the magnetic potential

(8)
fl
= 2/0 = 27 tan-
2,
is
460
[PT.
III.
CH. XI.
which we know to be a harmonic function, as

is conjugate to
the function log p, both being derived from the function log (x + iy).
In the substance of the conductors, there is no magnetic potential.
<
We may find
the force by evaluating the expression for the vectoror

as
above, except that now the line of force does not
potential,
surround the whole current, but only a portion of it.
If the
conductor is a solid cylinder of radius R,
%7TpP
4t7T
7Tp~W
= 4-7T/
-^
P-%P.
(9)
The integral (5) represents

only when //, has the same constant
value everywhere, for if it has discontinuities we must add a part
corresponding to the apparent current as shown in 228. In the
case just treated, however, the apparent current vanishes, for
the induction is tangent to the surfaces of the conductors.
In the general
(2)
and
(3)
p is constant in the space outside of

a magnetic potential, and the equations
case, if
the conductors there
is
become
"
dx
dy
showing that the function yu,fl is conjugate to the vector-potential

H, which is accordingly the flux-function for the magnetic induction.
The method
of functions of a complex variable
is
accordingly
applicable to problems connected with the field of cylindrical

conductors.
For instance, Fig. 65 represents for external points
the lines of force and equipotential lines of the field due to two
circular cylinders carrying equal currents in opposite directions.
No one of the circles in the figure however represents either of
The surface
the conductors, whose centers are at the points + a.
of a cylindrical conductor is tangent to lines of force only when
it is alone in the field, or accompanied by concentric conductors.
Within conductors, although there
equations (2) and (3) show that
is
is
no
still
magnetic
potential,
the flux-function for
the induction.
If
is
the area of the cross-section of any conductor, the
vector-potential at any point, whether external or internal,
is
by
ELECTROMAGNETISM.
234]
(5) equal to
where r
is
C-
ZfjiwS
461
\ogr=C- 2pl log r,
defined by the equation
S\ogr=
(u)
II
\ogrdadb.
But from the interpretation of a definite integral as a mean, 23,

we see that logr is the arithmetical mean of the logarithms of
the points of the cross-section from the fixed
defining the geometric mean of n quantities as
point x, y.
the nth root of their product, we see that r is the geometric
the distances of
all
Now
mean
of the distances of the points of the area from the point

x, y, for its logarithm is the arithmetical mean of their logarithms.
If T! and r.2 be the geometric mean distances of a point from two
areas 8 and $2 rs the geometric mean distance of the point from

both areas taken together, we have by the definition, (i i),
l
(12)
(8,
+ S ) log r, = 8, log r, + S
3
log r2
of this principle we may find the geometric mean

distance from a complex figure if we know it for the various
This method is due to Maxwell*.
shall
parts of the figure.
By means
We
find the geometrical mean distance from a circular ring of

infinitesimal width.
Let p be the radius, e the width of the ring,
and h the distance of the given point from its center.
Inserting
first
polar coordinates in the equation
(13)
2-7T/06
log r
1 1
),
ft*
2
\ log (h
Jo
+ p*-
2hp cos
This integral assumes different 'forms according as h is greater or

less than p.
Taking out from the parenthesis the square of the
greater of these, and integrating,
we get
h> P>
or
\
(15)
where
J is
og r
= logp +
-^j(^),
>h
>
the definite integral
J (a) =
f27T
+ a - 2a cos 0) d6,
2
Jo
log (1
Trans. Roy. Soc. Edinburgh,
18712.
462
which
is
a function only of the parameter

h > p and a = h/p when h < p.
= p/h when
that J(a) =
a
<
if
and we are
We
to
put
can easily show
For
1.
J(CL)
a,
[PT. III. CH. XI.
Jj (a)
I
+J
g(l
(a),
-f
a2
where
2a cos
cfa>,
</>)
<72 (a)
rfcr
Substituting
log (1
J, ( a )
The
in
/!
= Hog (1 +
Jo
a2
2a cos
J(a)
Now
integral being now.
we may add the
2</>
d<.
</>')
d$,
we may drop
is indifferent,
between the same
limits as
integrands, giving
= Tlog
J o
substituting
(/>)
gives
and since the variable of integration

the accent.
2a cos
JT
= TT + <'
</>
+a 2
J( a ) = I^ r\ g
Jo
{1
<'
a4
- 2a
2
(2 cos
1))
d(f>.
we obtain
+a 4
(1
f df = i J(a
2a 2 cos
2
).
Repeating the process we get
= i/(^) =
j
and letting
?i
j (>
increase indefinitely
But
= 0.
J"(0)
we
...
= i/co.
obtain, if a
<
1,
/() =
if
We
accordingly obtain from (14)

and (15) the result that the geometric mean distance from a
circular line is, for an outside point, its distance from the center,
/(0)
and
is finite.
an inside point, the radius of the
for
this result
we may
find the
mean
circle.
By means
of
distance from the area of a
For
ring of finite width, of internal radius E^ and external R%.
a point outside the ring the mean distance is its distance from
For a point in the space within the ring, by ( 1 1 )
the center.
or (12),
TT
2
(E 2 - Rf)
log r
r^ 2
27r
\ogp.pdp
JR,
= tt7r{R* (log
(
16)
log f
2
2
- 1) - R? (log R* - 1)},
ELECTROMAGNETISM.
234]
463
For a point in the area of the ring itself, we must divide the ring
into two, one within and one without the given point, so that
TT
(R? - Rfl \ogr
= 7r (If - Rf) log h

- h? log A - i (^ + TT
log R.
2
logr
(17)
The
l(
and since
/i
)},
is
vector-potential
always, for uniform flow,
this is the flux-function for the induction,
by reason of
the equation
,
we obtain the induction perpendicular

tiating according to
h,
which agrees with the result
The
226
electrokinetic
T= -
we
differen-
(9), in
which R^
is
equal to zero.
is,
by
(5),
H from
Cwdadbdz ~
r*
If
by
energy of the system of currents
and inserting the value of

(19)
to the radius,
so that
(5),
Mm fiww' log rdadbda'db'dz,
= (a- a')
'+ (b
integrate with respect to z from
but
bj.
oo
to oo
we
obtain an
length the energy

is
proportional to I, so that the energy per unit of length of the
conductors T/l is given by the above expressions omitting the
Each point of integration a, b
integration with respect to z.
infinite result for the energy,
and
a', b' is
for
finite
to traverse the cross-sections of all the conductors.
The
integral, containing the constant C, disappears, since to every

current there is a return current, in each of which the same value
first
of
C appears,
while for the two cross-sections the integral
wdadb,
464
[PT. III. CH. XI.
representing the total current in the positive direction, is zero.

If the flow is uniform in all the conductors, for any two conductors
there will be a term
- JU/1/, fog fM
where
log r l2
=jjjjlogrdadbda'db'
and r12 is the geometric mean distance of all pairs of points in

For instance let us consider a single
the two cross-sections.
circuit made up of conductors whose cross-section is denoted by
$! and conductors carrying the return current whose cross-section
is $2-
We
then have
T= 1
w'H'da'db'
Si
while
if
we
and H<2 due
divide
w'H'da'db',
i||
>S>2
into two parts,
jETj
H^
= (7
(?!
to the conductor 1,
/zw log rdadb,
/zw log rdadb,
T becomes
the
(20)
T=^(jw'da'db' -jjjj^ww'
sum
of the integrals
+ -&H w'da'db'
+^
2
(Iw'da'db'
JJ
11
IJLWW' log
r dadbda'db'
Sz
f
((f[t*0w log r dadbda'db'
JJJJ
(S'j
w'da'db'
log r dadbda'db'
s,
II
î
Si
+-
S2
The
due
to the conductor 2,
Si
Illl
pww'
log r dadbda'db'.
and third integrals, being the constant Ci/2 multiplied by

the direct and return currents respectively, cancel each other, and
The fourth and sixth are equal to
so do the fifth and seventh.
first
each other, and their
sum
2//,/1
is
/2
log TV"
= 2//,7
log ?v
ELECTROMAGNETISM.
234]
where r12
465
the mean distance between points

The second integral is
is
sections.
of the two cross-
the mean distance between pairs of points of the

and the eighth integral is the corresponding quantity
the area S2
Accordingly the self-inductance of the whole
where rn
is
section Si,
for
circuit per unit of length is
(21)
L=
7
fi
(4 log ru
2 log r u
r
- 2 log rn ) = 2p log ^-
Mi~aa
If the conductors are circular cylinders, we may use the formulae

If both the direct and return conductors are
already found.
single cylinders external to each other, their axes being a distance
d apart, rlz d. For infinitely thin tubes of radii
and R',
ru
= R,
=
=
r&
R',
so that
(22)
T-***J!B-
For tubes of
radii R^,
and
-R/,
RJ, integrating (17) over the
area of the ring,
{Sf (log
R? -l)-Bf (log jy -
1))
+ i (JZ, -
and making reductions

OE>2
lpg
(23)
From
we
this
ojrti
Z?2
7? 4
xi/o
7?
Ji
-/.i/i
'
Io8
we obtain r^ by
2 by .R/,
replacing RI,
obtain for the self-inductance of the circuit
(24)
For
f"
L=
f.
2^ log
^2
solid wires, since lim
(R
4
,
,R 2
log
log
R) =
0, this
^
2
2 '^
so that
Rf
/4
log
becomes
jB=0
w.
E.
30
THE ELECTROMAGNETIC
466
The
FIELD.
repulsion between the wires per unit length
dd
By means
dd
[PT. III. CH. XI.

is
of the principles here stated,
Maxwell has calculated
the inductances of coils of wire, by supposing the diameter of the

coil to be so great in comparison with the distance apart of the
may be treated like a group of

treatment
of the magnetic field due
The
straight conductors.
to currents even in straight conductors whose inductivity is
different
turns
that
the
coil
different from that of the surrounding
medium, except in the case

a problem of considerable complexity,
and the results given by Maxwell, for the case of two wires,
of concentric cylinders,
is
Art. 685, are only approximately correct.
CHAPTER
XII.
INDUCTION OF CURRENTS.
The
235.
Systems of Currents as Cyclic Systems.
phenomena of the induction of electric currents by changes in
The
the magnetic field were discovered by Faraday in 1831*.
results obtained experimentally by Faraday were deduced mathematically from the law of Lenz (see below), and from Ampere's
results regarding magnetic shells, together with the principle of
Conservation of Energy by F. E. Neumann f in 1845. The credit
due to Max well J of having had the idea of treating a system of

currents and the magnetic field belonging to them as a mechanical
system, subject to the ordinary laws of motion, and of thus de-
is
ducing the equations of induction from the generalized equations

of Lagrange and Hamilton.
The particular class of systems to
which currents may be assimilated is that studied by Helmholtz

under the name of cyclic systems, a detailed treatment of which
has been given in Chapter III.
We
have seen in the last chapter that if the strengths of a

of
currents be maintained constant, the currents tend to
system
move in such a way that the energy of the field produced by them
tends to increase.
This energy
is
a homogeneous quadratic function
of the strengths of the various currents, the coefficients, which we

have called inductances, being determined by the form and relative position of the circuits, and the nature of the medium in
which they are situated.
The medium being specified, these
of
the circuits may be made by giving
geometrical specifications
*
Faraday, Experimental Researches in Electricity, Vol. i. p. 1.

t F. E. Neumann, "Allgemeine Gesetze der inducirten Strome," Abh. Berl.
Akad., 1845.
J Maxwell,
CLV. 1864.
"A Dynamical
Sci. Papers, Vol.
Theory of the Electromagnetic Field," Phil. Trans.
i.
p. 526.
302
468
[PT.
III.
CH. XII.
finite or infinite number of geometrical parameters

qs
The electromagnetic forces due to the action of the currents may
be equilibrated by the action of certain impressed forces P s and
a certain
these forces
may be determined
as partial derivatives with respect

to the parameters of the potential energy, or of the energy of the
field.
These impressed forces we shall call the positional forces, and
since they are the negatives of the electromagnetic forces already

found, we have for any positional force
s
In order to specify the action of the system completely, we must

give, beside the values of the
parameters qs only the values of

the current-strength in every current-tube. If the currents are
distributed in three dimensions, this necessitates an infinite
number, but if there are a finite number of linear conductors,
,
The energy of
only a finite number of electrical parameters Is
the field is expressed as a homogeneous quadratic function of these
electrical parameters, the coefficients being functions of the posi.
tional parameters,
whose
velocities
do not occur.
If
we
consider
the negative energy of the field

as, instead of the negative
potential energy, the electrokinetic energy of the field, the current
strengths being considered as cyclic velocities, the analogy to a
mechanical cyclic system is complete. The cyclic coordinates q s)

being the time-integrals of the currents, represent the total
amounts of
electricity that
have traversed the respective
circuits
a fixed epoch. Since neither these coordinates, nor the

velocities of the positional coordinates occur in the expression for
the electrokinetic energy, all the conditions for a cyclic system are
since
fulfilled.
restriction must, however,
be made, which
is
of
no
importance in practice, namely that the velocities of the positional

coordinates must be small compared with a certain velocity, which
the velocity v, the ratio of the two units of elecall ordinary velocities, however, the
tricity.
is
electrokinetic energy
accurately represented in the form already
in this case
is
For the case of
found.
If
we have n
linear currents, the electrokinetic energy
T= \LJ? + M IJ
1Z
(2)
......
+ Mln IJn
is
235]
have the form obtained in 220 if a

coefficients L,
homogeneous medium is present, and in any case may be
where the
single
469
defined
as
magnetic
momentum
as
fluxes
in
The
231.
electrokinetic
of any circuit,
p =^- = Mls I + ...+L !s ...+M In)
(3)
1Vi
be defined as the total flux of magnetic induction through

all the currents.
We
may
that circuit in the positive direction due to
have already found
The
for
any positional
force,
equation
(i),
to any cyclic coordinate q s consists of the

s belonging
force
due to chemical, thermal, or
electromotive
s
impressed
other action, and the dissipative term given by Joule's law,
force
RS IS) where
we have
is
the resistance of the circuit.
Accordingly
dt
If
we write
this in the form
we see that the current in any circuit may be calculated by

Ohm's Law provided that we consider acting beside the electro-
8 an additional electromotive force

dpjdt. This
the electromotive force of induction, and from the above
definition of ps we see that it is equal to the time-rate of diminution of the flux of magnetic induction through the circuit in the
motive force
is called
The law
of induction was announced in virtually

and
was obtained from theoretical conby Faraday*,
siderations involving the idea of work by Neumann f, HelmholtzJ,
and Kelvin
The above equation is the general equation of an
positive direction.
this form
114, 3082.
Exp. Res.
t Neumann, loc. cit.
Helmholtz, Ueber die Erhaltung der Kraft.
Berlin, 1847.
p. 12.
B.A. Report, 1848.
Math, and Phys. Papers, Vol.
i.
p. 91.
Wiss. Abh., Bd.
i.
470
[PT. III. CH. XII.
and includes the steady state as a particular case,

for if the currents do not vary with the time, and there is no
motion of any circuit, every p s is constant, and we have for each
electric current,
circuit,
The statement is frequently made

does not hold for induced currents this is a mis-
the usual form of Ohm's Law.

that Ohm's
Law
conception, for in the statement of

electromotive forces of all kinds,
Ohm's Law we should include

including those due to in-
duction.
permanent magnets are present there will be terms in T

where each current is multiplied by the flux through it due to
magnets. These terms will be of the first order in the currents,
so that Twill not be homogeneous, and we have the case mentioned
If
We
have
66 each magnet acting like a concealed current.
in the previous chapter considered the possible replacement of a
magnet by currents, so that we may consider magnets replaced by
in
"
"
concealed
236.
or
"
"
apparent
Isocyclic
currents of unchangeable strength.
and Adiabatic Changes.
An
adiabatic
variation, being defined by the constancy of cyclic momenta, will

take place when in each circuit the electromotive force
s is just
large enough to maintain the current in the circuit steady, namely

IS If the current is varying, this necessitates the variaS
S
Such changes seldom occur in practice. Isocyclic

s
motions are such that all the currents remain unchanged. The
tion of
electrokinetic
momenta may be
varied by motion or deformation
of the circuits, involving change of the values of the parameters

qs or by motion of permanent magnets. The simplest phenomena
to observe experimentally, and those first discovered, are of this
,
class.
We may now apply to a system of currents the theorems which

have been demonstrated in
In particular may be
69, 70.
noticed the two theorems of
70,
which
may be
thus stated.
In any motion of currents or magnets during which the

strengths of all the currents are unchanged, the work done by the imthe work done
8 is equal to twice
pressed electromotive forces
I.
235
237]
against the positional forces, plus the

as heat.
For by
amount of energy
471
dissipated
70 (5) and (8)
and therefore
This theorem was stated by Lord Kelvin in 1860*.
II. Lenz's Law. In any system of conductors, induced currents
due to motion of the conductors are so directed as to oppose the
This law, stated by Lenzf in 1834, was, together with

Ampere's results, the basis of Neumann's deduction of the laws of
motion.
induction.
PARTICULAR CASES OF INDUCTION IN LINEAR CONDUCTORS.

Effect of sudden change of Electromotive Force
237.
or Resistance.
(1) SINGLE CIRCUIT.
resistance
Q)
containing
Let us
force
0)
first
a constant
consider a single circuit of

impressed electromotive
and accordingly traversed by the steady current
7 =
EQ/RQ.
Let now the electromotive force or the resistance, or both, be
E R
The current now varies

lf
suddenly changed to new values lt
from the initial value 7
in accordance with the differential
,
equation (5)
235, which becomes
we subtract 7l5 the steady value of the current under the

circumstances, from the total current, the difference
If
is
called the induced or extra-current.
The
new
differential equation
thus becomes
(2)
whose integral
is
*
Nichol's Cyclopedia, Article " Magnetism, Dynamical Relations of."
of Papers on Electrostatics and Magnetism, % 571.
t Lenz, Pogg. Ann. 31, p. 439, 1834.
Eeprint
THE ELECTROMAGNETIC
472
so that the induced current dies
FIELD.
[PT. III. CH. XII.
in geometrical ratio as the
away
time increases in arithmetical progression. Since after an infinite

interval of time the total current has attained the steady value Ilf
the value of the constant
The induced current
is
is
determined, and
we have
always in such a direction as to oppose the
change in the total current. The effect of self-induction is accordingly to make changes of strength less sudden. It is to be noticed
that the induced current varies in the same
manner
as the current
charging a condenser through a circuit without self-induction, as

207 (17). We shall here, as there, call the time in
treated in
which the current decreases in the ratio l/e the relaxation-time,
Both in the case of the condenser and in the present case
in-
creasing the capacity or the self-induction increases the relaxationtime, but whereas in the former case increasing the resistance
increases the relaxation-time in the latter
it
produces the opposite
effect.
In practical cases the relaxation-time is usually very short, so

that the induced current disappears almost entirely in a very
Under these circumstances the total quantity of

has passed may be measured by a ballistic galvanothat
electricity
For as the force exerted by the current on a magnet is
meter.
proportional to the strength of the current, the total quantity
short time.
rt
passing, or the time integral
Idt,
is
proportional to the time
J o
integral of the force on the magnet, or to the momentum imparted

If this momentum is all imparted before the
to the magnet.
magnet has had time to move, it may be easily shown that it
may be measured by
passing in a time
the
first
swing of the magnet.
f /# = f {/! + (7 - /O e-r] dt = I t-r (J

1
JO
Jo
The quantity
t is
- /O (e'r -
This formula was verified by Helmholtz* in 1851.

*
Helmholtz,
1).
The
total
" Ueber die Dauer und den Verlauf der durch StromesschwanWiss. Abh.
kungen inducirten elektrischen Strome," Pogg. Ann. Bd. 83, p. 505.

Bd. 1, p. 429.
237]
quantity due to the induction current
473
is
Idt = r(I -L\

n
If the current be passed through an electrodynamometer, that is

an instrument containing a fixed and a movable coil, the mechanical action between them is proportional to the square of the
current and the
momentum
imparted to the movable
coil is pro-
portional to the time integral of the square of the current.

effect due to the whole induced current is
The
/"
Jo
These two integrals have the same values that would be obtained
/x )/2 passing for a time 2r.
from a steady current of strength (/
Two
(2)
closed at the
In the case of two circuits which are

which have their electromotive
CIRCUITS.
same
instant, or
suddenly changed simultaneously, we have

the
subsequent period the differential equations
during
forces or resistances
(4)
Again
calling the final steady currents
we have
for
the induced currents
1^
=E /R
These equations are typical of all those in

readily integrated by the assumption
//>
where
A B
y
= AeP, /
(i)
(1)
,
I2
l,
(i)
72
may
IZ W
and are
Inserting
the factor eP appearbe omitted, giving us the simultaneous
(5),
+ R )A + M\B = 0,
M\A + L \ + R B = 0.
(L,\
this chapter,
and X are constants to be determined.
ing in every term
=E /R
(1)
=I
= BeV,
these values in the differential equations
equations
1^ = 7 - /!
THE ELECTROMAGNETIC
474
FIELD.
III.
[PT.
These equations can be satisfied for values of A and

from zero only if the determinant of the coefficients,
vanishes.
But
L\+R
2
expanded gives us the equation
this being
(L,L,
(7)
differing
M\
X + R1}
M\
CH. XII.
- M*) X + (RJ, + RiLJ X + R,R = 0,

2
a quadratic to determine X.
If
we
call its roots
X and X^ we have
x
_
(8)
,
Both roots are
we can
real, for
write
the quantity under the
radical sign
both terms of which are positive.

for since the electrokinetic energy
is
Both
roots are also negative,
we must have
intrinsically positive,
L^ -M
Having found the value
of
>0.
either of the equations (6) will give

If we choose the value
us the ratio of the constants A, B.

the first equation gives
B,_
~
If
/
we choose the value X 2 we
>.
obtain a different ratio
A,The theory
of linear differential equations shows that the sum of

is a solution, and that the general solution is
particular solutions
given by
where the constants A 1} B l A 2 B2 are connected by the equations

these
(9) and (10). We may now determine the absolute values of
,
237]
means
constants by
72
when t =
.
of the initial values of the currents
1^
and /2
/ô
=J (o)_/ a)=^ + ^
These being
475
(0)
we have
for the
/ and
a
induced currents
0,
l
+5
//o)=/2 (0)_/2 (D= B1

J
These equations with (9) and (10) determine the four constants,
The most important case is that
so that the solution is complete.
in which there is no electromotive force in one circuit, while the
other circuit originally open, and containing an electromotive force
E, is suddenly closed. The latter circuit is called the primary, and
will
be taken as that denoted by the
secondary, with the suffix
and
T
TTT/I.
R
2
*^J
J-
7?
7^
-I-
12 /
Since
\ and X
vanishes
has a
-1
_i_
7 ~*
T^
ii
r"
'
time goes on.
7"
-fî-t^2
z? r \2
7?
7"
'
the former the
suffix 1,
accordingly have
-["2-1^1
V \/ ( 7?
We
2.
\A/7~r
._ 7? r \2
_j_
2ip
p Ma
are negative, the induced currents die

function
away
as the
The
when
maximum
or
minimum when
and the curve representing
it
has a point of inflexion for
X22
These three points are equidistant, and, since \ and \^ have

This is
(72 have opposite signs.
the same sign, are real if C^ and

the case for the secondary current
/2 but
the primary current has

both coefficients negative, and consequently has no maximum nor
,
476
inflexion,
but
rises continuously,
[PT.
the appearance of
III.
its
CH. XII.
represen-
tative curve to the eye being the same as in the case of a

single
circuit.
The growth of the currents is represented in Fig. 92.
FIG. 92.
The
total quantity flowing in the

secondary is
EM
(13)
T> T)
Jo
'
-fi/iJ-â
This result
may
second of equations
also be obtained
(4),
with
E = 0,
by direct integration of the
r/
2 cft
0,
Jo
(14)
Jo
To
/i (0)
find the effect of breaking the primary current, we have

(1) =
/!
0, so that the whole quantity passing in the
= E/Ri,
secondary on breaking
is the same as on
making. This is one of
fundamental
results.
The
of variation of the
manner
Faraday's
on
the
different
from
that on making.
secondary is,
contrary, very
After the break we have to consider the primary circuit as sup-
pressed, so that the secondary is to be considered by itself, and

varies according to equation (3) above, where Ilt the final value, is
zero,
and 7
the initial value,
is
to
be found from the above value
of the time-integral of the secondary,
1^ Idt
JQ
(15)
=I
fY^' dt=^ = Ill 1*2
JO
R<2.
EM
237, 238]
The
fact that the
maximum
value
477
secondary jumps abruptly from zero to its

may be reached from considering
at starting
the preceding case with ^=00. The time between the secondary's taking the value zero and attaining its maximum and the
time from then to the inflexion is (log Xg/Xj)/^ \ 2 ), which is less
the greater
The
l}
effects
^=00.
vanishing for
here described
may be
illustrated
of the mechanical models described in
that the mass
by means of any
For instance suppose
71.
revolving with a uniform angular

velocity, the centrifugal force, which represents the electromagnetic
force, being just balanced by an applied force so that the distance
l5
Fig. 30,
is
of mj from the axis remains constant.
If ra2
is
at rest
and we
suddenly apply a force to the upper bar so as to increase its

angular velocity, the lower bar will begin to turn in the reverse
direction, the velocity representing the secondary induced current.
If on the other
lower begins to
hand the upper bar
move forward
from the
suddenly retarded, the
in the direct sense.
may be produced by suddenly changing
is
Similar effects
the distance of either
raj
corresponding to a relative motion of the two

have
circuits, producing a change in the mutual inductance.
not in this section explicitly considered this case, but since if the
or
axis,
We
change is made suddenly, and the circuits then remain at rest,

the differential equations are the same as those we have used, and
the solution is obtained from those here given.
238.
(1)
Periodically- varying Electromotive-force.

SINGLE CIRCUIT. Suppose that in the circuit is included
a variable electromotive-force varying proportionately to the cosine

of a linear function of the time, as would be the case if a coil of
wire should rotate in a uniform magnetic field about an axis in
the plane of the coil, and perpendicular to the direction of the
field.
(i)
Then the equation
for the current is
L-J-
RIE
cosa)t.
convenient way of treating such an equation is by replacing

the trigonometric term cos cat by the exponential e*, whose real
The value of / thus
part is the trigonometric part in question.
obtained will be complex, and its real part will be the solution of
the differential equation with the cosine term on the right, while
its imaginary part will have as the coefficient of i the solution of
478
[PT. III. CH. XII.
the equation with the sine on the right. In this way by separation of the real and the imaginary we are enabled to use the
exponential function, which retains its form on differentiation,
while the sine and cosine interchange.
Accordingly writing the
equation
L^ + RI=E^,
(2)
we may get a
by assuming /= Ae
on removing the factor e iut
io)t
particular solution
which in the equation
gives,
(Lico
(3)
+ R) A = E
as
+R
Lico
inserting
This determines the complex constant
so that the solution of the equation (2) is

.
_ E (R
Taking the
real part
Lico) (cos wt
we obtain
+ i sin
LW + R*
for
the solution of the equation
J so
(i),
+ La sin a)t)
LW + R*
EQ (R cos
cot
This assumes a more convenient form

constants a and
cat)
if
we determine two
that
cos
OL
Leo
sin a
giving
tan a
(4)
when the
solution
J = (Z a> +
2
becomes
(5)
/=^cos(a,<-a).
We may obtain
this result, and at the same time graphically

the
relations
of the current and electromotive-force by
represent
of
the fundamental properties of complex quantities.
making use
The complex quantity E eiMt has the modulus

cot, and is therefore represented by a vector
and the argument
of length
Q making
a vector revolving about the
an angle cot with the real axis, that is

The projection of this vector on
origin with angular velocity o>.
the real axis represents the impressed electromotive-force in the
238]
quantity varying in this manner

harmonic oscillation, with the amplitude
circuit.
The
electromotive-force takes on
and returns to
takes to
its original
make a complete
called the period,
is
unit time,
n=
and
CO/ZTT, is
479
is
said to perform a
values between
all
and
value in the time that the vector
revolution,
its reciprocal,
T= 2?r/&).
number
the
The time
of periods in
called the frequency.
In like manner the quantity Aei(at is represented by a vector

revolving with the same period, of length equal to the modulus of
Since the argument of a quotient is
the complex quantity A.
of
to
the
difference
the
arguments, the vector representing
equal
AeiMt lags behind that representing eitat by the constant angle
E
= arg.__.
But from the equation (3) we find that this ratio is the complex
1
The current,
quantity, R -f iLa), whose argument is tan" Lco/R.
of
the
second
vector
on the
the
projection
being represented by
real axis, is said to differ in phase from the electromotive-force by
the amount a, the difference in this case being a lag. The amplitude of the current, being the modulus of A, is the quotient of the
moduli
\R + iL<o\
Expressing these results analytically we obtain equation
(5).
The quantity J, by which it is necessary to divide the amplitude of the electromotive-force in order to obtain the amplitude of
the current, is called, as proposed by Heaviside, the impedance.
If the circuit has no self-inductance, or if the current is steady
(w
= 0),
it
It has
*
becomes the
resistance.
been proposed by Hospitalier* to
in the ratio
E /A,
call
the coefficient of
the reactance.
The mean value of a quantity varying harmonically taken

over any exact number of periods is zero, while in virtue of the
formulae
1
J-
rT
cos 2
JQ
ortcft
\ rT
|
-L
Jo
siu 2
c0tdt=n>
A
\
TT
JL
rT
si
sin
&>
cos o>cft
0,
JQ
Hospitaller, L'Industrie Electrique, May 10, 1893.

Bedell, Trans. Am. Inst. El. Eng. 1894, p. 640.
See
also,
Steinmetz and
480
the
mean
value of the square of such a quantity
[PT.
is
III.
CH. XII.
one-half the
square of its amplitude, and the mean of the product of two such
quantities of the same period and a difference of phase equal to a
right angle
is zero.
The quadratic mean,
or square root of the mean square of a

is called the effective or
variable current or electromotive-force

virtual current or electromotive-force.
a harmonically -vary ing quantity

divided by V2-
The
activity, or
is
Its value in the case of
accordingly the amplitude
power absorbed by the

Jf
EI= -~ cos o)t cos (cot

J
and
its
mean
value,
circuit, is
a),
by the above formulae,
solution (5) is to be added, in order to obtain the

the right
general solution of (i), the solution of the equation with
hand member equal to zero, obtained in the preceding section, but
To the
as the current thereby represented rapidly dies away, the resulting

state of the alternating current is that which we have found.
of circuits in parallel, to which a single harmonic

electromotive-force is applied, receive virtual currents inversely
number
if the frequency
proportional to their respective impedances
is great enough the distribution is almost independent of the
resistances of the branches, the
impedance being sensibly equal to
the reactance.
(2)
Two
CIRCUITS.
Suppose bhat we have two
circuits,
one
of which, the primary, contains a harmonic electromotive-force,

while the secondary contains no impressed electromotive-force,
except that due to induction.
(7)
The equations then
are
238]
or
making use
481
of complex variables as before,
+*;=
*
f*
(8)
,
(juJL I
particular solution
is
Lo
(\
-r
given as before by putting
IÂe^, I
giving
=BeP*,
+ R )A + MicoB = E
MicoA + (L ico + R )B = 0.'
(Lj,u
Eliminating
from these equations we get

\
do)
Comparing
this
T
ô, +
.
A+
with equation (3) above we find that the current
in the primary is the same as if the secondary were absent, and

the resistance and self-inductance of the primary were R' and L',
where
M = 2^! +
(ii)
Ll
These results were

paper
first
given by Maxwell in 1864 in his celebrated

of the Electromagnetic Field*."
"A Dynamical Theory
constitute the basis of the theory of the alternating current

transformer.
They
We see from equations (n) that the effect of the presence of

the secondary circuit is to cause an apparent increase of resistance
and decrease of self-inductance in the primary. Both of these
primary current
behind the electromotive-force, and accordingly, by (6), an increase of
effects cause a decrease in the angle of lag of the
power.
Inserting the values
( 1 1)
in (6)
we
obtain for the power
(12)
RJLf +
W. E.
2
ft)
Phil. Trans. Vol. CLV.
31
482
[PT. III. CH. XII.
increase the ratio L 2 a)/R2 the values of the apparent resistance and self-induction approach the limiting values
As we
These values are nearly approached in actual transformers, parwhen fully loaded with a number of lamps in parallel in
ticularly
Now
the secondary.
still
we have
when the primary and secondary
the same core
due
although
in general
are toroidal coils
wound on
231), so that very nearly the whole induction-flux
to either is linked with the other,
we have very
nearly
L.L.-MÔ.
The transformer
is then said to have no magnetic leakage.

In this
case the apparent inductance L^' is reduced to zero, the current
does not lag, and takes on the largest value that it can have,
namely
/=#/#'.
The
the power becomes,

square of R^/L^co in comparison with unity,
expression (12) for
__
+ R,M*IL*}
+ R,M*IL/Y + a>* (L, - WfLtf}
E<?(R,
2 {(R,
as
we
if
we
neglect the
'
see on dividing numerator and denominator
2
by L^co and
then adding and subtracting the term R^M^/L^ in the denominator. If there is no magnetic leakage, this increases as R 2
decreases, until it reaches the limiting value
\EI
2 R,
while
if
there
is
magnetic leakage, the power absorbed
maximum when
becoming equal to
-M
7**/
-j
and thence decreasing as
decreases.
is
238]
The power when
is
zero
which, for high frequencies,
when
is
483
only
may be much
less
than the maximum*,
maximum
being,
great enough, sensibly equal to the
value multiplied by
is
a>
The second
of equations (9) gives
B = - Mico
A -LT
<'*>
The modulus
of the quotient, being the quotient of the moduli,
._.._
I!
shows that the amplitude J2
of the secondary current is equal to

the amplitude of the primary IJ multiplied by Ma) divided by the
impedance of the secondary. Inserting the values of R, L', from
(ii) in /if,
(17)
(0)
A =L M
I
V/,
/2
gives for
(o)
,
|.B|=/
(18)
V+R
(0)
E Mco
n
AZ
2
2
2
+ M - 21
4
In the case of no magnetic leakage this becomes
J2<>=
(19)
and if we may neglect R^L^w or

we have the simple form
R /L
2
2 a)
in comparison with unity
(20)
This
and
the practical equation of the transformer.
is
(6),
J.
By
231, (5)
we have
J.
Magnetism,
Thomson, Elements of
the Mathematical
Theory of Electricity and
p. 409.
312
THE ELECTROMAGNETIC
484
FIELD.
and nz are the numbers of turns

* s so
small that
secondary coils. If w 2
we
have
neglected,
where
[PT. III.
CH. XII.
and
in the primary
its
square
be
may
E-
(21)
/.
The
ratio Wg/Wj is called the ratio of transformation.
The argument of the ratio B/A, being the difference of the

arguments of the numerator and denominator, shows that the
secondary lags behind the primary current by the phase-angle
which approaches two right angles as the ratio
The
efficiency of
in the secondary
^jft 2
a)/R 2 increases.
the transformation, or the ratio of the activity
/2
(o)2
,
^ na *
n ^ ne primary,
is,
by ( 1 8) and
2),
+ w (B^Lf +
2
which, neglecting RJtyM**?, becomes

(22)
that
is,
239.
in practical cases, nearly unity.
Circuit containing a Condenser.
In the
cases
heretofore considered the only energy of the system has been

If the circuits are connected with conductors
electrokinetic.
upon which charges of
electricity can accumulate,

addition have electrostatic, or potential energy.
we
shall in
As fche simplest case let us consider a single circuit whose ends

If the charge of one
are connected to a condenser of capacity K.
is
the current flowing
at
instant
then
condenser
any
q,
plate of the
into that plate
(0
The
is
defined as
'-%
electrostatic energy of the
system
is (
143)
238, 239]
which gives
rise
485
to the difference of potential, or electrostatic
electromotive-force impressed in the circuit in the direction of the

current,
'--
-j.
:
Accordingly the differential equation
from which, substituting from
(i),
for
we
the current
is
obtain the equation for the
charge,
Again, assuming q
= e xt we
obtain the quadratic for X
(6)
whose roots are
x=- 2Z + V4#
/^T3~
KL
(7)
x-
We
4Z
V/^
2Z
have now to consider two
CASE
I.
R* > ^LjK.
(8)
JL
'
.BTi
cases.
We
Both roots real
then have
and as
and Xg are both negative, the charge, and likewise the
current
I=\ Aet> +\ BeW,

t
(9)
die gradually away.

If there
motive-force
in
the
Q
circuit,
a permanent impressed electroto

we must add the quantity Q
is
EK
the charge, which, however, does not affect the current.
Determining the constants
and
by the conditions that
initially neither current nor impressed electromotiveand

that the initial charge is q we have
force,
there
is
(10)
while
force
q
if
0>
there
is
no
we obtain
initial charge,
but an impressed electromotive-
[PT. III. CH. XII.
In either
case, the curve representing the charge as a function of

the time has a point of inflexion distant from the origin by the
amount
6=
(12)
*gx
'
while the curve of current has one at an equal distance farther on.
of charge and of current are represented in Fig. 93.
The curves
FIG. 93.
CASE
II.
<
Both roots complex.
4tL/K.
^ = ~97'
LLi
we have
for
V=S
If
we
write
/"I
\/K7'~4T*'
V J\.Ju *Li
the roots
= 1* + iv,
- iv,
p,
^
X =
2
and we may write the solution

q
(14)
= e (A
fit
cos vt
+ B sin vt).
In this case the charge not only dies away, but periodically
changes sign, performing a damped harmonic oscillation of the
period
(.5)
We
(1
6)
^
2
JL _^
KL *&
have
the current
for
!=at
= e"* {(Aa + Bv) cos vt + (Bii
Determining the constants so that the

the charge ^ we have
Av)
initial
>
(17)
= q e^ (cos vt
sin
vtj
sin
vt}.
current
is zero,
and
487
239]
This case
is
The charge
represented in Fig. 94.
zero at times
is
such that
vt
= tan- - = 6,
1
FIG. 94.
which
is later
than the time of the vanishing of the current by

6, which approaches Tr/2 the smaller p.
the phase difference
We may
specify the damping, or decrease of the
charge or
by the relaxation-time of the damping factor e^ namely

2L/JR, or by the logarithmic decrement, that is the logarithm
f
current,
r=
of the ratio of a
maximum
value to the absolute value of the
next following minimum.

Since the maximum and minimum
values of the parenthesis in (17) are equal and opposite, and
separated by intervals of time T/2 = TT/V, the ratio of the absolute
values of q
/
is e
v
,
ox
and the logarithmic decrement
(15)
/ATT
=
KR*
If
R = 0,
there
is
no damping, X
Tt
(19)
X,
"
Introducing these values of
and the period
is
= *tr*/KL.
and X we may write
(15),
(ao)
so that if the
damping
is
small
quantities of the second order.
it
affects the period only
by small
THE ELECTKOMAGNETIC
488
FIELD.
[PT.
III.
CH. XII.
We
have in this case a type of the very important class of

phenomena known as electrical oscillations, of which we shall
The theory here given was

Kelvin*
Lord
in
1855
and
published by
by Kirchhofff in 1864.
The theory was confirmed experimentally in a qualitative manner
presently give the general theory.
by FeddersenJ in 1857, by observations on the electric spark

arising when a Ley den jar is discharged, and by Helmholtz in
1869, and Schiller in 1874, under conditions admitting of quanMore exact determinations in absolute measure
titative results.
have been made by Lodge and Glazebrook by a method involving
the spark, and by the author, by a method similar to that of
||
Helmholtz.
We
have seen that the occurrence of oscillations is due to the

If there is no
presence of both kinetic and potential energy.
kinetic energy, L = 0, and we reach the case treated in 207, while
if there is no potential energy, we have the case of
237, to which
K=
we may
oo
mechanical model of an
pass by putting
obtained
from
be
mechanical
may
any
system possessing
both potential and kinetic energy, such as a pendulum or a heavy
.
oscillation
The stronger the spring the quicker is

spring.
so
that
we
the oscillation,
may assimilate the reciprocal of the
body moved by a
The
capacity of the condenser to the elasticity of the spring.
self-inductance of the system, on the other hand, is the analogue
The analogy
of the mass, or inertia of the mechanical system.
of the resistance
may be
obtained by making the system move in

is retarded by a force
a viscous medium, so that the motion

proportional to the velocity.
240.
Periodic Electromotive force.
Resonance.
If
into a circuit joined to the plates of a condenser is introduced

a harmonically-varying electromotive force, we have for the current, instead of (4) of the preceding section the equation
(i)
Cut
K.
"
Thomson, On Transient Electric Currents," Phil. Mag. June 1853 Math.

and Physical Papers, Vol. i. p. 540.
t Kirchhoff, " Zur Theorie der Entladung einer Leydener Flasche," Pogg. Ann.
;
Bd. 121, 1864; Ges. Abh.
p. 168.
J Feddersen, "Beitrage zur Kentniss des elektrischen Funkens," Dissertation,

Kiel,
||
1857; Pogg. Ann. 103, p. 69.

Helmholtz, "Ueber elektrische Oscillationen," Wissensch. Abh. Bd.
Schiller, Pogg. Ann. 152, p. 535.
i.
p. 531.
239, 240]
Proceeding as in
238,
(2)
L\
and assume
for
we
489
write
the particular solution
Ae
iu>t
,
which inserted
in (2) gives
'
Kio))
From
this
we
get,
by comparison with
238, for the impedance,
'/.
(4)
and
for
the lag of the current behind the electromotive
ot
(5)
force,
= tan
so that the solution of (i)
is
E, cos
(tot
- a)
In order to obtain the general solution we must add to this result

= from the previous section.
the solution of the equation with
An
whose period
that of the force, as in our present

is
called
a
case,
forced oscillation or vibration, in contradistinction
to the case of the previous section, where, no force being applied,
oscillation
is
governed by the constants of the system, and the

If there is damping, the
free oscillation soon dies away/ leaving only the forced oscillathe period
is
oscillation is called a free oscillation.
tion.
We
see
L = 0,
other hand
no condenser,
and
current
the
238,
lags, while
(6) that if there is
by
obtain the case of
the lag
is
if
oo
we
on the
negative, or the current advances
by
the phase-angle
a
The reason
of this
is
= tan"
KtoR'
of course that in the differential equation
multiplied by the derivative, and the capacityreciprocal by the integral of the current, which, when the
electromotive force is an exponential with imaginary exponent,
the inductance
is
introduce
factor
the
ico
into
the
respectively, producing opposite effects
numerator or denominator
on the argument of A.
THE ELECTROMAGNETIC
490
FIELD.
[PT.
III.
CH. XII.
Thus the tendency of the inductance and capacity is to neutralise

each other's
there
is
Exact neutralization
effects.
neither lag nor advance,
is
produced, so that
when
2-7T
In this case the impedance is the smallest possible, and the

magnitude of the current is a maximum, being the same as would
be given by Ohm's Law for steady currents with a closed circuit.
The period of the electromotive force which gives this result is
exactly that of the free vibration which would be natural to the
Under these circumstances
system if there were no damping.
the system is said to be in resonance with the force.
The
magnitude of the current is inversely proportional to the resistFor this

ance, and if there were no damping would be infinite.
reason resonant oscillations, either mechanical or electrical,
may
be very intense.
By connecting two similar circuits with two
similar Ley den jars, Lodge has caused the oscillatory discharge of
one jar to produce such violent resonant oscillations in the other
circuit that a
phenomena
considerable
spark-discharge
is
interesting papers
The
number of
produced.
of resonance have been demonstrated in a
by Pupin*.
In order to show how the resonance depends on the agreement

of the frequency of the impressed force with that of the free
we give in Fig. 95 a graphical representation of the

If we call co m the value
current as a function of the frequency.
vibration,
of
co
which gives the
maximum
2
the amplitude of
current,
is
a>
In Fig. 95 are plotted the
values
of the
factor
of
The
E /R
as
different
the abscissas being those of o>/o> m

curves are, beginning at the outermost, for integral values of
ordinates,
the ratio
^ R
from
1 to 10.
The resonance
is
sharper the
larger this ratio.

*
Pupin, "Electrical Oscillations
Am. Journ.
Science, April,
of
May, 1893.
Low Frequency and
their
Resonance."
240, 241]
General Theory
241.
shall
now
of Electrical Oscillations.
491
We
consider the question of electrical oscillations in the
FIG. 95.
most general case of a network of linear conductors, conducted

which may carry electrostatic
with any number of conductors
in
be
These
pairs to form condensers, as
may
grouped
charges.
in the last section, or they may be entirely independent of one
Of the linear conductors, any one may form a closed

unconnected with the others, and affected only by current
induction, or may end at points of embranchment with other
another.
circuit
For brevity we
upon any of the conductors K.
the linear conductors wires, and the conductors
conductors, or
shall
call
We
suppose that the net contains p points

of embranchment, k of which are connected with accumulators,
for all wires which end on the same accumulator are to be
accumulators.
shall
Let the number of

considered as meeting in an embranchment.
Then if all the wires form a part of the same net,
wires be I.
the number of independent meshes is
+ 1, for we see at once
lp
number
of lines that can join p points to form

a closed net is p, giving one mesh, and that after the first mesh
every additional line adds a mesh*.
that the smallest
For every wire r between points a and

,
,,
dI^
dt
dlo
J-T*
dt
dt
we have an equation
where Eab is the impressed electromotive-force from a to b and Va

and Vb are the potentials of the points a and b. There are I
equations of this
*
sort.
By independent meshes we mean such
that circulation about any one is not

For instance the outer
the resultant of circulation about any number of others.

boundary of a plane net is not independent of its meshes.
492
[PT. III. CH. XII.
For every point of embranchment a we have an equation

(2)
the currents being
now marked with double
suffixes to
denote the
171, and ea denoting the

points between which they run, as in
of
the
accumulator
connected
with
the point, or zero if
charge
there
no accumulator.
These
equations are not all indepentogether, every current appears in both

so
that
the
left-hand
side in the sum is identically zero,
directions,
is
dent, for
adding them
all
giving
which
is
is
merely the statement that the total charge of the system

1
There are accordingly p
unaffected by the flow of currents.
independent equations
(2).
For every accumulator
Ka we have an equation,
138
(10),
Va=Piaei
(4)
From
the equations
hoff's principle,
179.
( I)
If,
the F's
may be
traversing any
eliminated by Kirch-
closed circuit,
we add
the equations (1) for each wire, every V appears with both signs,
so that on the right we obtain the sum of the E's around the
circuit.
We shall thus obtain as many equations as there are
independent meshes in the net, I p + 1. Other equations may

be obtained in the same manner by traversing any unclosed circuit
ending on two accumulators. All the potentials at embranch-
ments passed over are eliminated except those of the two ends.
The number of equations to be obtained in this manner is one less
than the number of accumulators, or k 1. We thus obtain in all
+ k = n equations, and there are the same number of inde-
lp
pendent variables. We may take as parameters to characterize

the system a set of currents, one circulating in each mesh, so that
the actual current in any wire is the sum or difference of the
The
currents in the two meshes to which that wire is common.
the
of
one
for
mesh-current
shall
be
taken
time-integral of any
lp
+ l q's thus denned, we will

parameters q. Besides the
I others, denoting the integral currents along any series
choose k
of wires joining the accumulators two and two, the whole series
is
forming a chain with two ends. The charge of any accumulator
241]
thus the difference of the two
493
of this sort whose wires
q's
it
The whole number of qs is now just equal to n, the

separates.
number of degrees of freedom of the system. The current in any
wire is the sum of two or three of the q's with the proper signs,
and
is a
homogeneous quadratic
becomes one also of the q"s. The
the electrokinetic energy
as
function of the currents,

7
derivative
it
'i/TTx
r hr
>] is
the electromotive-force of induction around
dt \dqs 'J
the circuit
for
s,
'
and every dlr /dqs is zero except in the case of the currents which
bound the circuit, for any of which dlr/dq8 is either plus or minus
'
The
unity.
becomes
dissipation function,
also a
64, (7)
homogenous quadratic function
the product terms will in general appear.

will also
which
be represented by
again the
The terms
is
sum
7 ,
of the q"s in
The
which
dissipative force
for
of the products
RI
around the
circuit.
ar
are accordingly what we get
wires bounding the mesh s.
Since any charge

the second
(i) for all the
equal to plus or minus one of the q's of

between two, W, the electro-
sort, or to the difference
static energy,
q's.
is
by adding the equations
Again
becomes a homogeneous quadratic function of these

^
vQs
is
the electrostatic electromotive-force belong-
ing to qs for
,
*r
dqs
Now by
lators at the
(4),
u6r
Vrt
while
'
der dqs
is
zero except for the
accumu-
OQs
beginning and end of qs where the derivative has the

,
494
values minus one and plus one respectively.
We
[PT. III. CH. XII.
shall write our
three functions
(5)
where the M' s are linear combinations of the inductances of the

wires, the R's linear combinations of their resistances, and the p's
linear combinations of the coefficients of electrostatic potential of
The values of the coefficients of the three
the accumulators.
functions are such that each of the functions
is
positive for all
possible choices of its variables.
We may
now apply Lagrange's

dpT\ dF
*Wv
where
circuit.
is
equations for any parameter qs
dW
99/
"
the total external electromotive-force around the
Performing the differentiations this becomes
a linear differential equation of the second order with constant

coefficients.
We have one such equation for each parameter qs
.
We
shall first find the free oscillations, that is the solutions
with every s = 0. As in the case of the simple examples of 237,

a particular solution may be obtained by assuming for every qs
,
= a^,
q,
(8)
where X
the same for
is
all
the qs.
Inserting these values in (7)
we obtain
(Jf21 X
(9)
......
+ (Mnn\ + Rnn\ + jpww ) dn = 0,

2
a set of linear equations to determine the ratios of the a's when X is

known. If these are to be satisfied by other than zero values of
241]
495
the as, however, the determinant of the coefficients must vanish,
namely
(10)
R nl \ +pu,
This
is
Mnn\* + Enn\ + pn
an equation of order 2n in
powers are absent
if
We
0.
^-1
If
......
X,
from which the odd
shall denote its roots
^-2 J ......
we multiply the rth equation

we obtain
by
^271
(9)
by ar and take the sum

,
for all rs,
\^
(1 1 )
^s
rs
a ras
+ \2,r2 Rrsar a + ^ r ^ prs a ra = 0.

s
The double sum by which X2 is multiplied is the value of the

We shall denote
function 2T when for every qg is substituted as
'
by 2T(a). Similarly the coefficient of X is 2F(a) and the term

independent of X is 2TT(a). But by the fundamental property of
the three functions, each must be positive. The equation (i i),
this
+ \F(a) 4- W(a) = 0,
X2
2
shows us at once that X can not be real and positive, for that
would involve the sum of three positive terms being equal to zero.
if
Secondly,
F = 0,
that
is,,
if
the resistance of every wire
is
zero,
is a pure imaginary.
In thi,s case &* and e~ M are trigonometric functions, representing an undamped oscillation of the
same period for all the parameters q.
and X
F is large
enough, X can be real and negative. In

parameter q gradually dies away to zero, the relaxation time being the same for all.
This corresponds to Case I of
Thirdly,
if
this case each
239.
Fourthly,
if
either
W or T is zero, instead of a pair of roots we
have a single one, which is real and negative, the cases corresponding respectively to 237 or 207.
Fifthly, in
vanish,
and
other cases, that is when neither T, F, nor

too large, X is complex.
We shall prove that
F is not
then the real part of
is
negative.
496
[PT.
III.
CH. XII.
When any value of X is determined, the equations (9) determine the quantities a except for a common factor. If complex
values enter, since any equation which involves i will also hold
i, changing any root X to its conjugate
good if i be changed to
every a to change to its conjugate a'. We shall denote
corresponding to the conjugate roots X and X' by a and a',
X' causes
the
a's
where
X = p + iv,
as
= a + i@
s
X'
iv,
fju
dg =OLS
- i&
Let us now apply the process that gave us equation (n),

except that
we multiply the equations
belonging to
\*2 r 2 s
(12)
rs
ar a s
'
+ X2r 2 #
s
In this equation, any

which r = a, s = b and r =
rs
= a,
or substituting the values of the

i/3 a ) (06
X by
the a"s
'
+ ^ r ^ prs ar a = 0.
f
ab
appears in the terms for
so that the
4-
sum
is
a b a a '),
a's,
- ify) + (06 + i
Using a notation similar
(13)
a r as
coefficient
b,
(12)
(9) containing
obtaining
X',
to that before employed, equation
is
X2
{T(oi)
T(0)}
Now performing
+\{F (a) + F(0)} +
W(a) + W(/3) =
0.
the same process on the equations (9) with

a's we obtain
X',
and multiplying by the

(14)
\'*{T( a ) + T(/3)}
so that
X and
fore for their
+ \'{F( a ) + F(f3)}+
X' are roots of the
same
TF()+TF(/3)
quadratic.
We
= 0,
have there-
sum
The solution therefore represents

negative.
in
as
the
second
case of 239, the period and
vibration,
Accordingly
a
//,
is
damped
damping being the same
Since for every root
for all
the qs.
X we obtain a
set of values of the a's,
shall distinguish the values for the different roots

suffixes, so
that a rg
for the sth period.
we
by a second set of
means the coefficient of e* in the coordinate qr

The theory of differential equations tells us
241]
that for the general solution

for all the roots, so that
ql
497
we must take the sum

we obtain
of the terms
(16)
We may now replace the exponentials by trigonometric

The appearance of the terms with conjugate imaginaries
a s e xt
+ a V7 = 2e^ (a
s
cos vt
terms.
- & sin vt)
leads to the disappearance of imaginaries from the result, so that
we
obtain,
g2
=2
=2
qn
=2
ql
J- it
(oi ll
cos vj
/3U
{e^'*(0ai
cos vj
/32 i
{e'
+
sin v-f) +
sin v-f)
. .
. . .
+ e* nt (oLln cos vn t
4- e* nt (<x
/3ln
sin vn t)},
cos vn t
- m sin vnt)},
cos
fi.
(17)
The
sn
ratios of the a's or of the
/3's
z/
sn
any column are given by the
in
equations (9), being different for the different columns. Since the
ratios of the /6's of any column are the same as those of the a's, we
may
q1
^2
otherwise write the equations as
- yj + A e^ cos (v t - 72) ... + A ln e^ cos (vnt - yn

)},
=2 [An** cos ( Vl t - 70 4- Aê^ cos e - 72) ... + J.flM cos (vnt - yn )},
=2
{A u ePt cos (vj
l2
(i/ 2
THE ELECTROMAGNETIC
498
FIELD.
[PT. III. CH. XII.
of any particular simple oscillation are the same for all

the coordinates, and the n factors of the amplitudes and the n
phases are to be determined from the initial values of the q's and
damping
of their
first
We
will
time derivatives.
now
consider the case of forced vibrations.
On
account of the linearity of the equations, if we find a solu(l)

for a particular set of values
of the right-hand
8
members of our equations (7), and a second solution q^ 2) for a
tion qr{1)
second set E}*\ then the

right-hand members
sum qr + qr
EM+E \
(l)
are
(z)
(2
will
be the solution when the
We
shall, therefore,
the effect of each impressed force by

force,
cos wt, all
Suppose
first
then
impressed a harmonic electromotive

of the same period. Then we have the equations
that in each circuit there

of which the 5th
itself.
consider
is
is
(20)
+
Assuming qr = a
i(at
these reduce to
+ Rlnia) -f Pin) an =
a>
+p m
0!+
- -
+ (-Mnntf+Rnnia + pnn) d n = En
a set of linear equations to determine the
a's.
we call the determinant of equation (10), D (X), and Drg (X)

minor of the element of the rth column and sth row, we have
If
- the
as the solution of (21),
Since
whose
(23)
D (X) =
is
roots are
the determinantal equation for the free vibration,
X l X2
,
...
X2w we have
D(\) = C(\- \)
(x
Accordingly the denominator
-x)
2
...
D (ico)
(x
\, n )
= cn
(x
is
D (ia) = CU (i& - X = GU {- + i(coThe minors D (ico) are rational integral functions of
(24)
rs
s)
- \).
fji s
v,)}.
iw,
and the
numerators are therefore complex quantities, which reduce to real
241, 242]
ones
if
and
its
is
Calling the modulus of a numerator
the R's are zero.
r,
argument 6rt
*J)n (ia)Es = Bj*r
(25)
Or
499
a small angle
if
We
the resistances are;small.
thus have
r
_
cn
A,
cn.[- & + %(-*.)}r
a"
where
'"
Retaining
now
only the real parts,
we have
for the solution
(28)
Thus
the resistances are small,
if
all
the oscillations are in
nearly the same phase. If the frequency of the impressed force

coincides with that of any one of the free oscillations, co
vs = 0,
and one
factor of the
denominator reduces to
/*,,
so that if the
of that oscillation is small, the amplitude is very large,

or infinite if there is no damping.
This is the case of resonance.
damping
(Resonance
may
also
which
all
the
J2's
be defined in a slightly different manner as
one of the roots of the equation D (X) =

in
have been put equal to zero. This corresponds
occurring when iw
is
with our example in
240.
In practical cases the difference
is
very
small.)
now we have
If
one of which
is
a system acted on by electromotive forces each

the sum of any number of harmonic components of
different periods, any component may cause resonance with any

free oscillation of the system, so that resonance may occur in a
large
number
242.
of ways.
Examples.
Two
Circuits.
We
shall illustrate the
principles of the preceding section, aside from the examples that

have already been given in
239, 240, involving one degree of
Consider an induction
freedom, by an example of two circuits.
which both the primary and secondary contain a condenser

This is the case of the so-called Tesla high-frequency
in which a Leyden jar produces an oscillatory discharge
coil in
in series.
coil,
through the primary, while the ends of the secondary are usually
connected with a small capacity, say a pair of knobs. We shall
322
500
[PT. III. CH. XII.
take for q t and q% the charges of the two condensers, so that the
currents are
T
/i=
(I)
We
dqz
dqj_
/2= 1-
i-
accordingly have
(2)
and the
differential equations for the free oscillations are
(3)
The equation
for the frequencies is
L^ + jftjX +
2
= 0,
-
(4)
or
(AA- -if V + (Aft + LA) x + (\21 + --i + ftft) V

2
(5)
0.
As this equation is of the fourth degree, we shall treat only the

case of no damping, which, as we have seen, will not cause a large
error in the determination of the frequencies.
Putting then
JJj
= _R =
the equation becomes
/^\
-N
-^1-^1
'
-^2-&2
i /r~\
or,
as
we may
1
(7)
-\2_i_
A< ~I~
ff
^
-T
otherwise write,
+ î-K. (A A - -M" ) = 0.
If the two roots of this quadric are

J.
!_
V' V'
242]
we have
for
501
the periods,
T _^ri
T ,_27ri
=
'
'
X,
X,
so that
T=TT s/2
[f
we
introduce the periods of the two circuits alone,
which
and a quantity
nearly a
is
mean
proportional between
them,
these periods
become
IT* +
T_
r + Vfo* - r
2
"V
(9)
- 7(^-
~V
2^=^,
^ = T, +
2
+40*
M +r
Incase
2 2
2 )
&>
= 47T
F* =T*-fr
a case of so-called resonance, though not the one that we have
We see that one of the periods is greater and the
other less than the common period of the separate circuits.
This
is
examined.
If the period of one of the circuits

of the other, so that both
T^T,
we
is
much
greater than that

t
and T*-T<?>20*,
have, developing the square roots by the binomial theorem, the
approximation,
7*2
*
rp'l
-L
In this case the longer period
"
T7 2
-LI
-r
rp 2
^
is
V
z
rpz>
2
nearly that of the longer individual
period, being somewhat longer, while the shorter period is somewhat shorter than the shorter individual period. This is probably
THE ELECTROMAGNETIC
502
FIELD.
III.
[PT.
CH. XII.
coil, where only the longer oscillation

For a further treatment of this example, the
plays much part.
reader is referred to articles by Oberbeck* and Bliimckef.
the usual case of the Tesla
We shall now consider the forced oscillation. Let there be

cos at in the primary circuit, there
an impressed force
being
none in the secondary. Then we have for the secondary
(12)
03
'
The amplitude
(13)
of the secondary current
72 =
(0)
to
.
\
a2
is
\
V+
We
l^^
get resonance
when
&>
(^-*')*
(14)
In case there
is
no condenser in the secondary, we have
and there
is
oo,
then but one frequency
(15)
for resonance,
^^K^LJ^-W)-
This
is
one of the roots of the quadratic
is
the practical case of a transformer or induction coil, and

by J. J. Thomson in his Recent Researches in Electricity
is
treated
and Magnetism, Chapter
VI., to
which the student
is
referred for
further examples of this subject. For a treatment at length of the

subject of oscillations, the student may consult Rayleigh, Theory
of Sound, Chapters IV, V. and X.B, and Routh, Advanced Rigid
Dynamics, Chapter
*
Oberbeck.
II.
"Ueber den Verlauf der
electrischen
Schwingungen
bei
den Tesla'-
Wied. Ann. 55, p. 623, 1895.

"Bemerkung zu der Abhandhmg des Hrn. A. Oberbeck."
schen Versuchen."
t Bliimcke.
Ann. 58, p. 405, 1896.
Wied.
CHAPTER
XIII.
EQUATIONS OF ELECTROMAGNETIC FIELD.
ELECTROMAGNETIC WAVES.
243.
Localized Electric Force of Induction.
preceding chapter
we have developed the theory
In the
of current induc-
tion in linear circuits, on the basis of the treatment of a set of
currents as a mechanical cyclic system, aod we have thus arrived

which are justified by experiment.
have found
We
at equations
for the electromotive force of induction in
d fdT\
any
circuit,
dp
where p, the electro-kinetic momentum corresponding to the circuit,

is by the results of
226 defined as the total flux of magnetic
induction through the circuit, that is the surface integral
p=
(2)
/T{8 cos (me)
+ W cos (ny) +
91 cos (nz)}
dS
over any cap bounded by the circuit.

i
If
we
consider the electromotive force around the circuit as
made up
as in
of electric forces acting at each point of the circuit, just

166 we considered the electromotive force due to electro-
static action as the line-integral of the electrostatic field-intensity,
we may here
consider the electromotive force as a line-integral
around the
circuit,
(3)
E<= l(Xdx + Ydy + Zdz) = JF cos (Fds) ds.
The vector
whose components are X, Y, Z is a quantity of
the same nature as the electric field-intensity, and we shall not in
future distinguish whether it is of electrostatic or electrodynamic
origin. If we apply Stokes's theorem to the line-integral in (3) we
504
[PT.
III.
CH. XIII.
it into a surface-integral which, in virtue of (i), must be

to
the negative time-derivative of the surface-integral in (2).
equal
convert
f[(fdz
(4)
dY\
fdx
-- dz\ cos(m/)
Uo*J)cos(nx)+ \dz
dxj
.
tt(JJ(\dy
^ cos
cos
+ ^ cos
/dY
dx\\
+ (^-^}\ dS
\dx
.
dy/)
d8t
As we assume that the circuit does not change geometrically

with the time the differentiation with respect to t may be passed
under the sign of integration, and operates only on the quantities
,
9R,
same
$1.
Since the two surface-integrals may be taken over the

and the equality holds for any portion of surface
(as we may choose any cap over any circuit), the
surface,
whatever
integrands are necessarily equal at

tating the equations
_as = a^_d_F
dt
points of space, necessi-
all
'
dz
dy
dt
=
~~
dt
dx
'
dy
These equations, which are more compactly expressed by

(6)
-f-ouriJ.
are the general equations of induction, and are justified because of

their leading, by the reverse process, to the equation (i), which is
direct experimental verification

directly verified by experiment.
of equations (5) has been given but recently.
we wish
to introduce the vector-potential belonging to the

magnetic induction, by 226 we have the alternative expression
If
forp*
p=
(7)
Comparing
( 8)
this
now with
l(Xdx + Ydy
*
(jftb
+ Qdy + Hdz).
the line-integral in (3) gives us
+ Zdz) = - ^
l(Fdx
+ Gdy + Hdz).
See the definition of p following equation
(3), p.
469.
243]
505
the equality of the line-integrals we must not conclude

the equality of the integrands, for the line-integral of any lamellar
From
We
vector point-function around a closed path vanishes.

ingly obtain
accord-
(9)
where
(X
whole
electric force,
it
changing
Y' Z')
is
becomes
a lamellar vector.
If
X,
Y,
denote the
when the
state of the magnetic field is not

the electrostatic force, so that the components
X', Y', Z' must be the negative derivatives of the electric potential.
Accordingly the equations are
r= _aF_8F
dt
dx
dG
dV
__
"
'
'
dz
dt
These are the equations as given by Maxwell*. We shall

however prefer the form (5), not containing either potential, as
introduced by Heavisidef and Hertz +
Since the electrostatic
field has no curl, it need not be considered
separately in equations (5).
If however there are impressed electromotive forces X', Y', Z'

not of electrostatic origin, such as those due to chemical or
thermal effects, and X, F, Z still denote the total field, we must
replace
X,
Y,
(Heaviside, Vol.
I.
in equations (5)
by
X-X
f
,
Y-Y', Z-Z'.
p. 449.)
In a closed conductor undergoing electromagnetic induction

there are not necessarily differences
of electric potential, for
Treatise, Art. 598, equations (B).

t "Electromagnetic Induction and
Papers, Vol.
I.,
"Die Krafte
Theorie."
its
Propagation."
Electrician, Feb. 1885,
p. 447, eq. (20).
elektrischer
Wied. Ann., 36,
Schwingungen behandelt nach der Maxwell'schen
p. 1, 1889.
Jones's trans., p. 138.
506
[PT. III. CH. XIII.
example in the case of a circular ring placed perpendicularly to the

force of a varying uniform magnetic field the electric potential is
If however the circuit is broken, current flows for a
constant.
very short time until the electric force vanishes; there is then
produced
disturbance
of
producing differences
charges
of
potential to be calculated from the equations
_dV ~
= d_G_
_d_v^dF
~
9#
'
'
dt
_9T ~
= aff
dz
dt
dy
'
dt
Conductors connected to the broken ends of the
circuit, for
instance the plates of an electrometer, will then show a difference

of potential.
244.
Displacement Currents.
tions (5) with the equations
222
=
4f7TV
we compare the equa-
---^
dy
(ll)
If
(2),
dz
-- dN
= dL
-^
dz
-5
da
-- dL
:r-
-=
dx
dy
we
notice that they are analogous in having the right-hand sides

equal to the curl of the electric and magnetic field respectively.
We
make the analogy
still more complete by introducing the

introduced
into
the theory by Maxwell of the electrical
conception
current
*.
displacement
Suppose that we have a condenser charged with electricity.

There is then a field of electric force, the lines of force running
from the positively charged plate to the negative. The electric
induction is, by 182 (16),
g=
47T(7.
the plates be connected by a conducting wire, the

the wire, until
positive charge passes from the positive plate along
a state of equilibrium is reached. During this period the electric
If
now
induction between the condenser plates is diminishing and finally

reaches zero. The hypothesis of Maxwell is that the change of the
*
"A
CLV. 1864.
Dynamical Theory of the Electromagnetic Field
(11)," Phil. Trans. Vol.
243, 244]
induction produces the same magnetic

by a current of current-density
effect
as
507
would be produced
at every point of the field, which together with the current in the
wire would form a closed circuit. As the equations 222 (2) were
deduced from the magnetic

pothesis
is
necessary
we
if
of closed currents,
effect
some hyand
are to deal with unclosed currents,
Maxwell's hypothesis is justified by its remarkable consequences.

Since Maxwell calls the vector g/4?r the electrical displacement, he
3^
terms the vector
tric
the displacement current.
The consequence of Maxwell's hypothesis is that in the dielecwe must introduce the components of the displacement
1
1 8
current -rtions
( 1 1 ),
8g)
4?r dt
-^r
4?r dt
-T
33
-^-
4?r dt
in place of u,
v,
in the equa-
giving
d$
= dN_dM
dt
'
csz
dy
d$_dL_dN
(12)
dt
~dz
dj$
= dM_dL
~
dx'
dx
dt
dy'
These equations are now completely analogous to the equations

(5) except for the difference of sign on the left, the two sets being
represented by
(13)
| = curl H.
01
is
conducting, we must introduce both the
conduction and the displacement current, so that the equations are
If the dielectric
aae
~dt
33
-g
dt
dN dM
+ 4nru = ^--- -5dz
dy
+ 4>7TW =
dM
-- dL
-=
dx
dy
508
Differentiating
these
[PT. III. CH. XIII.
equations respectively by
x,
y,
z and
adding we obtain
so that the total current is solenoidal, like the flow of
an incom-
Integrating (15) through a portion of space T

a closed surface S,
pressible fluid.
bounded by
i
a /Y/v83e
=
which by
j-
That
8g)
83\
[u cos (nx)
dv ,c
fff/du
dS,
182 (17) becomes

II
(17)
pdr
{u cos (nx)
dS.
the increase of charge of any portion of space is equal

This agrees perfectly
to the electricity brought in by conduction.
with our previous conceptions. Our statement made in 129 that
is,
electricity is
not incompressible
statement that the
is
total current,
and displacement currents,
is
also reconcilable
with Maxwell's
the resultant of the conduction
like the flow of
an incompressible
fluid.
analogy between
the
By
might
call
the vector j
the
equations (5)
and (12) we
pôa
-=- the
magnetic displacement current.
Magnetic conduction-currents do not exist, although they have

been introduced into the equations by Heaviside* for the sake
of symmetry.
Complete System of Equations for Media at rest.

may now collect all the fundamental equations of the theory
245.
We
has been developed. Before doing this it will be convenient

a slight change in our units. It will be recalled that in
as
it
to
make
the whole of Part III since the introduction of the electro-
magnetic system of units we have considered all quantities,

whether electrical or magnetic, to be measured in that system.
Up to the present this has been most convenient, and in practical cases
* "
p. 441.
dealing with electro-magnetism and electro-magnetic
Electromagnetic Induction and
its
Propagation."
Electrical Papers, Vol.
i.
244, 245]
induction this will generally be true.
new
We
509
are now, however, about
phenomena, and it will be convenient to

use the Gaussian system, that is, to measure all electrical quantities
in electrostatic units, and all magnetic ones in magnetic units.
We shall therefore be obliged to reintroduce the factor A, 210,
which will multiply the electric currents, and divide the electrical
212, equations (6) and (9). Equations (14) and
forces, according to
to consider a
(5) thus
A
,
class of
become*
dN
83E
+ 4*7rAu = -=
dt
dM
-5-
dy
dz
dN
^- + 4^ = dL
^---,
.
(A)
9g)
A* 93
"7TT
*_,_?*
--- dL
d%
dt
dZ
d%
A^-=dt
dy
(B)
dz
dZ
-_-^ dm = dX
^dY dX
A d3l
-
dY
dx
dt
dy
dy
These are the equations of cross-connection between the elecand magnetic fields and thus show that in non-conductors the
curl of the force of either field determines, or is determined by,
tric
the time-variation of the induction of the other.
If
we know
the
any instant we may accordingly find it at any

For we have the three sets of equations
expressing the Fourier-Ohm laws,
state of the field at
subsequent instant.
= eX,
(C)
The
letter
where we
eF,
=/*,
(D)'3R = /Jlf,
(E)
= \X,
v=\Y,
w=\Z.
denotes the electric inductivity, which in Chapter IX,

did not distinguish electric and magnetic quantities, was
denoted by p. It will be noticed that equations (A) and (B),

which are the fundamental equations of the theory, contain no
quantities that are intrinsic to the media, but only those quantities
which completely specify the electric and magnetic state of the
The equations (C), (D), and (E), on the contrary, contain

the quantities e, /JL and X, which denote properties of the media.
These latter equations are not fundamental to the theory, as they
may under certain circumstances be replaced by others. In addition
fields.
* In
Hertz's papers the right-hand members appear with the opposite sign, since
Hertz employs the left-handed arrangement of axes. (Cf. Fig. 1.)
510
we have
[PT.
III.
CH. XIII.
the electric and magnetic energies and the dissipativity,

or heat generated per unit of time,
for
T=
(F)
(ZS
+ Mm + NW) dr,
246.
Eolotropic Media. The equations (C), (D) and (E)
have been established on the supposition that the medium is
isotropic, that is that it has the same properties in all directions
at any point. In some bodies, such as certain natural crystals, this
is
is
The assumption next
not true.
made
Chapter IX
in order of simplicity to that
assume that the energy per unit volume

a homogeneous quadratic function of the components of the field
in
is
to
the six coefficients being properties of the medium. If we then

180, we find that our results are the
apply the reasoning of
same
as before, providing that
we
define the inductions
by the
equations
=
3 =
eu
e sl
-f e 12
F+e
13
X + F+ 633^,
91
e32
where
ers
= fj^
Z,
sr
prs = psr
The inductions thus defined have all the properties that we have
It has been pointed out
hitherto predicated with regard to them.
not
the
these
are
that
only possible generalizations of
by Pupin*
the equations (C) and (D).
Media which are not
may
w
*
isotropic are called eolotropic.
body
also be eolotropic with respect to conduction, in which case
Pupin.
X 31 X +
"Studies in the Electro-magnetic Theory."
Science, Vol. L., p. 326, 1895.
American Journal of
245
247]
We
shall in the future, as
we have done
511
in the past, consider only
isotropic bodies.
Consequences of the Equations of the Field.

If we differentiate the equations (A) respecPropagation.
tively by x, y, z and add, we obtain
247.
the consequences of which we have discussed in Chapter X.

medium is an insulator, the relaxation-time is infinite, and
+ a^ + 33
dz
a|
dx
is
If the
dy
independent of the time.
Applying the same process to the equations

8
dx
(B),
we
obtain
+ a^ + a^
dz
dy
independent of the time, and the value of this divergence

except in intrinsic magnets
We
is zero,
201).
now deduce
the more important consequences of the

from
the simpler to the more complicated
equations, proceeding
cases.
We shall first, therefore, consider the phenomena in insushall
which the equations (A) and (B) are exactly symmetrical.

account of the dual nature of the relations of the two fields it
lators, in
On
follows at once that every effect of electrodynamic induction in

producing electromotive forces has an analogous effect in the pro-
duction of magnetomotive forces by electric displacement currents.

For instance a closed iron ring placed in an electrostatic field
varying with the time would become magnetized. Effects of this

sort have not yet been observed, on account of the extreme small-
by which the displacement current is multiFor the same reason, electrostatic forces produced in
plied.
insulators by the variation of magnetic fields have not been
ness of the factor
successfully observed, although the attempt has been made by

Lodge*. The justification of the equations (A) has been given
by other
*
Phil.
results.
Lodge.
Mag.
"
On an Electrostatic Field produced by varying
(5) 27, p.
469, 1889.
Magnetic Induction."
THE ELECTROMAGNETIC
512
If
we perform upon the equations
which
FIELD.
[PT. III. CH. XIII.
(13) the operation of curl,
by the result of differentiating the third of

equations (B) by y and subtracting it from the second differentiated
is
typified
O2 V"
by z we
obtain, after adding
and subtracting
-^
C/36
d /m
m\ = v d /dx dY + dz\
-- -sr
4 57(5
A^-^\^-+^r -5dt \dy
dz J
dz J
dx\dx
dy
.
(i)
Now
>
supposing the
medium
to
be homogeneous, that
is
and
constant, making use of the equations (C) and (D), and

IJL
supposing there is originally no electrification, we have
and making use of the
first
of equations (A)
we transform
into
Proceeding in like manner we obtain for the other components,
(2)
We
thus find that in insulators each component of the two

a differential equation of the form
fields satisfies
(3)
|*-<**
where
l/A
an equation of great importance in mathematical

Let us multiply
physics,
both sides of the equation by the element of volume dr and
integrate throughout the volume bounded by a closed surface
S, applying the divergence theorem to the right-hand member,
Since this
we
>
is
shall investigate its general solution.
247]
If the surface
S is
a sphere of radius r with
its
513
center at the point
P. we have
where by
(f> r
we denote the
the sphere of radius
r,
values of
at points on the surface of
with center P.
Introducing polar coordinates into the left-hand side of equation

(4) also
we may write
Now
member
The
it
this and the transformed right-hand

the
upper limit r changes our equation (3) into
(5) by
differentiating
surface integral
jj&da),
which appears on both sides is 4?r times the mean value of the
on the surface of the sphere of radius r. Calling this
function
mean value r we have the equation
(/>
<j)
which, on performing the differentiations and dividing by

written
If
we now
introduce two
new independent
u=
we
have, putting
du
dt
r,
____
du
dv
__
du dr
_
w. E.
variables
dv
~~
dr*
= at
may be
r$ r = ty,
dt
dr
at 4- r,
r,
du2
'
dv dr
dudv
'
33
THE ELECTROMAGNETIC
514
so that our equation (9)
[PT. III. CH. XIII.
becomes
(.o)
Of
FIELD.
dudv
-a
this equation the general solution is
(11)
1r
= gi(u)+g*(v),
where gt and g2 are perfectly arbitrary functions of their arguments.

We consequently have for the solution of (9),
r$r = gl
(12)
When
(at
+ r) + g
we have
= g (at)+g
1
and
- r).
(at
this being true for all values of
independent, but one
is
(at),
the functions gl} g2 are not
the negative of the other, whatever the
Putting then
value of the argument.
gi
= g,
g*=-.9>
we have
r$r = g
(13)
Differentiating
by
(14)
$r + r
(at
+ r )-g(at- r).
r,
and again putting r
^=
g' (at
= 0, we
+ r) + g' (at - r\
obtain
$r=0 =2g'(at).
(15)
mean
over the surface of a sphere of radius

r with center at P, and the mean value over a sphere of radius
itself.
zero is the value at
Accordingly
But
is
<j> r
the
value of
<
^P =
(16)
Now
differentiating (13)
(r</v)
so that
>
and
<
for
= 0,
by r and
=a
{g (at
t,
+ r)-g
(at
- r)},
247]
Now
inserting the value of
Suppose that
take
ot
= 0,
<j> r ,
for a certain initial instant, for
the values of the function
and of
<
its
which we shall
time derivative
are given as functions of a point in space,
[=o =*>,</,*),
(20)
Inserting in the equation (19)
but when r
to
515
<j)
=a
it
becomes
the value of the left-hand side
is
by (16) equal
Accordingly
we have
finally,
This solution was given by Poisson*. It shows that the value of <f>
all times may be calculated for every point
if we know the
at
mean
value of
d</>/dt
at a time earlier
by the interval at for
all
points
on the surface of a sphere of radius at about P, as well as the rate

of variation of the
mean
Suppose that
altered.
value of
initially
<
</>
as the radius of the sphere
and
^-
is
are both zero except for
a certain region whose nearest point lies at a distance r: from P

Then as long as t<rl /a
and whose farthest at a distance r2
the mean value of <f> on the sphere of radius at is zero, and after
.
t > rja as
well.
Accordingly there is no disturbance except
is propagated
between the times râ and rja, or the quantity
in all directions with the velocity a.
It may be easily shown that
P is finite if F and f are finite everywhere.
<j>
(f>
We might have obtained the same result in a more simple

manner by transforming A$ to polar coordinates in equation (3),
making independent of the angular coordinates, when the equa<j>
tion
becomes
Nouveaux Memoires de V Academic
des Sciences,
t.
in.
332
THE ELECTKOMAGNETIC FIELD.
516
Of this
a solution
[PT. III.
CH. XIII.
is
shown (12). Accordingly for all points and times for

r has the same particular value we have the same
as has been
which at
value of
r<p,
or a particular value of
<f>
travels outwards with the
The value of is inversely proportional

velocity
r traversed. The solution makes the value of
a.
(/>
<
point from which
it
This
is
is
to the distance
infinite at
the
only an apparent
propagated.
the potential due to a single mass-point
infinite at the point, but is never infinite when the mass
difficulty, for just as
is
is
continuously distributed with finite density, so here if we consider

a finite region in which <j> is not zero, the infinite value will not
occur, as
is
shown by our general
solution (22).
We
see, accordingly, that a state of electric or magnetic field

existing in any region of space has its action propagated with the
finite velocity a = l/A \lpe in all directions, and inasmuch as by
the equations (A), (B), the time-variation of one field is proportional to the curl of the other, the second term of (22) shows that
a curl of one
field of
field in
any part of space causes a propagation of a
the other kind.
The conclusion that
and magnetic actions are prothe great and remarkable

velocity
consequence of Maxwell's theory, and was enunciated by him in
1864 in his celebrated paper on the Dynamical Theory of the
From this, and the other consequences
Electromagnetic Field.
of the equations, he was led to enunciate the theory that light
was an electromagnetic phenomenon. In fact, the equation (3)
is, as we have shown, the equation of wave motion, and is the
basis of any undulatory theory, whether of light or of sound.
The manner in which the equations give us a theory suitable for
249.
light and not for sound will be discussed in
pagated with
finite
electrical
is
Since the velocity of propagation is I/ A Ve/z, in air the velocity

should be 1/A = v, or the velocity which corresponds to the ratio
Determinations of this
of the two electrical units of quantity.
purely electrical quantity, as refinements in measurement increased, gave results showing a surprising agreement with the
247, 248]
517
determinations of the velocity of light, so that many German

as the reciprocal of the
authors are accustomed to speak of
velocity of light.
and
definition of
seems preferable, however, to keep the

v purely electrical, as we have given it in 212.
It
further confirmation of the electromagnetic theory of light

was sought in the fact that the index of refraction, being inversely
proportional to the velocity, should in non-magnetic bodies, for
which
fju
1,
inductivity.
be proportional to the square root of the electric

This relation was experimentally verified for a
number of transparent dielectrics to make it appear that

the agreement was not accidental, although many exceptions were
found.
sufficient
Nevertheless, although these considerations made the electromagnetic nature of light very probable, the theory of propagation
of actual electrical disturbances with finite velocity remained
unverified
by experiment
until
1887,
when Hertz began the
publication of his remarkable researches*, which have since carried

conviction of the truth of Maxwell's theory of electricity and
magnetism to the most conservative parts of the scientific world.

For an account of them the reader is referred to Hertz's collected
papers on "Die Ausbreitung der elektrischen Kraft," or to the
English translation by D. E. Jones.
Transfer of Energy. Poyn ting's Theorem.

now form the equation of activity for any portion r of
248.
shall
If
field.
E= W+ T
be the
total energy,
H the
dissipativity,
We
the
we
have
~+H
d
(i)
Since
and p do not vary with the time we have, by
--dfdi
*
gen."
"Ueber
(C), (D),
87-
die Ausbreitungsgeschwindigkeit der elektrodynamischen WirkunWied. Ann. 34, p. 551, 1888, trans, p. 107.
518
[PT.
CH. XIII.
III.
Inserting these and the corresponding values in the integrand,

and replacing the time derivatives by the curl-components from
equations (A) and (B), we have
dr.
dy
Now
(3)
We
but
vs )
integrating by the theorem of
dxj
\dz
H={(7N-ZM)
226
Veto?
(4), this
cos (nx)
becomes
+ (ZL - XN) cos (ny)
have supposed that there are no intrinsic electromotive forces,

if there are, the components X, Y, Z in ( I ) must be replaced
X-X', Y-Y', Z-Z'
(243), except in the last integral

the
dissipativity, consequently that integral will not
representing
be entirely cancelled as above, but there will remain the term
by
representing the activity of the impressed forces, in addition to

the surface-integral. If we extend the integral in (3) to a space
to whose boundaries electric
and magnetic actions do not extend,

we have
since the integrand in the surface-integral vanishes
f + *-
(5)
as the equation of activity (cf.

64 (6)), whose
of (5) is the 2F of 64, (8).)
while
is
the present E,
If the fields are not zero at the surface S, the equation (3)
shows that the energy in the volume
will
be accounted
for
by
supposing that a quantity of energy
per unit of surface S enters the volume r in unit time.

therefore call the vector
We may
248, 249]
519
-ZM)/4>7rA,
= (ZL the energy-current-density.
The equation
(3) accordingly states that the quantity of
energy
R is
transferred per unit of time across the unit of surface tangent

the direction of both the electric and the magnetic force. This
is
Poynting's* remarkable theorem.

It has
been remarked by
Thomson, Heaviside and Hertz
J. J.
that this determination of the energy current
is
not the only
possible one, since we may add to the above vector any solenoidal
vector without changing the surface-integral in (3).
Hertz has
makes energy flow at all points where

involves the continuous flow of energy
also pointed out that, as this

fields of
both kinds
(in closed tubes)
exist, it
when a magnetic
In
exist in each other's presence.
pole and an electrified point

many cases, however, the notion
of the motion of the energy here given is a very fruitful one.

has been further developed in several papers by Wienj*.
is sometimes called the radiant vector.
vector
It
The
Plane Waves. Let us again consider a perfect insuThe equations 247 (2) are all satisfied by any function of
249.
lator.
the argument
(i)
= Ix + my + nz
(f>
=
(f)
(Ix
at,
where a = l/A Ve/4.
+ my + nz
at).
For we have
-"*<>> *-"*'<>
-'*'
A<
g~"*'w
= (Z + m + <)<" 00,
2
and therefore
-*
(3)
*
Poynting, Phil. Trans.
t Wien.
685, 1892.
2, p. 343,
1884.
"Ueber den Begriff der Localisirung der Energie." Wied. Ann. 45, p.
"Ueber die Bewegung der Kraftlinien im electromagnetischen Felde."
Wied. Ann. 46, p. 352, 1892.
THE ELECTEOMAGNETIC FIELD.
520
if
[PT, III. CH. XIII.
we take
l
(4)
+ m + ri*=l.
The
quantities X, Y, Z, L, M,
values for all points for which
lx
(5 )
N accordingly have
+ my + nz
at
the same set of
= const.
But
this is the equation of a plane whose normal has the

direction cosines I, m, n, and whose distance from the origin is
The plane is accordingly travelling in the direction of
const.
at
its
a=
normal with the velocity
\JA
The disturbance
V/x-e.
is
accordingly a plane electromagnetic wave, whatever the nature of

the function 0.
The
six functions
* = &,
(6)
Then
</>i'
F=<fc, ^=4,3,
= mfa' -
a system
of linear
For
L = ^1} M=fr,
let
JV =
8.
and (B) we have
nfa',
& = ntf - %',
and
are not independent.
inserting these in the equations (A)
(7)
<
(8)
equations to determine the ratios of
c/>"s
ijr"s.
Multiplying the equations (7) or (8) in order by

adding either set, we obtain
I,
m,
n,
and
to
the
l# + mfa' + n& = 0, %' + ra^ + w^/ = 0.

'
(9)
These are two

variable
s,
(10)
equations with
regard
integrating which gives

/^!
that
differential
+ m< + nfa = C
2
lf
Ifa
is
lX + mY + nZ=Clt lL
This shows that the component of either field resolved parallel

to the direction of propagation is constant as we travel in that
direction as well as in the plane of the wave, and is therefore
249]
constant throughout
space.
521
But such a constant
field is
not
propagated at all, and we shall therefore disregard it, and put both
Both fields are consequently perpenconstants equal to zero.
dicular to the direction of propagation.
It is for this reason that
Maxwell's theory is appropriate for an explanation of light, which,

as the phenomena of polarization show, is due to transverse undulations.
Although the forces of the two fields lie in the wave-plane,

and are constant over any particular wave-plane, it does not
follow that their directions are the same in all wave-planes, that
We shall however assume that their
is for different values of s.
directions are the same in all wave-planes, and we will call the
direction cosines of F, a lt
Such a
y1} and of H, ctj, /32 <y2
wave is said to be plane-polarized. Then we have
X=
(11)
1 <fr
F=A&
and our equations
(12)
!<'
(7)
^=7i& L = CL^, Jf=A*, N=y ^
and
(8) take the
form
= O? - wA) ^r',
y ^ &&' = O?! - nA)
-/W-(fl*-J*)if/,
A*
(14)
7a
</>',
(13)
Multiplying the equations of the

2>
first
set
respectively
by
and adding, we get

a1 o2 +
&&+7i7 = 0,
2
or the electric and magnetic forces are mutually perpendicular, as

There are
well as perpendicular to the direction of propagation.
accordingly two directions, either of which might be chosen to
define the plane of polarization, and

decide between them.
it
rests with experiment to
Squaring and adding either equations (12) or equations (13)

gives
ef*=,np.
Extracting the square root and integrating,
522
[PT.
CH. XIII.
III.
putting the integration constant equal to zero for the same reason
as before.
Accordingly the fields are
r _i/00
r
X/
'
F _&/()
'
jf_A/W
_*/()
.
MJ.
fields are
-i-V
_*/()
--
V/A
V/L6
The two
~'
*_*/(*)
'
V/4
accordingly propagated together.
Comparing the energies of unit volume we
find
O7T
or the energy
is
equally divided between the two
The radiant
VFH
vector
is
of course
fields.
in the direction of
propagation.
250.
Propagation in a Conductor. In 247 in deducing
the equations of propagation we have supposed the conductivity to
be zero. If we do not make this assumption in substituting the
value of the curl-components on the
and in
like
manner
left
of equation (i)
we
obtain
the components of both fields satisfy the
all
equation
The general
solution
of this
equation has been given by
A
complicated to be given here.
In
the
below.
case
will
be
treated
particularly interesting special
mean time we shall content ourselves with the consideration of
Boussinesq, but
is
it
too
disturbances which are harmonic functions of the time.
purpose we
shall
assume
(3)
4>
= e**U(x
when our equation becomes
(4)
The equation
(5)
y,z\
For
this
249, 250]
523
We
made the
shall
subject of a treatise by Pockels.
here consider only the case in which
depends on a single
rectangular coordinate x, the circumstances being the same all
has been
In a conducting
over each plane perpendicular to the X-axis.
k2 is complex. In metals we know nothing
dielectric the value of
regarding the value of the electric inductivity e, for whereas

electrostatic phenomena may be explained by supposing it to be
infinite, in variable
states this is far
from being the
case.
In
experiments that have been performed with electric

waves thus far the value of co has not been great enough to make
fact in all
the influence of the term containing

with that containing X. (See 206.)
the
first
appreciable in comparison
shall therefore neglect
We
term, so that our equation of propagation
is
47T^V^=Ac/>.
Gli
(6)
This
is
shall
the equation for the conduction of heat, as given by

shall consider it in some detail below ( 254), but
We
Fourier.
now
return to the consideration of the equation
(7)
in which k2
is
the pure imaginary
(8)
The
solution of equation (7) is
U=C
(9)
Since
we have Vi =
(1
+ t)/V2
+ C.e-k
*.
the value of k
is
k=
(10)
k*
ie
Accordingly the real and imaginary parts of

i (o)t
A v/27rA/xa>
furnish us with particular solutions of (6).

<
We
thus obtain
= e -A*2ir\p.x cos (at- A \27rX/-t&> x),

= g-WsTAM". x s n (^ _ ^ V27rVo> #X
= eA N/2irAM * cos (cot + A \/27rXftw x),
.
x).
524
III.
[PT.
CH. XIII.
damped
d
two, being of the form e f(ccat), represent
waves travelling in the direction of increasing x with the
velocity
Of
these the
-r/v/
The
when x
values
its
first
periodic factor at
any time repeats
increased by the distance
is
A
A V/
r
\IJLCI)
which
called the
is
wave-length, the frequency being
An
n=
o)/27r.
^^' x which causes the

The damping factor e~
amplitude of
the wave to decrease in geometrical progression as the distance
travelled increases in arithmetical, has the relaxation-distance, or
the distance in which the amplitude diminishes in the ratio \\e
The
A V 2-77X^0)
two particular solutions represent waves travelling in the

with the same velocity and damping.
direction
opposite
last
Since the velocity depends on the frequency, there is no

definite velocity of propagation in a conductor. On account of the
damping, harmonic disturbances of high frequency rapidly die out,
consequently alternating fields penetrate but a short distance into

This was shown by Maxwell*, but its importance
conductors.
was brought out by the researches of

and Hertz
Heaviside*)*,
Lord RayleighJ
We
shall
now
consider the relations between the two
If the components
X,
complex constants
t+}cx
,
Y, Z, L,
lt
2)
M,
3)
N are
equal respectively to the

3) each multiplied
by
B B B
l}
inserting in the equations (A) and (B)

47T.AXJ.!
(1
4t7rA\A 2
2)
= 0,
= - kB
fields.
we
obtain
- AfîcoB^ = 0,
= kA
AfjLicoB
B = kA
2
3,
2.
Eliminating A 2/B3 or A 3 /B2 we obtain the value for k already

We thus see that the directional relations of the fields
found.
*
Treatise, Art. 689.
+ " The Induction of Currents in Cores."
Electrician, 1884.
Papers, Vol.
i.,
p. 353.
"
On
the Self-induction and Eesistance of Straight Conductors."
Phil.
Mag.
21, p. 381, 1886.
"Ueber
p.
395, 1889.
die Fortleitung elektrischer

Jones's trans, p. 160.
Wellen durch Drahte."
Wied. Ann. 37,
250, 251]
and the direction of propagation are the same as in

the ratio of the two fields is
525
insulators,
but
H
The magnetic field accordingly lags in phase by one-eighth of a
period behind the electric, while in an insulator the fields have the
same phase.
Reflection of
251.
Waves by a Conductor.
We
shall
now consider the effect of a train of plane waves in an insulator

striking the plane surface of a conductor which is parallel to their
We
shall suppose the conductor to extend to infinity in

plane.
one direction. Let us take the plane #
as the face of the
Let the waves be harmonic in the insulator, for which

and let the electric force be parallel to the F-axis, the
conductor.
x<
0,
magnetic to the Z-axis.

ductor we have
Then
in the
wave approaching the con-
V/x.
In the conductor we have
(2)
At the plane x = the boundary conditions to be satisfied are

that the tangential components of both forces and the normal
components of both inductions are continuous. The latter components being zero we have only the first two conditions to satisfy.
There are not enough constants to enable us to satisfy them both,
it is accordingly necessary to add to the terms representing the
disturbance in
the insulator other terms representing a wave

We
direction, or a reflected wave.
travelling in the opposite

therefore take
F=
()
v-H) + ^_>K)
Ve
V,
jf-.a/'HLw/K).
V/J,
VfJL
526
in
the insulator.
Our boundary
[PT.
III.
CH. XIII.
conditions are then, dividing
by e",
(4)
(1-t).
we have
Accordingly
(1-t)
c
2i7T\Ui
/-
A/
V
(5)
- (1 - l)
,~
.-.
COLLi
(1
t)
Since these two ratios are complex, at the surface of the conis a difference of phase between the direct, reflected
ductor there
As we
and transmitted waves.

conductor the ratio
C /C
/
increase the conductivity of the

1, in which case
approaches the value
the electric force vanishes at the boundary, which is a node for the
electric field, while the magnetic field is a maximum.
On the
other hand, as
we
increase the frequency of the oscillation or the

magnetic inductivity of the conductor the ratio C//Ci approaches
+ 1, or the magnetic force vanishes at the boundary, while the
electric is a
maximum.
If
we put
we find for the ratios of the amplitudes of the reflected and trans
mitted to that of the direct waves
'
r'J
0:
Ve
Accordingly whether r be very great or very small the reflected

waves tend to become as great as the direct, while the larger r the
less is
the magnitude of the transmitted waves.
251, 252]
527
The experiments of Hertz* confirmed the above results, the

boundary of the conductor being more nearly a node for the electric
than for the magnetic force. If the amplitude of the reflected
waves approaches that of the direct ones, the two systems will
interfere
and produce a
set of standing waves, with nodes at regular
distances from the conductor.
This
field
was explored by Hertz
by means of a
resonator, composed of a circle of wire with its ends

in
two small balls near together, constituting a conterminating
This
denser.
system has a certain period of its own, and what has
It was tuned to the period of
240 applies to it.
the waves, and being placed anywhere in the field would have
currents induced in it by the harmonic electromotive forces of
been said in
the
field.
Thus where the
force is a
maximum
sparks appear between the

For the further
balls of the resonator, disappearing at the nodes.
theory of the resonator the reader

Oscillations Electriques, J. J.
tricity
is
referred to Poincare, Les
Thomson, Recent Researches in Elec-
and Magnetism, and Drude, Physik des Aethers.
252.
Spherical Oscillator. We have hitherto considered
waves in insulators, without considering how they are produced.
In the experiments of Hertz the waves were produced by disturb-
ing the charges in a peculiar dumb-bell-shaped conductor, and

allowing oscillations to set in, which were propagated outward
through the air.
satisfactory theory of the oscillations in Hertz's
We may state the general problem

a
in
given
charge disposed
any manner not in equilibrium upon a
conductor of any form, find the nature of the oscillations that
oscillator
has not been given.
is settling down to its state of equiThis

librium, together with the fields to which they give rise.
problem is a very complicated one, and has been solved for very
ensue while the conductor
few
cases.
We
shall consider the case of a conducting sphere,
the charge upon which
is that induced by placing the sphere in a

uniform
electric
The field is supposed suddenly
field.
steady
and
the
oscillates until equilibrium is
then
destroyed,
charge
attained.
We
*
shall suppose e
=p =
1.
"Ueber elektrodynamische Wellen im Luftraume und deren Reflexion." Wied.
Ann. 34,
p. 609, 1888.
528
Since both
and
since
fields are
[PT. III. CH. XIII.
propagated according to the equation
any derivative of a solution
is
also a solution,
we may
take
which
(B).
and the equations (A) and

and t, we have for
satisfy the solenoidal condition
If
we assume
<j>
to be a function only of r
a diverging wave
<
Differentiating
dr
d
dz
drdz
by the coordinates we obtain
8(/>
drr
'
(3)
dr \r dr) dy
dr \r dr) r
8r \r 8r
~
8^2
BJ?
dr~dr
'
\r dr
r dr
'
so that the forces are
Y _d_ /19\
~8rVr8rJ
drrdr
IV
<->
Vy
r
1
dr \r dr) r
The
field is
'
'
r dr
thus the resultant of two parts, the
*dr(rdr')'
first,
equal to
252]
529
and the second, equal to
parallel to the radius,
At the
parallel to the Z-axis.
surface of the sphere, r
= R,
if
the
conductivity is large, the lines of force are normal to the surface,

so that this second component vanishes, and we have
dr 2
r dr
that
is
= 0.
r Or_| r=jR
When t = 0,
which
is,
by
the electric forces are derivable from a potential,
(i),
the potential
is,
equal to
==
But by
0).
194 (7)
in the case supposed, proportional to
r3
Consequently
(since A</>
.-(!),
dz\rj
initially
=
</>
Introducing the value of

<6)
Consequently the
<f>
(r>R)
from (2) this gives
/<)- a
function / is constant
ment greater than R.
Hence the value
for all values of its
of
argu-
<f>,
remains equal to C/r so long as r

at>R, and the field remains
=
When
t
the
wave arrives at the distance r,
(r
unchanged.
R)/a,
and to determine the field at subsequent instants we must deter-
mine the values of/ for values of

Let us make use of equation
Differentiating
<f>
8_^
w.
E.
less
than R.
(5).
_/>r
f(r-at)
r2
dr*~
argument
by r gives
80 = f'(r-at)
dr
its
at)
2f'(r-at)
2f(r-at)
r2
r3
34
530
and inserting these in the equation
f"(R-at)
is
an ordinary
~^~
R
CH. XIII.
f(B-at)
;
~~W~
we put
differential equation, which, if
u=
III.
we have
f'(R-at)
~BT
This
(5)
[PT.
at,
becomes
(o)
Rdu
du*
This has the solution f=
e*
where
-x +
CO
From
this
we
-.
obtain the general solution
f(u)
= e*R (A cos
r-at
gftU
(^
(r
damped harmonic
1
-^ wj
./o
J^
cos
representing a
+ B sin
-at)
+B
spherical
sin
wave
gj
(r
at))
of wave-length
= 4-7T.R/V3 = 7-255.R.
The logarithmic 4ecrement
is
7T/V3
1-814,
so that the oscillation almost ceases after a complete vibration.

This extreme damping is due to the radiation of the energy, and
not at
all
to the dissipation in the conductor, of
which we have
taken no account.
The nature
of the field radiated
by an
oscillation of this sort
has been discussed by Hertz*, making certain assumptions.
The preceding problem
is
the simplest that can be proposed
The
to represent a practical case of oscillations in a conductor.
It is evident, from
above demonstration is given by Poincare.
the investigation of oscillations in the last chapter, that a system
has as many possible periods as there are degrees of freedom. In
a conductor of three dimensions the currents have an infinite
*
Hertz.
"Die Krafte der
Maxwell'schen Theorie."
elektrischen
Wied. Ann. 36,
Schwingungen, behandelt nach der
p. 1, 1889.
Translation, p. 137.
252, 253]
531
of degrees of freedom, and there are an 'infinite number of

The above problem corresponds to the lowest possible
periods.
frequency for a sphere, when the surface -density of the electrifica-
number
a zonal surface-harmonic of degree one. For the general

treatment of oscillations in spheres and cylinders the reader is
tion
is
referred to J. J.
Thomson, Recent Researches in
Electricity
and
Magnetism.
Waves on
253.
Wires.
We
now come
to
what
is
the most important practical problem connected with
perhaps
electrical
waves, namely their propagation along wires, for upon this question
depends the theory of telegraphy and telephony. The subject has
been treated in great detail by Heaviside*, to whose papers the

reader
is referred.
We shall suppose that the direct and return conductors are either
cylindrical wires parallel to the .X-axis, or concentric tubes.
Let
be the sum of the resistances of the two wires per unit of length.
Let
be the capacity of the pair of conductors per unit of length,
L their self-inductance per unit of length. Let the total current
in one wire be / and the difference of potential between points
on the two wires having the same ^-coordinate be V. All these
quantities are supposed measured in the electromagnetic system.
We may
describe
the
phenomenon
as
follows.
When an
electromotive force is applied between any two corresponding

points on the wires, say by connecting them with the poles of a
battery, electricity of opposite signs flows out upon the surfaces
of the two wires, producing an electric field in the surrounding
space.
The
electrifications
then move along the wires, causing a
Both these fields imcurrent, thus producing a magnetic field.

to
into
the
conductor, and are there
mediately begin
penetrate
dissipated into heat.
As the
electric field,
whose
lines are in the
planes perpendicular to the conductors, rises from zero, it gives

rise to displacement currents in the planes perpendicular to the
conductors.
we
The magnetic
effect of these
displacement currents
ignore in comparison with those of the conduction

currents in the wire.
shall also ignore the penetration of the
shall
We
currents into the conductors, the theory of which would lead us

too far for the present purpose.
This we may safely do if the
*
Heaviside.
Wires."
" Contributions to the

Theory of the Propagation of Current in
Papers, xx. et al.
342
532
III.
[PT.
CH. XIII.
wires are small enough, or the tubes thin enough, or in any case
if the conductivity is great enough.
Ignoring the manner of
distribution of the current, then,
rent
point,
in the wire at
and
is,
like V,
any
point.
we
consider only the total curis variable from point to
This
a function of x and
t.
We
shall
suppose
are exactly symmetrical in the two conductors

as far as the currents go, so that / has equal values with opposite
that the
phenomena
signs at corresponding points in the two conductors.
Let us consider the charge that exists at any instant on the

dx.
Since
portion of one conductor between the points x and x
-\-
the capacity per unit length

we have for the charge dq,
is
K, the difference of potential V,
dq=VKdx.
If the current flowing in the positive direction at
x + dx
is /,
that at
is
so that the total gain of charge of the element in unit time
is
a/,
We
accordingly obtain the differential equation connecting the

current and difference of potential,
a/
^dv
>-*'
<x>
Considering now
the flow of the current, we have in the pair of

of the two wires the electrostatic electroelements
corresponding
motive force
dx and the electromotive
^v-
force of induction
dx
the resistance being Rdx.

Ldx-^-,
ot
equation
Accordingly the current
is
a/
The equations
Differentiating (i)
(3)
and (2) are the equations of the problem.

by t and (2) by x we may eliminate /, obtaining
(i)
253]
Differentiating (i)
by x and
(2)
by
we
533
eliminate F, obtaining
2
(4)
Consequently both the current and the potential are propagated

in accordance with the equation
be observed, is of the same form as the equation

the resistance of the wires here taking the place of the
We shall, with Poincare, refer to
conductivity of the medium.
which, as
250
it will
(2),
the equation (5) as the Telegraphic Equation.
rounded by a concentric tube, we have
for
while
If the wire
sur-
is
144), inserting the factor
electromagnetic measure,
if
the currents be supposed concentrated in the adjacent

we have by 234 (21), since r12 = r&,
surfaces of the tubes,
Consequently the coefficient of the
first
KL = e/^
term
is
2
.
we have two
wires of the same diameter, which is small with

159 (29) becomes
respect to their distance apart, the formula
If
while the formula of
so that
234 (22) gives
we have the same
relation as before.
This relation
is
of
course not accidental, depending upon the similar equations

satisfied by the electric and magnetic fields between the conductors,
The only reason for

which are brought out in
234.
in obtaining L the
deviation
from
is
that
the
above
relation
any
the
current density was supposed uniform, while in obtaining
surface density was not.
The theory of electromagnetic waves
relations
534
in free space
however shows us that in
all
cases
[PT. III. CH. XIII.
we must have the
-2fL
coefficient of
-^
ct
equal to epA*.
If accordingly the resistance of the wires is negligible, the

disturbances are propagated with the speed l/A^/e/ji or in air
with the velocity l/A=v.
The determination of this velocity
and the comparison of it with the results of determinations of

the ratio of the two units v is thus a matter of great importance.
Determinations of the velocity have been made by Blondlot*, by
Trowbridge and Duanef, and by SaundersJ, under the direction
of the author.
These have given a satisfactory agreement with
theory, and are the only direct experimental verifications, for it is
to be noticed that in the determinations
of Hertz, based upon

measurements of wave-lengths, the frequency was not observed,
but calculated, thus virtually begging the question.
The result just announced, and the statement that the velocity
of the waves should be approximately that of light, were contained
in a paper published by Kirchhoff in 1857, and were afterwards
rediscovered by Heavisid,e||.
The equations (i) and (2) were given
in 1876, and both researches remained singularly
unnoticed until recently.
by Heaviside||
254.
Particular Case.
Submarine cable.
The
first
case
of the telegraphic equation to be treated was that in which the

self-inductance of the circuit is negligible in comparison with its
This is the case in a submarine cable, in which the
resistance.
surrounded by a concentric tube of water, separated from

a
thin layer of dielectric. We may then neglect the first
by
term of (5), making use of the equation
wire
is
it
<
*
Blondlot.
Comptes Rend. 117, p. 543, 1893.
t Trowbridge and Duane. " The Velocity of Electric Waves."
Phil.
Mag.
(5),
40, p. 211, 1895.
Saunders.
"
On
the Velocity of Electrical Waves."
Physical Review,
(4), 2,
p. 81, 1896.
Kirchhoff.
||
p. 53.
"Ueber
die
Bewegung
der Electricitat in Drahten."
Ges. Abh. p. 131.

the Extra Current."
Heaviside.
Bd. 100, 1857
Pogg. Ann.
"On
Phil.
Mag. Aug. 1876; Papers,
Vol.
i.
253, 254]
535
and obtaining the current from the equation
dv
We
thus arrive at the so-called electrostatic theory of propagation,

"
given in 1855 by Lord Kelvin in his paper On the Theory of the
Electric Telegraph*," which established
bility of an Atlantic cable.
beyond question the
feasi-
As we have seen in 250, harmonic disturbances are propagated

with a velocity proportional to the square root of the frequency.
There is therefore no definite velocity of propagation in a cable,
and there
of signals mixing with each other and losing

are however more concerned with the question
is liability
We
their character.
If
we
factors
we
how a
single arbitrary short disturbance is propagated.

a
cable with different constants, for which
consider
of
by changing the
may make one
variables
by multiplying by constant
solution do for both.
= ax,
If
t'
we put
bt,
we have
K'R dV' ~
=
b
so that
dt
1 9 2 F'
'
a?
V'(x',t)=V(x,t\
*-K'K
"
KR
if
a2
that
dx*
'
is
Accordingly the time necessary to produce a given potential at a

distance x from the origin is proportional to
multiplied by
the square of the distance.
quote Lord Kelvin's words
KR
We
"We may
be sure beforehand that the American telegraph

with a battery sufficient to give a sensible current
at the remote end, when kept long enough in action; but the
time required for each deflection will be sixteen times as long as
will succeed,
would be with a wire a quarter of the length, such,

*
Proc. Roy. Soc.,
May
1855
for instance,
Math, and Phys. Papers, Vol. n.
p. 61.
536
[PT.
III.
CH. XIII.
French submarine telegraph to Sardinia and Africa.

One very important result is, that by increasing the diameter of
the wire and of the gutta-percha covering in proportion to the
whole length, the distinctness of utterance will be kept constant
as in the
for
varies
the
inversely as
diameter, and
square ,of the
(the electrostatical capacity of the unit of length) is unchanged

when the diameters of the wire and the covering are altered in
the same proportion."

"
"
(The so-called J72-Law has been applied to the theory of
telephony on long-distance land-lines, to which it is not at all
The use made of
applicable, as has been shown by Heaviside.
this law by the chief electrician of the English telegraphs would
have prevented long distance telephony in England, even had there
been any long distances.)
Guided by the conclusion announced above, we shall insert a

new variable, u x\*ft, and attempt to satisfy the equation (6) by
a function of u alone.
We
have
d_V =
~
dVdu =
IdVx^
du dt~
2 du
dt
vr"w9*'
mt W}L
~
du dx
dx
so that our equation
du v*
'
*Jt*
'
becomes
KR dV
d*V
Ê'"3*
(9)
or
d
<
The
integral of this equation
dV
KR
dV
/.
IO >
KR
is
"
or,
integrating from
to
= xj^Jt,
rx/4~t
(12)
V=C\J
KR
e~-T
dV =
254]
If
we now
write
u instead
of
u ^KR/2
537
this is
X.IKR
F=(
(13)
This definite integral

therefore of
x and
the value of the integral

to V,
we
will
a function of
is
its
upper
For x >
satisfying equation (6).
t,
is
As we may add any
VTT/S*.
and
limit,
and
constant
put
xjKR
*
F=F (l-4=P
(14)
e- u *du
VTT./O
Thus
for
x>
and
the definite integral
F= 0.
= 0,
For x =
zero, so that
is
F= F
0, t
.
> 0,
the value of
Consequently the
solution (14) represents the result of connecting one end of the

cable with a constant battery, and leaving it permanently connected.
The
definite integral in (14) is the transcendent
known
as the
which numerical tables have been calprobability-integral,

From these the values of F have been plotted, showing
culated.
for
the potential at the different points on the cable, Fig. 96, the
different curves being for times 1, 2, 3, 4, 5 times KR. It is to be
noticed that however small the interval of time from the instant of
connecting the battery, the disturbance is felt somewhat at all

Thus the velocity of propagation would
points, however remote.
be infinite, if we could speak of a velocity. This shows that the
*
This
may
We
be shown as follows.
have
J= fê'^dx^
Jo
Jo
Consequently
J2 =
{"'
Jo Jo
Changing to polar coordinates,
"
Therefore
J=* -=
2
Jo
e~ u2 du.
THE ELECTROMAGNETIC
538
FIELD.
[PT. III. CH. XIII.
theory is only an approximation, for it is hardly imaginable that

the velocity should be greater than in free space.
FIG. 96.
The current
with regard to
by
its
upper
is
x.
obtained, according to (7), by differentiating

must therefore differentiate the integral
We
limit,
multiplying
this
by
Accordingly
OS)
V5
=F
FIG. 97.
its
derivative
by
x.
254]
The values
Fig. 97
for
of the current for different points are
times
^, J,
-J-,
1,
4 times
when
When x
KR.
IT
so that
539
shown
in
we have
the value of the current
is
infinite, instan-
taneously.
The
x=
is
the potential at any given point other than

shown by the outside curve in Fig. 98, taken from Lord
rise
of
Fm.
98.
Kelvin's paper, the abscissas representing the time, the ordinates
the potential.
The potential rises asymptotically to its value at the origin, but

the current rises to a maximum, which occurs at the time
t
KRa?/2, and then dies away to
zero.
The maximum values
of
and
to
are inversely proportional to the distance from the origin,

the resistance R.
Since our differential equations are linear, disturbances due to

merely added. If the battery is connected, instead of permanently, for a definite time T, and the
different initial states are
cable then put to earth, the effect
ing case, after

at the origin.
will
be V(t)
a time r
is
the same as
if,
we permanently apply the
in the preced-
potential
If the preceding solution be called V(t) the present
V(t
T),
consequently
we may obtain
the graphical
540
[PT.
III.
CH. XIII.
representation by taking the difference of ordinates of the outside

curve in Fig. 98 and the same curve pushed to the right the
distance
T.
The other curves
x when the battery

KRx*.
at
is
in Fig.
98 represent the potential

1, 2, 3, 4, 5, 6, 7, times
applied for times
Since any derivative of a solution of (6)

by x is also a solution, and
is
a solution, the
derivative of (14)
represents the result of instantaneously connecting a battery and

then insulating the end of the cable. The distribution at any
is of course shown in
Fig. 97,
at the origin, the total charge
time
and while
V is
initially infinite
30
q
is finite,
255.
=K
f
I
Jo
Vdx = qQ
and remains constant throughout.
General
case
of
Telegraphic
Equation.
The
telegraphic equation (5) has been treated byHeaviside, Poincare*,

Picard (-, and BoussinesqJ. We shall give the solution of Boussinesq, not only because he has given the general solution of the
more general equation 250 (2), but because his method obtains
the solution by an ingenious artifice from Poisson's solution 247
(22), and the knowledge of other methods required by the processes
of Poincare and Picard is unnecessary.
Let us put
so that our equation
* Poincare.
is
" Sur la
propagation de 1'electricite."
Comptes Rendus, 117,
p.
1027, 1893.
t Picard.
" Sur
1'equation aux derives partielles qui se rencontre dans la
thôrie de la propagation de 1'electricite."
Comptes Rendus, 118, p. 16, 1894.
"Integration de 1'equation du son pour un fluide indefini."
J Boussinesq.
Comptes Rendus, 118, p. 162, 1894.
254, 255]
and
is
541
us suppose that initially the state of the line is given, that

the potential and current are given at all points, by
let
V=F(x), /-<?(*),
1
137
dV
(2)
Let us transform the equation by putting
V=e**u.
(3)
Accordingly
so that if
The
we put p =
b the equation
initial conditions
The method
becomes
now become
of Boussinesq
may be
applied to the more general
equation
which we shall accordingly consider, putting whenever we choose

u independent of y. The artifice employed is the introduction
into the function u of one more degree of freedom, by making it
depend upon another parameter z which is finally to be given any
constant value
we
please.
Let then u
satisfy the auxiliary equa-
tion
(8)
&-*>
so that the equation (7) becomes, taking this into account,
542
We
will
suppose that the
u=
[PT. III.
initial conditions are

(a?,
y,
CH. XIII.
now,
z\
du
(10)
S -*<***
Then the
solution of (9)
247
by
is,
(22), inserting explicitly
the rectangular coordinates and the direction cosines cos

cos 7, of the radius r in the functions <& at} <f) at
a,
cos ft
-i
(i i)
^ (a?,
j-g-
+v
The equation
y, s)
j$(0+ot
<
(a?
cos
a,
y 4- a cos ft2 + at cos 7)
+ at cos a, y + at cos ft 2 + at cosy)dc0.
(8) holds for all values of
t.
Solutions of
(a,
y,z)
(x*
V>z}
eta
F (x, y} cos
it
are
9J
+ H (x,
y) sin
y) sin
z,
(12)
<t>
and
for z
=9
<X y) cos
these reduce to <&
have the proper values of
u,
^vt
(x,
LI
= F,
We
-z
a
<#>=/, so that for
we
therefore insert the values
Now as we integrate over the whole

(12) in the integral (u).
sphere, the sine terms, being odd functions of z, disappear, while
the cosine terms, being even functions, give us double the value
that
we should
get by integrating over the hemisphere for which
Accordingly, giving z the constant value zero,

1
(13)
= 5-
r r
7)
,r-
ZTT vt J J
tF(x + at cos a, y + at cos @) cos
+ 2~ r ^ ^ + at cos a>
+ at cos ^^ cos
(ibt cos
7) dco
cos
In the telegraphic equation .F and g are independent of y, and

are therefore constant on all small circles of the sphere normal to
We
will therefore employ polar coordinates, a the

r with the JT-axis, and % the angle that the plane
of r and the X-axis makes with the XZ'-plane.
Then
the X-axis.
angle made by
cos
dco
= sin a cos ^,
= sin
255]
543
and
i
[* r*
o~^l
ir
2
JT
r* r*
t
"*"
2~~
~ ir
y( a) + at cos a ) cos (Mt sm
Let us put
The
sin a
cos
J!
definite integral
7T
(16)
is
(f)
=-
one of the set
(17)
which may be evaluated by developing the cosine into a

and integrating each term. We thus obtain
(is)
series
w-srj$*
2
an
infinite power-series in
f,
the coefficient of f 2 ? being
(2)!
where
(19)
7 =
.,,,
"
an integral which we can evaluate.

the integrand (sin 2^
% cos %)
Integrating by parts, writing
cos 2 ?" 1 %,
~ Cos2x) C0s2(s~
1'
xrfx
THE ELECTROMAGNETIC
544
FIELD.
[PT. III.
CH. XIII.
from which follows the reduction-formula

T
pq
~ 2g-
'
-^
Integrating (19) again by parts, writing the integrand

sin
giving the reduction-formula,

(23)
I,,'
By <? successive applications of the reduction-formula (21)

of the formula (23) we get
and
^
But we have
Accordingly, introducing even factors into
numerator and
denominator,
and our integral

(26)
If
we multiply
(27)
7)
(to
becomes
-~ "
AJ
this
by^
"
'
,.o
th<3
255]
is
545
called the Bessel's function of order p, and tables of its values

for real and pure imaginary values of the
have been calculated
argument
?*.
Our
integral (16)
*/o(f )
and (15)
=l
1 d
{*
t
= J (ibt sin a)
F (x + at cos a) J
1 F*
o I tg(x
*
Jo
The
therefore equal to
is
u=
(28)
is
+ atcosa) J
(ibt sin a) sin.
adz
(ibt sin a) sin ada.
Bessel's function i~*Jp (ix) of a pure imaginary
argument
usually denoted by Ip (x) (not the Ip of the preceding).

then
X2/a2
is
Putting
our solution becomes

(29)
-X /a
2
Performing the differentiation by
^â Jf_at
(*
dX.
26), since
^ + X)
(30)
7 (0)=
+ X) /
This solution was obtained by Heavisidef and by Poincare, by

entirely different processes.
We
now suppose
that initially there is no current in the

and that the potential is zero except between two points
line,
shall
<
That
asj.
is
F (x) = 0,
except
when
x <
^<
x.2 ,
for all #.
*
See Gray and Mathews, Treatise on Bessel Functions.

"The General Solution of Maxwell's Electromagnetic Equations in a
geneous Isotropic Medium." Phil. Mag. Jan. 1889, p. 30 Papers, Vol. n.
Homop. 478,
eq. (40).
w.
E.
35
THE ELECTROMAGNETIC
546
Then
FIELD.
[PT. III.
CH. XIII.
for
>
a?2
at or
x >
xl
at
all the functions

is zero.
occurring have the value zero, therefore
This at once distinguishes the solution from that for the case of
the cable, for the disturbance does not arrive at x until the time
=^
or
-on
When
the right,
on the
xl > at >
left.
#2
so
we have
(32)
ra
a Jx-x rfc-X)(*+)J.dX
\
vtj
for
This represents the disturbance while the wave is passing over,

In like manner for a point on the left,
a point on the right.
for
#x
/>
&
-x<
at
< #2
#,
rat
(33)
Finally, at later instants,
<
a?!
at or
a?2
a;
<
This represents the disturbance after the wave, travelling with

Accordingly the solution, while repre-
velocity a, has passed on.
senting a wave travelling with the velocity a, as in free space,

from that case in that there remains a residue, or tail to the
differs
wave, which does not

exponential factor
proper, and in the
fall
to zero
however great the time.
The
shows that the disturbance, both in the wave

tail, is
continually becoming attenuated.
Thus when
a
tail,
successive impulses are transmitted, each leaves

which interferes with all the succeeding waves, and the
possibility of telephonic speaking depends not only on the attenuation and distortion with the distance, but on the magnitude of
The tail also explains the discrepancies

tail of the wave.
that existed between the results of the attempts made to determine
the
255]
547
the velocity of electric waves by means of telegraph lines, what

was generally observed being more probably the maximum dis-
turbance than the front of the wave.
and
In order to give a concrete idea of the nature of the propagation,

to afford a means of comparison with the electrostatic theory,
and equal to F
time and length,
by taking the relaxation-time r = 1/6 for the unit of time, and the
relaxation distance, d = CLT
a/b, as the unit of length. Accordingly
we
is constant
shall suppose that the function
shall
also
our
x.
units of
to
2
#j
change
from
putting
,
We
---t>t,
#2
x-#
_
=
b (x
-#)
b\
-,
-j-
-=^,
we have
7-
(35)
a point on the right while the wave

equation was given by Heaviside in 1888,
for
This
passing over.
is
who
carefully refrained
from giving his method of deduction, remarking " since, although

they were very laboriously worked out by myself, yet as mathematical solutions, are more likely to have been given before in
some other physical problem than
to
be new*."
Inasmuch as not only Heaviside's results but any others were

overlooked by the three French mathematicians quoted, who
published results six years later, we may conclude that in the
English writer modesty and original productiveness were more
strongly developed than historical research. (This modesty is not
maintained on the same plane throughout.)

Inserting in the value of
V the series for /
(dropping accents),
and developing each term by the binomial theorem, we obtain
"Electromagnetic Waves."
Phil.
Mag.
1888;
Papers,
Vol.
n.
p.
eq. (52).
352
373,
THE ELECTROMAGNETIC
548
FIELD.
[PT.
III.
CH. XIII.
Changing the order of summation and putting
_
!.&--
(38)
q=q'+p,
p=oog' = oo
which by the definition of Ip *
Differentiating by
'P
equal to
is
and adding
to itself,
which by the formula connecting the derivative of Ip with the

contiguous function
Ip +i,
[Gray and Mathews (141)]
may be
written
2'j.
Inserting in the integral (35), and integrating from x to
(42)
+/p+1 (on
^
The terms
free
from x may, by writing out the sums

shown to be equal to e*.
for
Ip and
,
collecting terms, be
therefore, finally obtain for
We,
"Po
_ f
t
vl.3.5...2p-l
!/y(0
t>
i(0}
H-
From this the values of F have been calculated and plotted by

Mr W. P. Boynton in Fig. 99, which shows the distribution of
potential along the line to the right, for times
*
From
(27)
we have
for the present
^<n=n- j,
Ip
(f),
)=(!)"
1, 2, 3, 4,
()".
5 times
255]
the relaxation-time.
This
the electrostatic theory.
may be compared with
The
549
Fig. 96
showing
rise of potential at particular points,
FIG. 99.
as a function of the time,
is
shown
in Fig. 100, which is the

The different curves
analogue of the outer curve in Fig. 98.
FIG. 100.
for
from the
the points at
1,
2, 3,
4 times the relaxation-distance
start.
The
potential produced by connecting the battery for a definite

and
then removing it may, as before, be obtained by taking
time,
the differences of two curves relatively displaced. In this way the
effect of the initial potential shown by the rectangle in a is shown
in
6, c, d, e,
Fig. 101 for the times
'2, '4, *6,
1'6 times the relaxation-
550
time, the dotted lines showing the
no
resistance.
The
wave as
last figure,/, Fig. 101,
ening the duration of a signal, the
tail left
it
[PT.
would be
shows the
III.
if
CH. XIII.
there were
effect of short-
being noticeably smaller.
n
x
FIG. 101.
From
we may obtain an idea of the distance

possible, if we know the relaxation-distance
these figures
which telephony
is
to
of
With ordinary land-lines the relaxation-distance

line.
the order of several hundred kilometers. This has made speaking
the
is
of
possible between Boston and Chicago. Obviously it is of importance

v 2L/R as great as possible
to make the relaxation-distance d
wires great, and using large
between
the
distance
the
by making
copper wires.
256.
In the preceding examples we

be without end. In many practical
Terminal Conditions.
have considered the
line to
255, 256]
551
we wish to know what goes on in a line of finite length, when

the ends are connected to any electromagnetic systems whatever,
both when the systems are left to themselves and when electrocases
motive-forces are applied.
is
Space
lacking for more than the
briefest possible treatment of this matter,

treated in Heaviside's papers on wires to
which
is
very fully
which reference has
already been made.
The method
We
shall
of procedure
make use
"
is
the same in every case.
of the equations
d?
rr dV
87
(2)
dt
OL
ViK
Let us seek particular solutions of the form
(3)
Inserting in (i)
and
if
we have
we put
(5)
we have the equation
for u,
(6)
The
solution of this
is
u=A
(7)
where
and
cos
fjjc
4-
B sin fix,
are constants to be determined.
From
(2)
we
obtain
-T-
= K \u = K\ (A
w
The
(B
cos
/JUG
cosfjuv
+ B sin
A sin
functions (3) are solutions of the differential equations
whatever the value of X. The values of X that are admissible are

determined by the terminal conditions. We shall take as an
example one of the simplest cases possible. Let us suppose that
at one end,
where x =
I,
the two wires are connected, while at the
THE ELECTROMAGNETIC
552
FIELD.
[PT.
III.
CH. XIII.
other they are connected with the plates of a condenser of capacity

The conditions are then
x = l,
F=0,
Applying these to the solutions
cos
fjil
(3), (7), (8),
we have
+ B sin pi = 0,
< IO)
A/B we
Eliminating
obtain
rr
/Ktan/d=^r,
-&0
(I I)
a transcendental equation to determine p. This has an infinite

of roots, which may be real or complex.
When these are
number
determined,
Thus we
is
determined
find that there are
each root by the equation
for
an
(5).
number
of possible periods
for the free vibration, corresponding to the n periods for a system
with n degrees of freedom. The equation (n) corresponds to the
infinite
determinantal equation of 241 (10). The ratio B/A is determined

by (10). The determination of the absolute value of the coefficients
depends on the
Having found an
root
p s being
initial conditions.
number
infinite
distinguished by
F= 2eV (A
(12)
where we sum
for all
the roots.
of particular solutions, any

the general solution is
its suffix,
cos
+B
fi 8 o!
sin
fi s oo),
If the potential
is initially
given by
V=F(x), = Q,
t
(13)
we must have
F(x)
(14)
The problem
to
= 2, (A
cos
be solved
fji s
x + Bs sin ps x).
is
then that of developing an
arbitrary function of x in a trigonometric series of the form (14)

where the fig's are the roots of a certain transcendental equation,
namely (n). The problem is in general of considerable complexity,
who
and we
shall content ourselves
has treated
If there
equation
(n)
is
it
with referring to Heaviside,
at great length.
no condenser at # =
is
tan pi
oo
0,
but the circuit
is
open,
256]
553
and the roots are

IT
The
series (14)
then becomes
(15)
= we see
a Fourier's series, with the even terms omitted.
If
X
that
a
is
so
all
that
the
oscillations
are
by (5)
pure imaginary,
harmonic.
The wave-length
is
06)
or the length of the wires
an odd number of quarter wave-
is
lengths.
If on the other
hand the capacity K^
(17)
which
is infinite,
we get
/i,tan^ = 0,
is
the same as
origin also, putting
if
we had
considered the circuit closed at the
V= 0, for x = 0,
(10) would give

sin
fil
STT
(18)
of the line
is
first
of
= 0,
LT =
*-TThe length
from which (7) and the
21
^'
then a multiple of a half wave-
The two cases correspond to the cases of an organ pipe

length.
open at one end and closed at the other, or closed at both ends.
These conclusions have been
above theory applies,
verified
for instance, to the
by experiment.
The
experiments of Saunders
cited above.
The method employed in this example is typical of the general

process, for the terminal conditions, of whatever nature, are given
in the form of an ordinary linear differential equation in the time,
derivatives of V and I.
Applying this to our
assumed solutions (3) introduces algebraic functions of \, so that,
eliminating by means of (5), we obtain an equation of the form
involving the
tan/iZ=(OX
(19)
where
<j>
is
an algebraic function.
The
case
we have
the simplest case of this transcendental equation.
considered
is
THE ELECTROMAGNETIC
554
FIELD.
[PT. III. CH. XIII.
Equations for Bodies in Motion.
All the equations

of this chapter have been deduced on the supposition that all the
media were at rest. In deducing their extension to the case of
257.
media in motion we shall follow the method of Hertz, as given in

the last and crowning paper of his remarkable researches*.
We
media to be moving at every point with

v whose components at any point are a, @, 7. The
is not supposed necessarily to be moving like a rigid
shall suppose the
velocities
medium
in any manner.
At the surface of
the
between
two
media, although
separation
velocity may be
discontinuous its normal component must be continuous, in order
to preclude the occurrence of vacant spaces.
The fundamental
assumption made by Hertz is that as the medium moves or is
may be deformed
it
body
by the medium so as to pass

the
That
same
material
is, this would be the
points.
through
effect of the motion if it were the only influence at work to change
distorted, the lines of force are carried
the
Besides
field.
this,
we have
the usual effects that appear in
bodies at rest.
Let
3f,
g),
3,
3ft, $1,
represent the field at any point at rest
at a
with respect to the coordinate-axes. The total change in
point in motion will depend on several causes, the first being the
change that is instantaneously taking place at the fixed point
This
through which the material point happens to be passing.
we
ox
shall denote
by
ôt
Secondly the point
is
of the field where the forces are different.
parts
we
The sum
new parts
of these two
shall call
df_ctf
~ + ttidx + ddy
(l)
displaced to
dt
If a
a^ dt
dt
small
dy
a3E^ =
dt^dz
dt
a3E
a
3t
<&
8#
element normal to the JT-axis of area dS were
displaced parallel to itself, the flux through it would vary as just

But if the element rotates, it takes in new amounts of
stated.
flux.
it
At the
start the flux
through
it
was #dS, but
as
it
turns,
acquires a projection normal to the F-axis at the rate ^-, con-
sequently
*
its
"Ueber
Wied. Ann. 41,
die
flux in the positive direction decreases from this

Grundgleichungen der Elektrodynamik
p. 369,
1890; Trans,
p. 241.
fiir
bewegte Korper."
257]
cause at the rate
g)
^-,
555
and in the same manner from its Z-projection
^\
at the rate
But the area
3 o~
tne -^-projection of the element

3
r)/3
is
in the F-direction
also increasing, at the rate
^-
From
^-direction.
We
and
in the
this cause the flux increases at the rate
have therefore to replace the term

in equations (A) by
^-
sum
the
+ a~ + ^
,
(2)
+ 7r
_ +ir J.c +
'L_
Jg)
dy
a
We
first
i^ + 5 S + 931
i^ ay
have thus added in virtue of the motion two parts, the

is the component of the curl of the vector whose
of which
components are
73) -/33,
that
+33 dz
is
velocity.
*-,
3-7*.
the vector product of the induction of the field and the

'The last term is the component of the velocity times
the divergence of the induction.

We may therefore abbreviate
our equations which replace both (A) and (B) thus
(A")
curl
Vgu +
v div
+ 47rg = curl H,
-A
(B")
We may
{^ + curl V$v + v div 33J

interpret the
meaning
of the
= curl
F.
new terms
physically
In the equation (B") the term
thus.
produces a part of curljP. Two vectors having the same curl

differ only by a lamellar vector ( 223).
Consequently the motion
AV$$v = AVv^S perpengives rise to an electromotive force

dicular to the magnetic field and to the direction of the motion.
This
is
nitude
the ordinary electromotive force of induction, and its magmay be specified as equal in any element of conductor
556
to the
number
[PT.
III.
CH. XIII.
of unit tubes of induction cut

perpendicularly
by the element in unit time.
On
the
left
of (A") appears the term
|^
+ curl Vgv +
v div
+ 4<7rq\
222
of, first, the ordinary conduction current q, as in
the
as
in
second,
current,
(2),
displacement
equations (A), third,
a new term equal to the product of the velocity by the density of
consisting
In other words a charge in motion produces the

as a current.
This is verified by Rowland's celebrated
electric charge.
same
effect
experiment*. The corresponding quantity in equation (B") shows

the known electromotive forces produced by the motion of magnets,
and
also explains the
phenomena
of unipolar induction.
be noticed that the theory makes the magnets carry their

force with them.
It is to
lines of
The complete symmetry between electrical and magnetic

phenomena may now be said to have been completely verified by
experiment, with the exception of the second term of equation (A")
which would show the existence of a magnetic field due to the
electric field.
The existence of
made probable by an experiment made by
motion of insulating bodies in an

such magnetic
fields is
Rontgenj*.
The method of deducing the equations (A") and (B") may be

applied even when our axes of coordinates are in motion, if a, ft, y
be the velocities relative to the axes, and e,
X refer to the points
fixed with respect to the moving axes.
Thus the mutual actions
//.,
of bodies
depend only on their
relative motion.
Some
simple
considerations of this nature elucidating the phenomena of unipolar induction are found in a paper by the author in the Electrical
World^, the statements there
made being borne out by
experi-
ments by Lecher.
*
Helmholtz,
"Bericht
betreffend
Versuche
iiber
die
elektromagnetische
elektrischer Convection, ausgefuhrt von H. A. Kowland." Pogg. Ann.

148, p. 487, 1876 ; Wiss. Abh. Bd. i. p. 791.
t "Ueber die durch Bewegung eines im homogenen elektrischen Felde befind-
Wirkung
lichen Dielektricums hervorgerufene electrodynamische Kraft."

264, 1888.
Wied. Ann. 35,
p.
J "Unipolar Induction and Current without difference of Potential." Elec.

World (N.Y.), 23, pp. 491, 523, Apr. 1421, 1894.
" Eine Studie iiber
unipolare Induction." Wied. Ann. 54, p. 276, 1895.
257, 258]
557
The systems of units that

have been explained in this book are those in universal use. The
electromagnetic system is the one altogether used in practical
258.
Other Systems of Units.
measurements, but as we have seen when considering the mutual

and magnetic phenomena the Gaussian system
effects of electrical
is
least
liable
phenomena
produce confusion. When only electrostatic

under consideration the electrostatic system is
to
are
most convenient.
A change
of units has been proposed by Heaviside, who would

define the unit of electricity and magnetism in such a way that
the flux of force due to unit charge out from a closed surface in
This would have the
air should be unity in value, instead of 4?r.
convenient effect of causing the disappearance of the factor 4?r

from many of our equations, for instance from the equation
while the energy per unit volume would be
79
practical advantage would be the disappearance of 4-Tr in

the formula connecting current-turns with magnetomotive-force.
On the other hand the quantity 4?r would be introduced in certain
places where it is now absent. For instance the force at a distance

r from a charged point
would be
m
'
4 7rer2
<
It is rather singular that
Maxwell adopted
this
method
in his
definition of electrical displacement, making the density equal to

the divergence of the displacement, but did not do it in the case
He was
of the magnetic induction, nor even of the electric force.
therefore obliged to
and
make
the displacement equal to
times the
have an unfortunate appearance of disThis has been avoided by Hertz, and in the present
book, and it therefore seems merely a matter of convenience in
Heaviwriting whether we adopt Heaviside's proposition or not.
force,
his equations
symmetry.
side has called the
are
new
units rational, probably not because they

ones, but because of their avoid-
more reasonable than the old
ance in the majority of cases of the irrational number
4?r.
Of the
TABLE OF DIMENSIONS.
558
convenience of this there can be no question, but the question of

units seems now to be beyond the control of theoretical writers*.
A system of units has been proposed by Dr Johnstone Stoneyh,

who advocates the
choice of units so as to
make
the velocity
the
unit of velocity. This would make the numerical measure of all

quantities the same in both electrostatic and electromagnetic
which would be convenient, but inasmuch as the velocity

not yet accurately known, the proposal is hardly practical.
units,
is
table of dimensions of the principal electric
quantities
and magnetic
annexed.
is
Dimensions.
Quantities.
Fundamental
Units.
Length
Mass
Time
Derived Units.
Area
L-
Volume
L
L
Angle
Solid Angle
LT~
Velocity
Angular Velocity
T'
Acceleration
LT~'
Momentum
MLT~
MLT~
ML~ T~
l
Force
Pressure
ML*T~
Energy
Activity or
Power
Energy-Density
Energy-Current-Density
*
ML* T~*
ML~
MT~
T~*
"The position of 4tir in Electromagnetic Units."

Papers, Vol. n. p. 575.
t B. A. Report, 1891.
Heaviside.
28, 1892, p. 292
Nature, July
Electrical Units.
Charge
Volume-Density
Surface-Density
Moment
Polarization
Potential
Field-strength
Induction
Induction-Flux
Current
Current-Density
Capacity
Resistance
Conductance
L~ 1 T e~
Z T~ l c
Inductivity
Conductivity
T~ l e
Resistivity
e~ l
Magnetic Units.
Charge
M^l}
e~*
M*L~*-
e~*
Volume-Density
Surface-Density
Moment
Polarization
Potential
Field-strength
Induction
Induction-Flux
Vector-Potential
Inductivity
L~ 1 T2 c~ l
Z T~ z c
L~ 2 T2 e'
Reluctivity
Z 2 T~*c
Inductance
Reluctance
559
TABLE OF COMPAEATIVE NOTATION.
560
TABLE OF COMPARATIVE NOTATION.

This book
Electric Field
F,
Electric Induction
XYZ
ft,
H,
PQR
XYZ
XYZ
#g)3
Electric Current-density q,
Magnetic Field
Magnetic Induction
Vector Potential
Maxwell. Kirchhoff. Helmholtz.-* Heaviside.
uvw
uvw
LMN
XYZ
abc
FGH
belonging to mag
netic induction
f
'
Vector Potential of
magnetization
PQR
Polarization
Mechanical Force
Volume Density
Surface Density
Electrical Potential
Magnetic Potential
Electric Inductivity
Magnetic Inductivity
Susceptibility
Conductance
Capacity
Eesistance
Total Current
FGH
UVW (UVW
APPENDIX.
Note to
199, 200.
199,
According to the equations (6)
the force on an element of medium in an electric or magnetic field
becomes
infinite at
the surface between two media of different
inductivities, for there
this takes place is
on the surface
is
however
is finite.
The
discontinuous.
layer in which
infinitely thin, so that the total force

may find the force, as stated in 199,
We
by integrating throughout the space included
in
an
infinitely thin
We may
85.
layer containing the surface of discontinuity, as in
the
of
the
stress
also use the results of
200, finding
components
Ty
The six components
x
have discontinuities at the surfaces
of discontinuity of ^, and the forces on the unit of surface are
For instance let us consider a
equal to these discontinuities.
surrounded by a
surface bounding a medium of inductivity
2
in both
media by equations
Zz Yz Zx
,
Xy
(8).
will in general
yu,
medium
of inductivity /^ the surface being such that the lines of

force are normal to it in both media.
Then taking the direction
of the normal to a certain element for that of the JT-axis,
we have
in the
medium
(1)
1,
Z.-gL&.F,,
the force being a tension.
Yy = Z
=Y = Zx = Xy =
z
Q,
In the medium 2 we have

JT.
(2)
The two tensions being in opposite directions on the two sides

of the surface, the resultant force acting on unit of surface is the
difference,
3'=(g
(3)
w.
E.
1J
36
562
APPENDIX.
acting towards the

tinuous,
we have
medium
&=g
rrx =
{F -^
r=(jPl
)
(4)
But
1.
and the
since the induction
is
con-
force becomes,
We may
also obtain the formula (4) in a simple manner

by
the
energy-density in the two media. This is in the
considering
media
and 2
and
g2
respectively.
If now we consider the surface displaced normally a distance

dn toward the medium 1 the prism standing on the element dS
exchanges
its
energy
ft
dSdn,
ttTTft
for the
amount
dSdn,
so that there
is
loss of
energy
---
-4fl
vo,
dSdn,
during the motion through the distance dn. Accordingly the

force on the element dS is that given by (4) towards the medium
If there
1*.
a real charge on the surface, so that the induction
is
discontinuous,
r-c&iv
(5)
and
is
we have
if
^ = ^ this becomes
T-frW + FJ.
(6)
This
is
exemplified in
As a second
147.
particular case, let us consider a surface of disconis

tangential. Then we have in the
tinuity at
which the induction
medium
the pressure
*
factor
The deduction given by Maxwell,

/t 2
in the denominator.
art.
440, is only approximate, lacking the
APPENDIX.
563
and in the medium 2 the pressure
P = ^g
2
F2
2J
in the opposite direction.
There
is
accordingly on the element of surface the difference of
pressure
P=
(7)
If the two media lie side by side

acting towards the medium 1.
the same in
between the plates of a plane condenser we have
both media, so that the pressure is greater in the medium for

which fjL is greater, and the surface is impelled towards the other
This has been verified by Quincke*, who blew a large
bubble of air into a liquid contained between the plates of a con-
medium.
and observed the additional pressure necessary to be given

the air in order to resist the pressure due to the fluid. Quincke
also verified the tension in the direction of the lines of force by
denser,
an absolute electrometer with liquid. The stresses in

magnetized media have been similarly verified by experiments by
Quincke -f and Taylor Jones J.
filling
the principle of the sidewise force Mr A. P. Wills has

founded an accurate method for determining K for substances in
Upon
which it is extremely small, both for magnetic and diamagnetic

substances, by observing the attraction or repulsion on a slab with
its edge in a uniform field.
*
Quincke,
"
" Electrische
Untersuchungen, Wied. Ann. xix.
t Ibid. xxiv. p. 347, 1885.

J Jones,
"
On
p. 705,
1883.
'
Electromagnetic Stress," Phil. Mag. xxxix.
p. 254, 1895.
362
INDEX.
(The numbers refer
Absolute electrometers, 281
,,
value, 2
Absorption,
electric,
to
Pages.)
Betti, 256
Bichat, 282
Biot, 420
402
Acceleration, definition of, 94

unit of, 99
Blondlot, 282, 534
Action and reaction, 96

Activity, equation of, 125
Body
for steady flow, 334

of circuit,
479
motion, 133, 134

Africa, 536
d'Alembert's Principle, 107

Algebraic function, 14
Ampere, 409, 412, 416, 431, 449, 471

unit, 416
Amphigenic charge, 260
oscillation,
in motion, equations for, 555
du Bois, 375
du Bois-Eeymond, 41
Boltzmann, 139, 142
Bosanquet, 860
Boston, 550
Addition, geometric, 9
Adiabatic changes, 470
Amplitude of
Blumcke, 502
Bound
charge, 246
Boussinesq, 66, 522, 540, 541
Bowl, spherical, distribution on, 302

Boynton, 548
Branch-point, 89
Bridge, Wheatstone's, 342
British Association, 100, 416
479
Angle, solid, 76
Cable, submarine, 534
Apparent charge, 359

,,
current, 443
Cadmium
Argand,
8,
Argument
79
of
complex number, 8
Arithmetic, 1
Arithmetical mean, 39
ray, wave-length
Calculus of variations, 48
of ellipsoid,
274
,,
and inductance, neutralization
,,
instantaneous, 404
of,
480
Associative law, 1
of plane condenser,
Attenuation of waves, 546

Axioms, physical, 95
specific inductive,
Axis of polarized distribution, 242
279
357
of spherical condenser,
Cavendish, 251
Cavity, force in, 231
Bacharach, 171
Bedell, 479
Cayley, 85
Centimeter, 97
Beltrami, 199
Centrifugal force, 138

Centrobaric body, 155
Bessel's functions, 545
97
Capacity, 265, 360

of cylindrical condenser, 278
,,
Arons, 399, 400, 407

Association, British, 416
of,
275
566
INDEX.
C.G.S. system, 100
Condition, equations of, 92

Conditions, terminal, 550
Changes, isocyclic and adiabatic, 470

Charge, amphigenic, 260
Conductance, 333
apparent, 359
bound and
Conduction, electrification by, 245
246
free,
Conductivity, 326
induced, 360
Conductor, carrying current, forces on,

426
instantaneous, 404
monogenic, 260
,,
,,
definition of, 245
relation to induction, 358
Conductors, energy of systems of, 266

heat developed in, 333
Chasles, 165
theorem
of,
208
,,
336
mechanical forces on, 440
networks of, 337
non-linear, 344
,,
in parallel
,,
propagation
linear,
Chicago, 416, 550
,,
410
piimary and secondary, 475
rectangular, 455
Circuits, linking of,
Circular circuits, 456
harmonic functions, 87, 179

development
,,
in,
181
steady flow
series,
336
522
325
in,
quadrics, 28
56
Conformal representation,
Coefficients of induction, 265, 450
,,
in,
Cone, potential of, 158

Confocal ellipses and hyperbolas, 319
Clairaut, 101
Clifford,
and
,,
of potential, 266
of potential and induction,
properties
268
of,
Cohn, 399, 400

Coil, toroidal, 451
453
Comite international des poids
et
me-
,,
,,
in two-dimensional
Conservation of Energy, 110, 125

Conservative systems, 110
Constraint, equations of, 103
Continuity, 17, 81
Coordinates, 21
sures, 97
Commutative law, 1
Complex numbers, 4
cyclic,
in extended sense, 5
,,
on plane,
200
problems, 307
Conjugate functions, 82, 85
Connectivity, 59
Coercive force, 394
Coils, induction,
83, 199
of sphere
79
,,
,,
multiplication
,,
,,
variable, function of,
of,
cylindrical, 26
,,
ellipsoidal,
27
generalized, 116
80
orthogonal, 21, 83
function
,,
128
,,
of,
on
polar, 26
surface, 196
positional, 132
semi-polar, 26
Correction for end of wire, 346
Component, 9
,,
Concealed motions, 127

Concentration, 65
Coulomb, 251, 257, 352

unit, 416
of potential, 152
Condensers, 277
Couple experienced by polarized

soid, 370
484
,,
circuit containing,
,,
cylindrical, capacity of, 279
,,
discharge
hysteresis, 397
conducting, 401
of,
401
spherical, capacity
standard, 280
with two
ellip-
,,
moment
of,
due to polarization,
227
of,
dielectrics,
275
Critical point, 89
Curl, 53, 57
402
Current, apparent, 443
INDEX.
Current, displacement, 506
element
567
Damping
force due to, 422
factor,
487
extra, 471
Darboux, 193
Decrement, logarithmic, 487
Definite integrals, 36
,,
force due to straight, 420
Demagnetizing
,,
forces
,,
induced, 471
,,
of,
equivalent to magnetic shell, 412
,,
on conductor carrying,
426
due
421
to,
magnetic energy of
magnetic force due
to,
,,
total, 407,
Derivative, 17
409, 436
,,
in direction, 22
,,
partial, 19
,,
of potential function, 146
Derived units, 98
508
and
virtual
,,
magnetic, 352
surface, 159
,,
436
field of,
potential due to, 410

potential due to circular, 418
,,
effective,
Development in circular harmonics, 181
479
Current-sheets, cylindrical, 349, 454
in spherical harmonics, 191
,,
magnetic field due to, 449

Currents, circular, 456
as cyclic systems, 439, 467
,,
of
,,
, ,
potential
monics, 240
428
reciprocal distance in
spherical harmonics, 189
of lines of force, 201
of
energy
,,
and magnets, mutual energy
,,
non-linear, 432
Dicyclic systems, 139
,,
in parallel cylinders, 457
Dielectric,
,,
in three dimensions, 432
,,
of,
444
Diagrams
Curvilinear coordinates, Green's Theorem

in,
Cyclic coordinates, 128

forces,
Dielectrics,
128
momenta, 133
,,
motions, 128
Differential equations of motion, 97
of,
271
Differentiation of definite integral, 44
Dimensions, of
systems of currents
439, 467
systems, 132, 134
velocities, 128
,,
Cylinder, elliptic, polarization
of,
Cylinder, hollow, magnetization

potential
,,
parameter, 21, 64
,,
system, electrostatic analogy
,,
,,
245
conducting, 398
constant, 357
of slab, 365
condenser with two, 402
Differential, perfect, 58
171
,,
of polarized
body in spherical har-
,,
of,
375
definition of, 144
linear, force
,,
factor,
Density, 74
of,
as,
of
/x,
367
table of, 558
of units, 98
373
of,
418
e,
of electrical units, 251, 414
,,
Dirichlet, 12, 169, 183, 193, 212, 259
375
principle
158, 175
,,
problem
of,
of,
169, 171
169
Cylinders, eccentric, distribution on, 311
,,
for circle, 180
,,
electrification of concentric,
,,
for sphere, 186
,,
elliptic
,,
parallel
278
Disc, potential
and hyperbolic,
distri-
bution on, 315

non-linear currents
in,
457
Cylindrical condenser, capacity of, 279

,,
Damped
current-sheet, 454
oscillation,
486
,,
,,
of,
157
in zonal harmonics, 193
,,
repulsion of, 160
Discontinuity, 17
of derivative of potential,
162
of potential of shell, 238
Displacement, current, 408, 506

electrical. 408
568
INDEX.
Dissipation of energy in hysteresis, 375

Dissipation function, 125
Electromagnetic
field,
of,
Distance, reciprocal, 75
Distortion of waves, 546
on spherical bowl, 302

Distributions, double, 226
,,
polarized, energy of, 239
potential
,,
,,
,,
theorem, 66
for
,,
Lord
Kelvin's, 281
,,
symmetrical, 283
Electromotive force, 333
,,
,,
of induction, 469.
,,
,,
periodic, 477
two dimensions,
dinates, 173
555
Electrostatic system of units, 252
theory of cable, 535, 549
Double distributions, 228

of,
517
Quadrant, 284
Electrometers, 281
in curvilinear coor-
Doublet, potential
light,
Electromagnetism, equations of, 433

Electrometer, Bichat and Blondlot's, 282
177
,,
theory of
units, 413
,,
Distributive law, 1
Divergence, 66
of force, 154
,,
,,
of,
231
,,
515
equations of for
media at rest, 509
Distribution on ellipsoid, 273
,,
dynamical theory
units compared with prac-
,,
233
tical,
Drude, 527
417
problem of, 263

and magnetostatics, paral-
Electrostatics, general
Duane, 397, 534
,,
Duhem, 260
treatment
lel
Element of
Dynamics, 91
Dynamo-machine, 453
Dyne, 100
,,
circuit, force on,
Earth-plates, 345
Eccentric cylinders, 311
and hyperbolas,
,,
distribution on, 273

polarized, couple experienced
,,
potential
by, 370
Electrets, 389
Electric field, 244
of,
210
for internal point,
,,
Electrical equilibrium, 253
213
"fluid, "248
of,
,,
248
,,
in sphere, 298
inversion, 301
,,
phenomena, 243
,,
quantities, dimensions of, 251
Electrification,
Elliptic cylinder, polarization of, 373

,,
cylinders, distribution on, 315
Embranchment, point
End
252, 366
244
theorems on sign
Electrified sphere, energy of, 272
Electrodes, 336
homoaoid, 203
,,
358
,,
of,
uniform polarization of, 369

,,
Ellipsoidal coordinates, 31
Electricity, not incompressible, 248, 508
of,
polarization
,,
374
image, 294
unit
of revolution, attraction of, 219
,,
,,
true,
431
,,
Efficiency of transformer, 484
law
of,
confocal, 319
Ellipsoid, capacity of, 274
Effective current, 479
force,
353
of current, force due to, 422
Elements, parallel, attraction

Ellipses
of,
426
of,
338, 491
of wire, correction for, 346
Energy, 91
of,
260
,,
,,
conservation
of,
110
dissipation of in hysteresis, 395

of distributions, 219
439
Electrodynamic potential, 429
electrokinetic,
Electrokinetic energy, 439
equation
Electrokinetics, 325
in terms of field, 221
of,
109
INDEX.
Energy, Gauss's theorem
kinetic, 109
219
of,
Field, magnetic, 351
magnetic, 352
maximum theorem
,,
for,
223
mutual, of magnets and currents,

444
mutual
,,
due to polarization of sphere, 372

strength of, 144
Field-magnet, 453
,,
of magnetic field of currents, 436
, ,
569
Flow, steady, in conductors, 325

variable, 398
Flux of force through circuit, 440
potential, 220
of vector, 68
mutual, of two currents, 428

of polarized distributions, 239
system of conductors, 266

transfer of, 517
unit of, 111
Flux-function, 195
for
magnetic induction,
460
of
,,
,,
Force, centrifugal, 138
Eolotropic media, 510

Equation of activity, 125
for steady flow,
334
telegraphic, 533
,,
general case of, 540

Equations of condition, 92
of constraint, 103
,,
,,
,,
,,
differential, of
,,
of electromagnetic
field
,,
equations of lines
for.
at rest, 509
body, 387
,,
of induction, 504
,,
of lines of force, 194
,,
for
,,
of propagation, 512
moving
flux of
through
impressed, 97
,,
,,
electrical,
for,
electrical,
248
,,
magnetic, due to current, 409, 436

magnetomotive, 453
mechanical, on element of current, 426
104
tubes
unit
260
of,
of,
100
unit tube
of,
154
Forces acting on polarized body, 381
layers as screens, 289

surface, 145
Equipotentials,
440
lines of, 145
of stressed body, 387
,,
194
circuit,
,,
253
Equipotential layer, 165
,,
of,
lo-
503
of inertia, 108
law of
bodies, 555
Equilibrium, condition
induction,
calized,
of electromagnetism, 433
of equilibrium of stressed
,,
of
,,
,,
motion, 97
,,
media
coercive, 394
on conductor carrying current,
426
electrical, law of, 248
electromotive, 330
of induction, 469
,,
,,
condition
for
infinite
family of, 204

Everett, 101
elementary, various resolutions
,,
into,
430
mechanical, on conductors, 440

Force-function, 110
,,
Extra-current, 471
Newtonian, 111
Formulae, symbolic, 435
Fourier, 98, 183, 326, 523
Factor, demagnetizing, 375
Fourier-Ohm law, properties of vectors
Ewing, 375, 394
series of, 183
Family
of equipotentials, 204
Farad, 416
Faraday, 154, 243, 245, 251, 259, 335,

382, 384, 389, 467, 469
Feddersen, 488
Field, electric, 244
221
,,
energy in terms
,,
of current, magnetic energy of, 436
of,
obeying, 330
Free charge, 246

Freedom, degrees
Frequency, 479
of,
92
Function, algebraic, 14
,,
analytic, 81
,,
of
complex
variable, 80;
surface, 196
on
INDEX.
570
Function, conjugate, 82, 85
continuous, 17, 19
,,
Gyrostat, 129
dissipation, 125
Hamilton,
Green's, 290
harmonic, 65
,,
,,
holomorphic, 14, 88
homogeneous, 25
,,
irrational, 14
,,
monogenic, 81
,,
6, 8,
25, 57, 116, 123, 124
equations
Harmonic
,,
circular, 87, 179, 181

,,
,,
186,
240
,,
Functions, circular harmonic, 87, 179

force-,
183,
187, 188, 189, 191,
transcendental, 14
uniform, 13, 81
110
,,
hyperbolic and circular, 317

reciprocal, 123
,,
spherical harmonic, 183
,,
development
in, 181
spherical,
rational, 14
,,
467
functions, 65
multiform, 13
of point, 20
,,
of, 124,
Hamilton's Principle, 116, 142

Harkness and Morley, 18, 44, 171
oscillation,
479
Harnack, 179
Heat developed in conductors, 333
Heaviside, 56, 153, 360, 390, 391, 392,
393, 449, 479, 505, 508, 519, 524, 531,
534, 536, 540, 545, 547, 551, 557
Helmholtz, 56, 125, 128, 335, 355, 384,

Gauss,
8, 77,
100, 165, 199, 220, 238, 416
392, 418, 430, 435, 447, 448, 467, 469,

472, 488, 556
integral of, 77
theorem of, 75
theorem on energy, 219
,,
Gaussian system of units, 367, 413, 414,
418, 509, 557
General problem of electrostatics, 263
Helmholtz's theorem, 433
Generalized coordinates, 116
Holomorphic function, 14, 88

Homceoid, attraction of, 210
,,
equations of Lagrange, 118
,,
Hertz, 137, 392, 407, 418, 449, 505, 517,

519, 524, 527, 530, 534, 554, 557
Heterogeneous medium, effect of, 441
Hollow cylinder, magnetization of, 375
Geometric addition, 9
mean
,,
ellipsoidal,
distance, 461
Homogeneous
,,
multiplication, 10
Hopkinson, 394
,,
product, 11
Hospitalier, 479
Hyperbolas, orthogonal equilateral, 86

Hyperbolic and circular functions, 317
Gibbs, 153
Glazebrook, 488
Gram, 98
cylinders,
and Mathews, 545
,,
,,
,,
function
problem
theorem
,,
,,
couple, 397
dissipation of energy in, 375
for logarithmic
tial,
,,
on,
Hysteresis, 393
Green, 59, 164, 290

formulae of, 162
,,
,,
distribution
315
Gray, 457
,,
203
function, 25
of,
,,
178
loop, 394
static,
394
290
for plane, 293
Idiostatic method, 284
for sphere, 295
Image,
of,
of, 59,
63
for plane,
in
electrical,
294
in sphere, 298
290
Impedance, 479
Impressed force, 97
176
curvilinear
nates, 171
Guard-ring, 281
poten-
coordi-
Indefinite integral, 41
Index of refraction, 517
Induced charge, 360
INDEX.
Induced current, 471
571
Irrational, 2
Inductance, 450
algebraic, 4
,,
and capacity, neutralization
,,
of,
490
Induction, coefficients
of,
265
motions, 133, 134
453
coils,
function, 14
Isocyclic changes, 470
Ivory's theorem, 216
high-frequency, 499
,,
definition of, 230, 356
,,
machines, 286
electrification by, 244
,,
electromotive force
,,
general equations
electrical
of,
of,
469
504
Hertzian, 391
localized electromotive force
,,
,,
of,
Jacobi, 69, 73, 173

Jones, 517, app. 563
Jordan, 193
Joule, 416
Joule's law, 335, 469
Kelvin, 240, 259, 281, 288, 298, 302, 357,
503
382, 469, 471, 488, 535, 539

Dirichlet's Principle, 169
and
Maxwellian, 390
particular cases of, 471
relation of charge to, 358
,,
Kelvin's replenisher, 286
Kilogramme prototype, 98
Kinetic energy, 109
Inductive capacity, 357

Inductivity, 357
potential, 127
,,
theory of gases, 134, 399
Inertia, 91
,,
definition of, 95
,,
moment
Kirchhoff, 326, 331, 337, 339, 488, 534,
415
forces of, 108

of,
Kirchhoff 's laws, 337
130
Infinite integrand or limit,
43
straight current, force due to,
,,
420
Kohlrausch, 415
KR-law," 536
"
Kronecker, 44, 171
Infinitesimal, 16
arc, area,
Infinity,
and volume, 34
transformation, 49
.,
Lagrange, 108, 119, 123

,,
equations of, 118, 467
Lagrangian function, 118, 124
16
Influence, electrification by, 244
modified, 127
Instantaneous capacity, 404

,,
charge, 404
Lame,
,,
Integrability, 37
equation
46
definite, 36,
,,
of, 68, 85,
in
cylindrical coordi-
88
nates, 174
,,
50
Law,
surface-, 50
Integral form of Ohm's law, 332

Intrinsic polarization, 389
,,
,,
ment
of,
operator
of,
68
associative, 1
,,
commutative,
,,
distributive, 1
of electrical force, 248

of force, generalized, 354
Heaviside's treat-
Invariant of transformation, 309

Inverse points, 295
Iron, hard, 389
coordi-
nates, 173
indefinite, 41
line-,
148
spherical
double and multiple, 45

of function of complex variable,
,,
vectors, 59
Laplace, 189, 192, 251
Integral, 36
,,
173
Lamellar polarizations, 235
Insulators, 245
,,
22, 64,
391
Joule's, 335
,,
Kirchhoff 's, 337
KR-, 536
Lenz's, 137, 471
572
INDEX.
Law
Magnetomotive-force, 453
Layers, equipotential, as screens, 289
Magnetostatics and electrostatics, parallel treatment of, 353
of motion, Newton's, 95, 96

Layer, equipotential, 165
Leakage, magnetic, 482

Lecher, 556
Magnets, 226, 351
Length, unit
of,
189
Mascart, 457
Mass, definition of, 97

unit of, 98
jar, 488, 490,
499
Matter, 91
Light, electromagnetic theory

Limits, 2
of,
517
Maximum
of functions, 14
,,
theorem
resistance
,,
of,
343
145
of, force,
,,
Linking of
circuits,
harmonic
for energy,
233
251, 302, 324, 339, 384, 399, 407, 420,

431, 436, 439, 444, 446, 457, 461, 466,
467, 481, 505, 506, 507, 508, 516, 517,
equations
,,
of
Maxwell, 56, 59, 65, 67, 97, 139, 243,
current, force due to, 421
,,
minimum
or
function, 154
Linear conductors, 336
Lines
permanent, 389
Map, 199
surface, 20
,,
mutual energy
444
Magnification, 84
Lenz, 467, 471, 476

law of, 137, 471
Level sheet, 58
Leyden
currents,
of,
,,
97
of,
and
,,
Legendre, differential equation

,,
polynomial of, 188
of,
194
410
521, 524, 557
Mean, arithmetical, 39
Lipschitz, 302
,,
distance, geometrical, 461
Lodge, 488, 490, 511

Logarithm, 89, 90
,,
quadratic, 479
theories of the, 41
Logarithmic decrement, 487

,,
potential, 175
Mechanical force on conductor, 440

on element of circuit,
,,
,,
transformation, 322
,,
426
Medium, action
of,
354
Machines, electrical induction, 286

Maclaurin, 209, 210
,,
eolotropic, 510
Magnetic density, 352

,,
energy, 352
,,
heterogeneous, effect
,,
stress in,
351
field,
potential, 352
,,
due to current, 409

resistance, 360
, ,
shell,
,,
solenoid, 352
,,
Magnetism,
current equivalent to, 412
of,
,,
Minimum
,,
condition for equilibrium, 111

in Dirichlet's Prin,,
170
Fourier- Ohm law, 331

(
induced, of cylinder, 377

of hollow sphere,
,,
378
,,
Michelson, 97
Microfarad, 417
Models, mechanical, of induction, 13 J,

149, 477
366
residual, 389
,,
441
Mendenhall, 159
ciple,
Magnetization, 352
,,
of,
384
Mercator's projection, 200, 350

Metre prototype, 97
-in
true, 358
unit
247
Megohm, 417
due to current-sheet, 449

force due to current, 409, 436
leakage, 482
,,
effect of dielectric,
uniform, of ellipsoid, 370

of sphere, 371
,,
Modulus, 8
Moment of polarized distribution, 242

Momentum, generalized component of,
121
Monocyclic system, 137
Monogenic charge, 260
573
INDEX.
Motion, adiabatic, 133, 134
concealed, 127
Parameter,
first,
cyclic, 128
97
Paris, 97,
Multiform function, 14
Period, 479
Periodic electromotive force, 477
10
Permanent magnets, 389
undetermined, 105
,,
416
Path, dependence of integral on, 53

Pendulum, gyroscopic, 131
isocyclic, 133, 134
Multiplication, geometric,
Multiplier, Jacobi's, 69
vector, 22
second, scalar, 64
,,
differential equations of,
,,
21
differential,
Permeability, 357
Muraoka, 407
Mutual potential energy, 220
Permittance, 360
Physics, axioms of, 95
Nabla, 25
definition of, 91
Nachwirkung, 394
Networks of conductors, 337, 491
mathematical, 91
Picard, 56, 171, 179, 183, 540
Planes, parallel, distribution on, 279
Max well's treat-
,,
ment
of,
339
Pockels, 523
Poincare, 527, 530, 533, 540, 545

Point, branch or critical, 88
179, 290, 429, 431, 467, 469,
Neumann,
471
Newton, 108, 203

laws
,,
of,
theorem
Point-function, 20
charge in sphere, 366

Poisson, 193, 515
,,
equation of, 152
95
of,
203
forces, 113
Newtonian
law, electric, 251
,,
for logarithmic potential, 178
,,
for surface, 162, 169
Nichols, 407
Non-linear conductors, 344
Poisson 's theory of induction, 368

Polarization, 361
Notation, table of comparative, 560

Numbers, complex, 4
of ellipsoid, 370
,,
imaginary, 5
intensity of, 228
,,
integer, 1
intrinsic,
irrational, 2
389
Heaviside's treat-
ment
Numeric, 97
Oberbeck, 502
Oersted, 409
,,
solenoidal, 235
field
416
law, 326, 332

Operators, abbreviations for, 153
Order, right-handed, 8
,,
Ohm's
Orthogonal coordinates,
Orthomorphosis, 85
Oscillation,
21,
83
forced,
and
491
free,
489
of function, 37
,,
harmonic, 479
Oscillator, spherical, 527
Parallel, conductors in, 336

Parallel elements, attraction of, 413
Parallelogram, Wheatstone's, 342
due
to,
372
uniform, 368
potential due to,
234
Polarized body, 226, 228
distributions, energy of, 239
,,
damped, 486
electrical, 488,
,,
391
of sphere, 371
Ohm, 325
unit,
of,
lamellar, 235
forces on, 381
potential
of,
231
development
shells, 236
Positional coordinates, 132
,,
of,
Potential function, characteristics

coefficients of,
of,
240
154
266
completely defined by
characteristic properties,
167
574
INDEX.
Potential function, definition
,,
,,
of,
144
Eeactance, 479
of cone, 157
Eeaction, 96
of cylinder, 158
of disc, 157
Eeciprocal distance, 75
,,
of ellipsoid,
development of in
210
spherical
of polarized distribution, 231
,,
Potential due to circular current, 418
,,
,,
Eectangular
electrodynamic, 429
energy, 110, 114
Eefraction, index
kinetic, 127
Eelaxation-distance, 524
Eeflection by conductor, 525
-time, 399
magnetic, 352
mutual, 220
259
Eepresentation, conformal, 83
,,
circuit,
307
388
,,
Eesistivity,
Principle, d'Alembert's, 107
Hamilton's, 116
Kelvin and Dirichlet's, 169
475
Probability integral, 537
Problems, Dirichlet's, 169

Green's, 390
Problems, uniplanar, 308
Projection, Mercator's, 200, 350
stereographic, 201
of vector, 9
Propagation in conductor, 522

of disturbance, 515
,,
,,
of linear conductor, 343

magnetic, 360
,,
perpendicular to lines of force,
,,
equation
of,
512, 515
326
430
forces,
Eesolved part, 9
Eesonance, 488, 499, 501
Eesultant, 23
Eiemann,
82, 171, 183
Eight-handed order, 8
Eontgen, 556
Eotation of vector, 56
Eouth, 125, 502
Eowland, 407, 556

Sachse, 183
Sardinia, 536
Saturation, intensity
Quadrant electrometer, 284
Saunders, 534, 553

Savart, 420
Quadratic mean, 479
Scalar product, 10
Quaternion, 7
Scalars, 9
Quincke, app. 563
Schiller,
Bational functions, 14
numbers, 1
,,
units, 557
Eayleigh, 125, 139, 347, 502, 524
Series,
of,
393
488
Secondary
Eadiant vector, 519, 522
into
Eesolutions, various,
Pupin, 490, 510
,,
electric application,
Eesidual magnetism, 389

Eesistance, 333
Pressure, 385
,,
,,
479
Practical system of units, 416
circuit,
of induced current, 472
Eeluctance, 369, 453

Eeplenisher, 246, 286
Poynting's theorem, 517
Primary
,,
,,
Pot, operator, 153
Power absorbed by
517
of,
of lines of flow, 327
,,
logarithmic, 175
vector, 433
,,
theorem in electrostatics, 266

circuits, 455
,,
due to uniform polarization,

234
zero,
har-
monics, 189
functions, 123
circuit,
475
97
336
,,
conductors
,,
trigonometric, 179
Sheet, current, 349

vector, 199
in,
elementary
INDEX.
236
Shells, polarized,
Telegraphic equation, general case

circuits, 459
Singular point, 17
Slab in uniform field, 364
Telephonic speaking, possibility

Telephony, long-distance, 536
polarizations, 235
Solid angle, 76
Somoff, 173
Specific inductive capacity, 357
Sphere, attraction
157
of,
,,
hollow, magnetization
,,
potential
,,
uniform polarization
along lines of force, 388

Tensor, 8
Terminal conditions, 550
of,
378
mal, 386
Theorem, du Bois-Eeymond's, of the
mean, 41
Chasles's, 208
,,
divergence, 56
,,
Gauss's, 75
155
of,
on energy, 219
Green's, 59, 63
371
of,
Spheres, concentric, electrification of, 275

Spherical bowl, distribution on, 302
harmonic, 183
,,
,,
Helmholtz's, 433
Ivory's, 216
Maclaurin's, 209
,,
axis of, 187
,,
,,
development in, 191

development of po-
body in, 246

forms of, 186
zonal, 188
oscillator,
for energy, 223
,,
maximum,
,,
Newton's, 203
reciprocal, in electrostatics,
tential of polarized
266
Stokes's, 53, 56
,,
Taylor's, 49, 181
Theorems on sign of electrification, 260

Thomson, 171, 240, 488
Thomson and Tait, 95, 132
527
Spin, 56
Standard condensers, 280

Steinmetz, 479
Stokes's theorem, 53
Thomson, Elihu, 399

Thomson, J. J., 139, 381,
Stoney, 557
Time, relaxation-, 399

unit of, 98
,,
Tore, 451
Strength of
field,
444, 483, 502,
519, 527, 531
144
of polarized shell, 238
,,
medium, 384
Submarine cable, 534
Stress in
Toroidal
Surface-distributions, 159
Total current, 407, 508

Traction, 385
Transcendental function, 14
-tension, 272
coil,
451
Transformation of curve, 48
Susceptibility, 367
Symbolic formulae, 435
infinitesimal, 49
Symmetrical electrometers, 283
Systems of conductors, flow

,,
546
Tetrahedron, equilibrium of infinitesi-
potential of shell proportional to, 238
,,
of,
Tesla, 499, 502
vectors, 66
,,
540
Tension, 385
polarized, 235
Solenoidal property of induction, 230
,,
of,
Telephone
Soap-bubble, electrified, 272

Solenoid, magnetic, 352
,,
575
in,
327
conservative, 110
of currents as cyclic systems, 467
Transformer, 481, 483

,,
efficiency of,
484
Transverse undulations, 521

Trigonometric series, 179
Trowbridge, 534
Tannery, 44
Taylor's theorem, 49, 181
Telegraph, theory
logarithmic, 322
,,
of,
535
Telegraphic equation, 533
True electricity and magnetism, 358

Tubes of force, 260
unit, 154
576
INDEX.
Undulations, transverse, 521

Uniform function, 13
Vector, sheet, 199

radiant, 519, 522
polarization, 368
Vectors, 9
addition of, 9
,,
Uniplanar problems, 308

Unit of acceleration, 99
lamellar, 57
,,
of electricity, 252, 366
,,
multiplication
of force, 100
of length, 97
,,
of magnetism, 366
of mass, 98
,,
of time, 98
,,
of velocity, 99
,,
Velocity, definition
of,
93
of,
117
of propagation, 515
,,
derived, 98
,,
dimensions
unit
,,
tube of force, 154

Units, absolute, 100
of,
99
Virtual current, 479

Volt, 416
Vortex motion, 435

of,
98
414
Warburg, 394
,,
electrical,
,,
electromagnetic, 413
,,
electrostatic
tical,
10
generalized component
,,
,,
compared with prac-
417
,,
Gaussian, 367
practical, 411
,,
rational, 557
,,
various systems
,,
of,
obeying Fourier-Ohm law, 330

solenoidal, 66
,,
Water, dielectric constant

Watt, 416
Wave-length, 524
of,
402
as standard of length, 97
,,
Waves, plane, 519

Weber, 415, 416, 431
of,
557
Weierstrass, 18, 171
Wheatstone's bridge, 342

Value, absolute, 2
Wien, 519
Variable, complex, function

Variable flow, 398
of, 10,
196
Wills, 397, app. 563
Work, 102
Variation of multiple integral, 73

Variations, calculus of, 48
, ,
virtual,
Vector potentials, 433

,,
product, 11
CAMBRIDGE
PRINTED BY
in displacement of conductors, 269
103
Zero potential, 259
J.
AND
C. F.
CLAY, AT
THE UNIVERSITY PRESS.
University of Toronto
GO
Library
O
H
-P
1
-P
5
o o
J5 CH
-P O
SBO
v
h O
^!
-P EH
CD
DO NOT
REMOVE
THE
CARD
FROM
THIS
POCKET
Acme
Library Card Pocket
LOWE-MARTIN GO.
LIMITED

Webster Theory of Electricity and Magnetism, Being Lectures On Mathematical Physics (1897)

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Webster Theory of Electricity and Magnetism, Being Lectures On Mathematical Physics (1897)

Hochgeladen von

Copyright:

Verfügbare Formate

THE THBOEY

ELECTRICITY AND MAGNETISM.

ELECTRICITY AND MAGNETISM

.ECTURES ON MATHEMATICAL PHYSICS

ARTHUR GORDON WEBSTER,

A.B. (HARV.), PH.D. (BEROL.),

ASSISTANT-PROFESSOR OF PHYSICS. DIRECTOR OF THE PHYSICAL LABORATORY,

YORK: THE MACMILLAN COMPANY

AT THE UNIVERSITY PRESS.

justification is perhaps necessary for the

another treatise on Electricity and Magnetism, in view of the

numerous ones already existing

based upon various courses of lectures that

must always remain

the modern treatment of the subject,

at Clark University during the last six years.

the contents of the two volumes of Maxwell in a year, or even

years, is only to expose

Again, Maxwell's own views are there presented by

lent treatises of Mascart

and Joubert and of Watson and Burbury

Maxwell with considerable

admirable treatise, though containing much recent matter, suffers

Professor J. J. Thomson's altogether delightful Elements

of the Mathematical Theory, which

when the present

book was nearly ready

the press, while extremely modern as well

addressed to a somewhat different class of students

from that contemplated in writing the present book.

theoretical writings of Hertz, Heaviside,

have resulted in the systematization of Maxwell's theory and have

adapted to the use of the student on account of their

chapters on theoretical matters

by Hertz are hardly by way

present to the student the results of the theory as it stands to-

after the labors of Faraday, Maxwell, Helmholtz,

convenient to state what I consider

be the essentials of Maxwell's theory as distinguished from

in the opinion of the writer

Maxwell's system of equa-

fully the points of difference, they are

energy in the medium.

The magnetic action of displacement currents.

advisable to abolish the usual terms repugnant to so

who assuming the

deny the existence of

was not one of

Feeling that the consideration of the

old theory of action at a distance, but for the

introduces the student to

that he will need in various branches of mathematical physics, I

have prefixed to the treatment of Electricity a rather complete

been the custom of English writers

It will probably be admitted, however, that the inclusion of this

subject in a treatise mainly devoted to the consideration of rods,

devoted to Electricity and

no more appropriate than in one

colleges are as a rule insufficiently prepared in the departments

of mathematics necessary in approaching the subject of mathe-

of but three text-books on the

Calculus in English, those of Greenhill, Williamson and Byerly,

expedient to prefix a mathematical introduction