VECTOR AND TENSOR

ANALYSIS
by
G.E.HAY

HONG KONG POLYTEGH'NtC

LIBRARY

DOVER PUBLICATIONS, INC.

NEW YORK

Copyright 1953 by Dover Publications, Inc.

All rights reserved under Pan American and International Copyright Conventions.

Published in Canada by General Publishing Company, Ltd., 30 Lesmill Road, Don Mills, Toronto,
Ontario.
Published in the United Kingdom by Constable
and Company, Ltd., 10 Orange Street, London
WC 2.

Vector and Tensor Analysis is a new work, first

published by Dover Publications, Inc., in 1953.

International Standard Book Number: 0-486-60109-9

Library of Congress Catalog Card Number: 54-1621
Manufactured in the United States of America
Dover Publications, Inc.
180 Varick Street
New York, N. Y. 10014

TABLE OF CONTENTS

CHAPTER I. ELEMENTARY OPERATIONS

1. Definitions

2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.

.. . . . . . . . .
Addition of vectors. . . . . . .
Multiplication of a vector by a scalar
Subtraction of vectors. . . . . .
Linear functions . . . . . . . .
Rectangular cartesian coordinates
The scalar product . . . .
The vector product
Multiple products of vectors
Moment of a vector about a point
Moment of a vector about a directed line
Differentiation with respect to a scalar variable.
Integration with respect to a scalar variable
Linear vector differential equations
Problems . . . . . . . . . . . . . . .

Page

1
2
4'

6
6
7
10
11
15
18
20

22
25
26

28

CHAPTER II. ApPLICATIONS TO GEOMETRY

34
34

15. Introduction. . . . . . . . .
16. Some theorms of plane geometry

Solid Ana(ytic Geometry

Notation . . . . .
Division of a line segment in a given ratio
The distance between two points
The area of a triangle. . . . . . . .
21. The equation of a plane . . . . . .
22. The vector-perpendicular from a point to a plane.
23. The equation of a line . . . . . . . . . . . .

37

17.
18.
19.
20.

38
39
40
41
43
46
v

Page

24. The equation of a sphere . .

25. The tangent plane to a sphere

Differential Geometry
Introduction. . . . . . .
The principal triad. . . .
The Serret-Frenet formulas
Curvature and torsion
Problems . . . . . . . .

26.
27.
28.
29.

49
50

51

52
53

55
58

CHAPTER III. ApPLICATION OF VECTORS TO MECHANICS

Motion of a Particle
30. Kinematics of a particle. . . . . . . . . . . . . . . .
31. Newton's laws . . . . . . . . . . . . . . . . . . . .
32. Motion of a particle acted upon by a force which is a given
function of the time. . .
33. Simple harmonic motion .
34. Central orbits . . . . . .
35.
36.
37.
38.

39.
40.
41.
42.

Motion of a System of Particles

The center of mass of a system of particles . . . . . . . .
The moments and products of inertia of a system of particles
Kinematics of a rigid body
The time derivative of a vector. .
Linear and angular momentum
The motion of a system of particles
The motion of a rigid body with a fixed point
The general motion of a rigid body .
Problems . . . . . . . . . . . .

62
66
68
69
70
72
73
77
79

80
83
84
94
97

CHAPTER IV. PARTIAL DIFFERENTIATION

43.
44.
45.
46.

Scalar and vector fields . . . . . .

Directional derivatives. The operator del
Properties of the operator del
Some additional operators .

vi

102
102
105
107

Page

47.
48.
49.
50.

Invariance of the operator del

Differentiation formulas . . .
Curvilinear coordinates . . .
The expressions \Jf, \l. band \J X b in curvilinear coordinates. .
Problems . . . . . . . . . . . . . . . . . . . . . .

CHAPTER

V.

61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.

124
127

INTEGRATION

51. Line integrals .

52. Surface integrals
53. Triple integrals
Problems . . .
54. Green's theorem in the plane
55. Green's theorem in space . .
56. The symmetric form of Green's theorem.
57. Stokes's theorem. .
58. Integration formulas
59. Irrotational vectors .
60. Solenoidal vectors
..
Problems . . . . .
CHAPTER

III
117
120

VI.

130
134
138
139
140
143
145
146
149
151
152
154

TENSOR ANALYSIS

Introduction. . . . . . .
Transformation of coordinates . .
Contravariant vectors and tensors
Covariant vectors and tensors . .
Mixed tensors. Invariants . . . .
Addition and multiplication of tensors.
Some properties of tensors.
Tests for tensor character
The metric tensor . . . .
The conjugate tensor . . .
Lowering and raising of suffixes

157
157
159
160
161
162
163
164
166
167
169
Vll

Page

72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.

Magnitude of a vector. Angle between two vectors

Geodesics
.
Transformation of the Christoffel symbols
Absolute differentiation
Covariant derivatives
The curvature tensor .
Cartesian tensors .
Oriented cartesian tensors.
Relative tensors
Physical components of tensors .
Applications .
Problems
...

viii

170
170
173
174
177
177
178
180
181
184
186
189

VECTOR AND TENSOR ANALYSIS

CHAPTER

ELEMENTARY OPERATIONS
1. Definitions. Quantities which have magnitude only are called
scalars. The following are examples: mass, distance, area, volume. A
scalar can be represented by a number with an associated sign, which
indicates its magnitude to some convenient scale.
There are quantities which have not only magnitude but also direction. The following are examples: force, displacement of a point, velocity
of a point, acceleration of a point. Su<;h quantities are called vectors if
they obey a certain law of addition set forth in 2 below. A vector can
be represented by an arrow. The direction of the arrow indicates the
direction of the vector, and the length of the arrow indicates the
magnitude of the vector to some convenient scale.
Let us consider a vector represented by an arrow running from a
point P to a point Q, as shown in Figure 1. The straight line through
P and Q)s called the line of action of the vector, the point P is called the
origin of the vector, and the point Q.is called the terminus of the vector.
To denote a vector we write the letter indicating its origin followed
by the letter indicating its terminus, and place a bar over the two letters. The vector represented in Figure 1 is then represented by the
symbols PQ. In this book the superimposed bar will not be used in
any capacity other than the above, and hence its presence can always
Q

Figure 1

be interpreted as denoting vector character. This notation for vectors

is somewhat cumbersome. Hence when convenient we shall use a
simpler notation which consists in denoting a vector by a single symbol
in bold-faced type. Thus, the vector in Figure I might be denoted by
the symbol a. In this book no mathematical symbols will be printed
in bold-faced type except those denoting vectors.
The magnitude of a vector is a scalar which is never negative. The
magnitude of a vector PQ will be denoted by either PQ or IpQI. Similarly, the magnitude of a vector a will be denoted by either a or lal.
Two vectors are said to be equal if they have the same magnitudes
and the same directions. To denote the equality of two vectors the
usual sign is employed. Hence, if a and b are equal vectors, we write
a=b.

A vector a is said to be equal to zero ifits magnitude a is equal to

zero. Thus a = 0 if a = O. Such a vector is called a zero vector.

2. Addition of vectors. In I it was stated that vectors are quantities with magnitude and direction, and which obey a certain law of
addition. This law, which is called the law ofvector addition, is as follows.
R

Figure 2

Let a and b be two vectors, as shown in Figure 2. The origin and

terminus of a are P and Q. A vector equal to b is constructed with
It is difficult to write bold-faced symbols on the blackboard or in the exercise
book. When it is desired to write a single symbol denoting a vector, the reader
will find it convenient to write the symbol in the ordinary manner, and to place
a bar over it to indicate vector character.

its origin at Q. Its terminus falls at a point R. The sum a+ b is the

vector PR, and we write
a+b = PRo
Theorem 1.

Vectors satisfy the commutative law of addition; that is,

a+b = b+a.
Proof.

Let a and b be the two vectors shown in Figure 2. Then

(2.1)

a+b = PRo

We now construct a vector equal to b, with its origin at P. Its terminus falls at a point S. A vector equal to a is then constructed with
its origin at S. The terminus of this vector will fall at R, and Figure
3 results. Hence
R

Figure 3

b+a= PRo
(2.2)
From (2.1) and (2.2) it follows that a+b = b+a.
Theorem 2. Vectors satisfy the associative law of addition; that is,
(a+b) +c = a+(b+c).
Proof. Let us construct the polygon in Figure 4 having the vectors
a, b, c as consecutive sides. The corners of this polygon are labelled P,
Q, Rand S. It then appears that

(a+b) +c =-PR+c

PS,

a+(b+c) = a+QS

= PS.
Hence the theorem is true.
According to Theorem 2 the sum of three vectors a, b, and c is

independent of the order in which they are added. Hence we can

write a+b+c without ambiguity.
b

p~,
Figure 4

Figure 4 shows the construction of the vector a+b+c. The sum

of a larger number of vectors can be constructed similarly. Thus, to
find the vector a+b+c+d it is only necessary to construct the
polygon having a, b, c and d as consecutive sides. The required vector
is then the vector with its origin at the origin of a, and its terminus
at the terminus of d.
3. Mu(tiplication of a vector by a scalar. By definition, if m is a positive scalar and a is a vector, the expression ma is a vector with magnitude ma and pointing in the same direction as a; and if m is negative, ma is a vector with magnitude 1m I a, and pointing in the direction
opposite to a.
We note in particular that -a is a vector with the same magnitUde
as a but p0inting in the direction opposite to a. Figure 5 shows this
vector, and as further examples of the multiplication of a vector by a
scalar, the vectors .2a and -2a.

//
Figure 5

Theorem. The multiplication of a vector by a scalar satisfies the

distributive laws; that is,
4

= ma+na,

(3.1)

(m+n)a

(3.2)

m(a+b) = ma+mb.

Proof of (3.1). If m+n is positive, both sides of (3.1) represent

a vector with magnitude (m+n)a and pointing in the same direction
as a. If m+n is negative, both sides of (3.1) represent a vector with
magnitude Im+nla and pointing in the direction opposite to a.
Proof of (3.2). Let m be positive, and let a, b,ma and mb be as
shown in Figures 6 and 7. Then
R

Figure 6

ma

Figure 7

(3.3)

m(a+b)

mPR,

ma+mb

SUo

The two triangles PQ,R and STU are similar. Corresponding sides
are then proportional, the constant of proportionality being m. Thus
(3.4)

mPR=SU.

Since PR and SU have the same directions, and since m is positive,

then mPR = SUo Substitution in both sides of this equation from
(3.3) yields (3.2).
Now, let m be negative. Then Figure 7 is replaced by Figure 8.
Equations (3.3) apply in this case also. The triangles PQ,R and STU
are again similar, but the constant of proportionality is Iml, so ImlPR
= SUo Since PR and SU have opposite directions and m is negative,
5

rna

Figure 8

then mPR = SUo Substitution in both sides of this equation from

(3.3) again yields (3.2).

4. Subtraction of vectors. If a and b are two vectors, their difference

a - b is defined by the relation
a-b

a+(-b),

where the vector -b is as defined in the previous section. Figure 9

shows two vectors a and b. and also their differenc~ a- b.

Figure 9

5. Linear functions. If a and b are any two vectors, and m and n

are any two scalars, the expression ma+ nb is called a linear function
of a- and b. - Similarly, ma+nb+pc is a linear function of a, b,
and c. The extension of this to the cases involving more than three
vectors follows the obvious lines.
Theorem 1. If a and b are any two nonparallel vectors in a plane,
and if c is any third vector in the plane of a and b, then c can be
expressed as a linear function of a and b.
Proof. Since a and b are not parallel, there exists a parallelogram
with c as its diagonal and with edges parallel to a and b. Figure lO
shows this parallelogram. We note from this figure that
6

(5.1)

Figure 10

c =PQ+QR.

But PQ is parallel to a, and QR is parallel to b. Thus there exist

scalars m and n such that
PQ=ma,

QR

= nb.

Substitution from these relations in (5.1) yields

c = ma+nb.

Theorem 2. If a, band c are any three vectors not all parallel to

a single plane, and if d is any other vector, then d can be expressed
as a linear function of a, band c.
Proof. This theorem is the extension of Theorem I to space. Since
a, band c are not parallel to a single pla,ne, there exists a parallelepiped with d as its diagonal and with edges parallel to a, band c.
Hence there exist scalars m, nand p such that
d = ma+nb+pc.

6. Rectangular cartesian coordinates. In much ofthe theory and application of vectors it is convenient to introduce a set of rectangular
cartesian coordinates. We shall not denote these by the usual symbols
X,Y and .e, however, but shall use instead the symbols Xu X 2 and X3
These coordinates are said to have "right-handed orientation" or to be
"right-handed" if when the thumb of the right hand is made to point
in the direction of the positive X3 axis, the fingers point in the direction
of the 90 rotation which carries the positive Xl axis into coincidence
0

with the positive X2 axis. Otherwise the coordinates are "left-handed".

In V ector Analysis it is highly desirable to use the same orientation
always, for certain basic formulas are changed by a change in orientation. In this book we shall follow the usual practise of using righthanded coordinates throughout. Figure 11 contains the axes of such
a set of coordinates.

T--

J--,~l'-

a,

.L -

-!E--------..-Y
/Q
/ R

/"''''----- '" ------>J"

.J-----~~--------.,.

",

Figure 11

It is also convenient to introduce three vectors of unit magnitude,

one pointing in the direction of each of the three positive coordinate
axes. These vectors are denoted by iI' ~ and ia, and are shown in
Figure 11.
Let us consider a vector a. It has orthogonal projections in the
directions of the positive coordinate axes. These are denoted by ~, a2
and a3, as shown in Figure 11. They are called the 2~P5~!1~~Il"!s of a.
It should be noted that they can be positive or negative. Thus, for
example, al is positive when the angle between a and the direction of
the positive Xl axis (the angle QPS in the figure) is acute, and is negative when this angle is obtuse.
From Figure 11 it also appears that a is the diagonal of a rectangular

parallelepiped whose edges have lengths lall, la 21 and laal. Hence the
magnitude a of the vector a is given by the relation
(6.1)

= Va12+a22+aa2.

From the figure it also appears that

(6.2)

Now the vector PQ)s parallel to i l Because of the definitions of ~

and of the product of a scalar by a vector, we then have the lelation
Po.. = aliI' Similarly QR = a2i 2 and RS = aaia. Substitution in (6.2)
from these relations yields
(6.3)

This relation expresses the vector a as a linear function of the unit

vectors iI, i2 and i a . We note that the coefficients are the components
of a.
Theorem. The components of the sum of a number of vectors are
equal to the sums of the components of the vectors.
Proof. We consider two vectors a and b with components aI, ~, aa,
bl , b2 and ba. Then
a = alil+a~+aaia,
b = b1il +bJ2+baia .
Addition of both sides of these equations leads to the relation

a+b = alil+a2ia+a~ia+blil+b~+baia.
Now the sum of a number of vectors is independent of the order in
which the vectors are added, by Theorem 1 of 2. Hence we may
write the above equation in the form

a+b = lZtil+blil+a2ia+b2ia+aaia+baia.
By the theorem in 3 we may then write this in the form
a+ b = (al +bl)il + (~+b2)ia+ (a3+lh3)ia.
Hence the components of a+b are ~+bl' a2+b2 and aa+ba. This
proves the theorem when two vectors are added. The proof is similar
when more than two vectors are added.
9

7. The scalar product. Let us consider two vectors a and b with

magnitudes a and b, respectively. Let ex. be the smallest nonnegative
angle between a and b. as shown in Figure 12. Then 0 < ex. < 180.

Figure 12

The scalar ab cos ex. arises quite frequently, and hence it is convenient
to give it a name. It is called the scalar product of a and b. It is also
denoted by the symbols a b, and hence we have
a b = ab cos ex..

(7.1 )

The scalar product is sometimes referred to as the dot product.

If the components of a and b are denoted by a] , a2, aa, bl , b2 and ba
in the usual manner, the direction cosines of the directions of a and b
are respectively
aI, a2, aa.
a a a'

By a formula of analytic geometry, we then have

cos ex. =

al

bl

a b

+~
a

b2 + aa ha
b
a b

Su bstitution in (7.1) of this expression for cos ex. yields

(7.2)

a' b = albl+a2b2+aaba.

This relation expresses the scalar product of two vectors in terms of

the components of the vectors.
Theorem 1. The scalar product is commutative; that is,

ab =ba.
Proof.

Because of (7.2), we have

a b = ~bl +a2b2+aab3,
ba = blal+b2a2+bsaa.
10

Since albl = blal , etc., the truth of the theorem follows immediately.
Theorem 2. The scalar product is distributive', that is ,

a (b+c)

a.b+a.c.

Proof. According to the theorem in 6, the components of b+c

are bl+Cl , b2+C2 and b3+C3. Hence, by (7.2) we have

a (b+c) = al(bl +cl ) +a2(b 2 +c2) +a3(b 3+c3)

= albl +~b2+a3b3+alcl +a~2+a3c3
=a.b+ac.
This completes the proof.
If a and b are perpendicular, then

ab =0.
However; if it is given that a b = 0, it does not necessarily follow
that a is perpendicular to b. It can be said only that at least one of the
following must be true: a = 0; b = 0; a is perpendicular to b. Similarly, if it is given that

ab = ac,
it does not necessarily follow that b = c. For this relation can be
written in the form a (b - c) = 0, and hence it can be said only that
at least one of the following is true: a = 0; b = c; a is perpendicular
to the vector b-c.
We note the following expressions, in which a is any vector and
iI, ~ and ia are the unit vectors introduced in 6:

(7.3)

i l i l = 1,
i 2 il =0,
i 3 i l = 0,

a.a
il ~
~.~

ia ~

= a2,
= 0,
= 1,
= 0,

ia = 0,
ia = 0,
iaia=l.
il

~.

8. The vector product. Let us again consider two vectors a and b,

,the smallest nonnegative angle between then being denoted by r:J., as
shown in Figure 12. Then 0 .;;;; r:J. .;;;; 180. The vector product of a
and b is a third vector c defined in terms of a and b by the following
three conditions:
11

(i)
(ii)

c is perpendicular to both a and b;

the direction of c is that indicated by the thumb of the right
hand when the fingers point in the sense of the rotation a from
the direction of a to the direction of b;
(iii) c = ab sin 01:.
These conditions define c uniquely. Figure 13 shows c. The vector

Figure 13

product of a and b is also denoted by a X b. Hence

(8.1)
c = axb.
The vector product "is also called the cross product.
Theorem 1. The area A "of the parallelogram with the vectors a
and b forming adjacent edges is given by the relation
(8.2)
A = laxbl.
Proof. Figure 14 shows the parallelogram. Ifp is the perpendicular
distance from the terminus ofb to the line of action of a, then A = ap.
But p = b sin oc. Hence
A=absina
=

laxbl

Figure 11:

12

We shall now determine the components of the vector product c

in (8.1) in terms of the components ofa and h. Because of condition (i)
above, we have ac = 0, hc = 0. Because of (7.2), these equations
can take the form

+ +

alcl a2c2 aaCa = 0,

blcl + b2c2 + baca = 0.

If these equations are solved for

that
Cl

Cl

and

C2

in terms of Ca , it is found

C2

a2ba-aab2 = aabl-~ba

In order to preserve symmetry, we denote the common value

these three fractions by K, whence we have

Cj = K(a 2ba-aab2),
C2 = K(aabcalba),
Ca = K(a l b2-a2bj ).
Now c2 = C12+C22+C32. Hence
c2 = K2[(a 2ba-aab2)2+ (aa bl-al ba)2+ (a l b2-a2bl )2]
= K2[ a12(b22+ba2) +a22(b32+b12) +aa2 ( b12 + b22)
-2 (a2b2aaba+ a3baal bl + al bl a2b2)] .

(8.3)

The first term inside the square brackets can be written in the form
a12( b2-b 12). If the second and third terms are treated similarly it is
found that
C2 = K2[( a12+a22+aa2) b2- (alb l +a2b2+aaba)2]
= K2[a 2b2-(ab cos 0:)2]
= K 2a2b2(1-cos 2 0:)
= K 2a2 b2 sin 2 0:.
But by condition (iii) above, c2 = a2b2 sin 2 IX. Thus K = 1. If
these two values of K are inserted in (8.3) two vectors c result with
the same magnitude but pointing in opposite directions. Only one of
these vectors satisfies condition (ii) above. Now both values of K are
numerical, and are hence independent of a and h. Thus the same
value of K will satisfy condition (ii) for all vectors a and h. Hence it
13

is only necessary to find K for anyone special case in which e can be

found directly and with ease from conditions (i) - (iii) above. If we
take a = i l and b = ~, it is found from these conditions that e = is.
Thus ~ = b2 = C3 = I, a 2 = a3 = bi = b3 = CI = C2 = 0, and substitution in (8.3) yields K = 1. From (8.3) we then have in general,
(8.4)

CI

= a2b3 -asb2, C2 = aSbCalba ,

Thus the components oj the vector product a

~b2-a2bl' Hence

C3

~b2-~bl'

bare a2bs- a3 b2, a:A -

~b3 ,

or, in determinant form

(8.6)

axb =

Theorem 2.

is

it2

as
bs

b2

The vector product is not commutative, because

axb = -bxa.

(8.7)
Proof.

il

al
bl

By (8.6) it follows that

il

bXa

= bl
itl

is

b2
a2

b3
a3

Since this determinant differs from the determinant in (8.6) only

in that two rows are interchanged, the two determinants differ only
in sign. Hence (8.7) is true. The truth of this theorem can also be
seen easily by examining the three conditions which define the vector
product. According to these conditions the effect of interchanging the
order of a and b is only to reverse the direction of the vector product.
Theorem 3. The vector product is distributive; that is,

(8.8)
Proo

aX (b+e) = axb+axe.
Let us write

d
14

= aX (b+e), e = axb,

f = aXe.

Then

dl =

d 2 (b a+cs ) -lZa(b2 +C2)

(d 2ba-a sb2) + (a 2ca-aaC2)
= el+J;.

Similarly d2 = e2+f2 and da= es+};. Hence d = e+f, and so (8.8)

is true.
We note the following expressions, in which a is any vector and i l ,
i2 and ia are the unit vectors introduced in 6:
aXa = 0,

(8.9)

iIxil = 0,
i 2 xil = -ia,
ia xiI = i 2 ,

i l Xi2

= is,

~x~

= 0,

ia xi 2

= -iI,

iIxia = -i2'
i2 xia = iI,
iaxia=O.

If a and b are parallel, then

axb = 0.
Also, if it is given that a X b = 0, then at least one of the following
must be true: a = 0; b = 0; a is parallel to b. Similarly, if
then aX (b-e) =

axb = axe,

a =

and at least one of the following must be true:

0;

b = c;

a is parallel to b-e.

9. Multiple products of vectors. Let a, band e be any three vectors.

The expression
a (bxe)

is a sca~ar, and is called a scalar triple product of a, band e.

If the components of a, band e are denoted in the usual way, then
the components ofb X e are b2ca-bac2 , bacl-bIca , bIc2-b2cl , and we have
by (7.2)

a (b X e) =

d l ( b~a-baC2) +a2(baCI-bICa) +aa(b Ic2-

b2cl ),

or
(9.1)

a (b X e) =

a1

d2

aa

bi
C1

b2
C2

b3
Ca

15

Theorem 1. The permutation theorem for scalar triple products. If the

vectors in a scalar triple product are sUbjected to an odd number of
permutations, the value of this product is changed only in sign; and
if the number of permutations is even the value of the product is not
changed.
Proof. A permutation of the vectors in a scalar triple product is
defined as the interchange of any two vectors which appear in the
product. From (9.1) it appears that a single permutation produces an
interchange of two rows in the determinant. Since such an interchange
of rows results in a change of sign only, the truth of the theorem is
established.
Because of this theorem we have
a (b X c)

= b (c X a) = c . (a X b)
=

-c (b X a)

-a (c X b)

-b (a Xc).

Theorem 2. The volume V of the parallelepiped with the vectors a,

band c forming adjacent edges is given by the relation

(9.2)

V=

la. (bXc) I,

where the vertical lines here denote the absolute value.

Proof. Figure 15 shows the parallelepiped. Let d = b X c. Then

Figure 15

d = Ib X c I, and by Theorem 1 of 8 the area of the parallelogram

forming the base of the parallelepiped is then d. Hence V = hd,
16

where h is the altitude of the parallelepiped. But d is perpendicular

to the base, and if (J is the angle between a and d, then h = a I cos (J I.
(The absolute value signs are necessary here, since (J lies in the range
0
0 0 -< fJ -< 180 and hence cos (J may be negative.) Thus

I ad cos (J I
lad I
= I a (bxc) I.
=

The expression
aX (bxc)
is a vector, and is called a veclor triple product of a, band c. Let us write
d

b X c,

a X d.

Then e is equal to the vector triple product a X (b X c). By (8.5)

we have
e1 = a2da- aad2
= a2(blc2-bzCl) -aa(baccb1ca)
= b1 (a 2C2 aaca) -cl(a2b2+aaba)'

Because of (7.2), this can be written in the form

e1 = bl(ac-alcl)-cl(a.b-albl)
= b1(a. c) -c1(a. b).
Similarly
e2 = b2 (a.c)-c2 (ab),
ea = b3 (a. c) -ca(a. b).

Hence e
(9.3)

b(a.c)-c(a.b), and since e

ax(bxc)

aX(bxc), we have

= b(a.c)-c(a.b).

This is a rather important identity. It will be used frequently.

We note that the right side of (9.3) is a vector in the plane of b
and c. This is to be expected, since the vector a X (b Xc) is perpendicular to the vector b X c which is itself perpendicular to the plane
of band c.
Let us now consider the expression
(axb) X (cxd).
17

It is a vector. If we regard it as a vector triple product of aX b,

c and d, then by (9.3),

(9.4)

(axb) X (c xd) = c [(aXb) .d]-d [(axb) c].

Since an interchange of the order of the vectors in a vector product

only changes the sign,

(axb) X (c xd) = -(c xd) X (axb).

If we regard the right side of this equation as the vector triple product of c X d, a and b, then by (9.3),

(9.5)

(axb)x(cxd) =-a [(cxd)b]+b [(cxd).a].

We next consider the expression

(axb) (cxd).
It is a scalar. If we consider it as the scalar triple product of a X b,
c and d, and subject these three vectors to two permutations, then
according to Theorem I of 9, we have

(axb).(cxd) =c.[dx(axb)].
If the vector triple product on the right-hand side of this equation is
expanded by the identity in (9.3), we obtain

(axb).(cxd) = (ca)(d.b)-(c.b)(d.a),
or in a form more easily recalled,

(9.6)

(axb). (cxd) = (a.c)(bd)-(bc)(a.d).

There are many other multiple products of vectors. In general these

can be simplified by means of the theorems and formulas above. In
dealing with multiple products of vectors care must be exercised to
avoid writing down expressions which are ambiguous or have not
been defined. Thus, for example, the expression a X b X c is ambiguous since a X (b xc) =1= (a X b) X c, and the following expressions
have not been defined:

ab, a (b. c), a X (b c) .

10.
18

Moment of a vector about a point.

Let a be a vector with origin at

"
a

o
Figure 16

a point X, and let x be a vector with origin at a point 0 and terminus

at X, as shown in Figure 16. The moment of a about the point 0 is by
definition the vector P given by the relation

P
Theorem 1.

xXa.

If P is the moment of a about a point 0, then

P =pa,
where p is the perpendicular distance from 0 to the line of action ofa.
Proof. Now P = xa sin 0(, where 0( is the angle between x and a.
But p = x sin 0(. Hence P = pa.
Theorem 2. The moments. about a point of any two equal vectors
with the same line of action are equal.
Proof. Let a and a' be two equal vectors with the same line of ac-

o
Figure 17

19

tion, and with origins X and X', as shown in Figure 17. Let 0 be any
point, and let ox = x, ox' = x'. Let P and P' be the moments of
a and a' about o. Then
P = xXa,

P' =x'Xa'.

But x' = x+XX'. Hence

P' = (x+XX') xa'
= xxa'+XX'xa'.
Since XX' is parallel to a',XX'xa' = O. Hence, since a'
have finally
P'= xXa= P.

a we

11. Moment of a vector about a directed line. Each line defines two
directions which are opposite. A line is said to be directed when one of
these directions is labelled the positive direction and the other the negative direction.
Let us consider a directed line L, and let b denote a unit vector
pointing in the positive direction of the line, as shown in Figure 18. We

Figure 18

also introduce a point 0 on L and a vector a with origin at a point X.

If P denotes the moment of a about 0, the moment of a about L is by
definition the orthogonal projection ofP on L.1t is a scalar, and ifitis
denoted by Q we have
Q= P cos l>,

20

where <I> is the angle between P and the unit vector b. Hence Q, = b. P.
But P = x X a, where x = OR. Thus

(ILl)

Q,=b.(xxa).

Theorem 1. The above definition of the moment of a vector about

a directed line L is independent of the position of the point 0 on L.
Proof. Let 0' be a second point on L, as shown in Figure 19. Also,

--~------~~------~~------------L
o
0'

Figure 19

let P' be the moment of a about 0', and let Q; be the corresponding
moment of a about L. Then
Q, = b (x X a),

Q;

= b (x' X a) ,

where x' is as shown. But x' = O"O+x. Thus

Q; = b[(O"O+x) xa]
=b(O"Oxa)+Q,.
Since band 0"0 have the same line of action L, then b (0"0 X x) = 0
by Theorem 2 of 9. Thus Q; = Q" which proves the theorem.
Theorem 2. IfP denotes the moment ofa vector a about the origin
of the coordinates, then the three components of P are equal respectively to the moments of a about the three coordinate axes.
Proof. The truth of this theorem follows immediately from the
above definitions of the moments of a vector about a point and about
a line.

21

12. Differentiation with respect to a scalar variable. Let u be a scalar

variable. If there is a value of a vector a corresponding to each value
of the scalar u, a is said to be a function of u. When it is desired to indicate such a correspondence, we write a(u).
Let us consider a general v~lue of the scalar u and the corresponding
vector a(u). Let the vector OP in Figure 20 denote this vector. We

o
Figure 20

now increase the scalar u by an amount flu. The vector corresponding

to the scalar u + flu is a( u + flu). Let the vector OQ)n Figure 20 denote
this vector. The change in a(u) corresponding to the change flu in u
is then a(u+flu) - a(u). In the usual notation of calculus we denote
it by fla, so that
fla = a(u+flu) - a(u).
From the figure it is

s~en

that fla = PQ. Since flu is a scalar, the

" ' as P,><--'

7i Th e vector
vector -~a h as t h e same d
lrection
~u

. " ~a
1ImltLl.u .... 0 ~ U

is the rate of change of a with respect to u. It is also called the derivative of a with respect to u, and is denoted by the symbol
da
du

22

1"Imlt
. ~a
.
u

Ll.u -+ 0 ~

~:,

so that

In precisely the same way, we define the derivative with respect to u

of the vector

~:.

This vector is denoted by

d da
tPa
d)du) or du2

Higher derivatives of a with respect to u are defined similarly.

Let a(u) and b(u) be any two vectors which are functions ofa scalar
u, and let m be a scalar function of u. We shall now derive the following
formulas:
d
da db
(12.1)
du (a+b) = du + du '

da dm
m du + du a,

(12.2)

d
du (ma)

(12.3)

d
db da
d)a. b) = a du + du' b

(12.4)
Proof of (12.1).
the sum a+b is
( 12.5)

d
du (a X b)

db da
du + du

X b.

When u increases by an amount /lu, the change in

/l(a+b) =.(a+/la+b+/lb) - (a+b).

According to Theorem 2 of 2 ):he sum of a number of vectors is independent of the order of summ~tion. Thus the right side of (12.5) can
be written in the form a - a+b - b+Ila+/lb, which reduces to
/la+/lb. Thus
/l(a+b) = /la+/lb.
Ifboth sides of this equation are divided by /lu, and if /lu is then made
to approach zero, (12.1) is obtained.
Proof of (12.2).

When u increases by an amount /lu, the change in

ma is

(12.6)

/l(ma) = (m+/lm) (a+/la) - mao

According to the theorem in 3 the multiplication of a vector by a

scalar satisfies the distributive laws, as exemplified by Equations (3.1)
23

and (3.2). Because of the law exemplified by (3.1) we can then write
(12.6) in the form

(12.7)

Ll (ma) = m(a+Lla) +Llm(a+Lla) - ma,

and because of the law exemplified by (3.2), we can then write (12.7)
in the form
Ll(ma) = ma+mLla+Llm a+Llm Lla - ma
= m Lla+Llm a+Llm Lla.
If both sides of this equation are divided by Llu, and if Llu is then made
to approach zero, (12.2) results.
Proof of (12.3). When u increases by an amount Ll u, the change in
ab is
Ll(a. b) = (a+Lla). (b+Llb) - a b.
Since the scalar product is distributive, this equation may be written
in the form

Ll(a.b)

a b+aLlb+Lla b+LlaLlb - a b

= aLlb+Lla b+Lla. Llb.

If both sides of this equation are divided by the scalar .6.u, we have

.6.(a b) = a . .6.b + l1a. b+ .6.a. Llb .

.6.u
l1u l1u
.6.u
If we now let Llu approach zero, (12.3) results.
Proof of (12.4). This prooffollows exactly the same pattern as the
proof of ( 12.3), and hence will not be given here.
It is important to note that in (12.4) the order in which the vectors
a and b appear must be the same in all terms, since a X b is not equal
to bXa.
Ifa(u) is a vector with components aI, a2 and a3 , then
a =

alil+a2i2+a3i3'

By (12.1) and (12.2) we then have

da
du

d ( .)
du aliI

+dud

( .)

a 212 +

d ( .)
du aal 3

_
q~+ di2+ dia+da l
- al du a2 du a3 du du
24

II

+da2 +daa
du 12 du 13

Now aI' a2 and a3 are scalar functions of u. Also it> i2 and i3 are unit
vectors pointing in the directions of the positive coordinate axis. If they
are the same for all values of u, then
dil = di2 _ dia _ 0
du
du-du-'
and so
da _ dal +da2 +daa

du-du

11

dU

12

dU

13

From this equation we see that the components of the derivative of a

vector are equal to the derivatives of the components, provided the
directions of the coordinate axes are independent of the variable of
differentiation.
'
13. Integration with respect to a scalar variable . . Let a be a given function of a scalar u. We introduce orthogonal unit vectors i l , i2 and i3
with directions independent of u. Then
a(u) = alil+a2i2+a3i3'

We make the definition

(13.1 )
From this definition it follows that

;uI a(u) du = i l a1(u) +i2 a2(u) +i3 a3(u) = a,

as expected, since integration is the inverse of differentiation. It is to be
noted that each integral on the right side of(13.1) gives rise to a constant of integration.
Theorem. If a( u) is a linear function of constant vectors, with coefficients which are functions ofu, then a(u) du can be obtained by formal integration in which constant vectors are treated as are constants
in ordinary integration, and arbitrary constant vectors are inserted
where arbitrary constants would appear in ordinary integration.

25

Proof.

We have
a(u) = pf(u) +q g(u) + ... ,

where p, q, ... are constant vectors and feu), g(u),

functions ofu. By (13.1) it then follows that

are given

j a(u) du = i 1 [Pljf(u) dU+qlj g(u) du+ ... ]

+i2[P2jf(u) du+ "]+ia[P3jf(u) du+ .. ]
=p jf(u) du+qjg(u) du+ ....

If k, I, ... denote the integration constants of the integrals in the last

line, then the total contribution of these constants to a(u) du is the
single arbitrary constant vector C such that

pk+ql+. .

The following examples illustrate the above theorem:

j (pu+q)du = ~pU2+qU+C,
j P cos u du = p sin u+c.
14. Linear vector differential equations.

The equation

(14.1 )
in which a and Po, PI, "', Pn are given functions of the scalar u and x
is an unknown vector, is a linear vector differential equation of order'n.
Let F denote the differential operator in (14.1). Then (14.1) can be
expressed in the form
( 14.2)

F[x] = a.

Theorem. The general solution of the linear. vector differential

equation F[x] = a is x = Y +A, where A is a particular solution of
this differential equation, and
Y = c1Yl+c2Y2+CaY3+'" +cnYn,

26

... , Cn being arbitrary constant vectors and)'1' Y2, Ya, ... , Yn being
n linearly independent solutions of the homogeneous scalar differential
equation F[y] = 0.
Proof. Let us introduce the unit vectors iI' i2 and i3 with directions
independent of u. Then

C;, C2, C a,

lit

F[x]

Q1i1+a2i2+aai3'
xliI + x2i 2+ x3ia ,

= F[X1]i1+F[X2]i2+F[X3]ia.

Hence, from (14.2) we have

(14.3)
Let AI, A2 and A3 denote particular solutions of these three scalar differential equations, and let Y1 'Y2' ... ,Yn denote u linearly independent
particular solutions of the scalar differential equa tion F [y] = 0. Then
the general solutions of Equations (14.3) are

Xl = CllY1+ C12 Y2+ C13Y3+ . +c1nYn+AJ.'

X2= C21Y1+ C22)2+ C23Y3+ . +c2n Yn+ A 2,
X3= c3IY1+ c32Y2+ c33Ya+ +c3nYn+Aa,
where the c's are arbitrary constants. Let us multiply these three equations by iI, i2 and i 3, respectively, and then add. The result can be
written in the form
(14.4)
lit =Y+A,
where
Y = c;Y1+C2Y2+C3Y3+ +cnYn,
A = AliI + A2i2 Aaia,

the vectors Ct, C2' C 3 , .. , C n being arbitrary constant vectors. Equation (14.4) gives the general solution of Equation (14.1). We note that
Y is the general solution of the homogeneous equation F[ x] = 0, and
that A is a particular solution of Equation (14.1). The particular solution A can be found by procedures very similar to those used to find
particular solutions of linear scalar differential equations. This is demonstrated below.
27

As an example, let us find the general solution of the differential

equation
d 2 x dx
.
(14.5)
du 2 - du - 2x = lOp sm u+q(2u+ 1),
where p and q are constant vectors. We must first find two linearly
independent solutions of the equation
2

(14.6)

d y _ dy

du 2

du

_~ =

o.

The auxiliary equation of this differential equation is

m2 -m-2 =

o.

It has roots -1,2, whence the required solutions of (14.6) are eOu and
e2u . Thus
Y = cleou+c2e2u.
We now use the method of undetermined coefficients to find a particular solution A of Equation (14.5). The function on the right side of
(14.5), and the derivatives of this function, are linear functions of sin u,
cos u, u, 1, none of which are particular solutions of (14.6). Hence, we
look for a particular solution A in the form
A = b sin u+c cos u+du+e,

where b, c, d a~d e are constant vectors. By substitution in (14.5) we

readily find by equating coefficients that
3b-c.= -lOp,
b+3c = 0,
d = -q,
d+2e = -q.
Solving these four equations for b, c, d, e, we find that
b = -3p,

c =p,

d =-q,

e =0.

The general solution of Equation (14.5) is then

x = ~eou+c#2u+p(_3 sin u+cos u) -quo

Problems
1. The vectors a, b, c and d all lie in a horizontal plane. Their

28

magnitudes are 1, 2, 3 and 2, and their directions are east, northeast,

north and northwest, respectively. Construct these vectors.
2. If a, band c are defined as in Problem 1, construct the vectors
(a+b) +c, (b+a) +c, c+(a+b), and by measuring their magnitudes
and directions verify that they are equal.
3. Ifa and b are defined as in Problem 1, contruct the vectors a+2b,
2a+b, 3a-b, -2a-2b.
4. If a, band c are defined as in Problem 1, express each of these
vectors as a linear function of the other two, determining the coefficients graphically to two decimal places in each case.
5. Given that
a+2b=m,

2a- b = n,

where m and n are known vectors, solve for a and b.

6. If a and b are vectors with a common origin 0 and terminuses
A and B, in terms of a and b find the vector DC, where C is the middle
point of AB.
7. The vectors a and b form consecutive sides of a regular hexagon,
the terminus of a coinciding with the origin of b. In terms of a and b
find the vectors forming the other four sides.
8. If a, b, c and d have a common origin and terminuses A, B, C
and D, and if b-a = c-d, show that ABCD is a parallelogram.
9. The vectors a, band c have a.common origin and form adjacent
edges ofa parallelepiped. Show that a+b+c forms a diagonal.
10. Vectors are drawn from the center of a regular pentagon to its
vertices. Show that their sum is zero.
11. Consider the vectors a, b, c and d defined in Problem 1. Introduce rectangular cartesian coordinate axes such that the four vectors
lie in the X I X 2 plane with the Xl axis pointing east and the X 2 axis pointing north. Find the components of the vectors a, b, c, d, a+2b, and
3a - 2b; also, express these vectors in terms of their components and
the unit vectors iI, i2 and is.
12. Do Problem 4, making an exact determination of the coefficients
analytically by the use of components.
29

13. Given that

a =i1 +2i2+i a ,
b = 2i 1 +i 2 ,
c = 3i1 - 4i2 - Sip
verify that a.(b+c) =ab+ac, ax (b+c) =axb+axc. Also,
find laxbl, laxb+axcl.
14. Show that (a+b)(a-b) =a2 _b 2
15. Show that (a - b) X (a+b) = 2a X b.
16. The two vectors a = i 1 +3i c 2i a , b = 3i1 +2i 2 -2ia have a
common origin. Show that the line joining their terminuses is parallel
to the X1X 2 plane, and find its length.
17. Show that the vectors a = i 1 +4i 2 +3ia , b = 4i1 +2i2 -4ia are
perpendicular.
18. If a, band c are as defined in Problem 13, find a.(bxc),
(bxa) "C, aX (bxc), (a xb). (axe), (axb) X (a Xc).
19. If the vectors drawn from the origin to three points A, Band C
are respectively equal to the three vectors a, band c defined in Problem 13, find a unit vector n perpendicular to the plane ABC. Hence
find the distance from the origin to this plane.
20. Show that
ax (bxc)+bx (cXa) +c X (axb) = O.
21. Show that
ax[bx (cXd)] = [a. (cXd)]b-(a.b) (cxd)
= (b.d) (axc)-(b.c) (axd).
22. Show that
[a xb]. [(b X c) X (c X a)] = [a. (bx C)]2.
23. Show that
arb. (c xd)] - b[c. (d Xa)]+c[d (ax b)] - d[a (b xc)] = O.
24. Show that
[a.(bxc)](Cxg) =
30

a
Ca
ga

b
Cb
g.b

Cc
gc

25. Show that

n.(bxc)
n.(cxa)
n.(axb)
a+---b+
c.
a.(bxc)
b(cxa)
c.(axb)
This formula can be us~d to express any vector n as a linear function of
any three vectors a, band c not lying in the same plane.
26. Express the vector n = i 1 +2i 2+3ia as a linear function of the
vectors a, band c defined in Problem 13. (See Problem 25.)
27. Express the vector n = 2il - 2i2 - 3ia as a linear function of the
vectors a, band c defined in Problem 13. (See Problem 25.)
28. Show that
ax[(fxb) X (gxc)]+b)<[(fXc) X (gxa)]
+cx[(fXa) X (gxb)] = O.
29. If 0 is the origin of the coordinates and A, Band C are three
points such that
OA = 2i1 +2i 2-ia ,
AB = i1 - i 2+2i a ,
BC = -2il +2i2 - 3ia ,
find (i) the moment of BC about A, (ii) the moment of CB about 0,
(iii) the moment of BC about the directed line OA, (iv) the moments
of BC about the coordinate axes.
30. If a and b are two vectors, prove that a times the moment of b
about the line of action of a is equal to b times the moment of a about
the line of action ofb.
3l. If a(u) has a constant magnitude, show that
da
a-=O.
du
32. Ifa = p cos u+q sin u, where p and q are constant vectors and
u is a variable, show that
n=

a'. [da X d2aJ

= O.
du du 2
'
33. If a is a function of a variable u, show that

da). d3~
.
du

'!"[a. (da X d2a)]

=(ax
du
du du 2
du

31

34. Given that the unit vectors i l , i2 and i3 are independent of a

variable u, evaluate a du when

a =il +2ui 2 +8u 3i 3 ,

(i)
(ii)

a = i l cos

... )
(III

a =
.

2il
4-u2

+i2 sin u - i3 e

2i2 .
4+u 2

35. Evaluate the following integrals, in which p and q are constant

vectors:

(i)
(ii)
(iii)

f (p+qu2)du,
f (p cos u +q sec2u)du,
d
f p+qu
4 -u
U.

36. Find the vector x(u) in each of the following cases, given that
p, q and r are constant vectors:
dx
(i)
- = p u2 +q e2u

(ii)

du
d2 x
du 2
2

(iii)

d x
dw

'

P cos u+q SIn

(p

SIn

U,

u-q cos u) Xr.

37. Find the general solutions of the differential equations

d2 x dx
-2- --6x =0
(i)
du
du
'
d2x
dx
(ii)
du2 4 du
4x = 0,

(iii)
(iv)

d2 x

dx
du 2 - 2 du

d'x _
du'

+ 5x = 0,
6 d x + 11 d x _ 6 dx =
2
3

du 3

du

O.

du

38. Find the general solutions of the following differential equations,

given that p and q are constant vectors:

32

(i)

dx
du - 3x = p(3u2 + 1),

(ii)

d2 x
du 2 -4x

(iii)

dx
du2

2 dx
du

l6p cos 2u,

5 .
= - 6"
pe + q sm u.

39. Find the general solutions of the following differential equations, given that p and q are constant vectors:
d2x
dx
(i)
du 2 - 2 du + x = 2pe",
(ii)

(iii)

2
x
dx
u
- 2 - 3 - = 2pe
du
d2 x
dx
u2 - _
U - - 3x
du 2
du

au

l8qu2

6p.

'

33

CHAPTER

II

APPLICATIONS TO GEOMETRY

15. Introduction. This chapter contains a treatment by vector

methods of various elementary topics in geometry. This treatment is
included in the present volume for two reasons. First, it indicates the
ease and power which vector methods lend to studies in geometry.
Secondly, it affords the student an opportunity to gain additional skill
in the use of the vector operations introduced in the previous chapter.
Proofs of some well-known theorems of plane geometry will first be
given. Then a fairly broad treatment of solid analytic geometry will
be presented, in which some of the more familiar formulas will be deduced in vector form and by vector methods. Finally, the differential
geometry of curves in space will be considered briefly. For more complete treatments of analytic geometry and differential geometry by
vector methods, the reader is referred to the excellent books by F. D.
Murnaghan, 1 W. C. Graustein 2 and C. E. Weatherburn. 3
16. Some theorems of plane geometry. In this section we shall consider
proofs by vector methods of two well-known theorems of plane geometry.
Theorem I. The diagonals of a parallelogram bisect each other.
Proof. Let us consider the parallelogram OABG in Figure 21. The
diagonals cut each other at the point D. We must prove that D bisects
both of the line segments OB and AG.
For convenience we denote the vectors drawn from 0 to the points ,
F. D. Murnaghan, Analytic geometry, Prentice-Hall, New York, 1946.
W. C. Graustein, Differential geometry, The Macmillan Company, New York,
1935.
C. E. Weatherburn, Differential geometry of three dimensions, Cambridge
University Press, Cambridge, England. Vol. 1, 1927. Vol. 2, 1930.
1

34

c
~--------~------~B

o
Figure 21

A, B, C and D by a, b, c and d, respectively. Since D lies on the line

OB, there exists a scalar u such that
(16.1 )

d =ub.

Also, from the figure we see that

(16.2)
Since- D lies on the line AC there exists a scalar v such that

AD =vAC.
Hence we can write (16.2) in the form
(16.3)

= a+vAC.

We now equate the above two expressions given for d in (16.1) and
(16.3), obtaining

(16.4)

a+vAC = u b.

The next step is to express all vectors in this equation as linear functions
of any two vectors in the plane, say a and c. From the figure we see
that
b = a+c,
AC =-a+c,
whence (16.4) becomes

a+v (-a+c)

= u (a+c) ,

(l-u-v) a

= (u-v)

or
c.

35

Since a and c do not have the same line of action it then follows that
1 - u - v = 0,

u- v = 0.

We solve these equations foru and v, obtaining u = v = l. Thus Equation (16.1) becomes d = i b, from which it follows that D is the middle
point of OB.
We have now proved that the point of intersection D of the diagonals is the middle point of one of these diagonals. From symmetry, D
must also be the middle point of the other diagonal.
Theorem 2. The medians of a triangle meet in a single point which
trisects each of them.
Proof. Let us consider the triangle OAB in Figure 22. The points
B

O~-----~------">IA

E
Figure 22

C, D and E are the middle points of the sides, and F is the point of
intersection of the medians AD and BE. We must prove that F is a
point of trisection of each of the three medians AD, BE and OC.
For convenience we denote the vectors drawn from a to the points
A, B, C, D,E and F by a, b, c, d, e and f, respectively. Now F lies on
the median AD. Thus there exists a scalar u such that
(16.5)

DF=uDA

and we then have

f= OD+uDA.
36

Similarly, since F lies on the median BE, we have

f = OE+vEB,

where v is some scalar. We now equate these two expressions for f,

obtaining
( 16.6)
The next step is to express all vectors in this equation as linear functions of any two vectors in the plane, say a and b. From Figure 22 we
see that
(16.7)

OD = t b,
DA =-t b+a,

OE
EB

= ta,
=-ta+b.

Thus (16.6) becomes, after substitution from (16.7) and collection of

like terms,

(t-u-tv)a = (t-tu-v)b.
Since a and b do not have the same line of action, it follows that

t - u - tv = 0,

t-tu-v =0.

We solve these equations for u and v, obtaining u = v = t. Thus

Equation (16.5) becomes DF = 1 DA, from which it follows that F is
a point of trisection of the median DA.
We have now proved that the point of intersection F of two medians
is a point of trisection of one of these medians. From symmetry, Fmust
also be a point of trisection of the other of these medians. From symmetry F must also be a point of trisection of the third median.

Solid AnalYtic Geometry

17. Notation. We shall now consider the analytic geometry of
points in space. A right-handed set of rectangular cartesian coordinates is introduced, with origin at a point O. Just as in Chapter I, we
denote these coordinates by the symbols Xl' x 2 and x s , as shown in
Figure 23. Unit vectors il> i2 and i3 pointing in the directions of the
positive coordinate axes are also introduced, as shown.

37

X(x" x" x.)

~-------.------------.~

x.
Figure 23

To denote a general point in space we shall use the letter X, and we

shall call the vector OX the position-vector of X. For convenience we
shall denote this vector also by the symbol x, and its components I:>Y
the symbols (xu X 2 , xa). We then have
x = xliI +x2i 2+xai 3 .
The quantities Xl' x 2 and X3 are also the coordinates of the point X.
We shall use the letters A, B, C, ... to denote specific points in space,
and shall denote the position-vectors of these points by a, b, c, .... The
component of these vectors will be denoted in the usual way by the
symbols (aI' a2 , aa), (bl , b2 , ba), (CI , C2 , ca), ... We note that these
quantities are also the coordinates of the points A, B, C, .. '.

18. Division of a line segment in a given ratio. Let us suppose that we

are given two points A and B, and that it is desired to find a third
point C which divides the line segment AB in the given ratio m to n.
Figure 24 illustrates the problem. If C lies between "A and B, then
0< min < + 00; if C lies beyond B then - 00 < min < -1; if C lies
beyond A, then -1 < min < O. In any event we have

38

Figure 24

(18.l )

AC

CB

-=-.
m
n
If we now denote the position-vectors of A, Band C by a, band c,
respectively, then
AC = c - a,
CB = b - c,
and (18.1) can then be written in the form

n(c-a) = m(b-c).
Solving this equation for c, we obtain

mb+na
m+n
This formula expresses the position-vector c of the desired point C in
terms of the known quantities a, b, m and n.
In books on Analytic Geometry, formulas are usually given which
express the coordinates of C in terms of m, n and the coordinates of
A and B. It should be noted that (18.2) is entirely equivalent to these
formulas, for these formulas can be deduced from (18.2) simply by
equating the components of the left side of (18.2) to the components of
the right side of(18.2).
( 18.2)

c=;=

19. The distance between two points. Let us suppose that A and Bare
two given points, and that it is desired to find the distance d between
A and B in terms of the position-vectors a and b of A and B. Figure
25 illustrates the problem. Now
39

d= IABI.

But AB = b - a.
d2

Thus
= IABI2 =AB'AB = (b-a).(b-a)

whence

= V (b -

a) . (b - a) .
B

Figim 25

20. The area oj a triangle. Let us suppose that A, Band C are three
given points, and that it is desired to find the area 6.ahc of the triangle
ABC in terms of the position-vectors a, band c of A, Band C. Figure
26 illustraties the problem.

B
Figure 26

We first construct the parallelogram of which AB and AC form two

adjacent edges, as shown. Then 6.ahc is equal to one half the area of
this parallelogram. But, by Theorem I of 8, the area of this parallelogram is IAB X AC I. Hence we can write
(20.1)

40

6.ahc = t<P,

where
(20.2)
Now AB

cp

b-a, AC

ABXAC.

c-a. Thus

cp

(b-a)x(c-a).

This simplifies to
(20.3)

cp

bxc+cxa+axb.

The required area of the triangle is thus given by (20.1), qJ being the
magnitude of the vector given by (20.3).
A property of the vector cp will now be recorded, for future use.
Since this vector is equal to AB X AC we conclude that the vector cp is
perpendicular to the plane of the triangle ABC, and its direction is that indicated
by the thumb of the right hand when the fingers are placed to indicate the direction
of the passage around the triangle from A to B to C.

21. The equation of a plane. There are several ways in which a plane
can be specified. For example, three points which are on the plane and
do not lie on a single straight line can be given, or a line in the plane
and a point on the plane but not on the line can be given. In each of
several such cases we sh~ll now deduce the equation which must be
satisfied by the positiori-vector x of every point X on the plane. This
equation will be referred to simply as the equation of the plane. In books
on analytic geometry the equation of a plane usually appears as an
equation which involves scalars only, and is satisfied only by the coordinates of points on the plane. We shall refer to this latter equation
as the cartesian form of the equation of the plane.
(i) To find the equation of the plane through a given point and perpendicular
to a given vector. Let A be the given point and b be the given vector.
Figure 27 illustrates the problem, the plane P being the plane in
question.
Let X be a general point on P, and let a and x denote the positionvectors of A and X, respectively. Now AXis perpendicular to b. Thus
AXb =0.

41

But AX = x - a, whence it follows that

(x-a)b

(21.1 )

=0.

This is the desired equation of the plane P.

o~~

_____________
x,

x,
Figure 27

The cartesian form of the equation of the plane P can be obtained

readily from (21.1). It is only nt>t:~ssary to express (21.1) in terms of
the components of the vectors involved. In this way we obtain the
equation
(Xl -~) bl +(X2 - a 2) b2 +(X3 - a3 ) b3 = O.
(ii) To find the equation of the plane through three given points. Let A, B
and C be three given points. It is desired to find the equation of the
plane P containing these three points. Figure 28 illustrates the problem.
Let X be a general point on the plane P, and let a, b, c and x denote
the position-vectors of the points A, B, C and X, respectively. In 20
we saw that the vector cp given by the relation

42

cp = bxc+cxa+axb
is perpendicular to the plane P. Hence, by Problem (i) above the
equation of P is

(21.2)

(x-a).cp

= O.

Since acp =a.(bxc+cxa+axb) =a.(bxc),

Equation (21.2) can be written in the equivalent form

(21.3)

xcp = a.(b

c).

~:o-

_______

%.

Figure 28

22. The vector-perpendicular from a point to a plane. The vector-perpendicular from a point D to a plane P is the vector with origin at D and
terminus at the point on P nearest D.
(i) To find the vector-perpendicular from a point D to a plane P through a
given point and perpendicular to a given vector. Let A be the given point
and let b be the given vector. Figure 29 illustrates the problem. We
denote the position-vectors of the points A and [) by a and d, respectively. If the point E is the foot of the perpendicular from the point D
to the plane P, then DE is the vector-perpendicular from the point D to
the plane P. We shall denote it by the symbol p. It is required to find
p in terms of a and b.

43

Figure 29

Now p and b are parallel. Thus there exists a scalar constant K

such that
p =Kb.

(22.1 )
From the figure we see that
or
(22.2)

d+Kb+EA - a

O.

Now b is perpendicular to EA. Hence b'EA

plication of (22.2) by b yields
(22.3)

(d-a).b+Kb2 =

Thus
K=

44

(a-d)b
b2

o.

0, and so scalar multi-

and substitution for K in Equation (22.1) then yields

(22.4)

(a-d).bb
b2

In particular, if the point D is at the origin, then d = 0 and

(22.5)
(ii) To find the vector-perpendicular from a point D to a plane P through three
given points. Let A, Band C be the three given points, with positionvectors a, band c, respectively. Figure 30 illustrates the problem. It
is desired to find the vector-perpendicular p in terms of a, b, c and d.

~-------::Oi.

B
O~~~

_______I

f)

Figure 30

In 20 we saw that the vector

(j'

(j'

given by the relation

=bxc+cxa+axb

is perpendicular to the plane P. Hence we may regard P as the plane

through the given point A and perpendicular to the given vector (j'.
Thus, from Equation (22.4) it follows that the required vector-perpendicular is given by the relation

45

(a-d) cp
p=---cp.
cp2

Since a cp = a (b Xc) it follows that

(22.6)

a (b X c) - d cp
cp2

cpo

23. The equation of a line. There are several ways in which a line in
space can be specified. For example, two points on the line can be
given, or two planes through the line can be given. In each of several
such cases we shall now deduce the equation which must be satisfied
by the position-vector lit of every point X on the line. This equation will
be referred to simply as the equation of the line.
(i) To find the equation of the line through a given point and parallel to a
given vector. Let A be the given point, with position-vector a, and let b
be the given vector. Figure 31 illustrates the problem, L being the line
in question.

o~'~

/
x,

_______________

Figure 31

Let X be a general point on L, and let

of X. Now AX is parallel to b. Thus

lit

AX X b = O.

46

denote the position-vector

But AX

x - a, whence it follows that

(23.1 )

(x-a)Xb =0.

This is the desired equation of the line L.

(ii) To find the equation ofthe line through two given points. Let A and B be
the given points, with position-vectors a and b, respectively. Figure 32

0/E~

x,

_____________

Figure 32

illustrates the problem, L being the line in questIOn. Now L is parallel

to the vector AB, and AB = b - a. Thus, by Problem (i) above, the
desired equation of L is
(23.2)

(x - a) X (b - a)

O.

(iii) To find the equation of the line through a given point and perpendicular
to two given vectors. Let A be the given point with position-vector a, and
let band c be the given vectors. Figure 33 illustrates the problem, L
being the line in question. Now L is parallel to the vector b X c. Hence,
by Problem (i) above, the desired equation of Lis
(23.3)

(x-a)X(bXc) =0.

(iv) To find the equation of the line through a given point and perpendicular
to the plane through three given points. Let A be the given point on the line,
and let B, C and D be the given points on the plane. Figure 34 illustrates
47

o
Figure 33

the problem, Land P being the line and plane in question. We denote
the position-vectors of A, B, C and D in the usual manner. Let us consider the vector (j) given by the relation
(j) = cXd+dXb+bXc.

Figure 34

According to 20, this vector is perpendicular to the plane P. Thus L

is the line through A and parallel to (j), and hence by Problem (i) above
the desired equation ofL is
48

(23.4)

(x-a)xcp=O.

24. The equation of a sphere. Let S be a sphere of radius a with

center at a point C, as shown in Figure 35. If X is general point on
the sphere S, then
exex = !ex!2 = a2.

But

ex =

x-c.

Thus

(24.1 )

(x-c)(x-c) =a2

This is an equation satisfied by the position-vector of a general point X

on the sphere. It is thus the equation of the sphere.

o
We shall now prove the following well-known property of a sphere:
the angle at the surface of a sphere subtended by a diameter of the
sphere is a right angle. For convenience, the origin of the coordinate
system is chosen at the center of the sphere, as shown in Figure 36. Let
D .and E be points at the ends of a diameter, and let X be a general
point on the sphere. We denote the position-vectors of these points in
the usual manner. From the figure
DX

x-d,

EX

x-e.

49

Thus
DXEX= (x-d)(x-e) =x.x-x(d+e)+d.e.

But d+e=O, and if a denotes the radius of the sphere then xx = ai,
de = _ai. Hence it follows that
DXEX =0,

and so DX is perpendicular to EX.

j)

Figure 36

25. The tangent plane to a sphere. Let S be a sphere of radius a with

center at a point C, as shown in Figure 37. We shall now find

o
Figure 37

50

the equation of the plane P which touches S at a given point X.

Let Y be a general point on P. We denote the position-vectors of the
various points in the usual manner. From the figure it follows that
CY =CX+XY,

or
y-c =CX+XY.

We now multiply this equation scalarlybyCX. NowCX.CX = a2 , and

CX XY = 0 because CX is perpendicular to XY. Thus we have
(25.1)

But CX

CX(y-c) =a2

x- c. Thus Equation (25.1) becomes

(x - c) .(y - c)

= a2

This is the desired equation of the plane P.

Differential Geometry

26. Introduction. "'Ie shall consider here only a small portion of the
differential geometry of curves in space. Rectangular cartesian co-ordinates Xu X 2 and Xs are introduced, with origin at a point O. The quantities Xl' x2 and Xs denote the coordinates of a general point X with
position-vector x. If iI' i2 and is are unit vectors in the directions of
the positive coordinate axes, then as before,
(26.1)

= xlil+X2i2+X3i3'

A curve consists of the set of points the position-vectors of which satisfy

the relation
x = x(u),
where x(u) is a function of a scalar parameter u. We shall consider only
those parts of the curve which are free of singularities of all kinds.
If the set of points comprising a curve aU lie in a single plane, the
curve is said to be a plane curve. If this set of points does not lie in a
single plane, the curve is said to be a skew curve.
It is convenient to choose as the scalar parameter the length s of the
arc of the curve measured from some fixed point A. The quantity s is
51

positive for points on one side of A, and negative for points on the
other side of A. The equation of the curve may then take the form
x = x(s).
The derivatives with respect to s of the function x(s) will be denoted by
x', x", x"', etc.

27. The principal triad. Let us consider a general point X on a curve

C. The position-vector of X is x. We shall now define a set of three orthogonal unit vectors j1' j2 and j3 at X. They are functions of the parameter S.J and their derivatives respect: to s will be denoted in the usual

o
Figure 38

way by the symbols j/, j2 I and j3'. They are shown in Figure 38, and
are defined by the conditions:
(i) jl is tangent to the curve C, and points in the direction of s
increasing;
(ii) j2lies in the plane of the vectors jl and jl', and makes an acute
angle with jl' ;
(iii) js is such that the vectors jl' j2 and j3 form a right-handed triad 1.
1 At points on the curve where j/ is equal to zero, these conditions are not sufficient,for a unique determination of j. and j.. We exclude such points from consideration here.

52

The straight line through the point X and parallel to j2 is called the
principal normal to the curve. The straight line through X and parallel
to ja is called the binormal to the curve. The vectors jl, j2 and ja are
called the unit tangent vector, unit normal vector, and unit binormal vector,
respectively. The triad formed by these vectors is called the principal
triad. The plane through X and perpendicular to jl is called the normal
plane. The plane through X and perpendicular to ja is called the
osculating plane.

28. The Serret-Frenetformulas. Let Y be a point on the curve C and

near the point X, as shown in Figure 38. We denote the length of the
arc XY by the symbol ils, and the vector XY by the symbol ilx. Let m
now consider the vector
X/=

limit

!J.s ..... 0

ilx

ils

Now
limit lilxl
!J.s ..... 0
ils

1.

Thus x' is a unit vector. Further, the vector ilxjils lies along XY, and
its direction then tends to that of jl as ils tends to zero. Since jl is a unit
vector, we can then write
(28.1 )
The vectors jl" j2' and ja' can each be expressed as ~ linear function of any three non-coplaner vectors. In particular, they can be expressed as linear functions of the vectors jl, j2 and ja, and we then
have relations of the form
(28.2)

jl' =alljl +a 12 j2+ alaja ,

j2' =a 2dl +a 22 j2+ a 2aja ,
ja' =aadl +aa2j2+ aaaja ,

where the scalar coefficients are functions of the parameter s. Since

the vectors jl, j2 and ja are orthogonal unit vectors, they satisfy the
relations

53

(28.3)

jljl=l,
j2-j1 = 0,
ja -jl = 0,

jd2= 0,
jd2= 1,
ja-j2=0,

jd3=0,
j2-ja=0,
ja-ja= 1.

We differentiate with respect to s the first equation in the first line

of (28.3). This yields the relation
jljl' +jl' -jl = 0.
Since in a scalar product the order in which the vectors appear is immaterial, we can interchange the vectors in the second scalar product.
It then follows that
jl-jl' =0.
If we substitute here for jl' from the first equation in (28.2), and then
make use of Equations (28.3),we find that an =0. Similarlya22 =aaa=0,
and we may write
(28.4)
We now differentiate with respect to s the second relation in the first
line of (28.3). This yields
jl -j2' +jl' -j2 = 0.
If we substitute here for jl' and j2' from the first two equations in

(28.2), and then make use of Equations (28.3), we find that a12 +a 21 = 0.
Similarly we can find two similar relations, and we have altogether
(28.5)

. a I2 +a 21 =0,

a23 +aa2=0,

a31 +a I3 =0.

So far, only conditions (i) and (iii) above have been used. By condition (ii) the vector-jl' is to be in the plane of jl and j2. This can be
true only if

(28.6)
ala =
By condition (ii), the vector jl' is to make an acute angle with j2. If
this angle is denoted by oc, then cos oc must be positive. But
Ijl'l cos oc=jdl'
If we substitute here for jl' frrun Equations (28.2) and then use Equa-

tions (28.3) we find that

54

Thus
(28.7)

We now define two quantities x and" by the relations

(28.8)

Then, by (28.7) it follows that

(28.9)

> 0,

and because of Equations (28.4), (28.5), (28.6) and (28.8), we can now
express Equations (28.2) in the form

j1' = Xj2,
j2' = "ti3 - Xj1'
js' = - "ti2'
These are the Serret-Frenet formulas. They were given originally by
Serret (1851) and Frenet (1852) in an equivalent form which did not
involve vectors. The quantities x and ", which are functions of the arc
length s of the curve C, will be considered in some detail in the next
section.
(28.10)

29. Curvature and torsion. The quantity x appearing in Equations

(28.10) is called the curvature,ofthe curve. It can be shown that
x

. M
= 11m1t-,
/1, -+ 0

.:ls

where .:l6 is the angle between the tangents to the curve C at the points
X and Y in Figure 38. Thus x is the rate at which the tangent at the
pomt X rotates as X moves along the curve. The reciprocal of x is
called the radiuJ of curvature, and will be denoted by the symbol p.
The quantity" appearing in Equations (28.10) is called the torsiop.
of the curve. It can be shown that
" =

. M)
11m1t-,
/1t -+ 0

.:ls

where .:leI> is the angle between the binormals to the curve C at the
55

points X and Y in Figure 38. Thus l' is the rate at which the unit binormal at the point X rotates as X moves along the curve. The reciprocal of l' is called the radius of torsion, and will be denoted by cr.
To find x, we note from Equation (28.1) that
jl' = x".

jl = x',

(29.1)

Substitution from the second of these relations for jl' in the first of the
Serret-Frenet formulas then yields the equation
x" = xj2.

(29.2)

We now multiply each side of this equation scalarly by itself, obtaining

)(2

x" . ,,".

Because of (28.9), x is positive, and so

(29.3)

x. =

To find the torsion

to s. This yields

l'

vi,," .,,".

we differentiate Equation (29.2) with respect

x'" = xj2' +Xj2.

We now substitute for j2' from the second Serret-Frenet formula, obtaining
. . )+'.
(29.4)
x ", =x (1'Ja-XJl
XJ2.
From Equations (29.1), (29.2) and (29.4) it now follows that
x' (x" xx"') = jl[Xj2X (x-rja - X2jl+Xj2)]
= jl . [x 21' jl + x3jaJ
= X 21'.
Thus
(29.5)

't'

= "'2 x' (x" X x''') .

x

The curvature and torsion can be computed from Equations (29.3)

and (29.5). We can express these equations in different forms. Now
x = xlii +XJ2 +Xaia,
x' = xl'i l +x2'i2+xa'ia ,
x" = xI"il +x2"i2+xa"ia ,

,,'" =
56

xl'''il-t-X2'lfi2+Xa'''ia

Substitution from these relations in Equations (29.3) and (29.4) then

yields
(29.6)

It

= y;-,,2
+ x----:-:2+X
,,2
1
2
3,

(29.7)
Since It can now be found, we can obtain the unit tangent vector jl
and the unit normal vector j2 by use of Equations (29.1) and (29.2).
The unit binormal vector j3 can then be found easily, since it is equal
to j 1 X j 2. We have the collected results
(29.8)

Jl

= X,

J2

=;;:1 X

II

J3

= -It x' X

x" .

Let us now find the equation of the tangent to the curve at the point

X. If Y is ~ general point on this tangent, the desired equation is easily

seen to be

(y- x) Xjl = O.
Because of Equation (29.8), this can be written in the form
(29.9)

(y- x) Xx'

O.

In the same way, the equations of the principal normal and. binormal
can be found in the forms
(29.10)

(y-x) Xx" =0,

(29.11)

(y-x) X (x' xx") =0.

The equation of the normal plane at the point X is easily seen to be

(y - x) . jl

= o.

Because of Equation (29.8), this can be written in the form

(29.12)

(y-x)x'=O.

In the same way we can find the equation of the osculating plane in the
form
(y - x) . (x' X x") = o.
(29.13)

57

Problems

1. Prove that the line joining the middle points of any two sides of a
triangle is parallel to the third side, and is equal in length to one half
the length of the third side .
.2. Prove that the lines joining the middle points of the sides of a
quadrilateral form a parallelogram.
3. If 0 is a point in space, ABC is a triangle, and D, E and Fare
the middle points of the sides, prove that
OA+OB+OC = OD+OE+OF.

4. If 0 is a point in space, ABCD is a parallelogram, and E is the

point of intersection of the diagonals, prove that
OA+OB+OC+OD = 40E.

5. Prove that the line joining one vertex of a parallelogram to the

middle point of an opposite side trisects a diagonal of the parallelogram.
6. Prove that it is possible to construct a triangle with sides equal
and parallel to the medians of a given triangle.
7. Prove that the lines joining the middle points of opposite sides of
a skew: quadrilateral bisect each other. Prove also that the point
where these lines cross is the middle point of the line joining the middle
points of the diagonals of the quadrilateral.
8. Prove that the three perpendiculars from the vertices of a triangle
to the opposite sides meet in a point.
9. Prove that the bisectors of the angles of a triangle meet in a
point. Hint: the sum of unit vectors along two sides lies along the bisector of the contained angle.
10. Prove that the perpendicular bisectors of the sides of a triangle
meet in a point.
11. If 0 is a point in space and ABC is a triangle with sides oflengths
l, m and n, then
l OA+m OB+n OC

= (l+m+n)

OD,

where D is the center of the inscribed circle.

12. If ABC is a given triangle, the middle points of the sides BC, CA

58

and AB are denoted by D, E and F respectively, G is the point of intersection of the perpendiculars from the vertices to the opposite sides,
and H is the center of the circumscribed circle, prove that

iG

= 2 HD,

BG = 2 HE,

GG = 2 HF.

Hence prove that

In problems 13-22 the points A, B, C and D have the following coordinates: A(-l, 2, 3), B(2, 5, -3), C(4, 1, -1), D(l, 3, -3).
13. Find the position-vectors of the points of trisection of the line
segment AB.
14. Find the distance between the points A and B.
15. Find the area of the triangle ABC.
16. Find the cartesian form of the equation of the plane through A
and perpendicular to OB.
17. Find the cartesian form of the equation of the plane through
(i) the origin and the points A and B, (ii) the points A, Band C.
18. Find the distance from the point D to the plane through A and
perpendicular to OB.
19. Find the distance from D to the plane ABC.
20. Find the cartesian form of the equation of the line through A
and parallel to BG.
21. Find the cartesian form of the equation of the line through
A andB.
22. Find the cartesian form of the equation of the line through
D and (i) perpendicular to the plane through A, Band C, (ii) perpendicular to BG and OG.
23. A plane passes through a given point A with position-vector a,
and is parallel to each of two given vectors band c. Derive the equation of this plane in the form
(x-a).(bXc) =0.
24. A straight line L passes through a point A with position-vector
a, and is parallel to a vector b. A vector p has its origin at a point C

59

with position-vector e, its line of action along the perpendicular from

C to L, and its terminus on L. Show that

=a-

(a-e)h

e - --b2-- h .

25. A straight line L passes through a point A with position-vector

a, and is parallel to a vector h. A second straight line L' passes
through a point A' with position-vector a', and is parallel to a vector
hi. The vector p runs from L to L' along the common perpendicular.
Show that
(a'-a)e
p=
c2
e,
wheree =hxh'.
26. Prove that if the torsion of a curve is equal to zero, the curve is
a plane curve.
. 27. Prove that

,. +("
-2)"J2 +(2 x't'
'+ X't' ')"J3'
x "" =-3XXJl
X -x3 -x,
28. If the position-vector x of a general point on a curve is given as
a function of a parameter t, and if primes denote differentiations with
respect to t prove that
1

x = -S'2 V x" x" "

S"2

x'

JI = s' '
Also, derive the equations of the tangent, principal normal, binormal,
normal plane and osculating plane in the following forms:

60

tangent,

(y-x)xx' =0;

principal normal,

(y- x) X (x" -

binormal,
normal plane,
osculating plane,

(y-x) x (x'xx") =0;

(y- x) x' = 0;
(y-x)(x/xx") =0.

I;s x')

= 0;

29. The position-vector x of a general point on a circular helix is

given by the relation

= a cos t i1+a sin t i 2 +at cot rJ. is,

where a and rJ. are constants, and t is a parameter. Find p, (j and the
principal triad. Answer: p = a cosec 2 rJ., (j = 2a cosec 2rJ., jl = sin rJ.
(-il sin t+i2 cos t+ia cot rJ.), j2 = -il cos t- i2 sin t, ja = cos rJ. (i] sin t
- i2 cos t+ia tan rJ.).
30. The position-vector x of a general point on a curve is given by
the relation
x = a(3t - ta)i 1 +3at2i 2+a(3t+t3)ia,
where a is a constant and t is a parameter. Find p, (j and the principal
triad. Answer: p = (1 = 3a yS, v2j] = y-l(rJ. i 1+~i2+yia), j2 = y-l(-~il
+et i 2), v'Z""ja = y-l(_et i] - ~ i 2 +y is), where rJ. = 1 - t2 , ~ = 2t,
Y = 1+t2.

31. The position-vector x of a general point on a curve is given by

the relation
x =

a[(t-sin t)i 1 +(1- cos t)i2+tia],

where a is a constant and "t is a parameter. Find p and (j. Answer:

p = a et ai2 , [3-:tI2; (j = -a~, where et = 3 - 2 cos t, ~ = 2 - 2 cos t
+cos2 t.

61

CHAPTER

III

APPLICATION OF VECTORS TO MECHANICS

Motion of a particle
30. Kinematics of a particle. The phrase "kinematics of a particle"
refers to that portion of the study of the motion of a particle which is
not concerned with the forces producing the motion, but is concerned
rather with the mathematical concepts useful in describing the motion.
Let us consider a moving particle. It is necessary to introduce a
"frame of reference" relative to which the motion of the particle can
be measured. For a frame of reference we take a rigid body. Such a
body is one having the property that the distances between all pairs
of particles in it do not vary with the time. We then introduce a set of
rectangular cartesian coordinate axes fixed in the frame of reference.
Figure 39 shows these axes and the associated unit vectors iI' i2 and i a

~---"'-------'"
o

"
62

Figure 39

Let the curve C in this figure be the path of the particle, and let the
point X denote the position of the particle at time t. The vector OX
is the position-vector of the particle. We denote this vector also by x.
It is a function of the time t.
The velocity v of the particle relative to the frame of reference, and
the acceleration a of the particle relative to the frame of reference,
are defined by the relations
(30.1)

dv d2x
a= dt = dt 2 '

dx
v= dt'

The magnitude v of the velocity v is called the speed of the particle.

Thus, velocity is a vector and speed is a scalar. We shall now compute
various sets of components of the vectors v and a.
(i) The components of the velocity and acceleration in the directions of rectangular cartesian coordinate axes. Let Xl, X 2 , Xa denote the rectangular
cartesian coordinates of the point X in Figure 39. Then
x

xlil+X2i2+Xsis'

If we now adopt the convention that a single superimposed dot denotes a first time derivative, and a pair of superimposed dots denotes
a second time derivative, then
v =

dx
.
'.
'.
dt
= Xlll+X212+X313,

a =

de =

dv

..

".

..

XIII +X212+X31 3

Thus the desired components of v and a are

(30.2)
Xl, 2, X3;
Xl' X2, X3
(ii) The components of the velocity and acceleration in the directions of the
principal triad of the curve traced out by the particle. The curve C in Figure
39 is the path of the particle. Let jl' j2 and j3 denote the principal
triad at the general point X on C. The principal triad was discussed
in 27. If s denotes the arc length of C, then from Equations (28.1)
and (28.10) we have

dx
(30.3)
Jl = ds'
63

where x is the curvature of C. Now

dx

dx

v = dt = ds s,
and because of the first equation in (30.~) we then have
(30.4)
v = S jl'
Thus the velocity of the particle is directed along the tangent to its path, and the
speed is v = S.
Because of (30.4) we have
dv
djl
a

= dt =

00.

SJ1+ S

dt'

But because of the second equation in (30.3) we have

djl

Tt

djl .
ds s =

o.
XSJ2 ,

and hence
(30.5 )
Thus the acceleration a lies in the osculating plane of C. Also, the components
in the directions of the tangent, normal and binormal are

of a

(30.6)

dv
s=v=vds'

'0

v2

XS2

= xv2 = p ,

0,

where p is the radius of curvature of C.

(iii) The components of the velocity and acceleration in the directions q[ the
parametric liTles of cylindrical coordinates. Let r, 6, Xa be cylindrical coordinates of a general point X on the path C of the particle. Figure 40
shows these coordinates. We introduce a triad of unit vectors kl' k2'
k3 at 0 as shown; kl points toward the point X' which is the projection of X on the X1X 2 plane, ka is equal to ia, and k2 is such that the
triad is right-handed. It will be noted that kl' k2' ka point in the
directions of the parametric lines 1 of the cylindrical coordinates r, 6,
Xa at X.
1 It will be recalled ~hat the directions of the parametric lines of a coordinate
system at a point X are those directions in which one of the coordinates increases
while the other coordinates do not vary.

64

-"3
~~-----

---- -".

Figure 40

If iI, i2 and i3 are the usual unit vectors associated with the rectangular cartesian coordinate axes in Figure 40, then

kl = il cos 6+i2 sin 6,

k2 = -il sin 6+iz cos 6,
k3 = i 3,
and so

dkl = dkl
dt
de

(30o 7)

dk2
dt

(.
6
6) 6 = k 26 ,
-11 sm +12 COS

= dk2
de 6

(.
6

-11 cos
- 12
sin 6) 6 = - k 16 ,

dk3 -0
dt
From Figure 40 it follows that the position-vector x of the particle is
given by the relation x = rk1+x3k30 Thus
dx

= de

:1_

dkl

= rJl.l+ x3k 3+ r 7t+ X3

dk3

Tt .

Because of (30.7) we then obtain

65

(30.8)
Thus the desired components of v are
(30.9)
From (30.8) we find that
dv..
....
..
a = di = rk1 +(r6+r6)k 2+x}t3
. dk 1 6 dk2 . dk3
+r (ft+r (ft+X3"dt'
Substitution in this equation from (30.7) then yields
(30.10)

a = (r - r( 2)kl+(2rB+r6)k2+X3k 3'

Hence the desired components of a are

(30.11)

r- ,62,

2;6+r6

=;

;t (r 6),
2

.,

X3 .

Of course, if the particle is confined to the XIX 2 plane then X3

= 0,

r = x, and we find from (30.9) and (30.11) that the components ofv

and a in the directions of the parametric lines of the plane polar coordinates x, 6 are
.
. 1 d
.
x, x6', x - x6 2, -x -dt (x 26) .
(30.12)

31. Newton's laws. The concept of force is intuitive. We can define a unit force as that force which produces a standard deflection of
a standard spring. Hence we can assign a numerical value to the magnitude of any force.
We know that forces have magnitude and direction. It has been
verified experimentally to within the limits of experimental error that
forces obey. the law of vector addition. Hence we shall assume that
forces are vectors. The sum of two or more forces is sometimes called
the resultant of the forces.
The term "mass of a body" refers to the quantity of matter present
in the body. We can define a unit mass as that mass which, when sus66

pended from a standard spring at a standard place in the earth's

gravitational field, produces a standard deflection of the spring.
Hence we can assign a numerical value to the mass of any body.
We now introduce the laws governing the motion of a particle.
These laws, which were first stated by Isaac Newton and are called
Newton's laws, are as follows:
(i) Every particle continues in a state of rest or uniform motion in a
straight line unless compelled by some extenial force to change that
state.
.
(ii) The product of the mass and acceleration of a particle is proportional to the force applied to the particle, and the acceleration is in
the same direction as the force.
(iii) When two particles exert forces on each other, the forces have
the same magnitudes and act in opposite directions along the line
joining the two particles.
In the second law, the acceleration of the particle enters. This acceleration depends on the frame of reference employed. It thus appears
that Newton's second law cannot apply in all frames of reference.
Those frames of reference in which this law does apply are called
Newtonianframes of reference. A frame of reference fixed with respect to
the stars is Newtonian, and in making an accurate study of any motion
such a frame of reference should be used. However, for many problems
we may consider the earth as a Newtonian frame of reference, when
effects due to the motion of the earth are negligible.
Let us now consider a particle of mass m acted upon by a force F.
Let a denote the acceleration of the particle relative to a Newtonian
frame of reference. Then according to Newton's second law

k m a,

where k is a constant of proportionality. It is customary to choose units

of length, mass, time and force so that k is equal to unity. We then
have
F =ma.
(31.1 )
There are three such systems of units in general use. These are indicated in Table 1, together with abbreviations commonly used for these

67

units. Thus, for example, when a force of one pdl. acts on a particle
with a mass of one lb., the acceleration of the particle is one ft./sec. 2
The systems of units in the second and third columns of Table I are
called foot-pound-second sytems, or simply f.p.s. systems. The sy~
stem of units in the fourth column is called the centimeter-gramsecond system, or simply the c.g.s. system.
e.g.s.

f.p.s.
Unit
Unit
Unit
Unit

of length
of mass
of time
of force

foot (ft.)
pound (lb.)
second (sec.)
poundal (pdl.)
TABLE

foot (ft.)
slug
second (sec.)
pound-weight (Ib.wt.)

centimeter (em.)
gram (gm.)
second (sec.)
dyne

I. Systems of units used in mechanics.

The lb. wt. is the force exerted on a mass of one lb. by the earth's
gravitational field. If G denotes the acceleration due to gravity, expressed in ft./sec. 2, then
.
I lb. wt. = G pdl. ,
1 slug

G lb.

At points near the surface of the earth, G is approximately equal

to 32.
Equation (31.1), which governs the motion of a particle, may also
be written in the equivalent forms
(31.2)

dv
F=m-

dt '

32. Motion of a particle acted upon by a force which is a given function q(

the time. When the force F acting on a particle is a given function of
the time, Equations (31.2) can be solved by integration for the v~locity
v and position-vector :It of the particle.
As an example, let us suppose that
F = 12p+qcost,

68

where p and q are given constant vectors. Because of the first equation in (31.2) we then have
dv
m dt = 12p+q cos t,

whence

f( 12p+q cos t) dt.

mv =

We now carry out this integration in the manner outlined in 13,

obtaining
(32.1 )

mv = 12pt+q sin t+r,

where r is an arbitrary constant vector.

Now v = dx/dt. Thus from (32.1) we have
(32.2)

mx = f(l2pt+qsint+r) dt

= 6pt2 - q cos t+rt+s,

where s is an arbitrary constant vector. The arbitrary constant vectors rand s can be found if the initial values of x and v are known.
If these initial values are Xo and vo, it is readily found that
r = mvo ,

33. Simple harmonic motion. Let 0 be a point fixed in a Newtonian

frame of reference. Let us consider a particle moving under the action
of a force directed toward 0, the force having a magnitude proportional
to the distance from the particle to O. If x denotes the position-vector
of the particle relative to 0, then the force F acting on the particle
satisfies the relation
F = -kx,

where k is a constant. From Equation (31.2) we then have

d2 x
-kx = m dt 2 '

or
d2 x
-d2

k
+ m-x
=0.
69

This is a differential equation of the type considered in 14. According

to the procedure demonstrated there, the general solution of this differential equation is
x

C1

cos

11k

V;; t +

C2

. 11k

sm

V;; t ,

where C 1 and C 2 are arbitrary constant vectors. These arbitrary constant vectors can be found if the initial values of x and v are known.
If these initial values are Xo and vo, it is readily found that
.
C2

V~vo,

whence we have
x = Xo cos

. 1VII; t .
V;; t+vo 1VI;;I sm

11k

It will be noted that x is a linear function of Xo and Vo ; hence it follows

that the motion of the particle is confined to the plane P containing
the given vectors Xo and Vo. This result could have been anticipated,
for the force F acting on the particle has no component perpendicular to the plane P.

34. Central orbits. Let 0 be a point fixed in a Newtonian frame of

reference. Let us consider a particle acted upon by a force F directed
toward 0, the magnitude of F being a function of the distance from 0
to the particle. The path of the particle is called a central orbit. It will
be noted that the problem considered in 33 dealt with a special type
of central orbit.
Denoting the vector OX by x, we then have
F =F(x).
The equation of motion is

F =ma.
Let Xo and Vo be the initial values of x and the velocity v. The entire
path of the particle will be in the plane P containing the vectors
Xo and yo. Let x and 6 be polar coordinates in this plane. The components ofF in the directions of the parametric lines of these coordi70

nates are -F, O. Also, the components of a in these directions are given
in Equation (30.12). Hence we have
(34.1)

-F=m(x-xe2 ),

(34.2)

md

0 = -;

dt (x 26) .

These are two equations from which x and 6 can be determined as

functions of the time t. It is more convenient, however, to determine
from (34.1) and (34.2) a single equation by elimination of the time
variable. This single equation will now be deduced.
We first introduce a variab1ey defined by the relation
(34.3)

y = 1Jx.

Then from (34.2) we have

~-2

const.

whence
(34.4)
Then

(34.5)
By substitution in (34.1) for x, 6 and
we finally obtain

x from

(34.3), (34.4) and (34.5),

(34.6)
Now F is a function ofy alone. Once the form of this function has been
assigned, we can find the path of the particle by solving Equation
(34.6).
Let us now consider the special case when F varies inversely as the
square of x. Then we can write

= ymy2,

where y is a constant, and Equation (34.6) becomes

71

d2y

d6 2

+y

= h2

'

The general solution of this equation, expressed in terms of x, is

(34.7)

x1 = hY +
2

C1

Il'

6
cos v+c 2 sm ,

where C1 and C2 are arbitrary constants. These constants can be found

if the initial values Xo and Vo are known. It can be shown that (34.7)
represents either an ellipse, parabola or hyperbola, depending on the
values of Xo and Vo'

Motion of a system ofparticles

35. The center of mass of a system ofparticles. Let us consider a system

of N particles. We denote their masses by the symbols m1, m2, m 3 >
... ,mN' The total mass m of the system is then given by the relation
N

m=:Emj.

(35.1)

j= !

We denote the coordinates of the particle of mass mj by the symbols

(Xj1' Xj2, Xja)' The position-vector Xj of this particle then satisfies the
relation
Xj = Xj1 i l +Xj2 i 2 +xja ia
(j = 1, 2, ... , N).
We have then a set of 3N scalars Xjk (j = 1, 1, 3, "', N; k = 1, 2, 3)
which denotes the coordinates of the particles.
The center of mass of the system of particles is defined to be the
point C with position-vector Xc determined by the equation
N

(35.2)

Xc =

~
j=!

mj Xj.

The center of mass is sometimes called the mass center or centroid

or center of gravity.
If the distances between the individual particles in a system remain
unaltered, as mentioned previously the system is called a rigid body.
A rigid body often consists of a continuous distribution of matter, and
in this case the summations in Equations (35.1) and (35.2) above must
72

then be replaced by integrations. Thus, if p is the density of matter

in the body, V is the region occupied by the body, dV is the volume
of an element of the body and x is the position-vector of a point in
dV, then
(35.3)

=f

p dV,

=f

p x dV.

(35.4)

mxc

36. The moments and products of inertia of a system of particles. Let us

first consider a single particle of mass m. Let I denote the length of the
perpendicular from the particle to a line L, as shown in Figure 41.
m

L~

Figure 41

The moment of inertia of the particle about the line L is defined to be

the scalar I given by the relation
1= m l2.

Let p and q denote the lengths of the perpendiculars from the particle
to two perpendicular planes P and Q" as shown in Figure 42. The
product of inertia of the particle with respect to these two planes is
defined to be the scalar K given by the relation
K=mpq.

The moment of inertia about a line of a system of particles is defined

to be the sum of the moments of inertia about the line of the individual
particles. Also, the product of inertia with respect to two planes of

73

J:

Figure 42

a system of particles is defined to be the sum of the products of inertia

with respect to the two planes of the individual particles.
We now consider a set of N particles and introduce a set of rectangular cartesian coordinate axes with origin at a point O. As in
35 we denote the masses of these particles by mj U = 1,2,3, .. " N)
and their coordinates by the 3N symbols Xjk (j = 1, 2, "', N;
k = 1, 2, 3). The moments of inertia of this system of particles about
the three coordinate axes are denoted by 11 , 12 , and 13 , It is easily
seen that
N

11 =

~ mj (X2j2+X~'3) ,
j = I

(36.1)

12

= ~ mj (X2j3+X2it) ,

13

j= I

~ mj (X2i! +X2i2)
j= I

The products of inertia of this system of particles with respect to the

three coordinate planes, taken in pairs, are denoted by Kl , K2 and Ka.
I t is easily seen that
N

(36.2)

Kl

=j =~ I mj Xh Xj3

K2

=j =~ I mj Xia Xi!

K3

=. ~

N
J = I

74

mj xiI Xh .

Of course, if the system of particles forms a rigid body, the summations

in Equations (36.1) and (36.2) must be replaced by integrations.
If 11, 12 , 13 , K1 , K2 and K3 are known, the moment of inertia 1 of
the system about any line L through the origin can be found easily.
L

Xa

mj

o"....=-________ X.
x,

Figure 43

In order to prove this we let Pi denote the length of the perpendicular

from the particle of mass mj to the line L, as shown in Figure 43. Then
N

1=

~ mjPi2

j = 1

But from the figure we see that

pj

Xj

sin () =

lb X Xjl ,

where Xj is the position-vector of the particle of mass mi' xi is the

magnitude of xi' b is a unit vector on L, and () is the angle between
".i and b. The components of the vector b X Xj are
b 2 Xj3 - b 3 Xj2,

b 3 x jI -

bi

Xj3,

bi

Xi2 - b 2 XiI'

and hence p/ is equal to the sum of the squares of these three components. Thus
N

1= ~
j

mj [( b 2xj3 - b 3 Xj2)2

+ (b 3 XjI -

bi

Xj3)

+ (b

Xj2 - b 2 XjI)2]

= b12j ~= 1mj (Xj22+Xjl) +b 22j=~ 1mj (Xj32+Xjl)

75

N
+

ba2 L

N
mj (Xj12+Xj22) -

2 b2ba L

mj Xj2 Xja

j=l

j= 1

- 2 ba b1 L

j=l

N
mj Xj3 Xjl -

2 b1 b2 L

mj Xjl Xj2

j=l

Because of (36.1) and (36.2) we then have

(36.3)

1= 11 b12 +12 b22 +1a bi - 2 KI b2 ba - 2 K2 ba bi

2 Ka bi b2

It will be noted that bl , b2 and ba, which are the components of the
unit vector b on the line L, are also the direction cosines of L. Equation (36.3) is the desired equation which permits a simple determination of the moment of inertia I of a system about any line L through
the origin, once 11,12 , 1a., KI , K2 and Ka have been found.
If it happens that KI = K2 = Ka = 0, the coordinate axes are said
to be principal axes of inertia at the point O. It can be pr.oved that at
every point there is at least one set of principal axes of inertia. I In
many cases, principal axes of inertia can be deduced readily by considerations of symmetry of the system of particles. For example, at
the center of a rectangular parallelepiped the principal axes of inertia
are parallel to the edges of the body.
We shall now state without proof two theorems the proofs of which
are very simple and may be found in almost any text book on calculus.
The theorem of perpendicular axes. If a system of particles lies entirely
in a plane P, the moment of inertia of the system with respect to a line
L perpendicular to the plane P is equal to the sum of the moments of
inertia of the system with respect to any two perpendicular lines
intersecting L and lying in P.
The theorem of parallel axes. The moment of inertia I of a system of
particles about a line L satisfies the relation
1= 1'+m 12 ,

where l' is the moment of inertia of the system about a line L' parallel
to L and through the center of mass of the system, m is the total mass
1 See J. L. Synge and B. A. Griffith, Principles of Mechanics, McGraw-Hill Book
Co., 1942, pp. 311-321.

76

of the system, and l is the perpendicular distance between Land L'.

In most books of mathematical tables there are listed the moments
of inertia of many bodies with respect to certain axes associated with
these bodies. By the use of such tables together with Equation (36.3)
and the above two theorems, it is frequently possible to determine
rapidly the moment of inertia of a body with respect to any given line.
37. Kinematics of a rigid botly. Let us consider a rigid body which
is rotating about a line L at the rate of w radians per unit time. The
body is said to have an angular velocity. We can represent this angular
velocity completely by an arrow defined as follows: its length represents the scalar to some convenient scale; its origin is an arbitrary
point on L; its line of action coincides with L; it points in the direction
indicated by the thumb of the right hand when the fingers are placed
to indicate the sense of the rotation about L. To prove that angular
velocity is a vector, it is then only necessary to prove that it obeys the
law of vector addition. This will be done presently. We shall anticipate this result, and denote angular velocities by symbols in boldfaced type.
We shall first determine the velocity v of a general point X ina body
which has an angular velocity w. Let 0 denote a point on the line of
action of w, and let x denote the vector OX, as shown in Figure 44.

Figure 44

Let e denote the angle between wand x, and let p denote the length
of the perpendicular from X to the line of action of w. The displacement dx of the point X in time dt has the following properties:
(i)
(ii)

its direction is perpendicular to both wand x;

its direction is that indicated by the thumb of the right hand
77

when the fingers are placed to indicate the sense of the rotation
6 from w to x;
(iii) its magnitude is pwdt, which is equal to xwdt sin 6, x being the
magnitude of the vector x.
In view of the definition in 8 of the vector product of two vectors,
it then appears that d x = w X x dt. Thus

dx

Tt=wxx.
If the point 0 is fixed in a frame of reference S, the velocity v of the
point X relative to S is then

v = wxx.

(37.1)

On the other hand, if the point 0 has a velocity u relative to a frame

of reference S, the velocity of the point X relative to Sis
(37.2)

v=u+wXx.

We shall now prove that angular velocity obeys the law of vector
addition. Let us consider a body which is rotating simultaneously
about two lines Land L' which intersect at a point 0 fixed in a frame
of reference S. These angular velocities can be represented by the
arrows wand w' in Figure 45. Let X be a general point in the body,

LEz

/0

'"

Figure 45

with position-vector x relative to O. The two angular velocities impart

to X the two velocities w X x and w' X x which, being vectors, can be
added to yield the resultant velocity
(37.3)

= wXx+~'Xx.

To complete the proof we must show that (37.3) can be written in the
form v = w" X x, where w" is an arrow obtained by the application

78

of the law of vector addition to the arrows wand w'. Even though
angular velocity has not been assumed to satisfy the law of vector
addition, Equation (8.5) may be applied to the two products in (37.3)
to yield
VI

W 2 X3 - W3 X 2 +W'2 X3 - W'3 X 2

(W 2 +W'2) X3 -

and two similar expressions for

V2

(w 3 +w'3) X 2 ,

and v3 Hence we can write

v=w"XX,
where w"is an arrow having components wl+wl', w2+w'2, W 3 +W'3.
But these are the components of the vector obtained by applicai:ion
of the law of vector addition to the arrows wand w'. Hence w" is
equal to the vector sum of wand w', and so angular velocity is a
vector.
It will be noted that two angular velocities can be added only when
their lines of action have a point of intersectiol1, and that the line of
action of the sum passes through this point of intersection.

38. The time derivative of a vector. Let us consider a set of rectangular cartesian coordinate axes with origin 0 fixed in a frame of reference
S, and with axes rotating relative to S with angular velocity w. Then
the line of action of w passes though o.
If iI, i2 and i3 are the usual unit vectors associated with these coordinate axes, then the velocities relative to S of the terminuses of these
vectors are

But by the previous section these velocities are

w XiI'

WXi2,

W X i3

Hence

(38.1 )

di3

Tt =

WX13

Let a be a vector with components aI' a2 , a3 relative to the rotating

coordinate axes. Then

79

a = a l i l +a 2 i 2+a3 i3

and the time derivative of a, relative to S, is then

da
da l _ da2_ da 3_
di l
di2
di3
dt = d i l l +diI2+-~itI3+aldt+a2dt+a3 dt .

Because of Equations (38.1) we can write the iast three terms in the
form
which reduces to w X a.

Hence we can write

(38'.2)

da
aa
- =-+wXa
dt
at
'

where
_ dali + da 2i + da 3i
at - dt I dt 2 dt 3'

aa

(38.3)

Equation (38.2) expresses dajdt as the sum of two parts. The part
aajat is the time derivative of a relative to the moving coordinate
system. The part w X a is the time derivative of a relative to S when
a is fixed relative to the moving coordinate system.
When the origin of the coordinate system is not at rest relative to S
but has a velocity u, Equations (38.1) still hold, and hence also does
Equation (38.2).
39. Linear and angular momentum. Let us consider a particle of mass
m, with a position-vector x relative to a point 0 fixed in a frame of
m

o
Figure 46

reference S. Let v denote the velocity of the particle, as shown in

Figure 46. The linear momentum of the particle is a vector M defined
by the relation

80

M=mv.

The angular momentum of the particle about the point 0 is by definition the moment of M about O. We shall denote it by the symbol h.
Hence, by 10 where the moment of a vector about a point is considered, we have
h

xxM

xXmv

= mxXv.

Let us now consider a system of .N particles. As' before, we denote

the mass and position-vector of the j-th particle by mj and Xj, respectively. Also, we denote the velocity of this particle relative to S by Vj'
Then for this system the linear momentum M and the angular momentum h about 0 are defined by the relations
(39.1)

L mjVj,
j=1

L mjxjXvj.
j=1

Theorem. The iinear momentum of a system of particles is equal

to the product of the total mass of the system and the velocity of the
center of mass of the system.
Proof. The position-vector Xc of the center of mass of the system
is given by Equation (35.2). We differentiate this equation with
respect to the time t, obtaining

m dxc

dt

mj dXj.
dt

j= 1

But
dxc

Tt =

where

Vc

dXj
Vc,

dt =

Vj,

is the velocity of the center of mass C relative to S. Hence

N

(39.2)

mvc = L mjvj = M.
j=1

This completes the proof.

Let us now suppose that the'system of particles constitutes a rigid
body, and that the body is rotating about the point 0 which is fixed
in the frame of reference S. The body then has an angular velocity w

81

with a line of action which passes through O. The velocity relative to

S ofthej-th particle in the body is then
Vj = ooXXj,

and by Equation (39.1) the angular momentum of the system about

o is then
N

h = 2:

(39.3)

mj

Xj X (00 X Xj)

j= !

Because of the identity (9.3), we can then write

N

h = 2: mj[oox/-Xj(Xj. 00)].
j=!

Now let us introduce coordinate axes with origin at the point 0 fixed
in S. The directions of these coordinate axes need not be fixed in S.
As before we denote the coordinates of the j-th particle by (Xji' xj2 ,
Xja). The component hi of h then has the value
N

hI = 2:

mj

j=!

[<ill

(XjI2+Xj22+Xja2) - Xjl

(Xjl<ill +Xj 2 <il 2 +Xja<ila)]

<ill 2:

11 <il 1 - Ka<il2 - K 2<il a ,

<il2 2:

mj (Xj22+Xja2) -

N
mj Xjl Xj2 -

j=!

j=!

<ila 2:
j=l

mj Xja Xjl

where ~l' K2 and Ka are moments and products of inertia defined in

36. There are similar expressions for h2 and ha . We have finally
hI

(39.4 )

h2
ha

= 11 <il 1 - Ka<il2 - K 2<il a ,

= -Ka<ill + 12 <il 2 - Kl<il a ,
= -K2<ili - KI<il2+1a<ila .

Let us now consider a rigid body which is moving in a general

fashion relative to a frame of reference S. Let us introduce coordinate
axes with origin at the center of mass C of the body. The directions
of the coordinate axes need not be fixed in the body, however. We
may consider the body as having a velocity of translation Vc plus an
angular velocity about a line through C. Then, as seen in 37, the
velocity ofthej-th particle can be expressed in the form
Vj

82

vc+oo X "i.

Hence the angular momentum h of the system about the center of

mass C has the value
N

h = ~ mj XjX (Vc+W XXj)

j=!

(_~ mj Xj) Xvc+ i

} =!

mj XjX (WXXj).

j=!

By Equation (35.2) the first sum is equal to mxc. Since the origin of
the coordinate system and the center of mass C of the body coincide,
Xc = O. Thus
N

= j=!
~ m- xJ J

(w xx-)
.
J

The right side of this equation is the same as the right side of Equation
(39.3). Hence in the present case the components ofh are also given
by Equations (39.4).
We have then the important result: Equations (39.4) may be usedfor
the determination of the components of the angular momentum h of a rigid
body about either a fixed point 0 in the body or the center of mass C of the body.
In the two cases the origin of the coordinates is at 0 and C, respectively, the
directions of the coordinate axes being quite general. Equations (39.4)
cannot be used in the case of the angular momentum of a rigid body
about a moving point which is not the center of mass of the body.

40. The motion of a system of particles. Let us consider a genetal

system of N particles. Let mj denote the mass of the j-th particle, and
let Vj denote its velocity relative to a Newtonian system. The forces
acting on the j-th particle can be divided into two groups called
internal forces and external forces. Internal forces are those due to other
particles in the system. External forces include all other forces. Let
Fjk denote the internal force exerted on the j-th particle by the k-th
particle, and let Fj denote the total external force exerted on the
j-th particle.
Theorem 1. The rate of change of the linear momentum of the
system is equal to the sum the external forces acting on the system.
83

Proof. Applying to the j-th particle Newton's Second Law as

stated in 31, we have
(40.1 )
We now sum the.N equations in (40.1), obtaining
N

(40.2)

~
j = I

dv'
mj _ 1

dt

~ Fj+ ~
j= I

~ Fjk.

j= I k = I

Because of Newton's Third Law, as stated in 31, Fjk = -Fjk Thus

the double sum in Equation (40.2) vanishes, and we can then write
(40.2) in the form

dM
dt
'
where M is the linear momentum of the system and F is the sum of
the external forces acting on the system.
Theorem 2. The center of mass of a system of particles moves like
a particle with a mass equal to the total mass of the system acted upon
by a force equal to the sum of the external forces acting on the system.
Proof. In 39 we saw that M = mvc, where m is the total mass of
the system, and Vc is the velocity of the center of mass of the system.
Thus Equation (40.3) can be written in the form
(40.3)

--=F

(40.4)
This completes the proof.

41. The motion of a rigid botty with a fixed point. Let us now consider a system of particles which constitutes a rigid body with a point
o fixed relative to a Newtonian frame of reference.
Theorem 1. The rate of change of the angular momentum of the
body about 0 is equal to the total moment about 0 of the external
forces.
Proof. Let us introduce coordinates with origin at O. Then
(41.1)

84

v.
J

dXj,

dt

where vi> Xj and t have the usual meanings. By Equation (39.1), the
angular momentum h of the body about the fixed point 0 is
N

h=Lmxxv.
j=IJ]
J'
and so

dh

-=A+B
dt
'

(41.2)

where
N
h.
A = L mj -d1 X Vj ,
j= I
t

B =

N
h.
L m X X _ J
j= I J J
dt

Because of Equation (41.1) we have

N

A = L mjvjXvj = O.
j=1
Equation (40.1) gives an expression for mivj/dt. Because of this we
have
N

B = L XjX (Fj + L Fjk ) = G+H,

j=1
k=1
where
N

G = j=
L IxxF
]
J'

(41.3)

H = L L xjXFjk .
i=1 k=1
It will be recalled that Fj is the external force acting on the j-th particle and Fjk is the internal force exerted on the j-th particle by k-th
particle. We note that G is the sum of the moments about 0 of the
external forces. Now
N

H = L L XjXF;k = L L XkxFkj.
j=1 k=1
k=1 j=1
Thus
N

(41.4)

2 H = L L (Xj X Fjk+Xk X Fkj)

j= I k= I

85

But F kj = -Fjk

Thus (41.4) becomes

2H

J{

J{

1:

(Xj- Xk) X

Fjk

j= 1 k=l

Since the lines of action of the vectors Xj - Xk and Fjk coincide, their
vector product vanishes. Hence H = 0, and B = G, so Equation
(41.2) reduces to the form
(41.5)
where h is the angular momentum of the system about the fixed
point 0, and G is the total moment about 0 of the external forces.
This completes the proof.
We have placed the origin of the coordinate system at the fixed
point O. Let us now choose as coordinate axes a set of principal axes
of inertia of the body at O. (Principal axes of inertia are defined in
36.) Then the products of inertia K1, K 2, Ka all vanish, and from
Equations (39.4) we obtain for the components of the angular momentum h ofthe body about 0 the expressions
hI

(41.6)

= l 1 w1 ,

h2

= 12w2 ,

ha

= 13 w3'

where 11 ,12 ,13 are the moments of inertia ofthe body about the coordinate axes, and WI' W 2 , W3 are the components of the angular velocity W of the body about O.
In most cases the coordinate axes will be fixed in the body and will
hence have an angular velocity w about O. However, in a few special
cases when the body has a certain symmetry it will be found possible
and desirable to choose coordinate axes not fixed in the body. To
include such special cases we denote the angular velocity of the axes
about 0 by 02, which mayor may not differ from w. According to
Equation (38.3) we then have

dd~ = ~~ +o2xh

or
(41. 7)

dd~

iz1i1

+iz2i2+iz3i:i+(02h3 - 03h2) i l

+ (OA - 0lh3) i 2 +(Olh 2 - O:A) ia .

86

I'rom this equation we can read off the components of the vector
dh/dt. According to Equation (41.5) these components are equal to
the components of G. Hence we have the equations
~1+02ha - Oah2 = G1 ,
h2+0ahl - 0lha = G2 ,

ha+ O lh2 - 02hl

Ga ,

where G1 , G2 , Ga are the components of G. Because of Equations

(41.6) these relations can be written in the form
l/iJ 1 - 12w20a+1awa02 = G1 ,
12 w2 - 1awaOl +11w1 0 a = G2 ,
1a wa - 11WI02+12W201 = Ga

(41.8)

In the case when the coordinate axes are fixed in the rigid body,
then Q. = wand so (12.8) reduce to the form
11 w1 - (12 - 1a) W 2W a = G1 ,
12 2 - (la - 11) wawl = G2,
1awa - (11 - 12) W 1W 2 = Ga.

(41.9)

These equations are called Euler's equations of motion.

Theorem 3. The total moment about 0 of the gravity forces acting
on a system of particles is equal to the moment about 0 of a single
force equal to the resultant of the gravity forces and acting at the
center of mass of the system.
Proof. Let us introduce a coordinate system with origin at the
point O. Let k be a unit vector in the direction of the gravity forces.
Then the gravity force acting on thej-th particle is mjgk, and the total
moment about 0 of the gravity forces is
G'

= j -iI 'Xjxmjgk = (~mj

'Xj) xgk.
j= 1

But by Equation (35.2) we have

N

2: mx = m Xc

j-I

87

where m is the total mass and Xc is the position-vector of the center of

mass. Thus
G' = m:x:c xgk = Xc X (mgk).
But Xc X (mgk) is the moment about 0 of a single force mgk equal
to the resultant of the gravity forces and acting at the center of mass
C of the system. This completes the proof.
Example 1. A sphere of radius a is placed on a rough plane which makes
an angle rJ. with the horizontal, and is then released. Find the distance the
sphere moves down the plane in time t.
Figure 47 shows the configuration of the system at a general time t.

mgj

Figure 47

The center of mass of the sphere is at its geometrical center. The

point Q is the initial point of contact of the sphere and plane, and in
the time t the sphere has rolled through an angle <P, as shown. The
point on the sphere which is in contact with the plane is at rest.
Hence the sphere has a fixed point, and we select this point as the origin
o of the coordinate system. We must select the coordinate axes to
coincide with principal directions of inertia of the sphere at O. This
requirement is satisfied if the unit vectors i 1 and i3 are chosen as
shown in the figure. The unit vector i2 is then perpendicular to i 1 and
i a , and points up from the page. We note that the coordinate axes
are not fixed in the body, and that consequently Equations (41.8)
apply.
The external forces acting on the sphere consist of gravity and the
reaction of the plane. Because of Theorem 3 above, the forces exerted
88

by gravity on all the particles of the sphere may be replaced by a single

force mgj acting at C, as shown, wherej is a unit vector. The reaction
of the plane is a force which may be resolved into a force N i3 normal
to the plane and a force - Til along the plane, as shown. The moment
G of the external forces about 0 is given by the relation
G

= OCxmgj,

since the moments of Nand T about 0 are equal to zero. But

OC

= ai3, j = il sin

('J. -

i3 cos

('J..

Thus
=

ai3 X mg (il sin

mgai2 sin ex,

('J. -

i3 cos ex)

so

(41.10)

G1 =0,

G2 = mga sin ('J.,

Ga =0.

The coordinate axes have no angular velocity, so

(41.11)
The angular velocity w of the sphere about

is

w = <1> i 2 ,

where the superimposed dot denotes differentiation with respect to t.

Thus
(41.12)

WI

= 0,

W2

= <1>,

W3

= 0.

The moments of inertia of the sphere about the coordinate axes are
(41.13)
II = 12 = tma2, 13 = tma2.
We now substitute in Euler's equations (41.8) from Equations
(41.10), (41.11), (41.12) and (41.13) to obtain the relation

tma2<1>i2 = mgai2 sin ex.

Thus

i> _
-

5g sin ('J.
7a
'

and two integrations then yield

89

'" = 5g sin QI;

'V

14a

t,

-since <I> = <I> = 0 when t = O.

If z is the distance the sphere has rolled down the plane in time t,
then z = act> and we have

z = !4 gt2 sin

(41.14)

QI;.

Example 2. Two shafts are attached to the comers A and G of a rectangular

plate ABGD, in such a way that the axes of the shafts are continuations of
the diagonal AG. The shafts tum in two bearings each at a distance c from the
center of the plate. The system is made to rotate at a constant rate of <xl radians
per unit time. Find the forces exerted on the bearings.

The system is shown in Figure 48. We choose the center 0 of the

D

I'

-R

Figure 48

plate as the origin of the coordinate system. We also choose the unit
vectors il and i2 parallel to edges of the plate, as shown. The unit
vector i3 is then perpendicular to the plane of the plate. The directions
of these three vectors are principal directions of inertia of the plate
at o. The moments of inertia of the plate about the coordinate axes
are

II = i ma 22, 12 = imal2, 13 = im(a I2+ai) ,

where m is the mass of the plate, and 2a1 and 2a 2 are the lengths of
the edges.
(41.15)

90

The plate has an angular velocity w the line of action of which is

the diagonal AC and the magnitude of which is the given constant
w. If IX is the angle between wand iI' then
(41.16)
and
Thus
(41.17)

WI

cos IX,

w2

= w sin

IX,

W3

= 0.

Since the coordinate axes are fixed in the body, Euler's Equations
(41.9) apply. We substitute in these equations from (41.15) and
(41.17), obtaining the relations

G1 =0,

G2 =0,

G3 = tm(a I2 -a 22)w2 sin IX cos

IX.

Thus the moment G about 0 of the external forces is normal to the

plate and rotates with it. Hence the forces exerted on the shaft by the
bearings must be in the plane of the plate. Let us denote these forces
by Rand -R, as shown in Figure 48. We must then have

2cR

= G3

whence we find that

(41.18)

R = ic m (al- a22 ) w 2 sin IX cos IX

= .1 mw2 a1 a2 (a 12 - a22) .
6
c(aI2+a22)

By Newton's third law ( 31), the forces exerted on the right and
left bearings are -R and R, with magnitudes R given in Equation
(41.18) above.

Example 3. A gyroscope is mounted so that one point on its axis is fixed.

Investigate those motions under gravify in which the axis of the gyroscope makes
a constant angle with the vertical.
A gyroscope is a body with an axis of symmetry, the shape of the
body being such that the moment of inertia of the body about its
91

axis of symmetry is large. For example, the disc and shaft in Figure
49 constitute a gyroscope.

Figure 49

In Figure 50 the line OA is the axis of the gyroscope, the fixed point
being at 0 and the center of mass being at C. We introduce a fixed
unit vector j pointing up from 0, and a set of moving orthogonal
A

o
Figure 50

unit vectors iv i2 and ia defined as follows: ia is along OA; i 1 is in the

plane of j and i a , as shown; i z completes the triad. We note that i z is
horizontal.
The angle between OA and j is denoted by (); it is constant. The
plane of j and ia rotates about j at a rate of p radians per unit time;
p is called the precession. The gyroscope spins about its axis at the
rate of s radians per. unit time; s .is called the spin. We denote the
mass of the gyroscope by m and the distance OC by l.
The only external forces are the reaction of the pivot support at
and the gravity forces. The former force has no moment about O.
The latter forces may be replaced by a single force -mgj at C. Hence

G = (lia) X (-mgj).

92

But
(41.19)

j = i l sin a+ia cos a,

whence we get
G = -lmg sin a i 2.
Thus
(41.20)

G2 = -lmg sin 6, Ga = O.
The angular velocity w of the gyroscope is
Gl = 0,

= sia+pj.

Because of Equation (41.19) we then have

w

= p sin ail + (s+p cos a) i a ,

whence
(41.21)

<ill =psin6,

<il2 =0,

<ila =s+pcos6.

The angular velocity .Q. of the coordinate axes is given by the relation
.Q.

= pj
= P (il sin 6+ia cos a).

Thus
(41.22)

0 1 = p sin a,

O 2 = 0,

Oa =

P cos a.

The coordinate axes associated with i l , i2 and ia are principal axes

of inertia at 0, and we have
(41.23)
We now substitute in Euler's equations (41.8) from Equations
(41.20), (41.21), (41.22) and (41.23) to obtain the relations
(41.24)

11P sin a = 0,

(41.25 )

[las + (la - 11) P cos a] p sin a = lmg sin a,

(41.26)

la (S -

Pcos a) = o.

One solution of these equations is a = O. In this case the axis of the

93

gyroscope is vertical, and the gyroscope is said to be "sleeping". If

6 is not equal to zero, then Equations (41.24) and (41.26) yield

P = constant,

= constant,

and Equation (41.25) takes the form

(41.27)

(Ia - II) cos 6 p2 +Iasp -lmg

O.

This is a relation among the three constants p, sand 6. Hence it

appears that we may assign arbitrarily values for two of these constants
and there will exist a corresponding motion of the top with 6 a constant, provided of course the value of third constant, as obtained from
Equation (41.27), is real.
We note from (41.27) that

(13 - II) P cos e .

13
The quantities P and e may be observed readily. The corresponding
spin S may be computed by means of this relation. If the precession
is small, we note from Equation (41.28) that the spin is large and has
the approximate value
lmg
S=-'
Ia
(41.28)

lmg
- laP

S---

42. The general motion ofa rigid body. We now consider a rigid body
moving in a general manner. It mayor may not have a fixed point.
The motion of its mass center can be determined from Theorem 2 of
40, which applies to the motion of any system of particles. This
theorem yields
m dvc = F
dt
'
where m is the total mass of the body, Vc is the velocity of its center of
mass, and F is the sum of the external forces acting on the body.
Integration of (42.1) gives the position-vector Xc of the center of mass
C of the body as a function of the time t.
To determine the complete motion of the body it is then only necessary to find the angular velocity of the body about its center of mass.
(42.1 )

94

To do this, we choose the origin 0 of the coordinate system at the

center of mass C of the body. We then consider the body as having
a velocity of translation Vc plus an angular velocity w with a line of
action through C. The velocity Vj of the j-th particle is then given by
the relations
d,,
(42.2)
Vj = vc+
= vc+w X "i.

d/

But by definition the angular momentum h of the body about the

point 0 is
N

= j=1
2: mj"j X Vj'

and we then have, just as in 41,

dh
-=A+B
dt
'

(42.3)
where
(42.4)

N
d,,'
= j=1
2: m _ 1 XV
1 dt
},

B =.2: mj"j
J=I

dv.
-d} .
t

From Equation (42.2) we then have

N

= j=1
2: m.(v1 J

vd XvJ

=.2: mjVj X Vj - v c X .2: mj Vi

}=I

}=I

= O-vc xM ,

where M is the linear momentum of the body. But M = mvc by

Equation (39.2). Thus A = O. Just as in 41 we find that B = G,
where G denotes the total moment about 0 of all the external forces.
Thus Equation (42.3) takes the form
(42.5)

dh
dt -- G

We have thus the result: the rate qf change qf the angular momentum qf
a body about its center qf mass is equal to the total moment qf the external
forces about the center qf mass.
95

We have placed the origin of the coordinate system at the center of

mass C of the body. If we choose the coordinate axes to coincide with
principal axes of inertia of the body at C, then just as in 41 we obtain
Equations (41.6) and finally Euler's Equations (41.8) from which we
can find the unknown quantities (U1' (U2 and (Us which characterize the
rotation of the body about its center of mass.
In conclusion, it should be noted particularly that the equation
dhjdt = G can be used only in the two following cases: (i) the body
has a fixed point and the origin is at this fixed point; (ii) the origin
is at the center of mass of the body.
Example 1. A gyroscope with a constant spin is carried along a horizontal
circular path at a cons'tant speed, with its axis tangent to the path of its center
of mass. Find the forces exerted on the axle of the gyroscope by the bearings in
which the axle turns, neglecting gravity.
We choose the center of mass of the gyroscope as the origin 0 of the
coordinate system. The path of 0 is shown in Figure 51; it is a circle

Figure 51

with center A and radius a. We introduce an orthogonal right triad

of unit vectors at 0, as shown; i1 points vertically up; i2 points towards
A; is is tangent to the circle and hence lies along the axis of the gyroscope.
Let us suppose that 0 travels at a speed v in the direction opposite
to is, and that s is the rate at which the gyroscope spins about its axis,
s being positive when it is in the sense of the 90 rotation from i 1
to i 2
The time required for 0 to go around the circle is 2 7t alv. In this
time the gyroscope has turned about i 1 through an angle of 2 7t
radians. Hence the angular velocity W of the gyroscope is

96

 + -27t2/11

W =S13

7taV

Thus
(42.6)

W3

S.

The angular velocity Q of the coordinate axes. is

Q

v.
all

whence
(42.7)
Also
(42.8)
We now substitute in Euler's equations (41.8) from Equations
(42.6), (42.7) and (42.8) to obtain the relations
SV

G2 = -13 -,
a

G3

= 0.

Hence the forces exerted on the gyroscope by the bearings must be

in the X 3X 1 plane. If Rand -R are these forces, and 2b is the distance
between the bearings, then
2bR = 13 ~
a

whence

Problems
1. A particle moves on the curve x 2 = h tan kx 1 , X3 = 0, where h
and k are constants. The X 2 component of the velocity is constantFind the acceleration.
97

2. A particle moves with constant speed. Prove that its acceleration

is perpendicular to its velocity.
3. A particle moves on an elliptical path with constant speed. At
what points is the magnitude of its acceleration (i) a maximum, (ii)
a minimum?
4. A particle moves in space. Find the components of its velocity and
acceleration along the parametric lines of spherical polar coordinates.
5. A particle moves in space. Its position-vector x relative to the
origin of a fixed set of rectangular cartesian coordinate axes is given in
terms of the time t by the relation
x = h(il cos t+i z sin t+ia t),
where h is a constant and i l , i z and ia are the usual unit vectors in the
directions of the coordinate axes. Find the components of the velocity and acceleration in the directions of (i) the coordinate axes mentioned above, (ii) the principal triad of the path of the particle, (iii)
the parametric lines of spherical polar coordinates. Find the speed
and the magnitude of the acceleration.
6. A particle describes a rhumb line on a sphere in such a way that
its longitude increases uniformly. Prove that the resultant acceleration
varies as the cosine of the latitude, and that its direction makes with
the inner normal an angle equal to the latitude.
7. Two forces A and B act at a point. If IX is the angle between
their lines of action, prove that the magnitude of the resultant R is
given by the relation

R2 = AZ+.82+ 2AB cos IX.

8. Four forces A, B, C and D act at a point 0 and are in equilibrium' the forces C andD being perpendicular and having equal magnitudes. Find C in terms of A, B and the angle IX between A and B.
9. Forces with magnitudes 1, 4, 4 and 2V31b. wt. act at a point.
The directions of the first three forces are respectively the directions
of the positive axes of Xl, 'x z and Xa. The direction of the fourth force
makes equal acute angles with these axes. Find the magnitude and
direction of the resultant.

98

10. A force F acts on a particle of mass m. Find the magnitude of

the acceleration, given that (i) F = 6 poundals, m = 3 lb., (ii) F =
6 lb. wt., m = 2 slugs, (iii) F = 6 poundals, m = 2 slugs, (iv) F =
6 lb. wt., m. = 3 lb., (v) F = 5 dynes, m = 10 gm.
II. A particle of mass m is acted upon by a force F given by the
relation

where p and q are constant vectors and t is the time. Find the velocity v and position-vector x of the particle in terms of t, given that v
= 0 and x = 0 when t = O.
12. A particle of mass m is acted upon by two forces P and Q.
The force P acts in the direction of the Xl axis. The force Q makes
angles of 45 with the axes of X 2 and X 3 Also P = P sin kt and Q. =
q cos kt, where p, q and k are constants and t is the time. At time
t = the particle has coordinates (b,O, 0) and is moving towards the
origin with a speed pjmk. Find the position-vector x of the particle.
Prove that the particle moves on an ellipse, and find the center and
lengths of the axes of the ellipse.
13. A particle of mass m moves under the action of a force pe-qt
and a resistance -lv, where p is a constant vector, q and l are positive
constants, t is the time, and v is the velocity of the particle. Prove that

xoo - x.

rq1 (p+ m q u)

where u is the velocity when t = 0, and lito and Xoo are respectively the
position-vectors of the particle when t = 0 and when t becomes infinite. Is the above result true when l = mq?
14_ A particle of mass m moves under the action of a force p cos qt
- kx, where p is a constant vector, q and k are positive constants,
t is the time, and x is the position-vector of the particle relative to a
fixed point O. When t = 0, the particle is at 0 and has a velocity u.
Find x in terms of t when (i) k =1= mq2, (ii) k = mq2.
IS. A particle of mass m is acted upon by a single force y mjxZ
directed towards a fixed point 0, where y is a constant and X is the

99

distance from 0 to the particle. At time t = 0 the particle is at a point

B and has a velocity of magnitude u in a direction perpendicular to the
line OB. Prove that the orbit is (i) an ellipse if bu2 < 2y, (ii) a parabola if bu2 = 2y, (iii) an hyperbola if bu2 > 2y, where b = OB.
16. Find the moment of inertia of a circular disk of mass m and
radius a about (i) the axis of the disk, (ii) a diameter of the disk.
[Answer: (i) tma 2; (ii) lma2 .]
17. Using the result of Problem 16, find the moment of inertia of
a circular cylinder of mass m, length 2l and radius a about (i) the
axis of the cylinder, (ii) a generator of the cylinder, (iii) a line
through the center of the cylinder perpendicular to its axis, (iv) a
diameter of one end of the cylinder. [Answer: (i) tma2 ; (ii) !ma2 ;
(iii) /2 m (4l2+3a 2); (iv) T\ m (16l 2 +3a2).]
18. A circular cylinder has a mass m, length 2l and radius a. Rectangular cartesian coordinates are introduced, with origin 0 at the
center of the cylinder, and the X3 axis coinciding with the axis of the
cylinder. Two particles each of mass m' are attached to the cylinder
at the points (0, a, l) and. (0, -a, -l). Find the moments and products
of inertia II, 12 , 13, K l , K2 , and K3 for the system consisting of the
cylinder and the two particles.
19. A circular disk of mass m and radius a spins with angular speed
Ul about a line through its center 0, making an angle oc with its axis.
Find the angular momentum ofthe disk about O.
20. For the system of masses in Problem 18, find the angular
momentum about the point 0 when the system has an angular speed
Ul about (i) the Xl axis, (ii) the X 2 axis, (iii) the X3 axis.
21. A circular cylinder of mass m, length 2l and radius a turns
freely about its axis which is horizontal. A light inextensible cord is
wrapped around the cylinder several times. A constant force F is
applied to the end of the cord. If the cylinder starts from rest at time
t = 0, show that at time t it has turned through the angle Ft 2 /ma.
22. The circular cylinder of Problem 21 is again mounted with its
axis horizontal, and has a light inextensible cord wrapped around it.
A body with a mass m' is attached to the end of the cord. If the cylinder
starts from rest at time t = 0 show that at time t the cylinder has
100

mgt2
h
. hI
d
turne d t h rough t h e ang1e a(m+2m')
' were g IS t e acce eratlOn ue
to gravity.

23. A circular cylinder of mass m, length 2l and radius a is placed

on a rough plane which makes an angle rJ. with the horizontal, and is
then released. Find the distance the cylinder moves down the plane
in time t.
24. In Problem 18 there was introduced a system consisting of a
circular cylinder of mass m with two attached particles each of mass
m'. This system is mounted so it can turn about the axis of the cylinder
in two smooth bearings each at a distance c from the center of the cylinder. The system is made to rotate with constant angular speed w.
Find the reactions of the bearings.
25. A circular disk of mass m and radius a turns with constant angular speed w about an axis through the center 0 of the disk and making a
constant angle rJ. with the axis of the disk. The disk turns in two
smooth bearings each at a distance c from the point O. Find the reactions of the bearings.
26. A uniform rod of length 2l is free to turn about an axis L
perpendicular to it and througl: ~(S center. The center of the rod
moves at constant speed v around a circular track of radius a, the
axis L being always tangent to the track. Deduce the equations of
motion of system.

101

CHAPTER

IV

PARTIAL DIFFERENTIATION

43. Scalar and vector fields. Let V denote a region in space, and let
X be a general point in V. Let Xl, X 2 , X3 denote the rectangular cartesian coordinates of X, and let x denote the position-vector of X. Then
x = xlil+X2i2+X3i3'

Let us now consider the case when there is associated with each
point in the region Va value of a scalar j Then we write f = f(x l ,
X 2 , X 3 ), or more compactly
f=f(x).
The values off associated with all the points in V constitute a scalar
field.
Let us now consider the case when there is associated with each
point in the region V a value of a vector a. The:r;t we write a = a (Xl'
X 2 , x a), or more compactly
a = a(x).

The values of a associated with the points in V constitute a vector field.

It is frequently necessary to consider scalar and vector fields which
vary with a parameter, such as the time t. In such cases we write

f=f(x, t),

= a(x, t).

44. Directional derivatives. The operator del. Let C be a curve in a

region V, as shown in Figure 52. Let X be a general point on C, with
position-vector x as shown, and let s be the arc length of C measured
from some fixed point Q. on C. It was seen in 28 that the vector ~
is a unit vector tangent {Q C in the direction of s increasing. This
vector is shown in Figure 52.
102

;,

Figure 52

We now introduce a function J(Xl' X2 , xa) defined everywhere in

the region V. Then J is defined at all points on the curve C, and the
rate of change ofJ with respect to the arc length s at the general point
X on C is given by the well-known formula of differential calculus,

dj = dJ dX l + dJ dX 2 + dJ dXa.
ds
dX 1 ds
dX 2 ds
dX a ds
Let us now consider all curves through X having the same tangent at
X as C has there. For each of these curves the value at X of the right
side of Equation (44.1) is the same. Thus at each point X there is

(4tl)

a unique value of

associated with each direction. This value is

called the directional derivative off

We now introduce an operator called del, which we denote by the
symbol \l and define by the relation

(44.2)-

"(7.d
v = 11 -::;vX 1

od
+ 12od-::;+ 1a-::;-
vX
vXa
2

If we wish to operate on a function with this operator, we write the

symbol denoting the function immediately to the right of the symbol
\l. Thus, ifJis a scalar field, then

J= (il:vX + i2:VX + i3 vXa

: )J

\l

or
1I"

\lJ =

~f
vX 1

11 -::;-

dJ
~f
+ 12 -::;+la -::;-.
dX
vXa
0

103

The expression VJ is often called the gradient of j, and is denoted

by grad j.
We note that the right side of Equation (44.1) is equal to the scalar
productofVJand dx/ds, by Equation (7.2). Thus
(44.3)

dJ =
ds

Vf dx,
ds

or
(44.4)

-dJ
ds = vrt
J'

where t is the unit vector in the direction in which the directional

derivative is taken. Some theorems will now be proved.
Theorem 1. The component of VJ in the direction of a unit vector
t is equal to the directional derivative ofJin that direction.
Proof. The component of VJ in the direction 0f t is /VJ/ cos 6,
where fl is the angle between the vectors VJ and t. But, since t is a
unit vector,

/VJ/

cos 6 =

Vft

or
(44.5)

/VJ/

df
cos 6 = ds'

by Equation (44.4).

Theorem 2. The vector VJ points in the direction in which df/ds

has a maximum value; also, this maximum value is equal to /VJ/.
Proof. Both parts of this theorem follow from Equation (44.5)
which shows that when 6 = 0, df/ds attains its maximum value which
is /VJ/.
Theorem 3. The vector field VJ is normal to the surfaces J =
constant.
Proof. Let X be a point on a surface J = constant, and let t be
a unit vector tangent at X to this surface. Then the value of df/ds at
X corresponding to the direction of t vanishes. By Equation (44.4)
we then have
Vft = O.
104

This implies that vf is perpendicular to all vectors at X tangent

there to the surfacef = constant, which completes the proof.

45. Properties of the operator del. This operator, which is denoted

by the symbol V, is defined in the previous section. For convenience
we express it in the form
3

(45.1)

V =2: i r
r=1

0
-

oX r

Then

Vf=

(45.2)

r=1

ir

of.

OXr

We shall now prove a number of theorems relating to this operator.

Theorem 1. Iff and g are scalar fields, then

Proof.

= Vf+Vg.

V (f+g)

(45.3)

Now, by Equation (45.2) we have

3

()

V(f+g) =r~1 ir

oXr (f+g)

~f + ~ ir {)e
oX r r= I oXr

= ; ir
r= 1

= Vf + Vg.
Theorem 2.

Iff is a function of a single scalar field u, then

Vf=J~Vu.

(45.4)
Proof.

By Equation (45.2), we have

3

of

r=l

OX T

Vf=2: i r -
Butf is a function of u, which is a function of X r Thus
of dj ou
~ = -d"""l
(r = 1,2,3),
oX
U OX
r

so
105

V']=~jrdjdU
r=\

dj

du

du

dX r

~ i.
r=\

dU
dX r

dj

= du V'u.

Theorem 3.

If] is a function of n scalar fields

U1 ,

U2 , , Un,

then

(45.5)

Proof. Since] is a function of the n variables

are themselves functions of Xl' X 2 , X3 then

U1 , U2 , , Un

which

Hence

As mentioned in the previous section, \1 operates only on functions

written on its immediate right. Thus, if] and g are scalar fields, the,n
3
dg
]\lg =]2:. i r r=1

106

dXr

Theorem 4.

If f and g are scalar fields, then

v (fg) = fv g+gV f

(45.6)
Proof.

We have
3
d
V(fg) =r~li,. dX (fg)
r
g
f
ir(fd +gd )
r= 1
dX r
dX r
=fVg+ gVf

=f

46. Some additional operators. Let a be any vector field. We first

consider the operator a V. This operator has the obvious meaning

(46.1)

aV = (alil+a2i2+aaia).(il

dX 1
d
d
d
= a 1 -::j + a2 -::j + aa ~
uX 1
UX 2
UXa
3

+ i2

uX 2

+ ia

UXa

=:E ar -
r=1 dX r
This operator is a scalar, and can be applied to a scalar field or to a
vector field. Thus, iff and b are two fields, then
3
df
3
db
(46.2)
(aV)f=:E ar-::j, (aV)b=:E a,-::j'
r= 1 UX r
r=1 UX r
We note that
c

or

or
107

it
at

i3
i2
a3
a2
(46.5)
ax\7 =
d
d
d
dX t
dX 2 dX 3
The operator a X \7 is a vector operator. It can be applied to scalar
fields. Thus, ifJis a scalar field, then

(46.6)

(aX\7)J=

We note that

it
at

i2
a2

i3
a3

dJ
dX t

d.f
dXz

d.f
dX 3

(aX \7)J = ax \7J.

Inwritingtheexpressions (a\7)J, (a.\7)b and (aX\7)J, onemust
exercise care in the matter of the order in which the symbols appear,
since the operator \7 and all operators constructed from it operate
only on the functions on their immediate right. Thus, for example,

Ca. \7)b =1= b(a. \7).

The left side of this expression is a vector field, while the right side
is a vector operator.
We now introduce the operators \7. and \7 x. These operators can
be applied to vector fields, the vector portion of the operator \7
operating on the vector field with scalar or vector multiplication. If
b is a vector field, then

(46.7)

\7 . b = (it

~ + i2 dX2
~ + i3 dX~ 3 ). b .

uX t

Thus we may write

(46.8)

\7b

=(~ 1 i.~).b
=~ 1 i.. dX
db
dX r
r
r=

r=

But

i.. b = br
Thus

(46.9)
108

(r = 1,2,3).

=~ ~(iTb).
dX
r= 1

The expression V b is often called the divergence of b, and is

written div b.
We also have
(46.10)

v X

J)

d+
o12d
0
= (011 ~
~ + 13 -;;OX
oX
oX
l

Thus we may write

(46.11)

Vxb =

d)
3
db
3 d
3
:E Ir-;;- xb =:E lrX -;;- = :E -;;- (Irxb).
0

( r=1

OXr

r=1

OXr

r=10Xr

But
ilxb

ilx (blil+b2i2+b3i3)
= b2ia - b3i2
by Equations (8.9). Similarly we have
i 2xb = bail - bli 3, i3Xb
Hence (46.11) becomes

= bli 2 - b2ip

(46.12)
This can be written conveniently in the form
il
i2
i3
d
d
d
(46.13)
Vxb =
dX l
dX 2
dX 3
b2
bl
b3
The expression V X b is often called the curl of b or the rotation of
b, and is written curl b or rot b.
Theorem. 1. If a and b are two vector fields, then
(46.14)

V (a+b)

V . a+V . b,

Vx(a+b) = Vxa+Vxb.
Proof of Equation (46.14). From Equation (46.9), we have
3
d
V (a+b) =r~1 dX (ar+br)
r
_ ~ (dar + db r)
-r=1 ,dXr
dX r
= Va+Vb.

(46.15)

109

Proof of Equation (46.15).

\7 X (a+b) =

From (46.13) we have

il
d
dX l
a1 +b l
i1
d
dX l
a1

i2
i3
d
J
dX 3
dX 2
a2+b 2 a3+b3
i2
i3
d
d
dX 3
dX 2
a3
a2
il
i2
d
d
+
dXl
dX2
b2
bl

i3
d
dX3
b3

\7 xa+ \7xb.

Theorem 2.
(46.16)
Proof.
(46.17)

If a and b are two vector fields, then

(aX\7)b =a(\7xb).

Because of (46.4) we have

(ax\7)b =

(a 2;X3-a3;X2) (iIb)

(a ~-a ~) (~b)
3

dX 1

dX 3

+ (a 1;X2- a2~J (i3 b ) .

But
Thus (46.17) can be written in the form
(a X \7) .b = a l (db 3 _ db 2) + a2 (db l _ db 3) + a3 (db 2 _ db l )
dX 2 dX 3
dX 3 dX 1
dX 1 dX 2
Because of (46.12), we see that the right side of this equation is
~qual to a (\7 xb).
110

47. Ivariance oj the operator del. Let us consider two rectangular

cartesian coordinate systems with a common origin O. We denote
them by the symbols Sand S'. For the coordinate system S the coordinates and associated unit vectors will be denoted by Xl' X 2 , Xa and
i 1 , i 2 , is, respectively, while for the system S' the corresponding

it

Figure 53

quantities will be denoted by Xl" X 2', xs' and i 1 ', i 2', is'. Figure 53
shows the coordinate axes and the associated unit vectors.
We now introduce the two operators
'\7

'\7,

oJ

+ 12oJ-::;- + ISoJ-::;-,
J +
J +
J
:1--;
12

11 -::;OX 1
0,

= 11

OX 2

uX 1

OX 2

uXs

IS~'

dX s

LetJbe a scalar field, and let b be a vector field. We shall now consider proofs of the relations
(47.1)

v1= VJ,

(47.2)

V' . b = V . b,

(47.3)

V' xb = V xb.

The implications of these equations are the following. The operator

del includes differentiations with respect to the coordinates of the
system S and multiplications by the three unit vectors associated
with the system S. It thus appears that when del operates on any
III

field, the resultant field is likely to depend on the particular choice

of coordinates. However, Equations (47.1), (47.2) and (47.3) state
that this is not the case. It is this property of del which is termed
"invariance of the operator del".
Before proceeding to a proof of the above three formulas, we shall
develop some preliminary formulas. Let us consider the three angles
which the Xl' axis makes with the axes of the system S. These are the
angles 611 , 612 and 6la in Figure 53. We denote their cosines by all>
a12 and a 13 These three quantities are the components of iI' relative
to the system S, and so
3

iI'

au i l +a 12 i 2 +a13 ia

= s=1
~ als is

There are similar expressions for i 2 ' and i a'. The three expressions
can be written in the compact form
(47.4)

i,.'

(r

au is

1, 2, 3) .

s=1

We note that the constant a,s is the cosine of the angle between the
two vectors i,.' and is.
The components of the vector i l relative to the system S' are seen
to be all , a 2l and aal. Thus
3

il =

~
s=1

asl is'.

Proceeding similarly for i2 and i a , we obtain the three equations

(47.5 )

i,. =

~ asr

s= 1

is'

(r

1,2,3).

Throughout the remainder of this section we will adopt the convention that latin subscripts range over the values 1, 2 and 3, as in the
above equations.
We now introduce a set of nine quantities called the Kronecker
delta. This set is denoted by the symbol 8", and is defined by the
equation
(47.6)
112

8,s = 1 ifr = s
=Oifr=s.

Thus &11 = &22 = &33 = 1 and &23 = &31 = &12 = &32 = &13 = &21 = O.
Since iI, i2 and i3 are unit orthogonal vectors, they satisfy the nine
relations (7.3). These nine relations can now be written compactly
in the form
(47.7)
We also have the nine relations
(47.8)
Let us now substitute in Equation (47.8) from (47.4). This gives
the relation
&rs

= (

art
t= 1

it) .

(~anl i.)
.=1

= t=1
~ ~ art ani iti..
u= 1
But by (47.7) we have iti.

= &tu

so

(47.9)

&rs=

t=1 u=1

art a su &tu

= t=1
~ art ~ ani &t.
.=1

The last summation here is

The right side reduces to asl when t = 1, to a,2 when t = 2, and to

as3 when t = 3. Thus we may write
3

(47.lO)

~ ani tu

u=1

a st ,

and so (47.9) reduces to the relations

3

(47.11)

ars = ~ art a st '

1=1

The nine equations in (47.U) are called the orthogonality conditions.

We could also run through the above derivation of (47.11) but with
the roles of the primed and unprimed quantities interchanged. This
would entail substituting in Equations (47.7) from (47.5). In this
way we would obtain the orthogonality conditions in the form

113

(47.12)

3rs

= 1=1
1: a"

als'

Let b be any vector field with components b, relative to the system

S, and with components br' relative to the system S'. Then
3

= ,=1
1: bs' is' .

b = 1: b, i,.

(47.13)

,=1

We now substitute here for i,. from (47.5), obtaining the relation
3

1: b/ is'
r=1

= r=l
1: b,
3

= 1:

1: asr is'

s=l

1: b, asr is' .

,=1,=1

Hence we must have

3

(47.14)

bs' = 1: asr b,.

,=1

If in an analogous fashion we substitute in Equation (47.13) for is'

from (47.4), we obtain the relations
(47.15)

b,

= r=l
1: ars b/.

Equations (47.14) and (47.15) are the equations of transformation

of the components of a vector field b.
If we choose b = x, where as usual x is the position-vector of a
general point X with coordinates x, relative to the system Sand
coordinates x/ relative to the system S,' then Equations (47.14) and
(47.15) yield
(47.16)

Xs '

= L asr x"
,=1

x,

= ,=1
1: ars x/.

These are the equations of transformation from a set of rectangular

cartesian coordinates to a second set whose axes are obtained from
those of the first set by a rotation about the origin. From Equations
(4 7.16) we have the relations
114

(47.17)

{)X',

()X s
""::)"7
oX r

-:s-= an ,
oX
r

aTS

Weare now in a position to turn to proofs of Equations (47.1), (47.2)

and (47.3).
Proof of Equation (47.1). This equation reads

\11= \1f
Now

\I1=~i'r{)!'
1
{)X

(47.18)

r=

and

Because of this relation and (47.4) we can write (47.18) in the form
(47.19)

\11 = ~ ~ ~
r=1

ars is {)j {)x', .

,=1 '=1

{)Xt {)X r

But, from Equations (47.17) we have

{)x t
""::)"7
oX T

art

Thus (47.19) becomes

,

\11 = T=l
~ ~
s=1
3

t=1

3.

a"

{)j

{)j
art Is '""l
ox,
3

= s=1
~ ~ 1, '""l ~ a" aTt
t=1
oX r=1
t

Because of the orthogonality conditions (47.12), the last summation

here is equal to as,' Thus
,
3
3 . {)j

\11 =

~ Is'""l

,=1 t=1
3

oX,

a,t

{)j

= s=l
~ Is ~ '""l a" .
t=1 oXt
The last summation here reduces to {)jj {)x1 when s
when s = 2, and to {)jj {)xa when s = 3. Thus

1, to {)jj {)X'I.
115

(47.20)
and so
3
df
\I'j=L is s=1

= \If.

dx,

The truth of Equation (47.1) also follows from Theorem 2 of 44,

as this theorem states that \If points in the direction in which the
directional derivative offhas a ma~mum value, and that the magnitude of \If is equal to this maximum value. This theorem then implies that both the direction and magnitude of \If are independent
of the coordinate system.
Proof of Equation (47.2). This equation reads
\I'b

\Ib.

Now
3 db'
\I'b =L - ' .
,= 1 dx's

Because of (47.14) we then have

3

\I'b =L L -

$==1 r=1 ()X's

=~ ~
.1=

1 r= 1

(a b)
.rr

a db,
ST

()x's

But by Equations (47.17), the last partial derivative here is equal to

a,t. Thus
3
3
3
db
\I'. b = L L L a" a,1 T
s=1 ,=1 1=1

= L

db

oX I

L ~ L a" ast

r=1 1=1 OX/S=I

Because of the orthogonality conditions (47.12) the last summation

here reduces to 8,1. Thus
116

\7'.b =

~ ~ ob, ~'I
,=1 1=1

OXt

~ ob"

OX,
by a repetition of the arguments leading up to Equations (47.10) and
(47.20). This completes the proof.
Proof of Equation (47.3). This is left as an exercise for the reader
(Problem 13 at the end of the present chapter).
.=1

48. Differentiation formulas. We shall consider here a group of

well-known formulas involving the operator del. Iff is a scalar field,
a and b are vector fields, and x is the usual position-vector of a general point X, these.formulas are the following:

(48.1 )

\7.(fa) =f(\7a)+(\7f).a,

(48.2)

\7 X (fa) =f(\7xa)+(\7f)xa,

(48.3)

\7.(axb) =b(\7xa)-a(\7xb),

(48.4)

\7x (axb) =a(\7.b)+(b.\7)a-b(\7a)-(a.\7)b,

(48.5 )

\7(a.b) = (a. \7) b+(b. \7) a+a <(\7 Xb)+bx (\7 xa),

(48.6)

\7 X (\7f) = 0,

(48.7)

\7(\7xa) =0,

(48.8)

\7X(\7xa) =\7(\7.a)-(\7.\7)a,

(48.9)

\7.x=3,

(48.10)

\7xx=O.

(48.11)

(a. \7)x = a.

Direct proofs of all eleven of these formulas follow similar lines. We

shall present here only proofs of (48.1), l48.3), (48.5) and (48.8).
The proofs of the remaining formulas are left as exercises for the
reader (Problems 14, 15 and 16 at the end of the present chapter).
Proof of Equation (48.1). Because of Equation (46.8), which gives
some equivalent forms for \7 . b, we have

117

\7.(ja) = L

,=1

=L3

,=1

1,- (fa)
dx,

1,'

f a)
~Jda +ddx,
dx,

3 . da
1,'-

=fL

,=1

dx,

3.d!
+ ,=1
L 1 , - " 'a
dX,

= f(\7a) +(\71) a.
Proof of Equation (48.3). Because of Equation (46.8) we have
3
d
\7 . (~ X b) =,~ 1 i,.. dx, (a X b)

= L3 i, (da
- Xb )
dX,

,=1

+,=1
L3 i,.. ( a X-db) .
dx,

Because of the permutation theorem for scalar triple products (Theorem I of 9), we can write this last equation in the form

da'
3
(
db)
\7(axb) =L3 b ( i,.X-)La'
i,.xdx,

,=1

dx,

,=1

da
3.
db
=bL 1,x--a1: I,X3.

dx,

,=1

,=1

dx,

= b(\7 xa) - a(\7 xb).

Proof of Equation (48.5).

Because of Equation (46.11) we have

3
db
ax(\7xb) =aX L i,x""")
,=1

uX,

We now apply to the right side of this equation the identity (9.3)
for vector triple products, obtaining
(48.12)

d
ax(\7xb) =L3 i,. (db)
a-- - L3 - b (ai,)
,=1

dx,

,=1 dX,

3.(a 'db)
- - L3 a -db

=L

1
,=1'

dX,

,=1' dx,

=L3 i, ( a 'db)
- -(a\7) b.
,=1

118

dX,

Similarly, by an interchange of a and b we have

(48.13)

(da )

3
b X (\7 xa) =r~l
iT dxT',b - (b \7) a.

Addition of (48.12) and (48.13) yields

3

()

aX (\7xb)+bX(\7xa) =T~li, dX, (ab) - (a\7) b- (b\7a)

= \7 (ab) - (a\7)b- (b\7)a.

If the last two terms here are taken to the left side of the equation,
(48.5) results.
In addition to this manner of proving formulas (48.1)-(48.11),
there is another manner of proof which applies to some of these formulas, and which is quite expeditious. We mention it here because it
affords one an opportunity to acquire additional facility in manipulations involving the operator del. This manner of proof consists in
applying to expressions involving del the permutation theorem for
scalar triple products (Theorem 1 of 9), and the identity (9.3) for
vector triple products which is

(48.14)

aX (bxe)

b(a.e) - e(a.b).

Of course, when such an application is made, a symbol denoting a

scalar or vector field must never be moved from one side of del to the
other, since del operates only on those quantities on its right. Also,
when such application is made, the resulting expression must have the
same scalar or vector character as the original expression.
For example, let us prove Equation (48.8) in this way. If we apply
the identity (48.14) formally to \7 X (\7 xa), we obtain the difference
of two 'terms. The first term must be one of the following:
(48.15)

\7 (\7.a),

\7 (a.\7),

(\7.a)\7,

(a. \7) \7.

Now, in the expression \7 X (\7 X a), the symbol a appears to the

right of both symbols \7. Hence we must select from the terms in
(48.15) the one with this same property. Hence we select \7(\7 a).
As a check, we note that \7 X (\7 X a) is a vector, and that the ol\ly
term in (48.15) which is a vector is \7 (\7.a). In a similar fashion
119

we find tha,t the term corresponding to the second term on the right
side of (48.14) is (V, V) a. Thus

Vx(Va) = V (Va) - (VV)a.

We note that the operator V V satisfies the relation
{)2

V V =

{)2

(()X )2
1

{)2

+ (()X 2)2 + (()X 3)2'

This operator is called the Laplacian operator, and is denoted in

many books by the symbol V 2
As a second example, let us prove Equation (48.3) by application
of the permutation theorem for scalar triple products. To do this,
it is convenient to introduce an operator Va which is just the operator
del with the added restriction that the partical differentiation is to be
applied to no vector field other than a. The operator Vb is defined
similarly. Equation (12.4) states the differentiation rule for vector
products. Because of this rule we have
(48.16)

V(aXb)

Va(axb)+Vdaxb).

We now apply the permutation theorem for scalar triple products

to the first term on the right side of Equation (48.16), remembering
that a must always remain on the right side of Va' and that b need
not do so. We then get b. (Va xa). If the second term on the right
side of Equation (48.16) is treated analogously, we obtain the relation

V(axb) =b(Vaxa)-a(Vbxb).
But

Vaxa = V Xa,
Thus Equation (483) is proved.

49. Curvilinear coordinates.

(49.1)

Zl =11 (Xl'

Let us write down the three equations

X 2 , X 3 ),' Z2

=12 (Xl'

X 2 , X 3 ),

Z3

=13 (Xl' X 2 ,

X 3 ),

where1l (Xl' X 2 , X 3 ), 12 (Xl' X2, X 3 ) and13 (Xl' X 2 , X 3 ) are any functions

which are single valued and differentiable throughout some region V.
These equations prescribe for any point X with rectangular cartesian
120

coordinates Xl' x 2 and Xa a new set of coordinates Zl, Z2 and Z3' These
new coordinates are called curvilinear coordinates, and Equations
(49.1) are equations of transformation of coordinates.
We now propose to compute, in terms of quantities pertaining to
curvilinear coordinates only, the expressions \7J, \7. b and \} xb,
where 1 and b denote respectively a scalar and a vector field. With
this goal in mind, we shall devote the rest of this section to some preliminary considerations, and shall complete the final computations
in the next section.
The Jacobian of the transformation (49.1) is the determinant

(49.2)

]' =

dZ l
dX 1

dZ l
dX 2

dZ l
dX a

dZ 2
dX 1

dZ 2
dX 2

dZ 2
dXa

dZ a
dX 1

dZ 3
dX2

dZ a
dX a

We shall consider only the case when ], does not vanish anywhere
in V, so that Equations (49.1) may be solved 1 for Xl, x2 and Xa to
yield the relations
(49.3)

Xl =gl (Zl,Z2,Za), X2 = g2 (Zl,Z2,Z3),

Xa =ga (Zl,Z2,Za)'

Let us write down the equation

Zl =11 (Xl' X2 , Xa) = constant.

The locus of this equation is a surface. If we vary the constant in this

equation, we get a family of surfaces called the parametric surfaces
of Zl' Similarly, we have parametric surfaces of Z2 and Za' Let us
now consider the curves of intersection of the parametric surfaces of
Z2 and Za' Along each of these curves Z2 and Za are constant, and Zl
alone varies. These curves are called the parametric lines of Zl' Simi1 See almost any book on Advanced Calculus; for example, I. S. Sokolnikoff,
Advanced Calculus, McGraw-Hill Book Co., New York, 1939, pp. 430-438.

121

iarly, we have parametric lines of Z2 and ZS' For example, in the case
of cylindrical coordinates r, e, xs , the parametric lines of r, e and Xs
are respectively horizontal straight lines cutting the Xa axis, horizontal
circles with centers on the Xs axis, and vertical straight lines.
Let X be a general point in a region V. Through X there passes a
parametric line of each of the curvilinear coordinates ZI, Z2' Zs' We
now introduce three unit vectors kl' k2 and ks with origins at X,
defined as follows: kl is tangent at X to the parametric line of ~l'
and points in the direction of Zl increasing; k2 and ks are defined analogously with respect to the par_ametric lines of Z2 and Zs' Figure 54

k,

X~_""""=-+

Figure 54

shows these parametric lines and associated unit vectors. In the

case of cylindrical coordinates, the associated unit vectors kl' k2 and
ka were introduced in 30.
In the remainder of the present section we shall consider only the
case when the three vectors kl' k2 and ks are mutually perpendicular.
The curvilinear coordinates are then said to be orthogonal. We shall
also require that kl' k2 and ka form a right-handed triad, the term
"right-handed" being as defined in 6.
The distance ds between two adjacent points is given by the relation
3

(49.4)

(ds)2 = (dX 1 )2+(dx2)2+(dxs)2 = 2: (dX )2,

T

T=l

where dx 1 , dX 2 and dx s are the infinitesimal differences between the

rectangular cartesian coordinates of the two points. From Equations.
(49.3) we have
.

122

(r

= 1,2,3).

Substitution from these relations in (49.4) then yields for (ds)2 a

homogeneous quadratic expression in the quantities dz 1 , dZ 2 and
dz 3 Because the coordinate system ZI' Z2' Z3 is orthogonal, it turns
out that no terms involving products of these differentials appear in
this quadratic expression, and we can then write
(49.5)
where hI' h2 and h3 are known positive functions of ZI' Z2 and Z3'
The right side of (49.5) is called the fundamental quadratic form, or the
metric form.
Les S1> S2 and S3 denote the arc lengths of the three parametric
lines in Figure 54. From (49.5) we then have
(49.6)
Let us consider ZI' It is a scalar field. We shall now consider 'i7 ZI.
Because of Theorems 2 and 3 of 44 we see that
(i)
(ii)
(iii)

'i7Z1> being normal to the surface ZI = constant, has the same

direction as kl'
1'i7 zll is equal to the maximum value of the directional deritive tkl/ds,
this maximum value arises when tl).e directional derivative is
taken in the direction ofkl' and is hence equal to dz1 /ds p

Because of (ii), (iii) and Equation (49.6) it follows that l'i7z11 = ljh 1 ;
and because of (i) we then have 'i7z 1 = k 1 /h 1 Similar observations
regarding Z2 and Z3 then permit us to write
.
(49.7)
From these equations we get the relation
k 1 (k2 X k s)

hI h2 h S 'i7 Zl' ('i7 Z;? X 'i7z 3)

But the left side of this equation is equal to one since kl' k2 and ks
form a right-handed orthogonal triad of unit vectors. Thus

123

(49.8)

50. The expressions Vj, V b and V X b in curvilinear coordinates.

We consider first the expression Vf Now f is a function of Zl' Z2 and
Z3' Hence by Equantion (45.5) we have
df
df
Vf = ~ VZ1 + -;:;- VZ 2
uZ1
uZ2

+ df
CJ VZa
uZa

Because of Equation (49.7) we may then write this relation in the

form
(50.1 )
or
(50.2)

3 1 df
Vf = ~ -h -;:;- k,.

,-1

uz,

We now turn to V . b . We first express b as a linear function of the

unit vectors k 1, k2 and k a , in the form
(50.3)
It should be noted that b1 , b 2 and ba are here the orthogonal projections of b on the lines of action of the unit vectors k 1, k2 and .ka,
respectively. In order to deduce conveniently the desired expression
for V b, we now make a rather unusual step. Since

k1 = k 2 xka ,

k2 = kaxk1,

ka = k 1 xk 2 ,

then substitution from Equations (49.7) yields

k1 = h2h a V Z2 X V Za ,
k2 = hah1 VZa X VZ 1,
ka = h1h2 VZ1 X Vz 2
We now substitute from these expressions in Equation (50.3), and
operate on the resultant expression with the operator V to obtain
the relation
124

(50.4)

\7. b = \7. (b1h2ha \7Z2 X \7Za) \7. (b 2hil.l \7za X \7ZI)

+\7. (b ahlh2 \7z1 X \7Z2)'

For the first of the three terms on the right side of (50.4) we can write,
because of relation (48.1),
(50.5)

\7. (b 1h2ha \7z 2 X \7za) = \7 (b 1h2ha) . (\7Z2 X \7 za)

+b 1h2ha \7. (\7 Z2 X \7 Za).

But because of Equation (45.5) we have

d
\7 (b 1h2 ha) =""'l (b 1h2ha) \7 Zl
~

d
+ ")
(b 1h2ha)
~

Thus, since \7 Zl> \7 Z2 and \7 Za are mutually

d
\7 (b1h2ha)' (\7 Z2 X \7 Za) = ") (b 1h2ha)
oZI
or, by Equation (49.8),
1
(50.6)
\7(b 1h2ha)(\7Z2 X \/'za) = hhh
I 2

\7Z2

d
+ ")
(b 1h2ha)
~

\7 Za'

perpendicular, we obtain
\7 zd\7 Z2 X \7 za)

d
")(b 1h2ha)'
a oZI

Also, because of Equation (48.3) we have

(50.7) \7. (\7Z2X\7Za) =\7za(\7X\7Z2)-\7Z2(\7X\7Za) =0

because of Equation (48.6). The Equations (50.6) and (50.7) now

permit us to write (50.5) in the form
1
d
\7 . (b 1h2ha \7 Z2 X \7 Za) = h h h ""'l (b 1h2ha)'
I 2 a oZI
This relation and two similar relations involving the second and third
terms on the right side of Equation (50.4) then permit us to write
(50.4) in the form
(50.8)

\7. b

=.

h h1h [d
-=j (b 1h2 ha)
I 2 a oZI

d (b 2hahl) + Td (b ahlh2) ]
+ ToZ2
oZa.

We note that if the curvilinear coordinates happen to be rectangular

cartesian coordinates, then hI = h2 = ha = 1 and (50.8) reduces to
(46.9), as expected.
Finally, we turn to the expression \7 X b. Because of Equations
(49.7) and (50.3) we have
125

or
(50.9)

\l xb = \l X (b 1hl \lZI) +'17 X (b 2h2 \lZ2)

+\l X (bah a \lza) .

For the first term on the right side we can then write
(50.10)

\l X (b 1hl \lZI) = \l (b 1hl)

\l ZI + b1hl (\l

\l ZI) ,

because of Equation (48.2). But \l X \lZI = 0 by Equation (48.6),

and by Equation (45.5) we have
\l (b 1hl)

d
";) (b hl ) \l ZI
[ uZI 1

\lZI =

+ lza (b hl)
1

d
+ ";)
(b hl) \l Z2
uZ2
1

\lza]

\lZI'

Now Equations (49.7) yield

\lZIX \lZI

\lZ2 X \lZI
\lZ3 X \lZI

0,
1
h2hl k2 Xkl

=-

1
hahl ka X kl

= hahl k 2

h2hl ka,
1

Thus Equation (50.10) reduces to

k2 d
ka d
\lx(b 1hl \lZI) = h-h '::) (b 1hl) -h-h '::) (b 1hl)'
a 1 uZa
2 1 uZ2

This relation and two similar relations involving the second and third
terms on the right side of Equation (50.9) permit us to write Equation
(50.9) in the form
(50.11)

[d

kl
\l X b = h~a dZ 2 (bah a) - dZa (b 2h2 )

d ]
+ hak2hl [d
dZ a (b hl) - dZ I (baha)
1

ka[d
,.
d
]
+ hlh2
dZ I (b2h2) - dZ 2 (b hl)
1

126

We note that if the curvilinear coordinates happen to be rectangular

cartesian coordinates, then (50.11) reduces to (46.12), as expected.
Problems
1. If f = (X l )2X2 + (X2)2X3 - X l X 2X 3 ' find the directional derivative
of f at the point A(1,-4,8) in the direction of the position-vector
a of A.
2. Iff = Xl sin (7tX2 ) +X3 tan (7tX 2), find the directional derivative
off at the point A (1, 0, - 2) in the direction of the vector drawn from
A to the point B (3, -3,4).
3. Find a unit vector normal to the surface X~s - X 3X l +Xl X 2 - 1 = 0
at the point A (1,2, -1).
4. Two surfaces X l X 2 - (X3)2+ 15 = 0 and (X2)2 - 3xa+5 = 0 intersect in a curve G. At the point 4. (3,-2,3) on G find (i) the angle
between the normals to the two surfaces, (ii) a unit vector tangent
toG.
5. If f = (X l )2+2 (X 2 )2+2 (X3)2, find the maximum value of the
directional derivative offat the point (1, -2, -4).
6. If x = xlil +xJ2+x3ia, prove that 'Vx = "/x, and that

Vx = nX'2",
where n is a constant.
7. Find 'Vr and 'V6, where rand 6 are the usual plane polar coordinates. Also, find the magnitudes and directions of 'Vr and 'V6.
8. If f = r3 - cos 2 6, where rand 6 are plane polar coordinates,
find 'V f in terms of r, 6 and the unit vectors i l and i2 associated with
the corresponding rectangular cartesian coordinates.
9. Iff and g are scalar fields, prove that
'V(j/g)

= g.2 (g 'Vf- f'Vg).

10. Iff = (Xl)2+X3 V(Xl)2+ (X2)2 and g = X l X 2Xa, find at the point
A (3,4,5) the expressions 'V (jg) and 'V (j/g). Note Theorem 4 of
45, and Problem 9 above.
11. Iff = XlX~3' a = xlil - x 2i 2 and b = xsxli2 - xlxJa, compute
the following: (i) (a 'V)f, (ii) (a. 'V) b, (iii) (a X 'V)f, (iv) (a X 'V)' b,

127

(v) (ax\7)xb, (vi)\7.b, (vii)\7xb, (viii)a(\7xb).

12. Let Sand S' be two rectangular cartesian coordinate systems.
The axes of S can be moved into coincidence with the axes of S' by
a positive rotation of
radians about the X3 axis, followed by a
second positive rotation of
about the bisector of the angle between
the positive axes of Xl and X 2 (i) Express the coordinates of S' in terms
of the coordinates of S, and conversely. (ii) If 1 = X 2X3 +XaX I and
b = (Xl +X2 ) i l + (Xl - X 2 ) i2+Xaia, express 1 and b in terms of quantities pertaining to the system S'.
13. Prove Equation (47.3).
14. Starting from Equations (46.8) and (46.11), verify Equatiors
(48.2), (48.4), (48.6), (48.7), (48.8), (48.9), (48.10) and (48.11).
15. Using the identity (48.14), verify Equations (48.4) and (48.5).
16. Using the permutation theorem for scalar triple products,
verify Equation (48.7).
17. If a is a constant vector, prove that \7 (a. x) = a.
18. Provethatax(\7Xx) =0.
19. If a is a constant vector, prove that \7 X (a X x) = 2a.
20. Prove that
\7 (a. b) = a(\7. b) +b (\7 .a) +(a X \7) xb+(b X \7) xa.
21. Prove that
aX (\7 xb) - (ax \7) xb = a(\7 b) - (a \7) b.
22. Prove that
(a.\7)a =!\7a2 -ax(\7xa).
23. Prove that
a(\7.a) =!\7a2 -(ax\7)xa.
24. If1(xl , X 2 , xa) is a homogeneous polynomial of degree n, prove
that (x \7)1 = nj.
25. Prove that if a is a constant vector, then
(a X \7) . (b xc) = (a . b) (\7. c) + (c \7) (a. b)
- (ac) (\7b) - (b. \7) (a.c).
26. If r, G, Z are cylindrical coordinates, describe their parametric
surfaces and show that

t7t

128

t7t

27. If r, e, cp are spherical polar coordinates, describe the parametric surfaces and lines, and show that
(dS)2 = (dr)2+ r2(de)2+r2 sin2e (dcp)2.
28. Prove that for the transformation from rectangular cartesian
coordinates to orthogonal curvilinear coordinates Zl' Z2, Za for which
the metric form is as given in Equation (49.5), the Jacobian I satisfies
the relation hlh2hsl = 1.
29. Express Equation (50.11) in terms ofa determinant.
30. Write out the expressions VI, V b and V xb in the case of
cylindrical coordinates r, e, z.
31. W ri te out the expressions VI, V b and V X b in the case of
spherical polar coordinates.
32. By setting b = Vf in Equation (50.8), deduce an expression
for V2f in terms of general orthogonal curvilinear coordinates.
33. Show that, in cylindrical coordinates r, e, z, we have
2
V2f = d2f !. d2J
df
N.
dr2
r2 de 2
dZ 2 r dr

+1

34. Solve the differential equation V2J =

when f is a function
only of the cylindrical coordinate r.
35. Show that, in spherical polar coordinates r, e, cp, we have

V21" = d2f + !. d2J + _1_ d2f + ~ df + cot e df.

J
dr2
r2 de2 r2sin2e dcp2
r dr
r2 de
36. Solve the differential equation v2f =
when f is a function
only of the spherical polar coordinate r.
37. If kl' k2 and ka are the unit vectors associated with the spherical polar coordinates r, e and cp, prove that
dk 1
dk1
k . e
1
Tr
= 0,
J6 = k 2, dk
dcp = a SIn ,

d;2

d;a

= 0,

0,

~2 = -kl>

~2 = ka cos e,

~~a =

~a =

0,

-kl sine -k2 cos

e.

Note: the corresponding problem in the case of plane polar coordinates is worked out in Chapter III, 30.

129

CHAPTER

lNTEGRATION
51. Line integrals. A curve is called a regular arc it can be represented
in some rectangular cartesian coordinate system by the equation
x

xI(u)il+X2(U)i2+X3(U)i3'

where x is the position-vector of a general point X on the curve, u

is a parameter with the range (l.';;;; u .;;;; ~, and for this range of u the
functions xl(u), x2(u) and X3(U) are continuous with continuous first
derivatives. A curve which consists of a finite number of regular arcs
joined end to end, and which does not intersect itself, is called a
regular curve. Throughout this chapter we shall consider only regular
curves, and shall refer to them simply as curves.
Let us consider a curve C with terminal points A and B, as shown in
Figure 55. Let1 (Xl' X 2 , x3 ) be a function which is single valued and

-~B
K.

Q..

Q,

Q.

Figurtl 55

continuous on the curve C. We divide C into N parts by the N + 1

points Qo, QI' Q2' "', Qx, as shown. The length of the line segment Qp.!Qp) (p = 1,2, "', N) is denoted'by Asp" Let Xp be a point
on the arc QP.IQp, and let its coordinates be (XPI' XP2 , Xp3)' The line
integral of1 over C is then defined to be
N

(51.1)

lim ~ 1 (XPI , Xp2 , xp3 ) Asp =

N .... ooP=!

f 1 ds
C

t>.sp .... 0

This limit is independent of the manner in which the curve C is divided

into parts, since 1 is continuous and single valued on C. If 1 = 1
130

everywhere on C, then Equation (51.1) defines the arc length of C.

Let X be a general point on the curve C, as shown in Figure 56.

Figure 56

Let s denote the arc length of C measured from the end A of C. The
vector dx/ds was seen in 28 to be a unit vector tangent to C in the
direction of s increasing. Denoting this vector by t, we have
(51.2)

dx

= ds .

Let b(x 1 , x 2 , x 3 ) be a vector field defined over C. The orthogonal

projection of b on the unit tangent vector t is caned the tangential
component ofb. Ifwe denote it by b" we have

bt = bt.

(51.3)

The line integral ofb over C is defined to be

(51.4)

f b, ds
c
Because of (51.2), we have

= Jb.t ds.
c

(51.5)
Thus
fbtds = fbodx = f(bldxl+b2dx2+b3dx3)'
c
c
c
We maY,also consider line integrals over C with integrands which
are vectors. The integration of vectors was defined in 13. Following
this definition, we have

(51.6)

(51. 7)

1c b ds = i

f b1ds+i 2 f b2ds+i3 f b3ds,

c
c
c

131

(51.8)

jbxtds = jbXd:x
c
= i1

!c (b dxa-badx
2

2)

+i 2! (badxl-bldxa) +ia! (b 1dx 2-b 2dx 1 ).

c

To apply the above considerations to two-dimensional problem;

involving line integrals along curves in the X 1X 2 plane, it is only necessary to set ba = Xa = 0 in the above formulas. A few examples will
now be worked out.
Example 1. Let i = (X 1)2+ (x 2)a, and let us evalute the line integral
ofi along the straight line X2 = 2Xl in the X1X2 plane from the origin
to the point B(2,4). This problem is two-dimensional. There are
several ways of solving this problem, since a curve may be represented
parametrically in many different ways. We present here two solutions.
(i) The curve C can be represented by the equations
Thus
d" = i1tix1+i2tix2 = (ic1- :ii2) du,
ds = Id"l = V5 duo
Also, i = u2 +8u3, whence
2

jids

= !(u2 +8ua) VS du = lO~V5.

(ii)

On C we have

ds

X2

= 2x1,

V1+ (:::Y dX

= (X1)2+8(X1)3,
1

Vs dx

Thus
2

jids
c

= ![(x1)2+8(x1)a] Vs dX 1

= lO;v'5.

Example 2. Let b = XJl (Xa+X1)2i2+xlia, and let us evaluate

the line integral of b over the curve C in Example 1 above. We
present two solutions.

132

(i)
(51.9)

We have
bdx = bldxl+b2dx2+badxa
= x2dx I + (xa +XI )2 dx2+xl dxa
=

But on C we have

[X 2+(Xa+X I)2
X2

2x 1, Xa

~:: +xI ~::J dx

0, so

jb.dx
C

(ii)

= j[2X I +2(X1 )2] dX I =

8
23

From (51.9) we have

b . dx = [ X2 dXI
dX + (X 3+X I )2 +X I dX3]
dX dX2
2
2

But on C we have Xl

tX2' Xa = 0, so
4

j bdx
C

= j[tX2+!-(X2)2] dX2 =

8
23

Example 3. Let b = a2xlil +ax~ai2+XI(x3)2 ia where a is a constant,

and let us evalute the line integral ofb along the curve C in Figure 57
"
B (0,

G,a)

"
Figure 57

from the point A to the point B, the curve C being a portion of the
intersection of the cylinder (X I )2+(X2)2 = a2 and the plane xI+xa = a.
133

Now

But on C we have
(X2)2

= a2 -(Xl)2,

Xa =
Xl

dX a

dX l

X2

dX l

-=--,

Thus

a-

dX 2

Xl

= -1.

j bdx = j[a2x l
C

a(a -

Xl)Xl - Xl

(a -

Xl)2]

dX l = - ia4..

52. Surface integrals. A regular surface element is defined to be a

portion of a suIface which, for some orientation of the coordinate
axes can be projected onto a region S' in the X l X 2 plane enclosed by a
regular closed curve, and which can be represented by the equation
Xa = g(Xl' x 2), where g(Xl' x 2) is continuous and has continuous first
derivatives-in S'. In this chapter we shall consider only surfaces composed of a finite number of regular surface elements.
Let us consider a surface S bounded by a closed curve C, as shown
in Figure 58. Letf (xu X 2 , xa) be a function which is single valued and

Figure 58

continuous on S. We divide the region S into N parts with areas

aSp (p = 1, 2, ... , N). Let Xp be a point on the element of area aSp,
as shown, and let its coordinates be (XPl' x p2 , xpa). The surface
integral off over S is defined to be
134

(52.1)

lim

2;

N-+oop=l
ASp -+ 0

f(xpl> xp2 , xpa) tl.Sp = !IdS.

S

This limit is independent of the manner in which S is divided into

parts, since f is continuous and single valued throughout the region
S. Iff = 1, then Equation (52.1) yields the surface area of S.
Let us suppose that S is a regular surface element, as shown in Figure
59. To evaluate the surface integral in Equation (52.1), we let X be

1--+---',

. . . ----5'---X'

"

OS'

c'-----Figure 59

a general point on S and let dS be an element of area at X. We now

project S onto the region S' in the X 1X 2 plane, X and dS projecting into
X' and dS', respectively. Let n be the unit vector normal at X to the
surface S, making an acute angle with the xa axis. Then

dS' = nadS.
Let the equation of the surface S be
(52.2)
Denoting the left side of this equati.on by C, we have by Theorems 2
and 3 of 44,
DCI DC dC. dC. dC.
n

Iv

= -

= -::;11 +
oX
I

dXa
Til -

uXI

dXa

-:J

uX 2

~ 12
uX 2

+ oXa
.;:;- la

1 2 +1 a

135

Thus
(52.3)
We then have
(52.4)

[fdS ~ [f[X" x".~(x" x,)]

[1+(;:)' + (};JJ ~S"

To evaluate the integral on the right side, we may write dS' = dX l

dx 2 , and then perform a double integration with respect to Xl and
-;.~- over the region S'. Or we may use polar coordinates rand e in
the X 1X2 plane, writing dS' -: r dr de.
If S is not a regular surface element, we divide it into regular surface elements. The surface integral off over S is then found as the
sum of the surface integrals off over these regular surface elements.
To distinguish between the two sides of a surface S, let us designate
one side as the positive side and the other as the negative side. Let
D be the unit vector normal to S at a general point X, and lying on the
positive side of S. Let b(xl' X 2 , x3 ) be a vector field defined over S.
The orthogonal projection of b on n is called the normal component
ofb. Ifwe denote it by bn , we have
bn = bn.

(52.5)

The surface integral ofb over S is defined to be

(52.6)

fbndS=fb.ndS.
s
s
It is often convenient to introduce an infinitesimal vector dS defined
by the relation

(52.7)

dS

= n. dS,

so Equation (52.6) may take the form

(52.8)

fbndS = fb.dS.
s
s

It is sometimes necessary to consider surface integrals with integrands which are vectors. For example, we have, following the definition of integration of vectors in 13,
136

(52.9)

jb dS = iljbldS+i2Ib2dS+iajbadS,
s
s
s
jb xndS = ilj (b2na-ban2)dS+i2j (bancblna)dS
s

(52.10)

+ia j (b l n2 - b2nl )dS.

s

We shall now work out some examples.

Example 1. Let] = (X I )2+2x2+xa-1, and let us evaluate the
surface integral of] over a region S consisting of that part of the plane
2XI +2x2 +xa = 2 lying in the first octant.
The region S is shown in Figure 60. It is a regular plane element.

-,

"

Figure 60

rJxa =

We have

{}xa

dX 1
'dX 2
whence (52.3) yields na = 1/3. Thus dS

= _2
= 3 dx l dx 2, and so

j]dS = j 3] dXI dx 2
S

But on S, we have
Xa
]

= 2 - 2XI - 2x2 ,
= (X I )2+2x2+ (2 - 2x l

2x 2) - 1

= (Xl - 1)2.

137

Thus, if S' denotes the projection of S on the

X IX2

plane, we have

I-x,

1jdS =31 (Xl - 1)2dx dx =31 1(Xl - 1)2dx dx I = {.

l

S'

Example 2. If b = x 2i l +x ai 2 , evaluate the surface integral of b

over the region S in Example 1, the origin being on the negative side
of S.
We have
1
2 n-2
n
nI -- 3'
2 - 3'
a -- 3'
bn = bInI +b 2n2+b ana = !(X 2+Xa) ,
dS = 3dx I dx 2.
Thus

1bndS = 21 (x 2+xa)dxI dx 2

But on S we have Xa

= 2 - 2XI -

2x2 , whence
1

I-XII

1bndS = 2 1(2 - 2x -x2)dxI dX2 = 2 11(2 -2XI -x2)dxI dX2 = 1.

l

s'

0 0

53. Triple integrals. Let V be a region in space enclosed by a surface S as shown in Figure 61. Let.f (Xl' X 2 , xa) be a function which is

Figure 61

single valued and continuous throughout V. We divide V into N

parts with volumes ~ Vp (p = 1, 2, ... , N). Let Xp be a point in the
element of volume ~Vp, as shown, and let its coordinates be (xPl>
xp2 , xpa). The triple integral of j over V is defined to be
(53.1)

138

This limit is independent of the manner in which V is divided into

parts, since]is single valued and continuous in V.
We may also consider triple integrals with integrands which are
vectors. Thus, if b is a vector field which is single valued and continuous in V, we have, following the definition of integration of vectors
in 13,
(53.2)

fb dV= ilf b dV+i f b dV+ia f badV.

l

v
v
v
v
To evalute triple integrals, one may divide the region V into elements by means of three systems of planes parallel to the coordinate
planes. The value of the integral is then found by the performance of
three integrations with respect to the rectangular cartesian coordinates
Xu X2 and Xa. Or we may divide V using parametric surfaces of a
curvilinear coordinate system, in which case we evaluate the triple
integral by performing three integrations with respect to the three
curvilinear coordinates. Since most readers will have had considerable
experience in evaluating triple integrals in connection with elementary
calculus, no example need be given here.

Problems
1. If x is the prosition-vector of a general point on a circle C of
radius a, and t is the unit tangent vector to C, evaluate
dx.

ft.

2. If] = (XI)2 - (X2)2, evaluate the line integral of] along the line
= 2 from the point A(O,I) to the point B(2,0).
3. Evaluate the line integral in Problem 2 when the curve C consists
of (i) the two line segments AD and DB, where D has coordinates
(1,1), (ii) the two line segments AO and OB, where 0 is the origin.
4. If ] = 8lxI - 9, evaluate the line integral of] along the curve
(X2)2 = (X I)3 from the origin to the point (1,1).
5. tf] = X2Xa+X3XI +XIX2' evaluate the line.integral of] from the
origin 0 to the point B( 1,2,3) along the path consisting of (i) the line
segment OB; (ii) the three line segments OD, DE and EB, where D
and E have coordinates (1,0,0) and (1,2,0), respectively.

Xl +2X2

139

~.

If J =

xI +X z+X3 ,

evaluate the line integral

ofJ along the curve

x = a cos u il+a sin u iz+a u cot oc i 3 , (O<u<t7t),

where a and oc are constants.
7. If b = (Xl - xZ)i l +x 2i z, evaluate the line integral of b along the
following curves: (i) the path in Problem 2 above, (ii) the two
paths in Problem 3 above, (iii) the path in Problem 4 above.
8. If b = 2xlxJI+[(X I)Z- (xz)2]i2, evaluate the line integral ofb
from the point A(O,O) to the point B(l,l) along the following curves:
(i) X 2 =XU (ii) (Xl)2 =X2 , (iii) Xl = (X 2)2.
9. Ifb is as given in Problem 8, evaluate the line integral ofb from
the point A(a, 0) to the point B(O,a) alollgthecircle (X I)2+(X Z)2 = a2 ,
using the polar angle 6 as the parameter varying along the curve.
10. Evaluate the line integral in Example 3 of 51, using for the
curve C the parametric representation
x = a cos u il+a sin u i 2+a(1-cos u) is, (O<u<t7t).
11. If b = XJI +X3iZ - xli a , evaluate the line integral of b along
the curve in Problem 6.
12. If C is the curve in Problem 6, evaluate the following:

(i)fxds,
(ii)fxxtds.
c
c
13. If J = Xl +X a , evaluate the surface integral ofJ over the region
S consisting of the triangle cut from the plane 6x 1 +3x2 +2x3 = 6 by
the three planes Xl = 0, X 2 = 0, Xl +X2 = 1.
14. If b = xliI + (X2) 2ia, evaluate the surface integral ofb over the
region S in Problem 13, the origin being on the negative side of S.
15. If J = (X I)2+(X Z)2, evaluate its surface integral over the region
S consisting of the part of the surface X3 = 2 - (XI)2 - (X2)2 in the: first
octant. Use polar coordinates in the X I X2 plane.
16. Find the area of the region S in Problem 15.
17. If b = (x l)2i 1 +x3i a ,evaluate the surface integral ofb over the
region S of Problem 15, the origin being on the positive side of S.

54. Green's theorem in the plane. This theorem is as follows. Let S

be a closed region in the X I X 2 plane bounded by a curve C, as shown in
140

"
i,

c
. 0 ' - - - - + _ - - x,
i,

Figure 62, and let b be a vector field continuous and with continuous
first derivatives in the region S. Then

!i (\7Xb) dS = c!b.tds,

(54.1)

where the integration over G is carried out in the positive direction,

that is, in that direction of travel around C in which the interior of S
lies on the left. This direction is indicated by the arrow on C in Figure
62. In Equation (54.1), is is as usual a unit vector which forms with
i l and i2 a right handed triad, and t is a unit vector tangent to G in
the positive direction.
Equation (54.1) is often written in the form
(54.2)

db2)
j (bldxl~b2dx2)'
j (dbl
dX - dX dS = I
2
c

We first prove that

(54.3)

jsdbdx. dS = - cj bIdxl>
l

in the case when G can be cut by a line parallel to the X 2 axis in two
points at most. Figure 63 illustrates the situation. On G there are
two points D and E where the tangent to G is parallel to the X 2 axis.
Let d and e be the abscissas of D and E, respectively. These points
divide G into two parts G' and Gil. At a general point X(x l , x 2 ) in S
we introduce an element of area lying in a strip parallel to the X 2 axis,
the left edge of the strip cutting G' and Gil at the points X' (Xl' x 2')
and X"(XI, X2/1), as shown. Then
141

Jb1(X1,

X2') dX 1 -

f b1(X1,

= -

X 2") dX 1

f b1(X1,

X2') dX 1 -

Jb1(x'1,

X2") dXl

Let us now consider the case when C can be cut by a line parallel to
the X 2 axis in more than two points, such as the case of the curve C
in Figure 62. Here we have only to join the points F and G where
there are tangents parallel to the X 2 axis by a curve K which is contained

O;M-L---d.~--1------------+I-'"

Figure 63

in S and which cannot be cut by a line parallel to the X 2 axis in more

than one point. The curve K divides S into two portions for both of
which the above proof of (54.3) holds. Hence, is we apply (54.3) to
both portions, and add, the two line integrals over K cancel, and we
hence establish (54.3) for the entire region S. In a similar fashion we
can prove Equation (54.3) for the case when several curves such as

142

K are required. And we can do likewise when S is multiply connected,

that is, S has holes in it and C then consists of several isolated parts.
In an analogous fashion we can prove that
(54.4)
Subtraction of (54.4) from (54.3) then yields (54.2). This completes
the proof.
55. Green's theorem in space. This theorem is as follows. Let V be
a closed region bounded by a surface S, and let b be a vector field
continuous and with continuous first derivatives in V. Then

(55.1)

!\l.bdV=!b.ndS,
s

where n is the unit outer normal vector to S.

This theorem is also called the divergence theorem, and can be
written in the form
(55.2)
We shall first prove that
(55.3)

f ;;3

dV =

h3na dS ,

in the case when S can be cut by a line parallel to the Xa axis in two
points at most. Figure 64 illustrates the situation, T being the projection of S on the X 1 X 2 plane. On S there is a curve C consisting of points
where the tangent plane to S is parallel to the X3 axis. The curve C
cuts S into two portions S' and S". At a point X(Xl> X2 , xa) in V we
introduce an element of volume lying in a prism parallel to the X3
axis, the vertical line through X meeting S' and S" at the points X'
(Xl> X2, x3 ') and X" (xl> X2 , xt), as shown. Thus
143

(55.4)

=!

![b 3(X I , X2 , X'3) - b3(X I , X2 , x3T] dX 2 dXl"

S'

S"

o.

_________ ---x,
--------T

Figure 64

Let n' be the unit outer normal vector at X', and let dS' be the area of
of the element cut from S' by the vertical prism. Let us define n" and
dS" at X" similarly. Then we have
dX 2 dX 1 = n3' dS' = -na" dS",
and so we may write (55.4) in the form

! ~:3 = !
dV

S'

ba(XI' X2 , x3')n 3'dS,+j b3(X I , X2 , x 3")n 3"dS"

S"

=f b3nadS.
S

Let us now consider the case when S can be cut by a vertical line in
more than two points. In such cases we can always divide V into a
number of regions VI' V2 , by cutting V by a number of surfaces
Xl> K 2 , so chosen that the boundary of each of the regions VI)
144

V2 , can be cut by a vertical line in two points at most. The above

proof of (55.3) then applies to the regions VI' V 2, '. If we apply
(55.3) to VI' V2 , and add, the surface integrals over Kl> K 2 ,
cancel, and we hence establish (55.3) for the entire region V.
In a manner similar to the above we can prove that
(55.5).

!~!:dV= !b n

2 2

dS,

!~!3dV= !b n

S 3

dS.

Addition of Equations. (55.3) and (55.5) then .yields (55.2). This

completes the proof.
If] is a scalar field with continuous second order derivatives, then
we can set b = \7 ] and substitute in Equation (55.1) to obtain

f \7. (\7 f) dV = fs \7f

dS,

or
(55.6)

!(\7.\7)]dV = ! t dS,

v
s
where \7. \7 is the familiar Laplacian operator, often denoted by \7 2 ,
and d]jdn is the directional derivative of]in the direction of the outer
normal to the surface S.

56. The symmetric form of green's theorem. Let] and g be scalar fields
with continuous second derivatives in a closed region V bounded by
a surface S. We may then apply Green's theorem as stated in Equation (55.1), but with the vector b replaced by ] \7 g. This yields
(56.1)

fv \7. (j\7g) dV = sf f\7gn dS.

But

\7. (j\7g) =] (\7. \7)g+\7 f \7g

=]\72g+\7 f \7g,
because of Equation (48.1). Also, \7 g. n is equal to the directional
derivative dgjdn of g in the direction of the outer normal D to S. Thus
Equation (56.1) becomes
145

fj~! dS.

fUV 2g+V!-Vg) dV =
v
s

(56.2)

Similarly, by an interchange ofj and g in the above, we obtain

(56.3)

f (g

vy+ Vg vj) dV =

fg

v
s
Subtraction of (56.3) from (56.2) then yields

dS.

f(i~!

f (jV2g - gVY) dV =
- g Z)dS.
v
s
This equation is called the vmmetric jorm oj Green's theorem.
(56.4)

57. Stokes's theorem. This theorem is as follows. Let S be a closed

region on a surface, the boundary of S being a curve C. We choose
a positive side for S, and let n be the unit vector normal to S on the
positive side. The positive direction on C is defined to be that in which
an observer on the positive side of S would travel to have the interior
of S on his left. Let t be the unit vector tangent to C in the positive
direction, and let b be a vector field with continuous first derivatives
in the closed region S. Then Stokes's theorem states that
f n(v Xb) dS

(57.1)

fb.t ds,

where the integration around C is carried out in the positive direction.

This theorem can also be written in the form
(57.2)

f
S

[nl (dbdX32- dbdX32'), + n2(~bl_

db 3) + n3 (~b2 - ~bl)JdS
()x 3 dX I
()X2.
()x I

= f (bIdxI+b2dx2+b3dx3)

We shall first prove that

(57.3)

f n (V X bIil ) dS
S

fbI dXl>
c

in the case when S is a regular surface element and the positive side of
146

S is that side on which the unit normal vector n points in the direction
of increasing Xa' Figure 65 illustrates the situation, and shows the

S'

C'

Figure 65

unit tangent vector t of C and also the region S' in the

which S projects.
Now

jn.('\1Xb l i 1 )dS= jn.(i2

(57.4)

X 1X 2

plane into

~!l-ia~!l)dS.

sa

.. 2

Let the equation of the surface Sbe X3 = g(Xl' x 2). Then on S, we have

bl[Xl,X2,xa(XI,X2)]
dC 1
dX 2

(57.5 )

__

db 1
dX 2

C1 (XI,X2) '

db 1 dXa.
dX a dx 2,

We now substitute from this equation for db 1 jdX 2 in Equation (57.4)

to obtain the relation

(57.6)

j
s

n ('\1 X b1i l )

dS = - j

ni3~:~ dS+ j

s
= - /1 +/2 ,

n'

(i2 +i3 ~::)

~!~ dS

where 11 and 12 denote the two integrals on the right side of this
equation.
Let us consider 11' We have n.i a dS = ns dS = dS', where dS' is
the projection of dS on the X 1X 2 plane. Since C1 is a function of Xl and
X 2 only, we can then write
147

11 = - f;~ dS'.
s'

By Green's theorem in the plane, as stated in Equation (54.2), we

then have
(57.7)

11

=f

C1 (X 1 ,x2 )dx 1

=f

C'

b[x1> x
1

2,

X3 (X1>x 2)]dx 1

C'

!b dx
1

We now consider 12 , The position-vector of the general point X

on Sis
Hence

and so
(57.8)

12

dx db

1
n"')"')
dX 2 uX 3

dS.

But the vector dX! dX 2 is tangent at X to the curve of intersection of S

and a plane parallel to the X 2X 3 plane. Hence this vector is tangent to
S and is then perpendicular to the unit normal vector n, so that

dx

n-

dX 2

=0.

Thus Equation (57.8) yields 12 = 0, and from Equations (57.6) and

(57.7) we can then conclude that Equation (57.1) is true.
When the positive si.de of S is chosen so that the unit normal vector
n points in the direction of decreasing X3, the proof of Equation (57.3)
is similar to the above, the only differences in the proofs being that
in the present case n3 is negative and the direction of integration around
the curve C is opposite that in the above proof.
When the surface S is not a regular surface element, we divide it
into a number of regular surface elements S1> S2' ... by a number of
curves L 1 , L 2 , '. The above proof of (57.3) then applies to the
regions Sl' S2' .. '. If we apply (57.3) to these regions, and add, the
line integrals over L 1 , L 2 , cancel, and we hence establish Equation
(57.3) for the entire region S.
148

In a manner similar to the above we can prove that

b2dx 2 , f n (\7 X b3i 3)dS = f b3dx 3

e s c
Addition of Equations (57.3) and (57.9) then yields (57.1). This
completes the proof.
(57.9)

f n (\7 X b2i 2 )dS

=f

58. IntegrationJormulas. Green's theorem in space and Stokes's

theorem were considered in 55 and 57, respectively. These theorems are integration formulas which we may write in the form

(58.1 )

f \7.b dV = sf nb dS,

(58.2)

f(nX\7).bdS= ft.bds.
s

Both of these theorems involve a vector field b; Equation (58.1)

presents a transformation from a triple integral to a surface integral,
while (58.2) presents a transformation from a surface integral to a
line integral. We shall now introduce four other integration formulas,
which we shall state as two theorems.
Theorem 1. Let V be a closed region in space bounded by a surface S with the unit outer normal vector n, as in the case of Green's
theorem in space. LetJ and b be two fields with continuous firstderivatives in V. Then
(58.3)

f \7JdV = f nJdS,
v
s

(58.4)

f \7 X b dV = f n X b dS.
v
s

Theorem 2. Let S be a closed region lying on a surface and bounded

by a curve C, n being the unit positive normal vector to Sand t being
the unit positive tangent vector to C, as in the case of Stokes's theorem.
LetJ and b be two fields with continuous first derivatives in S. Then

(58.5)

f (n X \7)J dS = f tJ ds,
s
c
149

(nx \7) xb dS = txb tis.

s
c
Proof of Equation (58.3). Let e be a constant vector field. If in
Equation (58.1) we set b =.ie, we obtain

(58.6)

(58.7)

f \7. (fe) dV = f n-fe dS.

But we have

\7. (fe) = \7 Ie,

by Equation (48.1), since e is a constant vector. Thus Equation
(58.7) may be written in the form

e'[f\7JdV-/ nJdS] =0.

Since e is an arbitrary constant vector, the expression in the square
brackets must vanish, whence Equation (58.3) is proved.
Proof of Equation (58.4). We again introduce the constant vector
field e, but in Equation (58.1) we replace b by bXe to obtain the relation

(58.8)

f \7 . (b

c) dV =

f n (b

c) dS.

v
s
Since e is a constant vector, we have by the permutation theorem for
scalar triple products the relations

\7.(bxe) = e(\7xb), n.(bxe) = e.(nxb.)

Thus Equation (58.8) may be written in the form
e.[!\7Xbdv-/nXbdS] =0.
Since e is an arbitrary constant vector, we conclude that (58.4) IS
true.
Proofs of Equations (58.5) and (58.6). To prove these twoequations we replace b in Equation (58.2) first by Je and then by b X e,
where e is a constant vector field. The procedure then follows that in
the previous two proofs. The details are left as exercises for the reader
(Problem 11 at the end of this chapter).
150

The six integration formulas (58.1)-(58.6) may be written compactly in the form

! \7* T dV = s! n* T dS,

(58.9)

(58.10)

!(nX\7)*TdS=!t*Tds,
c

where T can denote a scalar field or a vector field, and the asterisk has
the following meanings: if T is a scalar field, it denotes the multiplication of a vector and a scalar; and if T denotes a vector field, it
denotes either scalar or vector multiplication. Thus, for example,
if T denotes a vector field b and the asterisk denotes scalar multiplication, then Equation (58.10) becomes (58.2).
59. Irrotational vectors. A vector field b (xu x 2 , x 3 ) is said to be
irrotational in a region V in space if everywhere in V we have

(59.1 )

\7xb =0.

Let Cjl be any scalar field with continuous second derivatives; and
let us write b = \7Cjl. Then

\7xb =.\7x\7Cjl =0,

so' a vector b defined as the gradient of a scalar field is irrotational.
We shall now show that an irrotational vector field b has the following properties:
(i)
Its integral around every reducible circuit in V vanishes.
When V is simply connected, b is the gradient of a scalar
(ii)
field.
To verify the first property, we consider a general circuit in V
which is reducible, that is, it can be contracted to a point without
leaving V. Let S be a surface entirely in V and bounded by C. If we
assume that b has continuous first derivatives, then Stokes's theorem
(57.1) yields

!b.tds =! n(\7xb) dS =
c

by Equation (59.1).

151

To verify the second property, we let X be a general point in V,

and let Xo be a given point. We also let G' and GIf be any two paths
III V from Xo to X, as shown in Figure 66. Because of property (i)
x

C'

Xo

Figure 66

above, the line integral of b from Xo to X is the same for paths G' and
Gil and hence has the same value for all paths in V from Xo to X. Thus,
if we write
x

(59.2)

cp = jb.dx,
Xo

then cp depends only on the coordinates (Xl' X 2 , x3 ) of X.

tiation of Equation (59.2) yields dcp = bdx, or
(59,3)

drp
ds

Differen-

= b.dx.
ds

But drp/ds is the directional derivative of cp, and by Equation (44.3) is

equal to \}cp.(dx/ds). Thus Equation (59.3) may be written in the
form
dx
(\}cp-b)'ds =0.
Since dx/ds is an arbitrary vector, then
(59,4)
b = \}cp.
This completes the proof.
The function rp is called a scalar potential function.
60. Solenoidal vectors. A vector field b(Xl' X 2 , x 3 ) is said to be solenoidal in a region V in space if everywhere in V we have
(60.1)
\}b=O.

152

Let ~ be any vector field with continuous second derivatives, and

let us write b = \j X ~. Then
\jb =\j(\jX~) =0.
We shall now show that, ifb is any solenoidal vector field, there exists
a vector field ~ such that b = \j X ~.
To prove the proposition, we must solve the scalar equations
(60.2)

b - d'Pa _ d'P2 ,
I dX 2 dX a
(60.3)
b - ?'PI _ d'Pa ,
2 dX a dX I
(60.4)
b - d'P2 _ d'Pl ,
a - dX I
dX 2
for 'PI, 'P2 and 'Pa, where bl , b 2 and ba are given functions subject to
the condition

(60.5)
Let us choose 'PI = O. Then we have from Equations (60.3) and
(60.4) by partial integrations with respect to Xl'
(60.6)
a,

(60.7)
a,

where a l is a constant and ~2 and \j!a are arbitrary functions of X2

and X 3 To satisfy Equation (60.2) we must have

_J
x

b =
I

a,

'(db 2 + db 3) dX I
dX 2
dX a

+ d~a _ d~2.
dX 2

dX a

Because of Equation (60.5) we can then write

153

This equation is satisfied if we choose

h =

jb (al> X

2,

X:i)

0,

dx 2 ,

a,

where a2 is a constant. The final result is then

({II

= 0,

({I2

jbg(Xl> X

Xg) dXl>

2,

a,

({Ig

= - f"b 2 (Xl> X2 , Xg) dX I + jb l (al> X2 , X3 ) dx 2 ,

a:a

a1

where all integrations are partial integrations, and al and a2 are constants. The function cp is called a vector potential function.
In the above proof, several arbitrary selections have been made.
This indicates that a given solenoidal vector field b does not possess a
unique vector potential function. In order to see this more clearly,
we let cp be one vector potential function corresponding to the solenoidal vector field b, and let j be any scalar field. Then

\7x(cp+\7j) = \7 xcp+\7 X\7j= \7xcp =b.

Thus, cp \7 j is also a vector potential function corresponding to the

field b.
If b is any vector field having continuous second derivatives in a
region V, then b can be expressed as the sum of an irrotational vector
and a solenoidal vector. The proof of this will not be given here.

Problems
1. Let C be a closed curve in the

X IX2

plane. Prove that the area

A of the region S enclosed by C is given by the relation

A =

t f (Xl dX
c

2 -

x 2 dx l )

where the integration over C is carried out in the direction of travel

around C in which the interior of S is on the left.
2. Ifx is the position-vector of a general point X on a closed surface
154

S, and n is the unit outer normal vector to S, prove that the volume
V of the region enclosed by S is given by the relation

V=tf nxdS.
s
3. If S is a closed surface with a unit outer normal vector n, prove
that
nxxdS =0.
s

4. If V is a region bounded by a surface S, and n is the unit outer

normal vector to S, prove that

13 =

!f
s

[(X I)2+ (X2)2] (xliI +x 2i 2 )n dS,

where 13 is the moment of inertia of V about the Xs axis.

5. Ifb has continuous first derivatives in a closed region Vbounded
by a surface S, prove that

fs n(\7xb) dS =0.
6. If b = a l (xI)2iI+aZ(x2)2i2+aS(xs)2is, where al> a2 and as are
constants, evaluate the surface integral of b over the sphere through
the origin with center at the point A (al> a2 , as).
7. If b = (XI)2il+XIXJZ+xsis, evaluate the surface integral of b
over the cube bounded by the planes Xl = 2, X 2 = 2, Xs = 2 and the
coordinate planes.
8. If b = [(XI)2 - xJ i l +[2(x l )z+3xJ i z - 2Xlxaia, evaluate the
surface integral of b over the sphere S with cent~r at the point E
(1,0,2) and passing through the point F(3, -2, 1).
9. If C is any closed curve, prove that dx = o.

lO. Let C be the circle with the equations (X I)2+(xz)2

and let

= 4,

Xs

0,

b = [(X I)2+ X2 ] il+[(XI)z+xa] i 2+xJa.

Evaluate the line integral of b around C in the direction indicated by

155

the fingers of the right hand when the thumb points in the direction
of the positive Xa axis.
11. Prove Equations (58.5) and (58.6).
12. A vector field b has continuous first derivatives in a closed
region V. On the bounding surface S of V, b is normal to S. Prove
that

!VXbdV=O.
v
13. If j and b are fields with continuous first derivatives in a region
V bounded by a surface S, prove that

!jnxb

! xb

dS = j\7
dV+ \7jxb dV.
s
v
v
14. If band care irrotational vector fields, prove that b X c is
solenoidal.
15. If b = x2xail+X3Xli2+XlX2i3, show that b is solenoidal, and
find its vector potential.
16. Show that the vector field
x1i 1+xJ2
(Xl)2+ (X2)2

is solenoidal in any region which does not contain the origin.

17. If r, e and z are cylindrical coordinates, show that V6 and Vln r
are solenoidal vectors, and find their vector potentials.
18. If V b = <II and V X b = f, prove that
dI
Jja
V 2b1 - -JX
()X
1

Jj2
+-,
JX3
2

and derive similar equations involving b2 and ba

156

CHAPTER

VI

TENSOR ANALYSIS

61. Introduction. Tensors are mathematical or physical concepts

which have certain specific laws of transformation when there is a
change in the coordinate system. As we shall see, vectors are just one
of the many types of tensors.
Tensor analysis is a study of tensors. It has many applications. One
of these is in the field of classical differential geometry of the curve and
surface in our ordinary space, as well as the more significant generalizations of this geometry to spaces of higher dimensionality; this generalization is often referred to as Riemannian geometry. Another application is to mathematical physics. Here tensor analysis permits us to
express easily in terms of curvilinear coordinates the fundamental
equations of the various subjects such as hydrodynamics, elasticity,
electricity and magnetism. It also aids in the formulation of the
natural laws of mathematical physics, since such laws when expressed
in terms of tensors are independent of anyone particular coordinate
system.
62. Transformation of coordinates. Let us consider a set of curvilinear
coordinates which we denote by the symbols Z\;:.2 and e. We use
the supercripts to agree with a convention to be introduced later.
We shall also refer to these coordinates by writing Z (r = 1,2,3).
Let us write
(62.1 )
where the three functions l' (r = 1, 2, 3) are single valued and
differentiable for some range of values of z\ Z2 and e. These equations represent a coordinate transformation to new curvilinear coordinates Z,\;:.'2 and Z'3. The Jacobian l' of a transformation such as
157

this one was defined in 49. We express it conveniently here by writing

(62.2)

l'

= I ~~' I '

the ranges for rand s being 1, 2, 3. We assume that l' does not vanish,
whence, as mentioned in 49, Equations (62.1) may be solved to
yield
(62.3)

From Equation (62.1) we have

3

(62.4)

{}

I,

dz" = ~ ~ dz'
s= 1 {}z'

(r=I,2,3).

We now introduce two conventions, as follows:

Range fonvention. A small latin suffix which occurs just once in a
term is understood to assume all the values 1,2,3.
Summation convention. A small latin suffix which occurs just twice
in a term implies summation with respect to that suffix over the range
1,2,3.
Because of these conventions, we may now write Equations (62.4)
in the form
(62.5)

I,

'+Z

{}Z',

d'"

{}z' (",

the repeated suffix s implying the summation. In this equation, the

suffix r is called a free suffix, since we obtain from (62.5) a different
equation for each value of r; on the other hand, the suffix s in (62.5)
is called a dummy suffix, since a change of both of the suffixes s into
any other latin suffix does not alter Equation (62.5).
The Kronecker delta ~TS was introduced in 47. We shall require
it in the present chapter, but shall find it convenient to denote it by
the symbol ~;, so we have
ifr=s

(62.6)
=

158

ifr. =1= s.

We note the identity

dZ dZ's
dZ's dZt '
T

8t

(62.7)

which is true because its right side is equal to di/dZI, which, because
of the independence of the coordinates i, satisfies the relations

dZ

dz l = 1
= 0

if r = t,

ifr =1=

t.

Of course the primed and unprimed coordinates in Equation (62.7)

can be interchanged.
Throughout this section we have considered three curvilinear coordinates z\ z2,;:;1 in space. The considerations here can be applied
equally well to the case of two curvilinear coordinates Z1, Z2 on a
surface, the only modifications being in the range and summation
conventions above, where it becomes necessary to assume that latin
suffixes have the range 1,2. This same modification, when applied to
the rest of this chapter, yields results applicable to two-dimensional
problems. In an analogous manner, by the assumption that latin
suffixes have the range 1, 2, "', N, where N is any positive integer,
the results in this chapter can be generalized to yield the Riemannian
geometry of the N-dimensional Riemannian space. There are many
excellent books on this subject. 1

63. Contravariant vectors and tensors. Let X be a general point in

space with curvilinear coordinates ZT. Let AT denote a set of three
quantities associated with the point X, such as for example components ofa velocity. We now introduce a second curvilinear coordinate
system relative to which the coordinates of X are zIT, and denote the
transforms of AT by Air. We then make the definition: a set of quantities AT associated with a point X is said to be the components of a contravariant vector if they transform, on change of coordinates, according to the equation
1 See, for example: J. L. Synge and A. Schild, Tensor calculus, University of
Toronto Press, Toronto, 1949.

159

(63.1)

{} I,

~
AS ,
(}ZS

A"

where the partial derivatives are evaluated at X.

Because of Equation (62.5) we see that the quantities dz;' are the
components of a contravariant vector. Also, we shall see later on that
when the transformation is from rectangular cartesian coordinates to
rectangular cartesian coordinates, the components of a vector as defined in 6 have the law of transformation (63.1) and are hence also
the components of a contravariant vector.
We shall now define contravariant tensors. Let A" be a set of nine
quantities associated with the point X with curvilinear coordinates
z', and let A" transform into Airs when the coordinates are transformed to zIT. We then have the definition: a set f!f quantities Ars is called
the components of a contravariant tensor of the second order if they transform
according to the equation
{}

(63.2)

I, {}

IS

A'rs~ ~ A tu .
- {}Zl {}z"

Contravariant tensors of higher order are defined analogously. Contravariant vectors are often called contravariant tensors of the first order.

64. Covariant vectors and tensors. The definition of covariant vectors

is as follows: a set oj three quantities A, associated with a point X is said
to be the components oj a covariant vector if they transJorm on change oj coordinates, according to the equation
(64.1)

A,=

{}ZS
(}Z"

As-

As an example, let us consider a function <I>(z1,

(64.2)

{}<I>
{}Z"

Z2,

z3). We have

{}<I> {}ZS
{}ZS {}Zl"

whence we note that the three expressions {}<I> I {}z' are the components
of a covariant vector.
The definition of a covariant tensor of order two is as follows:
160

a set of nine quantities A,s is said to be the components of a covariant tensor

of order two if they traniform according to the equation
(64.3)
Covariant tensors of higher order are defined analogously. Covariant
vectors are often called covariant tensors of order one.
In 63 we have used superscripts in-writing symbols denoting contravariant tensors, and in the present section we have used subscripts in
writing symbols denoting covariant character. This convention will
be followed throughout the present chapter.

65. Mixed tensors. Invariants. We can also define tensors whose law
of transformation involves a combination of both contravariant and
covariant properties. Such tensors are called mixed tensors. Thus,
for example, a set of twenty-seven quantities A:t is said to be the components
of a mixed tensor of the third order with one contravariant suffix and two covariant suffixes (f they transform according to the equation
(65.1 )
It should be noted that (65.1) represents twenty-seven equations,
and that the right side of each of these equations consists of the sum
of twenty-seven terms.
A single quantity A is said to be an invarinat ifit transforms according
to the equation

A' =A.
Invariants may be called contravariant tensors of order zero, or covariant tensors of order zero. An example of an invariant is the temperature in a fluid.
We shall now prove that the Kronecker delta has the tensor character indicated by its suffixes, that is, that it is a mixed tensor of the
second order. We must establish the equation

a:

(65.2)
161

Now

{}z' t
{}Zl t
{}Z" 8. = {}Z" 8,

{}Z 81

+ {}z'S

+ {}Z

81

{}Z'S

3'

When t is given the values 1, 2 and 3, the right side of this relation
reduces respectively to the expressions

{}z'
{}z'S'

{}z'
{}z'S'

{}z'
{}z's'

Thus we may write

and so the right side of Equation (65.2) reduces to

{}z,r {}Zl
{}zt {}z's

But this is equal to the left side of Equation (65.2) because of Equation
(62.7).

66. Addition and multiplication of tensors. If we have two tensors of the

same order and type, their sum is defined to be the set of quantities
obtained by the addition of corresponding components of the two tensors. Thus the sum of the tensors I and I is a set of twenty-seven
quantities C;I given by the relations

A:

B:

(66.1)
I t is easily proved that the sum of two tensors is a tensor of the same
order and type as the two tensors added.
There are two products of tensors, called the outer product and the
inner product. The outer product of two tensors is defined to be the
set of quantities obtained by multiplication of each component of the
first tensor by each component of the second tensor. Thus, for example, the outer product of the tensors
and
is the set of 243 quantities C;:. ~iven by the relations

A:

B:,

(66.2)
In writing this equation, we have been careful to keep each particular
162

suffix at the same level on both sides of the equation. When this
convention if followed, it is easily proved that the outer product of
two tensors has the tensor character indicated by the number and
positions of its suffixes. Thus, for example, in Equation (66.2), C;:., is
a mixed tensor of the fifth order, with two contravariant suffixes and
three covariant suffixes.
We now introduce an operation called contraction. It consists in
identifying a superscript and a subscript of a tensor. Thus, for example,
if from the tensor A:t we form the set of quantities B t defined by the
relation
(66.3)
we are performing a contraction. We note that each component
of B t is equal to the sum of three components of A:t It is easily
proved that contraction of a tensor yields a tensor Thus, for example,
B t in Equation (66.3) is a covariant vector. It should be noted that
we do not perform contractions by identifying suffixes at the same level,
since such operations do not yield terisors, in general. For example,
if A:t is a tensor, A:s is not a tensor, in general.
We now define the inner product of two tensors. To obtain it, we
form an outer product of the two tensors and then perform a contraction involving a superscript of one tensor and a subscript of the
other tensor. Thus, for example, an inner product of two tensors A~
and B" is the set of quantities Cst given by the relation

Cst = A:B,t
Of course the inner product of two tensors is not unique. Since outer
multiplication and contraction of tensors both yield tensors, the inner
multiplication of tensors also yields tensors.

67. Some properties of tensors. One of the most important properties

of tensors is the following: if a tensor equation is true for one coordinate
system, it is true for all coordinate systems. To prove this, let us consider,
for example, a tensor having components t for a specific coordinate
system z', and let us suppose that

A:

163

A:, =0.

(67.1)

Let <." be any other coordinate system, and let A ':, denote the components of this tensor for the coordinate system z". We must show
that
(67.2)
By the laws of tensor transformation we have
"
A sl

d<." d<.v d<.w u _

dZu d<.'s d<." Avw - 0,

by Equation (67.1). The proof is similar in the case of other tensor

equations.
and t denote the components of two tensors for a
Now let
specific coordinate system <." and let us suppose that

A:,

B:

A;, =

B~"

Then, by the previous section, A~, - B~, is a tensor which vanishes

for the coordinate system <.', and hence for any other coordinate system zIT we have
Tensors also have the property of being transitive. To explain this
property, we introduce three coordinate systems z;', <.IT and <.'IT. The
property of being transitive is then the following: if a certain set of
quantities is a tensor for the transformation from coordinates <.' to
zIT, and is a tensor of the same order and type for the transformation
from coordinates <.IT to <."': then for the overall transformation from
the coordinates <.' to lIT the set of quantities is a tensor of this same
order and type. It is easily proved that a tensor of any order or type
is transitive.
68. Tests for tensor character. We shall now demonstrate by some
examples a useful test for establishing the tensor character of a set
of quantities.
Example 1. Let ATS be a set of nine quantities such that AUX.
is a contravariant vector, where X, is an arbitrary covariant vector.

164

We shall now prove that A rs is a contravariant tensor of the second

order. Since A"Xs is a contravariant vector, we have
d 'T
A'rsX's -- ~
(68.1 )
dZ t AtuXU'
But Xu is a covariant vector, so
(68.2)

dZ'V ,
Xu = dZ u Xv'

We now rewrite Equation (68.1), replacing the dummy suffix s on

the left side by D, and substituting for Xu on the right side from (68.2),
to obtain the relation

or
(68.3)

dz" AtU)X' - 0
(A'TV _ dz'T
dZ t dZ u
v

Since X T is arbitrary, so is X'v' Hence the expressions in the brackets

in Equations (68.3) vanish, whence we conclude that A" is a contravariant tensor of the second order.
Example 2. Let A TS be a set of quantities such that ArsXTXS is
an invariant, where XT is an arbitrary contravariant vector. We shall
now prove that Ars is a covariant tensor of the second order provided
it is symmetric in all coordinate systems, that is, provided we have
for every coordinate system relations of the form Ars = A". We
proceed much as in Example 1. Since A"X" XS is an invarint, we have
(68.3)
Since XT is a contravariant vector we can express the XT and XS in
this equation in terms of x,t, obtaining
T
A' X'TX'S = A dZ dz' X'IX'"
"
rs dll dZ'u

We now replace the dummy suffixes rand s on the left side by t and u,
whence we obtain the equation
(68.4)
165

where
(68.5)
Equation (68.4) can be written in the form
(68.6)

IS

bu (X',),+b (X12 )+b (X )+(b.3 +b ) X"X"

+(b., +b 13 ) X 13X" + (b12 +b 21 ) X"X"

O.

Since X'is arbitrary, so are the three quantities X". Hence the coefficients in Equation (68.6) must vanish, which leads to the equation
(68.7)
From (68.5) we then have
I

A tu

+ Aut =

{}ZT {};t
{}Z'U

{}z't

{};t {}z'

ATS

+ {}Z'U {}Z/l Ars

In the last term on the right side, we interchange the dummy suffixes
rand s, obtaining the relation

(68.8)

A ' tu

{}z' {};t (A
A )
+ A'ut = {}Z't
{}z'u rs + sr'

Since we are given that Ars

reduces to the form
I

A tu

= Asr , A'tu = A'ul> then Equation (68.8)

{}z' (}z'

~ ~Ars,
oZ oZ .

which establishes the tensor character of the set of quantities A,s.

69. The metric tensor. Let x, be the three rectangular cartesian
coordinates. Then the distance ds between two adjacent points is
given by the relation
(69.1)

(ds)'

(dx,)'+(dx,)"+(dx.)'

If we introduce curvilinear coordinates

j'(x" x., x.) or x, = g,(z" z', z), then
dx,

dx,dx,.

z: defined by relations

z'

= {}x'd
{}z' ;t,

and substitution in Equation (69.1) yields an expression of the form

166

(69.2)

(ds)2 = bll (dlY+b .. (dl)2+b,,(dt)'+b'3dtdt

+b 3,dldl +b"dz'dz2,
where brs are six functions of z', l, z'. We now define nirie quantities
grs by the relations
b2, = g2.'

bll = gll'

b33 = g",

(69.3)

l.b
2 23 =g23 -g
32,

Ib 31
"2

-g
31 -g
13)

t b"

= g12

g., .

We may now express (69.2) in the form

(dS)2 = gll(dz')2+g2.(dz2)'+g3,(dz')2 (g .. +g.. ) dldt
3
+(g3,+g'3) dz dl+(g,.+g.,) dldz",
or

(69.4)

(ds)'

= grsdz'dz'.

Let us observe the right side of this equation. It is an invariant, since

ds is an invariant; also dl is an arbitrary contravariant vector; also,
from Equations (69.3) defining grs, we see that grs is symmetric.
Hence, by the test for tensor character presented in Example 2 of
68 we can conclude that grs is a covariant tensor of the second order. It
is called the metric tensor.

70. The conjugate tensor. Let g denote the detern;tinant whose elements are the components of the metric tensor. Then

g = Ig,sl
In the expanded form of this determinant, the coefficient of anyone
element grs is called the cofactor of g,S" We denote it by the symbol
f:::.TS. We note in passing that the minor ofgrs is equal to
(-1 y+s /j.rs.
(70.1 )

We shall now prove that

(70.2)

grs

/j.,t =

g 8!,

(70.3)

gs,

/j.t, =

g 8!.

Proof of Equation (70.2). Let us suppose that, when we write the

167

determinant Igrsl, the subscript r varies over the rows, while the subscript s varies over the columns. We now consider Equation (70.2)
when s = t = 1. The right side is equal to g. The left side is

which is just the expansion of the determinant g by elements of the

first column. The proof of Equation (70.2) when r = s = 2 and
r =s = 3 is similar. Now let us consider Equation (70.2) when s= 2
and t = 1. The right side is equal to zero. The left side is

which is just the expansion of a determinant obtained from g by a

replacement of the elements of the first column by those of the second
column; in this determinant the first two columns are identical, and
so the value of the determinant is zero. The proof of Equation (70.2)
in all other cases when rand s differ is similar.
Proof of Equation (70.3). This proof is quite similar to the proof of
(70.2), differing from it only in that the various expansions of determinants are by elements of rows rather than columns.
We now define the conjugate tensor g" by the relation
(70.4)

jj."

g"=-'

Then Equations (70.2) and (70.3) yield the expressions

(70.5 )

g"il

(70.6)

g"i'

= 3~.

3:,

We shall now prove that is is a contravariant tensor of the second

order. Even though Equation (70.5) indicates that g"l' is a mixed
tensor of the second order, and gTS had known tensor character, we
cannot make any direct deductions from this as to the tensor character
of it, since we cannot assign g" arbitrarily without changing gr<. Let
X' be an arbitrary contravariant vector. If we write

168

then Yr is an arbitrary covariant vector. Hence we have

gPrYr = gprgsrX' =

~:X'

= XP,

and so gprYr is a contravariant vector, Yr being an arbitrary covariant

vector. Thus, by the test for tensor character presented in Example
1 of 68, we can conclude that Ir is a contravariant tensor of the
second order.
71. Lowering and raising oj suffixes. In order to avoid ambiguity
in certain steps to follow, we shall adopt the convention of placing
but one suffix in anyone vertical line. We shall use dots to denote
vacant spaces in the rows of suffixes. Thus, for example, we might
write A~, to denote a mixed tensor of the second order. Other examples are B;:t, C~~ t.
If we form inner products of any given tensor with the metric
tensor or the conjugate tensor, we obtain a number of new tensors.
It is customary to use the same principal letter in writing these tensors.
For example, from a contravariant tensor A", we can generate another
second order tensor A:, as follows:

In appearance, this new tensor differs from the original one only in
that the suffix in the second position has dropped from the upper level
to the lower one. We say that we have lowered a suffix. Other tensors
obtained from A" by this lowering process are the following:

In an analogous fashion, the inner multiplication of a tensor by the

conjugate tensor results in a raising of a suffix. Thus for example,
from a given tensor B" we can generate other tensors by this raising
process, as follows:

B:, = gt Btn

B;'

= it B rt , B" = gtt" Btu.

If a certain subscript is raised, and is then lowered, it is easily seen

that the final tensor is the same as the original one. The convention
169

of writing no two suffixes in anyone vertical line was introduced solely

to ensure this property.
72. Magnitude of a vector. Angle between two vectors. Let A' be a
contravariant vector. Its magnitude A is defined by the relation

(72.1 )

A =Vg,)'A'.

From this relation it readily follows that

(72.2)

A = VA.A' =

Vg" A,A,.

A unit vector is one whose magnitude is equal to one. We note that,

if the coordinates are rectangular cartesian, then g" = a,s> and
(72.1) then reduces to the familiar form
(72.3)

= V(A')'+(A')'+(A3)'.

Let X' and Y' be two unit vectors at a point. The angle 6 between
them is defined by the relation
(72.4)

cos 6 = g"X'Y'.

If the coordinates are rectangular cartesian, this equation reduces to

the familiar form
(72.5)

cos 6

= X'Y'+X'Y'+X'Y'.

73. Geodesics. The parametric equations of a curve C may be

written in the form

(73.1)

z' =

1'(s) ,

the arc length s of the curve being used as the parameter. Because of
Equation (62.5) we see that dz'/ds are the components of a contravariant vector. We denote it by P', so
(73.2)

p' = dz'.
ds

Further, by a definition of the previous section, the magnitude

this vector is given by the relation
170

p of

p2 = gmnpmp. = gm.

dz m dz

dS di

(dS)2
(ds)'

l.

Thus p' is a unit vector. It is called the unit tangent vector of the
curve C.
A geodesic may be defined as the curve of shortest length joining
two points. In our three-dimensional space, the geodesics are straight
lines. If we consider surfaces, which are of course two-dimensional
spaces, the geodesics are not necessarily straight lines. For example,
in the case of a spherical surface, the geodesics are the great circles,
that is, those circles on the sphere whose centers coincide with the
center of the sphere.
Let X and Y be two points. The distance L between them, measured
along some curve, is given by the line integral
y

(73.3)

L =

f ds = xf Vgmn d~dzn = xf V; ds,

x

where
(73.4
According to the Calculus of Variations, L is an extremum if the path
joining X and Y is such that
(73.5)

d
ds

(dW\ dW

dP'l- d~ = o.

The actual derivation of these equations is beyond the scope of this

book.
Now W is a function of p' and gm. which are known functions of ;:'.
Thus W = w(~, l, z', p" p', pS). The two partial derivatives in
Equation (73.5) are computed as though the quantities p' and ;:'
are independent. Thus

dW

m n
dp' = gmn Il,'P

+ gmnpm

Il,n

+ gmrpm = grnp + grmpm

grnpn + grnp = 2grnpn,

= grnpn
=

171

d ({}w) 2 dp 2 {}gm m.
{}p' = grn ds + dz m P 'P ,
{}w = {}gm. pmp.

ds

{}z'

{}z'

Hence equations (73.5) become

(73.6)

+ {}grn
pm'P. _.1 {}gmn pm'P. =
{}Zm
2 (}Z'

g dp
rn

ds

But, by an interchange of the dummy suffixes m and n we have

(73.7)

{}gm
{}zm

pmp. =

{}grm P"P m

(}z".

Substituting for one half of the middle term in Equation (73.6) from
Equation (73.7), we then obtain the relation
(73.8)

gm

dll

ds + [mn, r]pmp- = 0,

where
(73.9)

[mn, r] =

t( ~!r; + ~!: -1;)

The quantities [mn,r] are the Christoffel symbols of the first kind. We
note that
[mn,r] = [nm,r].
(73.10)
Ifwe multiply Equation (73.8) by g't, we obtain
(73.11)
where
(73.12)
F!.,. = tTmn,r].
The quantities F~. are the Christoffel symbols of the second kind.
We note that
(73.13)
The first term in Equation (73.11) reduces to
t dpdpt
d' z'
8.-=- =-.
ds
ds
ds'
172

Thus we may write the differential equations of a geodesic in the form

(73.14)

74. Transformation of the Christoffel symbols. Let.t and zIT be two

sets of curvilinear coordinates. In terms of these coordinates the equations of the geodesics are

(74.1)
and

' IS
d~
ds'

(74.2)

d d'.q
+ F'spqds
~ -~ds
Ip

- 0
-.

But

dz'
ds

dz' dz 'P
dz'PTs'

d'z'
dZ' d'z'P
ds' = dz'P ds'-

d'Z' dz 'P d/ q
dz'P dZ'q dS Ts

Substitution in Equation (74.1) then yields

dZ' d'z'P
dz'P ds'

d'Z'
dz'P dz ,q
dz m dZ) dz 'P dz,q
+ F'",. dZ'P dz ,q dS dS =

o.

We now multiply this equation by dZ"! dZ', obtaining the relation

d'z's
(d Z'S d'z'
dz's dz m dZ") dz'P dz'q
ds' + dZ' dz'P dz,q + Pm. dz' dz'P dz'q dS dS = o.
Comparison of this equation with (74.2) then yields
(74.3)

't _
dz's dz m dz,
F pq - dz' d/P dz,qF m

dz's

+ dz'

d'z"
dz'P dz'q

This is the equation of transformation of the Christoffel symbols of the

second kind. We note that these symbols are not tensors, but would
be if the last term on the right side were missing.
173

75. Absolute differentiation. Let Ap be a covariant vector defined

over a curve C with equations i = l'(u), u being a parameter on C.
Then Ap = Ap(u). The absolute derivative of Ap along C is
q
Mp = dA p _ p A dz .
(75.1)
~u
du
pq
du
S

We shall now prove that

(75.2)

~Ap/~u

is a covariant vector. We have

M'p = dA'p _ F's". A' dz'q

~u
du
"" sTu'

Now
(75.3)

dA'p _ ~ ({}z, A)
du - du {}z'P ,

{}i dA,
= {}z'P du

{}' z' dz'q

{}z,q Tu A,.

+ {}z'P

Also, because of Equation (74.3), the second term on the right side of
Equation (75.2) satisfies the relation

(75.4)
Substitution from Equations (75.3) and (75.4) in (75.2) then yields
M'p = {};(" (dAm _ F' A dZn )
~u
{}z'P du
mn, du
{}zm Mm
= {}z'P

au'

This completes the proof.

The covariant vector A, is said to be propagated parallelly along
Cif
(75.5)
174

~A, = 0
~u

In this case the components Ar satisfy first order differential equations,

and can hence be assigned arbitrarily at anyone point on C. When
the space is three dimensional and the coordinates are rectangular
cartesian, Equations (75.5) reduce to dArldu = 0, so that Ar are
constant along C.
If Ar is a contravariant vector defined on a curve C, its absolute
derivative along C is
(75.6)

3Ar _ dAr
3u - du

+P
m.

Am d::.
du

We could prove that 3Ar/3u is a contravariant vector in a manner

similar to that used above for 3Ar/3u. However, it is more convenient
to proceed as follows. Let Br be any vector propagated parallelly
along C. Then
~ -PB dz'.
du m du

r.

But AmBm is an invariant, and hence so is

Since Bm can be assigned arbitrarily at anyone point, the coefficient

of Bm here, which is 3A m/3u, is then a contravariant vector.
The contravariant vector Ar is said to be propagated parallelly
along C if
r
aA
--0

3u -

Let Am", A~. and Am. be any second order tensors. Their absolute
derivatives along a curve Care
(75.7)
175

( 75.8)
(75.9)

q
+ pq n du - FPnq Am. l'
q
aAmn _ dAmn _ FP A dz _ P A
au - du
mq pn du
nq mp

OA~'n _ dA'!'n
au - ~

Fm AP dz

q
dz
du'
q

dz
du

The patterns exhibited by the suffixes in these equations should be

noted. Just as in the case of the absolute derivatives of the vectors,
each of the above three absolute derivatives has the same tensor
characters as the original tensor. We can prove this in a manner
similar to that used above. For example in the case of aA~n/au, we
form the invariant A':'nXm yn, where Xm and yn are any vectors propagated parallelly along C.
The absolute derivative of any tensor of higher order is defined
similarly, and has the tensor char.acter of the original tensor. For
example, we have
q
q
q
aA~sl = dA: st + F' AI> dz _p AT dz _p AT dz
au
ds
pq
sl dt
sq '1'1 dt
Iq .sp dt
If A is an invariant, we make the definition

aA
dA
au = du-

(75.lO)

I t can be proved in a direct manner that

(75.11)

ag

~
au

= 0'au amn -- 0 ,

agmn
"'u
0

o.

The proofs of these are left as exercises for the reader (Problem 28 at
the end of this chapter).
The rule for the absolute derivative of a product oftwo tensors is the
same as for the ordinary derivative of a product. Thus, for example
(75.12)

!au (ATs B) =
I

aA:s B
au
I

+ AT

's

aB I.
OU

This can be proved in a direct manner which is somewhat lengthy. It

can also be proved concisely by the use of special coordinates called
Riemannian coordinates.
176

76. Covariant derivatives. Let A, be a covariant vector defined over

some region V in space, and let C be a curve in V. If u is a parameter
on C, then
q
oAp = dAp -F' As dz = ({}A p -F' A) dz~.
ou
du
pq
du
\(}zq
pq s du
The expression on the right side is a covariant vector, and dzq/du is
a contravariant vector which may be assigned arbitrarily. Hence the
coefficient of dzq/du is a covariant tensor of the second order. We
denote it by A p1q and call it the covariant derivative of Ap , so we have
{}Ap
s
APlq = {}zq -Fpq As.
(76.1 )
In a similar manner we can arrive at the covariant derivatives of tensors of all orders and types. We have, for example,

Ap"Iq = {}AP
{}zq

(76 2)
(76.3)

A p'qrls --

+ FP,q A' ,

~
AP'qr + FPts At'qr - Ftqs AP'lr - Ftrs APqt
()ZS

These two covariant derivatives have the tensor character indicated

by their suffixes.
Of course, we have from Equations (75.10) and (75.1 L),
( 76.4)

(76.5)

gmnlq

=0,

O~q

=0,

g~~lq

=0.

Because of these relations, the operation of raising and lowering suffixes can be permuted with the operation of taking the covariant
derivative. Thus, for example, we have

grs A ~It = (grs AS) It = A'it.

It can be shown that covariant differentiation has the same product
rule as ordinary differentiation.

77. The curvature tensor.

(77.1)

Now

A"m

= ~:~-FfmAp.
177

Since ATlm is a covariant tensor of the second order, then

(77.2)
By use of Equations (77.1) and (77.2), we can arrive after a straightforward but lengthy calculation at the relation
(77.3)
where

(77.4)

Wrmn =

~m F:" -

;zrF:"

+ P;"Fpm - Ff.,Fpn

From Equation (77.3) it follows by the tests for tensor character that
rmn has the tensor character indicated by its suffixes. It is called the
mixed curvature tensor.
The covariant curvature tensor is

R:

(77.5 )
This tensor plays an important role in mathematical physics. For
any coordinate system in three dimensional space, this tensor vanishes,
since it vanishes for rectangular cartesian coordinates. On the other
hand, if for anyone coordinate system on a surface this tensor vanishes,
then there exists for this surface a curvilinear coordinate system such
that
g" = 1 if r = s
= 0 if r = s.
78. Cartesian tensors. A tensor is said to be cartesian when the transformations involved are from one set of rectangular cartesian coordinates to another. In 47 the special case of such transformations was
considered when the transformation is a rotation about the origin.
In the general case, the transformation is a rotation plus a translation.
If we superimpose a translation on the rotation considered in 47,
the equations of transformation are
(78.1 )
178

where ars are the constants considered in 47 satisfying the orthogonality conditions
(78.2)
Also, as and a's are constants such that a's = - aTS aT' Just as in 47,
we have the relations
(78.3)

dX's
dXT
dx' = aST = dX's'

The Jacobian I of the transformation is

I

\~;,J =

laSTI

Hence by the rule for the multiplication of determinants, we have

(78.4)

soI=l.
Theorem 1. For cartesian tensors there is no distinction between
contravariant and covariant character.
Proof. Because of Equation (78.3) it follows that the laws of transformation of contravariant components and covariant components
are the same. Further
so that the raising or lowering of a suffix does not change the values
of the components.
Because of this theorem, when dealing with cartesian tensors we
do not need both superscripts and subscripts, so subscripts will be used
exclusively.
In 6 we introduced the orthogonal projections of a vector on rectangular cartesian coordinate axes, calling these projections the components of a vector. Throughout the rest of this book we shall refer
to these as the physical components of a vector for rectangular cartesian
coordinates.
Theorem 2, The physical components of a vector for rectangular
cartesian coordinates constitute a cartesian tensor of the first order.

179

Proof. The nine constants ar, in Equations (78.1) are the cosines
of the angles between the axes of two rectangular cartesian coordinate
systems. Hence, if b is a vector with physical components br and b'r
for these two systems of coordinates, then just as in 47 we have

b/

= asr br
= uX
i!:~ br.
r

This is the law of transformation of a cartesian tensor of the first order,

so the proof is complete.
When the coordinates are rectangular cartesian,. the Christoffel
symbols vanish, and so the absolute derivative becomes the ordinary
derivative and the covariant derivative becomes the partial derivative.
For example, 8br/3u becomes dbr/du, and brl' reduces to Jbr/Jx,.
Thus we conclude that the directional derivatives of the physical
components of a vector for rectangular cartesian coordinates constitute
a cartesian tensor of the first order, and the nine partial derivatives
of the physical components of a vector constitute a cartesian tensor
of the second order.

79. Oriented cartesian tensors. Equation (78.4) shows that the Jacobian I of a transformation from one set of rectangular cartesian coordinates to another is equal to-either plus one or minus one. The former
case arises when the transformation is between coordinates whose
axes have the same ori~ntation (both right-handed or both lefthanded), and the latter case arises when the orientations are opposite.
A set of quantities is said to constitute an oriented cartesian tensor if
it is a cartesian tensor when I = 1 and is not a cartesian tensor when
1= -1.
In two-dimensional problems, suffixes have the range 1, 2. For
such problems we introduce a permutation symbol c" defined as follows:
(79.1)

cl l

= c,. = 0,

eu = 1,

Cn

= -1 .

In the three dimensional case, the permutation symbol

nition
180

Crs'

has the defi-

Crsl

(79.2)

= 0 if two suffixes are equal

= 1 if (rst) is an even permutation of (123)

= -1 if (rst) is an ood permutation of (123),

a single permutation of (rst) being an interchange of any two of r, S
and t, and an even or odd permutation meaning an even or odd
number of single permutations. Thus, for example,
C123

C312

= 1,

C321

-1,

CU3

= 0.

In Theorem 4 of the next section we shall see that both of these permutation symbols are oriented cartesian tensors.
We can now express in tensor notation many of the formulas and
equations of the earlier chapters of this book. Thus, the scalar and
vector products of two vectors a and b are respectively
a, b" Crsl as b,
We note that this vector product is an oriented cartesian tensor. The
scalar triple product a (b X c) is
The expressioas \l j, \l. band \l X b become

dj
dX, '

db,
dXr '

Crsl

db ,
dXs'

respectively. The differentiation formulas (48.1 )-( 48.11) may also be

expressed in this notation; when this is done (Problem 36 at the end
of this chapter), some of the formulas become trivial, and the truth of
others follows at once from the identity
(79.3)
C1m Crsl = ams a
nl - ami am,
the proof of which is left to the reader (Problem 34 at the end of this
chapter).

80. Relative tensors. The Jacobian I of a transformation from one

set of curvilinear coordinates i to another set z" is given by the
relation
Z'
1= ddZ's'

181

We now introduce the following definition: a set of quantities A' is a

relative contravariant vector of weight W, if it transforms according
to the equation
{}

"

A" = ~
AS I
{}~

W being an integer. Relative contravariant tensors of higher order

are defined analogously, as are relative covariant tensors, relative
mixed tensors, and relative invariants. For example, A~sl is a relative
mixed tensor of weight W if it has the law of transformation
A "os! =

{}z"

{}Z'

{}zP

{}zm {}z'S {}z'!

Am

IW

O'P

A relative tensor of weight one is called a tensor density.

When it becomes necessary to distinguish between the kind of
tensors considered heretofore and relative tensors we refer to the former
type as absolute tensors. The following properties of relative tensors
are easily established: (i) If a relative tensor vanishes in one coordinate
system, it vanishes in all coordinate systems. (ii) The transformation
of relative tensors is transitive. (iii) The sum of two relative tensors
of the same order and weight is a relative tensor of this same weight.
(iv) Both the inner and outer products of two relative tensors yield
relative tensors whose weights are equal to the sum of the weights
of the two tensors. (v) Contraction does not change the weight of a
relative tensor.
Theorem l. The determinant g = !grs[ is a relative tensor of weight
two.
Proo Because of the rule for the multiplication of determinants,
we have
(80.1)

g, = !g, ,s! =

which completes the proof.

182

{}Z'

{}z

{}z" {}z,;

g,. 1

1;;:,11 ;~I!gvw!

J2g,

Let us now introduce a covariant permutation symbol erst and a

contravariant permutation symbol erst, both defined numerically
just as in Equation (79.2).

Theorem 2. The covariant permutation symbol is a relative covariant tensor of weight -1.
Proof. We must show that
(80.2)

erst =

{}zm

{};C' {}zP

-1

{}Z,r {}Z,s {}Z,t emnp 1

Let us denote the right side of this equation by 1-1 <I> rst. Then
(80.3)
But the expression on the right side here is just the expansion of the
determinant I, so <1>123 = 1. In a similar manner, we see that each
component of <l>rst is equal to a determinant. If the suffixes on any
one component are subjected to a single permutation, two rows or
columns of the corresponding determinant are interchanged, and so
the sign of the component is changed. We thus have the results:
(i) <l>rst = 0 if two suffixes are equal, (ii) <l>rst = 1 if (rst) is an even
permutation of (123), (iii) <l>rst = -lif (rst) is an odd permutation of
(123). Thus we may write
.
and so the proof is complete.

Theorem 3. The contravariant permutation symbol is a relative

contravariant tensor of weight one.
Proof. This proof is quite similar to that of Theorem 2 above,
and is hence omitted.
Theorem 4. The permutation symbols are oriented cartesian tensors.
Proo When the coordinates are rectangular cartesian, we see
from Equation (80.2) that erst is an absolute tensor only when 1 = 1.
Hence it is an oriented- cartesian tensor. The proof is similar for erst.
Because of Theorem 1 above, we may change the weight ofa tensor
by multiplying the tensor by a power of g. Let us assume that g is
183

positive. We can then construct the absolute permutation symbols as

follows:
(80.4 )

'lJTSt

vg CTSt ,

Because of Theorems 1,3 and 4 above, we see that llTSt is an absolute

covariant tensor, while llTSt is an absolute contravariant tensor.
Let A::: denote a general relative tensor of weight W. Then giwA:::
is an absolute tensor. The covariant derivative of A::: is defined to be
(80.5)

- tW ( g-tWA"')
A
"'IT -g
... IT'

We note that A::: IT has the same weight as A:::. The absolute derivatives of relative tensors are defined analogously. It can be proved
that both the covariant and absolute derivatives of relative tensors
obey the same product rule as do ordinary derivatives.

81. Physical components ~f tensors. In Theorem 2 of 78 it was

stated that the physical components b of a vector b for rectangular
cartesian coordinates constitute a cartesian tensor of the first order.
We shall now define the contravariant and covariant component of
the vector b for curvilinear coordinates l. Denoting these components by E and B T , we make the definitions
T

(81.1 )

BT

dz:: b
dX s S,

B _ dXs b
T -

dz::

It is easily proved that BT = gTfi'. The quantities BT and BT describe

the vector b in a certain manner, and have the tensor character indicated by their suffixes.
In 72 of the present chapter we defined abstractly the magnitude
of a contravariant vector. According to this definition, the magnitude
of BT is
B

= VgTS BT B S

Since the right side of this equation is an invariant, and since

the metric tensor of rectangular cartesian coordinates, we have

B = vamnbmb n
184

= vbmb m = b.

aTS

IS

Thus the abstract definition of magnitude of a vector given in 72

agrees with the definition of magnitude given in 1. In a similar
fashion, the abstract definition given in 72 for the angle between
two vectors agrees with our physical notions of angle.
Let us consider a unit vector having components A' fot a curvilinear coordinate system zr. The physical component of B' in the direction of A' is defined to be the invariant

B" = g" H A'.

Ifa is the angle between B' and

J(,

then by 72 we have

cos 6 = g"

B'

A' .

Thus B" = b cos a, as might be expected.

The physical components of a vector B' in the directions of the
parametric lines of the coordinates Z are called the physical components of B' for the curvilinear coordinates z'. Let us denote them by
B(,). When the curvilinear coordinates are orthogonal, then as mentioned in 49, we have
(81.2)

If A(;) is a unit vector in the direction of the parametric line of Zl, then
(81.3)

A(I)

dz1

= ds = h/

By lowering suffixes, we also have

(81.4)

Thus

B(1l = g,fi'A(:) = HAcI), = Blhl

In a similar dashion we get B(.) and B(a), so that the physical compo- .
nents of the vector b relative to the curvilinear coordinate system
z' are
BIh" B'h., B8h.
By lowering suffixes, we may also express these physical components
in the form

185

(81.5)

There is a similar procedure for tensors of higher order. For example, let tTS be a set of nine quantities which are the components of a
cartesian tensor. The stress components of elasticity form such a set.
We define the tensorial components of this tensor for curvilinear coordinates z' in terms of trs by the appropriate laws of tensor transformation. Let us denote these components by the symbols Trs, T:"
T;s and T TS. The physical component of this tensor along the directions of two unit vectors ),: and [.tT is defined to be TT'ATfLs. The
physical components of this tensor for the curvilinear coordinates z'
are defined to be the physical components obtained by taking ),: and
[.t' in the directions of the parametric lines of the coordinates. Let us
denote these nine physical components by T(TS). Then

T(TS) = ymnA(T)mA(S)n,
AIs) being the three unit vectors in the directions of the parametric
lines of the coordinates. When the coordinates z' are orthogonal, then
\~) is as given in Equation (81.3), and there are two similar relations
for A(;) andA(;). We then get for T(TS) the expressions

h:T",

hlh, T 12 ,

hlh3 T 13 ,

h3hl T 31 ,

h3h, T 32 ,

h: T 33

(81.6)

There are other expressions for T(TS) in terms of the covariant components TTS' as well as in terms of mixed components.
Physical components of tensors of higher order can be defined
analogously.

82. Applications. We shall consider now the problem of expressing

the fundamental equations of mathematical physics in terms ofquantities pertaining to curvilinear coordinates.
Dynamics of a particle. Let X be the rectangular cartesian coordinates of a particle at time t. Its velocity and acceleration are then
T

186

(82 . 1)

v,

dx,

= dt '

a,

dv,

= dt .

If f, is the force acting on the particle and m

Newton's second law we have
(82.2)

IS

its mass, then by

rna, =j,.

Let l be curvilinear coordinates of the particle, V', A' and F' being
contravariant component of velocity, acceleration and force for this
coordinate system. We have by definition
(82.3)

dZ'

dZ' dx,

dz'

= dX, v, = ax, dt = dt

We could define A' similarly in terms of a" but it is simpler to write

(82.4)

A' = i3V' = dV'

at
dt

+ F' vm dz n
mn

dt

We could define F' in terms of j" but it is easier to use the relation
(82.5 )

dW

= j,dx, = F,dz' ,

where dW is the work done in an infinitesimal displacement with

components dx, and dl. By substitution in this equation for dx, in
terms of d;:.', we get F, by equating coefficients of dz'.
We now write the following expressions tentatively for the equations of motion in terms of curvilinear coordinates:
(82.6)

rnA' =F'.

To check these we note that they are tensor equations, and are true
when the coordinates are rectangular cartesian since in this case they
reduce to (82.2). Hence they are true for all coordinate systems.
The mathematical theory of elasticity. Using rectangular cartesian
coordinates x" we have the displacement u" the strain components
ers and the stress components Trs. For the determination of these,
we have the equations
(82.7)
(82.8)
187

(82.9)
JT"
Jx-

(82.lO)

+X

r
pddt
-U '
-

together with certain boundary conditions. In the above, k rsmn are

elastic constants, Xr is the external force per unit volume, p is the
density and t is the time. If the body is at rest, the right side of (82.lO)
vanishes. If the body is isotropic, that is, it has no preferred directions
elastically, Equation (82.9) reduces to
(82.11)
where "and f.I. are elastic cOl,1stants, and

8=e =dUr
rr
dX r

(82.12)

The quantities in the above equations are all cartesian tensors.

Let us now introduce curvilinear coordinates l. For this coordinate
system the quantities corresponding to u" ern t", krsmn and Xr are
defined !by use of the laws of tensor transformation, as before. There
is no difficulty if we use the same principal letter to designate these
quantities for both coordinate systems. We now wish to convert
Equations (82.7)-(82.12) to curvilinear coordinate. We write tentatively:

(82.13)

ers

(82.14)

ernlsm +esmlrn

(82.15)
(82.16)
(82.17)

T rs

t(ur!s+uslr)'

ermlsn + esn1rm ,

k,smnemn ,

's

T .. !s

+X =
T

(j2U'

P Jt 2

'

TTs = "8grs +2f.1.e",

(82.18)
These equations are tensor equations, and reduce to Equations
(82.7)-(82.12) when the coordinates are rectangular cartesian coordinates. Hence Equations (82.13)-(82.18) are true for all coordinate
systems, and are the desired equations. (Note that the term JV/dt"
188

in Equation (82.16) is a tensor; see Problem 45 at the end of this

chapter.)
Hydrodynamics. In terms of rectangular cartesian coordinates, the
fundamental equations for a perfect fluid are
.

dp
dt

(82.19)

+ dXd r (pur ) =

0,

dV r + dV r Vs = Xr _! dP ,
dt
dx,
p dXr
where p is the density, t is the time, Vr is the velocity, Xr is the external
force per unit volume, and p is the pressure. We convert these to
curvilinear coordinates in a manner analogous to that used for elasticity, obtaining the equations
(82.20)

~~ + (PVr)lr =

(82.21 )

(82.22)

du + r ,
Jt
VI'U =

xr

0,

I rs dP
-pg
dZ:

In a similar manner, we can obtain in terms of curvilinear coordinates the fundamental equations of other branches of mathematical
physics, such as electricity and magnetism, geometrical optics and
heat conduction.

Problems
1. Prove that

(Ars+Asr)lZ: = 2A rslz'.
2. If A = ArsZrz:, where Ars are constants, prove that

3. Evaluate 8~, 8:~, 8:a:8~.

4. Prove that

8: Art = A,t,

8:8~A~

A~ .

5. By differentiating Equation (62.7) with respect to Zo, prove that

189

d 2Z"
dZ" dZ'v dZ'w d'Z"
d;:! dl = - dZ" d;:! dZ t dZ'v dZ'w

6. Let b, denote the covariant components of a vector for rectangular cartesian coordinates. Find the covariant components of this
vector for cylindrical coordinates r, e, Z in terms of b, and r , e, z.
7. If z' are curvilinear coordinates and cD is an invariant, do the
.
d2i J > .
expressIOns dZ'd;:! constItute a tensorr
'l

B. Prove that the sum of two tensors A~t and B;t is a tensor.
9. Prove that the outer product of two tensors
and B;t is a tensor.
10. If A;t is a tensor, prove that A;t is a tensor.
11. In three-dimensional space, how many different expressions are
represented by A:B;qC;'? When each such expression is written out
explicitly, how many terms does it contain?
12. Prove that the tensor A;' is transitive.
13. Let A~ be a set of nine quantities such that A~X: is an invariant,
where
is an arbitrary mixed tensor of the second order. Prove that
A; is a mixed tensor of the second order.
14. Let A"t be a set of quantities such that ATSt;rt is a covariant
tensor of the second order, where X t is an arbitrary contravariant
vector. Establish the tensor character of A"t.
15. If g" = 0 for = r s, prove that
1
1
3.
1
gIl = _,
g22 = _,

A:

X:

=-,

g
g23 = (1 =g12 = O.
16. Find the components of g" for cylindrical coordinates r, e, z.
17. Find the components of g" for spherical polar coordinates
gn

r,

e, 'P'

g2.

lB. Let z' and z be plane oblique cartesian coordinates whose

axes have an angle ~ between them. For these coordinates find grs
and i'.
19. Prove that g"i' = 3.
20. Prove that

190

21. Using the results of Problem 20, prove that

1 d
_
F~s = Vg dZ' V g.
22. Prove that

+ [rn, m] =

[Tm, n]

t,:., [mn, r]

g"F~.

23. Compute the Christoffel symbols for cylindrical coordinates

6, z.
24. Compute the Christoffel symbols for spherical polar coordinates.
25. By using rectangular cartesian coordinates and Equations
(73.14), show that the geodesics in three-dimensional space are straight
lines.
26. Deduce the differential equations of the geodesics in three dimensional space, in terms of cylindrical coordinates.
27. Deduce the law of transformation of the Christoffel symbols
of the first kind in the form

T,

[ pq, s]'

dZ m dZ' dZ T
dZ'P dZ'q dZ'S [mn, T]

d 2 Zm

dZ n

+ gmn dz'P dZ'q dZ's'

28. Prove that

Sgm. = 0
S
Su
'Su

sm
n

= ,

29. Prove that

S (A B )

SU

.S

SA ~s BAr SBt
= s;;t +
.s &;'.

30. Using the results of Problem 21, prove that

A~I' =

Jg ~, (Vi A').

31. IfJis an invariant, use the results of Problem 30 to prove that

the invariant g'.lrs = \1'1, where v' is the Laplacian operator discussed
in the second last paragraph of 48; hence show that

n'f
v

1 d (. In
Vg
d,t v g

's dJ)
g d,t .
191

32. IfJis an invariant, use the results of Problem 31 to evaluate

terms of cylindrical coordinates r, e, z, comparing the result
with that in Problem 33 of Chapter IV.
33. Prove that the permutation symbols ers for two dimensional
problems satisfies the identity

'\7'1 in

34. Prove Equation (79.3).

35. Express in cartesian tensor form the vector identity
aX (b Xc) = b(a.c)-c(a. b),
and prove it by use of Equation (79.3).
36. Express the eleven equations (48.1)-(48.11) in cartesian tensor
form, and then verify them.
37. If Ars is an absolute tensor, prove that the determinant IArsl is a
relative invariant of weight -2. Establish the tensor characters of IA:sl
and IArsl.
38. Prove that CrstCrst = 6, where Crst and erst are the permutation
symbols.

39. Prove that

~: = 0,

where g is the determinant of the metric

tensor.
40. If Ar is a relative tensor of weight one, prove that
dAr
A: 1r = (Jzr'
41. If Br are the covariant components of a vector for orthogonal
curvilinear coordinates zr whose metric form is given in Equation
(81.2), show that the physical components of this vector for the curvilinear coordinates ;( are as given in (81.5).
42. If T' are the components of a tensor for the coordinates ;( in
Problem 41, find the physical components of T' for these coordinates
in terms of mixed components.
43. Let r, e and <p be the spherical polar coordinates of a particle.
Prove that for these coordinates the contravariant, covariant and
physical components of velocity are respectively

192

dr dfJ drp) (dr dfJ

fJ drp)
(dt' dt' dt' dt'r dt' r sm dt'
(dr, ,dfJ, rsin fJ drp).
2

dt

dt

dt

44. Let r, fJ and z be the cylindrical coordinates of a particle. Find

the physical components of acceleration for these coordinates, in terms
of r, fJ, z and their time derivatives.
45. Let A::: be a general tensor which is a function ofa parameter
t as well as of the coordinates. Prove that, for any transformation
independent of t, the derivatives of A::: with respect to t are tensors
with the same tensor character as A:::.
46. In the theory of electricity and magnetism, Maxwell's equations
for free space are
\J

JE
= k-,--,
dt

"V

JH

-[L- ,

dt

\JH=O,
\JE = 0,
where H is the magnetic intensity, k is the dielectric constant, E is the
electric intensity, t.J. is the permeability and t is the time. Express these
equations in tensor form for general curvilinear coordinates i.

193

A CATALOGUE OF SELECTED DOVER BOOKS

IN ALL FIELDS OF INTEREST

A CATALOGUE OF SELECTED DOVER

BOOKS IN ALL FIELDS OF INTEREST
CELESTIAL OBJECTS FOR COMMON TELESCOPES, T. W. Webb. The
most used book in amateur astronomy: inestimable aid for locating and
identifying nearly 4,000 celestial objects. Edited, updated by Margaret W.
Mayall. 77 illustrations. Total of 645pp. 5% x BJf.!.
20917-2, 2091B-0 Pa., Two-vol. set $9.00
HISTORICAL STUDIES IN THE LANGUAGE OF CHEMISTRY, M. P.
Crosland. The important part language has played in the development of
chemistry from the symbolism of alchemy to the adoption of systematic
nomenclature in 1892. ". . . wholeheartedly recommended,"-Science. 15
63702-6 Pa. $6.00
illustrations. 416pp. of text. 5% x BI4.
BURNHAM'S CELESTIAL HANDBOOK, Robert Burnham, Jr. Thorough,
readable guide to the stars beyond our solar system. Exhaustive treatment,
fully illustrated. Breakdown is alphabetical by constellation: Andromeda
to Cetus in Vol. 1; Chamaeleon to Orion in Vol. 2; and Pavo to Vulpecula
in Vol. 3. Hundreds of illustrations. Total of about 2000pp. 6% x 914.
23567-X, 23568-B, 23673-0 Pa., Three-vol. set $26.85
THEORY OF WING SECfIONS: INCLUDING A SUMMARY OF AIRFOIL DATA, Ira H. Abbott and A. E. von Doenhoff. Concise compilation
of subatomic aerodynamic characteristics of modem NASA wing sections,
plus description of theory. 35Opp. of tables. 693pp. 5% x BJf.!.
60586-8 Pa. $-7.00
DE RE METALLlCA, Georgius Agricola. Translated by Herbert C. Hoover
and Lou H. Hoover. The famous Hoover translation of greatest treatise on
technological chemistry, engineering, geology, mining of early modem
times (1556). All 2B9 original woodcuts. 63Bpp. 6% x 11.
60006-B Clothbd. $17.95
THE ORIGIN OF CONTINENTS AND OCEANS, Alfred Wegener. One
of the most inHuentia~ most controversial books in science, the classic
statement for continental drift. Full 1966 translation of Wegener's final
6170B-4 Pa. '4.50
(1929) version. 64 illustrations. 246pp. 5% x BJf.!.
THE PRINCIPLES OF PSYCHOLOGY, William James. Famous long
course complete, unabridged. Stream of thought, time perception, memory,
experimental methods; great work decades ahead of its time. Still valid,
useful; read in many classes. 94 figures. Total of 1391pp. 5% x BJf.!.
203BI-6, 203B2-4 Pa., Two-vol. set $13.00

CATALOGUE OF DOVER BOOKS

ART FORMS IN NATURE, Ernst Haeckel. Multitude of strangely beautiful natural fonns: Radiolaria, Foraminifera, jellyfishes, fungi, turtles, bats,.
etc. All 100 plates of the 19th-century evolutionist's Kunstformen der
Natur (1904). 1OOpp. 9% x 12114.
22987-4 Pa. $4.50
CHILDREN: A PICTORIAL ARCHIVE FROM NINETEENTH-CENTURY SOURCES, edited by Carol Belanger Grafton. 242 rare, copyrightfree wood engravings for artists and designers. Widest such selection
available. All illustrations in line. 119pp. 8% x 11114.
23694-3 Pa. $3.50
WOMEN: A PICTORIAL ARCHIVE FROM NINETEENTH-CENTURY
SOURCES, edited by Jim Harter. 391 copyright-free wood engravings for
artists and designers selected from rare periodicals. Most extensive such
collection available. All illustrations in line. 128pp. 9 x 12.
23703-6 Pa. $4.50
ARABIC ART IN COLOR, Prisse d'Avennes. From the greatest
mentalists of all time--50 plates in color, rarely seen outside the
East, rich in suggestion and stimulus. Includes 4 plates on covers.
9% x 124.
23658-7 Pa.

omaNear
46pp.
$6.00

AUTHENTIC ALGERIAN CARPET DESIGNS AND MOTIFS, edited by

June Beveridge. Algerian carpets are world famous. Dozens of geometrical
motifs are charted on grids, color-coded, for weavers, needleworkers, craftsmen, designers. 53 illustrations plus 4 in color. 48pp. 8Y4 x 11. (Available
in U.S. only)
23650-1 Pa. $1.75
DICTIONARY OF AMERICAN PORTRAITS, edited by Hayward and
Blanche Cirker. 4000 important Americans, earliest times to 1905, mostly
in clear line. Politicians, writers, soldiers, scientists, inventors, industrialists, Indians, Blacks, women, outlaws, etc. Identilicatory infonnation.
21823-6 Clothbd. $40.00
756pp. 94 x 120/4.
HOW THE OTHER HALF LIVES, Jacob A. Riis. Journalistic record of
filth, degradation, upward drive in New YoQrk immigrant slums, shops,
around 1900. New edition includes 100 original Riis photos, monuments of
22012-5 Pa. $6.00
early photography. 233pp. 10 x 7%.
NEW YORK IN THE THIRTIES, B~renice Abbott. Noted photographer's
fascinating study of city shows new buildings that have become famous
and old sights that have disappeared forever. Insightful commentary. 97
22967-X Pa. $5.00
photographs. 97pp. 11 % x 10.
MEN AT WORK, Lewis W. Hine. Famous photographic studies of construction workers, railroad men, factory workers and coal miners. New
supplement of 18 photos on Empire State building construction. New
introduction by Jonathan L. Doherty. Total of 69 photos. 63pp. 8 x 100/".
23475-4 Pa. $3.00

CATALOGUE OF DOVER BOOKS

DRAWINGS OF WILLIAM BLAKE, William Blake. 92 plates from
Book of Job, Divine Comedy, Paradise Lost, visionary heads, mythological
figures, Laocoon, etc. Selection, introduction, commentary by Sir Geoffrey
Keynes. 178pp. 8% x 11.
22303-5 Pa. $4.00
ENGRAVINGS OF HOGARTH, William Hogarth. 101 of Hogarth's
greatest works: Rake's Progress, Harlot's Progress, Illustrations for Hudibras,
Before and After, Beer Street and Gin Lone, many more. Full commentary.
256pp. 11 x 13%.
22479-1 Pa. $7.95
DAUMIER: 120 GREAT LITHOGRAPHS, Honore Daumier. Wide-ranging
collection of lithographs by the greatest caricaturist of the 19th century.
Concentrates on eternally popular series on lawyers, on married life, on
liberated women, etc. Selection, introduction, and notes on plates by
Charles F. Ramus. Total of 158pp. 9% x 12,.
23512-2 Pa. $5.50
DRAWINGS OF MUCHA, Alphonse Maria Mucha. Work reveals draftsman of highest caliber: studies for famous posters and paintings, renderings for book illustrations and ads, etc. 70 works, 9 in color; including 6
items not drawings. Introduction. List of illustrations. 72pp. 9 3/s x 12%.
(Available in U.S. only)
23672-2 Pa. $4.00
GIOVANNI BATTISTA PIRANESI: DRAWINGS IN THE PIERPONT
MORGAN LIBRARY, Giovanni Battista Piranesi. For first time ever all
of Morgan Library's collection, world's largest. 167 illustrations of rare
Piranesi drawings-archeological, architectural, decorative and visionary.
Essay, detailed list of drawings, chronology, captions. Edited by Felice
23714-1 Pa. $7.50
Stampfle. 144pp. 9% x 124.
NEW YORK ETCHINGS (1905-1949), John Sloan. All of important
American artist's N.Y. life etchings. 67 works include some of his best
art; also lively historical record-Creenwich Village, tenement scenes.
Edited by Sloan's widow. Introduction and captions. 79pp. 8% x 11%.
23651-X Pa. $4.00
CHINESE PAINTING AND CALLIGRAPHY: A PICTORIAL SURVEY,
Wan-go Weng. 69 fine examples from John M. Crawford's matchless private
collection: landscapes, birds, Howers, human figures, etc., plus calligraphy.
Every basic form included: hanging scrolls, handscrolIs, album leaves, fans,
etc. 109 illustrations. Introduction. Captions. 192pp. 8% x 11%.
23707-9 Pa. $7.95
DRAWINGS OF REMBRANDT, edited by Seymour Slive. Updated Lippmann, Hofstede de Groot edition, with definitive scholarly apparatus. All
portraits, biblical sketches, landscapes, nudes, Oriental figures, classical
studies, together with selection of work by followers. 550 illustrations.
Total of 63Opp. 9% x 12,. 21485-0, 21486-9 Pa., Two-vol. set $15.00
THE DISASTERS OF WAR, Francisco Goya. 83 etchings record horrors
of Napoleonic wars in Spain and war in general. Reprint of 1st edition, plus
3 additional plates. Introduction by Philip Hofer. 97pp. 9% x 8,.
21872-4 Pa. $3.75

CATALOGUE OF DOVER BOOKS

HISTORY OF BACTERIOLOGY, William Bulloch. The only comprehensive history of bacteriology from the beginnings through the 19th century. Special emphasis is given to biography-Leeuwenhoek, etc. Brief
accounts of 350 bacteriologists form a separate section. No clearer, fuller
study, suitable to scientists and general readers, has yet been written. 52
illustrations. 448pp. 5 5/8 x 84.
23761-3 Pa. $6.50
THE COMPLETE NONSENSE OF EDWARD LEAR, Edward Lear. All
nonsense limericks, zany alphabets, Owl and Pussycat, songs, nonsense
botany, etc., illustrated by Lear. Total of 321pp. 5% x 8Jh. (Available
in U.S. only)
20167-8 Pa. $3.00
INGENIOUS MATHEMATICAL PROBLEMS AND METHODS, Louis
A. Graham. Sophisticated material from Graham Dial, applied and pure;
stresses solution methods. Logic, number theory, networks, inversions, etc.
20545-2 Pa. $3.50
237pp. 5% x 8Jh.
BEST MATHEMATICAL PUZZLES OF SAM LOYD, edited by Martin
Gardner. Bizarre, original, whimsical puzzles by America's greatest puzzler.
From fabulously rare Cyclopedia, including famous 14-15 puzzles, the
Horse of a Different Color, 115 more. Elementary math. 150 illustrations.
167pp. 5% x 8Jh.
20498-7 Pa. $2.75
THE BASIS OF COMBINATION IN CHESS, J. du Mont. Easy-to-follow,
instructive book on elements of combination play, with chapters on each
piece and every powerful combination team-two knights, bishop and
knight, rook and bishop, etc. 250 diagrams. 218pp. 5% x 8Jh. (Available
in U.S. only)
23644-7 Pa. $3.50
MODERN CHESS STRATEGY, Ludek Pachman. The use of the queen,
the active king, exchanges, pawn play, the center, weak squares, etc.
Section on rook alone worth price of the book. Stress on the moderns.
Often considered the most important book on strategy. 314pp. 5% x 8Jh.
20290-9 Pa. $4.50
LASKER'S MANUAL OF CHESS, Dr. Emanuel Lasker. Great world
champion offers very thorough coverage of all aspects of chess. Combinations, position play, openings, end game, aesthetics of chess, philosophy of
struggle, much more. Filled with analyzed games. 39Opp. 50/8 x 8Jh.
20640-8 Pa. $5.00
500 MASTER GAMES OF CHESS, S. Tartakower, J. du Mont. Vast
collection of great chess games from 179B-1938, with much material nowhere else readily available. Fully annoted, arranged by opening for
easier study. 664pp. 5% x BJh.
23208-5 Pa. $7.50
A GUIDE TO CHESS ENDINGS, Dr. Max Euwe, David Hooper. One
of the finest modern works on chess endings. Thorough analysis of the
most frequently encountered endings by former world champion. 331
23332-4 Pa. $3.50
examples, each with diagram. 248pp. 5% x 8lh.

CATALOGUE OF DOVER BOOKS

HOLLYWOOD GLAMOUR PORTRAITS, edited by John Kobal. 145
photos capture the stars from 1926-49, the high point in portrait photography. Gable, Harlow, Bogart, Bacall, Hedy Lamarr, Marlene Dietrich,
Robert Montgomery, Marlon Brando, Veronica Lake; 94 stars in all. Full
background on photographers, technical aspects, much more. Total of
160pp. 8% x 114.
.
23352-9 Pa. $6.00
THE NEW YORK STAGE: FAMOUS PRODUCTIONS IN PHOTOGRAPHS, edited by Stanley Appelbaum. 148 photographs from Museum
of City of New York show 142 plays, 1883-1939. Peter Pan, The Front
Page, Dead End, Our Town, O'Neill, hundreds of actors and actresses, etc.
23241-7 Pa. $6.00
Full mdexes. 154pp. 9:Jh x 10.
MASTERS OF THE DRAMA, John Gassner. Most comprehensive history
of the drama, every tradition from Greeks to modern Europe and America,
including Orient. Covers 800 dramatists, 2000 plays; biography, plot summaries, criticism, theatre history, etc. 77 illustrations. 890pp. 5:ys x 8:Jh.
20100-7 Clothbd. $10.00
THE GREAT OPERA STARS IN HISTORIC PHOTOGRAPHS, edited by
James Camner. 343 portraits from the 1850s to the 1940s: Tamburini,
Mario, Caliapin, Jeritza, Melchior, Melba, Patti, Pinza, Schipa, Caruso,
Fanar, . Steber, Gobbi, and many more-270 performers in all. Index.
199pp. 8% x 111/4.
23575-0 Pa. $6.50
J. S. BACH, Albert Schweitzer. Great full-length study of Bach, life,
background to music, music, by foremost modern scholar. Ernest Newman
translation. 650 musical examples. Total.of 928pp. 5% x 8:Jh. (Available
in U.S. only)
21631-4, 21632-2 Pa., Two-vol. set $10.00
COMPLETE PIANO SONATAS, Ludwig van Beethoven. All sonatas in
the fine Schenker edition, with fingering, analytical material. One of best
modem editions. Total of 615pp. 9 x 12. (Available in U.S. only)
23134-8, 23135-6 Pa., Two-vol. set $15.00
KEYBOARD MUSIC, J. S. Bach. Bach-Gesellschaft edition. For harpsichord, piano, other keyboard instruments. English Suites, French Suites,
Six Partitas, Goldberg Variations, Two-Part Inventions, Tluee-Part Sinfonias. 312pp. 81(s x 11. (Available in U.S. only)
22360-4 Pa. $6.95
FOUR SYMPHONIES IN FULL SCORE, Franz Schubert. Schubert's four
most popular symphonies: No.4 in C Minor ("Tragic"); No.5 in B-Hat
Major; No.8 in B Minor ("Unfinished"); No.9 in C Major ("Great").
Breitkopf & Hartel edition. Study score. 261pp. 9:Ys x 121(4.
23681-1 Pa. $6.50
THE AUTHENTIC GILBERT & SULLIVAN SONGBOOK, W. S. Gilbert,
A. S. Sullivan. Largest selection available; 92 songs, uncut, original keys,
in piano rendering approved by Sullivan. Favorites and lesser-known fine
numbers. Edited with plot synopses by James Spero. 3 illustrations. 399pp.
9 x 12.
23482-7 Pa. $7.95

CATALOGUE OF DOVER BOOKS

"OSCAR" OF THE WALDORF'S COOKBOOK, Oscar Tschirky. Famous
American chef reveals 3455 recipes that made Waldorf great; cream of
French, German, American cooking, in all categories. Full instructions,
easy home use. 1896 edition. 907pp. 60/8 x 9%.
20790-0 Clothbd. $15.00
COOKING WITH BEER, Carole Fahy. Beer has as superb an effect on
food as wine, and at fraction of cost. Over 250 recipes for appetizers,
soups, main dishes, desserts, breads, etc. Index. 144pp. 5% x B72. (Available in U.S. only)
23661-7 Pa. $2.50
STEWS AND RAGOUTS, Kay Shaw Nelson. This international cookbook
offers wide range of lOB recipes perfect for everyday, special occasions,
meals-in-themselves, main dishes. Economical, nutritious, easy-to-prepare:
goulash, Irish stew, boeuf bourguignon, etc. Index. 134pp. 53;8 x B72.
23662-5 Pa. $2.50
DELICIOUS MAIN COURSE DISHES, Marian Tracy. Main courses are
the most important part of any meal. These 200 nutritious, economical
recipes from around the world make every meal a delight. "I . . . have
found it so useful in my own household,"-N.Y. Times. Index. 219pp.
5% x 872.
23664-1 Pa. $3.00
FIVE ACRES AND INDEPENDENCE, Maurice G. Kains. Great backto-the-land classic explains basics of self-sufficient farming: economics,
plants, crops, animals, orchards, soils, land selection, host of other necessary things. Do not confuse with skimpy faddist literature; Kains was
one of America's greatest agriculturalists. 95 illustrations. 397pp. 53;8 x 872.
20974-1 Pa.$3.95
A PRACTICAL GUIDE FOR THE BEGINNING FARMER, Herbert
Jacobs. Basic, extremely useful first book for anyone thinking about moving
to the country and starting a farm. Simpler than Kains, with greater emphasis on country living in general. 246pp. 5% x 872.
23675-7 Pa. $3.50
A GARDEN OF PLEASANT FLOWERS (PARADISI IN SOLE: PARADISUS TERRESTRIS), John Parkinson. Complete, unabridged reprint of
first (1629) edition of earliest great English book on gardens and gardening. More than 1000 plants & flowers of Elizabethan, Jacobean garden
fully deSCribed, most with woodcut illustrations. Botanically very reliable,
a "speaking garden" of exceeding charm. 812 illustrations. 628pp. B72
x 124.
23392-8 Clothbd. $25.00
ACKERMANN'S COSTUME PLATES, Rudolph Ackermann. Selection of
96 plates from the Repository of Arts, best published source of costume
for English fashion during the early 19th century. 12 plates also in color.
Captions, glossary and introduction by editor Stella Blum. Total of 12Opp.
8% x lllJ4.
23690-0 Pa. $4.50

CATALOGUE OF DOVER BOOKS

SECOND PIATlGORSKY CUP, edited by Isaac Kashdan. One of the
greatest tournament books ever produced in the English language. All 90
games of the 1966 tournament, annotated by players, most annotated by
both players. Features Petrosian, Spassky, Fischer, Larsen, six others.
23572-6 Pa. $3.50
22Bpp. 5% x Blh.
ENCYCLOPEDIA OF CARD TRICKS, revised and edited by Jean Hugard.
How to perform over 600 card tricks, devised by the world's greatest
magicians: impromptus, spelling tricks, key cards, using special packs,
much, much more. Additional chapter on card technique. 66 illustrations.
402pp. 5% x Slh. (Available in U.S. only)
21252-1 Pa. $3.95
MAGIC: STAGE ILLUSIONS, SPECIAL EFFECTS AND TRICK PHOTOGRAPHY, Albert A. Hopkins, Henry R. Evans. One of the great classics;
fullest, most authorative explanation of vanishing lady, levitations, scores
of other great stage effects. Also small magic, automata, stunts. 446 illustrations. 556pp. 5% x Slh.
23344-B Pa. $6.95
THE SECRETS OF HOUDINI, J. C. Cannell. Classic study of Houdini's
incredible magic, exposing closely-kept professional secrets and revealing,
in general terms, the whole art of stage magic. 67 illustrations. 279pp.
5% x Blh.
22913-0 Pa. $3.00
HOFFMANN'S MODERN MAGIC, Professor Hoffmann. One of the best,
and best-known, magicians' manuals of the past century. Hundreds of
tricks from card tricks and simple sleight of hand to elaborate illusiollS
involving construction of complicated machinery .. 332 illustrations. 563pp.
5% x Slh.
23623-4 Pa. $6.00
MADAME PRUNIER'S FISH COOKERY BOOK, Mme. S. B. Prunier.
More than 1000 recipes from world famous Prunier's of Paris and London,
specially adapted here for American kitchen. Grilled tournedos with
anchovy butter, Lobst.er a la Bordelaise, Prunier's prized desserts, more.
Glossary. 34Opp. 5% x Slh. (Available in U.S. only)
22679-4 Pa. $3.00
FRENCH COUNTRY COOKING FOR AMERICANS, Louis Diat. 500
easy-to-make, authentic provincial recipes compiled by former head chef
at New York's Fitz-Carlton Hotel: onion soup, lamb stew, potato pie, more.
309pp. 5% x SJh.
23665-X Pa. $3.95
SAUCES, FRENCH AND FAMOUS, Louis Diat. Complete book gives over
200 specific recipes: bechamel, Bordelaise, hollandaise, Cumberland, apricot, etc. Author was one of this century's finest chefs, originator of
vichyssoise and many other dishes. Index. 156pp. 5% x S.
23663-3 Pa. $2.50
TOLL HOUSE TRIED AND TRUE RECIPES, Ruth Graves Wakefield.
Authentic recipes from the famous Mass. restaurant: popovers, veal and
ham loaf, Toll House baked beans, chocolate cake crumb pudding, much
more. Many helpful hints. Nearly 700 recipes. Index. 376pp. 5% x Slh.
23560-2 Pa. $4.50

CATALOGUE OF DOVER BOOKS

THE AMERICAN SENATOR, Anthony Trollope. Little known, long unavailable Trollope novel on a grand scale. Here are humorous comment
on American vs. English culture, and stunning portrayal of a heroine/
villainess. Superb evocation of Victorian village life. 561pp. 5% x BJh.
23801-6 Pa. $6.00
WAS IT MURDER? James Hilton. The author of Lost Horizon and Goodbye, Mr. Chips wrote one detective novel (under a pen-name) which was
quickly forgotten and virtually lost, even at the height of Hilton's fame.
This edition brings it back-a finely crafted public school puzzle resplendent with Hilton's stylish atmosphere. A thoroughly English thriller by
the creator of Shangri-la. 252pp. 5% x 8. (Available in U.S. only)
23774-5 Pa. $3.00
CENTRAL PARK: A PHOTOGRAPHIC GUIDE, Victor Laredo and
Henry Hope Reed. 121 superb photographs show dramatic views of
Central Park: Bethesda Fountain, Cleopatra's Needle, Sheep Meadow, the
Blockhouse, plus people engaged in many park activities: ice skating, bike
riding, etc. Captions by former Curator of Central Park, Henry Hope
Reed, provide historical view, changes, etc. Also photos of N.Y. landmarks
on park's periphery. 96pp. 8Jh x 11.
23750-8 Pa. $4..50
NANTUCKET IN THE NINETEENTH CENTURY, Clay Lancaster. 180
rare photographs, stereographs, maps, drawings and floor plans recreate
unique American island society. Authentic scenes of shipwreck, lighthouses, streets, homes are arranged in geographic sequence to provide
walking-tour guide to old Nantucket existing today. Introduction, captions.
16Opp. 8% x 11314.
23747-B Pa. $6.95
STONE AND MAN: A PHOTOGRAPHIC EXPLORATION, Andreas
Feininger. 106 photographs by Lite photographer Feininger portray man's
deep passion for stone through the ages. Stonehenge-like megaliths, fortified towns, sculpted marble and crumbling tenements show textures, beau23756-7 Pa. $5.95
ties, fascination. 128pp. 9lf4 x 100/4.
CIRCLES, A MATHEMATICAL VIEW, D. Pedoe. Fundamental aspects
of college geometry, non-Euclidean geometry, and other branches of mathematics: representing circle by point. Poincare model, isoperimetric property, etc. Stimulating recreational reading. 66 figures. 96pp. 5% x 8lf4.
63698-4 Pa. $2.75
THE DISCOVERY OF NEPTUNE, Morton Grosser. Dramatic scientific
history of the investigations leading up to the actual discovery of the
eighth planet of our solar system. Lucid, well-researched book by well23726-5 Pa. $3.00
known historian of science. 172pp. 5% x BJh.
THE DEVIL'S DICTIONARY. Ambrose Bierce. Barbed, bitter, brilliant
witticisms in the form of a dictionary. Best, most ferocious satire America
20487-1 Pa. $2.00
has produced. 145pp. 5% x 8Jh.

CATALOGUE OF DOVER BOOKS

THE ANATOMY OF THE HORSE, George Stubbs. Often considered the
great masterpiece of animal anatomy. Full reproduction of 1766 edition,
plus prospectus; original text and modernized text. 36 plates. Introduction
by Eleanor Garvey. 121pp. 11 x 14%.
23402-9 Pa. $6.00
BRIDGMAN'S LIFE DRAWING, George B. Bridgman. More than 500
illustrative drawings and text teach you to abstract the body into its major
masses, use light and shade, proportion; as well as specific areas of anatomy,
of which Bridgman is master. 192pp. 61,2 x 94. (Available in U.S. only)
22710-3 Pa. $3.00
ART NOUVEAU DESIGNS IN COLOR, Alphonse Mucha, Maurice
Verneuil, Georges Auriol. Full-color reproduction of Combinaisons ornementales (c. 1900) by Art Nouveau masters. Floral, animal, geometric,
interlacings, swashes-borders, frames, spots-all incredibly beautiful. 60
22885-1 Pa. $4.00
plates, hundreds of designs. 9% x 8-1/16.
FULL-COLOR FLORAL DESIGNS IN THE ART NOUVEAU STYLE,
E. A. Seguy. 166 motifs, on 40 plates, from Les lleurs et leurs applications
decoratives (1902): borders, circular deSigns, repeats, allovers, "spots."
All in authentic Art Nouveau colors. 48pp. 9% x 12lj4.
23439-8 Pa. $5.00
A DIDEROT PICTORIAL ENCYCLOPEDIA OF TRADES AND INDUSTRY, edited by Charles C. Gillispie. 485 most interesting plates from
the great French Encyclopedia of the 18th century show hundreds of
working figures, artifacts, process, land and cityscapes; glassmaking, papermaking, metal" extraction, construction, weaving, making furniture, clothing,
wigs, dozens of other activities. Plates fully explained. 92Opp. 9 x 12.
22284-5, 22285-3 Clothbd., Two-vol. set $40.00
HANDBOOK OF EARLY ADVERTISING ART, Clarence P. Hornung.
Largest collection of copyright-free early and antique advertising art ever
compiled. Over 6,000 illustrations, from Franklin's time to the 1890's for
special effects, novelty. Valuable source, almost inexhaustible.
Pictorial Volume. Agriculture, the zodiac, animals, autos, birds, Christmas,
fire engines, flowers, trees, musical instruments, ships, games and sports,
much more. Arranged by subject matter and use. 237 plates. 288pp. 9 x 12.
20122-8 Clothbd. $13.50
Typographical Volume. Roman and Gothic faces ranging from 10 point to
300 point, "Barnum," German and Old English faces, script, logotypes,
scrolls and flourishes, 1115 ornamental initials, 67 complete alphabets,
more. 310 plates. 32Opp. 9 x 12.
20123-6 Clothbd. $15.00
CALLIGRAPHY (CALLIGRAPHIA LATINA), J. G. Schwandner. High
point of 18th-century ornamental calligraphy. Very ornate initials, scrolls,
borders, cherubs, birds, lettered examples. 172pp. 9 x 13.
20475-8 Pa. $6.00

CATALOGUE OF DOVER BOOKS

THE COMPLETE BOOK OF DOLL MAKING AND COLLECTING,
Catherine Christopher. Instructions, patterns for dozens of dolls, from rag
doll on up to elaborate, historically accurate figures. Mould faces, sew
clothing, make doll houses, etc. Also collecting information. Many illustrations. 288pp. 6 x 9.
22066-4 Pa. $4.50
THE DAGUERREOTYPE IN AMERICA, Beaumont Newhall. Wonderful
portraits, 1850's townscapes, landscapes; full text plus 104 photographs.
The basic book. Enlarged 1976 edition. 272pp. 8% x 111/4.
23322-7 Pa. $Q.95
CRAFTSMAN HOMES, Gustav Stickley. 296 architectural drawings, floor
plans, and photographs illustrate 40 different kinds of "Mission-style"
homes from The Craftsman (1901-16), voice of American style of simplicity
and organic harmony. Thorough coverage of Craftsman idea in text and
picture, now collector's item. 224pp. 8% x 11.
23791-5 Pa. $6.00
PEWTER-WORKING: INSTRUCTIONS AND PROJECTS, Burl N. Osborn. & Gordon O. Wilber. Introduction to pewter-working for amateur
craftsman. History and characteristics of pewter; tools, materials, step-bystep instructions. Photos, line drawings, diagrams. Total of 160pp.
7% x 10%.
23786-9 Pa. $3.50
THE GREAT CHICAGO FIRE, edited by David Lowe. 10 dramatic, eyewitness accounts of the 1871 disaster, including one of the aftermath and
rebuilding, plus 70 contemporary photographs and illustrations of the
ruins-courthouse, Palmer House, Great Central Depot, etc. Introduction
by David Lowe. 87pp. 8% x 11.
23771-0 Pa. $4.00
SILHOUETTES: A PICTORIAL ARCHIVE OF VARIED ILLUSTRATIONS, edited by Carol Belanger Grafton. Over 600 silhouettes from the
18th to 20th centuries include profiles and full figures of men and women,
children, birds and animals, groups and scenes, nature, ships, an alphabet.
Dozens of uses for commercial artists and craftspeople. 144pp. 8% x 11 %.
23781-8 Pa. $4.00
ANIMALS: 1,419 COPYRIGHT-FREE ILLUSTRATIONS OF MAMMALS, BIRDS, FISH, INSECTS, ETC., edited by Jim Harter. Clear wood
engravings present, in extremely lifelike poses, over 1,000 species of animals. One of the most extensive copyright-free pictOrial sourcebooks of its
23766-4 Pa. $7.95
kind. Captions. Index. 284pp. 9 x 12.
INDIAN DESIGNS FROM ANCIENT ECUADOR, Frederick W. Shaffer.
282 original designs by pre-Columbian Indians of Ecuador (500-1500 A.D.).
Designs include people, mammals, birds, reptiles, fish, plants, heads, geometric designs. Use as is or alter for advertising, textiles, leathercraft, etc.
23764-8 Pa. $3.50
Introduction. 95pp. 8% x 11%.
SZIGETI ON THE VIOLIN, Joseph Szigeti. Genial, loosely structured
tour by premier violinist, featuring a pleasant mixture of reminiscenes,
insights into great music and musicians, innumerable tips for practicing
23763-X Pa. $3.50
violinists. 385 musical passages. 256pp. 50/8 x 8%.

CATALOGUE OF DOVER BOOKS

THE COMPLETE WOODCUTS OF ALBRECHT DURER,
Dr. W. Kurth. 346 in all: "Old Testament," "St. Jerome,"
"Life of Virgin," Apocalypse," many others. Introduction by
21097-9
Dodgson. 285pp. 81h x 12V4.

edited by
''Passion,''
Campbell
Pa. $7.50

DRAWINGS OF ALBRECHT DURER, edited by Heinrich WolHlin. 81

plates show development from youth to full style. Many favorites; many
new. Introduction by Alfred Werner. 96pp. 8Ve x 11. 22352-3 Pa. $5.00
THE HUMAN FIGURE, Albrecht Durer. Experiments in various techniques-stereometric, progressive proportional, and others. Also life studies
that rank among finest ever done. Complete reprinting of Dresden Sketchbook. 170 plates. 355pp. 8% x llV4.
21042-1 Pa. $7.95
OF THE JUST SHAPING OF LETTERS, Albrecht Diirer. Renaissance
artist explains design of Roman majuscules by geometry, also Gothic lower
and capitals. GroUer Club edition. 43pp. 7% x 103.4 21306-4 Pa. $8.00
TEN BOOKS ON ARCHITECTURE, Vitruvius. The most important'book
ever written on architecture. Early Roman aesthetics, technology, classical
orders, site selection, all other aspects. Stands behind everything since.
20645-9 Pa. $4.00
Morgan translation. 331pp. 5% x 81,2.
THE FOUR BOOKS OF ARCHITECfURE, Andrea Palladio. 16th-century
classic responsible for Palladian movement and style. Covers classical architectural remains, Renaissance revivals, classical orders, etc. 1738 Ware
English edition. Introduction by A. Placzek. 216 plates. llOpp. of text.
91h x 12%.
21308-0 Pa. $8.95
HORIZONS, Norman Bel Geddes. GreaUndustrialist stage deSigner, "father
of streamlining," on application of aesthetics to transportation, amusement,
architecture, etc. 1932 prophetic account; function, theory, specific projects.
222 illustrations. 312pp. 7% x 103.4.
23514-9 Pa. $6.95
FRANK LLOYD WRIGHT'S FALLINGWATER, Donald Hoffmann. Full,
illustrated story of conception and building of Wright's masterwork at
Bear Run, Pa. 100 photographs of site, construction, and details of com23671-4 Pa. $5.50
pleted structure. 112pp. 914 x 10.
THE ELEMENTS OF DRAWING, John Ruskin. Timeless classic by great
Viltorian; starts with basic ideas, works through more difficult. Many
practical exercises. 48 illustrations. Introduction by Lawrence Campbell.
22730-8 Pa. $2.75
228pp. 5% x 81h.
GIST OF ART, John Sloan. Greatest modern American teacher, Art Students League, offers innumerable hints, instructions, guided comments to
help you in painting. Not a formal course. 46 illustrations. Introduction
23435-5 Pa. $4.00
by Helen Sloan. 200pp. 5% x 81h.

CATALOGUE OF DOVER BOOKS

THE DEPRESSION YEARS AS PHOTOGRAPHED BY ARTHUR ROTHSTEIN, Arthur Rothstein. First collection devoted entirely to the work of
outstanding 1930s photographer: famous dust storm photo, ragged children,
unemployed., etc. 120 photographs. Captions. 119pp. 91f4 x 103,4.
23590-4 Pa. $5.00
CAMERA WORK: A PICTORIAL GUIDE, Alfred Stieglitz. All 559 illustrations and plates from the most important periodical in the history of
art photography, Camera Work (1903-17). Presented four to a page, reduced in size but still clear, in strict chronological order, with complete
captions. Three indexes. Glossary. Bibliography. 176pp. 8;'s x 111f4.
23591-2 Pa. $6.95
ALVIN LANGDON COBURN, PHOTOGRAPHER, Alvin L. Coburn. Revealing autobiography by one of greatest photographers of 20th century
gives insider's version of Photo-Secession, plus comments on his own work.
77 photographs by Coburn. Edited by Helmut and Alison Gernsheim.
16Opp. 81fs x 11.
23685-4 Pa. $6.00
NEW YORK IN THE FORTIES, Andreas Feininger. 162 brilliant photographs by the well-known photographer, formerly with Life magazine, show
commuters, shoppers, Times Square at night, Harlem nightclub, Lower
East Side, etc. Introduction and full captions by John von Hartz. 181pp.
91f4 x 10%.
23585-8 Pa. $6.00
GREAT NEWS PHOTOS AND THE STORIES BEHIND THEM, John
Faber. Dramatic volume of 140 great news photos, 1855 through 1976,
and revealing stories behind them, with both historical and technical information. Hindenburg disaster, shooting of Oswald, nomination of Jimmy
23667-6 Pa. $5.00
Carter, etc. 160pp. 81f4 x ll.
THE ART OF THE CINEMATOGRAPHER, I,.eonard Maltin. Survey of
American cinematography history and anecdotal interviews with 5 mastersArthur Miller, Hal Mohr, Hal Rosson, Lucien Ballard, and Conrad Hall.
Very large selection of behind-the-scenes production photos. 105 photographs. Filmographies. Index. Originally Behind the Camera. 144pp.
81f4 x ll.
23686-2 Pa. $5.00
DESIGNS FOR THE THREE-CORNERED HAT (LE TRICORNE),
Pablo Picasso. 32 fabulously rare drawings-including 31 color illustrations
of costumes and accessories-for 1919 production of famous ballet. Edited
by Parmenia Migel, who has written new introduction. 48pp. 90/s x 121f4.
(Available in U.S. only)
23709-5 Pa. $5.00
NOTES OF A FILM DIRECTOR, Sergei Eisenstein. Greatest Russian
filmmaker explains montage, making of Alexander Nevsky, aesthetics; comments on self, associates, great rivals (Chaplin), similar material. 78 illus22392-2 Pa. $4.50
trations. 24Opp. 5% x 8Jh.

CATALOGUE OF DOVER BOOKS

THE EARLY WORK OF AUBREY BEARDSLEY, Aubrey Beardsley. 157
plates, 2 in color: Manon Lescaut, Madame Bovary, Morte Darthur, Salome,
other. Introduction by H. Marillier. 182pp. 81fs x 11. 21816-3 Pa. $4.50
THE LATER WORK OF AUBREY BEARDSLEY, Aubrey Beardsley.
Exotic masterPieces of full maturity: Venus and Tannhauser, Lysistrata,
Rape of the Lock, Volpone, Savoy material, etc. 174 plates, 2 in color.
21817-1 Pa. $4.50
186pp. 8% xlI.
THOMAS NAST'S CHRISTMAS DRAWINGS, Thomas Nast. Almost all
Christmas drawings by creator of image of Santa Claus as we know it,
and one of America's foremost illustrators and political cartoonists. 66
illustrations. 3 illustrations in color on covers. 96pp. 8% x 111f4.
23660-9 Pa. $3.50
THE DORE ILLUSTRATIONS FOR DANTE'S DIVINE COMEDY,
Gustave Dore. All 135 plates from Inferno, Purgatory, Paradise; fantastic
tortures, infernal landscapes, celestial wonders. Each plate with appropriate
(translated) verses. 141pp. 9 x 12.
23231-X Pa. $4;50
OORE'S ILLUSTRATlONS FOR RABELAlS, Gustave Dore. 252 striking
illustrations of Gargantua and Pantagruel books by foremost 19th-century
illustrator. Including 60 plates, 192 delightful smaller illustrations. 153pp.
9 x 12.
2.3656-0 Pa. $5.00
LONDON: A PILGRIMAGE, Gustave Dore, Blanchard Jerrold. Squalor,
riches, misery, beauty of mid-Victorian metropolis; 55 wonderful plates,
125 other illustrations, full social, cultural text by Jerrold. 191pp. of text.
9% x 12lj4.
22306-X Pa. $6.00
THE RIME OF THE ANCIENT MARINER, Gustave Dore, S. T. Coleridge.
Dore's finest work, 34 plates capture moods, subtleties of poem. Full text.
22305-1 Pa. $3.50
Introduction by Millicent Rose. 77pp. 91f4 x 12.
THE DORE BIBLE ILLUSTRATIONS, Gustave Dore. All wonderful, de-.
tailed plates: Adam and Eve, Flood, Babylon, Life of Jesus, etc. Brief
King James text with each plate. Introduction by Millicent Rose. 241
plates. 241pp. 9 x 12.
23004-X Pa. $6.00
THE COMPLETE ENGRAVINGS, ETCHINGS AND DRYPOINTS OF
ALBRECHT DURER. "Knight, Death and Devil"; "Melencolia," and
more-all Diirer's known works in all three media, including 6 works
fonnerly attributed to him. 120 plates. 235pp. 8% x 11Jt&.
22851-7 Pa. $6.50
MAXIMILIAN'S TRIUMPHAL ARCH, Albrecht Diirer and others. Incredible monument of woodcut art: 8 foot high elaborate arch-heraldic
figures, humans, battle scenes, fantastic elements-that you can assemble
yourself. Printed on one side, layout for assembly. 143pp. 11 x 16.
21451-6 Pa. $5.00

CATALOGUE OF DOVER BOOKS

THE PHILOSOPHY OF HISTORY, Georg W. Hegel. Great classic of
Western thought develops concept that history is not chance but a rational
20112-0 Pa. $4.50
process, the evolution of freedom. 457pp. 5% x 8.!.
LANGUAGE, TRUTH AND LOGIC, Alfred J. Ayer. Famous, clear introduction to Vienna, Cambridge schools of Logical Positivism. Role of philosophy, elimination of metaphysics, nature of analysis, etc. 16Opp.
5% x 81f.1. (Available in U.S. only)
20010-8 Pa. $2.00
A PREFACE TO LOGIC, Morris R. Cohen. Great City College teacher
in renowned, easily followed exposition of formal logic, probability, values,
logic and world order and similar topics; no previous background needed.
209pp. 5% x 8.!.
23517-3 Pa. $3.50
REASON AND NATURE, Morris R. Cohen. Brilliant analysis of reason and
its multitudinous ramifications by charismatic teacher. Interdisciplinary, synthesizing work widely praised when it first appeared in 1931. Second
(1953) edition. Indexes. 496pp. 5% x 81f.1.
23633-1 Pa. $6.50
AN ESSAY CONCERNING HUMAN UNDERSTANDING, John Locke.
The only complete edition of enormously important classic, with authoritative editorial material by A. C. Fraser. Total of 1176pp. 5% x 8.!.
20530-4, 20531-2 Pa., Two-vol. set $14.00
HANDBOOK OF MATHEMATICAL FUNCTIONS WITH FORMULAS,
GRAPHS, AND MATHEMATICAL TABLES, edited by Milton Abramowitz and Irene A. Stegun. Vast compendium: 29 sets of tables, some to
as high as 20 places. 1,046pp. 8 x 10.!.
61272-4 Pa. $14.95
MATHEMATICS FOR THE PHYSICAL SCIENCES, Herbert S. Wilf.
Highly acclaimed work offers clear presentations of vector spaces and
matrices, orthogonal functions, roots of polynomial equations, conformal
mapping, calculus of variations, etc. Knowledge of theory of functions of
real and complex variables is assumed. Exercises and solutions. Index.
63635-6 Pa. $5.00
284pp. 50/8 x 84.
THE PRINCIPLE OF RELATIVITY, Albert Einstein et aI. Eleven most
important original papers on special and general theories. Seven by Einstein, two by Lorentz, one each by Minkowski and Weyl. All translated,
60081-5 Pa. $3.00
unabridged. 216pp. 50/8 x 8.!.
THERMODYNAMICS, Enrico Fermi. A classic of modem science. Clear,
organized treatment of systems, first and second laws, entropy, thermodynamic potentials, gaseous reactions, dilute solutions, entropy constant. No
math beyond calculus required. Problems. 16Opp. 5% x 8.!.
60361-X Pa. $3.00
ELEMENTARY MECHANICS OF FLUIDS, Hunter Rouse. Classic undergraduate text widely considered to be far better than many later books.
Ranges from fluid velocity and acceleration to role of compressibility in
fluid motion. Numerous examples, questions, problems. 224 illustrations.
63699-2 Pa. $5.00
376pp. 50/8 x 84.

CATALOGUE OF DOVER BOOKS

THE SENSE OF BEAUTY, George Santayana. Masterfully written discussion of nature of beauty, materials of beauty, fonn, expression; art, literature, social sciences all involved. 16Bpp. 5% x B1,2.
2023B-0 Pa. $2.50
ON THE IMPROVEMENT OF THE UNDERSTANDING, Benedict
Spinoza. Also contains Ethics, Correspondence, all in excellent R. Elwes
translation. Basic works on entry to philosophy, pantheism, exchange of
ideas with great contemporaries. 402pp. 5 3/8 x B1,2.
20250-X Pa. $4.50
THE TRAGIC SENSE OF LIFE, Miguel de Unamuno. Acknowledged
masterpiece of existential literature, one of most important books of 20th
century. Introduction by Madariaga. 367pp. 5% x B1,2.
20257-7 Pa. $4.50
THE GUIDE FOR THE PERPLEXED, Moses Maimonides. Great classic
of medieval Judaism attempts to reconcile revealed religion (Pentateuch,
commentaries) with Aristotelian philosophy. Important historically, still
relevant in problems. Unabridged Friedlander translation. Total of 473pp.
5% x B1,2.
20351-4 Pa. $6.00
THE I CHING (THE BOOK OF CHANGES), translated by James Legge.
Complete translation of basic text plus appendices by Confucius, and
Chinese commentary of most penetrating divination manual ever prepared.
Indispensable to study of early Oriental ciVilizations, to modem inquiring
reader. 44Bpp. 5% x B1,2.
21062-6 Pa. $4.00
THE EGYPTIAN BOOK OF THE DEAD, E. A. Wallis Budge. Complete
reproduction of Ani's papyrus, :finest ever found. Full hieroglyphic text, interlinear transliteration, word for word translation, smooth translation.
Basic work, for Egyptology, for modem study of psychic matters. Total of
533pp. 61,2 x 91/4. (Available in U.S. only)
21B66-X Pa. $5.95
THE GODS OF THE EGYPTIANS, E. A. Wallis Budge. Never excelled
for richness, fullness: all gods, goddesses, demons, mythical :figures of
Ancient Egypt; their legends, rites, incarnations, variations, powers, etc.
Many hieroglyphic texts cited. Over 225 illustrations, plus 6 color plates.
Total of 9BBpp. 6% x 91/4. (Available in U.S. only)
22055-9, 22056-7 Pa., Two-vol. set $12.00
THE ENGLISH AND SCOTTISH POPULAR BALLADS, Francis J.
Child. Monumental, still unsuperseded; all known variants of Child ballads,
commentary on origins, literary references, Continental parallels, other
features. Added: papers by G. L. Kittredge, W. M. Hart. Total of 2761pp.
61,2 x 94.
21409-5, 21410-9, 21411-7, 21412-5, 21413-3 Pa., Five-vol. set $37.50
CORAL GARDENS AND THEIR MAGIC, Bronsilaw Malinowski. Classic
study of the methods of tilling the soil and of agricul~ural rites in the
Trobriand Islands of Melanesia. Author is one of the most important figures
in the field of modern social anthropology. 143 illustrations. Indexes. Total
of 911pp. of text. 5% x B1/4. (Available in U.S. only)
23597-1 Pa. $12.95

CATALOGUE OF DOVER BOOKS

TONE POEMS, SERIES II: TILL EULENSPIEGELS LUSTIGE
STREICHE, ALSO SPRACH ZARATHUSTRA, AND EIN HELDENLEBEN, Richard Strauss. Three important orchestral works, including very
popular Till Eulenspiegel's Marry Pranks, reproduced in full Score from
original editions. Study score. 315pp. 90/8 x 121/4. (Available in U.S. only)
23755-9 Pa. $7.50
TONE POEMS, SERIES I: DON JUAN, TOD UND VERKLARUNG
AND DON QUIXOTE, Richard Strauss. Three of the most often performed and recorded works in entire orchestral repertoire, reproduced in
full score from original editions. Study score. 286pp. 90/8 x 124. ~Avail
able in U.S. only)
23754-0 Pa. $7.50
11 LATE STRING QUARTETS, Franz Joseph Haydn. The form which
Haydn defined and "brought to perfection." (Grove's). 11 string quartets
in complete score, his last and his best. The first in a projected series of
the complete Haydn string quartets. Reliable modern Eulenberg edition,
otherwise difficult to obtain. 320pp. 80/8 x 111/4. (Available in U.S. only)
23753-2 Pa. $6.95
FOURTH, FIFTH AND SIXTH SYMPHONIES IN FULL SCORE, Peter
Ilyitch Tchaikovsky. Complete orchestral scores of Symphony No. 4 in
F Minor, Op. 36; Symphony No. 5 in E Minor, Op. 64; Symphony No. 6
in B Minor, "Pathetique," Op. 74. Bretikopf & Hartel eds. Study score.
480pp. 90/8 x 121/4.
23861-X Pa. $10.95
THE MARRIAGE OF FIGARO: COMPLETE SCORE, Wolfgang A.
Mozart. Finest comic opera ever written. Full score, not to be confused
with piano renderings. Peters edition. Study score. 448pp. 9% x 121/4.
(Available in U.S. only)
23751-6 Pa. $11.95
"IMAGE" ON THE ART AND EVOLUTION OF THE FILM, edited by
Marshall Deutelbaum. Pioneering uook brings together for first time 38
groundbreaking articles on early silent films from Image and 263 illustrations newly shot from rare prints in the collection of the International
Museum of Photography. A landmark work. Index. 256pp. 81/4 x 11.
23777 -X Pa. $8.95
AROUND-THE-WORLD COOKY BOOK, Lois Lintner Sumption and
Marguerite Lintner Ashbrook. 373 cooky and frosting recipes from 28
countries (America, Austria, China, Russia, Italy, etc.) include Viennese
kisses, rice wafers, London strips, lady fingers, hony, sugar spice, maple
cookies, etc. Clear instructions. All tested. 38 drawings. 182pp. 5% x 8.
23802-4 Pa. $2.50
THE ART NOUVEAU STYLE, edited by Roberta Waddell. 579 rare
photographs, not available elsew'lere, of works in jewelry, metalwork, glass,
ceramics, textiles, architecture and furniture by 175 artists-Mucha, Seguy,
Lalique, Tiffany, Gaudin, Hohlwein, Saarinen, and many others. 288pp.
8% x 111/4.
23515-7 Pa. $6.95

CATALOGUE OF DOVER BOOKS

UNCLE SILAS, J. Sheridan LeFanu. Victorian Gothic mystery novel, considered by many best of period, even better than Collins or Dickens.
, Wondedul psychological terror. Introduction by Frederick Shroyer. 436pp.
5% x B3h.
21715-9 Pa. $6.00
JURGEN, James Branch CabeIl. The great erotic fantasy of the 1920's
that delighted thousands, shocked thousands more. FuIl final text, Lane
edition with 13 plates by Frank Pape. 346pp. 5% x B3h.
23507-6 Pa. $4.50
THE CLAVE RINGS, Anthony Trollope. Major novel, chronicling aspects
of British Victorian society, personalities. Reprint of Cornhill serialization,
16 plates by M. Edwards; first reprint of full text. Introduction by Norman
23464-9 Pa. $5.00
Donaldson. 412pp. 5% x B3h.
KEPT IN THE DARK, Anthony Trollope. Unusual short novel about
Victorian morality and abnormal psychology by the great English author.
Probably the first American publication. Frontispiece by Sir John Millais.
23609-9 Pa. $2.50
92pp. 63h x 9If4.
RALPH THE HEIR, Anthony Trollope. Forgotten tale of illegitimacy,
inheritance. Master novel of Trollope's later years. Victorian country estates, clubs, Parliament, fox hunting, world of fully realized characters.
Reprint of 1B71 edition. 12 illustrations by F. A. Faser. 434pp. of text.
5% x Blh.
23642-0 Pa. $5.00
YEKL and THE IMPORTED BRIDEGROOM AND OTHER STORIES
OF THE NEW YORK GHETTO, Abraham Cahan. Film Hester Street
based on Yekl (lB96). Novel, other stories among first about Jewish immigrants of N.Y:s East Side. Highly praised by W. D. Howel1s-Cahan "a
new star of realism." New introduction by Bernard G. Richards. 24Opp.
5% x B3h.
22427-9 Pa. $3.50
THE HIGH PLACE, James Branch Cabell. Great fantasy writer's enchanting comedy of disenchantment set in 1Bth-century France. Considered by some critics to be even better than his famous lurgen. 10 illustrations and numerous, vignettes by noted fantasy artist Frank C. Pape.
32Opp. 5% x B3h.
23670-6 Pa. $4.00
ALICE'S ADVENTURES UNDER GROUND, Lewis Carroll. Facsimile
of ms. Carroll gave Alice Liddell in 1B64. DiHerent in many ways from
final Alice. Handlettered, illustrated by Carroll. Introduction by Martin
214B2-6 Pa. $2.00
Gardner. 12Bpp. 5% x B3h.
FAVORITE ANDREW LANG FAIRY TALE BOOKS IN MANY
COLORS, Andrew Lang. The four Lang favorites in a boxed set-the
complete Red, Green, Yellow and Blue Fairy Books. 164 stories; 439 illustrations by Lance10t Speed, Henry Ford and G. P. Jacomb Hood. Total of
23407-X Boxed set, Pa. $14.95
about 150Opp. 5% x B3h.

CATALOGUE OF DOVER BOOKS

AMERICAN ANTIQUE FURNITURE, Edgar G. Miller, Jr. The basic
coverage of all American furniture before 1840; chapters per item chronologically cover all types of furniture, with more than 2100 photos. Total
21599-7, 21600-4 Pa., Two-vol. set $17.90
of 1106pp. 7% x 10%.
ILLUSTRATED GUIDE TQ SHAKER FURNITURE, Robert Meader.
Director, Shaker Museum, .old Chatham, presents up-to-date coverage of
all furniture and appurtenances, with much on local styles not available
elsewhere. 235 photos. 146pp. 9 x 12.
22819-3 Pa. $5.00
,ORIENTAL RUGS, ANTIQUE AND MODERN, Walter A. Hawley. Persia,
Turkey, Caucasus, Central Asia, China, other traditions. Best general survey of all aspects; styles and periods, manufacture, 'uses, symbols and their
interpretation, and identification. 96 illustrations, 11 in color. 32Opp.
6lfs x 94.
22366-3 Pa. $6.95
CHINESE PQTTERY AND PQRCELAIN, R. L. Hobson. Detailed descriptions and analyses by former Keeper of the Department of .oriental Antiquities and Ethnography at the British Museum. Covers hundreds of
pieces from primitive times to 1915. Still the standard text for most periOds.
136 plates, 40 in full color. Total of 750pp. 5% x 81,2.
23253-0 Pa. $10.00
THE WARES ,OF THE MING DYNASTY, R. L. Hobson. Foremost scholar
examines and illustrates many varieties of Ming (1368-1644). Famous blue
and white, polychrome, lesser-known styles and shapes. 117 illustrations,
9 full color, of outstanding pieces. Total of 263pp. 6lfs x 94. (Available
23652-8 Pa. $6.00
in U.S. only)

Prices subject to change without notice.

Available at your book dealer or write for free catalogue to Dept. GI, Dover
Publications, Inc., 180 Varick St., N.Y., N.Y. 10014. Dover publishes more
than 175 books each year on science, elementary and advanced mathematics,
biology, music, art, literary history, social sciences and other areas.

QA433.H33 c.4
HK POLYTECHNIC LIBRARY

QA

Hay, G. E.

433
.1:.33

c.4
Vector

and

tensor

analaysis.

pm;lffFJ>~ UBRA~y "

Reserve

oJI.eGtiou

(continued from front flap)

ApPLICATIONS OF TENSOR ANALYSIS, A.J. McConnell. (60373-3) $6.00 NUMERI.

CAL SOLUTION OF DIFFERENTIAL EQUATIONS, William E. Milne.
(62437-4) $6.00
TABLES OF INDEFINITE INTEGRALS, G. Petit Bois. (60225-7) $5.00
MATRICES AND TRANSFORMATIONS, Anthony J. Pettofrezzo. (63634-8) $3.00
CoNCEPTS OF PROBABILITY THEORY, Paul E. Pfeiffer. (63677-1) $6.50
PROBABILITY THEORY; A CONCISE CoURSE, Y.A. Rozanov. (63544-9) $3.00
ORDINARY DIFFERENTIAL EQUATIONS AND STABILITY THEORY: AN INTRODUC.
TION, David A. Sanchez. (63828-6) $3.00,
FOUNDATI?NS OF STATISTICS, Leonard J. Savage. (62349-1) $4.50
GEOMETRY OF CoMPLEX NUMBERS, Hans Schwerdtfeger. (63830-8) $4.00
A TABLE OF THE INCOMPLETE ELLIPTIC INTEGRAL OF THE THIRD KIND, R.G. Selfridge and S.E. Maxfield. (60501-9) $10.00
INTEGRAL, MEASURE AND DERIVATIVE, Georgi E. Shilov and B.L. Gurevich.
(63519-8) $4.50
LINEAR ALGEBRA, Georgi E. Shilov. (63518-X) $6.00
PROBLEMS IN PROBABILITY THEORY, MATHEMATICAL STATISTICS AND THEORY 01"
RANDOM FUNCTIONS, A.A. Sveshnikov. (63717-4) $6.50
TENSOR CALCULUS, J.L. Synge and A. Schild. (63612-7) $5.50
SEQUENTIAl. ANALYSIS, Abraham Wald. (61579-0) $4.00
CALCULUS OF VARIATIONS WITH ApPLICATIONS TO PHYSICS At-Io. ENGINEERING,
Robert Weinstock. (63069-2) $5.00
. '
INTRODUCTION TO VECTOR AND TENSOR ANALYSIS, Robert C. Wrede ..
(61879-X) $6.00

Paperbound unless otherwise indicated. Pr~ces subject to change without

notice. Available at your book dealer or wnte for free catalogues to Dept.
Mathematics. Dover Publications, Inc., 180 Varick Street, New York, N.Y.
10014. Please indicate field of interest. Each year Dover publish,es over 200
books on fine art, music, crafts and needlework, antiques, languages, literature,children's books, chess, cookery, nature, anthropology, science,
,
mathematics, and other areas.
'
Manufactured in the U.S.A.