SUMMARY OF IMPORTANT THEOREMS AND FORMULAE
Chain Rule for Functions of Several Variables:
If

with corresponding fomulae for uxy and uyy.
If F(x, y) = 0,

If x = f(x, y), u = y(x, y),


2. CONVERGENCE OF DOUBLE SEQUENCES
Convergence Test for Double Sequences:
The sequence anm converges, or, in symbols,

3. UNIFORM CONVERGENCE AND INTERCHANGE OF INFINITE OPERATIONS
If a series of positive continuous functions converges to a continuous limit function in a closed region, it converges uniformly to that limit.
Interchange of Differentiation and Integration:
Differentiation of an integral with respect to a parameter

provided that f(x, y) and fx(x, y) are continuous in the interval under consideration
Interchange of Differentiation and Integration in Improper Integrals:

provided that fx(x, y) is continuous in the interval under consideration and the integrals

converge uniformly in that interval.
Interchange of Two Integrations:
If f(x, y) is continuous and a, b, a, b are constants,

The order of integration may also be reversed when the limits are not constants, provided that both integrations are performed over entire the region concerned and corresponding new limits arc introduced
Interchange of Two Integrations in Improper Integrals.

provided that the integral
converges uniformly in
the interval a £ x £
b.
| 4.5.4 | 8.5.6 | I, 6.9.5 | ![]() |
| 4.4.3 | 8.5.1 | I, 4.8.5 | I, A8.2 | I, 9.5.1 | ![]() |

If f(x) is sectionally smooth and
converges and if f(x+ 0)+ f(x - 0) =
2f(x), then

If x > 0, the gamma function G(x) is defined by the equation

It satisfies the functional equation
![]()
hence, if x is a positive integer n,
![]()
For all values of x other than 0, -1, -2, ··· , it may- he expressed by the formulae

where
is Euler's constant. More over, for every integer m
³ 2,

Again,
, Extension
Theorem.
Hence, in particular,

The beta function B(x, y) is defined as follows for positive values of x and y:

The beta and gamma functions are connected by the relation

For any complex z,
![]()
where C denotes a path which surrounds the positive real axis and approaches it asymptotically on either side.
Mean Value Theorem for Functions of Two Variables:

Taylor's Theorem for Funtions of Two Variables:

where the remainder Rn (in the symbolical notation is given by

If, as n increases, this remainder tends to zero, we have the infinite Taylor series

Mean Value Theorems for Multiple lntegrals :

where DR is the area of R and m a value intermediate between the maximum and minimum of f(x, y) in R.
Similarly, if p(x, y) ³ 0,

Notation: A vector v in three dimensions has the components v1, v2, v3.
Length of a Vector:
![]()
Addition of Vectors:
![]()
is the vector with the components
![]()

where d is the angle between u and v.
![]()
is the vector with the components


If the coordinate axes are rotated, the vector components are transformed in the same way as x, y, z, the components of the position vector.
By the derivative of the function f(x, y) in the direction of the unit vector n with the components cos a, sin a we mean the limit

Hence

In particular,

and hence, in general,

In the same way, in three dimensions, the derivative in the direction of the vector n with components cos a, cos b, cos g is

The Differential Operations: With every scalar function f(x1, x2, x3) is associated a vector grad f with the components fx , fy, fz. The derivative of f in the direction of the unit vector n is n gradf.
There is associated with every vector field u(x1, x2, x3) a vector curl u with the components

and a scalar function

Using the symbolic
vector
with the components
we have
![]()
Moreover,

Definition of multiple integral:
Transformation of a Multiple Integral: If the oriented region R of the xy-plane is mapped onto a correspondingly oriented region R' of the uv-plane by a reversible (1.1) transformation the Jacobian

of which does not vanish anywhere, then

An analogous formula holds for any number of dimensions.
In particular, transformations to polar co-ordinates
![]()
or
![]()


Reduction of a Multiple Integral to Ordinary Integrals:
Let a £ y £ b in R and for every y a = a(y) £ x £ b(y) = b; then

8. INTEGRAL THEOREMS OF GAUSS, GREEN AND STOKES
Definition of curvilinear (line) integral.
1. Two Dimensions:
If the region R is simply connected, the line integral
is independent of the path C joining two points in R if, and only if, the condition of integrability holds at every point of R. In this case, if the initial point is fixed, the integral is a function U(x, h) of the end-point, such that the vector A with components a, b satisfies the relation

Gauss' Theorem:
Let R be a simply-connected region and C its boundary. Then

or, in vector notation,

where n is the unit vector in the direction of the outward normal, An the normal component of the vector A with components f, g and ds the arc element of the boundary curve.

In vector notation, the first form of the theorem is

where
![]()
and
denotes differentiation in the direction of the
outward normal.
2. Three Dimensions:
The necessary and sufficient condition that the line integral

shall be independent of the path C joining two points in a simply-connected region R is
![]()
or, in full,
![]()
Surface Integral: This is given by

or

if x = x(u, v), y = y(u, v), z = z(u, v) and the oriented region B in the xy-plane corre»ponds to the surface S.
Gauss' Theorem:
Let n be the unit vector in the direction
of the outward normal and An the normal
component of the vector A with components
a, b, c; moreover, let
denote differentiation
in the direction of the outward normal. Then

or, in vector notation,

the integrals on the right hand side being taken over the closed surface S bounding the region R.

where
and S have the same meaning as before and
![]()
Stokes' Theorem: Let the oriented surface S be bounded by the correspondingly oriented curve C. Then

In vector notation: Let At be the tangential component of the vector A = (f, y, c) in the direction in which the curve C is described, (curl A)n the component of curl A in the direction of the outward normal and ds the arc element on C, measured in the direction in which the curve is described: Then

The following rules hold only for maxima and minima inside region under consideration.
Free Maxima and Minima of a function of Two Variables:
The necessary conditions for an extreme value of the function u = f(x, y) are
![]()
If these conditions are satisfied and if
![]()
there is» an extreme value at the point in question. It is a maximum or a minimum according to whether fxx(and hence also fyy) is negative or positive. If
![]()
the point is a saddle point.
Maxima and Minima subject to Subsidiary Conditions (Method of undetermined Multipliers):
If in the function u = f(x1, ··· , xn) the n variables are linked by the m subsidiary conditions (m < n)
![]()
we introduce m multipliers l1, ··· , lm and form the function
![]()
Then, the m conditions and the n additional equations

yield (m + n) necessary conditions for the extreme points.
In what follows (x, h) or (x, h, z) are current co-ordinates.
1. Plane Curves.
Equation of the curve:
![]()
Equation of the tangent at the point (x, y) (Volume I, 5.1.3, 3.2.1):

Equation of the normal at the point (x, y) (Volume I, 5.1.3; 3.2.1):

Curvature (Volume I, 5.2.6, 3.2.3):

Radius of curvature (Volume I, 5.2.6; 3.2.3):

Evolute (locus of centres of curvature) (Volume I, 5.2.6, A5.1:

Involute (Volume I, 5.2.6):
![]()
where a is an arbitrary constant and s the arc length measured from a given point (s being the parameter).
Point of inflection (Volume I, 3.5.1, 5.1.5; 3.2.3):
The necessary condition for a point of inflection is

Angle between two curves (Volume I, 5.1.3; 3.2.3):

In particular, the curves are orthogonal if
![]()
the curves touch if
![]()
Two curves y = f(x), y = g(x have contact of order n at a point x, if

2. Curves in Space:
Equation of the curve:
![]()
Direction cosines of the tangent:


where ds is the arc element.
3. Surfaces:
Equation of the surface:

Equation of the tangent plane:

Direction cosines of the normal as well

where
![]()
![]()
in particular, the condition that the surface are orthogonal is
![]()
In order to obtain the envelope of the family of plane curves
![]()
or of the family of surfaces
![]()
we calculate the discriminant by eliminating c from the equations
![]()
The discrimmant contains the envelope and also the geometrical locus of the singular points.
If the family of curves is given by the parametric equations x = f (t,c), y = y(t, c), the discriminant is obtained by eliminating c and t from the equations

The envelope of a two-parameter family of surfaces
![]()
is contained in the equation obtained by eliminating the two parameters c1, c2 from the equations
![]()
Let a plane curve be given by the equations

The length of arc is

The length of arc of the three-dimensional curve
![]()

Area of Plane Surface:. The area bounded by the curve
![]()
and two radius vectors q 0, q1, where r, ql are polar co-ordinates, is given by

The area enclosed by the curve
![]()
the two co-ordinates x = x0, x = x1 and the x-axis is

Let R be a positively-oriented plane surface and C its boundary (orientation and sign of an area ). Then the area of the surface is

Area of Curved Surface Let the equation of the surface be

In the case (c), let E, F, G be the so-called fundamental quantities of the surface, i.e., let

![]()
The arc length of the curve
![]()
drawn on the surface is then

The area of the curved surface lying vertically above the region R in the xy-plane is

the last integral being taken over the region B of the uv-plane which corresponds to the region R .
The area of the surface of revolution
![]()
produced when the curve
![]()
is rotated about the z-axis, is

where is the arc of the meridian curve z = f(x). Volume I, 5.2.8.
The surface wn of the unit sphere in n dimensions
![]()
is given by

The volume bounded below by the region R and above by the surface S with the equation
![]()
is given by

If the surface S is closed and forms the entire boundary of the region V, the volume of this region is given by

In polar co-ordinates, the same volume is given by

where R is the region of rqf -space corresponding to the region V.
The volume of the surface of revolution
![]()
which is produced, when the curve
![]()
is rotated about the z-axis, is (also Volume I, 5.2.8)

The volume vn of the unit sphere in n dimensions
![]()

The volume swept out by a moving plane area P of area A is

where dn/dt is the component of the velocity of the mean centre of P perpendicular to the plane of P.
The necessary and sufficient condition for the integral

to be stationary is Euler's equation

or
![]()
If F involves several functions u1(x), u2(x), ··· , un(x) and their derivatives, then a necessary and sufficient condition that the integral

shall he stationary is that u1(x), u2(x), ··· , un(x) satisfy the system of Euler's equations

If F depends on x, u(x), u'(x), u"(x), Euler's equation is

If
is to be made stationary
subject to the subsidiary condition G(x, y, z)
= 0, then a necessary
condition is

where l is Lagrange's multiplier.
The necessary and sufficient condition for
![]()
to be analytic in a regionR is that in R the Cauchy-Riemann differential equations
![]()
hold.
Cauchy's theorem: If f(t) is analytic in a simply-connected region R, then

if C is a closed curve inside R.
Cauchy's formula: Under the same condition as Cauchy's theorem, the formula

holds if z is a point in side C.
If f(z) is analytic inside and on the boundary of a circle |z - z0| £ R, it can be expanded in a power series

which converges inside the circle. Here
