and when ,
according as lies in the 1st, 2nd, 3rd, or 4th quadrants. Here
The principal value of corresponds to , that is, . It is single-valued on , except on the interval where it is discontinuous and two-valued. Unless indicated otherwise, these principal values are assumed throughout the DLMF. (However, if we require a principal value to be single-valued, then we can restrict .)
A function is continuous at a point if . That is, given any positive number , however small, we can find a positive number such that for all in the open disk .
A neighborhood of a point is a disk . An open set in is one in which each point has a neighborhood that is contained in the set.
A point is a limit point (limiting point or accumulation point) of a set of points in (or ) if every neighborhood of contains a point of distinct from . ( may or may not belong to .) As a consequence, every neighborhood of a limit point of contains an infinite number of points of . Also, the union of and its limit points is the closure of .
A domain , say, is an open set in that is connected, that is, any two points can be joined by a polygonal arc (a finite chain of straight-line segments) lying in the set. Any point whose neighborhoods always contain members and nonmembers of is a boundary point of . When its boundary points are added the domain is said to be closed, but unless specified otherwise a domain is assumed to be open.
A region is an open domain together with none, some, or all of its boundary points. Points of a region that are not boundary points are called interior points.
A function is continuous on a region if for each point in and any given number () we can find a neighborhood of such that for all points in the intersection of the neighborhood with .
A function is differentiable at a point if the following limit exists:
Differentiability automatically implies continuity.
If exists at and , then
Conversely, if at a given point the partial derivatives , , , and exist, are continuous, and satisfy (1.9.25), then is differentiable at .
A function is said to be analytic (holomorphic) at if it is differentiable in a neighborhood of .
A function is analytic in a domain if it is analytic at each point of . A function analytic at every point of is said to be entire.
If is analytic in an open domain , then each of its derivatives , , exists and is analytic in .
If is analytic in an open domain , then and are harmonic in , that is,
or in polar form ((1.9.3)) and satisfy
at all points of .
An arc is given by , , where and are continuously differentiable. If and are continuous and and are piecewise continuous, then defines a contour.
A contour is simple if it contains no multiple points, that is, for every pair of distinct values of , . A simple closed contour is a simple contour, except that .
for a contour and continuous, . If , , then the integral is defined analogously to the infinite integrals in §1.4(v). Similarly when or .
Any simple closed contour divides into two open domains that have as common boundary. One of these domains is bounded and is called the interior domain of ; the other is unbounded and is called the exterior domain of .
If is continuous within and on a simple closed contour and analytic within , then
If is continuous within and on a simple closed contour and analytic within , and if is a point within , then
provided that in both cases is described in the positive rotational (anticlockwise) sense.
Any bounded entire function is a constant.
If is a closed contour, and , then
where is an integer called the winding number of with respect to . If is simple and oriented in the positive rotational sense, then is or depending whether is inside or outside .
If is continuous on , then with
is harmonic in . Also with , as within .
The extended complex plane, , consists of the points of the complex plane together with an ideal point called the point at infinity. A system of open disks around infinity is given by
Each is a neighborhood of . Also,
A function is analytic at if is analytic at , and we set .
Suppose is analytic in a domain and are two arcs in passing through . Let be the images of and under the mapping . The angle between and at is the angle between the tangents to the two arcs at , that is, the difference of the signed angles that the tangents make with the positive direction of the real axis. If , then the angle between and equals the angle between and both in magnitude and sense. We then say that the mapping is conformal (angle-preserving) at .
The linear transformation , , has and maps conformally onto .
The transformation (1.9.40) is a one-to-one conformal mapping of onto itself.
The cross ratio of is defined by
or its limiting form, and is invariant under bilinear transformations.
Other names for the bilinear transformation are fractional linear transformation, homographic transformation, and Möbius transformation.
A sequence converges to if . For , the sequence converges iff the sequences and separately converge. A series converges if the sequence converges. The series is divergent if does not converge. The series converges absolutely if converges. A series converges (diverges) absolutely when (), or when (). Absolutely convergent series are also convergent.
Let be a sequence of functions defined on a set . This sequence converges pointwise to a function if
for each . The sequence converges uniformly on , if for every there exists an integer , independent of , such that
for all and .
A series converges uniformly on , if the sequence converges uniformly on .
Suppose is a sequence of real numbers such that converges and for all and all . Then the series converges uniformly on .
A doubly-infinite series converges (uniformly) on iff each of the series and converges (uniformly) on .
For a series there is a number , , such that the series converges for all in and diverges for in . The circle is called the circle of convergence of the series, and is the radius of convergence. Inside the circle the sum of the series is an analytic function . For in (), the convergence is absolute and uniform. Moreover,
For the converse of this result see §1.10(i).
When and both converge
(principal value), where
(principal value), where ,
Lastly, a power series can be differentiated any number of times within its circle of convergence:
A set of complex numbers where and take all positive integer values is called a double sequence. It converges to if for every , there is an integer such that
for all . Suppose converges to and the repeated limits
exist. Then both repeated limits equal .
A double series is the limit of the double sequence
If the limit exists, then the double series is convergent; otherwise it is divergent. The double series is absolutely convergent if it is convergent when is replaced by .
If a double series is absolutely convergent, then it is also convergent and its sum is given by either of the repeated sums
Suppose the series , where is continuous, converges uniformly on every compact set of a domain , that is, every closed and bounded set in . Then
for any finite contour in .
Let be a finite or infinite interval, and be real or complex continuous functions, . Suppose converges uniformly in any compact interval in , and at least one of the following two conditions is satisfied: