CHAPTER

2
ANALYTIC FUNCTIONS

We now consider functions of a complex variable and develop a theory of differentiation

for them. The main goal of the chapter is to introduce analytic functions, which play
a central role in complex analysis.

13. FUNCTIONS AND MAPPINGS

Let S be a set of complex numbers. A function f defined on S is a rule that assigns
to each z in S a complex number w. The number w is called the value of f at z and is
denoted by f (z), so that w = f (z). The set S is called the domain of definition of f .∗
It must be emphasized that both a domain of definition and a rule are needed in
order for a function to be well defined. When the domain of definition is not mentioned,
we agree that the largest possible set is to be taken. Also, it is not always convenient
to use notation that distinguishes between a given function and its values.

EXAMPLE 1. If f is defined on the set z =  0 by means of the equation

w = 1/z, it may be referred to only as the function w = 1/z, or simply the function 1/z.

Suppose that u + iv is the value of a function f at z = x + i y ; that is,

u + iv = f (x + i y).

∗
Although the domain of definition is often a domain as defined in Sec. 12, it need not be.

37
38 ANALYTIC FUNCTIONS CHAP. 2

Each of the real numbers u and v depends on the real variables x and y, and it follows
that f (z) can be expressed in terms of a pair of real-valued functions of the real
variables x and y:
(1) f (z) = u(x, y) + iv(x, y).

EXAMPLE 2. If f (z) = z 2 , then

f (x + i y) = (x + i y)2 = x 2 − y 2 + i2x y.
Hence
u(x, y) = x 2 − y 2 and v(x, y) = 2x y.

If the function v in equation (1) always has value zero, then the value of f is
always real. Thus f is a real-valued function of a complex variable.

EXAMPLE 3. A real-valued function that is used to illustrate some important

concepts later in this chapter is
f (z) = |z|2 = x 2 + y 2 + i0.

If n is a positive integer and if a0 , a1 , a2 , . . . , an are complex constants, where

an = 0, the function
P(z) = a0 + a1 z + a2 z 2 + · · · + an z n
is a polynomial of degree n. Note that the sum here has a finite number of terms and that
the domain of definition is the entire z plane. Quotients P(z)/Q(z) of polynomials are
called rational functions and are defined at each point z where Q(z) = 0. Polynomials
and rational functions constitute elementary, but important, classes of functions of a
complex variable.
If the polar coordinates r and θ are used instead of x and y, then
u + iv = f (r eiθ )
where w = u + iv and z = r eiθ . In that case, we may write
(2) f (z) = u(r, θ) + iv(r, θ).

EXAMPLE 4. Consider the function w = z 2 when z = r eiθ . Here

w = (r eiθ )2 = r 2 ei2θ = r 2 cos 2θ + ir 2 sin 2θ.
Hence
u(r, θ) = r 2 cos 2θ and v(r, θ) = r 2 sin 2θ.

A generalization of the concept of function is a rule that assigns more than one
value to a point z in the domain of definition. These multiple-valued functions occur
SEC. 14 THE MAPPING w = z 2 43

The transformation w = z 2 also maps the upper half plane r ≥ 0, 0 ≤ θ ≤ π onto

the entire w plane. However, in this case, the transformation is not one to one since
both the positive and negative real axes in the z plane are mapped onto the positive
real axis in the w plane.
When n is a positive integer greater than 2, various mapping properties of the
transformation w = z n , or w = r n einθ , are similar to those of w = z 2 . Such a
transformation maps the entire z plane onto the entire w plane, where each nonzero
point in the w plane is the image of n distinct points in the z plane. The circle r = r0
is mapped onto the circle ρ = r0n ; and the sector r ≤ r0 , 0 ≤ θ ≤ 2π/n is mapped
onto the disk ρ ≤ r0n , but not in a one to one manner.

Other, but somewhat more involved, mappings by w = z 2 appear in Example 1,

Sec. 107, and Exercises 1 through 4 Sec. 108.

EXERCISES
1. For each of the functions below, describe the domain of definition that is understood:
 
1 1
(a) f (z) = 2 ; (b) f (z) = Arg ;
z +1 z
z 1
(c) f (z) = ; (d) f (z) = .
z+z 1 − |z|2
Ans. (a) z = ±i; (b) Re z = 0.
2. In each case, write the function f (z) in the form f (z) = u(x, y) + iv(x, y):
z̄ 2
(a) f (z) = z 3 + z + 1; (b) f (z) = (z = 0).
z
Suggestion: In part (b), start by multiplying the numerator and denominator by z̄.

Ans. (a) f (z) = (x 3 − 3x y 2 + x + 1) + i(3x 2 y − y 3 + y);

x 3 − 3x y 2 y 3 − 3x 2 y
(b) f (z) = + i .
x 2 + y2 x 2 + y2

3. Suppose that f (z) = x 2 − y 2 − 2y +i(2x −2x y), where z = x +i y. Use the expressions
(see Sec. 6)
z+z z−z
x= and y=
2 2i
to write f (z) in terms of z, and simplify the result.
Ans. f (z) = z 2 + 2i z.
4. Write the function
1
f (z) = z + (z = 0)
z
44 ANALYTIC FUNCTIONS CHAP. 2

in the form f (z) = u(r, θ) + iv(r, θ).

   
1 1
Ans. f (z) = r + cos θ + i r − sin θ.
r r
5. By referring to the discussion in Sec. 14 related to Fig. 19 there, find a domain in the
z plane whose image under the transformation w = z 2 is the square domain in the w
plane bounded by the lines u = 1, u = 2, v = 1, and v = 2. (See Fig. 2, Appendix 2.)
6. Find and sketch, showing corresponding orientations, the images of the hyperbolas
x 2 − y 2 = c1 (c1 < 0) and 2x y = c2 (c2 < 0)
under the transformation w = z 2 .
7. Use rays indicated by dashed half lines in Fig. 21 to show that the transformation w = z 2
maps the first quadrant onto the upper half plane, as shown in Fig. 21.
8. Sketch the region onto which the sector r ≤ 1, 0 ≤ θ ≤ π/4 is mapped by the transfor-
mation (a) w = z 2 ; (b) w = z 3 ; (c) w = z 4 .
9. One interpretation of a function w = f (z) = u(x, y)+iv(x, y) is that of a vector field in
the domain of definition of f . The function assigns a vector w, with components u(x, y)
and v(x, y), to each point z at which it is defined. Indicate graphically the vector fields
represented by
z
(a) w = i z; (b) w = .
|z|

15. LIMITS
Let a function f be defined at all points z in some deleted neighborhood of a point z 0 .
The statement that f (z) has a limit w0 as z approaches z 0 , or that
(1) lim f (z) = w0 ,
z→z 0

means that the point w = f (z) can be made arbitrarily close to w0 if we choose the
point z close enough to z 0 but distinct from it. We now express the definition of limit
in a precise and usable form.
Statement (1) means that for each positive number ε, there is a positive number
δ such that
(2) | f (z) − w0 | < ε whenever 0 < |z − z 0 | < δ.
Geometrically, this definition says that for each ε neighborhood |w − w0 | < ε of
w0 , there is a deleted δ neighborhood 0 < |z − z 0 | < δ of z 0 such that every point
z in it has an image w lying in the ε neighborhood (Fig. 22). Note that even though
all points in the deleted neighborhood 0 < |z − z 0 | < δ are to be considered, their
images need not fill up the entire neighborhood |w − w0 | < ε. If f has the constant
value w0 , for instance, the image of z is always the center of that neighborhood. Note,
too, that once a δ has been found, it can be replaced by any smaller positive number,
such as δ/2.
SEC. 15 LIMITS 45

y v

w
w0
z0 z
O x O u FIGURE 22

The following theorem on uniqueness of limits is central to much of this chapter,

especially the material in Sec. 21.

Theorem. When a limit of a function f (z) exists at a point z 0 , it is unique.

To prove this, we suppose that
lim f (z) = w0 and lim f (z) = w1 .
z→z 0 z→z 0

Then, for each positive number ε, there are positive numbers δ0 and δ1 such that
| f (z) − w0 | < ε whenever 0 < |z − z 0 | < δ0
and
| f (z) − w1 | < ε whenever 0 < |z − z 0 | < δ1 .
Since
w1 − w0 = [ f (z) − w0 ] + [w1 − f (z)],
the triangle inequality tells us that
|w1 − w0 | ≤ [ f (z) − w0 ] + [w1 − f (z)] = | f (z) − w0 | + | f (z) − w1 |.
So if 0 < |z − z 0 | < δ where δ is any positive number smaller than δ0 and δ1 , we find that
|w1 − w0 | < ε + ε < 2ε.
But |w1 − w0 | is a nonnegative constant, and ε can be chosen arbitrarily small. Hence
w1 − w0 = 0, or w1 = w0 .

Definition (2) requires that f be defined at all points in some deleted neighborhood
of z 0 . Such a deleted neighborhood, of course, always exists when z 0 is an interior
point of a region on which f is defined. We can extend the definition of limit to the case
in which z 0 is a boundary point of the region by agreeing that the first of inequalities
(2) need be satisfied by only those points z that lie in both the region and the deleted
neighborhood.

EXAMPLE 1. Let us show that if f (z) = i z̄/2 in the open disk |z| < 1, then
i
(3) lim f (z) = ,
z→1 2
46 ANALYTIC FUNCTIONS CHAP. 2

the point 1 being on the boundary of the domain of definition of f . Observe that when
z is in the disk |z| < 1,
i iz i |z − 1|
f (z) −
= − = .
2 2 2 2
Hence, for any such z and each positive number ε (see Fig. 23),
i
f (z) −< ε whenever 0 < |z − 1| < 2ε.
2
Thus condition (2) is satisfied by points in the region |z| < 1 when δ is equal to 2ε or
any smaller positive number.
y v

–i f(z)
z =2 2

O 1 x O u

FIGURE 23

If limit (1) exists, the symbol z → z 0 implies that z is allowed to approach z 0

in an arbitrary manner, not just from some particular direction. The next example
emphasizes this.

EXAMPLE 2. If
z
(4) f (z) = ,
z
the limit
(5) lim f (z)
z→0
does not exist. For, if it did exist, it could be found by letting the point z = (x, y)
approach the origin in any manner. But when z = (x, 0) is a nonzero point on the real
axis (Fig. 24),
x + i0
f (z) = = 1;
x − i0
y
z = (0, y)

(0, 0) z = (x, 0) x
FIGURE 24
SEC. 16 THEOREMS ON LIMITS 47

and when z = (0, y) is a nonzero point on the imaginary axis,

0 + iy
f (z) = = −1.
0 − iy
Thus, by letting z approach the origin along the real axis, we would find that the desired
limit is 1. An approach along the imaginary axis would, on the other hand, yield the
limit −1. Since a limit is unique, we must conclude that limit (5) does not exist.

While definition (2) provides a means of testing whether a given point w0 is a

limit, it does not directly provide a method for determining that limit. Theorems on
limits, presented in the next section, will enable us to actually find many limits.

16. THEOREMS ON LIMITS

We can expedite our treatment of limits by establishing a connection between limits
of functions of a complex variable and limits of real-valued functions of two real
variables. Since limits of the latter type are studied in calculus, we may use their
definition and properties freely.

Theorem 1. Suppose that

f (z) = u(x, y) + iv(x, y) (z = x + i y)
and
z 0 = x0 + i y0 , w0 = u 0 + iv0 .
If
(1) lim u(x, y) = u 0 and lim v(x, y) = v0 ,
(x,y)→(x0 ,y0 ) (x,y)→(x0 ,y0 )

Then
(2) lim f (z) = w0 ;
z→z 0

and, conversely, if statement (2) is true, then so is statement (1).

To prove the theorem, we first assume that limits (1) hold and obtain limit (2).
Limits (1) tell us that for each positive number ε, there exist positive numbers δ1 and
δ2 such that
ε 
(3) |u − u 0 | < whenever 0 < (x − x0 )2 + (y − y0 )2 < δ1
2
and
ε 
(4) |v − v0 | < whenever 0 < (x − x0 )2 + (y − y0 )2 < δ2 .
2
48 ANALYTIC FUNCTIONS CHAP. 2

Let δ be any positive number smaller than δ1 and δ2 . Since

|(u + iv) − (u 0 + iv0 )| = |(u − u 0 ) + i(v − v0 )| ≤ |u − u 0 | + |v − v0 |
and

(x − x0 )2 + (y − y0 )2 = |(x − x0 ) + i(y − y0 )| = |(x + i y) − (x0 + i y0 )|,
it follows from statements (3) and (4) that
ε ε
|(u + iv) − (u 0 + iv0 )| < + =ε
2 2
whenever
0 < |(x + i y) − (x0 + i y0 )| < δ.
That is, limit (2) holds.
Let us now start with the assumption that limit (2) holds. With that assump-
tion, we know that for each positive number ε, there is a positive number δ such
that
(5) |(u + iv) − (u 0 + iv0 )| < ε
whenever
(6) 0 < |(x + i y) − (x0 + i y0 )| < δ.
But
|u − u 0 | ≤ |(u − u 0 ) + i(v − v0 )| = |(u + iv) − (u 0 + iv0 )|,
|v − v0 | ≤ |(u − u 0 ) + i(v − v0 )| = |(u + iv) − (u 0 + iv0 )|,
and

|(x + i y) − (x0 + i y0 )| = |(x − x0 ) + i(y − y0 )| = (x − x0 )2 + (y − y0 )2 .
Hence it follows from inequalities (5) and (6) that
|u − u 0 | < ε and |v − v0 | < ε
whenever

0< (x − x0 )2 + (y − y0 )2 < δ.
This establishes limits (1), and the proof of the theorem is complete.

Theorem 2. Suppose that

(7) lim f (z) = w0 and lim F(z) = W0 .
z→z 0 z→z 0
SEC. 16 THEOREMS ON LIMITS 49

Then
(8) lim [ f (z) + F(z)] = w0 + W0 ,
z→z 0

(9) lim [ f (z)F(z)] = w0 W0 ;

z→z 0

and, if W0 = 0 ,
f (z) w0
(10) lim = .
z→z 0 F(z) W0

This important theorem can be proved directly by using the definition of the limit
of a function of a complex variable. But, with the aid of Theorem 1, it follows almost
immediately from theorems on limits of real-valued functions of two real variables.
To verify property (9), for example, we write
f (z) = u(x, y) + iv(x, y), F(z) = U (x, y) + i V (x, y),
z 0 = x0 + i y0 , w0 = u 0 + iv0 , W0 = U0 + i V0 .
Then, according to hypotheses (7) and Theorem 1, the limits as (x, y) approaches
(x0 , y0 ) of the functions u, v, U , and V exist and have the values u 0 , v0 , U0 , and V0 ,
respectively. So the real and imaginary components of the product
f (z)F(z) = (uU − vV ) + i(vU + uV )
have the limits u 0 U0 − v0 V0 and v0 U0 + u 0 V0 , respectively, as (x, y) approaches
(x0 , y0 ). Hence, by Theorem 1 again, f (z)F(z) has the limit
(u 0 U0 − v0 V0 ) + i(v0 U0 + u 0 V0 )
as z approaches z 0 ; and this is equal to w0 W0 . Property (9) is thus established. Corre-
sponding verifications of properties (8) and (10) can be given.
It is easy to see from definition (2), Sec. 15, of limit that
lim c = c and lim z = z 0 ,
z→z 0 z→z 0

where z 0 and c are any complex numbers; and, by property (9) and mathematical
induction, it follows that
lim z n = z 0n (n = 1, 2, . . .).
z→z 0

So, in view of properties (8) and (9), the limit of a polynomial

P(z) = a0 + a1 z + a2 z 2 + · · · + an z n
as z approaches a point z 0 is the value of the polynomial at that point:
(11) lim P(z) = P(z 0 ).
z→z 0
52 ANALYTIC FUNCTIONS CHAP. 2

Furthermore,
2z 3 − 1 (1/z 2 ) + 1 z + z3
lim =∞ since lim = lim = 0.
z→∞ z 2 + 1 z→0 (2/z 3 ) − 1 z→0 2 − z 3

18. CONTINUITY
A function f is continuous at a point z 0 if all three of the following conditions are
satisfied:
(1) lim f (z) exists,
z→z 0
(2) f (z 0 ) exists,
(3) lim f (z) = f (z 0 ).
z→z 0

Observe that statement (3) actually contains statements (1) and (2), since the existence
of the quantity on each side of the equation in that statement is needed. Statement (3)
says, of course, that for each positive number ε, there is a positive number δ such that
(4) | f (z) − f (z 0 )| < ε whenever |z − z 0 | < δ.
A function of a complex variable is said to be continuous in a region R if it is
continuous at each point in R.
If two functions are continuous at a point, their sum and product are also contin-
uous at that point; their quotient is continuous at any such point if the denominator
is not zero there. These observations are direct consequences of Theorem 2, Sec. 16.
Note, too, that a polynomial is continuous in the entire plane because of limit (11) in
Sec. 16.
We turn now to two expected properties of continuous functions whose verifica-
tions are not so immediate. Our proofs depend on definition (4) of continuity, and we
present the results as theorems.

Theorem 1. A composition of continuous functions is itself continuous.

A precise statement of this theorem is contained in the proof to follow. We let

w = f (z) be a function that is defined for all z in a neighborhood |z − z 0 | < δ of a
point z 0 , and we let W = g(w) be a function whose domain of definition contains the
image (Sec. 13) of that neighborhood under f . The composition W = g[ f (z)] is, then,
defined for all z in the neighborhood |z − z 0 | < δ. Suppose now that f is continuous at
z 0 and that g is continuous at the point f (z 0 ) in the w plane. In view of the continuity
of g at f (z 0 ), there is, for each positive number ε, a positive number γ such that
|g[ f (z)] − g[ f (z 0 )]| < ε whenever | f (z) − f (z 0 )| < γ .
(See Fig. 26.) But the continuity of f at z 0 ensures that the neighborhood |z − z 0 | < δ
can be made small enough that the second of these inequalities holds. The continuity
of the composition g[ f (z)] is, therefore, established.
SEC. 18 CONTINUITY 53

y v V

g[ f(z)]
z z0
f(z0) g[ f(z0)]
O x O u O U
f(z)

FIGURE 26

Theorem 2. If a function f (z) is continuous and nonzero at a point z 0 , then

f (z) = 0 throughout some neighborhood of that point.

Assuming that f (z) is, in fact, continuous and nonzero at z 0 , we can prove Theo-
rem 2 by assigning the positive value | f (z 0 )|/2 to the number ε in statement (4). This
tells us that there is a positive number δ such that
| f (z 0 )|
| f (z) − f (z 0 )| < whenever |z − z 0 | < δ.
2
So if there is a point z in the neighborhood |z − z 0 | < δ at which f (z) = 0, we have
the contradiction

| f (z 0 )|
| f (z 0 )| < ;
2
and the theorem is proved.
The continuity of a function
(5) f (z) = u(x, y) + iv(x, y)
is closely related to the continuity of its component functions u(x, y) and v(x, y), as
the following theorem indicates.

Theorem 3. If the component functions u and v in expression (5) are continuous

at a point z 0 = (x0 , y0 ), then so is f. Conversely, if f is continuous at z 0 , the same is
true of u and v at that point.

The proof follows immediately from Theorem 1 in Sec. 16, regarding the con-
nection between limits of f and limits of u and v.
The next theorem is extremely important and will be used often in later chapters,
especially in applications. Before stating the theorem, whose proof is based on Theo-
rem 3, we recall from Sec. 12 that a region R is closed if it contains all of its boundary
points and that it is bounded if it lies inside some circle centered at the origin.
54 ANALYTIC FUNCTIONS CHAP. 2

Theorem 4. If a function f is continuous throughout a region R that is both

closed and bounded, there exists a nonnegative real number M such that
(6) | f (z)| ≤ M for all points z in R,
where equality holds for at least one such z.

To prove this, we assume that the function f in equation (5) is continuous and
note how it follows that the function

[u(x, y)]2 + [v(x, y)]2
is continuous throughout R and thus reaches a maximum value M somewhere in R.∗
Inequality (6) thus holds, and we say that f is bounded on R.

EXERCISES
1. Use definition (2), Sec. 15, of limit to prove that
z2
(a) lim Re z = Re z 0 ; (b) lim z = z 0 ; (c) lim = 0.
z→z 0 z→z 0 z→0 z
2. Let a, b, and c denote complex constants. Then use definition (2), Sec. 15, of limit to
show that
(a) lim (az + b) = az 0 + b; (b) lim (z 2 + c) = z 02 + c;
z→z 0 z→z 0
(c) lim [x + i(2x + y)] = 1 + i (z = x + i y).
z→1−i
3. Let n be a positive integer and let P(z) and Q(z) be polynomials, where Q(z 0 ) = 0. Use
Theorem 2 in Sec. 16, as well as limits appearing in that section, to find
1 i z3 − 1 P(z)
(a) lim n (z 0 = 0); (b) lim ; (c) lim .
z→z 0 z z→i z + i z→z 0 Q(z)

Ans. (a) 1/z 0n ; (b) 0; (c) P(z 0 )/Q(z 0 ).

4. Use mathematical induction and property (9), Sec. 16, of limits to show that
lim z n = z 0n
z→z 0

when n is a positive integer (n = 1, 2, . . .).

5. Show that the function
 
z 2
f (z) =
z
has the value 1 at all nonzero points on the real and imaginary axes, where z = (x, 0)
and z = (0, y), respectively, but that it has the value −1 at all nonzero points on the
line y = x, where z = (x, x). Thus show that the limit of f (z) as z tends to 0 does

∗
See, for instance, A. E. Taylor and W. R. Mann, “Advanced Calculus,” 3d ed., pp. 125–126 and p. 529,
1983.