3 Contour integrals and Cauchy’s Theorem

3.1 Line integrals of complex functions

Our goal here will be to discuss integration of complex functions f (z) =
u + iv, with particular regard to analytic functions. Of course, one way to
think of integration is as antidifferentiation. But there is also the definite
integral. For a function f (x) of a real variable x, we have the integral
� b
f (x) dx. In case f (x) = u(x) + iv(x) is a complex-valued function of
a
a real variable x, the definite integral is the complex number obtained by
� b
integrating the real and imaginary parts of f (x) separately, i.e. f (x) dx =
� b � b a

u(x) dx + i v(x) dx. For vector fields F = (P, Q) in the plane we have
a a�
the line integral P dx+Q dy, where C is an oriented curve. In case P and
C
Q are complex-valued, in which case we call P dx + Q dy a complex 1-form,
we again define the line integral by integrating the real and imaginary parts
separately. Next we recall the basics of line integrals in the plane:

1. The vector field F = (P, Q) is a gradient vector field ∇g, which we

can
� write in terms of 1-forms as P dx + Q dy = dg, if and only if
C P dx+Q
� dy only depends on the endpoints of C, equivalently if and
only if C P dx+Q dy = 0 for every closed curve C. If P dx+Q dy = dg,
and C has endpoints z0 and z1 , then we have the formula
� �
P dx + Q dy = dg = g(z1 ) − g(z0 ).
C C

2. If D is a plane region with oriented boundary ∂D = C, then

� �� � �
∂Q ∂P
P dx + Q dy = − dxdy.
C D ∂x ∂y

3. If D is a simply connected plane region, then F = (P, Q) is a gradient

vector field ∇g if and only if F satisfies the mixed partials condition
∂Q ∂P
= .
∂x ∂y

(Recall that a region D is simply connected if every simple closed curve in

D is the boundary of a region contained in D. Thus a disk {z ∈ C : |z| < 1}

1
is simply connected, whereas a “ring” such as {z ∈ C : 1 < |z| < 2} is not.)
In case P dx + Q dy is a complex 1-form, all of the above still makes sense,
and in particular Green’s theorem is still true.
We will be interested in the following integrals. Let dz = dx + idy, a
complex 1-form (with P = 1 and Q = i), and let f (z) = u + iv. The
expression
f (z) dz = (u + iv)(dx + idy) = (u + iv) dx + (iu − v) dy
= (udx − vdy) + i(vdx + udy)
is
� also a complex 1-form, of a very special type. Then we can define
f (z) dz for any reasonable closed oriented curve C. If C is a parametrized
C
curve given by r(t), a ≤ t ≤ b, then we can view r� (t) as a complex-valued
curve, and then
� � b
f (z) dz = f (r(t)) · r� (t) dt,
C a
where the indicated multiplication is multiplication of complex numbers
(and not the dot product). Another notation which is frequently used is
the following. We denote a parametrized curve in the complex plane by z(t),
a ≤ t ≤ b, and its derivative by z � (t). Then
� � b
f (z) dz = f (z(t))z � (t) dt.
C a

For example, let C be the curve parametrized by r(t) = t + 2t2 i, 0 ≤ t ≤ 1,

and let f (z) = z 2 . Then
� � 1 � 1
z 2 dz = (t + 2t2 i)2 (1 + 4ti) dt = (t2 − 4t4 + 4t3 i)(1 + 4ti) dt
C 0 0
� 1
= [(t2 − 4t4 − 16t4 ) + i(4t3 + 4t3 − 16t5 )] dt
0
= t3 /3 − 4t5 + i(2t4 − 8t6 /3)]10 = −11/3 + (−2/3)i.
For another example, let let C be the unit circle, which can be efficiently
parametrized as r(t) = eit = cos t + i sin t, 0 ≤ t ≤ 2π, and let f (z) = z̄.
Then
r� (t) = − sin t + i cos t = i(cos t + i sin t) = ieit .
d
Note that this is what we would get by the usual calculation of eit . Then
dt
� � 2π � 2π � 2π
−it
z̄ dz = e · ie dt =
it it
e · ie dt =
it
i dt = 2πi.
C 0 0 0

2
 One final point in this section: let f (z) = u + iv be any complex valued
function. Then we can compute ∇f , or equivalently df . This computation
is important, among other reasons, because of the chain rule: if r(t) =
(x(t), y(t)) is a parametrized curve in the plane, then

d ∂f dx ∂f dy
f (r(t)) = ∇f · r� (t) = + .
dt ∂x dt ∂y dt
d
(Here · means the dot product.) We can think of obtaining f (r(t)) roughly
dt
∂f ∂f
by taking the formal definition df = dx + dy and dividing both sides
∂x ∂y
by dt.
Of course we expect that df should have a particularly nice form if f (z)
is analytic. In fact, for a general function f (z) = u + iv, we have
� � � �
∂u ∂v ∂u ∂v
df = +i dx + +i dy
∂x ∂x ∂y ∂y

and thus, if f (z) is analytic,

� � � �
∂u ∂v ∂v ∂u
df = +i dx + − +i dy
∂x ∂x ∂x ∂x
� � � � � �
∂u ∂v ∂u ∂v ∂u ∂v
= +i dx + +i idy = +i (dx + idy) = f � (z) dz.
∂x ∂x ∂x ∂x ∂x ∂x

Hence: if f (z) is analytic, then

df = f � (z) dz

and thus, if z(t) = (x(t), y(t)) is a parametrized curve, then

d
f (z(t)) = f � (z(t))z � (t)
dt

This is sometimes called the chain rule for analytic functions. For example,
if α = a + bi is a complex number, then applying the chain rule to the
analytic function f (z) = ez and z(t) = αt = at + (bt)i, we see that

d αt
e = αeαt .
dt

3
3.2 Cauchy’s theorem
Suppose now that C is a simple closed curve which is the boundary
� ∂D of a
region in C. We want to apply Green’s theorem to the integral f (z) dz.
C
Working this out, since

f (z) dz = (u + iv)(dx + idy) = (u dx − v dy) + i(v dx + u dy),

we see that
� �� � � �� � �
∂v ∂u ∂u ∂v
f (z) dz = − − dA + i − dA.
C D ∂x ∂y D ∂x ∂y

Thus, the integrand is always zero if and only if the following equations hold:
∂v ∂u ∂u ∂v
=− ; = .
∂x ∂y ∂x ∂y
Of course, these are just the Cauchy-Riemann equations! This gives:
Theorem (Cauchy’s integral theorem): Let C be a simple closed curve
which is the boundary ∂D of a region in C. Let f (z) be analytic in D. Then
�
f (z) dz = 0.
C

Actually, there is a stronger result, which we shall prove in the next

section:
Theorem (Cauchy’s integral theorem 2): Let D be a simply connected
region in C and let C be a closed curve (not necessarily simple) contained
in D. Let f (z) be analytic in D. Then
�
f (z) dz = 0.
C

Example: let D = C and let f (z) be the function z 2 + z + 1. Let C be

the unit circle. Then as before we use the parametrization of the unit circle
given by r(t) = eit , 0 ≤ t ≤ 2π, and r� (t) = ieit . Thus
� � 2π � 2π
f (z) dz = (e2it + eit + 1)ieit dt = i (e3it + e2it + eit ) dt.
C 0 0

4
It is easy� to check directly that this integral is 0, for example because terms
2π
such as 0 cos 3t dt (or the same integral with cos 3t replaced by sin 3t or
cos 2t, etc.) are all zero.
On the other hand, again with C the unit circle,
� � 2π � 2π
1
dz = e−it ieit dt = i dt = 2πi �= 0.
C z 0 0

The difference is that 1/z is analytic in the region C−{0} = {z ∈ C : z �= 0},

but this region is not simply connected. (Why not?) �
Actually, the converse to Cauchy’s theorem is also true: if f (z) dz = 0
C
for every closed curve in a region D (simply connected or not), then f (z) is
analytic in D. We will see this later.

3.3 Antiderivatives
If D is a simply connected region, C is a curve contained
� in D, P , Q are de-
∂Q ∂P
fined in D and = , then the line integral P dx+Q dy only depends
∂x ∂y C �
on the endpoints of C. However, if P dx + Q dy = dF , then P dx + Q dy
C
only depends on the endpoints of C whether or not D is simply connected.
We see what this condition means in terms of complex function theory: Let
f (z) = u + iv and suppose that f (z) dz = dF , where we write F in terms of
its real and imaginary parts as F = U + iV . This says that
� � � �
∂U ∂U ∂V ∂V
(u dx − v dy) + i(v dx + u dy) = dx + dy + i dx + dy .
∂x ∂y ∂x ∂y

Equating terms, this says that

∂U ∂V
u= =
∂x ∂y
∂U ∂V
v=− = .
∂y ∂x
In particular, we see that F satisfies the Cauchy-Riemann equations, and
its complex derivative is
∂U ∂V
F � (z) = +i = u + iv = f (z).
∂x ∂x

5
We say that F (z) is a complex antiderivative for f (z), i.e. F � (z) = f (z). In
this case
∂u ∂2U ∂2V ∂v
= 2
= = ;
∂x ∂x ∂x∂y ∂y
∂u ∂2U ∂2V ∂v
= 2
= − =− .
∂y ∂y ∂x∂y ∂x
It follows that, if f (z) has a complex antiderivative, then f (z) satisfies the
Cauchy-Riemann equations: f (z) is necessarily analytic.
Thus we see:
Theorem: If the 1-form f (z) dz is of the form dF , or equivalently the
vector field (u + iv, −v + iu) is a gradient vector field ∇(U + iV ), then both
F (z) and f (z) are analytic, and F (z) is a complex antiderivative for f (z):
F � (z) = f (z). Conversely, if F (z) is a complex antiderivative for f (z), then
F (z) and f (z) are analytic and f (z) dz = dF .
The theorem tells us a little more: Suppose that F (z) is a complex
antiderivative for f (z), i.e. F � (z) = f (z). If C has endpoints z0 and z1 , and
is oriented so that z0 is the starting point and z1 the endpoint, then we have
the formula � �
f (z) dz = dF = F (z1 ) − F (z0 ).
C C
For example, we have seen that, if C�is the curve parametrized by r(t) =
t+2t2 i, 0 ≤ t ≤ 1 and f (z) = z 2 , then z 2 dz = −11/3+(−2/3)i. But z 3 /3
C
is clearly an antiderivative for z 2 , and C has starting point 0 and endpoint
1 + 2i. Hence
�
z 2 dz = (1 + 2i)3 /3 − 0 = (1 + 6i − 12 − 8i)/3 = (−11 − 2i)/3,
C

which agrees with the previous calculation.

When does an analytic function have a complex antiderivative?
� From
vector calculus, we know that f (z) dz = dF �if and only if C f (z) dz only
depends on the endpoints �of C, if and only if C f (z) dz = 0 for every closed
curve C. In particular, if C f (z) dz = 0 for every closed curve C then f (z)
is analytic (converse to Cauchy’s theorem).
If f (z) is analytic in a simply connected region D, then the fact that
f (z) dz = P dx + Q dy satisfies ∂Q/∂x = ∂P/∂y (here P and Q are com-
plex valued) says that (P, Q) is a gradient vector field, or equivalently that
f (z) dz = dF , in other words that f (z) has an antiderivative. Hence:

6
Theorem: Let D be a simply connected region and let f (z) be an analytic
function in D. Then there exists a complex antiderivative F (z) for f (z).
Fixing
� a base point p0 ∈ D, a complex antiderivative F (z) for f (z) is given
by f (z) dz, where f (z) is any curve in D joining p0 to z.
C
As a consequence, we see that, if D is �
simply connected, f (z) is analytic
in D and C is a closed curve in D, then f (z) dz = 0 (Cauchy’s integral
C
formula 2), since f (z) dz = dF , where F is a complex antiderivative for f (z),
and hence � �
f (z) dz = dF = 0,
C C
by the Fundamental Theorem for line integrals. �
1
From this point of view, we can see why dz = 2πi �= 0, where C
C z
is the unit circle. The antiderivative of 1/z is log z, and so the expected
answer (viewing the unit circle as starting at 1 = e0 and ending at e2πi = 1
is log 1 − log 1. But log is not a single-valued function, and in fact as z = eit
turns along the unit circle, the value of log changes by 2πi. So the correct
answer is really log 1 − log 1, viewed as log e2πi − log e0 = 2πi − 0 = 2πi.
Of course, 1/z is analytic except at the origin, but {z ∈ C : z �= 0} is not
simply connected, and so 1/z need not have an antiderivative.
The real point, however, in the above example is something special about
log z, or 1/z, but not the fact that 1/z fails to be defined at the origin. We
could have looked at other negative powers of z, say z n where n is a negative
integer less than −1, or in fact any integer �= −1. In this case, z n has an
antiderivative
� z n+1 /(n + 1), and so by the fundamental theorem for line
integrals z n dz = 0 for every closed curve C. To see this directly for the
C
case n = −2 and the unit circle C,
� � 2π � 2π
z −2 dz = e−2it ieit dt = i e−it dt = 0.
C 0 0

This calculation can be done somewhat differently as follows. Let r(t) = eαt ,
where α is a nonzero complex number. Then, by the chain rule for analytic
functions, an antiderivative for the complex curve r(t) is checked to be
�
1
s(t) = eαt dt = eαt .
α

7
Hence,
� b
1 � αb �
eαt dt = e − eαa .
a α
�
In general, we have seen that z n dz = 0 for every integer n �= −1, where
C
C is a closed curve. To verify this for the case of the unit circle, we have
� � 2π � 2π � �2π
i
z dz =
n
e ie dt = i
nit it
e(n+1)it dt = e(n+1)it
C 0 0 i(n + 1) 0
i � � 1
= e2(n+1)πi − e0 = (1 − 1) = 0.
i(n + 1) n+1

Finally, returning to 1/z, a calculation shows that

� � � �
1 x dx y dy −y dx x dy
dz = + + i + .
z x2 + y 2 x2 + y 2 x2 + y 2 x2 + y 2

The real part is the gradient of the function 12 ln(x2 + y 2 ) = d ln r. But the
imaginary part corresponds to the vector field
� �
−y x
F= , ,
x2 + y 2 x2 + y 2

which is a standard example of a vector field F for which Green’s theorem

fails, because F is undefined at the origin. In fact, in terms of 1-forms,
−y dx x dy
+ = d arg z = dθ.
x2 + y 2 x2 + y 2
In the next section, we will see how to systematically use the fact that the
integral of 1/z dz around a closed curve enclosing the origin to get a formula
for the value of an analytic function in terms of an integral.

3.4 Cauchy’s integral formula

Let C be a simple closed curve in C. Then C = ∂R for some region R (in
other words, C is simply connected). If z0 is a point which does not lie on
C, we say that C encloses z0 if z0 ∈ R, and that C does not enclose z0 if
/ R. For example, if C is the unit circle, then C is the boundary of the
z0 ∈
unit disk B = {z : |z| < 1}. Thus C encloses a point z0 if z0 lies inside the
unit disk (|z0 | < 1), and C does not enclose z0 if z0 lies outside the unit disk
(|z0 | > 1).

8
Theorem (Cauchy’s integral formula): Let D be a simply connected
region in C and let C be a simple closed curve contained in D. Let f (z) be
analytic in D. Suppose that z0 is a point enclosed by C. Then
�
1 f (z)
f (z0 ) = dz.
2πi C z − z0

For example, if C is a circle of radius 5 about 0, then

� 2
ez
dz = 2πie4 .
C z−2
� 2
ez
But if C is instead the unit circle, then dz = 0, as follows from
C z−2
Cauchy’s integral theorem.
Before we discuss the proof of Cauchy’s integral formula, let us look at
the special case where f (z) is the constant function 1, C is the unit circle,
and z0 = 0. The theorem says in this case that
�
1 1
1 = f (0) = dz,
2πi C z

as we have seen. In fact, the theorem is true for a circle of any radius: if
Cr is a circle of radius r centered at 0, then Cr can be parametrized by reit ,
0 ≤ t ≤ 2π. Then
� � 2π � 2π
1 1
dz = ire dt = i
it
dt = 2πi,
Cr z 0 reit 0
�
1
independent of r. The fact that dz is independent of r also follows
Cr z
from Green’s theorem.
The general case is obtained from this special case as follows. Let C =
∂R, with R ⊆ D since D is simply connected. We know that C encloses z0 ,
which says that z0 ∈ R. Let Cr be a circle of radius r with center z0 . If
r is small enough, Cr will be contained in R, as will the ball Br of radius
r with center z0 . Let Rr be the region obtained by deleting Br from R.
Then ∂Rr = C − Cr , where this is to be understood as saying that the
boundary of Rr has two pieces: one is C with the usual orientation coming
from the fact that C is the boundary of R, and the other is Cr with the
clockwise orientation, which we record by putting a minus sign in front

9
of Cr . Now z0 does not lie in Rr , so we can apply Green’s theorem to the
function f (z)/(z − z0 ) which is analytic in D except at z0 and hence in Rr :
�
f (z)
dz = 0.
∂Rr z − z0

But we have seen that ∂Rr = C − Cr , so this says that

� �
f (z) f (z)
dz − dz = 0,
C z − z 0 Cr z − z0

or in other words that

� �
f (z) f (z)
dz = dz.
C z − z0 Cr z − z0

Now suppose that r is small, so � that f (z) is approximately equal to f (z0 )

f (z)
on Cr . Then the second integral dz is approximately equal to
C r z − z0
� �
f (z0 ) 1
dz = f (z0 ) dz,
C r z − z0 Cr z − z0

where Cr is a circle of radius r centered at z0 . Thus we can parametrize Cr

by z0 + reit , 0 ≤ t ≤ 2π, and
� � 2π � 2π
1 1
dz = ire dt = i
it
dt = 2πi,
C r z − z0 0 reit 0

as before. Thus �
1
f (z0 ) dz = 2πif (z0 ),
Cr z − z0
� �
f (z) f (z)
and so dz = dz is approximately equal to 2πif (z0 ). In
C z − z0 C r z − z0 �
f (z)
fact, this becomes an equality in the limit as r → 0. But dz is
C z − z0
independent of r, and so in fact
�
f (z)
dz = 2πif (z0 ).
C z − z0

Dividing through by 2πi gives Cauchy’s formula.

The main theoretical application of Cauchy’s theorem is to think of the
point z0 as a variable point inside of the region R such that C = ∂R; note

10
that the z in the formula is a dummy variable. Thus we could equally well
write: �
1 f (w)
f (z) = dw,
2πi C w − z
for all z enclosed by C. This description of the analytic function f (z) by
an integral depending only on its values on the boundary curve of R turns
out to have many very surprising consequences. For example, it turns out
that an analytic function actually has derivatives of all orders, not just first
derivatives, which is very unlike the situation for functions of a real variable.
In fact, every analytic function can be expressed as a power series. This fact
can be seen by rewriting Cauchy’s formula above as
�
1 f (w)
f (z) = dw,
2πi C w(1 − z/w)
1
and then expanding as a geometric series. The fact that every
1 − z/w
analytic function is given by a convergent power series is yet another way of
characterizing analytic functions.

3.5 Homework
�
1. Let f (z) = x2 + iy 2 . Evaluate f (z) dz, where (a) C is the straight
C
line joining 1 to 2 + i; (b) C is the curve (1 + t) + t2 i, 0 ≤ t ≤ 1. Are
the results the same? Why or why not might you expect this?
2. Let α = c + di be a complex number. Verify directly that
d αt
e = αeαt .
dt
3. Let D be a region in C and let u(x, y) be a real-valued function on D.
We seek another real-valued function v(x, y) such that f (z) = u + iv
is analytic, i.e. satisfies the Cauchy-Riemann equations. Equivalently,
we want to find a function v such that
∂v ∂u ∂v ∂u
=− and = ,
∂x ∂y ∂y ∂x
� �
∂u ∂u
which says that ∇v is the vector field F = − , . Show that
∂y ∂x
F satisfies the mixed partials condition exactly when u is harmonic.
Conclude that, if D is simply connected, then F is a gradient vector
field ∇v and hence that u is the real part of an analytic function.

11
4. Let C be a circle
� centered at 4+i of radius 1. Without any calculation,
1
explain why dz = 0.
C z
5. Let C be the curve defined parametrically as follows:

z(t) = t(1 − t)et + [cos(2πt3 )]i, 0 ≤ t ≤ 1.

�
2
Evaluate the integral ez dz. Be sure to explain your reasoning!
C

6. Let D be a simply connected region in C and let C be a simple closed

curve contained in D. Let f (z) be analytic in D.� Suppose that z0 is a
1 f (z)
point which is not enclosed by C. What is dz?
2πi C z − z0
�
ez
7. Use Cauchy’s formula to evaluate dz, where C is a circle of
C z+1
radius 4 centered at the origin (and oriented counterclockwise).

8. Let C be the unit circle centered at 0 in the complex plane C and

oriented counterclockwise. Evaluate each of the following integrals,
and be sure that you can justify your answer by a calculation or a
clear and concise explanation.
� � 2 �
e−z
(a) 4
z dz; (b) dz; (c) z −5 dz;
C C z − i/2 C
� 2 � �
z − 1/3 1 e−2z
(d) dz; (e) dz; (f) dz.
C z+5 C (12z − 5) C 3z + 2
2

9. Let D be a simply connected region in C and let C be a simple closed

curve contained in D. Let f (z) be analytic in D. Suppose that z0 is a
point enclosed by C.

(a) By the usual formulas, show that

� �
d f (z) f � (z) f (z)
= − .
dz z − z0 z − z0 (z − z0 )2

(b) By using the fact that the line integral of a complex function with
an antiderivative is zero and the above, conclude that
� �
f � (z) f (z)
dz = dz.
C z − z 0 C (z − z0 )2

12
(c) Now apply Cauchy’s formula to conclude that
�
1 f (z)
f � (z0 ) = dz.
2πi C (z − z0 )2

13