Concise Complex Analysis

Solution of Exercise Problems

Ai Shu Xue
March 9, 2008
1
Contents
1 Calculus 3
2 Cauchy Integral Theorem and Cauchy Integral Formula 14
3 Theory of Series of Weierstrass 27
4 Riemann Mapping Theorem 52
5 Dierential Geometry and Picards Theorem 53
6 A First Taste of Function Theory of Several Complex Variables 53
7 Elliptic Functions 53
8 The Riemann -Function and The Prime Number Theory 53
2
This is a solution manual of selected exercise problems from Concise Complex Analysis, Second Edition,
by Gong Sheng (World Scientic, 2007). This version solves the exercise problems in Chapter 1-3, except
the following: Chapter 1 problem 37-42; Chapter 2 problem 47, 49; Chapter 3 problem 15 (xi).
1 Calculus
1.
Proof. See, for example, Munkres [4], 38.
2. (1)
Proof. |2i| = 2, arg(2i) =

2
. |1 i| =

2, arg(1 i) =
7
4
. |3 + 4i| = 5, arg(3 + 4i) = arctan
4
3
0.9273
(Matlab command: atan(4/3)). | 5+12i| = 13, arg(5+12i) = arccos
_

5
13
_
1.9656 (Matlab command:
acos(5/13)).
(2)
Proof. (1 + 3i)
3
= 26 18i.
10
43i
=
8
5
+
6
5
i.
23i
4+i
=
5
17

14
17
i. (1 +i)
n
+ (1 i)
n
= 2
n
2
+1
cos
n
4
.
(3)
Proof. | 3i(2 i)(3 + 2i)(1 +i)| = 3

130.

(43i)(2i)
(1+i)(1+3i)

=
5

10
=
5
2
.
3.
Proof. Let = arctan
1
5
, = arctan
1
239
. Then 5 i =

26e
i
, 1 + i =

2e
i

4
and (5 i)
4
(1 + i) =
676

2e
i(

4
4)
. Meanwhile (5 i)
4
(1 + i) = 956 4i = 676

2e
i
. So we must have

4
4 = + 2k,
k Z, i.e.

4
= 4 arctan
1
5
arctan
1
239
+2k, k Z. Since 4 arctan
1
5
arctan
1
239
+2 > 0

2
+2 =
7
4
>

4
and 4 arctan
1
5
arctan
1
239
2 < 4

2
2 = 0 <

4
, we must have k = 0. Therefore

4
= 4 arctan
1
5

arctan
1
239
.
4.
Proof. If z = x + yi,
1
z
=
x
x
2
+y
2
+
y
x
2
+y
2
i. z
2
= x
2
y
2
+ 2xyi.
1+z
1z
=
1x
2
y
2
(1x)
2
+y
2
+
2y
(1x)
2
+y
2
i.
z
z
2
+1
=
x
3
+xy
2
+x+i(y
3
x
2
y+y)
(x
2
y
2
+1)
2
+4x
2
y
2
.
5.
Proof. a = 1, b = +i, c = +i. So = b
2
4ac = (
2

2
4) +i(2 4) and
z =
( +i)

2
.
6.
Proof. Denote arg z by , then z +
r
2
z
= re
i
+
r
2
re
i
= r(e
i
+ e
i
) = 2r cos = 2Rez, and z
r
2
z
=
re
i

r
2
re
i
= r(e
i
e
i
) = 2ir sin = 2iImz.
7.
3
Proof. If a = r
1
e
i
1
and b = r
2
e
i
2
, then

a b
1 ab

r
1
r
2
e
i(
2

1
)
1 r
1
r
2
e
i(
2

1
)

.
Denote
2

1
by , we can reduce the problem to comparing |r
1
r
2
e
i
|
2
and |1 r
1
r
2
e
i
|
2
. Note
|r
1
r
2
e
i
|
2
= (r
1
r
2
cos )
2
+r
2
2
sin
2
= r
2
1
2r
1
r
2
cos +r
2
2
and
|1 r
1
r
2
e
i
|
2
= (1 r
1
r
2
cos )
2
+r
2
1
r
2
2
sin
2
= 1 2r
1
r
2
cos +r
2
1
r
2
2
.
So |1 r
1
r
2
e
i
|
2
|r
1
r
2
e
i
|
2
= (r
2
1
1)(r
2
2
1). This observation shows

ab
1 ab

= 1 if and only if at least

one of a and b has modulus 1;

ab
1 ab

< 1 if |a| < 1 and |b| < 1.

8.
Proof. We prove by induction. The equation clearly holds for n = 1. Assume it holds for all n with n N.
Then

N+1

i=1
a
i
b
i

2
=

i=1
a
i
b
i
+a
N+1
b
N+1

2
=
_
N

i=1
a
i
b
i
+a
N+1
b
N+1
__
N

i=1
a
i

b
i
+ a
N+1

b
N+1
_
=

i=1
a
i
b
i

2
+|a
N+1
|
2
|b
N+1
|
2
+ a
N+1

b
N+1
N

i=1
a
i
b
i
+a
N+1
b
N+1
N

i=1
a
i

b
i
=
N

i=1
|a
i
|
2
N

i=1
|b
i
|
2

ii<jN
|a
i

b
j
a
j

b
i
|
2
+|a
N+1
|
2
|b
N+1
|
2
+|b
N+1
|
2

1i<j=N+1
|a
i
|
2
+|a
N+1
|
2

1i<j=N+1
|b
i
|
2

1i<j=N+1
_
|a
i
|
2
|b
N+1
|
2
+|a
N+1
|
2
|b
i
|
2
a
i
a
N+1
b
i

b
N+1
a
i
a
N+1

b
i
b
N+1

=
N+1

i=1
|a
i
|
2
N+1

i=1
|b
i
|
2

1i<jN
|a
i

b
j
a
j

b
i
|
2

1i<j=N+1
|a
i

b
j
a
j

b
i
|
2
=
N+1

i=1
|a
i
|
2
N+1

i=1
|b
i
|
2

1i<jN+1
|a
i

b
j
a
j

b
i
|
2
.
By method of mathematical induction, |

n
i=1
a
i
b
i
|
2
=

n
i=1
|a
i
|
2

n
i=1
|b
i
|
2

1i<jn
|a
i

b
j
a
j

b
i
|
2
holds
for all n N. Cauchys inequality follows directly from this equation.
9.
Proof. We rst consider the special case where a
1
= 0 and a
2
= 1. Then for a
3
= re
i
, a
1
, a
2
, a
3
are vertices
of an equilateral triangle if and only if r = 1 and cos =
1
2
. Meanwhile the equation 1 + re
2i
= re
i
is
equivalent to
_
1 +r cos 2 = r cos
r sin2 = r sin,
which implies r = 1 and cos =
1
2
. So the equality is proven for the special case a
1
= 0, a
2
= 1.
For general case, note a
2
1
+ a
2
2
+ a
2
3
= a
1
a
2
+ a
2
a
3
+ a
3
a
1
if and only if (a
2
a
1
)
2
+ (a
3
a
1
)
2
=
(a
2
a
1
)(a
3
a
1
), which is further equivalent to 1 +
_
a
3
a
1
a
2
a
1
_
2
=
a
3
a
1
a
2
a
1
. This shows the general case can be
reduced to the special case of a
1
= 0 and a
2
= 1. Essentially, these reductions correspond to the composition
of coordinate translation and coordinate rotation.
4
10.
Proof.
D :=

1
1

1


1
0 0 1

.
Suppose = r
1
e
i
1
and = r
2
e
i
2
. Then

r
1
e
i
1
r
1
e
i
1
r
2
e
i
2
r
2
e
i
2

= r
1
r
2
e
i(
1

2
)
r
1
r
2
e
i(
1

2
)
.
So D = 0 if and only if
1

2
= k, which is equivalent to , , being colinear. Note the above reduction
corresponds to the coordination transformation that places at origin.
11.
Proof. We apply formula (1.11) (correction: the formula for x
2
should be x
2
=
z z
(1+|z|
2
)i
). If z = 1 i, then
x
1
=
2
3
, x
2
=
2
3
, and x
3
=
2
3
. If z = 4 + 3i, then x
1
=
4
13
, x
2
=
3
13
, and x
3
=
12
13
.
12.
Proof. Denote by x = (x
1
, x
2
, x
3
) and y = (y
1
, y
2
, y
3
) the points on the Riemann sphere S
2
corresponding
to z
1
and z
2
, respectively. Then x and y are two ends points of a diagonal of S
2
if and only if x +y = 0. By
formula (1.11), x +y = 0 is equivalent to
_

_
z
1
+ z
1
1+|z
1
|
2
+
z
2
+ z
2
1+|z
2
|
2
= 0
z
1
z
1
1+|z
1
|
2
+
z
2
z
2
1+|z
2
|
2
= 0
|z
1
|
2
1
|z
1
|
2
+1
+
|z
2
|
2
1
|z
2
|
2
+1
= 0.
Solving the third equation gives |z
1
z
2
| = 1. So we can assume z
1
= re
i
and z
2
=
1
r
e
i
(r > 0). Then
1 +|z
1
|
2
= 1 +r
2
and 1 +|z
2
|
2
= 1 +
1
r
2
=
1
r
2
(1 +|z
1
|
2
). So the other two equations become
_
2r cos + 2r cos = 0
2r sin + 2r sin = 0,
which is equivalent to
_
cos
+
2
cos

2
= 0
sin
+
2
cos

2
= 0.
So we must have
1
2
() = k +

2
(k Z), i.e. = 2k +. So z
1
z
2
= e
i()
= e
i(2k+)
= 1.
13.
Proof. Suppose z C is on a circle. Then there exists z
0
C and R > 0, such that |z z
0
| = R. Suppose
z = x+iy and z
0
= x
0
+iy
0
. Then |z z
0
| = R implies (xx
0
)
2
+(y y
0
)
2
= R
2
. After simplication, this
can be written as |z|
2
+ z
0
z +z
0
z +|z
0
|
2
R
2
= 0. Let A = 1, B = z
0
and C = |z
0
|
2
R
2
. Then A, C R
and |B|
2
> AC.
Conversely, if A|z|
2
+Bz +

B z +C = 0 is the equation satised by z, we have two cases to consider: (i)
A = 0; (ii) A = 0.
In case (i), A|z|
2
+ Bz +

B z + C = 0 can be written as |z|
2
+
B
A
z +

B
A
z +
|B|
2
A
2
=
|B|
2
A
2

C
A
> 0. Dene
z
0
=

B
A
and R =
_
|B|
2
A
2

C
A
. The equation can be written as |z|
2
+ z
0
z+z
0
z+|z
0
|
2
= R
2
, which is equivalent
to |z z
0
| = R. So z is on a circle.
5
In case (ii), the equation Bz +

B z + C = 0 stands for a straight line in C. Indeed, suppose B = a + ib
and z = x + iy, then Bz +

B z + C = 0 becomes 2ax 2by + C = 0. If we regard straight lines as circles,
A|z|
2
+Bz +

B

+C = 0 represents a circle in case (ii) as well.

To see the necessary and sucient condition for the image of this circle on the Riemann sphere S
2
to be
a great circle, we suppose a generic point z on this circle corresponds to a point (x
1
, x
2
, x
3
) on the Riemann
sphere S
2
. Then by formula (1.10) z =
x
1
+ix
2
1x
3
, we can write the equation A|z|
2
+ Bz +

B z + C = 0 as
(B+

B)x
1
+i(

BB)x
2
+(AC)x
3
= (A+C). (x
1
, x
2
, x
3
) falls on a great circle if and only if (x
1
, x
2
, x
3
)
lies on the intersection of S
2
and a plane that contains (0, 0, 0), which is equivalent to A + C = 0 by the
equation (B +

B)x
1
+i(

B B)x
2
+ (AC)x
3
= (A+C).
14.
Proof. Suppose Z
1
= (x
1
, x
2
, x
3
) and Z
2
= (y
1
, y
2
, y
3
). Then by formula (1.11),
d(z
1
, z
2
)
2
= (x
1
y
1
)
2
+ (x
2
y
2
)
2
+ (x
3
y
3
)
2
= 2 2(x
1
y
1
+x
2
y
2
+x
3
y
3
)
= 2 2
_
z
1
+ z
1
1 +|z
1
|
2

z
2
+ z
2
1 +|z
2
|
2
+
z
1
z
1
i(1 +|z
1
|
2
)

z
2
z
2
i(1 +|z
2
|
2
)
+
|z
1
|
2
1
|z
1
|
2
+ 1

|z
2
|
2
1
|z
2
|
2
+ 1
_
= 4
|z
1
|
2
+|z
2
|
2
z
1
z
2
z
1
z
2
(1 +|z
1
|
2
)(1 +|z
2
|
2
)
= 4
(z
1
z
2
)( z
1
z
2
)
(1 +|z
1
|
2
)(1 +|z
2
|
2
)
.
So d(z
1
, z
2
) =
2|z
1
z
2
|

(1+|z
1
|
2
)(1+|z
2
|
2
)
. Consequently,
d(z
1
, ) = 2 lim
z
2

z
1
z
2
1

_
(1 +|z
1
|
2
)
_
1
|z
2
|
2
+ 1
_
=
2
_
1 +|z
1
|
2
.
15.
Proof. (i) The distance between z and z
1
is no greater than the distance between z and z
2
.
(ii) z z
1
and z z
2
are perpendicular to each other.
(iii) The angle between z +i and z i is less than

4
.
(iv) z is on an ellipse with c and c as the foci.
16.
Proof. Regard C as R
2
and apply Heine-Borel theorem and Bolzano-Weierstrass theorem for R
n
(n N).
17.
Proof. Note max{|Rez
n
Rez
0
|, |Imz
n
Imz
0
|} |z z
0
| |Rez
n
Rez
0
| +|Imz
n
Imz
0
|.
18. (1)
6
Proof. We suppose |a|, |b| < . Then (z
n
)
n
and (z

n
)
n
are bounded sequences. Denote by M a common
bound for both sequences. > 0, N, so that for any n > N, max{|z
n
a|, |z

n
b|} <

M
. Then for any n
satisfying n > N
_
1 +
M
2
2
_
(dene z
0
= z

0
= 0)

1
n
n

k=1
z
k
z

nk
ab

<

1
n
N

k=1
z
k
z

nk

1
n
n

k=N+1
z
k
z

nk
ab

<
NM
2
n
+

1
n
n

k=N+1
(z
k
a)z

nk

a
n
n

k=N+1
z

nk
ab

< + +|a|

1
n
nN1

k=0
z

k
b

= 2 +|a|

n N 1
n

1
n N 1
nN1

k=0
z

k
b

.
Taking upper limits on both sides, we get
lim
n

1
n
n

k=1
z
k
z

nk
ab

2 + 0 = 2.
Since is arbitrary, we conclude lim
n

1
n

n
k=1
z
k
z

nk
ab

= 0, i.e.
lim
n
1
n
n

k=1
z
k
z

nk
= ab.
(2)
Proof. Apply the result of (1) with z

n
1.
19.
Proof. (i) f(z) =

z z is a function of both z and z. So f is not dierentiable.
(ii) f(z) = z is a function of z, so it is not dierentiable.
(iii) f(z) = Rez =
1
2
(z + z) is a function of both z and z. So f is not dierentiable.
20.
Proof. By chain rule,

z
g(f(z)) = g

(f(z))
f(z)
z
= 0. So g(f(z)) is also holomorphic.
21. (1)
Proof. Let u(x, y) = Ref(z) and v(x, y) = Imf(z). Then by C-R equations f

(z) = u
x
(x, y) + iv
x
(x, y) =
v
y
(x, y) iu
y
(x, y). So f

(z) 0 if and only if v

x
= v
y
= u
x
= u
y
0. By results for functions of real
variables, we conclude u and v are constants on D. so f is a constant function on D.
(2)
Proof. (i) and (ii) are direct corollaries of C-R equations.
For (iii), we assume without loss of generality that f(z) 0 on D. We note |f(z)|
2
= u
2
(z) + v
2
(z). So
by the C-R equations, 0 =

x
|f(z)|
2
= 2uu
x
+ 2vv
x
= 2uu
x
2vu
y
and 0 =

y
|f(z)|
2
= 2uu
y
+ 2vv
y
=
2uu
y
+ 2vu
x
, i.e.
_
u v
v u
_ _
u
x
u
y
_
= 0. Since f 0,
_
u v
v u
_
is invertible on D
1
:= {z D : f(z) = 0}. So
u
x
= u
y
0 on D
1
. By the C-R equations v
x
= v
y
0 on D
1
. So f is a constant function on each simply
7
connected component of D
1
. Since D
2
:= {z D : f(z) = 0} has no accumulation points in D (see Theorem
2.13), f must be identical to the same constant throughout D.
For (iv), we assume arg f(z) . Then g(z) := e
i
f(z) is also holomorphic and f is a constant function
if and only if g is a constant function. So without loss of generality, we can assume arg f(z) 0. Then
Imf(z) 0. By C-R equations, Ref(z) is a constant too. Combined, we can conclude f is identically equal
to a constant on D.
22.
Proof. The tangent vector a(x, y) of the curve u(x, y) = c
1
at point (x, y) is perpendicular to (u
x
, u
y
);
the tangent vector b(x, y) of the curve v(x, y) = c
2
at point (x, y) is perpendicular to (v
x
, v
y
). Since
(v
y
, v
x
) (v
x
, v
y
) and (u
x
, u
y
) = (v
y
, v
x
) by C-R equations, we must have a(x, y) b(x, y), which
means the curves u(x, y) = c
1
and v(x, y) = c
2
are orthogonal.
23. (1)
Proof. Since
_
x = r cos
y = r sin
, we have
_

r

_
=
_
cos sin
r sin r cos
_ _

x

y
_
:= A(, r)
_

x

y
_
.
Its easy to see A
1
(, r) =
1
r
_
r cos sin
r sin cos
_
. Writing the Cauchy-Riemann equations in matrix form, we
get
_

x

y
_
u =
_
0 1
1 0
_ _

x

y
_
v.
Therefore, under the polar coordinate, the Cauchy-Riemann equations become
_

r

_
u = A(, r)
_

x

y
_
u = A(, r)
_
0 1
1 0
_ _

x

y
_
v = A(, r)
_
0 1
1 0
_
A
1
(, r)
_

r

_
v =
_
0
1
r
r 0
_ _

r

_
v.
Fix z = re
i
, then
f

(z) = lim
r0
u(r + r, ) +iv(r + r, ) [u(r, ) +iv(r, )]
r e
i
=
u
r
e
i
+i
v
v
e
i
=
r
z
_
u
r
+i
v
r
_
.
(2)
Proof. If f(z) = R(cos +i sin), then u = Rcos and v = Rsin. So
u
r
=
R
r
cos Rsin

r
, v
r
=
R
r
sin+Rcos

r
, u

=
R

cos Rsin

, v

=
R

sin+Rcos

.
Writing the above equalities in matrix form, we have
_
u
r
v
r
_
= A
_
R
r

r
_
,
_
u

_
= A
_
R

_
, where A =
_
cos Rsin
sin Rcos
_
.
Therefore the C-R equations become
A
_
R
r

r
_
=
_
u
r
v
r
_
=
_
0
1
r

1
r
0
_ _
u

_
=
_
0
1
r

1
r
0
_
A
_
R

_
.
8
Note A
1
=
1
R
_
Rcos Rsin
sin Rcos
_
. So A
1
_
0
1
r

1
r
0
_
A =
_
0
R
r

1
Rr
0
_
. Therefore
_
R
r

r
_
=
_
0
R
r

1
Rr
0
_ _
R

_
, i.e.
_
R
r
=
R
r

= Rr

r
.
24.
Proof. By the C-R equations, J =

u
x
u
y
v
x
v
y

= u
x
v
y
v
x
u
y
= u
x
u
x
v
x
(v
x
) = u
2
x
+v
2
x
= |f

(z)|
2
. J is
the area of f(D), see, for example, Munkres [4], 21 The volume of a parallelepiped.
25. (1)
Proof. We can parameterize with (t) = t +ti (0 t 1). So
_

xdz =
_
1
0
t(dt +idt) =
1
2
+
1
2
i.
(2)
Proof.
_

|z 1||dz| =
_
2
0
|e
it
1|dt =
_
2
0

2 2 cos tdt =

2
_
2
0

sin
_

4

t
2
_
cos
_

4

t
2
_

dt
=

2
_
4

3
4

|sin cos | d =

2
_
4

3
4

(cos sin) d4.

(3)
Proof.
_

dz
z a
=
_
2
0
iRe
it
Re
it
dt = 2i.
26.
Proof. We rst recall sinh z =
e
z
e
z
2
and coshz =
e
z
+e
z
2
. Then straightforward calculations show
cos(x +iy) = coshy cos x i sinhy sinx, sin(x +iy) = coshy sinx +i sinhy cos x.
By this result, we have
siniz =
e
i(iz)
e
i(iz)
2i
=
e
z
e
z
2i
= sinhz, cos iz =
e
i(iz)
+e
i(iz)
2
=
e
z
+e
z
2
= cosh z,
(sin z)

=
_
e
iz
e
iz
2i
_

=
e
iz
+e
iz
2
= cos z.
9
and
cos z
1
cos z
2
sinz
1
sinz
2
=
e
iz
1
+e
iz
1
2
e
iz
2
+e
iz
2
2

e
iz
1
e
iz
1
2i
e
iz
2
e
iz
2
2i
=
e
i(z
1
+z
2
)
+e
i(z
1
+z
2
)
+e
i(z
1
z
2
)
+e
i(z
1
z
2
)
4

e
i(z
1
+z
2
)
+e
i(z
1
+z
2
)
e
i(z
1
z
2
)
e
i(z
1
z
2
)
4
=
2e
i(z
1
+z
2
)
+ 2e
i(z
1
+z
2
)
4
= cos(z
1
+z
2
).
27.
Proof. sini = sinh1.
cos(2 +i) = cosh1 cos 2 i sinh1 sin2.
tan(1 +i) =
sin 1 cos 1+i sinh 1 cosh 1 cos 2
cosh
2
1 cos
2
1+sinh
2
1 sin
2
1
.
2
i
= e
2k
[cos(log 2) +i sin(log 2)], k Z.
i
i
= e

2
+2k
, k Z.
(1)
2i
= e
2(2k+1)
, k Z.
log(2 3i) = log

13 +i[2k arctan
3
2
], k Z.
arccos
1
4
(3 +i) =
3+i

10e
i(

1
2
arctan
3
4
)
4
.
28.
Proof. e

2
i
= i, e

2
3
i
=
1
2

3
2
i. log i = (2k +

2
)i, k Z.
29.
Proof. z
z
= e
z log z
= e
(x+iy) log(x+iy)
= e
x
2
log(x
2
+y
2
)y(2k+arctan
y
x
)
e
i[
y
2
log(x
2
+y
2
)+x(2k+arctan
y
x
)]
, k Z.
30.
Proof. The root of z
n
= a are |a|
1
n
e
i
2k+arg a
n
, k = 0, 1, , n 1.
31. (1)
Proof. If u
n
(z) = Ref
n
(z) and v
n
(z) = Imf
n
(z), then by max{|u
n
(z) u
m
(z)|, |v
n
(z) v
m
(z)|} |f
n
(z)
f
m
(z)| |u
n
(z) u
m
(z)| +|v
n
(z) v
m
(z)|, the Cauchy criterion for convergence for complex eld is reduced
to that for real numbers.
(2)
Proof. Note |

m
n=k
f
n
(z)| M

m
n=k
a
n
. So

n=1
f
n
(z) converges uniformly if

n=1
a
n
converges.
32. (i)
Proof. Since lim
n
n
1
n
2
= e
lim
n
ln n
n
2
= 1, Hadamards formula implies R = 1.
(ii)
Proof. Since lim
n
n
ln n
n
= e
lim
n
[
ln
n
ln n]
1, Hadamards formula implies R = 1.
(iii)
10
Proof. By Stirlings formula n!

2n
_
n
e
_
n
, we have
lim
n
_
n!
n
n
_ 1
n
= lim
n
e
1
n
ln
n!
n
n
= lim
n
e
1
n
[ln(n!)ln

2n(
n
e
)
n
]
lim
n
e
1
n
ln

2n(
n
e
)
n
ln n
= e
1
.
So Hadamards formula implies R = e.
(iv)
Proof. Since lim
n
n = , Hadamards formula implies R = 0.
33.
Proof. Let u(z) = Ref(z) and v(z) = Imf(z). Then f is uniformly convergent on a set A if and only if u
and v are uniformly convergent on A. Since dz = dx+idy, we can reduce the theorem to that of real-valued
functions of real variables.
34. (1)
Proof. [f(z)e
z
]

= f

(z)e
z
f(z)e
z
= 0. So f(z) = Ce
z
for some constant C. f(0) = 1 dictates
C = 1.
(2)
Proof. f

(z) = lim
0
f(z+)f(z)

= lim
0
f(z)
f()1

= f(z)f

(0) = f(z). By part (1), we conclude

f(z) = e
z
.
35. (1)
Proof. [e
f(z)
]

= f

(z)e
f(z)
= 1. So e
f(z)
= 1.
(2)
Proof. Its clear f(1) = 0. So
f

(z) = lim
0
f(z +z) f(z)
z
= lim
0
f(z(1 +)) f(z)
z
=
1
z
lim
0
f(1 +)

=
1
z
f

(1) =
1
z
.
Therefore
_
1
z
e
f(z)
_

=
1
z
2
e
f(z)
+
1
z
e
f(z)
f

(z) = 0.
This shows e
f(z)
= z.
36.
Proof. 2[(z
1
) (z
2
)] = (z
1
z
2
)
_
1
1
z
1
z
2
_
. So (z) is univalent in any region that makes z
1
z
2
= 1
whenever z
1
= z
2
. Since z
1
z
2
= 1 implies |z
1
| =
1
|z
2
|
and arg z
1
= arg z
2
, we conclude is univalent in any
of the four regions in the problem.
43.
Proof. This is straightforward from Problem 24.
44.
11
Proof. We rst use Abels Lemma on partial sum: for n > m,
n

k=m+1
a
k
b
k
=
n

k=m+1
(S
k
S
k1
)b
k
= S
b
b
n1
+
n

k=m+1
S
k
(b
k
b
k+1
) S
m
b
m+1
.
Denote by K a bound of the sequence (S
n
)

n=1
, then |S
n
b
n+1
| K|b
n+1
|, |S
m
b
m+1
| K|b
m+1
|, and
|

n
k=m+1
S
k
(b
k
b
k+1
)| K

n
k=m+1
|b
k
b
k+1
|. Since lim
n
b
n
= 0 and

n=1
|b
n
b
n+1
| < ,
we conclude

n
k=m+1
a
k
b
k
can be arbitrarily small if m is suciently large. So under conditions (1)-(3),

n=1
a
k
b
k
is convergent.
In the Dirichlet criterion, we require
(1) (S
n
)

n=1
is bounded;
(2) lim
n
b
n
= 0;
(3) (b
n
)

n=1
is monotone.
Clearly, (2)+(3) imply (3), so the Dirichlet criterion is a special case of the result in current problem.
In the Abel criterion, we require
(1) (S
n
)

n=1
is convergent;
(2) (b
n
)

n=1
is bounded;
(3) (b
n
)

n=1
is monotone.
Dene b

n
= b
n
b, where b = lim
n
b
n
is a nite number. Then (S
n
)

n=1
satises (1) and (b

n
)

n=1
satises (2) and (3). By the Dirichlet criterion,

n=1
a
n
b

n
converges. Note

N
n=1
a
n
b
n
=

N
n=1
a
n
b

n
+bS
N
.
By condition (1), we conclude

n=1
a
n
b
n
converges.
Remark 1. The result in this problem is the so-called Abel-Dedekind-Dirichlet Theorem.
45. (1)
Proof. For any (0, 1 q), there exists N N, such that for any n N, |a
n
|
1
n
< q + < 1. By comparing

n=1
|a
n
| and

n=1
(q +)
n
, we conclude

n=1
a
n
is absolutely convergent.
(2)
Proof. By denition of upper limit and the fact q > 1, we can nd > 0 and innitely many n, such that
|a
n
|
1
n
> q > 1. Therefore lim
n
|a
n
| = and

n=1
a
n
is divergent.
46.
Proof. If q < 1, for any (0, 1 q), we can nd N N, such that for any n N,
|a
n+1
|
|a
n
|
< q + < 1. So
|a
N+k
| |a
N
|(q +)
k
, k 0. Since

k=1
(q +)
k
is convergent, we conclude

n=1
a
n
is also convergent.
If q > 1, we have two cases to consider. In the rst case, the upper limit is assumed to be a limit. Then
we can nd > 0 such that q > 1 and for n suciently large, |a
n+1
| > |a
n
|(q ) > |a
n
|. This implies
lim
n
a
n
= 0, so the series is divergent. In the second case, the upper limit is assumed not to be a limit.
Then we can manufacture counter examples where the series is convergent. Indeed, consider (k 0)
a
n
=
_
1
(k+1)
2
, if n = 2k + 1
2
(k+1)
2
, if n = 2k + 2.
Then

n=1
a
n
is convergent and lim
n

a
n+1
a
n

= 2.
47.
12
Proof. We choose > 0, such that lim
n
n
_
|
a
n+1
a
n
| 1
_
< 1 2. By the denition of upper limit, there
exists N
1
N, such that for any n N
1
, n
_
|
a
n+1
a
n
| 1
_
< 1 2, i.e.
|a
n+1
|
|a
n
|
< 1
1+2
n
. Dene b
n
=
1
n
1+
.
Then
b
n+1
b
n
=
n

(n + 1)
1+
=
_
1
1
n + 1
_
1+
= 1
1 +
n + 1
+O
_
1
(n + 1)
2
_
.
So there exists N
2
N, such that for n N
2
,
b
n+1
b
n
> 1
1+2
n+1
. Dene N = max{N
1
, N
2
}, then for any
n N,
a
n+1
a
n
<
b
n+1
b
n
. In particular, we have

a
N+1
a
N

<
b
N+1
b
N
,

a
N+2
a
N+1

<
b
N+2
b
N+1
, ,

a
N+k
a
N+k1

<
b
N+k
b
N+k1
.
Multiply these inequalities, we get
|a
N+k
|
|a
N
|
b
N
b
N+k
.
Since

n=1
b
n
is convergent, we conclude

n=1
a
n
must also converge.
1
48. (1)
Proof. Since (a
n
)

n=1
monotonically decreases to 0, for n suciently large, lna
n
< 0 and lim
n
ln a
n
n
0.
So lim
n
a
1
n
n
= e
lim
n
ln a
n
n
e
0
= 1. By Hadamards formula, R 1.
(2)
Proof. The claims seems problematic. For a counter example, assume R > 1 and a
n
=
1
R
n
. If z = Re
i
with
= 0, we have
N

n=0
a
n
z
n
=
N

n=0
1
R
n
R
n
e
in
=
N

n=0
cos n +i
N

n=0
sinn =
sin
(N+1)
2
cos
N
2
sin

2
+i
sin
(N+1)
2
sin
N
2
sin

2
.
So

n=0
a
n
z
n
is not convergent.
49.
Proof. By denition of uniform convergence, for any > 0, there exists N N, such that for any m N,

n=m
|a
n
||z z
0
|
n
< , z D. In particular, for any z

D, by letting z z

, we get

n=m
|a
n
||z

z
0
|
n
. That is, for this given , we can pick up N, such that for any m N,

n=m
|a
n
||z z
0
|
n
,
z

D. This is exactly the denition of uniform convergence on

D.
50.
Proof. We choose a sequence (r
n
)

n=1
such that r
n
(0, 1) and lim
n
r
n
= 1. Dene
n
=

n
k=0
a
k
z
k
0

f(r
n
z
0
). We show lim
n

n
= 0. Indeed, we note
|
n
| =

k=0
a
k
z
k
0

k=0
a
k
(r
n
z
0
)
k

k=0
a
k
(1 r
k
n
)z
k
0

k=n+1
a
k
(r
n
z
0
)
k

k=0
|a
k
|(1 r
n
)(1 +r
n
+ +r
k1
n
) +

k=n+1
k
n
|a
k
|r
k
n

k=0
|a
k
|(1 r
n
)k +

n+1
n
1
1 r
n
,
1
It seems the condition lim
n

a
n+1
a
n

= 1 is redundant.
13
where
n
=

kn
k|a
k
|. We choose r
n
in such a way that lim
n
n(1 r
n
) = 1, e.g. r
n
= 1
1
n
. Then
lim
n

n+1
n(1r
n
)
= 0 since lim
n
na
n
= 0. By Problem 18 (2), we have
n

k=0
|a
k
|(1 r
n
)k = (1 r
n
)n

n
k=0
|a
k
|k
n
0, as n .
Combined, we conclude lim
n

n
= 0.
Remark 2. The result in this problem is a special case of the so-called Landaus Theorem, see, for example,
Boos [1], 5.1 Boundary behavior of power series.
2 Cauchy Integral Theorem and Cauchy Integral Formula
1.
Proof. (i) The integral is equal to 2i sini = (e
1
e).
(ii) The integral is equal to 2ei.
(iii) For suciently small > 0, the integral is equal to
_
|z|=
cos z
(z +)(z )
dz +
_
|z+|=
cos z
(z +)(z )
dz = 2i
_
cos
2
+
cos()
2
_
= 0.
(iv) The integral is equal to 2i
_
1
z3
_
(n1)

z=1
= 2i
(n1)!
2
n
=
(n1)!
2
n1
i.
(v) For suciently small > 0he integral is equal to
_
|z+i|=
1
(z i)(z
2
+ 4)

dz
(z +i)
+
_
|zi|=
1
(z +i)(z
2
+ 4)

dz
(z i)
= 2i
_
1
(2i) 3
+
1
2i 3
_
= 0.
(vi) By the Fundamental Theorem of Algebra, the equation z
5
1 = 0 has ve solutions z
1
, z
2
, z
3
, z
4
, z
5
.
For any i {1, 2, 3, 4, 5}, we have
(z
i
z
1
) (z
i
z
i1
)(z
i
z
i+1
) (z
i
z
5
) = lim
zz
i
z
5
1
z z
i
= lim
zz
i
z
5
z
5
i
z z
i
= 5z
4
i
=
5
z
i
.
So for suciently small > 0, the integral is equal to
5

i=1
_
|zz
i
|=
dz
(z z
1
)(z z
2
)(z z
3
)(z z
4
)(z z
5
)
= 2i
5

i=1
1
(z
i
z
1
) (z
i
z
i1
)(z
i
z
i+1
) (z
i
z
5
)
= 2i
6

i=1
z
i
5
= 0,
where the last equality comes from the fact that the expansion of (zz
1
) (zz
5
) is z
5
1 and (z
1
+ +z
5
)
is the coecient of z
4
.
(vii) If |a|, |b| > R, the integral is equal to 0. If |a| > R > |b|, the integral is equal to
2i
(ba)
n
. If
|b| > R > |a|, the integral is equal to 2i
_
1
zb
_
(n1)

z=a
= 2i (1)
n1
(a b)
n
. If R > |a|, |b|, the
integral is equal to 2i
_
1
(ba)
n
+ (1)
n1 1
(ab)
n
_
= 0.
14
(viii) Denote by z
1
, z
2
, z
3
the three roots of the equation z
3
1 = 0. Then the integral is equal to
2i
_
1
(z
1
z
2
)(z
1
z
3
)(z
1
2)
2
+
1
(z
2
z
1
)(z
2
z
3
)(z
2
2)
2
+
1
(z
3
z
1
)(z
3
z
2
)(z
3
2)
2
+
_
1
z
3
1
_
(1)

z=2
_
= 2i
_
z
1
3(z
1
2)
2
+
z
2
3(z
2
2)
2
+
z
3
3(z
3
2)
2

12
49
_
.
2.
Proof.
1
2i
_
C
e
z

n

d

=
1
2i
_
C

k=0
z
k
k!
1

nk

d

k=0
z
k
k!

1
2i
_
C
1

nk

d

=
z
n
n!
,
where the last equality is due to Cauchys integral formula. We need to justify the exchange of integration
and innite series summation. Indeed, for m > n, we have

k=0
z
k
k!

1
2i
_
C
1

nk

d

1
2i
_
C

k=0
z
k
k!
1

nk

d

1
2i
_
C

k=m+1
z
k
k!
1

nk

d

1
2
_
C

k=m+1
|z|
k
k!
|d|
||
n+1

1
2
_
C

k=m+1
M
k
k!
|d|
||
n+1
,
where M = |z| max
C
||. Since

k=m+1
M
k
k!
is the remainder of e
M
, for m suciently large,

k=m+1
M
k
k!
can be smaller than any given positive number . So for m suciently large,

k=0
z
k
k!

1
2i
_
C
1

nk

d

1
2i
_
C

k=0
z
k
k!
1

nk

d

1
2
_
C
|d|
||
n+1
.
This shows

k=0
z
k
k!

1
2i
_
C
1

nk

d

converges to
1
2i
_
C

k=0
z
k
k!
1

nk

d

.
3.
Proof. We note
g()
z1
=
g()

1
z
. If |z| < 1, we apply Cauchys integral formula to
1
2i
_
||=1
f()
z
d and apply
Cauchys integral theorem to
1
2i
_
||=1
g()

1
z
d (regarding
1
z
as a parameter and
g()

1
z
an analytic function of
). If |z| > 1, we apply Cauchys integral theorem to
1
2i
_
||=1
f()
z
d (regarding z as a parameter and
f()
z
an analytic function of ) and apply Cauchys integral formula to
1
2i
_
||=1
g()

1
z
d.
4.
Proof. For z {z : |z| < 3}, by Cauchys integral formula, f(z) = 2i(3z
3
+ 7z + 1). So f

(1 + i) =
2i[6(1 +i) + 7] = 12 + 26i.
5.
15
Proof. We note
_
|z|=1
(z +
1
z
)
2n
dz
z
=
_
2
0
(e
i
+e
i
)
2n
id = 2
2n
i
_
2
0
cos
2n
d.
On the other hand,
(z +
1
z
)
2n

1
z
=
2n

k=0
C
k
2n
z
k
(z
1
)
2nk
z
1
=
2n

k=0
C
k
2n
z
2n+2k1
.
Therefore
_
2
0
cos
2n
d =
1
2
2n
i
_
|z|=1
(z +
1
z
)
2n
dz
z
=
1
2
2n
i
_
|z|=1
C
n
2n
dz
z
=
2C
n
2n
2
2n
= 2
1 3 5 (2n 1)
2 4 6 2n
.
6.
Proof. This is straightforward from Cauchys integral formula.
7.
Proof. For any R > 0, we have f
(n)
(0) =
n!
2i
_
|z|=R
f()

n+1
d. So

f
(n)
(0)
n!

=
1
2

_
2
0
|f(Re
i
)|
R
n
d

1
2

_
2
0
Me
|Re
i
|
R
n
d

= M
e
R
R
n
.
Let R = n, we get the desired inequality.
Remark 3. Note h(R) = ln
e
R
R
n
= R nlnR takes minimal value at n. This is the reason why we choose
R = n, to get the best possible estimate. Another perspective is to assume f(z) = e
z
. Then

f
(n)
(0)
n!

=
1
n!

1

2n(
n
e
)
n =
1

2n
_
e
n
_
n
by Stirlings formula. So the bound M
e
R
R
n
is not far from the best one.
8. (1)
Proof. For any z D
2
, we can nd R > 0 suciently large, so that z is in the region enclosed by and
{ : | z| = R}. By Cauchys integral formula
f(z) =
1
2i
_

f()
z
d +
1
2i
_
||=R
f()
z
d.
So
1
2i
_

f()
z
d = f(z) +
1
2i
_
||=R
f()
z
d
= f(z) +
1
2i
_
||=R
f() f()
z
d +
1
2i
_
||=R
f()
z
d.
We note

1
2i
_
||=R
f() f()
z
d

sup
D(z,R)
|f() A| 0, as R ,
and Cauchys integral formula implies
1
2i
_
||=R
f()
z
d = f() 1 = A.
16
So by letting R in the above equality, we have
1
2i
_

f()
z
d = f(z) +A.
If z D
1
, h() =
f()
z
is a holomorphic function in D
2
, with z a parameter. So for R suciently large,
Cauchys integral theorem implies
0 =
1
2i
_

f()
z
d +
1
2i
_
||=R
f()
z
d.
An argument similar to (1) can show
1
2i
_

f()
z
d = A.
(2)
Proof. We note
z
z
2
=
1

1
z
. For R suciently large, Cauchys integral theorem implies
0 =
1
2i
_

f()

d +
1
2i
_
||=R
f()

d.
So
1
2i
_

f()

d =
1
2i
_
||=R
f()

d. Similarly,
1
2i
_

f()
z
d =
_
1
2i
_
||=R
f()
z
d f(z) z D
2
1
2i
_
||=R
f()
z
d z D
1
.
Therefore
1
2i
_

zf()
z
2
d =
1
2i
_

_
f()

f()
z
_
d =
_
f(z) z D
2
0 z D
1
.
9.
Proof. Since f(z) = 0 in D, 1/f(z) is holomorphic in D. Applying Maximum Modulus Principle to 1/f(z),
we conclude |f(z)| achieves its minimum on D. Applying Maximum Modulus Principle to f(z), we conclude
|f(z)| achieves its maximum on D. Since |f(z)| M on D, |f(z)| must be a constant on

D. By Chapter
1, exercise problem 21 (iii), we conclude f(z) is a constant function.
10.
Proof. For R large enough, |a|, |b| < R. So we can nd > 0 such that (assume a = b)
_
|z|=R
f(z)
(z a)(z b)
dz =
_
|za|=
f(z)
(z a)(z b)
dz +
_
|zb|=
f(z)
(z a)(z b)
dz =
2i
a b
[f(a) f(b)].
Meanwhile, denote by M a bound of f, we have

_
|z|=R
f(z)
(z a)(z b)
dz

M
_
|z|=R
|dz|
|(z a)(z b)|
M
_
|z|=R
|dz|
(R |a|)(R |b|)
=
2MR
(R |a|)(R |b|)
0, as R .
Combined, we conclude 0 =
2i
ab
[f(a) f(b)], a, b C. This shows f is a constant function.
17
11.
Proof. By Cauchys integral formula,
|f
(n)
(0)|
n!
2

_
|z|=r
f()

n+1
d

n!
2
_
2
0
1
1r
r
n
d =
n!
r
n
(1 r)
.
12.
Proof. We note
lim
zz
0
(z) (z
0
)
z z
0
=
1
2i
lim
zz
0
_

()
( z)( z
0
)
d =
1
2i
_

()
( z
0
)
2
d,
where the exchange of limit and integration can be seen as an application of Lebesgues Dominated Conver-
gence Theorem (We can nd > 0 so that z in a neighborhood of z
0
, dist(, z) . Then
1
(z)(z
0
)
)
1

2
).
This shows is holomorphic in any region D which does not contain any point of .
The formula for n-th derivative of can be proven similarly and by induction. For details, we refer the
reader to Fang [3], Chapter 4, 3, Lemma 2.
Remark 4. The result still holds if is piecewise continuous on . See, for example, Fang [3], Chapter 7,
proof of Theorem 7.
13.
Proof. Since f(z) = 0 in D, 1/f(z) is holomorphic in D. Applying Maximum Modulus Principle to 1/f(z),
we conclude |f(z)| does not achieve its minimum in D unless its a constant function. Applying Maximum
Modulus Principle to f(z), we conclude f(z) does not achieve its maximum in D unless its a constant
function. Combined, we have m < |f(z)| < M for any point z D.
14.
Proof. Dene f(z) =
p
n
(z)
z
n
and g(z) = f(
1
z
). Then g(z) is holomorphic and bounded on D(0, 1) \ {0}, since
lim
z0
g(z) = lim
z
p
n
(z)
z
n
is a nite number and g(z) is clearly continuous in a neighborhood of D(0, 1).
By Theorem 2.11, Riemanns Theorem, g(z) can be analytically continued to D(0, 1). By a corollary of
Theorem 2.18, Maximum Modulus Principle, the maximum value of |g(z)| can only be reached on D(0, 1).
So |g(z)| max
|z|=1
|g(z)| = max
|z|=1
|f(z)| = max
|z|=1
|p
n
(z)| M, z D(0, 1), i.e. |f(z)| =
|p
n
(z)|
|z
n
|
M
for 1 |z| < .
15.
Proof. Dene g(z) =
f(Rz)
M
(z D(0, 1)) and apply Theorem 2.19, Schwarz Lemma, to g(z), we have
|g(z)| |z| (z D(0, 1)) and |g

(0)| 1. So for any z D(0, R), f(z) = f(R

z
R
) = Mg(
z
R
) M
|z|
R
and
|f

(0)| = |g

(0)|
M
R

M
R
.
16.
Proof. Dene = dist(C \ V, K). Then (0, ). By Theorem 2.7, for any z K, we have
|f
(n)
(z)|
n!
2
_
V
|f()|
| z|
n+1
|d|
n!
2
_
V
|d|

n+1
sup
zV
|f(z)|.
Dene c
n
=
n!
2
_
V
|d|

n+1
and take supremum on the left side of the above inequality, we have
sup
zK
|f
(n)
(z)| c
n
sup
zV
|f(z)|.
18
17.
Proof. We have to assume has nite length, that is, L() :=
_

|dz| < . Then by the denition of uniform

convergence, for any > 0, there exists n
0
such that n n
0
, sup
zD
|f(z) f
n
(z)|

L()
. So for any
n n
0
,

f(z)dz
_

f
n
(z)dz

|f(z) f
n
(z)||dz|
_

L()
|dz| = .
Therefore lim
n
_

f
n
(z)dz =
_

f(z)dz.
18.
Proof. We note the linear fractional transformation (z) =
z
z+
maps {z : Rez > 0} to D(0, 1). So
h(z) = (f(z)) maps D(0, 1) to itself (see, for example, Fang [3], Chapter 3, 7) and h(0) = 0. By Schwarz
Lemma, h(z)| = |z| and |h

(0)| 1. This is equivalent to

f(z)
f(z) +

|z|, and |f

(0)| 2.
19.
Proof. The linear fractional transformation (z) =
z
z2A
maps {z : Rez < A} to D(0, 1). So h(z) =
(f(z)) =
f(z)
f(z)2A
maps D(0, 1) to itself and h(0) = 0. By Schwarz Lemma, |h(z)| |z|. So z D(0, 1)
|f(z)| |z| |f(z) 2A| |z||f(z)| + 2A|z|, i.e. |f(z)|
2A|f(z)|
1 |z|
.
20. (1)
Proof. We note
e
z
+e
z
+ 2 cos z
4
=
1
4
[e
z
+e
z
+e
iz
+e
iz
] =
1
4

k=0
1
k!
[z
k
+ (z)
k
+ (iz)
k
+ (iz)
k
] =

m=0
z
4m
(4m)!
.
The radius of convergence is .
(2)
Proof. By induction, it is easy to show
_
1
(1)
3
_
(n)
=
(n+2)!
2!
(1 )
(n+3)
. So the Taylor expansion of
1
(1)
3
is
1
(1)
3
=

n=0
(n+1)(n+2)
2

n
, with radius of convergence equal to 1. This implies
+ 4
2
+
3
(1 )
3
= + 7
4
+

n=3
(3n
2
3n + 1)
n
.
Replace with z
2
, we can get the Taylor expansion at z = 0: z
2
+ 7z
8
+

n=3
(3n
2
3n + 1)z
2n
, and the
radius of convergence is equal to 1 (this can be veried by calculating lim
n
(3n
2
3n + 1)
1
n
= 1).
(3)
19
Proof. By induction, it is easy to show
_
(1 )
4

(n)
=
(n+3)!
3!
(1 )
(n+4)
. So for D(0, 1), (1 )
4
=

n=0
(n+1)(n+2)(n+3)
6

n
. Replace with z
5
, we get for z D(0, 1)
1
(1 z
5
)
4
=

n=0
(n + 1)(n + 2)(n + 3)
6
z
5n
.
Therefore
(1 z
5
)
4
=
z
20
(1 z
5
)
4
=

n=0
(n + 1)(n + 2)(n + 3)
6
z
5n+20
.
By Hadamards formula, the radius of convergence is 1.
(4)
Proof. By induction, we can show
_
(1 +z)
2

(n)
= (1)
n
(n + 1)!(1 +z)
(n+2)
. So z D(0, 1),
1
(1 +z)
2
=

n=0
(1)
n
(n + 1)z
n
.
It is also easy to verify that
1
(z + 1)
2
(z 1)
=
1
4(z + 1)
+
1
4(z 1)

1
2(z + 1)
2
=
1
4

n=0
(z)
n

1
4

n=0
z
n

1
2

n=0
(1)
n
(n + 1)z
n
.
Therefore
z
6
(z + 1)(z
2
1)
=
z
6
4

n=0
[(1)
n
+ 1 + 2 (1)
n
(n + 1)] z
n
=
1
4

n=0
[(1)
n
(2n + 3) + 1] z
n+6
.
By Hadamards formula, the radius of convergence is 1.
21. (1)
Proof. We note
1
2
_

|f(re
i
)|
2
d =
1
2
_

n=0
a
n
r
n
e
in
_

m=0
a
m
r
m
e
im
_
d
=
1
2

m,n=0
_

a
n
a
m
r
n+m
e
i(nm)
d
=

m,n=0
a
n
a
m
r
n+m

nm
=

n=0
|a
n
|
2
r
2n
,
where
mn
=
_
1 if m = n
0 if m = n
and the exchange of integration and summation is justied by Theorem 1.14
(Abel Theorem) and Problem 17.
(2)
20
Proof. In the result of part (1), multiply both sides by r and integrate with respect to r from 0 to 1, we have

n=0
|a
n
|
2
_
1
0
r
2n+1
dr =

n=0
|a
n
|
2
2n + 2
=
1
2
_
1
0
_

|f(re
i
)|
2
rdrd =
1
2
_ _
D
|f(z)|
2
dA.
So

n=0
|a
n
|
2
n+1
=
1

_ _
D
|f(z)|
2
dA.
22.
Proof. Suppose u is harmonic on D(0, R) and continuous

D(0, R), then z

D(0, 1), v(z) := u(zR) is
harmonic on D(0, 1) and continuous on

D(0, 1). By formula (2.31), for any z D(0, R),
u(z) = v(
z
R
) =
1
2
_
2
0
v(e
i
)
1

z
R

1
z
R
e
i

2
d =
1
2
_
2
0
u(Re
i
)
R
2
|z|
2
|R ze
i
|
2
d
=
1
2
_
2
0
u(Re
i
)
R
2
|z|
2
|R ze
i
|
2
d =
1
2
_
2
0
u()
R
2
|z|
2
| z|
2
d.
This is formula (2.32). To verify formula (2.33), note
Re
Re
i
+z
Re
i
z
= Re
(Re
i
+z)(Re
i
z)
|Re
i
z|
2
= Re
R
2
|z|
2
+R(ze
i
ze
i
)
|Re
i
z|
2
=
R
2
|z|
2
|Re
i
z|
2
.
Therefore,
u(z) =
1
2
_
2
0
u(Re
i
)
R
2
|z|
2
|Re
i
z|
2
d =
1
2
_
2
0
u(Re
i
)Re
Re
i
+z
Re
i
z
d = Re
_
1
2
_
2
0
u(Re
i
)
Re
i
+z
Re
i
z
d
_
= Re
_
1
2i
_
||=R
u()
+z
z
d

_
.
This is formula (2.33).
23.
Proof. On D(0, 1), |(6z)(z
4
6z +3)| = |z
4
+3| 4 < | 6z|. By Rouche Theorem, z
4
6z +3 and 6z
have the same number of zeros in D(0, 1), which is one. On D(0, 2), |z
4
(z
4
6z+3)| = |6z3| 15 < |z
4
|.
So z
4
6z + 3 and z
4
have the same number of zeros in D(0, 2), which is four. Combined, we conclude
z
4
6z + 3 = 0 has one root in D(0, 1) and three roots in the annulus {z : 1 < |z| < 2}.
24.
Proof. On D(0, 1), |(5z
4
) (z
7
5z
4
z +2)| = |z
7
z +2| 4 < | 5z
4
|. So z
7
5z
4
z +2 and 5z
4
have the same number of zeros in D(0, 1), which is four (counting multiplicity).
25.
Proof. Let P(z) = z
4
+ 2z
3
2z + 10. We note P(z) = (z
2
1)(z + 1)
2
+ 11. If z R and |z| 1,
P(z) 11; if z R and |z| 1, P(z) 1 (1 + 1)
2
+ 11 = 7. So P(z) = 0 has no root on the real
axis. If z = iy with y R, we have P(iy) = y
4
+ 10 2iy(y
2
+ 1) = 0. So P(z) = 0 has no root on the
imaginary axis. Consider the region D enclosed by the curve =
1

3
, where (R is a positive number)

1
= {z : 0 Rez R, Imz = 0},
2
= {z : |z| = R, arg z [0,

2
]}, and
3
= {z : 0 Imz R, Rez = 0}.
On
1
,

1
arg P(z) = 0. On
2
, P(z) = z
4
_
1 +
2z
3
2z+10
z
4
_
. So

2
arg P(z) = 4

2
+o(1) = 2 +o(1)
(R ). On
3
,

3
arg P(z) = arg P(0) arg P(iR) = arg 10 arg(R
4
+ 10 2iR(R
2
+ 1)) = 0
arg
_
1 2i
R(R
2
+1)
R
4
+10
_
= o(1) (R ). Combined, we have

arg P(z) =

3
k=1

k
arg P(z) = 2 + o(1)
(R ).
21
By The Argument Principle, P(z) = 0 has only one root in the rst quadrant. The root conjugate to
this root must lie in the fourth quadrant. The other two roots are conjugate to each other and are located in
the second and third quadrant. So one must lie in the second quadrant and the other must lie in the third
quadrant.
Remark 5. The above solution can be found in Fang [3], Chapter 6, 3 Example 5.
26.
Proof. On D(0, 1), |(8z +10) (z
4
8z +10)| = |z
4
| = 1 < |10 8|z|| | 8z +10|. So z
4
8z +10 and
8z + 10 have the same number of zeros in D(0, 1), which is zero.
On D(0, 3), |z
4
(z
4
8z +10)| = |8z 10| < 34 < |z
4
|. So z
4
8z +10 and z
4
have the same number
of zeros in D(0, 3), which is four (counting multiplicity). So z
4
8z +10 = 0 has no root in D(0, 1) and four
roots in the annulus {z : 1 < |z| < 3}.
27.
Proof. If a > e, on D(0, 1), |az
n
(az
n
e
z
)| = |e
z
| e
|z|
= e < |az
n
|. By Rouche Theorem, e
z
az
n
and
az
n
have the same number of zeros in D(0, 1), which is n.
28.
Proof. On D(0, 1), |z (z f(z))| = |f(z)| < 1 = |z|. So z f(z) and z have the same number of zeros in
D(0, 1), which is one. So there is a unique xed point of f in D(0, 1).
Remark 6. This is a special case of Browers Fixed Point Theorem.
29. (1)
Proof. Re(ze
z
) = e
x
(xcos y y siny).
(2)
Proof. Let u(x, y) = ax
3
+bx
2
y +cxy
2
+dy
3
. Then u = (6a + 2c)x + (2b + 6d)y. So u is harmonic if and
only if 3a+c = 0 and b +3d = 0. So the most general harmonic function of the form ax
3
+bx
2
y +cxy
2
+dy
3
is ax
3
3dx
2
y 3axy
2
+dy
3
.
30.
Proof. Let u(r, ) = ln(12r cos +r
2
). Then using the fact =
1
r

r
_
r

r
_
+
1
r
2

2
, we can verify u = 0.
So by the mean-value property of harmonic functions, we have
1
2
_
2
0
u(r, )d = u(0, 0) = 0. Since cos
is an even function of , its not hard to show
_
2
0
ln(1 2r cos + r
2
)d = 2
_

0
ln(1 2r cos + r
2
)d. So
_

0
ln(1 2r cos +r
2
)d = 0.
31.
Proof. Let M = sup
zC
|u(z)|. Then z C and R > |z|, the holomorphic function dened in formula (2.34)
satises
|f(z)|
1
2
_
2
0
|u(Re
i
)|

Re
i
+z
Re
i
z

d
M
2
_
2
0
R +|z|
R |z|
d M, as R .
So f(z) is bounded on the entire complex plane, and by Liouvilles Theorem, f is a constant function.
Therefore u = Re(f) is a constant function.
32.
22
Proof. If the arc is D(0, 1), the desired harmonic function u 1. So without loss of generality, we assume
the arc = {z D(0, 1) :
1
arg z
2
} with 0 <
2

1
< 2.
For n suciently large, we consider
n
= {z D(0, 1) :
1

1
n
arg z
2
+
1
n
}. By the Partition of
Unity Theorem for R
1
, we can nd
n
C(D(0, 1)) such that 0
n
1,
n
1 on , and
n
0 on
D(0, 1) \
n
. Dene u
n
(z) =
_
2
0
P(, z)
n
()d where P(, ) is the Poisson kernel and = Re
i
. Then u
n
is the unique solution of the Dirichlet problem with boundary value
n
. So
n
is harmonic in D(0, 1).
It is easy to see that if (
n
)
n
uniformly converges to the indicator function 1

(z), (u
n
)
n
uniformly
converges to u(z) :=
_
2
0
P(, z)1

()d on D(0, ) ( (0, 1)). Since each u

n
satises the local mean value
property, u must also satises the local mean value property in D(0, 1) and is continuous. By Theorem 2.22,
u is harmonic on D(0, 1). This suggests the harmonic function we are looking for is exactly u.
What is left to prove is that for any z
0
D(0, 1), lim
|z|=1,zz
0
u(z) = 1

(z
0
). Indeed, in general,
we have the following classical result: Suppose function is piecewise continuous on D(0, 1), then the
function u(z) :=
_
2
0
P(, z)()d is harmonic in D(0, 1) and for any continuity point z
0
of , we have
lim
zz
0
,|z|<1
u(z) = (z
0
).
For the proof of continuity on the boundary, we refer to Fang [3] Chapter 7, Theorem 7.
33.
Proof. > 0, there exists N, such that for any n N, p N, we have
sup
U

n+p

k=n+1
f
k
()

< .
By Maximum Modulus Principle,
sup
z

n+p

k=n+1
f
k
(z)

sup
U

n+p

k=n+1
f
k
()

< .
So Cauchy criterion implies

n=1
f
n
(z) converges uniformly on

U.
Remark 7. This result is the so-called Weierstrass Second Theorem.
34.
Proof. Dene g(z) =
f(z)f(z
0
)
zz
0
=
f(z)
zz
0
, z

D(0, R) \ {z
0
}. Then g(z) can be analytically continued to
D(0, R) (Theorem 2.11, Riemann Theorem) and is therefore continuous on

D(0, R). By Cauchys integral
formula, we have
f(0)
z
0
= g(0) =
1
2i
_
||=R
g()

d =
1
2i
_
||=R
f()
( z
0
)
d.
Therefore

f(0)
z
0

1
2
_
2
0
|f(Re
i
)
|Re
i
z
0
|
d
M
R |z
0
|
,
which is equivalent to R|f(0)| (M +|f(0)|)|z
0
|.
Remark 8. The point of this problem is to give an estimate of a holomorphic functions modulus at z = 0
in terms of its zeros.
35.
Proof. This problem seems suspicious. For example, when z
0
= 0, of course

n
k=1
|z
0
z
k
| =

n
k=1
|z
k
| > 1.
When z
0
is suciently large, it is also clear

n
k=1
|z
0
z
k
| > 1. So the point z
0
can be both inside and
outside D(0, 1). Im not sure what proof is needed for such a trivial claim.
23
36.
Proof. By Maximum Modulus Principle, M(r) = max
|z|r
|f(z)|. So M(r) is an increasing function on
[0, R).
37.
Proof. Suppose there is an n-th degree polynomial p(z) = a
0
+a
1
z + +a
n
z
n
such that n 1, a
n
= 0, and
p(z) = 0 for any z C. Then q(z) = 1/p(z) is holomorphic over the whole complex plane. For any R > 0,
we have by Maximum Modulus Principle
max
|z|R
|q(z)| max
|z|=R
|q(z)|.
Since on {z : |z| = R},
|p(z)| |a
n
|R
n
|a
n1
|R
n1
|a
1
|R|a
0
| = R
n
_
|a
n
|
|a
n1
|
R

|a
1
|
R
n1

|a
0
|
R
n
_
as R ,
we must have lim
R
max
|z|=R
|q(z)| = 0. This implies sup
zC
|q(z)| = 0, i.e. p(z) . This is a
contradiction and our assumption must be incorrect.
38.
Proof. Apply the Maximum Modulus Principle to the function g(z) = 1/f(z), which is holomorphic on
U.
39.
Proof. Note the Laplace operator = 4

, so log |z| is a harmonic function on C \ {0} and log |f(z)| is a

harmonic function outside the zeros of f(z). Dene K = {z U : f(z) = 0} and V

=
zK
D(z, ). Then
for any R, log |z| + log |f(z)| is harmonic in U \

V

and continuous on

U \ V

. By Maximum Modulus
Principle for harmonic functions, max
z

U\V

(log |z| + log |f(z)|) = max

z(

U\V

)
(log |z| + log |f(z)|).
When z approaches to zeros of f(z), log |f(z)| , so by letting 0, we can further deduce that
max
zU
(log |z| + log |f(z)|) = max
zU
(log |z| + log |f(z)|). Therefore
log |z| + log |f(z)| max{log r
1
+ log M(r
1
), log r
2
+ log M(r
2
)}, z U,
which is the same as
log r + log M(r) max{log r
1
+ log M(r
1
), log r
2
+ log M(r
2
)}, r [r
1
, r
2
].
Now let be such that the two values inside the parentheses on the right are equal, that is
=
log M(r
2
) log M(r
1
)
log r
1
log r
2
.
Then from the previous inequality, we get
log M(r) log r
1
+ log M(r
1
) log r,
which upon substituting value for gives
log M(r) (1 s) log M(r
1
) +s log M(r
2
),
where s =
log r
1
log r
log r
2
log r
1
.
40.
24
Proof. Fix r (0, 1). For any (0, r) and any z D(0, ), we have z
n
D(0, ) (n N). By Cachys
integral formula, n, p N, we have
n+p

k=n+1
|f(z
k
)| =
n+p

k=n+1
|f(z
k
) f(0)| =
n+p

k=n+1
1
2

_
D(0,r)
_
f()
z
k

f()

_
d

n+p

k=n+1
1
2

_
D(0,r)
f()
( z
k
)
d z
k

n+p

k=n+1
1
2
_
2
0
|f(re
i
)|d
r
k

k

1
2
_
2
0
|f(re
i
)|d
r
n+p

k=n+1

k
0
as n . So

n=1
f(z
n
) converges absolutely and uniformly on

D(0, ).
Remark 9. In general, there is no Mean Value Theorem for holomorphic functions. For example, f(z) =
exp
_
i
2z(z
1
+z
2
)
z
2
z
1

_
is analytic but f(z
2
) f(z
1
) = f

(z)(z
2
z
1
), z C (for more details, see Qazi [5]).
But as the proof of Theorem 2.7 shows, the formula
f(z) f(z
0
) =
z z
0
2i
_
U
f()d
( z)( z
0
)
more or less lls the void. In particular, it shows holomorphic functions are locally Lipschitz.
41.
Proof. This problem is the same as Problem 15.
42.
Proof. This problem is the same as Problem 19.
43. (1)
Proof. By Problem 18,

f(z)1
f(z)+1

|z|. So |f(z)| 1 |f(z)1| |z||f(z)+1| |z||f(z)| +|z|. Since |z| < 1,

the above inequality implies |f(z)|
1+|z|
1|z|
.
From the same inequality

f(z)1
f(z)+1

|z|, we have
(Ref(z) 1)
2
+ (Imf(z))
2
|z|
2
[(Ref(z) + 1)
2
+ (Imf(z))
2
] |z|
2
(Ref(z) + 1)
2
+ (Imf(z))
2
.
So |Ref(z) 1| |z|(Ref(z) +1). If Ref(z) > 1, we can deduce from this inequality
1|z|
1+|z|
1 < Ref(z). If
Ref(z) 1, we can deduce from this inequality 1 Ref(z) |z|Ref(z) +|z|, i.e.
1|z|
1+|z|
Ref(z).
(2)
Proof. We note f(z) =
1+e
i
z
1e
i
z
if and only if
f(z)1
f(z)+1
= e
i
. So the claim is straightforward from Schwarz
Lemma.
44.
25
Proof. Because D(0, 1) is pre-compact, by Theorem 1.11 (Bolzano-Weierstrass Theorem), it suces to nd a
sequence (z
n
)

n=1
D(0, 1), such that lim
n
z
n
= z
0
D(0, 1) and (f(z
n
))

n=1
is bounded. Assume this
does not hold. Then z
0
D(0, 1) and n N, there exists (z
0
) > 0 such that z D(z
0
, (z
0
)) D(0, 1),
|f(z)| n. The family of these open sets (D(z, (z))
zD(0,1)
is an open covering of the compact set
D(0, 1). By Theorem 1.10 (Heine-Borel Theorem), there is a nite sub-covering (D(z
k
, (z
k
))
p
k=1
of D(0, 1).
Consequently, for each n N, we can nd
n
> 0 such that z {z : 1
n
|z| < 1}, |f(z)| n. Without
loss of generality, we can assume (
n
)

n=1
monotonically decreases to 0.
Since f(z) uniformly approaches to as z D(0, 1), by Theorem 2.13, f has nitely many zeros in
D(0, 1). So f can be written as f(z) =

m
i=1
(z z
i
)
k
i
h(z), where h is a holomorphic function on D(0, 1)
with no zeros in D(0, 1). So h satises the Minimum Modulus Principle: > 0, |h(z)| min
||=
|h()| for
any z D(0, ). Therefore, for any z D(0, 1
n
),
|h(z)| min
||=1
n
|h()| = min
||=1
n
|f()|
|

m
i=1
( z
i
)
k
i
|

n
max
||=1
n

m
i=1
| z
i
|
k
i

n
2

m
i=1
k
i
.
Fix z D(0, 1
n
) and let n , we get h(z) = . Contradiction.
45.
Proof. By assumption, we can write f(z) as f(z) =

n
j=1
(z z
j
)
k
j
h(z), where h(z) is a holomorphic
function in D(0, 1) with no zeros in D(0, 1). Dene G(z) = h(z)

n
j=1
(1 z
j
z)
k
j
. Then G is holomorphic
on D(0, 1). Apply Maximum Modulus Principle to G(z) on {z : |z| 1 } ( > 0), we get
|G(z)| max
|z|=1
|h(z)|
n

j=1
|1 z
j
z|
k
j
= max
|z|=1
|f(z)|

n
j=1

zz
j
1 z
j
z

k
j
max
|z|=1
1

n
j=1

zz
j
1 z
j
z

k
j
.
Since each
zz
j
1 z
j
z
maps D(0, 1) to D(0, 1), by letting 0, we get G(z) 1, i.e. |f(z)|

n
j=1

zz
j
1 z
j
z

k
j
.
46.
Proof. Let a = f(0) and dene
a
() =
+a
1 a
. Then h(z) =
a
(f(z)) maps D(0, 1) to D(0, 1) with
h(0) =
a
(a) = 0. So Schwarz Lemma implies |h

(0)| 1. We note h

(z) =
a
(f(z)) f

(z). Since

a
() =
(1 a) ( +a)( a)
(1 a)
2
=
1 +|a|
2
(1 a)
2
,
we have h

(0) =

a
(a)f

(0) =
f

(0)
|a|
2
1
. So we have
|f

(0)|
1 |a|
2
1, i.e. |f

(0)| 1 |a|
2
.
48.
Proof. The Mobius transformation w(z) =
zi
z+i
maps C
+
to D = D(0, 1). So f Aut(C
+
) if and only if
w f w
1
Aut(D). By Theorem 2.20, a D and R, such that w f w
1
() =
a

(). Plain
calculation shows
f(z) = w
1

a

w(z) =
(1 +a)(z +i) (1 + a)e
i
(z i)
(1 a)(z +i) ( a 1)e
i
(z i)
.
50.
Proof. This problem is the same as Problem 10.
26
3 Theory of Series of Weierstrass
Throughout this chapter, all the integration paths will take the following convention on orientation: all the
arcs take counterclockwise as their orientation and all the segments take the natural orientation of R
1
as
their orientation.
1.
Proof. This is just Mittag-Leer Theorem (Theorem 3.8).
2.
Proof. To prove Theorem 3.1, choose varepsilon > 0 suciently small so that the circles {z : |z z
k
| = }
(1 k n) dont intersect with each other or with U. Then by Cauchys integration theorem,
_
U
f(z)dz
n

k=1
_
|zz
k
|=
f(z)dz = 0,
which is
_
U
f(z)dz = 2i

n
k=1
Res(f, z
k
).
To prove Theorem 3.12, we choose R > 0 suciently large so that

U D(0, R). Then
_
U
f(z)dz = 2i
n

k=1
Res(f, z
k
) and
_
D(0,R)
f(z)dz
_
U
f(z)dz = 0.
Since
_
D(0,R)
f(z)dz = 2iRes(f, ), we conclude

n
k=1
Res(f, z
k
) + Res(f, ) = 0.
3. (i)
Proof. Suppose
1
z
3
(z+i)
has Laurent series

n=
c
n
(z + i)
n
. Then by Theorem 3.2 and Cauchy integral
theorem, for (0, 1),
c
n
=
1
2i
_
|+i|=
1
( +i)
n+1

1

3
( +i)
d =
1
(n + 1)!
d
n+1
d
n+1
(
3
)

=i
=
1
(n + 1)!
(1)
n+1
(n + 3)!
2!
(i)
(n+4)
= (1)
n+1
(n + 2)(n + 3)
2
i
n
.
(ii)
Proof. We note
z
2
(z + 1)(z + 2)
=
(z + 1)(z + 2) 3z 2
(z + 1)(z + 2)
= 1
3(z + 1) 1
(z + 1)(z + 2)
= 1 +
1
z + 1

4
z + 2
= 1 +
1
z

1
1 +
1
z
2
1
1 +
z
2
=

k=0
(1)
k
z
k+1
1

k=1
(1)
k
2
k1
z
k
.
(iii)
Proof. We note
log
_
z a
z b
_
= log
_
1
a
z
1
b
z
_
= log
_
1
a
z
_
log
_
1
b
z
_
=

n=1
1
n
_
a
z
_
n
+

n=1
1
n
_
b
z
_
n
=

n=1
1
n
(b
n
a
n
)z
n
.
27
(iv)
Proof. We note
z
2
e
1
z
= z
2

k=0
z
k
k!
= z
2
+z +
1
2
+

k=1
z
k
(k + 2)!
.
(v)
Proof. We note sinz =

k=0
(1)
k z
2k+1
(2k+1)!
. So sin
z
z+1
=

k=0
(1)
k
(2k+1)!
z
2k+1
(z+1)
2k+1
. Suppose sin
z
z+1
has Laurent
series

n=
c
n
(z + 1)
n
. Then by Theorem 3.2, for > 0,
c
n
=
1
2i
_
|+1|=
1
( + 1)
n+1

k=0
(1)
k
(2k + 1)!

2k+1
( + 1)
2k+1
d
=

k=0
(1)
k
(2k + 1)!

1
2i
_
|+1|=

2k+1
( + 1)
2k+n+2
d
=

k
n1
2
(1)
k
(2k + 1)!
1
(2k +n + 1)!
d
2k+n+1
d
2k+n+1
_

2k+1
_

=1
.
Since
d
2k+n+1
d
2k+n+1
_

2k+1
_

=1
=
_

_
0 n 1,
(2k + 1)! n = 0,
(2k+1)!
(n)!
(1)
n
n 1,
we have
c
n
=
_

_
0 n 1,

k0
(1)
k
(2k+1)!
n = 0,
(1)
n
(n)!

k
n1
2
(1)
k
(2k+n+1)!
n 1
=
_

_
0 n 1,
sin1 n = 0,
1
(n)!
cos 1 n 1, n is odd,
1
(n)!
sin1 n 1, n is even.
Therefore
sin
z
z + 1
=

k=0
_
sin1
(2k)!
(z + 1)
2k
+
cos 1
(2k + 1)!
(z + 1)
(2k+1)
_
.
4. (i)
Proof. z = 0 is a removable singularity.
(ii)
Proof. z = 1 is a pole of order 1. The Laurent series of the function in a neighborhood of z = 1 can be
obtained by
1
z
2
1
cos
z
z + 1
=
1
(z 1)(z + 1)

n=0
(1)
n
(2n)!
_
z
z + 1
_
2n
.
By discussion (3) of page 91, z = 1 is an essential singularity.
28
(iii)
Proof. Since e
1
z
=

n=0
z
n
, we have z(e
1
z
1) =

n=1
z
n+1
. Therefore z = 0 is an essential singularity.
(iv)
Proof. z = 1 is an essential singularity.
(v)
Proof. z = 1 is an essential singularity. z = 0 is a pole of order 1.
(vi)
Proof. z = k +

2
(k Z) are poles of order 1.
5. (1)
Proof. The limit lim
za
f(z) exists (nite or ) if and only if lim
za
1
f(z)
exists. By discussion (3) on page
93, a is an essential singularity of f(z) if and only if a is an essential singularity of
1
f(z)
.
(2)
Proof. Clearly, a is an isolated singularity of P(f(z)). We note by Fundamental Theorem of Algebra, P()
maps C to C. Consequently, P() maps a dense subset of C to a dense subset of C. By Weierstrass
Theorem (Theorem 3.3), f(z) maps a neighborhood of a to a dense subset of C. Therefore, P(f(z)) maps a
neighborhood of a to a dense subset of C. So it is impossible for a to become a removable singularity or a
pole of P(f()). Hence a must be also an essential singularity of P(f(z)).
6. (i)
Proof. We note any positive integer can be uniquely represented in the form

k=0
a
k
2
k
, where each
a
k
{0, 1} and only nitely many a
k
s are non-zero. In the expansion of

n=0
(1+z
2
n
), each representation
z

k=0
a
k
2
k
appears once and only once. Therefore, after a rearrangement of the terms, we must have

n=0
(1 +z
2
n
) = 1 +z +z
2
+z
3
+z
4
+ =
1
1 z
.
(ii)
Proof. sinhz =
1
2
[e
z
e
z
] has zeros a
n
= ni (n Z), where a
0
= 0 is a zero of order 1. Since for any
R > 0, we have

n=,n=0
_
R
|a
n
|
_
2
=

n=,n=0
R
2
n
2
< , by Weierstrass Factorization Theorem, we can
nd an entire function h(z) such that
sinhz = ze
h(z)
n=

n=,n=0
__
1
z
ni
_
e
z
ni
_
= ze
h(z)

n=1
_
1 +
z
2
n
2
_
.
Since lim
z0
sinh z
z
= , we conclude e
h(0)
= . Meanwhile we have
cothz =
(sinhz)

sinhz
=
1
sinh z
_
ze
h(z)

n=1
_
1 +
z
2
n
2
_
_

=
1
z
+h

(z) +

n=1
2z
n
2
1 +
z
2
n
2
=
1
z
+h

(z) +

n=1
2z
n
2
+z
2
.
29
Let
n
be the rectangular path [n +(n +
1
2
)i, n +(n +
1
2
)i, n (n +
1
2
)i, n (n +
1
2
)i, n +(n +
1
2
)i]. Then
for any given a, when n is large enough, we have
_

n
cothz
z
2
a
2
dz = I +II +III +IV,
where
I =
_
n
n
e
2[x+(n+
1
2
)i]
+1
e
2[x+(n+
1
2
)i]
1
_
x + (n +
1
2
)i

2
a
2
dx =
_
n
n
e
2x
+1
e
2x
1
_
x + (n +
1
2
)i

2
a
2
dx,
II =
_
(n+
1
2
)
n+
1
2
e
2(n+yi)
+1
e
2(n+yi)
1
(n +yi)
2
a
2
idy,
III =
_
n
n
e
2[x(n+
1
2
)i]
+1
e
2[x(n+
1
2
)i]
1
[x (n +
1
2
)i]
2
a
2
dx =
_
n
n
e
2x
+1
e
2x
1
[x (n +
1
2
)i]
2
a
2
dx
and
IV =
_
n+
1
2
(n+
1
2
)
e
2(n+yi)
+1
e
2(n+yi)
1
(n +iy)
2
a
2
idy.
We note
e
x
1
e
x
+1
(1, 1) for any x R. So we have
|I|
_
n
n
1
x
2
+ (n +
1
2
)
2
a
2
dx =
2
_
(n +
1
2
)
2
a
2
arctan
n
_
(n +
1
2
)
2
a
2
0,
as n . Similarly we can conclude lim
n
III = 0. We also note for x R large enough,
e
x
+1
e
x
1
< 2. So
for n large enough,
|II|
_
n+
1
2
(n+
1
2
)
e
1+e
2n
1e
2n
n
2
+y
2
a
2
dy
4

n
2
a
2
arctan
n +
1
2

n
2
a
2
0,
as n . Similarly we can conclude lim
n
IV = 0. Combined, we have shown
lim
n
_

n
cothz
z
2
a
2
dz = 0.
Dene f(z) =
coth z
z
2
a
2
. By Residue Theorem, for a iZ, we have
_

n
f(z)dz = 2iRes(f; a) + 2iRes(f; a) + 2i
n

k=n
Res(f; a
n
).
Its easy to see Res(f; a) = Res(f; a) =
coth a
2a
. Since
lim
za
n
(z a
n
)f(z) = lim
za
n
e
2z
+ 1
z
2
a
2
z a
n
e
2z
1
=
2
2(a
2
n
a
2
)
lim
za
n
2(z a
n
)
e
2z
e
2a
n
=
1
(n
2
+a
2
)
,
we must have Res(f; a
n
) =
1
(n
2
+a
2
)
. Therefore
0 =
1
2i
lim
n
_

n
f(z)dz = lim
n
_
cotha
a

n

k=n
1
(a
2
+k
2
)
_
=
cotha
a

k=
1
(a
2
+k
2
)
This implies for a iZ,
coth a =
1
a
+

k=1
2a
a
2
+n
2
.
30
Plugging this back into the formula cothz =
1
z
+ h

(z) +

n=1
2z
n
2
+z
2
, we conclude h

(z) = 0 for z iZ.

By Theorem 2.13, h

(z) 0 for any z C. So

sinh z = ze
h(0)

n=1
_
1 +
z
2
n
2
_
= z

n=1
_
1 +
z
2
n
2
_
.
Remark 10. The above solution is a variant of the proof for factorization formula of sinz. See Conway[2]
VII, 6 for more details.
(iii)
Proof. From problem (v), we have

2

n=1
_
1
1
4n
2
_
= 1 and
cos z = sin
_
1
2
z
_
=
_
1
2
z
_

n=1
_
1
_
1
2
z
_
2
n
2
_
=
_
1
2
z
_

n=1
__
1 +
1
2
z
n
__
1
1
2
z
n
__
=
_
1
2
z
_

n=1
__
1 +
1
2n
__
1
z
n +
1
2
__
1
1
2n
__
1 +
z
n
1
2
__
= lim
N

2
_
1
z
1
2
_
N

n=1
_
1
1
4n
2
_

n=1
_
1
z
n +
1
2
_

n=1
_
1 +
z
n
1
2
_
= lim
N
_

2
N

n=1
_
1
1
4n
2
_
_

_
1
z
N +
1
2
_

n=1
_
1
z
2
(n
1
2
)
2
_
=

n=0
_
1
_
z
n +
1
2
_
2
_
.
(iv)
Proof. The function e
z
1 has zeros a
n
= 2ni (n Z), where a
0
= 0 is a zero of order 1. Since for any
R > 0, we have

n=,n=0
_
R
|a
n
|
_
2
=

n=,n=0
R
2
4n
2
< , by Weierstrass Factorization Theorem, we can
nd an entire function h(z) such that
e
z
1 = ze
h(z)

n=,n=0
__
1
z
2ni
_
e
z
2ni
_
= ze
h(z)

n=1
_
1 +
z
2
4n
2
_
.
To determine h(z), we note
e
z
e
z
1
=
(e
z
1)

e
z
1
=
1
z
+h

(z) +

n=1
2z
4n
2
1 +
z
2
4n
2
=
1
z
+h

(z) +

n=1
2z
4n
2
+z
2
.
Let
n
be the rectangular path [2n+(2n+1)i, 2n+(2n+1)i, 2n(2n+1)i, 2n(2n+1)i, 2n+(2n+1)i].
Dene f(z) =
e
z
(z
2
a
2
)(e
z
1)
. Then for any given a, when n is large enough, we have
_

n
f(z)dz = I +II +III +IV,
31
where
I =
_
2n
2n
f(x + (2n + 1)i)dx =
_
2n
2n
e
e
x
[(x + (2n + 1)i)
2
a
2
](e
x
+ 1)
dx,
II =
_
(2n+1)
2n+1
f(2n +yi)idy =
_
(2n+1)
2n+1
e
(2n+yi)
[(2n +yi)
2
a
2
](e
(2n+yi)
1)
idy,
III =
_
2n
2n
f(x (2n + 1)i)dx =
_
2n
2n
e
e
x
[(x (2n + 1)i)
2
a
2
](e
x
+ 1)
dx,
and
IV =
_
2n+1
(2n+1)
f(2n +yi)idy =
_
2n+1
(2n+1)
e
(2n+yi)
[(2n +yi)
2
a
2
](e
(2n+yi)
1)
idy.
It is easy to see
|I|
_
2n
2n
dx
x
2
+ (2n + 1)
2
a
2

2
_
(2n + 1)
2
a
2
arctan
2n
_
(2n + 1)
2
a
2
0
as n . Similarly, we can show lim
n
III = 0. Also, we note
|II|
_
2n+1
(2n+1)
dy
(4n
2
+y
2
a
2
)(e
2n
1)

2
(e
2n
1)

4n
2
a
2
arctan
2n + 1

4n
2
a
2
0,
as n . For n suciently large,
e
2n
e
2n
1
2, so
|IV |
_
2n+1
(2n+1)
e
2n
dy
(4n
2
+y
2
a
2
)(e
2n
1)

4

4n
2
a
2
arctan
2n + 1

4n
2
a
2
0,
as n . Combined, we have shown
lim
n
_

n
f(z)dz = 0.
By Residue Theorem, for a 2Zi, we have
_

n
f(z)dz = 2iRes(f; a) + 2iRes(f; a) + 2i
n

k=n
Res(f; a
n
).
Its easy to see Res(f; a) =
e
a
2a(e
a
1)
and Res(f; a) =

2a(e
a
1)
. Since
lim
za
n
(z a
n
)f(z) = lim
za
n
e
z
(z
2
a
2
)

z a
n
e
z
1
=
1
4n
2
+a
2
,
Res(f; a
n
) =
1
4n
2
+a
2
. Therefore
0 =
1
2i
lim
n
_

n
f(z)dz = lim
n
_

2a
e
a
+ 1
e
a
1

n

k=n
1
4n
2
+a
2
_
.
So

2a
e
a
+1
e
a
1
=
1
a
2
+

n=1
2
4n
2
+a
2
for a 2Zi. Therefore for z 2Zi, we have
h

(z) =
e
z
e
z
1

1
z

n=1
2z
4n
2
+z
2
=
e
z
e
z
1


2
e
z
+ 1
e
z
1
=

2
.
So h(z) =
z
2
+h(0). Its easy to see lim
z0
e
z
1
z
= . So e
h(0)
= . Combined, we conclude e
h(z)
= e
z
2
.
So
e
z
1 = ze
z
2

n=1
_
1 +
z
2
4n
2
_
.
32
Equivalently, we have
e
z
1 = ze
z
2

n=1
_
1 +
z
2
4
2
n
2
_
.
(v)
Proof. Using the formula sinz = i sinh(iz) and the factorization formula for sinhz, we have
sinz = i(iz)

n=1
_
1
z
2
n
2
_
= z

n=1
_
1
z
2
n
2
_
.
7. For a clear presentation of convergence of innite products, we refer to Conway [2], Chapter VII, 5. In
particular, we quote the following theorem (Conway [2], Chapter VII, 5, Theorem 5.9)
Let G be an open subset of the complex plane. Denote by H(G) the collection of analytic functions on
G, equipped with the topology determined by uniform convergence on compact subsets of G. Then H(G) is
a complete metric space. Furthermore, we have the following theorem (proof omitted).
Theorem 11. Let G be a region in C and let (f
n
)

n=1
be a sequence in H(G) such that no f
n
is identically
zero. If

[f
n
(z)1] converges absolutely and uniformly on compact subsets of G, then

n=1
f
n
(z) converges
in H(G) to an analytic function f(z). If a is a zero of f then a is a zero of only a nite number of the
functions f
n
, and the multiplicity of the zero of f at a is the sum of the multiplicities of the zeros of the
functions f
n
at a.
Proof. (Proof of Blaschke Product) Fix r (0, 1). Since

k=1
(1 |a
k
|) < , lim
k
|a
k
| = 1. So there
exists k
0
N, such that for any k k
0
,
1+r
2
< |a
k
| < 1. So for any k k
0
,

a
k
z
1 a
k
z

|a
k
|
a
k
1

|a
k
|(1 |a
k
|) +|z||a
k
|(1 |a
k
|)
|1 a
k
z||a
k
|

(1 +r)(1 |a
k
|)
1+r
2
(1 r)
=
2
1 r
(1 |a
k
|).
Since

k=1
(1 |a
k
|) < , we conclude

k=1

a
k
z
1 a
k
z

|a
k
|
a
k
1

is absolutely and uniformly convergent on

{z : |z| r}. Therefore

k=1
a
k
z
1 a
k
z

|a
k
|
a
k
is absolutely and uniformly convergent on {z : |z| r}. By
Weierstrass Theorem (Theorem 3.1), f(z) =

k=1
a
k
z
1 a
k
z

|a
k
|
a
k
represents a non-zero holomorphic function on
{z : |z| < 1}. Since the mapping z
a
k
z
1 a
k
z
(0 < |a
k
| < 1) maps D(0, 1) to D(0, 1), we conclude |f(z)| 1.
By the theorem quoted at the beginning of the solution, its clear that (a
k
)

k=1
are the only zeros of f(z).
8.
Proof. Let R

= R , (0, R). Dene

F

(z) = f(z)

t
j=1
R

(zb
j
)
R
2

b
j
z

s
i=1
R

(za
i
)
R
2

a
i
z
, z D(0, R

).
Its easy to verify that each of
R
2

(zb
j
)
R
2

b
j
z
and
R
2

(za
i
)
R
2

a
i
z
maps D(0, R

) onto itself and takes the boundary to

the boundary. Therefore F

(z) is analytic on D(0, R

), has no zeros in D(0, R

), and |F

(z)| = |f(z)| for

|z| = R

. Since log |F

(z)| is a harmonic function, by formula (2.33) on page 71,

log |F

(z)| =
1
2
_
2
0
log |F

(R

e
i
)|Re
R

e
i
+z
R

e
i
z
d =
1
2
_
2
0
log |f(R

e
i
)|Re
R

e
i
+z
R

e
i
z
d.
33
Plugging the formula of F

into the above equality, we get

log |f(z)| =
1
2
_
2
0
log |f(R

e
i
)|Re
R

e
i
+z
R

e
i
z
d +
t

j=1
log

R
2

b
j
z
R

(z b
j
)

i=1
log

R
2

a
i
z
R

(z a
i
)

.
Letting 0 yields the desired formula.
9.
Proof. By Theorem 3.4 (correction: holomorphic function f(z) in the theorems statement should be
meromorphic function f(z)), f(z) has the form of
1
z+1
+
2
z2
+
3
(z2)
2
+c, where c is a constant. f(0) =
7
4
implies c = 1. Since for any D(0, 1),
1
(1 )
2
=
_
1
1
_

=
_

n=0

n
_

n=0
(n + 1)
n
,
the Laurent expansion of f(z) in 1 < |z| < 2 is
f(z) =
1
z + 1
+
2
z 2
+
3
(z 2)
2
+ 1
=
1
z
1
1 +
1
z

1
1
z
2
+
3
4

1
(1
z
2
)
2
+ 1
=
1
z

n=0
_

1
z
_
n

n=0
_
z
2
_
n
+
3
4

n=0
(n + 1)
_
z
2
_
n
+ 1
=

n=1
(1)
n
z
n
+
3
4
+

n=1
3n 1
2
n+2
z
n
.
10.
Proof. We follow the construction outlined in the proof of Mittag-Leer Theorem (Theorem 3.9). For each
n N, when |z| <
n
2
,
n
(z) has Taylor expansion

n
(z) =
n
(z n)
2
=
1
n

1
_
1
z
n
_
2
=
1
n

k=0
(k + 1)
_
z
n
_
k
.
Let
n
be a positive integer to be determined later. We estimate the tail error obtained by retaining only
the rst
n
terms of the Taylor expansion of
n
(z):

n
(z)
1
n

k=0
(k + 1)
_
z
n
_
k

1
n

k=
n
+1
(k + 1)
2
k
=
1
n

k=
n
(k + 2)
2
k+1
=
1
n
_

k=
n
k
2
k+1
+
1
2

n
1
_

1
n 2

n
1
+
1
n

k=
n
_
k+1
k
x
2
x
dx =
1
n 2

n
1
+
1
n
_

n
x
log
1
2
d
_
1
2
_
x
=
1
n 2

n
1
+
1
n 2

n
_

n
log 2
+
1
(log 2)
2
_
.
If we let
n
= n, then
n
:=
1
n2

n
1
+
1
n2

n
_

n
log 2
+
1
(log 2)
2
_
=
1
n2
n1
+
1
n2
n
_
n
log 2
+
1
(log 2)
2
_
satises

n=1

n
< . By the proof of Mittag-Leer Theorem, we can write f(z) as
f(z) = U(z) +

n=1
_
n
(z n)
2

1
n
n

k=0
(k + 1)
_
z
n
_
k
_
,
where U(z) is an entire function.
34
11. (i)
Proof. Dene f(z) =
1
(z
2
a
2
)(e
z
1)
, where a C\2iZ. Let
n
be the rectangular path [2n+(2n+1)i, 2n+
(2n +1)i, 2n (2n +1)i, 2n (2n +1)i, 2n +(2n +1)i]. Then for any given a, when n is large enough, we
have
_

n
f(z)dz = I +II +III +IV,
where
I =
_
2n
2n
f(x + (2n + 1)i)dx =
_
2n
2n
dx
[(x + (2n + 1)i)
2
a
2
](e
x
+ 1)
,
II =
_
(2n+1)
2n+1
f(2n +yi)idy =
_
(2n+1)
2n+1
idy
[(2n +yi)
2
a
2
](e
(2n+yi
1)
,
III =
_
2n
2n
f(x (2n + 1)i)dx =
_
2n
2n
dx
[(x (2n + 1)i)
2
a
2
](e
x
+ 1)
,
and
IV =
_
2n+1
(2n+1)
f(2n +iy)idy =
_
2n+1
(2n+1)
idy
[(2n +iy)
2
a
2
](e
(2n+iy)
1)
.
Its easy to see
|I|
_
2n
2n
dx
x
2
+ (2n + 1)
2
a
2

2
_
(2n + 1)
2
a
2
arctan
2n
_
(2n + 1)
2
a
2
0
as n . Similarly, we can show lim
n
III = 0. Also, we note
|II|
_
2n+1
(2n+1)
dy
(4n
2
+y
2
a
2
)(1 e
2n
)
=
2
(1 e
2n
)

4n
2
a
2
arctan
2n + 1

4n
2
a
2
0
as n , and
|IV |
_
2n+1
(2n+1)
dy
(4n
2
+y
2
a
2
)(e
2n
1)
=
2
(e
2n
1)

4n
2
a
2
arctan
2n + 1
4n
2
a
2
0
as n . Combined, we have shown lim
n
_

n
f(z)dz = 0. By Residue Theorem, for a 2iZ, we have
_

n
f(z)dz = 2iRes(f; a) + 2iRes(f; a) + 2i
n

k=n
Res(f; 2ni).
Its easy to see Res(f; a) =
1
2a(e
a
1)
and Res(f; a) =
1
2a(1e
a
)
. Since
lim
z2ni
(z 2ni)f(z) = lim
z2ni
(z 2ni)
(z
2
a
2
)(e
z
1)
=
1
(4n
2
+a
2
)
,
we have Res(f; 2ni) =
1
(4n
2
+a
2
)
. Therefore
0 =
1
2i
lim
n
_

n
f(z)dz = lim
n
_
1
2a

e
a
+ 1
e
a
1

1

k=n
1
4k
2
+a
2
_
=
1
2a

e
a
+ 1
e
a
1

1

k=
1
4k
2
+a
2
.
Replacing a with z, we get after simplication
1
e
z
1
=
1
z

1
2
+

k=1
2z
4k
2

2
+z
2
, z C \ 2Zi.
This is the partial fraction of 1/(e
z
1).
35
(ii)
Proof. Dene f(z) =
1
(za) sin
2
(z)
, where a C\Z. Let
n
be the rectangular path [
_
n +
1
2
_
+ni,
_
n +
1
2
_
+
ni,
_
n +
1
2
_
ni,
_
n +
1
2
_
ni,
_
n +
1
2
_
+ni]. Then when n is large enough, we have
_

n
f(z)dz = I +II +III +IV,
where
I =
_
(n+
1
2
)
n+
1
2
dx
(x +ni a) sin
2
[(x +ni)]
=
_
(n+
1
2
)
(n+
1
2
)
4dx
(x +ni a)(e
2n+2ix
+e
2n2ix
2)
,
II =
_
n
n
idy
[(n +
1
2
) +yi a] sin
2
[(n +
1
2
) +yi]
=
_
n
n
4idy
[(n +
1
2
) a +yi](e
2y
+e
2y
+ 2)
,
III =
_
(n+
1
2
)
(n+
1
2
)
dx
(x ni a) sin
2
[(x ni)]
=
_
(n+
1
2
)
(n+
1
2
)
4dx
(x a ni)(e
2n+2ix
+e
2n2ix
2)
,
and
IV =
_
n
n
idy
[(n +
1
2
) +yi a] sin
2
[(n +
1
2
) +yi]
=
_
n
n
4idy
[(n +
1
2
) a +yi](e
2y
+e
2y
+ 2)
.
Its easy to see
|I|
_
(n+
1
2
)
(n+
1
2
)
4dx
_
(x a)
2
+n
2
(e
2n
e
2n
2)

4(2n + 1)
n(e
2n
e
2n
2)
0
as n ,
|II|
_
n
n
4dy
_
(n +
1
2
+a)
2
+y
2
(e
2y
+e
2y
+ 2)
=
_
n
0
8dy
_
(n +
1
2
+a)
2
+y
2
(e
2y
+e
2y
+ 2)

_
n
0
8dy
(n +
1
2
+a)e
2y
=
8
(n +
1
2
+a)
1 e
2n
2
0
as n ,
|III|
_
(n+
1
2
)
(n+
1
2
)
4dx
_
(x a)
2
+n
2
(e
2n
e
2n
2)

4(2n + 1)
n(e
2n
e
2n
2)
0
as n , and lastly
|IV |
_
n
n
4dy
_
(n +
1
2
a)
2
+y
2
(e
2y
+e
2y
+ 2)
=
_
n
0
8dy
_
(n +
1
2
a)
2
+y
2
(e
2y
+e
2y
+ 2)

_
n
0
8
(n +
1
2
a)e
2y
=
8
(n +
1
2
a)
1 e
2n
2
0
as n . Combined, we have shown lim
n
_

n
f(z)dz = 0. By Residue Theorem, we have
_

n
f(z)dz = 2iRes(f; a) + 2i
n

k=n
Res(f; k).
36
Its easy to see Res(f; a) =
1
sin
2
(a)
. Dene h
k
(z) =
(zk)
2
sin
2
(z)
. Since lim
zk
h
k
(z) =
1

2
, h
k
is holomorphic in
a neighborhood of k and for > 0 small enough, and we have
Res(f; k) =
1
2i
_
|zk|=
dz
(z a) sin
2
(z)
=
1
2i
_
|zk|=
h
k
(z)
(z a)

dz
(z k)
2
=
d
dz
_
h
k
(z)
z a
_

z=k
.
We note
d
dz
_
h
k
(z)
z a
_

z=k
= lim
zk
h

k
(z)(z a) h(z)
(z a)
2
= lim
zk
h

k
(z)
z a

1
(k a)
2

2
,
and
lim
zk
h

k
(z)
z a
= lim
zk
2(z k) sin(z) 2(z k)
2
cos(z)
sin
3
(z)
= lim
zk
2 sin(z) + 2(z k) cos(z) 4(z k) cos(z) + 2
2
(z k)
2
sin(z)
3 sin
2
(z) cos(z)
= lim
zk
2 sin(z) 2(z k) cos(z)
3 sin
2
(z)
= lim
zk
2 cos(z) 2 cos(z) + 2
2
(z k) sin(z)
6
2
sin(z) cos(z)
= 0.
So for a C \ Z,
_

n
f(z)dz =
2i
sin
2
(a)

n

k=n
2i
(k a)
2

2
.
Let n , we get

2
sin
2
(z)
=

n=
1
(z n)
2
.
Remark 12. Another choice is to let f(z) =
cot(z)
(z+a)
2
, see Conway [2], page 122, Exercise 6. The trick used in
this problem and problem 6 (ii) will be generalized in problem 12.
(iii)
Proof. By the solution of problem 6 (ii), we have
coth z =
1
z
+

n=1
2z
z
2
+
2
n
2
.
Since cothz = i cot(iz), we have
cot z = i coth(iz) =
1
z

n=1
2z

2
n
2
z
2
=
1
z

n=1
_
1
n z

1
n +z
_
.
Therefore

cot

=
1

n=1
_
1
n

1
n +
_
= lim
N
_
1

n=1
_
1
n

1
n +
_
_
= lim
N
_
1

_
1

+
1
2
+ +
1
N
_
+
_
1
+
+
1
2 +
+ +
1
N +
__
= lim
N
__
1

1

_
+
_
1
+

1
2
_
+ +
_
1
(N 1) +

1
N
_
+
1
N +
_
=

n=0
_
1
n +

1
n + ( )
_
=

n=0
2
(n +)[n + ( )]
.
37
By letting = 3 and = 1, we get

3
=

3
cot

3
=

n=1
1
(3n 2)(3n 1)
.
Remark 13. For a direct proof without using problem 6 (ii), we can let f(z) =
cot(z)
z
2
a
2
and apply Residue
Theorem to it. For details, see Conway [2], page 122, Exercise 8. The line of reasoning used in this approach
will be generalized in problem 12 below.
12. (1)
Proof. Since z = is a zero of f(z) with multiplicity p, f(z) can be written as
h(z)
z
p
where h is holomorphic
in a neighborhood of . Consequently, h is bounded in a neighborhood of . In the below, we shall work
with a sequence of neighborhoods of that shrinks to . So without loss of generality, we can assume h
is bounded and denote its bounded by M.
Let
n
be the rectangular path [(n +
1
2
) + ni, (n +
1
2
) + ni, (n +
1
2
) ni, (n +
1
2
) ni, (n +
1
2
) + ni].
Then
_

n
f(z) cot(z)dz = I +II +III +IV,
where
I =
_
(n+
1
2
)
(n+
1
2
)
h(x +ni)
(x +ni)
p
e
i(x+ni)
+e
i(x+ni)
2
e
i(x+ni)
e
i(x+ni)
2i
dx =
_
(n+
1
2
)
(n+
1
2
)
h(x +ni)
(x +ni)
p
e
ixn
+e
ix+n
e
ixn
e
ix+n
idx,
II =
_
n
n
h((n +
1
2
) +yi)
[(n +
1
2
) +yi]
p
e
i[(n+
1
2
)+yi]
+e
i[(n+
1
2
)+yi]
2
e
i[(n+
1
2
)+yi]
e
i[(n+
1
2
)+yi]
2i
idy
=
_
n
n
h((n +
1
2
) +yi)
[(n +
1
2
) +yi]
p
(1)
n
e
y
(i) + (1)
n
e
y
i
(1)
n
e
y
(i) (1)
n
e
y
i
(1)dy,
III =
_
(n+
1
2
)
(n+
1
2
)
h(x ni)
(x ni)
p
e
ix+n
+e
ixn
e
ix+n
e
ixn
idx,
and
IV =
_
n
n
h((n +
1
2
) +yi)
[(n +
1
2
) +yi]
p
(1)
n
e
y
i + (1)
n
e
y
(i)
(1)
n
e
y
i (1)
n
e
y
(i)
(1)dy.
We note
|I|
_
(n+
1
2
)
(n+
1
2
)
M
(x
2
+n
2
)
p
2
e
n
+e
n
e
n
e
n
dx =
e
2n
+ 1
e
2n
1
2M
n
arctan
n +
1
2
n
0
as n ,
|II|
_
n
n
M
_
(n +
1
2
)
2
+y
2

p
2

e
y
e
y
e
y
+e
y

dy
2M
n +
1
2
arctan
n
n +
1
2
0
as n ,
|III|
_
(n+
1
2
)
(n+
1
2
)
M
(x
2
+n
2
)
p
2
e
n
+e
n
e
n
e
n
dx =
e
n
+e
n
e
n
e
n
2M
n
arctan
n +
1
2
n
0
as n , and
|IV |
_
n
n
M
_
(n +
1
2
)
2
+y
2

p
2

e
y
e
y
e
y
+e
y

dy
2M
n +
1
2
arctan
n
n +
1
2
0
38
as n . Combined, we can conclude lim
n
_

n
f(z) cot(z)dz = 0. Meanwhile, for n suciently large,
by Residue Theorem, we have
_

n
f(z) cot(z)dz = 2i
m

k=1
Res(f(z) cot(z),
k
) + 2i
n

k=n
Res(f(z) cot(z), k).
We note
Res(f(z) cot(z), k) =
1
2i
_
|zk|=
f(z)
cos(z)
sin(z)
dz =
1
2i
_
|zk|=
(1)
k
f(z)
z k

cos(z)(z k)
sin[(z k)]
dz =
f(k)

.
So
_

n
f(z) cot(z)dz = 2i
m

k=1
Res(f(z) cot(z),
k
) + 2i
n

k=n
f(k)

.
Letting n gives us
lim
n
n

k=n
f(k) =
n

k=1
Res(f(z) cot(z),
k
).
(2)
Proof. As argued in the solution of (1), we can assume f(z) =
h(z)
z
p
where h is holomorphic in a neighborhood
of and has bound M in that neighborhood. Let
n
denote the same rectangular path as in our solution
of (1). Then
_

n
f(z)
sin(z)
dz = I +II +III +IV,
where
I =
_
(n+
1
2
)
n+
1
2
h(x +ni)
(x +ni)
p
2idx
e
ixn
e
ix+n
,
II =
_
n
n
h((n +
1
2
) +yi)
[(n +
1
2
) +yi]
p
2i
e
y
(1)
n
(i) e
y
(1)
n
i
idy,
III =
_
n+
1
2
(n+
1
2
)
h(x ni)
(x ni)
p
2idx
e
ix+n
e
ixn
,
and
IV =
_
n
n
h((n +
1
2
) +yi)
_
(n +
1
2
) +yi

p
2i
e
y
(1)
n
i e
y
(1)
n
(i)
idy.
We note
|I|
2M
e
n
e
n
_
n+
1
2
(n+
1
2
)
dx
(x
2
+n
2
)
p
2

4M
e
n
e
n
_
n+
1
2
0
dx
x
2
+n
2
=
2M
(e
n
e
n
)n
arctan
n +
1
2
n
0
as n , and
|II|
_
n
n
1
_
(n +
1
2
)
2
+y
2

p
2
2Mdy
e
y
+e
y

_
n
n
Mdy
(n +
1
2
)
2
+y
2
=
2M
n +
1
2
arctan
n
n +
1
2
0
as n . By similar argument, we can prove lim
n
III = lim
n
IV = 0. Combined, we can conclude
that lim
n
_

n
f(z)
sin(z)
dz = 0. Meanwhile, for n suciently large, by Residue Theorem, we have
_

n
f(z)
sin(z)
dz = 2i
m

k=1
Res
_
f(z)
sin(z)
,
k
_
+ 2i
n

k=n
Res
_
f(z)
sin(z)
, k
_
.
39
Its easy to see
Res
_
f(z)
sin(z)
, k
_
=
1
2i
_
|zk|=
(1)
k
z k

f(z)(z k)
sin[(z k)]
dz =
(1)
k
f(k)

.
Therefore,
_

n
f(z)
sin(z)
dz = 2i
m

k=1
Res
_
f(z)
sin(z)
,
k
_
+ 2i
n

k=n
(1)
k
f(k)

.
Let n , we get lim
n

n
k=n
(1)
k
f(k) =

m
k=1
Res
_
f(z)
sin(z)
,
k
_
.
(3) (i)
Proof. Let f(z) =
1
(z+a)
2
. Then by result of (1),

n=
1
(n +a)
2
= Res
_
cot(z)
(z +a)
2
, a
_
=
d
dz
[cot(z)]|
z=a
=

2
sin
2
(a)
.
This result is veried by problem 11 (ii).
(ii)
Proof. Let f(z) =
1
z
2
+a
2
. Then by result of (2),

n=
(1)
n
n
2
+a
2
=
_
Res
_
1
(z
2
+a
2
) sin(z)
, ai
_
+ Res
_
1
(z
2
+a
2
) sin(z)
, ai
__
=

ai sin(ai)
=
csch(a)
a
.
This result can be veried by Mathematica.
13. (i)
Proof. f(z) =
1
z
2
z
4
has poles 0, 1, and 1. is a removable singularity.
Res(f; 0) =
1
2i
_
|z|=
dz
z
2
(1 z
2
)
=
d
dz
_
1
1 z
2
_
|
z=0
=
(2z)
(1 z
2
)
2
|
z=0
= 0,
Res(f; 1) =
1
2i
_
|z1|=
dz
(1 z)(1 +z)z
2
=
1
(1 +z)z
2
|
z=1
=
1
2
,
and
Res(f; 1) =
1
2i
_
|z+1|=
dz
(1 +z)(1 z)z
2
=
1
2
.
(ii)
Proof. We note
f(z) :=
z
2
+z + 2
z(z
2
+ 1)
2
=
2
z
+
1 +i
4(z i)
2
+
4 i
4(z i)
+
1 i
4(z +i)
2
+
4 +i
4(z +i)
.
Therefore, the function f(z) has poles 0, i and i. is a removable singularity of f(z). Furthermore, we
have
Res(f; 0) = 2, Res(f; i) =
4 i
4
= 1
i
4
, Res(f; i) =
4 +i
4
= 1 +
i
4
.
40
(iii)
Proof. Suppose a
1
, , a
n
are the roots of the equation z
n
+ a
n
= 0. Then each a
i
is a pole of order 1 for
the function f(z) =
z
n1
z
n
+a
n
. is a removable singularity of f(z). Then
Res(f; a
i
) =
a
n1
i

n
j=1,j=i
(a
i
a
j
)
= lim
za
i
a
n1
i
(z a
i
)

n
j=1
(z a
j
)
= lim
za
i
a
n1
i
(z a
i
)
z
n
+a
n
=
1
n
.
(iv)
Proof. f(z) =
1
sin z
has Z as poles of order 1. is not an isolated singularity of f(z). And
Res(f, n) =
1
2i
_
|zn|=
dz
sinz
=
1
2i
_
|zn|=
1
z n
(1)
n
(z n)
sin(z n)
dz = (1)
n
.
(v)
Proof. First, we note as cos
_
1
z2
_
=
1
2
_
e
i
z2
+e

i
z2
_
=

n=0
(1)
n
(z2)
2n
. So f(z) := z
3
cos
_
1
z2
_
=

n=0
(1)
n
z
3
(z2)
2n
. This shows z = 2 is an essential singularity of f(z), and
Res(f; 2) =

n=0
(1)
n
2i
_
|z2|=
z
3
(z 2)
2n
dz = (1) 3z
2
|
z=2
+ (1)
2

1
3!
6|
z=2
= 11.
Also, is an isolated singularity of f(z), and we have
Res(f; ) =
1
2i
_
|z|=R
z
3
cos
_
1
z 2
_
dz =
1
2i
_
|z2|=R
z
3
cos
_
1
z 2
_
dz
=
1
2i
_
||=
1
R
_
1

+ 2
_
3
cos
_

2
_
=
1
2i
_
|=|
( + 2)
3

5
cos d
=
1
4!
d
4
d
4
_
( + 2)
3
cos

|
=0
=
8
3
.
(vi)
Proof. Let f(z) =
e
z
z(z+1)
= e
z
_
1
z

1
z+1
_
. Then Res(f; 0) = 1, Res(f; 1) = e
1
, and Res(f; ) =
1 e
1
.
14.
Proof. We can write g(z) as (z a)
2
h(z) where h is holomorphic and h(a) = 0. Then h(a) =
g(a)
(za)
2
and
h

(a) =
g

(a)(z a) 2g(a)
(z a)
3
.
41
Therefore
Res
_
f(z)
g(z)
, a
_
= Res
_
1
(z a)
2

f(z)
h(z)
, a
_
=
d
dz
_
f(z)
h(z)
_

z=a
=
f

(a)h(a) f(a)h

(a)
h
2
(a)
=
f

(a)
g(a)
(za)
2
f(a)
g

(a)(za)2g(a)
(za)
3
g
2
(a)
(za)
4
=
g(a)(z a)[f

(a)(z a) + 2f(a)] f(a)g

(a)(z a)
2
g
2
(a)
.
15. We make an observation of some simple rules that facilitate the evaluation of integrals via Residue
Theorem.
Rule 1 (rule for integrand function). The poles of the integrand function should be easy to nd, such
that the integrand function can be easily represented as
f(z)
(za)
n
, where f is holomorphic.
The holomorphic function f(z) can be multi-valued function like logarithm function and power function.
When its dicult to make f holomorphic in the desired region, check if it is the real or imaginary part of a
holomorphic function.
Rule 2 (rule for integration path). The choice of integration path is highly dependent on the properties
of the integrand function, where symmetry and multi-valued functions (log and power functions) are often
helpful.
Oftentimes, the upper and lower limits of integration either form a full circle (i.e. [0, 2]) so that
substitution for trigonometric functions can be easily done or have as one of the end points.
(i)
Proof. Let
1
= {z : R Rez R, Imz = 0},
2
= {z : |z| = R, 0 arg z }, and f(z) =
z
2
(z
2
+1)
2
. Then
_

1
f(z)dz +
_

2
f(z)dz = 2i Res(f, i) = 2i lim
zi
d
dz
_
z
2
(z +i)
2
_
=

2
,
and

2
f(z)dz

_

0
R
2
e
2i
(R
2
e
2i
+ 1)
2
Re
i
id

_

0
R
3
(R
2
1)
2
d 0
as R . So by letting R , we have
_

0
x
2
dx
(x
2
+1)
2
=
1
2
_

x
2
dx
(x
2
+1)
2
=

4
.
(ii)
Proof. We note sin
2
x =
1cos 2x
2
, so
_
2
0
dx
a + sin
2
x
=
1
2
_

0
dx
a + sin
2
x
=
1
2
_
2
0
d
(2a + 1) cos
.
Let b = 2a + 1, then
_
2
0
dx
a + sin
2
x
=
1
2
_
|z|=1
dz
iz
b
1
2
(z +z
1
)
=
1
2i
_
|z|=1
dz

1
2
z
2
+bz
1
2
.
The equation
1
2
z
2
+ bz
1
2
= 0 has two roots: z
1
= b

b
2
1 and z
2
= b +

b
2
1. Since b > 1, we
have z
1
D(0, 1) and z
2


D(0, 1). Therefore,
_
2
0
dx
a + sin
2
x
= i
_
|z|=1
dz
(z z
1
)(z z
2
)
= i 2i
1
z
1
z
2
=

b
2
1
=

2
_
a(a + 1)
.
42
Remark 14. If R(x, y) is a rational function of two variables x and y, for z = e
i
, we have
R(sin, cos ) = R
_
1
2
_
z +
1
z
_
,
1
2i
_
z
1
z
__
, d =
dz
iz
.
Therefore
_
2
0
R(sin, cos )d =
_
|z|=1
R
_
1
2
(z +
1
z
),
1
2i
(z
1
z
)
_
dz
iz
.
Remark 15. When the integrand function is a rational function of trigonometric functions, in view of the
previous remark, it is desirable to have [0, 2] as the integration interval. For this reason, we rst use
symmetry to expand the integration interval from [0,

2
] to [0, ], and then use double angle formula to
expand [0, ] to [0, 2]. This solution is motivated by Rule 2.
(iii)
Proof. Let
1
= {z : R Rez R, Imz = 0},
2
= {z : |z| = R, 0 arg z }, and f(z) =
ze
iz
z
2
+1
. Then
_

1
f(z)dz +
_

2
f(z)dz = 2iRes(f, i) =
_
|zi|=
ze
iz
dz
(z i)(z +i)
= 2i
ie
1
2i
=

e
i,
and

2
f(z)dz

_

0
Re
iRe
i
R
2
e
2i
+ 1
Re
i
id

_

0
R
2
e
Rsin
R
2
1
d 0
as R by Lebesgues Dominated Convergence Theorem. Therefore by letting R , we have
_

xe
ix
x
2
+ 1
dx =
_

xcos x
x
2
+ 1
dx +i
_

xsinx
x
2
+ 1
dx =

e
i,
which implies
_

0
xsin x
x
2
+1
dx =

2e
.
(iv)
Proof. Let r, R be two positive numbers such that r < 1 < R. Let
1
= {z : r |z| R, arg z = 0},

2
= {z : r |z| R, arg z = },
R
= {z : |z| = R, 0 arg z }, and
r
= {z : |z| = r, 0 arg z }.
Dene f(z) =
log z
(1+z
2
)
2
where log z is dened on C \ [0, ) and takes the principle branch of Logz with
arg z (0, 2) (see page 23).
By Residue Theorem,
_

1
+
R
+
2

r
f(z)dz = 2iRes(f, i) = 2i lim
zi
d
dz
_
log z
(z +i)
2
_
=

2
+

2
4
i.
We note

R
f(z)dz

_

0
log(Re
i
)
(R
2
e
2i
+ 1)
2
Re
i
id

_

0
R
_
(log R)
2
+
2
(R
2
1)
2
d
R(log R +)
(R
2
1)
2
0
as R ,

r
f(z)dz

_

0
log(re
i
)
(r
2
e
2i
+ 1)
2
re
i
id

_

0
r
_
(log r)
2
+
2
(1 r
2
)
2
d
r(log r +)
(1 r
2
)
2
0
as r 0, and
_

2
f(z)dz =
_
r
R
log x
(1 +x
2
)
2
dx =
_
R
r
log(x)
(1 +x
2
)
2
dx =
_
R
r
log x +i
(1 +x
2
)
2
dx.
43
Therefore by letting r 0 and R , we have
_

0
2 log x +i
(1 +x
2
)
2
dx =

2
+

2
4
i,
i.e.
_

0
log x
(1+x
2
)
2
dx =

4
.
(v)
Proof. Let
1
,
2
,
r
and
R
be dened as in (iv). Dene f(z) =
z
1
1+z
2
, where z
1
= e
(1) log z
is dened
on C \ [0, ) with log z taking the principle branch of Logz. By Residue Theorem,
_

1
+
R
+
2

r
f(z)dz = 2iRes(f, i) = 2i
e
(1) log z
z +i
|
z=i
= e
(1)

2
i
.
We note

R
f(z)dz

_

0
R
1
e
i(1)
1 +R
2
e
2i
Re
i
id

R
2
R
2
1
0
as R ,

r
f(z)dz

_

0
r
1
e
i(1)
1 +r
2
e
2i
re
i
id

r
2
1 r
2
0
as r 0, and
_

2
f(z)dz =
_
r
R
x
1
1 +x
2
dx =
_
R
r
e
(1) log(x)
1 +x
2
dx =
_
R
r
x
1
e
i(1)
1 +x
2
dx.
Therefore by letting r 0 and R , we have
_

0
x
1
(1 +e
i(1)
)
1 +x
2
dx =
_

0
x
1
1 +x
2
dx 2 cos
(1 )
2
e
i(1)

2
= e
i(1)

2
,
i.e.
_

0
x
1
1+x
2
dx =

2
csc
_

2
_
.
(vi)
Proof. This problem is a special of problem (x). See the solution there.
(vii)
Proof. We use the method outlined in the remark of Problem (ii).
_

0
d
a + cos
=
1
2
_

d
a + cos
=
1
2
_
|z|=1
dz
iz
a +
z+z
1
2
=
1
2i
_
|z|=1
dz
1
2
z
2
+az +
1
2
.
The equation
1
2
z
2
+ az +
1
2
= 0 has two roots: z
1
= a +

a
2
1 and z
2
= a

a
2
1. Clearly,
|z
1
| =
1
a+

a
2
1
< 1 and |z
2
| > 1. So
_

0
d
a + cos
=
1
i
_
|z|=1
dz
(z z
1
)(z z
2
)
= 2
1
z
1
z
2
=

a
2
1
.
Remark 16. The expansion of integration interval from [0, ] to [, ] is motivated by Rule 2.
(viii)
44
Proof. Using the result on Dirichlet integral:
_

0
sin x
x
dx =

2
, we have
_

0
_
sinx
x
_
2
dx =
_

0
sin
2
xd
_

1
x
_
=
_

0
2 sin xcos xdx
x
=
_

0
sin(2x)
2x
d(2x) =

2
.
(ix)
Proof. It is easy to see that when = p = 0, the integral is equal to 1. So without loss of generality, we only
consider the cases where and p are not simultaneously equal to 0.
Choose r, R so that 0 < r < R, and r is suciently small and R is suciently large. Dene
1
= {z :
r |z| R, arg z = 0},
2
= {z : r |z| R, arg z = 2},
r
= {z : |z| = r, 0 < arg z < 2}, and
R
= {z :
|z| = R, 0 < arg z < 2}. Let = cos , then (1, 1]. Finally, dene f(z) =
z
p
z
2
+2z cos +1
=
e
p log z
z
2
+2z+1
.
For R suciently large and r suciently small, the two roots of z
2
+ 2rz + 1 = 0 are contained in
{z : r < |z| < R}. These two roots are, respectively, z
1
= + i

1 r
2
= cos + i| sin| and z
2
=
i

1 r
2
= cos i| sin|. So |z
1
| = |z
2
| = 1, and arg z
1
+arg z
2
= 2. Denote arg z
1
by . If = 0,
we have by Residue Theorem
_

1
+
R

r
f(z)dz = 2i[Res(f, z
1
) + Res(f, z
2
)] = 2i
_
e
p log z
1
z
1
z
2
+
e
p log z
2
z
2
z
1
_
=

| sin|
[e
pi
e
p(2)i
] =
2i
sin
sin(p)e
pi
.
If = 0, we have by Residue Theorem
_

1
+
R

r
f(z)dz = 2iRes(f, 1) = 2pi e
pi
.
Meanwhile, we have the estimates

R
f(z)dz

R
p
R
2
2R 1
R 2 0
as R ,

r
f(z)dz

r
p
1 2r r
2
r 2 0
as r 0, and
_

2
f(z)dz =
_
R
0
(xe
2i
)
p
dx
x
2
+ 2x + 1
=
_
R
0
x
p
dx
x
2
+ 2x + 1
e
2pi
.
Since 1 e
2pi
= 2 sin
2
(p) 2 sin(p) cos(p)i = 2i sin(p)e
pi
, by letting R and r 0, we have
_

0
x
p
dx
x
2
+ 2xcos + 1
=
_
sin(p)
sin sin(p)
if = 0
p csc (p) if = 0 and p = 0.
If we take the convention that

sin
= 1 for = 0, then the above three formulas can be unied into a
single one:
sin(p)
sin sin(p)
.
Remark 17. We choose the above integration path in order to take advantage of the multi-valued function
x
p
(Rule 1). The no symmetry in x
2
+ 2xcos + 1 is handled by using the full circle instead of half circle.
(x)
45
Proof. Choose two positive numbers r and R such that 0 < r < 1 < R. Let
1
= {z : r |z| R, arg z = 0},

2
= {z : r |z| R, arg z = 2},
R
= {z : |z| = R, 0 < arg z < 2} and
r
= {z : |z| = r, 0 < arg z < 2}.
Dene f(z) =
z
1/p
p(z+1)z
where z
1/p
= e
log z
p
is dened on C \ [0, ). Note by substituting y
1
p
for x, we get
_

0
dx
1 +x
p
=
_

0
y
1
p
dy
p(y + 1)y
.
By Residue Theorem,
_

1
+
R

r
f(z)dz = 2Res(f, 1) = 2i
(1)
1
p
p
=
2i
p
e
log e
i
p
= 2ie
i
,
where =

p
. We have the estimates

R
f(z)dz

_
2
0
e
1
p
log(Re
i
)
p(Re
i
+ 1)Re
i
Re
i
id

2R
1
p
p(R 1)
0
as R ,

r
f(z)dz

_
2
0
e
1
p
log(re
i
)
p(re
i
+ 1)re
i
re
i
id

2r
1
p
p(1 r)
0
as r 0, and
_

2
f(z)dz =
_
R
r
(xe
2i
)
1
p
dx
p(x + 1)x
=
_
R
r
x
1
p
dx
p(x + 1)x
e
2i
.
Therefore by letting r 0 and R , we have
_

0
dx
1 +x
p
=
_

0
x
1
p
dx
p(x + 1)x
=
2ie
i
1 e
2i
=
2ie
i
2 sin e
(

2
+)i
=

sin
=

p
csc
_

p
_
.
Remark 18. The roots of 1 + x
p
= 0 are not very expressible. So following Rule 1, we make the change of
variable x
p
= y. Then we choose the above integration path to take advantage of the multi-valued function
y
1
p
. Since there is no symmetry in the denominator (y + 1)y, we used a full circle instead of a half circle.
The change of variable x
p
= y is equivalent to integration on an arc of angle 2/p instead of 2.
(xi)
Proof. (Oops! looks like my solution has a bug. Catch it if you can.) By the change of variable x =
1
y+1
, we
have
_
1
0
x
1p
(1 x)
p
1 +x
2
dx =
_

0
y
p
dy
(y + 1)
3
+ (y + 1)
.
The equation (y + 1)
3
+ (y + 1) = 0 has three roots: z
0
= 1, z
1
= 1 + i, and z
2
= 1 i. Dene
f(z) =
z
p
(z+1)
3
+(z+1)
. Then
2

i=0
Res(f, z
i
) =
1
(z + 1)(z z
1
)

z=z
2
+
1
(z + 1)(z z
2
)

z=z
1
+
1
(z z
1
)(z z
2
)

z=z
0
=
1
i (2i)
+
1
i 2i
+
1
i i
= 0.
46
Let R and r be two positive numbers with R > r. Dene
1
= {z : r < |z| R, arg z = 0},
2
= {z : r <
|z| R, arg z = 2},
R
= {z : |z| = R, 0 < arg z < 2}, and
r
= {z : |z| = r, 0 < arg z < 2}. Then by
Residue Theorem
_

1
+
R

r
f(z)dz = 2i
2

i=0
Res(f, z
i
) = 0.
We note

R
f(z)dz

_
2
0
(Re
i
)
p
(Re
i
+ 1)
3
+ (Re
i
+ 1)
Re
i
id

2R
p+1
R
3
3R
2
4R 2
0
as R , and

_
2
0
(re
i
)
p
(re
i
+ 1)
3
+ (re
i
+ 1)
re
i
id

2r
p+1
2 4r 3r
2
r
3
0
as r 0. So by letting r 0 and R , we have
_

0
y
p
(y + 1)
3
+ (y + 1)
dy + 0
_

0
y
p
(y + 1)
3
+ (y + 1)
dy e
2pi
0 = 0.
Therefore..., Houston, we got a problem here.
(xii)
Proof. Choose r, R such that 0 < r < R and r is suciently small and R is suciently large. Let
1
=
{z : r |z| R, arg z = 0},
2
= {z : r |z| R, arg z = },
R
= {z : |z| = R, 0 < arg z < } and

r
= {z : |z| = r, 0 < arg z < }. Dene f(z) =
log z
z
2
+2z+2
. By Residue Theorem,
_

1
+
R
+
2

r
f(z)dz = 2i Res(f, 1 +i) =
_
log 2
2
+
3
4
i
_
.
We note

R
f(z)dz

_
2
0
log(Re
i
)
R
2
e
2i
+ 2Re
i
+ 2
Re
i
id

log R + 2
R
2
2R 2
2R 0
as R ,

r
f(z)dz

_
2
0
log(re
i
)
r
2
e
2i
+ 2re
i
+ 2
re
i
id

log r + 2
2 2r r
2
2r 0
as r 0, and
_

2
f(z)dz =
_
R
r
log(x)
x
2
2x + 2
dx =
_
R
r
log x +i
x
2
2x + 2
dx.
Therefore by letting r 0 and R , we have
_

0
_
log x
x
2
2x + 2
+
log x
x
2
+ 2x + 2
_
dx +i
_

0
dx
x
2
2x + 2
=

2
log 2 +
3
4

2
i,
i.e.
_

0
_
log x
x
2
2x+2
+
log x
x
2
+2x+2
_
dx =

2
log 2.
Meanwhile, we note
_

0
_
log x
x
2
2x+2

log x
x
2
+2x+2
_
dx =
_

0
log x
4x
(x
2
+2)
2
4x
2
dx =
_

0
log y
y
2
+4
dy. To calculate
the value of
_

0
log y
y
2
+4
dy, we apply again Residue Theorem. Lets dene g(z) =
log z
z
2
+4
and then we have
_

1
+
R
+
2

r
g(z)dz = 2i Res(g, 2i) =

2
_
log 2 +

2
i
_
.
We note

R
g(z)dz

_
2
0
log(Re
i
)
R
2
e
2i
+ 4
Re
i
id

log R + 2
R
2
4
2R 0
47
as R ,

r
g(z)dz

_
2
0
log(re
i
)
r
2
e
2i
+ 4
re
i
id

log r + 2
4 r
2
2r 0
as r 0, and
_

2
g(z)dz =
_
R
r
log(x)
x
2
+ 4
dx =
_
R
r
log x +i
x
2
+ 4
dx.
So by letting r 0 and R , we have
2
_

0
log x
x
2
+ 4
dx +i
_

0
dx
x
2
+ 4
=

2
log 2 +

2
4
i.
Hence
_

0
log x
x
2
+4
dx =

4
log 2. Solving the equation
_
_

0
log x
x
2
2x+2
dx +
_

0
log x
x
2
+2x+2
dx =

2
log 2
_

0
log x
x
2
2x+2
dx
_

0
log x
x
2
+2x+2
dx =

4
log 2,
we get
_

0
log x
x
2
+ 2x + 2
dx =

8
log 2.
Remark 19. Note how we handled the combined diculty caused by no symmetry in the denominator of the
integrand function and log function: because x
2
+ 2x + 2 is not symmetric, we want to use full circle; but
this will cause the integrals produced by dierent integration paths to cancel with each other. Therefore we
are forced to choose half circle and use two equations to solve for the desired integral.
(xiii)
Proof. We choose r and R such that 0 < r < R, and r is suciently small and R is suciently large. Let

1
= {z : r |z| R, arg z = 0},
2
= {z : r |z| R, arg z = },
R
= {z : |z| = R, 0 < arg z < }, and

r
= {z : |z| = r, 0 < arg z < }. Dene f(z) =

z log z
z
2
+1
. Then
_

1
+
R
+
2

r
f(z)dz = 2iRes(f, i) =

2
2
ie

4
i
=

2
2

2
(1 +i).
We note

R
f(z)dz

_
2
0
(Re
i
)
1
2
log(Re
i
)
R
2
e
2i
+ 1
Re
i
id

2R

R(log R + 2)
R
2
1
0
as R ,

r
f(z)dz

_
2
0
(re
i
)
1
2
log(re
i
)
r
2
e
2i
+ 1
re
i
id

2r

r(log r + 2)
1 r
2
0
as r 0, and
_

2
f(z)dz =
_
R
r

xlog(x)
x
2
+ 1
dx =
_
R
r

xe

2
i
(log x +i)
x
2
+ 1
dx = e

2
i
_
R
r

xlog x
x
2
+ 1
dx +e

2
i
i
_
R
r

x
x
2
+ 1
dx.
By letting r 0 and R , we have
(1 +i)
_

0

xlog x
x
2
+ 1
dx
_

0

x
x
2
+ 1
dx =

2
2

2
(1 +i).
Therefore,
_

x log x
x
2
+1
dx =

2
2

2
.
48
(xiv)
Proof. We note the power series expansion ln(1 z) =

n=1
z
n
n
and ln(1 +z) =

n=1
(1)
n+1 z
n
n
hold in
|z| < 1 and the convergence is uniform on |z| 1 , for any (0, 1). Therefore, for any > 0, we have
_

log
_
e
x
+ 1
e
x
1
_
dx =
_

n=1
(1)
n+1
e
nx
n
+

n=1
e
nx
n
_
dx = 2
_

k=1
e
(2k1)x
2k 1
_
dx = 2

k=1
e
(2k1)
(2k 1)
2
.
Therefore
_

0
log
_
e
x
+ 1
e
x
1
_
dx = lim
0
2

k=1
e
(2k1)
(2k 1)
2
= 2

k=1
1
(2k 1)
2
,
where the last equality can be seen as an example of Lebesgues Dominated Convergence Theorem applied
to counting measure. By Problem 11 (ii) with z =
1
2
, we know

k=1
1
(2k1)
2
=

2
8
. So the integral is equal
to

2
4
.
(xv)
Proof. Since we have argued rather rigorously in (xiv), we will argue non-rigorously in this problem.
_

0
x
e
x
+ 1
dx =
_

0
e
x
x
1 +e
x
dx =
_

0
e
x
x

n=0
(1)
n
e
nx
dx =

n=0
(1)
n
_

0
e
(n+1)x
xdx =

n=0
(1)
n
(n + 1)
2
.
Using the function

2
sin z
and imitating what we did in Problem 12, we can easily get

n=1
(1)
n1
n
2
=

2
12
.
Therefore
_

0
x
e
x
+1
dx =

2
12
.
(xvi)
Proof. We provide two solutions, which are essentially the same one.
Solution 1. In problem (xvii), we shall prove (let a = 1)
_

0
xsinx
2 2 cos x
dx = log 2.
Note
_

0
xsinx
2 2 cos x
dx =
_

0
x 2 sin
x
2
cos
x
2
dx
2 2 sin
2 x
2
= 2
_
2
0
cos d
sin
= 2
_
2
0
log(sin)d.
So
_
2
0
log(sin )d =

2
log.
Solution 2. We provide a heuristic proof which discloses the essence of the calculation. Note how Rule
1-3 lead us to this solution.
_
2
0
log(sin)d =
_
2
0
log(cos )d =
1
2
_
2

2
log(cos )d
=
1
4
_
2

2
log
_
1 + cos 2
2
_
d
=
1
8
_

log(1 + cos )d

4
log 2.
(Note how Rule 2 leads us to extended the integration interval from [0,

2
] to [, ].) Motivated by the
similar diculty explained in the remark of Problem (xvii), we try Rule 1 and note
_

log(1 +e
i
)d =
_
|z|=1
log(1 +z)
iz
dz = 2 log 1 = 0.
49
For the above application of Cauchys integral formula to be rigorous, we need to take a branch of logarithm
function that is dened on C \ (, 0] and take a small bypass around 0 for the integration contour, and
then take limit.
Using the above result, we have
0 = Re
_

log(1 +e
i
)d =
1
2
_

log
_
(1 + cos )
2
+ sin
2

d =
1
2
_

[log 2 + log(1 + cos )] d.

Therefore
_

log(1 + cos )d = 2 log 2 and

_
2
0
log(sin)d =
1
8
_

log(1 + cos )d

4
log 2 =

2
log 2.
(xvii)
Proof. We rst assume a > 1. Then by integration-by-parts formula we have
_

0
xsinx
1 2a cos x +a
2
dx =
1
2
_

xsinxdx
1 2a cos x +a
2
=
1
2
_

xd(1 2a cos x +a
2
)
(1 2a cos x +a
2
) 2a
=
1
4a
_
2 log(1 +a)
2

log(1 2a cos x +a
2
)dx
_
.
Since a > 1, a {z : Rez > 0} for any D(0, 1). Therefore we can take a branch of the logarithm
function such that log(az) is a holomorphic function on D(0, 1) and is continuous on

D(0, 1). For example,
we can take the branch log z such that it is dened on C \ (, 0] with log(e
i
) = and log(e
i
) = .
By Cauchys integral formula,
_

log(a e
i
)d =
_
|z|=1
log(a z)
iz
dz = 2 log a.
Therefore,
_

log(1 2a cos x + a
2
)dx = 2Re
_

log(a e
i
)d = 4 log a. Plug this equality into the
calculation of the original integral, we get

a
log
_
1+a
a
_
.
If 0 < a < 1, we have
_

0
xsinx
1 2a cos x +a
2
dx =
1
a
2
_

0
xsinx
1
2
a
cos x +
1
a
2
dx =
1
a
2


1
a
log
_
1 +
1
a
1
a
_
=

a
log(a + 1).
Assuming the integral as a function of a is continuous at 1, we can conclude for a = 1, the integral is
equal to log 2. The result agrees with that of Mathematica. To prove the continuity rigorously, we split
the integral into
_
2
0
x sin xdx
12a cos x+a
2
and
_

2
xsin xdx
12a cos x+a
2
. For the second integral, we have (x [

2
, ])
xsinx
1 2a cos x +a
2
=
xsinx
|a e
ix
|

xsinx
|a e
i

2
|
xsinx.
So by Lebesgues dominated convergence theorem, the second integral is a continuous function of a for
a (0, ). For the rst integral, we note (1 2a cos x +a
2
) takes its minimum at cos x. So
xsinx
1 2a cos x +a
2

x
sinx
.
Again by Lebesgues dominated convergence theorem, the rst integral is a continuous function of a for
a (0, ). Combined, we conclude the integral
_

0
xsinx
1 2a cos x +a
2
dx
as a function of a (0, ) is a continuous function.
50
Remark 20. We did not take the transform cos x =
z+z
1
2
because log(1 2a cos x +a
2
) would then become
log[w(a) w(z)] log(2a), where w(z) =
z+z
1
2
maps D(0, 1) to U := C \ [1, 1] and we cannot nd a
holomorphic branch of logarithm function on U.
The application of integration-by-parts formula and the extension of integration interval from [0, ] to
[, ] are motivated by Rule 2. But here a new trick emerged: if the integrand function or part of it
could come from the real or imaginary part of a holomorphic function, apply Cauchys integral formula (or
integration theorem) to that holomorphic function and try to relate the result to our original integral (Rule
1).
(xviii)
Proof. The function log(z +i) is well-dened on C
+
:= {z : Imz 0}. We take the branch of log(z +i) that
evaluates to log 2 +

2
i at i. Dene
R
= {z : |z| = R, 0 arg z }. Then for R > 1, Residue Theorem
implies
_
R
R
log(z +i)
1 +z
2
dz +
_

R
log(z +i)
1 +z
2
dz = 2i
i +i
i +i
= log 2 +

2
2
i.
Note

R
log(z +i)
1 +z
2
dz

_

0
log(Re
i
+ 1)
1 +R
2
e
2i
Re
i
id

R[log(R + 1) + 2]
R
2
1
0
as R . So
_

log(x+i)
1+x
2
dx = log 2 +

2
2
i. Hence
log 2 = Re
_

log(x +i)
1 +x
2
dx =
_

log

x
2
+ 1
1 +x
2
dx =
_

0
log(1 +x
2
)
1 +x
2
dx.
Remark 21. The main diculty of this problem is that the poles of
1
1+x
2
coincide with the branch points
of log(x
2
+ 1), so that we cannot apply Residue Theorem directly. An attempt to avoid this diculty
might be writing
log(x
2
+1)
1+x
2
as the sum of
log(x+i)
1+x
2
and
log(xi)
1+x
2
and integrate them separately along dierent
paths which do not contain their respective branch points. But log(x + i) and log(x i) cannot be dened
simultaneously in a region of a Riemann surface which contains both integration paths. But the observation
that Re[log(x i)] = log

x
2
+ 1 gives us a simple solution (Rule 1).
16.
Proof. The power series

n=0
z
n
is divergent on every point of D(0, 1): if z = 1,

n=0
z
n
= 1+1+1+ ;
if z = 1 and z = e
i
,
N

n=1
z
n
=
N

n=0
cos(n) +i
N

n=0
sin(n) =
sin
(N+1)
2
cos
N
2
sin

2
+i
sin
(N+1)
2
sin
N
2
sin

2
.
Similarly, the series

n=0
z
n2
is divergent on every point of D(0, 1). So the functions represented by
these two power series cannot be analytic continuation of each other.
17.
Proof. Its clear we need the assumption that = 0. The series

n=0
(z)
n
is convergent in U
1
= {z : |z| <
1
||
}. The series
1
1z

n=0
(1)
n
[(1)z]
n
(1z)
n
is convergent in U
2
= {z : |1 ||z| < |1 z|}. On U
3
= U
1
U
2
,
both of the series represent the analytic function
1
1z
. So the functions represented by these two series are
analytic continuation of each other.
18.
51
Proof. On the line segment {z : 0 < Rez < 1, Imz = 0}, Taylor series of ln(1 + z) is z
1
2
z
2
+
1
3
z
3
.
On the same line segment, 0 <
1z
2
<
1
2
. So Taylor series of ln(1 x) (|x| 1, x = 1) gives
ln2
1 z
2

(1 z)
2
2 2
2

(1 z)
3
3 2
3
= ln2 + ln
_
1
1 z
2
_
= ln(1 +z).
Since the two holomorphic functions represented by these two series agree on a line segment, they must agree
on the intersection of their respective domains (Theorem 2.13). Therefore they are analytic continuations of
each other.
19.
Proof. Clearly the series is convergent in D(0, 1) and is divergent for z = 1. So its radius of convergent
R = 1. Let f(z) =

n=0
z
2
n
. Then f(z) is analytic in D(0, 1) and z = 1 is a singularity of f(z). We note
f(z) = z +

n=1
z
2
n
= z +

n=1
z
22
n1
= z +

n=1
(z
2
)
2
n1
= z +f(z
2
). Therefore, we have
f(z) = z +f(z
2
) = z +z
2
+f(z
4
) = z +z
2
+z
4
+f(z
8
) = .
So the roots of equations z
2
= 1, z
4
= 1, z
8
= 1, , z
2
n
= 1, , etc. are all singularities of f. These
roots form a dense subset of D(0, 1), so f(z) can not be analytically continued to the outside of its circle
of convergence D(0, 1).
4 Riemann Mapping Theorem
3.
Proof. Let g(z) =

n=0
c
n
z
n
. Then g(z) is holomorphic and has the same radius of convergence as f(z).
z

R convergence domain of g(z), we have

g(z

) =

n=0
c
n
z
n

n=0
c
n
(

)
n
=

f( z

) =

f(z

) = f(z

),
where the next to last equality is due to the fact that z

R and the last equality is due to the fact that

f(R) R. By Theorem 2.13, f g in the intersection of their respective domains of convergence. So
c
n
= c
n
, i.e. c
n
(n = 0, 1, ) are real numbers.
11.
Proof. Let g(z) =

f( z) (rst take conjugate of z, then take conjugate of f( z)). Then
g(z)
z
=


f( z)
z
= conjugate
_
f( z)
z
_
= 0.
So g is a holomorphic function. Clearly g is univalent. Since U is symmetric w.r.t. real axis, g(U) =

f(

U) =

f(U) = conjugate(D(0, 1)) = D(0, 1). Furthermore, g(z

0
) =

f( z
0
) =

f(z
0
) = 0, and
g

(z
0
) = lim
zz
0

f( z)

f( z
0
)
z z
0
= lim
zz
0
conjugate
_
f( z) f( z
0
)
z z
0
_
= conjugate(f

( z
0
))
= conjugate(f

(z
0
))
= f

(z
0
) > 0.
By the uniqueness of Riemann Mapping Theorem, g f.
52
5 Dierential Geometry and Picards Theorem
6 A First Taste of Function Theory of Several Complex Variables
There are no exercise problems for this chapter.
7 Elliptic Functions
There are no exercise problems for this chapter.
8 The Riemann -Function and The Prime Number Theory
There are no exercise problems for this chapter.
References
[1] Johann Boos. Classical and modern methods in summability, Oxford University Press, 2001.
[2] John B. Conway. Functions of one complex variable, 2nd Edition, Springer, 1978.
[3] Fang Qi-Qin. A course on complex analysis (in Chinese), Peking University Press, 1996.
[4] James Munkres. Analysis on manifolds, Westview Press, 1997.
[5] M. A. Qazi. The mean value theorem and analytic functions of a complex variable. J. Math. Anal. Appl.
324(2006) 30-38.
53