IV 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

1 X X X X X X X X X X X
2 X X X X X X X X
3 X X X X X X X
4 X X X X
5 X X X X
6 X X X X
7 X X X X X X X X X X X
8 X X X X X X X X

1
I.1.1
Identify and sketch the set of points satisfying.
(a) jz 1 ij = 1 (f) 0 < Im z <
(b) 1 < j2z 6j < 2 (g) < Re z <
(c) jz 1j2 + jz + 1j2 < 8 (h) jRe zj < jzj
(d) jz 1j + jz + 1j 2 (i) Re (iz + 2) > 0
(e) jz 1j < jzj (j) jz ij2 + jz + ij2 < 2

Solution
Let z = x + iy, where x; y 2 R.
(a) Circle, centre 1 + i, radius 1.

jz 1 ij = 1 , j(x 1) + i (y 1)j = 1 , (x 1)2 + (y 1)2 = 12

(b) Annulus with centre 3, inner radius 1=2, outer radius 1.

1 < j2z 6j < 2 , 1 < 2 jz 3j < 2 ,

, 1=2 < jz 3j < 1 , (1=2)2 < (x 3)2 + y 2 < 12
p
(c) Disk, centre 0, radius 3.

jx + iy 1j2 + jx + iy + 1j2 < 8 ,

p 2
, (x 1)2 + y 2 + (x + 1)2 + y 2 < 8 , x2 + y 2 < 3

(d) Interval [ 1; 1].

q q
2
jz 1j + jz + 1j 2 , (x 1) + y2 2 (x + 1)2 + y 2 ,
q 2 q 2
2 2
, (x 1) + y2 2 (x + 1) + y 2 ,
q q 2
, (x + 1)2 + y 2 x+1, (x + 1) +2
y2 (x + 1)2 , y = 0

Now, take y = 0 in the inequality, and compute the three intervals

2
 x < 1; then jx 1j + jx + 1j = (x 1) (x + 1) = 2x 2;
1 x 1 then jx 1j + jx + 1j = (x 1) + (x + 1) = 2 2
x > 1; then jx 1j + jx + 1j = (x 1) + (x + 1) = 2x 2:

(e) Half–plane x > 1=2.

jz 1j < jzj , jz 1j2 < jzj2 , jx + iy 1j2 < jx + iyj2 ,

, (x 1)2 + y 2 < x2 + y 2 , x > 1=2

(f) Horizontal strip, 0 < y < .

(g) Vertical strip, <x< .
(h) CnR.

jRe zj < jzj , jRe (x + iy)j2 < jx + iyj2 , x2 < x2 + y 2 , jyj > 0
(i) Half plane y < 2.

Re (iz + 2) > 0 , Re (i (x + iy) + 2) > 0 , y+2>0,y <2

(j) Empty set.

jz ij2 + jz + ij2 < 2 , jx + iy ij2 + jx + iy + ij2 < 2 ,

, x2 + (y 1)2 + x2 + (y + 1)2 < 2 , x2 + y 2 < 0

3
I.1.2
Verify from the de…nitions each of the identities
(a) z + w = z + w (b) zw = z w (c) jzj = jzj (d) jzj2 = z z
Draw sketches to illustrate (a) and (c).

Solution
Substitute z = x + iy and w = u + iv, and use the de…nitions.
(a)

z + w = (x + iy) + (u + iv) = (x + u) + (y + v) i =
= (x + u) (y + v) i = (x iy) + (u iv) = z + w:

(b)

zw = (x + iy) (u + iv) = (xu yv) + (xv + yu) i =

= (xu yv) (xv + yu) i = (x iy) (u iv) = z w:

(c)

q p
jzj = x + iy = jx iyj = x2 + ( y)2 = x2 + y 2 = jx + iyj = jzj :

(d)

p 2
jzj2 = jx + iyj2 = x2 + y 2 = x2 + y 2 =
= x2 i2 y 2 = (x + iy) (x iy) = z z:

4
I.1.3
Show that the equation jzj2 2 Re (az) + jaj2 = 2
represents a circle
centered at a with radius .

Solution
Let z = x + iy and a = + i , we have

jzj2 2 Re (az) + jaj2 =

2
= x2 + y 2 2 Re (( i ) (x + iy)) + 2
+ =
2 2 2 2
=x +y 2 Re (( x + y) + i ( y x)) + + =
2 2 2 2
=x +y 2 ( x + y) + + =
= (x )2 + (y )2 :

Thus the equation jzj2 2 Re (az) + jaj2 = 2

, becomes

(x )2 + (y )2 = 2
;
which is the equation for a circle of radius centered at ( ; ), which in
complex notation is the point a = + i .

5
I.1.4
Show that jzj jRe zj + jIm zj, and sketch the set of points for which
equality holds.

Solution
Apply triangle inequality to z = Re z + i Im z, to obtain jzj 6 jRe zj + jIm zj.
Now set z = x + iy, and see then equality holds

p p p p 2 p p 2
jzj = jRe zj+jIm zj , x2 + y 2 = x2 + y 2 , x2 + y 2 = x2 + y 2 ,
p 2 p 2 p p
, x2 +y 2 = x2 + x2 +2 x2 y 2 , x2 y 2 = 0 , x = 0 or y = 0:

Equality holds only when z is real or pure imaginary, which are all the points
on the real and imaginary axis.

6
I.1.5
Show that

jRe zj jzj ; jIm zj jzj :

Show that

jz + wj2 = jzj2 + jwj2 + 2 Re (z w) :

Use this to prove the triangle inequality jz + wj jzj + jwj.

Solution
Let z = x + iy. Then since the square root function is monotone, we have
p p
jRe zj = jxj = x2 6 x2 + y 2 = jzj ;
p p
jIm zj = jyj = y 2 6 x2 + y 2 = jzj :

Now for z; w we have

jz + wj2 =
= (z + w) (z + w) = (z + w) (z + w) = z z + z w + wz + ww =
= jzj2 + 2 Re (z w) + jwj2 ;

where we have used the fact that 2 Re (z w) = z w +z w = z w + wz = z w +wz.

We use both of the above facts and the trivial identities jzj = jzj and jzwj =
jzj jwj to prove the triangle inequality for z; w. We have

jz + wj2 = jzj2 + 2 Re (z w) + jwj2 jzj2 + 2 jRe (z w)j + jwj2

jzj2 + 2 jz wj + jwj2 = jzj2 + 2 jzj jwj + jwj2 = (jzj + jwj)2 :

The desired inequality now follows by taking square root of both sides.

7
I.1.6
For …xed a 2 C, show that jz aj = j1 azj = 1 if jzj = 1 and 1 az 6= 0.

Solution
If jzj = 1, then jzj = 1 and z z = 1. Use this and get

jz aj = jz aj jzj = jz z azj = j1 azj = j1 azj :

We have

jz aj
= 1;
j1 azj
as was to be shown.

8
I.1.7
Fix > 0, = 6 1, and …x z0 ; z1 2 C. Show that the set of z satisfying
jz z0 j = jz z1 j is a circle. Sketch it for = 12 and = 2, with
z0 = 0 and z1 = 1. What happens when = 1?

Solution
Recall that a circle in R2 centered at (a; b) with radius r is given by the
equation

(x a)2 + (y b)2 = r2 :
We manipulate the equation

jz z0 j = jz z1 j :
The solutions set of the equation above remains the same if we square both
sides,

jz z0 j2 = 2
jz z1 j2 :
Let z = x + iy, z0 = x0 + iy0 and z1 = x1 + iy1 . Thus our equation becomes

(x x0 )2 + (y y0 )2 = 2
(x x1 )2 + (y y1 )2 :
Expanding the squares, and grouping terms, we have

2
1 x2 2 x0 2
x1 x+ x20 2 2
x1 + 1 2
y 2 2 y0 2
y1 y+ y02 2 2
y1 =0
2
Dividing both sides by (1 ), we have

2
(x0 x1 ) (x20 2 2
x1 ) (y0 2
y1 ) (y02 2 2
y1 )
x2 2 2
x+ 2
+ y2 2 2
y+ 2
=0
1 1 1 1
Now complete the squares for both the x and y terms. Recall that

x2 2ax + b = (x a)2 a2 + b:
So we have

9
 2 2 2 2
x0 x1 (1 ) (x20 2 2
x1 ) (x0 2
x1 )
x + +
1 2
(1 2 )2

2 2 2 2
y0 y1 (1 ) (y02 2 2
y1 ) (y0 2
y1 )
+ y + = 0:
1 2
(1 2 )2

This becomes

2 2 2 2
x0 x1 y0 y1
x 2
+ y 2
=
1 1
2 2 2 2 2
(x0 x1 ) + (y0 y1 ) (1 ) (x20 + y02 2
(x21 + y12 ))
= =
(1 2 )2

2
(x1 x0 )2 + (y1 y0 )2
= ;
(1 2 )2

which is the equation for a circle. If z0 = 0, and z1 = 1, we have

2 2 2
2
x 2
+y = 2
:
1 1
When = 21 we have a circle of radius 23 centered at 1
3
; 0 , and when = 2,
2 4
we have a circle of radius 3 centered at 3 ; 0 . When = 1, we have the
equation

jz z0 j = jz z1 j ;
which is the line bisecting the two points. When z0 = 0; z1 = 1, this is the
line Re z = 21 .

10
I.1.7 I.1.7 I.1.7
2 2 2
Im z Im z Im z
1 1 1

-1 1 2 -1 1 2 -1 1 2
-1
Re z -1
Re z -1
Re z

-2 -2 -2

1
= 2
=1 =2

11
I.1.8
Let p (z) be a polynomial of degree n 1 and let z0 2 C. Show
that there is a polynomial h (z) of degree n 1 such that p (z) =
(z z0 ) h (z)+p (z0 ). In particular, if p (z0 ) = 0, then p (z) = (z z0 ) h (z).

Solution
Set

p (z) = an z n + an 1 z n 1
+ an 2 z n 2
+ + a2 z 2 + a1 z + a0 :
and

h (z) = bn 1 z n 1
+ bn 2 z n 2
+ bn 3 z n 3
+ : : : + b2 z 2 + b1 z + b0 :
We equate coe¢ cients in the polynomial identity p (z) = (z z0 ) h (z) +
p (z0 ), and get

an z n + an 1 z n 1
+ an 2 z n 2
+ + a2 z 2 + a1 z + a0 =
= (z z0 ) bn 1 z n 1 + bn 2 z n 2 + bn 3 z n 3 + : : : + b2 z 2 + b1 z + b0 +p (z0 ) =
= bn 1 z n + bn 2 z n 1 + bn 3 z n 2 + : : : + b2 z 3 + b1 z 2 + b0 z + p (z0 )
bn 1 z0 z n 1 bn 2 z0 z n 2 bn 3 z0 z n 3 : : : b2 z0 z 2 b1 z0 z b0 z0 =
bn 1 z n +(bn 2 bn 1 z0 ) z n 1 +(bn 3 bn 2 z0 ) z n 2 +(bn 4 bn 3 z0 ) z n 3 +
+ (b2 b3 z0 ) z 3 + (b1 b2 z0 ) z 2 + (b0 b1 z0 ) z + p (z0 ) b0 z0 :

Equate and solve for the bj ´s in terms of aj ´s.

8
8 >
> b n 1 = an
>
> nP k 1
< an = b n 1 <
bk = ak+1+i z0i ; 0 k n 2
ak = bk 1 bk z0 ; 0 k n 1 ) i=0
: >
> Pn
a0 = p (z0 ) b0 z0 >
>
: p (z0 ) = ai z0i
i=0

Proof by induction on degree n of p (z), set

p (z) = an z n + an 1 z n 1
: : : + a0 ;
where an 6= 0.

12
Fix z0 and write

p (z) = an (z z0 ) z n 1
+ r (z) ;
where deg r (z) n 1.
By using the induction hypothesis, we can assume that

r (z) = q (z) (z z0 ) + c;
where deg q (z) deg r (z). Then

p (z) = an z n 1
+ q (z) (z z0 ) + c = h (z) (z z0 ) + c
Since deg q (z) n 2, deg r (z) n 1.
Plug in z0 , get p (z0 ) = c.

13
I.1.9
Find the polynomial h (z) in the preceding exercise for the following
choices of p (z) and z0
(a) p (z) = z 2 and z0 = i
(b) p (z) = z 3 + z 2 + z and z0 = 1
(c) p (z) = 1 + z + z 2 + + z m and z0 = 1

Solution
From the preceding exercise we have

p (z) = (z z0 ) h (z) + p (z0 ) :

We solve for h (z) then

p (z) p (z0 )
h (z) = :
z z0
(a)
We have that p (z) = z 2 and z0 = i, thus p (z0 ) = p (i) = 1.
Thus

p (z) p (z0 ) z2 + 1
h (z) = = = z + i;
z z0 z i
and

z 2 = (z i) (z + i) 1:
(b)
We have that p (z) = z 3 + z 2 + z and z0 = 1, thus p (z0 ) = p ( 1) = 1.
Thus

p (z) p (z0 ) z3 + z2 + z + 1 (z + 1) (z 2 + 1)
h (z) = = = = z 2 + 1;
z z0 z+1 z+1
and

z 3 + z 2 + z = (z + 1) z 2 + 1 1:
(c)
We have that p (z) = 1 + z + z 2 + + z m and z0 = 1, thus

14
 0; m odd
p (z0 ) = p ( 1) =
1; m even
Thus

p (z) p (z0 )
h (z) = =
8 z z0
< (zm 1 +z m 3 +z 2 +1)(z+1)
= zm 1
+ zm 3
+ + z2 + 1 if m is odd
= z+1
: (zm 1 +z m 3 +z 3 +z )(z+1)+1
= zm 1
+ zm 3
+ : : : + z 3 + z if m is even
z+1

and

zm + zm 1
+ z2 + z + 1 =
(z + 1) (z m 1
+ zm 3
+ + z 2 + 1) if m is odd
=
(z + 1) (z m 1
+ zm 3
+ 3
+ z + z) + 1 if m is even

15
I.1.10
Let q (z) be a polynomial of degree m 1. Show that any polynomial
p (z) can be expressed in the form

p (z) = h (z) q (z) + r (z) ;

where h (z) and r (z) are polynomials and the degree of the remain-
der r (z) is strictly less than m.
Hint. Proceed by induction on the degree of p (z). The resulting
method is called the division algorithm.

Solution
First suppose p (z) is the zero polynomial. (So, the degree of p (z) is 1.)
The degree of r (z) must be less than the degree of q (z). If h (z) 6= 0, it
follows that the degree of h (z) q (z) is greater than the degree of r (z). This
then implies that h (z) q (z)+r (z) 6= 0. So, h (z) q (z)+r (z) = 0 implies that
h (z) = 0, and thus r (z) = 0. So, the polynomials are h (z) = 0 and r (z) = 0,
and these polynomials are the only ones that satisfy both conditions.

Now assume that the division algorithm is true for all polynomials p (z) of
degree less than n. (Where n 0.) If the degree of q (z) is greater than the
degree of p (z), and h (z) is nonzero, then h (z) q (z) + r (z) has degree greater
than p (z). So, if the degree of q (z) is greater than the degree of p (z), then
h (z) = 0 and thus r (z) = p (z). This proves both existence and uniqueness
of h (z) and r (z), in this case.

Now, suppose that the degree of q (z) is less than or equal to the degree of
p (z). Set

p (z) = an z n + an 1 z n 1
+ + a1 z + a0
and

q (z) = bm z m + bm 1 z m 1
+ + b1 z + b0 ;
where an 6= 0, bm 6= 0 and m n.
Let
an n m
p1 (z) = z q (z) p (z) ;
bm

16
then

an n m
p1 (z) = z bm z m + bm 1 z m 1
+ + b1 z + b0 an z n + an 1 z n 1
+ + a1 z + a0 :
bm
The monomials of degree n cancel, and therefore p1 (z) is a polynomial of de-
gree at most n 1. It follows, by assumption, that p1 (z) = h1 (z) q (z)+r1 (z),
where h1 (z) and r1 (z) are the unique polynomials satisfying the conditions
above.
Let
an n m
h (z) = z h1 (z) ;
bm
and let

r (z) = r1 (z) :
Then

h (z) q (z) + r (z) =

an n m
= z h1 (z) q (z) r1 (z) =
bm
an n m
= z q (z) (h1 (z) q (z) + r1 (z)) =
bm
an n m
= z q (z) p1 (z) =
bm
= p (z) :

Thus given p (z) and q (z), there exist polynomials h (z) and r (z) satisfying
the above two conditions.

17
I.1.11
Find the polynomials h (z) and r (z) in the preceding exercise for
p (z) = z n and q (z) = z 2 1.

Solution
Require

z n = h (z) z 2 1 + r (z) ; deg r (z) 1:

If n is even

zn = zn 2
+ zn 4
+ zn 6
+ + z2 + 1 z2 1 + 1:
If n is odd

zn = zn 2
+ zn 4
+ zn 6
+ + z3 + z z2 1 + z:
Thus

zn 2
+ zn 4
+ zn 6
+ + z 2 + 1; if n is even,
h (z) = ;
zn 2
+ zn 4
+ zn 6
+ + z 3 + z; if n is odd.
and

1; if n is even,
r (z) = :
z; if n is odd.

18
I.2.1
Express all values of the following expressions in both polar and
Cartesian
p coordinates,
p and plot them.
(a) i (c) 4
1 (e) ( 8)1=3 (g) (1 + i)8
p p 25
(b) i 1 (d) 4 i (f) (3 4i)1=8 (h) 1+i
p
2

Solution
(a)

p n o n p o
i( =2+2k ) 1=2 i( =4+k )
i= e =e ; k = 0; 1 = ei =4 i5 =4
;e = (1 + i) = 2 :

(b)

p p 1=2
i 1 = 2ei(3 =4+2k )
= 21=4 ei(3 =8+k )
; k = 0; 1 =

= 21=4 ei3 =8
; 21=4 ei11 =8
= 21=4 (cos (3 =8) + i sin (3 =8)) :

(c)

p n o
+2k ) 1=4
4
1 = ei( = ei( =4+k =2)
; k = 0; 1; 2; 3 =
n p p o
i =4 i3 i=4 i5 =4 i7 =4
= e ;e ;e ;e = (1 i)= 2; ( 1 i) = 2 :

(d)
p n o
4 =2+2k ) 1=4 i( =8+k =2)
i= ei( e ; k = 0; 1; 2; 3 = ei =8
; ei5 =8
; ei9 =8
; ei13 =8
=
= f (cos ( =8) + i sin ( =8)) ; (cos (5 =8) + i sin (5 =8))g :

(e)
n o
+2k ) 1=3
( 8)1=3 = 23 ei( = 2ei( =3+2k =3)
; k = 0; 1; 2 =
n p p o
= 2ei =3
; 2ei ; 2ei5 =3
= 1 + i 3; 2; 1 i 3 :

(f)

19
 1 4
Draw …gure and get tan 0 = 4=3 ) 0 = tan 3
.

n o
) 1=8
(3 4i)1=8 = 5ei( 0 +2k
= 51=8 ei( 0 =8+k =4)
; k = 0; 1; : : : ; 7 =
= 51=8 ei( 0 =8)
; 51=8 ei( 0 =8+ =4)
; 51=8 ei( 0 =8+ =2)
; 51=8 ei( 0 =8+3 =4)
=
1=8 1=8
5 (cos ( 0 =8) + i sin ( 0 =8)) ; 5 (cos ( 0 =8 + =4) + i sin ( 0 =8 + =4)) ;
51=8 (cos ( 0 =8 + =2) + i sin ( 0 =8 + =2)) ; 51=8 (cos ( 0 =8 + 3 =4) + i sin ( 0 =8 + 3 =4)) :
(g)
p 8
(1 + i)8 = 2ei( =4+2k )
= 16ei(2 +16k )
= 16ei2 = 16:
(h)

25
1+i =4+2k ) 25 1+i
p = ei( = ei(25 =4+50k )
= ei( =4+6 +50k )
= ei =4
= p :
2 2

I.2.1a I.2.1b I.2.1c

1 1 1

-1 1 -1 1 -1 1

-1 -1 -1

I.2.1d I.2.1e I.2.1f

2
1 1

-1 1 -2 2 -1 1

-1 -1
-2

20
I.2.1g I.2.1h
20
1

-20 20 -1 1

-1
-20

21
I.2.2
Sketch the following sets
(a) jarg zj < =4 (c) jzj = arg z
(b) 0 < arg (z 1 i) < =3 (d) log jzj = 2 arg z

Solution
a) Sector. b) Sector. c) Is a spiral curve starting at 0, spiraling to 1. (d) Is
a spiral curve, spiraling to 0, and to 1.

I.1.2a I.1.2b I.2.2c

5 5

π/3 (π/2,0)
π/4
π/4 (−π,0) (2π,0)
-5 5 -5 5

(−3π/2,0)

-5 -5

I.2.2d (-3 < x < 3) I.2.2d (-30 < x < 30) I.2.2d (-800 < x < 800)

(1,0)

22
I.2.3
For a …xed complex number b, sketch the curve ei + be i : 0 2 .
Di¤erentiate between the cases jbj < 1; jbj = 1 and jbj < 1.
Hint. First consider the case b > 0, and then reduce the general
case to this case by a rotation.

Solution
For 0 < b < 1, an ellipse x2 = (1 + b)2 + y 2 = (1 b)2 = 1, traversed in positive
direction with increasing . For b = 1, an interval [ 2; 2]. For 1 < b < +1,
an ellipse traversed in negative direction. For b = ei' , express equation as
ei'=2 ei( '=2) + e i( '=2) to see that curve is rotate of ellipse or interval
by '=2.

I.2.7 (b = 0.5) I.2.7 (b = 1) I.2.7 (b = 1.5)

3 3 3
y y y
2 2 2
1 1 1

-3 -2 -1 1 2 3 -3 -2 -1 1 2 3 -3 -2 -1 1 2 3
-1 x -1 x -1 x
-2 -2 -2
-3 -3 -3

23
I.2.4
For which n is i an n-th root of unity?

Solution.
i is an nth root of unity for in = 1, n = 4k, k = 1; 2; 3; : : :

24
I.2.5
For n 1, show that
(a) 1 + z + z 2 + + z n = (1 z n+1 ) = (1 z), z 6= 1
1
sin n+
1 2
(b) 1 + cos + cos 2 + + cos n = 2
+ 2 sin =2
.

Solution
(a)
Set
sn = 1 + z + z 2 + + zn;
and multiply sn with z and have

zsn = z + z 2 + z 3 + + z n+1 :
Now subtract zsn from the sum sn , and have

sn (1 z) = 1 z n+1 :
If z 6= 1 we have after division by 1 z;

1 z n+1
sn = :
1 z
(b)
Apply (a) to z = ei and to z = e i
;

1ei(n+1)
1 + ei + ei2 + + ein = ;
1 ei
i i2 in 1 e i(n+1)
1+e +e + +e = :
1 e i
Add the identities, and use the de…nitions of sine and cosine.

25
 2 (1 + cos + cos 2 + + cos n ) =
1ei(n+1) 1 e i(n+1)
= + =
1 ei 1 e i
1 ei(n+1) e i =2 1 e i(n+1) ei =2
= + =
e i =2 ei =2 ei =2 e i =2
1 e i =2 1 ei(n+1) 1 ei =2 1 e i(n+1)
= + =
2i sin ( =2) 2i sin ( =2)
" 1 1
#
1 ei =2 e i =2 + ei(n+ 2 ) e i(n+ 2 ) sin ( =2) + sin n + 1
2
= = =
2i sin ( =2) sin ( =2)
sin n + 12
=1+ :
sin ( =2)

Divide both sides with 2 and get

1 sin n + 21
1 + cos + cos 2 + + cos n = + :
2 2 sin =2

26
I.2.6
Fix n 1. Show that the n th roots of unity w0 ; : : : ; wn 1 satisfy
(a) (z w0 ) (z w1 ) : : : (z wn 1 ) = z n 1:
(b) w0 + + wn 1 = 0 if n 2:
(c) w0 wn 1 = ( 1)n 1 :
nP1
0; 1 k n 1;
(d) wjk =
j=0 n; k = n:

Solution
(a)
Let w0 ; : : : ; wn 1 be the n th roots of unity then wj = e2 ji=n
, and wj are
a roots of z n 1 since
ji=n n
wjn 1 = e2 1 = e2 ji
1 = 0:

By the fundamental theorem of algebra, and since the root are simple and
the coe¢ cient for the z n term is 1 it follows that

zn 1 = (z w0 ) (z w1 ) (z wn 1 ) :
(b)
Using exercise (a) and multiply the factors in the product we get

zn 1 = zn (w0 + w1 + wn 1 ) z n 1
+c2 z n 2
+ +( 1)n w0 w1 wn 1 :

By identi…cation of the coe¢ cients we see that

w0 + w1 + + wn 1 =0 if n 2:
(c)
From (b) follows that

1 = ( 1)n w0 w1 wn 1 , w0 w1 wn 1 = ( 1)n 1
:
(d)
If 1 k n 1

n
X
n 1 X
n 1 X
n 1
1 e2 ki=n 1 e2 ki
2 ji=n k 2 ki=n j
wjk = e = e = = = 0:
j=0 j=0 j=0
1 e2 ki=n 1 e2 ki=n

27
If k = n

X
n 1 X
n 1 X
n 1 X
n 1
2 ji=n n
wjk = e = 2 ji
e = 1 = n:
j=0 j=0 j=0 j=0

28
I.2.7
Fix R > 1 and n 1, m 0. Show that

zm Rm
; jzj = R:
zn + 1 Rn 1
Sketch the set where equality holds. Hint. See (1.1) p.2.

Solution

I.2.7 I.2.7 (n = 1) I.2.7 (n = 2)

1 1

a+b b
-1 1 -1 1
a
-1 -1

I.2.7 (n = 3) I.2.7 (n = 4) I.2.7 (n = 5)

1 1 1

-1 1 -1 1 -1 1

-1 -1 -1

We use that jz wj jzj jwj see (1.1) on page 2 in CA, and have that

jz n + 1j > jz n j 1 = Rn 1;
where the last equaltity is due to that jzj = R.
Some rearrangement gives
1 1
;
jz n + 1j (Rn 1)

29
and mulitplication by jz m j = Rm gives

zm Rm
; jzj = R:
zn + 1 Rn 1
For equality, we must have

jz n + 1j = Rn 1;
because jzjm = Rm .
We rearrarange this equality to

j 1j + jz n + 1j = jz n j

and we conclude that 1 and z n +1 must lie on the same ray. (We have used
the fact that jaj+jbj = ja + bj implies that a; b lie on the same ray, see …rst …gure I.2.7)
If

z = Rei
then

z n = Rn ein :
Since R > 1, we require that ein = 1 thus

i=n
z = wk Re ;
where wk is an n th root of unity.
If

i=n
z = wk Re ;
then

zn + 1 = Rn + 1;
and

zm Rm
= :
zn + 1 Rn 1

30
I.2.8
Show that cos 2 = cos2 sin2 and sin 2 = 2 cos sin using de
Moivre’s formulae. Find formulae for cos 4 and sin 4 in terms of
cos and sin .

Solution
Let 2 R be given. Then by de Moivre’s formulae (on page 8 in CA) we
have for all n 2 Z

cos n + i sin n = (cos + i sin )n

Hence for n = 2 we get

cos 2 + i sin 2 =
= (cos + i sin )2 = cos2 + i2 cos sin sin2 =
= cos2 sin2 + i (2 cos sin ) ;

then, by setting the real and imaginary parts equal to each other we obtain

cos 2 = cos2 sin2

sin 2 = 2 cos sin :

Similarly, applying de Moivre’s formulae for n = 4 we get (using the Binomial

Theorem)

cos 4 + i sin 4 = (cos + i sin )4 =

= cos4 + i4 cos3 sin 6 cos2 sin2 i4 cos sin3 + sin4 =
= cos4 6 cos2 sin2 + sin4 + i 4 cos3 sin 4 cos sin3 ;

then, by setting the real and imaginary parts equal to each other we obtain

cos 4 = cos4 6 cos2 sin2 + sin4

sin 4 = 4 cos3 sin 4 cos sin3 :

as required.

31
I.3.1
Sketch the image under the spherical projection of the following
sets on the sphere
(a) the lower hemisphere Z 0
(b) the polar cap 43 Z 1
p p
(c) lines of lattitude X = 1 Z 2 cos , Y = 1 Z 2 sin , for Z …xed
and 0 2 p p
(d) lines of longitude X = 1 Z 2 cos , Y = 1 Z 2 sin , for …xed
and 1 Z 1
(e) the spherical cap A X 1, with center lying on the equator,
for …xed A. Separate into cases, according to various ranges of A.

Solution

I.3.1a I.3.1b I.3.1c

2 4 2
y y y
1 2 1

-2 -1 1 2 -4 -2 2 4 -2 -1 1 2
-1 -2 -1

-2 -4 -2

I.3.1d I.3.1e (A = -1/2) I.3.1e (A = 0)

2 4 2
y y y
1 2 1
θ

-2 -1 1 2 -6 -4 -2 2 -2 -1 1 2
-1 -2 -1

-2 -4 -2

32
 I.3.1e (A = 1/2)
4
y
2

-2 2 4 6
-2

-4

(a)
Image is the unit disk.
(b) p
Set Y = 0 and Z = 3=4 in X 2 + Y 2 + Zp 2
= 1, we have X = 7=4. We have
the formula x = X= (1 p Z), thus x = 7. Image is the exterior of a disk,
with centre 0 and radius 7. p p
(c) Image
p is the disk, with centre
p 0 and radius 1 + z= 1 z. We have that
2
X = 1 Z cos and 2
Y = p 1 Z sin , we can take the radius to image
p
1 Z2
for = 0, thus r = 1 Z = p1+Z 1 Z
.
(d) Image is lines of longitude issuing from 0.
(e)
Case 1: 1 < A < 0 p
Image is the exterior of the disk centered at 1=A with radius 1 A2 = jAj.
Case 2: A = 0
Image is the left half–plane.
Case 3: 0 < A < 1 p
Image is a disk centered at 1=A p with radius p 1 A2 = jAj.
Set Y = 0 then Z goes from 1 A2 top 1 A2 . We have p the formula
x = X= (1 Z) thus x goes from A= 1 + 1 A = 1 2 1 A2 =A to
p p
A= 1 1 A2 = 1 + 1 A2 =A.

33
I.3.2
If the point P on the sphere corresponds to z under the stereo-
graphic projection, show that the antipodal point P on the sphere
corresponds to 1=z.

Solution
For z = x + iy the corresponding point on the sphere under transformation
given on p. 12 in CA is given by (X; Y; Z) where

2x
X = ;
jzj2 + 1
2y
Y = 2 ;
jzj + 1
jzj2 1
Z = :
jzj2 + 1
For z 6= 0, we have
1 1 x + iy x + iy
x y
= = = 2 = i 2
z x iy (x iy) (x + iy) x2 + y 2
jzj jzj
which corresponds to the point on the sphere given by (X 0 ; Y 0 ; Z 0 ) where

2x
0 jzj2 2x
X = = = X;
1 2 2
jzj + 1
z
+ 1
2y
jzj2 2y
Y0 = = = Y;
1 2 2
jzj + 1
z
+ 1
1 2
0 z
1 jzj2 1
Z = = = Z:
1 2 jzj2 + 1
z
+ 1
Hence, the map (X; Y; Z) 7 ! ( X; Y; Z) on the sphere corresponds to
z 7 ! 1=z of C.

34
I.3.3
Show that as z travers a small circle in the complex plane in the
positive (counterclockwise) direction, the corresponding point P
on the sphere traverses a small circle in the negative (clockwise)
direction with respect to someone standing at the center of the
circle and with body outside the sphere. (Thus the stereographic
projection is orientation reversing, as a map from the sphere with
orientation determined by the unit outer normal vector to the com-
plex plane with the usual orientation.)

Solution
Draw the picture. Or argue as follows. The orientation of the image circle
is the same for all circles on the sphere orientated so that N is outside the
circle. This can be seen by moving one circle continuously to the other, and
seeing that the image circles moves continuously. Thus we need to shrink it
only for the equator of the sphere oriented as indicated ( ), if the South Pole
is inside it. And the image circle is the unity and positive direction of the
unit circle ( ) is the converse. So the orientation of the image is clockwise
(negative).

35
I.3.4
Show that a rotation of the sphere of 180 about the X-axis corre-
sponds under stereographic projection to the inversion z 7 ! 1=z of
C.

Solution
For z = x + iy the corresponding point on the sphere under transformation
given on p. 12 in CA is given by (X; Y; Z) where

2x
X = ;
jzj2 + 1
2y
Y = 2 ;
jzj + 1
jzj2 1
Z = :
jzj2 + 1

For z 6= 0, we have
1 1 x iy x iy x y
= = = 2 = i
z x + iy (x + iy) (x iy) x + y2 jzj2 jzj2
which corresponds to the point on the sphere given by (X 0 ; Y 0 ; Z 0 ) where

2x
0 jzj2 2x
X = = = X;
1 2 jzj2 + 1
z
+ 1
2y
jzj2 2y
Y0 = = = Y;
1 2 2
jzj + 1
z
+ 1
1
jzj2
1 jzj2 1
Z0 = 1 = = Z:
jzj 2 + 1 jzj2 + 1

Hence, the map (X; Y; Z) 7 ! (X; Y; Z) on the sphere corresponds to

inversion z 7 ! 1=z of C. Moreover, the map (X; Y; Z) 7 ! (X; Y; Z) is
given by rotation of the sphere by 180 about the X-axis.

36
I.3.5
Suppose (x; y; 0) is the spherical projection of (X; Y; Z). Show that
the product of the distances from the north pole N to (X; Y; Z) and
from N to (x; y; 0) is 2. What is the situation when (X; Y; Z) lies on
the equator on the sphere?

Solution
The distance from from the north pole N = (0; 0; 1) to (X; Y; Z) is
q
X 2 + Y 2 + (Z 1)2 ;
X
and the distance from the north pole N = (0; 0; 1) to (x; y; 0) = ; Y ;0
1 Z 1 Z
is
s
X2 Y2
+ + 1:
(1 Z)2 (1 Z)2
The product of distances is

1=2
1=2 X2 Y2 X 2 + Y 2 + (Z 1)2
X 2 + Y 2 + (Z 1)2 + +1 = = 2:
(1 Z)2 (1 Z)2 1 Z

When (X; Y; Z) lie on the equator, the product is simply the square of the
distance from N to a point on the equator. By the Pythagorean Law, this is
1 + 1 = 2.

37
I.3.6
We de…ne the chordal distance d (z; w) between two points z; w 2 C
to be the length of the straight line segment joining to the points
P and Q on the unit sphere whose stereographic projections are z
and w, respectively.
(a) Show that the chordal distance is a metric, that is, it is sym-
metric, d (z; w) = d (w; z); satis…es the triangle inequality d (z; w)
d (z; ) + d ( ; w); and d (z; w) = 0 if and only if z = w.
(b) Show that the chordal distance from z to w is given by

2 jz wj
d (z; w) = q q ; z; w 2 C:
2 2
1 + jzj 1 + jwj
(c) What is d (z; 1)? Remark. The expression for d (z; w) shows
that in…nitesimal arc length corresponding to the chordal metric
is given by
2ds
d (z) = ;
1 + jzj2
where ds = jdzj is the usual Euclidean in…nitesimal arc length. The
in…nitesimal arc length d (z) determines another metric, the spher-
ical metric (z; w), on the extended complex plane. See Section
IX.3.

Solution
(a)
Follows from fact that Euclidean distance in R3 is a metric on the sphere.
(b)
Set z = x1 + iy1 ; w = x2 + iy2 , we have

jz wj2 =
= (z w) (z w) = (z w) (z w) =
2 2
= zz zw zw + ww = jzj + jwj zw zw =
2 2
= jzj + jwj (x1 + iy1 ) (x2 iy2 ) (x1 iy1 ) (x2 + iy2 ) =
= jzj + jwj2
2
2x1 x2 2y1 y2 : (1)

38
We take square of distance between P and Q

2 2 2
d (z; w)2 = (X X 0 ) + (Y Y 0 ) + (Z Z 0) =
= X 2 + Y 2 + Z 2 + X 02 + Y 02 + Z 02
2 (XX 0 + Y Y 0 + ZZ 0 ) =
!
4x1 x2 + 4y1 y2 + jzj2 1 jwj2 1
=2 2 =
jzj2 + 1 jwj2 + 1
!
jzj2 + 1 jwj2 + 1 4x1 x2 4y1 y2 jzj2 1 jwj2 1
=2 =
jzj2 + 1 jwj2 + 1
!
2jzj2 + 2 jwj2 4x1 x2 4y1 y2 (1) 4 jz wj2
=2 = :
jzj2 + 1 2
jwj + 1 2
jzj + 1 jwj2 + 1

Taking the positive square root we have

2 jz wj
d (z; w) = q q ; z; w 2 C:
1 + jzj2 1 + jwj2
(c)

2 jz wj 2 jz=w 1j 2
d (z; 1) = lim q q = lim q q =q :
w!1 2 2 w!1 2 2 2
1 + jzj 1 + jwj 1 + jzj 1= jwj + 1 1 + jzj

39
I.3.7
Consider the sphere of radius 21 in (X; Y; Z) space, resting on the
(X; Y; 0) plane, with south pole at the origin (0; 0; 0) and north pole
at (0; 0; 1). We de…ne a stereographic projection of the sphere onto
the complex plane as before, so that corresponding points (X; Y; Z)
and z (x; y; 0) lie on the same line through the north pole. Find
the equations for z = x + iy in terms of X; Y; Z, and the equations
for X; Y; Z in terms of z. What is the corresponding formula for the
chordal distance? Note. In this case, the equation of the sphere is
2
X 2 + Y 2 + Z 12 = 41 .

Solution
The line through P = (X; Y; Z) and N = (0; 0; 1) is on the form N +
t (P N ), and it meats the xy plane when

(x; y; 0) = (0; 0; 1) + t ((X; Y; Z) (0; 0; 1)) = (tX; tY; 1 + t (Z 1)) :

By simultaneous equations we have, and solve for X, Y and Z

8 8
< x = tX < X = xt
y = tY ) Y = yt
: :
1 + t (Z 1) = 0 Z = 1 1t
1 2 1
We solve for the t that is a point on the sphere X 2 + Y 2 + Z 2
= 4
we
have

2 2
x2 y 2 1 1 1 t t2
+ 2 + = , jzj2 + 1 = ,
t2 t 2 t 4 2 4
, jzj2 t + 1 = 0 , t = jzj2 + 1

thus we have
8 x
> X = 1+jzj
< y
2

Y = 1+jzj2
>
: Z = jzj2 :
1+jzj2

40
Set z = x1 + iy1 ; w = x2 + iy2 , we have

jz wj2 =
= (z w) (z w) = (z w) (z w) =
2 2
= zz zw zw + ww = jzj + jwj zw zw =
2 2
= jzj + jwj (x1 + iy1 ) (x2 iy2 ) (x1 iy1 ) (x2 + iy2 ) =
= jzj2 + jwj2 2 (x1 x2 + y1 y2 ) : (1)

And the coordical distance

41
 d (z; w)2 = (X1 X 2 )2 + (Y1 Y 2 )2 + (Z1 Z 2 )2 =
!2
2 2
x1 x2 y1 y2 jz1 j2 jz2 j2
= + + =
1 + jzj2 1 + jwj2 1 + jzj2 1 + jwj2 1 + jzj2 1 + jwj2
2 2 2
x1 1 + jwj2 x2 1 + jzj2 + y1 1 + jwj2 y2 1 + jzj2 + jzj2 jwj2
= 2 2 =
1 + jzj2 1 + jwj2
2 2
(x21 + y12 ) 1 + jwj2 + (x22 + y22 ) 1 + jzj2 + jzj4 2 jzj2 jwj2 + jwj4
= 2 2 +
1 + jzj2 1 + jwj2
2x1 x2 1 + jwj2 1 + jzj2 2y1 y2 1 + jwj2 1 + jzj2
+ 2 2 =
1 + jzj2 1 + jwj2
2 2
jzj2 1 + jwj2 + jwj2 1 + jzj2 + jzj4 2 jzj2 jwj2 + jwj4
= 2 2 +
1 + jzj2 1 + jwj2
2 (x1 x2 + y1 y2 ) 1 + jwj2 1 + jzj2
+ 2 2 =
1 + jzj2 1 + jwj2
jzj2 + 2 jzj2 jwj2 + jzj2 jwj4 + jwj2 + 2 jwj2 jzj2 + jwj2 jzj4 + jzj4 2 jzj2 jwj2 + jwj4
= 2 2 +
1 + jzj2 1 + jwj2
2 (x1 x2 + y1 y2 ) 1 + jwj2 1 + jzj2
+ 2 2 =
1 + jzj2 1 + jwj2
jzj2 + jwj2 1 + jwj2 1 + jzj2 2 (x1 x2 + y1 y2 ) 1 + jwj2 1 + jzj2
= 2 2 =
1 + jzj2 1 + jwj2
jzj2 + jwj2 2 (x1 x2 + y1 y2 ) 1 + jwj2 1 + jzj2
= 2 2 =
1 + jzj2 1 + jwj2
jzj2 + jwj2 2 (x1 x2 + y1 y2 ) (1)
= 2 2 =
1 + jzj 1 + jwj
(1) jz wj2
= :
jzj2 + 1 jwj2 + 1

This gives

42
 jzwj
d (z; w) = q q :
2 2
jzj + 1 jwj + 1

43
I.4.1
Sketch each curve and its image under w = z 2 .
(a) jz 1j = 1 (c) y = 1 (e) y 2 = x2 1; x > 0
(b) x = 1 (d) y = x + 1 (f) y = 1=x; x 6= 0

Solution

I.4.1a z-plane I.4.1b z-plane I.4.1c z-plane

y 4 y 4 y 4

2 2 2

-4 -2 2 4 -4 -2 2 4 -4 -2 2 4
-2 x -2 x -2 x
-4 -4 -4

I.4.1a w-plane I.4.1b w-plane I.4.1c w-plane

y 4 y 4 y 4

2 2 2

-4 -2 2 4 -4 -2 2 4 -4 -2 2 4
-2 x -2 x -2 x
-4 -4 -4

I.4.1d z-plane I.4.1e z-plane I.4.1f z-plane

y 4 y 4 y 4

2 2 2

-4 -2 2 4 -4 -2 2 4 -4 -2 2 4
-2 x -2 x -2 x
-4 -4 -4

44
I.4.1d w-plane I.4.1e w-plane I.4.1f w-plane
y 4 y 4 y 4

2 2 2

-4 -2 2 4 -4 -2 2 4 -4 -2 2 4
-2 x -2 x -2 x
-4 -4 -4

45
I.4.2
Sketch the image of eachpcurve in the preceding problem under the
principal branch of w = z, and also sketch, on the same grid but
in apdi¤erent color, the image of each curve under the other branch
of z.

Solution
1=2 n 1=2
p
w = z 1=2 = jzj ei argjzj = jzj ei Argjzj+i2 = jzjei Arg z=2

I.4.2a z-plane I.4.2b z-plane I.4.2c z-plane

y 4 y 4 y 4

2 2 2

-4 -2 2 4 -4 -2 2 4 -4 -2 2 4
-2 x -2 x -2 x
-4 -4 -4

I.4.2a w-plane I.4.2b w-plane I.4.2c w-plane

y 4 y 4 y 4

2 2 2

-4 -2 2 4 -4 -2 2 4 -4 -2 2 4
-2 x -2 x -2 x
-4 -4 -4

46
I.4.2d z-plane I.4.2e z-plane I.4.2f z-plane
y 4 y 4 y 4

2 2 2

-4 -2 2 4 -4 -2 2 4 -4 -2 2 4
-2 x -2 x -2 x
-4 -4 -4

I.4.2d w-plane I.4.2e w-plane I.4.2f w-plane

y 4 y 4 y 4

2 2 2

-4 -2 2 4 -4 -2 2 4 -4 -2 2 4
-2 x -2 x -2 x
-4 -4 -4

47
I.4.3
(a) Give a brief description of the function z 7 ! w = z 3 , considered
as a mapping from the z plane to the w- plane. (Describe what
happens to w as z traverses a ray emanating from the origin, and as
z traverses a ray a circle centered at the origin.) (b) Make branch
cuts and de…ne explicitly three branches of the inverse mapping.
(c) Describe the construction of the Riemann surface of z 1=3 .

Solution
(a)
The function w = f (z) = z 3 . For z = rei , we have z 3 = r3 ei3 . The radical
rays at angle are mapped to rays at angle 3 , that is, arg w = 3 arg z. The
magnitude of a complex number is cubed, jwj = jzj3 . Circles, centered at the
origin of radius r, are mapped to cocentric circles with radius r3 .
(b)
We make branch cuts at ( 1; 0],

1=3 n 1=3
p
z = w1=3 = jwj ei arg w = jwj ei Arg w+i2 = ei2 n=3
jwjei Arg w=3

we choose
g (z) = z 1=3 = r1=3 ei =3 , < < .
Sheet 1 : Take f1 (z) = g (z),
Sheet 2 : Take f2 (z) = e2 i=3 g (z),
Sheet 3 : Take f3 (z) = e4 i=3 g (z).

(c) Top edge of cut on sheet 1 to bottom edge of cut on sheet 2. Top edge of
cut on sheet 2 to bottom edge of cut on sheet 3. Top edge of cut on sheet 3
to bottom edge of cut on sheet 1. The endpoints for f (z) is continuous on
surface

Spara

48
 I.4.3a ±
I.2.1b I.2.1c

±
±
±±

1 1

±
±

±
±
±

±
±±

±
±

±
±

±±±±±±±

±
±

-1 1 -1 1
±±
±± ±

-1 -1
±

Spara

49
I.4.4
Describe how to construct thep Riemann surfaces for the 2=5
following functions
(a) w = z 1=4 ; (b) w = z i; (c) w = (z 1) :
Remark. To describe the Riemann surface of a multivalued function, begin
with one sheet for each branch of the function, make branch cuts so that the
branches are de…ned continuously on each sheet, and identify each edge of
a cut on one sheet to another appropriate edge so that the function values
match up continuously.

Solution
(a) Use four sheets, can make branch cuts along real axis from 1 to 0.
(b) Use two sheets, can make branch cuts along horizontal line from 1 + i
to i.
(c) Use …ve sheets, can make branch cuts along real axis from 1 to 1.

50
I.5.1
Calculate and plot for ez for the following points z.
(a) 0 (c) (i 1) =3 (e) i=m; m = 1; 2; 3; : : :
(b) i + 1 (d) 37 i (f) m (i 1) m = 1; 2; 3; : : :

Solution

I.5.1e I.5.1f

1 0.2

-1 1 -0.4 -0.2 0.2

-1 -0.2

(a)

e0 = 1:
(b)

i+1
e = e i e1 = e:
(c)
p !
(i 1)3 =3 i=3 =3 1 3
e =e e =e + i = 0:351 + 0:006i:
2 2
(d)

e37 i = e36 i e i = 1:
(e)
i=m
We take the limit for the sequence e , m = 1; 2; 3; : : : as m ! 1, and
have

i=m
e ! 1;

51
as m ! 1. Because e i=m = 1 the sequence approaches its limit along a
circle with radius 1.
(f)
We take the limit for the sequence em(i 1) = e m emi , m = 1; 2; 3; : : : as
m ! 1, and have

m mi
e e ! 0;
m mi m
as m ! 1. Because je e j=e the sequence spiraling to its limit in
origo.

52
I.5.2
Sketch each of the following …gures and its image under the expo-
nential map w = ez . Indicate the images of horizontal and vertical
lines in your sketch.
(a) the vertical strip 0 < Re z < 1,
(b) the horizontal strip 5 =3 < Im z < 8 =3,
(c) the rectangle 0 < x < 1, 0 < y < =4,
(d) the disk jzj =2,
(e) the disk jzj ,
(f) the disk jzj 3 =2.

Solution

I.5.2a z-plane I.5.2b z-plane I.5.2c z-plane

4 3
8

2 6 2

4
-4 -2 2 4 1
-2 2

-4 -4 -2 0 2 4 1 2 3

I.5.2a w-plane I.5.2b w-plane I.5.2c w-plane

4 4 3

2 2
2

-4 -2 2 4 -4 -2 2 4 1
-2 -2

-4 -4 1 2 3

53
I..5.2d z-plane I.5.2e z-plane I.5.2f z-plane
4 4 4

2 2 2

-4 -2 2 4 -4 -2 2 4 -4 -2 2 4
-2 -2 -2

-4 -4 -4

I..5.2d w-plane I.5.2e w-plane I.5.2f w-plane

20 100
4

2 10

-4 -2 2 4 -10 10 20 30 100
-2 -10
-4
-20 -100

54
I.5.3
Show that ez = ez .

Solution

ez = ex iy
= ex e iy
= ex (cos ( y) + i sin ( y)) = ex (cos y i sin y) =
= ex (cos y + i sin y) = ex (cos y + i sin y) = ex eiy = ex+iy = ez :

55
I.5.4
Show that the only periods of ez are the integral multiples of 2 i,
that is, if ez+ = ez for all z, then is an integer times 2 i.

Solution
Show that the only periods of ez are the integral muliples of 2 i that is, if
ez+ = ez for all z, then is an integer times 2 i.

ez = ez+ = ez e ) e = 1 ) = 2 mi

56
I.6.1
Find and plot log z for the following complex numbers z. Specify
the principal value. p
(a) 2 (b) i (c) 1 + i (d) 1 + i 3 =2

Solution
(a)
Suppose that n = 0; 1; 2; : : :

log 2 = log j2j + i Arg 2 + 2 ni = log 2 + i2 n:

(b)
Suppose that n = 0; 1; 2; : : :

log i = log jij + i Arg i + 2 ni = i =2 + i2 n:

(c)
Suppose that n = 0; 1; 2; : : :

log (1 + i) =
p
= log j1 + ij + i Arg (1 + i) + 2 ni = log 2 + i =4 + i2 n =
= log 2=2 + i =4 + i2 n:

(d)
Suppose that n = 0; 1; 2; : : :

p
1+i 3
log 2
=
p p
1+i 3 1+i 3
= log 2
+ i Arg 2
+ 2 ni =
= i=3 + i2 n:

57
I.6.1a I.6.1b I.6.1c
20 20 20

10 10 10

-1 1 2 -1 1 2 -1 1 2
-10 -10 -10

-20 -20 -20

I.6.1d
20

10

-1 1 2
-10

-20

58
I.6.2
Sketch the image under the map w = Log z of each of the following
…gures.
(a) the right halv-plane Re z > 0,
(b) the half-disk jzj < 1, Re z > 0,
(c) the unit circle jzjp= 1, p
(d) the slit annulus e < jzj < e2 , z 62 ( e2 ; e),
(e) the horizontal line y = e,
(f) the vertical line x = e.

Solution
(b) We have a disk with radius less then 1, this means jzj < 1, thus log jzj
< 0 and is unbounded, therefore, it goes from 0 to 1. Since Re (z) > 0,
the polar angle is between 2 and 2 .
(d) Here, we have an annulus, the polar angle is from to , thus the argu-
ment is in this range. Since, the log function is the inverse of the exponential
map, circles in z plane are stright lines in the w plane (w = log z).
Therefore, the image p of this annulus under the log function is the rectangle,
bounded by x = log j ej = 12 and x = log je2 j = 2.

I.6.2a z-plane I.6.2b z-plane I.6.2c z-plane

3 4
2
2
2
1

-3 -2 -1 1 2 3 -2 2 -4 -2 2 4
-1
-2
-2 -2
-3 -4

59
I.6.2a w-plane I.6.2b w-plane I.6.2c w-plane
3
π
2 π/2
1

-3 -2 -1 1 2 3
-1
−π/2
-2
−π
-3

I.6.2d z-plane I.6.2e z-plane I.6.2f z-plane

8 4 4
6
4 2 2
2

-8 -6 -4 -2
-2 2 4 6 8 -4 -2 2 4 -4 -2 2 4
-4 -2 -2
-6
-8 -4 -4

I.6.2d w-plane I.6.2e w-plane I.6.2f w-plane

4 4 4

2 2 2

-4 -2 2 4 -4 -2 2 4 -4 -2 2 4
-2 -2 -2

-4 -4 -4

60
I.6.3
De…ne explicitly a continuous branch of log z in the complex plane
slit along the negative imaginary axis, Cn [0; i1).

Solution

I.6.3

We have log z = log rei = log r + i . To avoid the negative imaginary axis
we chose =2 < < 3 =2. We use the branch

f rei = log r + i ; =2 < < 3 =2;

of log z.

61
I.6.4
How would you make a branch cut to de…ne a single-valued branch
of the function log (z + i 1)? How about log (z z0 )?

Solution
Any straight line cut from z0 to 1, in any direction, will do.

62
I.7.1
Find all values and plot them.
p (1 i)
(a) (1 + i)i (b) ( i)1+i (c) 2 1=2
(d) 1+i 3

Solution
(a)
Suppose that m; n = 0; 1; 2; : : :

(1 + i)i =
p p
= ei log(1+i) = ei(logj1+ij+i Arg(1+i)+i2 m)
= ei(log 2+i =4+i2 m)
=e 2 m
e =4 i log
e 2
= [m = n] =
p
= e2 n e =4 i log
e 2
:

(b)
Suppose that m; n = 0; 1; 2; : : :

( i)1+i =
= e(1+i) log( i)
= e(1+i)(logj ij+i Arg( i)+i2 m)
= e(1+i)( i =2+i2 m)
=
2 m =2 i( =2+2 m)
=e e e = [m = n] = e2 n e =2 i(
e =2 2 n)
=
= ie2 n e =2
:

(c)
Suppose that m = 0; 1; 2; : : :

1=2
2 =
log 2=2 (logj2j+i Arg 2+i2 m)=2 (log 2+i2 m)=2 log 2=2 i m
=e =e =e =e e =
p
= 1= 2:

(d)
Suppose that n = 0; 1; 2; : : :

63
 p 1 i
1+i 3 =
p p p
i) log(1+i 3) i)(logj1+i 3j+i Arg(1+i 3)+i2 n)
= e(1 = e(1 =
= e(1 i)(log 2+i =3+i2 n)
= e2 n elog 2+ =3 i(
e log 2+ =3+2 n)
=
= e2 n elog 2+ =3 i(
e log 2+ =3)
:

64
I.7.2 h i
2i
Compute and plot log (1 + i) .

Solution
We rewrite the expression, suppose that k; m; n = 0; 1; 2; : : :

h i
log (1 + i)2i =
h i
log(1+i) 2i
= log e = log e2i log(1+i) = log e2i(logj1+ij+i Arg(1+i)+i2 k) =
h p i
= log e2i(log 2+i =4+i2 k) = log e 4 k e =2 ei log 2 = [k = m] = log e4 m e =2 i log 2
e =
= log e4 m
e =2 i log 2
e +i Arg e4 m
e =2 i log 2
e +i2 n = 4 m =2+i log 2+i2 n =
= =2 + i log 2 + 4 m + 2 in:

Thus the set is a square lattice.

I.7.2
20

10

-10 10 20
-10

-20

65
I.7.3
Sketch the image of the sector f0 < arg z < =6g under the map w =
z a for
(a) a = 32 (b) a = i (c) a = i + 2
Use only the principal branch of z a .

Solution

I.7.3a z-plane I.7.3b z-plane I.7.3c z-plane

5 5 5

-5 5 -5 5 -5 5

-5 -5 -5

I.7.3a w-plane I.7.3b w-plane I.7.3c w-plane

5 2

-5 5 -2 2

1
-5 -2

Some points on the y-axis is marked, and the point (1; 0) too.

66
I.7.4
Show that (zw)a = z a wa , where on the right we take all possible
products.

Solution
Let 2 (zw)a , and k; m; n = 0; 1; 2; : : :

=
= ea log(zw) = ea(logjzwj+i Arg(zw)+i2 n)
= ea(logjzj+logjwj+i Arg z+i Arg w+i2 m+i2 n)
=
a(logjzj+i Arg z+i2 (m+n)) a(logjwj+i Arg w) a a
=e e 2z w :

Conversely, if 2 z a wa , say

=
= ea log z ea log w = ea(logjzj+i Arg z+i2 m) a(logjwj+i Arg w+i2 n)
e =
= ea(logjzj+logjwj+i Arg z+i Arg w+i2 (m+n))
= ea(logjzwj+i Arg(zw)+i2 k+i2 (m+n))
=
a(logjzwj+i Arg(zw)+i2 (k+m+n))
=e =
= ea log(zw) 2 (zw)a

We have that (zw)a = z a wa as sets

67
I.7.5
i
Find ii . Show that is does not coincide with ii i = i 1 .

Solution
We rewrite the expression ii , suppose that k = 0; 1; 2; : : :

ii =
= ei log i = ei(logjij+i Arg(i)+i2 k)
= ei(i =2+i2 k)
=
2 k =2
=e :
i
Now we can …nd ii we use that we have ii , suppose that k; m = 0; 1; 2; : : :

i
ii =
= ii
i
= ei log(i ) = ei(log(e
i 2 k =2
)) = ei(logje 2 k =2
j+i Arg(e 2 k =2
)+i2 m)
=
i( 2 k =2+i2 m) 2 m i( 2 k =2) 2 m
=e =e e = ie =
= ie2 m
:

Which not coincide with ii i = i 1

= 1=i = i which is smaller than
f ie2 m ; m = 0; 1; 2; : : :g.

68
I.7.6
Determine the phase factors of the function z a (1 z)b at the branch
points z = 0 and z = 1. What conditions on a and b guarantee that
z a (1 z)b can be de…ned as a (continuous) single-valued function
on Cn [0; 1]?

Solution
Phase factors is e2 ia at 0 and e2 ib at 1. Require e2 ia e2 ib = 1, or a + b = n,
where n 2 Z , to have a continuous single–valued determination of z a (1 z)b
on Cn [0; 1].

69
I.7.7
Let x1 < x2 < < xn be n consecutive
p points on the real axis.
Describe the Riemann surface of (z x1 ) (z xn ). Show that
for n = 1 and n = 2 the surface is topologically a sphere with certain
punctures corresponding to the branch points and 1. What is it
when n = 3 or n = 4? Can you say anything for general n? (Any
compact Riemann surface is topologically a sphere with handles.
Thus a torus is topologically a sphere with one handle. For a given
n, how many handles are there, and where do they come from?)

Solution p p
For n = 1 and n = 2, the functions f1 p= z x1 and f2 = (z x1 ) (z x2 ),
and for a general n we have fn (x) = (z x1 ) (z x2 ) : : : (z xn ). To con-
stuct the Riemann Suface we use two sheets with slits [x1 ; x2 ] ; [x3 ; x4 ] ; : : :.
If n is odd, we also need a slit [xn ; +1) . Identify top edge of slit on one
sheet with bottom edge of slit on other sheet. If n = 1 and n = 2 the surface
is topologically a sphere with certain punctures corresponding to the branch
points and 1. If n = 3 or n = 4, surface is a torus. For a general even
n, fn (x) will be continuous on Cn ([x1 ; x2 ] [ [x3 ; x4 ] [ : : : [ [xn 1 ; xn ]), and
its Riemann surface will be a sphere with n2 1 holes. For a general odd n
though, fn (x) will be continuous on Cn ([x1 ; x2 ] [ [x3 ; x4 ] [ : : : [ [xn ; +1]),
its Riemann surface will be a sphere with n 2 1 holes.

70
I.7.8 p
Show that z 2 1=z can be de…ned as a (single-valued) continuous
function outside the unit disk, that is, for jzj > 1. Draw branch
cuts so that the function can be de…ned continuous o¤ the branch
cuts. Describe the Riemann surface of the function.

Solution p
The function is z 1 1=z 3 . If jzj > 1, can use the principal value of the
square root to de…ne a branch of the function. There are branch points at
z = 0 and z 3 = 1, that is at 0; 1; e2 i=3 and e 2 i=3 . Make two branch cuts by
connecting any two pairs of point by curves; for instance, connect 0 to 1 by
a straight line, and the other cube roots of unity by a straight line or arc of
unit circle. The resulting two-sheeted surface with identi…cation of cuts and
with points at in…nity is a torus.

71
I.7.9 q
Consider the branch of the function z (z 3 1) (z + 1)3 that is posi-
tive at z = 2. Draw branch cuts so that this branch of the function
can be de…ned continuously o¤ the branch cuts. Describe the Rie-
mann surface of the function. To what value at z = 2 does this
branch return if it is continued continuously once counterclockwise
around the circle fjzj = 2g?

Solution
We have that the function is
p p
(z + 1) z 2 z (1 1=z 3 ) (1 + 1=z):
If jzj > 1, the second square root can be de…ned to be single-valued for
jzj > 1. The value of the function at z = 2 returns to thepnegative of their
initial value then we travers the circle jzj = 2 , because z does. We can
construct a Riemann Surface by making cuts at 1; 0; 1; e2 i=3 ; e 2 i=3 and
1. The function for the Riemann Surface is
q
(z + 1) z (z 1) (z + 1) (z e2 i=3 ) (z e 2 i=3 ):

72
I.7.10 q
Consider the branch of the function z (z 3 1) (z + 1)3 (z 1) that is
positive at z = 2. Draw branch cut so that this branch of the func-
tion can be de…ned continuously o¤ the branch cuts. Describe the
Riemann surface of the function. To what value at z = 2 does this
branch return if it is continued continuously once counterclockwise
around the circle fjzj = 2g?

Solution
We have that the function is
q
z (1 1=z 3 ) (1 + 1=z)3 (1
4
1=z)
The branch that is positive for z = 2 is, it is de…ned continuously for jzj > 1.
Branch return to original value around circle jzj = 2. We can construct a
Riemann Surface by making cuts at 0; 1; e2 i=3 and e 2 i=3 .
The function for the Riemann Surface is
q
(z 1) (z + 1) z (z + 1) (z e2 i=3 ) (z e 2 i=3 ):

73
I.7.11 p
Find the branch points of 3 z 3 1 and describe the Riemann surface
of the function.

Solution

I.7.11a I.7.11b I.7.11c

1 1 1

-1 1 -1 1 -1 1

-1 -1 -1

We rewrite as follows
p q
3 3
z3 1= (z 1) (z e2 i=3 ) (z e 2 i=3 ):

This equation will have 3 branch points, which are the cube root of unity.
So the phase factor is
e2 i=3 . The Riemann surface is obtained ty pasting three sheets with the
corresponding branch cuts, we end up with a one hole torus.
Make two cuts, from 1 to e 2 i=3 , on each sheet. In this case the cuts share
common endpoint.
We nee three shets where f0 (z), f1 (z) = e2 i=3 f0 (z) and f2 (z) = e 2 i=3 f0 (z)
Note: Can use Riemann–formula to see that surface is a torus.
Cheek by going around each little tip what phase change to, which of for
your sheet.

74
I.8.1
Establish the following addition formulae
(a) cos (z + w) = cos z cos w sin z sin w;
(b) sin (z + w) = sin z cos w + cos z sin w;
(c) cosh (z + w) = cosh z sinh w + sinh z sinh w;
(d) sinh (z + w) = sinh z cosh w + cosh z sinh w;

Solution
(a)
Using the de…nitions of sine and cosine functions given on page 29 in CA we
have,

cos (z + w) =
ei(z+w) + e i(z+w) 1
= = 2ei(z+w) + 2e i(z+w) =
2 4
1 i(z+w) i(z w)
= e +e + e i(z w) + e i(z+w) +
4
+ ei(z+w) ei(z w) e i(z w) + e i(z+w) =
eiz + e iz eiw + e iw eiz e iz eiw e iw
= =
2 2 2i 2i
= cos z cos w sin z sin w:

(b)
Using the addition formula (a), and using that sin z = cos (z =2) and
cos z = sin (z =2) we have,

sin (z + w) = cos z + w =
2
= cos z cos w sin z sin w =
2 2
= cos z sin w + sin z cos w = sin z cos w + cos z sin w:

(c)
Using the addition formula (a) and the formulas cos (iz) = cosh z and sin (iz) =
i sinh z given on page 30 in CA we have,

75
 cosh (z + w) = cos (i (z + w)) = cos (iz) cos (iw) sin (iz) sin (iw) =
= cosh z cosh w i sinh z i sinh w = cosh z cosh w + sinh z sinh w:

(d)
Using the addition formula (b) and the formulas cos (iz) = cosh z and sin (iz) =
i sinh z given on page 30 in CA we have,

sinh (z + w) = i sin (i (z + w)) = i [sin (iz) cos (iw) + cos (iz) sin (iw)] =
= i [i sinh z cosh w + cosh z i sinh w] = sinh z cosh w + cosh z sinh w:

76
I.8.2
Show that jcos zj2 = cos2 x + sinh2 y, where z = x + iy. Find all zeros
and periods of cos z.

Solution.
Use trigonometric formulas from page 29 and 30 in CA we have,

cos z = cos (x + iy) = cos x cos (iy) sin x sin (iy) = cos x cosh y i sin x sinh y:

Now take the modulus squared, and use cosh2 y = 1 + sinh2 y,

jcos zj2 = cos2 x cosh2 y + sin2 x sinh2 y =

= cos2 x 1 + sinh2 y + sin2 x sinh2 y =
= cos2 x + cos2 x + sin2 x sinh2 y = cos2 x + sinh2 y:

The identity for jcos zj2 shows that the only zeros of cos z are the zeros of
cos z on the real axis, because

cos x = 0 , x = 2 + m ; m = 0; 1; 2; : : : ;
cos z = 0 ,
sinh y = 0 , y = 0:
Translation by any period of cos z sends zeros to zeros. Thus any period
is an integral multiple of , and since odd integral multiples are not periods,
the only periods of cos z are 2 n; 1 < n < 1.

77
I.8.3
Find all zeros and periods of cosh z and sinh z.

Solution
We have that cosh z = cos (iz), thus the zeros of cosh z are at z = i =2 +im ,
m = 0; 1; 2; : : :, and the periods of cosh z are 2 mi, m = 0; 1; 2; : : :.

We have that sinh z = i sin (iz), thus the zeros of sinh z are at z = m i,
m = 0; 1; 2; : : :, and the periods of cosh z are 2 mi, m = 0; 1; 2; : : :.

78
I.8.4
Show that

1 1 1 + iz
tan z= log ;
2i 1 iz
where both sides of the identity are to be interpreted as subsets of
the complex plane. In other words, show that tan w = z if and only
if 2iw is one of the values of the logarithm featured on the right.

Solution
Set z = tan w, we have that

sin w eiw e iw
e2iw 1
z = tan w = = = :
cos w i (eiw + e iw ) i (e2iw + 1)
Solve for e2iw
1 + iz
e2iw = ;
1 iz
and take logarithm and solve for w

1 1 + iz
w= log :
2i 1 iz
1
Because z = tan w then w 2 tan z and we have the identity

1 1 1 + iz
tan z= log :
2i 1 iz

79
I.8.5
Let S denote the two slits along the imaginary axis in the complex
plane, one running from i to +i1, the other from i to i1. Show
that (1 + iz) = (1 iz) lies on the negative real axis ( 1; 0] if and
only if z 2 S. Show that the principal branch

1 1 1 + iz
Tan z= Log
2i 1 iz
maps the slit plane CnS one-to-one onto the vertical strip fjRe wj < =2g.

Solution
I.8.5 z-plane I.8.5 I.8.5
π

i
-1
-i
−π

1+iz
w1 = 1 iz
w2 = Log w1

I.8.5 w-plane

−π/2 π/2

w = 2i1 w2 = 2i1 Log 11+iz

iz
We begin to show that 1+iz1 iz
2 ( 1; 0] if and only if z 2 ( i1; i] [ [i; i1).
Set
1 + iz
= w1
1 iz
and solve for z, thus

80
 1 w1
z= i
1 + w1
1 w1
As w1 goes from 1 to 1, 1+w 1
i goes from i to i1 along iR, and when
1 w1
w1 goes from 1 to 0, 1+w1 i goes from i1 to i along iR.
Since we have that
1 + iz 1 w1
= w1 ) z = i
1 iz 1 + w1
the map is one to one.
Remark that the map makes correspondence between the interval [ i: i1]
in z plane and [ 1; 1] in w1 plane, and between interval [i; i1] in z
plane and [ 1; 0] in w1 plane.
We have that w2 = Log w1 maps Cn( 1; 0) onto fjIm wj < g, thus w = w2i2
maps jIm w2 j < onto jRe wj < 2 , se …gures.
We have that the function

1 1 1 + iz
Tan z= Log
2i 1 iz
maps CnS onto jRe wj < 2 , where Tan 1 z is the principal branch for
tan 1 z. The other branches are given by fn (z) = Tan 1 z + n , 1 < n <
1.

81
I.8.6
1
Describe the Riemann surface for tan z.

Solution

I.8.6 z-plane I.8.6 I.8.5

+++++++++++
++++

π
____

i
++++++
_ _ _ _-1
__
-i
++++
____

_ _ _ _ −π
_______

1+iz
w1 = 1 iz
w2 = Log w1

I.8.6 w-plane
___________

+++++++++++

−π/2 π/2

1 1 1+iz
w= w
2i 2
= 2i
Log 1 iz

We have that

1 1 1 + iz
f0 (z) = Tan z= Log ;
2i 1 iz
other branches of tan 1 z are fn (z) = f0 (z) + n , where 1 < n < 1.
Use one copy of the double slit plane S for each integer n, and de…ne fn (z) =
Tan 1 z + n on the nth sheet to the (n + 1) th sheet along one of the cuts,
so that fn (z) and fn+1 (z) have same value at the junction.

82
Make in…nite many copies of S and call them Sn . De…ne fn (z) = Tan 1 z+n
on Sn . Identify "+" side of cut on Sn to " " side of cut on Sn+1 . Then
fn on Sn continuous to fn+1 on Sn+1 , and fn maps Sn onto vertical strip
n 21 < Re z < n + 12 . Note that composite function is not de…ned at
i and 1 (endpoints of slits), and its image omits the sequence 12 + n,
1 < n < 1.

83
I.8.7 p
Set w = cos z and = eiz . Show that = w w2 1. Show that
h p i
1
cos w = i log w w2 1 ;
where both sides of the identity are to be interpreted as subsets of
the complex plane.

Solution
Set w = cos z and = eiz , we have that

eiz + e iz
+ 1=
w = cos z = = :
2 2
Solve for
p
=w w2 1;
and set = eiz and take the logarithm and solve for z
p
z= i log w w2 1 :
1
Because w = cos z then z 2 cos w we have the identity

1
p
cos w= i log w w2 1 :

84
I.8.8
Show that the vertical strip jRe (w)j < =2 is mapped by the function
z (w) = sin w one-to-one onto the complex z plane with two slits
( 1; 1] and [+1; +1) on the real axis. Show p that the inverse
function is the branch of sin 1 z = i Log iz + 1 z 2 obtained by
taking the principal value of the square root. Hint. First show
that the function 1 z 2 on the slit plane omits the negative real
axis, so that the principal value of the square root is de…ned and
continuous on the slit plane, with argument in the open interval
between =2 and =2.

Solution
Set w = x + iy, by formulas on page 30 in CA we have that

sin w = sin (x + iy) = sin (x) cos (iy) + cos (x) sin (iy) =
= sin x cosh (y) + i cos (x) sinh y:

If jRe wj = jxj < =2 then sin w is mapped on Cn ( 1; 1] [ [+1; +1),

because

Im (sin w) = 0 ) y = 0 ) sin w = sin (x)

and

1 < sin (x) < 1

for these x.
1
Let now z = sin w. The value of w is referred to as sin z, that is, the
complex number whose sine is z. Note that

eiw e iw
z= : (1)
2i
Now with = eiw and 1= = e iw
in (1), we have

1=
z= :
2i
Mulitplying the above by 2i and doing some rearranging, we …nd that

85
 2 2
2iz = 1 or 2iz 1 = 0:
With the quadratic formula, we solve this equation for and …nd
p p
= iz + 1 z2 or eiw = iz + 1 z2
We now take the logarithm of both sides of the last equation and divide the
result by i to obtain
1 p
w= log iz + 1 z2
i
1
and, since w 2 sin z,

1
p
sin z= i log iz 1 z2 :

86