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) [. 1 i[ = 1 (f) 0 < Im. < :
(b) 1 < [2. 6[ < 2 (g) : < Re . < :
(c) [. 1[
2
+[. + 1[
2
< 8 (h) [Re .[ < [.[
(d) [. 1[ +[. + 1[ _ 2 (i) Re (i. + 2) 0
(e) [. 1[ < [.[ (j) [. i[
2
+[. + i[
2
< 2
Solution
Let . = r + i, where r. R.
(a) Circle, centre 1 + i, radius 1.
[. 1 i[ = 1 =[(r 1) + i ( 1)[ = 1 =(r 1)
2
+ ( 1)
2
= 1
2
(b) Annulus with centre 3, inner radius 1,2, outer radius 1.
1 < [2. 6[ < 2 =1 < 2 [. 3[ < 2 =
=1,2 < [. 3[ < 1 =(1,2)
2
< (r 3)
2
+
2
< 1
2
(c) Disk, centre 0, radius
_
3.
[r + i 1[
2
+[r + i + 1[
2
< 8 =
=(r 1)
2
+
2
+ (r + 1)
2
+
2
< 8 =r
2
+
2
<
_
_
3
_
2
(d) Interval [1. 1].
[. 1[ +[. + 1[ _ 2 =
_
(r 1)
2
+
2
_ 2
_
(r + 1)
2
+
2
=
=
_
_
(r 1)
2
+
2
_
2
_
_
2
_
(r + 1)
2
+
2
_
2
=
=
_
(r + 1)
2
+
2
_ r + 1 =
_
_
(r + 1)
2
+
2
_
2
_ (r + 1)
2
= = 0
Now, take = 0 in the inequality, and compute the three intervals
2
r < 1. then [r 1[ +[r + 1[ = (r 1) (r + 1) = 2r _ 2.
1 _ r _ 1 then [r 1[ +[r + 1[ = (r 1) + (r + 1) = 2 _ 2
r 1. then [r 1[ +[r + 1[ = (r 1) + (r + 1) = 2r _ 2.
(e) Halfplane r 1,2.
[. 1[ < [.[ =[. 1[
2
< [.[
2
=[r + i 1[
2
< [r + i[
2
=
=(r 1)
2
+
2
< r
2
+
2
=r 1,2
(f) Horizontal strip, 0 < < :.
(g) Vertical strip, : < r < :.
(h) CR.
[Re .[ < [.[ =[Re (r + i)[
2
< [r + i[
2
=r
2
< r
2
+
2
=[[ 0
(i) Half plane < 2.
Re (i. + 2) 0 =Re (i (r + i) + 2) 0 = + 2 0 = < 2
(j) Empty set.
[. i[
2
+[. + i[
2
< 2 =[r + i i[
2
+[r + i + i[
2
< 2 =
=r
2
+ ( 1)
2
+ r
2
+ ( + 1)
2
< 2 =r
2
+
2
< 0
3
I.1.2
Verify from the denitions each of the identities
(a) . + n = . + n (b) .n = . n (c) [ .[ = [.[ (d) [.[
2
= . .
Draw sketches to illustrate (a) and (c).
Solution
Substitute . = r + i and n = n + i, and use the denitions.
(a)
. + n = (r + i) + (n + i) = (r + n) + ( + ) i =
= (r + n) ( + ) i = (r i) + (n i) = . + n.
(b)
.n = (r + i) (n + i) = (rn ) + (r + n) i =
= (rn ) (r + n) i = (r i) (n i) = . n.
(c)
[ .[ =

r + i

= [r i[ =
_
r
2
+ ()
2
=
_
r
2
+
2
= [r + i[ = [.[ .
(d)
[.[
2
= [r + i[
2
=
_
_
r
2
+
2
_
2
= r
2
+
2
=
= r
2
i
2

2
= (r + i) (r i) = . ..
4
I.1.3
Show that the equation [.[
2
2 Re ( c.) +[c[
2
= j
2
represents a circle
centered at c with radius j.
Solution
Let . = r + i and c = c + i,, we have
[.[
2
2 Re ( c.) +[c[
2
=
= r
2
+
2
2 Re ((c i,) (r + i)) + c
2
+ ,
2
=
= r
2
+
2
2 Re ((cr + ,) + i (c ,r)) + c
2
+ ,
2
=
= r
2
+
2
2 (cr + ,) + c
2
+ ,
2
=
= (r c)
2
+ ( ,)
2
.
Thus the equation [.[
2
2 Re ( c.) +[c[
2
= j
2
, becomes
(r c)
2
+ ( ,)
2
= j
2
.
which is the equation for a circle of radius j centered at (c. ,), which in
complex notation is the point c = c + i,.
5
I.1.4
Show that [.[ _ [Re .[ +[Im.[, and sketch the set of points for which
equality holds.
Solution
Apply triangle inequality to . = Re . +i Im., to obtain [.[ 6 [Re .[ +[Im.[.
Now set . = r + i, and see then equality holds
[.[ = [Re .[+[Im.[ =
_
r
2
+
2
=
_
r
2
+
_

2
=
_
_
r
2
+
2
_
2
=
_
_
r
2
+
_

2
_
2
=
=r
2
+
2
=
_
_
r
2
_
2
+
_
_
r
2
_
2
+2
_
r
2
_

2
=r
2

2
= 0 =r = 0 or = 0.
Equality holds only when . is real or pure imaginary, which are all the points
on the real and imaginary axis.
6
I.1.5
Show that
[Re .[ _ [.[ . [Im.[ _ [.[ .
Show that
[. + n[
2
= [.[
2
+[n[
2
+ 2 Re (. n) .
Use this to prove the triangle inequality [. + n[ _ [.[ +[n[.
Solution
Let . = r + i. Then since the square root function is monotone, we have
[Re .[ = [r[ =
_
r
2
6
_
r
2
+
2
= [.[ .
[Im.[ = [[ =
_

2
6
_
r
2
+
2
= [.[ .
Now for .. n we have
[. + n[
2
=
= (. + n) (. + n) = (. + n) ( . + n) = . . + . n + n . + n n =
= [.[
2
+ 2 Re (. n) +[n[
2
.
where we have used the fact that 2 Re (. n) = . n+. n = . n+ n. = . n+n ..
We use both of the above facts and the trivial identities [.[ = [ .[ and [.n[ =
[.[ [n[ to prove the triangle inequality for .. n. We have
[. + n[
2
= [.[
2
+ 2 Re (. n) +[n[
2
_ [.[
2
+ 2 [Re (. n)[ +[n[
2
_
_ [.[
2
+ 2 [. n[ +[n[
2
= [.[
2
+ 2 [.[ [n[ +[n[
2
= ([.[ +[n[)
2
.
The desired inequality now follows by taking square root of both sides.
7
I.1.6
For xed c C, show that [. c[ , [1 c.[ = 1 if [.[ = 1 and 1 c. ,= 0.
Solution
If [.[ = 1, then [ .[ = 1 and . . = 1. Use this and get
[. c[ = [. c[ [ .[ = [. . c .[ = [1 c .[ = [1 c.[ .
We have
[. c[
[1 c.[
= 1.
as was to be shown.
8
I.1.7
Fix j 0, j ,= 1, and x .
0
. .
1
C. Show that the set of . satisfying
[. .
0
[ = j [. .
1
[ is a circle. Sketch it for j =
1
2
and j = 2, with
.
0
= 0 and .
1
= 1. What happens when j = 1?
Solution
Recall that a circle in R
2
centered at (c. /) with radius : is given by the
equation
(r c)
2
+ ( /)
2
= :
2
.
We manipulate the equation
[. .
0
[ = j [. .
1
[ .
The solutions set of the equation above remains the same if we square both
sides,
[. .
0
[
2
= j
2
[. .
1
[
2
.
Let . = r +i, .
0
= r
0
+i
0
and .
1
= r
1
+i
1
. Thus our equation becomes
(r r
0
)
2
+ (
0
)
2
= j
2
_
(r r
1
)
2
+ (
1
)
2
_
.
Expanding the squares, and grouping terms, we have
_
1 j
2
_
r
2
2
_
r
0
j
2
r
1
_
r+
_
r
2
0
j
2
r
2
1
_
+
_
1 j
2
_

2
2
_

0
j
2

1
_
+
_

2
0
j
2

2
1
_
= 0
Dividing both sides by (1 j
2
), we have
r
2
2
(r
0
j
2
r
1
)
1 j
2
r +
(r
2
0
j
2
r
2
1
)
1 j
2
+
2
2
(
0
j
2

1
)
1 j
2
+
(
2
0
j
2

2
1
)
1 j
2
= 0
Now complete the squares for both the r and terms. Recall that
r
2
2cr + / = (r c)
2
c
2
+ /.
So we have
9
_
r
r
0
j
2
r
1
1 j
2
_
2
+
(1 j
2
) (r
2
0
j
2
r
2
1
) (r
0
j
2
r
1
)
2
(1 j
2
)
2
+
+
_


0
j
2

1
1 j
2
_
2
+
(1 j
2
) (
2
0
j
2

2
1
) (
0
j
2

1
)
2
(1 j
2
)
2
= 0.
This becomes
_
r
r
0
j
2
r
1
1 j
2
_
2
+
_


0
j
2

1
1 j
2
_
2
=
=
(r
0
j
2
r
1
)
2
+ (
0
j
2

1
)
2
(1 j
2
) (r
2
0
+
2
0
j
2
(r
2
1
+
2
1
))
(1 j
2
)
2
=
=
j
2
_
(r
1
r
0
)
2
+ (
1

0
)
2
_
(1 j
2
)
2
.
which is the equation for a circle. If .
0
= 0, and .
1
= 1, we have
_
r
j
2
1 j
2
_
2
+
2
=
_
j
1 j
2
_
2
.
When j =
1
2
we have a circle of radius
2
3
centered at
_

1
3
. 0
_
, and when j = 2,
we have a circle of radius
2
3
centered at
_
4
3
. 0
_
. When j = 1, we have the
equation
[. .
0
[ = [. .
1
[ .
which is the line bisecting the two points. When .
0
= 0. .
1
= 1, this is the
line Re . =
1
2
.
10
-1 1 2
-2
-1
1
2
Re z
Im z
I.1.7
-1 1 2
-2
-1
1
2
Re z
Im z
I.1.7
-1 1 2
-2
-1
1
2
Re z
Im z
I.1.7
j =
1
2
j = 1 j = 2
11
I.1.8
Let j (.) be a polynomial of degree : _ 1 and let .
0
C. Show
that there is a polynomial /(.) of degree : 1 such that j (.) =
(. .
0
) /(.)+j (.
0
). In particular, if j (.
0
) = 0, then j (.) = (. .
0
) /(.).
Solution
Set
j (.) = c
n
.
n
+ c
n1
.
n1
+ c
n2
.
n2
+ + c
2
.
2
+ c
1
. + c
0
.
and
/(.) = /
n1
.
n1
+ /
n2
.
n2
+ /
n3
.
n3
+ . . . + /
2
.
2
+ /
1
. + /
0
.
We equate coecients in the polynomial identity j (.) = (. .
0
) /(.) +
j (.
0
), and get
c
n
.
n
+ c
n1
.
n1
+ c
n2
.
n2
+ + c
2
.
2
+ c
1
. + c
0
=
= (. .
0
)
_
/
n1
.
n1
+ /
n2
.
n2
+ /
n3
.
n3
+ . . . + /
2
.
2
+ /
1
. + /
0
_
+j (.
0
) =
= /
n1
.
n
+ /
n2
.
n1
+ /
n3
.
n2
+ . . . + /
2
.
3
+ /
1
.
2
+ /
0
. + j (.
0
)
/
n1
.
0
.
n1
/
n2
.
0
.
n2
/
n3
.
0
.
n3
. . . /
2
.
0
.
2
/
1
.
0
. /
0
.
0
=
/
n1
.
n
+(/
n2
/
n1
.
0
) .
n1
+(/
n3
/
n2
.
0
) .
n2
+(/
n4
/
n3
.
0
) .
n3
+
+ (/
2
/
3
.
0
) .
3
+ (/
1
/
2
.
0
) .
2
+ (/
0
/
1
.
0
) . + j (.
0
) /
0
.
0
.
Equate and solve for the /
j
s in terms of c
j
s.
_
_
_
c
n
= /
n1
c
k
= /
k1
/
k
.
0
. 0 _ / _ : 1
c
0
= j (.
0
) /
0
.
0
=
_

_
/
n1
= c
n
/
k
=
nk1

i=0
c
k+1+i
.
i
0
. 0 _ / _ : 2
j (.
0
) =
n

i=0
c
i
.
i
0
Proof by induction on degree : of j (.), set
j (.) = c
n
.
n
+ c
n1
.
n1
. . . + c
0
.
where c
n
,= 0.
12
Fix .
0
and write
j (.) = c
n
(. .
0
) .
n1
+ : (.) .
where deg : (.) _ : 1.
By using the induction hypothesis, we can assume that
: (.) = (.) (. .
0
) + c.
where deg (.) _ deg : (.). Then
j (.) =
_
c
n
.
n1
+ (.)
_
(. .
0
) + c = /(.) (. .
0
) + c
Since deg (.) _ : 2, deg : (.) _ : 1.
Plug in .
0
, get j (.
0
) = c.
13
I.1.9
Find the polynomial /(.) in the preceding exercise for the following
choices of j (.) and .
0
(a) j (.) = .
2
and .
0
= i
(b) j (.) = .
3
+ .
2
+ . and .
0
= 1
(c) j (.) = 1 + . + .
2
+ + .
m
and .
0
= 1
Solution
From the preceding exercise we have
j (.) = (. .
0
) /(.) + j (.
0
) .
We solve for /(.) then
/(.) =
j (.) j (.
0
)
. .
0
.
(a)
We have that j (.) = .
2
and .
0
= i, thus j (.
0
) = j (i) = 1.
Thus
/(.) =
j (.) j (.
0
)
. .
0
=
.
2
+ 1
. i
= . + i.
and
.
2
= (. i) (. + i) 1.
(b)
We have that j (.) = .
3
+ .
2
+ . and .
0
= 1, thus j (.
0
) = j (1) = 1.
Thus
/(.) =
j (.) j (.
0
)
. .
0
=
.
3
+ .
2
+ . + 1
. + 1
=
(. + 1) (.
2
+ 1)
. + 1
= .
2
+ 1.
and
.
3
+ .
2
+ . = (. + 1)
_
.
2
+ 1
_
1.
(c)
We have that j (.) = 1 + . + .
2
+ + .
m
and .
0
= 1, thus
14
j (.
0
) = j (1) =
_
0. : odd
1. : even
Thus
/(.) =
j (.) j (.
0
)
. .
0
=
=
_
_
_
(z
m1
+z
m3
+z
2
+1)(z+1)
z+1
= .
m1
+ .
m3
+ + .
2
+ 1 if : is odd
(z
m1
+z
m3
+z
3
+z)(z+1)+1
z+1
= .
m1
+ .
m3
+ . . . + .
3
+ . if : is even
and
.
m
+ .
m1
+ .
2
+ . + 1 =
=
_
(. + 1) (.
m1
+ .
m3
+ + .
2
+ 1) if : is odd
(. + 1) (.
m1
+ .
m3
+ + .
3
+ .) + 1 if : is even
15
I.1.10
Let (.) be a polynomial of degree : _ 1. Showthat any polynomial
j (.) can be expressed in the form
j (.) = /(.) (.) + : (.) .
where /(.) and : (.) are polynomials and the degree of the remain-
der : (.) is strictly less than :.
Hint. Proceed by induction on the degree of j (.). The resulting
method is called the division algorithm.
Solution
First suppose j (.) is the zero polynomial. (So, the degree of j (.) is .)
The degree of : (.) must be less than the degree of (.). If /(.) ,= 0, it
follows that the degree of /(.) (.) is greater than the degree of : (.). This
then implies that /(.) (.)+: (.) ,= 0. So, /(.) (.)+: (.) = 0 implies that
/(.) = 0, and thus : (.) = 0. So, the polynomials are /(.) = 0 and : (.) = 0,
and these polynomials are the only ones that satisfy both conditions.
Now assume that the division algorithm is true for all polynomials j (.) of
degree less than :. (Where : _ 0.) If the degree of (.) is greater than the
degree of j (.), and /(.) is nonzero, then /(.) (.)+: (.) has degree greater
than j (.). So, if the degree of (.) is greater than the degree of j (.), then
/(.) = 0 and thus : (.) = j (.). This proves both existence and uniqueness
of /(.) and : (.), in this case.
Now, suppose that the degree of (.) is less than or equal to the degree of
j (.). Set
j (.) = c
n
.
n
+ c
n1
.
n1
+ + c
1
. + c
0
and
(.) = /
m
.
m
+ /
m1
.
m1
+ + /
1
. + /
0
.
where c
n
,= 0, /
m
,= 0 and : _ :.
Let
j
1
(.) =
c
n
/
m
.
nm
(.) j (.) .
16
then
j
1
(.) =
c
n
/
m
.
nm
_
/
m
.
m
+ /
m1
.
m1
+ + /
1
. + /
0
_

_
c
n
.
n
+ c
n1
.
n1
+ + c
1
. + c
0
_
.
The monomials of degree : cancel, and therefore j
1
(.) is a polynomial of de-
gree at most :1. It follows, by assumption, that j
1
(.) = /
1
(.) (.)+:
1
(.),
where /
1
(.) and :
1
(.) are the unique polynomials satisfying the conditions
above.
Let
/(.) =
c
n
/
m
.
nm
/
1
(.) .
and let
: (.) = :
1
(.) .
Then
/(.) (.) + : (.) =
=
_
c
n
/
m
.
nm
/
1
(.)
_
(.) :
1
(.) =
=
c
n
/
m
.
nm
(.) (/
1
(.) (.) + :
1
(.)) =
=
c
n
/
m
.
nm
(.) j
1
(.) =
= j (.) .
Thus given j (.) and (.), there exist polynomials /(.) and : (.) satisfying
the above two conditions.
17
I.1.11
Find the polynomials /(.) and : (.) in the preceding exercise for
j (.) = .
n
and (.) = .
2
1.
Solution
Require
.
n
= /(.)
_
.
2
1
_
+ : (.) . deg : (.) _ 1.
If : is even
.
n
=
_
.
n2
+ .
n4
+ .
n6
+ + .
2
+ 1
_ _
.
2
1
_
+ 1.
If : is odd
.
n
=
_
.
n2
+ .
n4
+ .
n6
+ + .
3
+ .
_ _
.
2
1
_
+ ..
Thus
/(.) =
_
.
n2
+ .
n4
+ .
n6
+ + .
2
+ 1. if : is even,
.
n2
+ .
n4
+ .
n6
+ + .
3
+ .. if : is odd.
.
and
: (.) =
_
1. if : is even,
.. if : is odd.
.
18
I.2.1
Express all values of the following expressions in both polar and
Cartesian coordinates, and plot them.
(a)
_
i (c)
4
_
1 (e) (8)
1=3
(g) (1 + i)
8
(b)
_
i 1 (d)
4
_
i (f) (3 4i)
1=8
(h)
_
1+i
_
2
_
25
Solution
(a)
_
i =
_
_
c
i(=2+2k)
_
1=2
= c
i(=4+k)
. / = 0. 1
_
=
_
c
i=4
. c
i5=4
_
=
_
(1 + i) ,
_
2
_
.
(b)
_
i 1 =
_
_
_
2c
i(3=4+2k)
_
1=2
= 2
1=4
c
i(3=8+k)
. / = 0. 1
_
=
=
_
2
1=4
c
i3=8
. 2
1=4
c
i11=8
_
=
_
2
1=4
(cos (3:,8) + i sin (3:,8))
_
.
(c)
4
_
1 =
_
_
c
i(+2k)
_
1=4
= c
i(=4+k=2)
. / = 0. 1. 2. 3
_
=
=
_
c
i=4
. c
i3i=4
. c
i5=4
. c
i7=4
_
=
_
(1 i),
_
2. (1 i) ,
_
2
_
.
(d)
4
_
i =
_
_
c
i(=2+2k)
_
1=4
c
i(=8+k=2)
. / = 0. 1. 2. 3 =
_
c
i=8
. c
i5=8
. c
i9=8
. c
i13=8
_
_
=
= (cos (:,8) + i sin (:,8)) . (cos (5:,8) + i sin (5:,8)) .
(e)
(8)
1=3
=
_
_
2
3
c
i(+2k)
_
1=3
= 2c
i(=3+2k=3)
. / = 0. 1. 2
_
=
=
_
2c
i=3
. 2c
i
. 2c
i5=3
_
=
_
1 + i
_
3. 2. 1 i
_
3
_
.
(f)
19
Draw gure and get tan o
0
= 4,3 = o
0
= tan
1
_

4
3
_
.
(3 4i)
1=8
=
_
_
5c
i(
0
+2k)
_
1=8
= 5
1=8
c
i(
0
=8+k=4)
. / = 0. 1. . . . . 7
_
=
=
_
5
1=8
c
i(
0
=8)
. 5
1=8
c
i(
0
=8+=4)
. 5
1=8
c
i(
0
=8+=2)
. 5
1=8
c
i(
0
=8+3=4)
_
=
_
5
1=8
(cos (o
0
,8) + i sin (o
0
,8)) . 5
1=8
(cos (o
0
,8 + :,4) + i sin (o
0
,8 + :,4)) .
5
1=8
(cos (o
0
,8 + :,2) + i sin (o
0
,8 + :,2)) . 5
1=8
(cos (o
0
,8 + 3:,4) + i sin (o
0
,8 + 3:,4))
_
.
(g)
(1 + i)
8
=
_
_
2c
i(=4+2k)
_
8
= 16c
i(2+16k)
= 16c
i2
= 16.
(h)
_
1 + i
_
2
_
25
=
_
c
i(=4+2k)
_
25
= c
i(25=4+50k)
= c
i(=4+6+50k)
= c
i=4
=
1 + i
_
2
.
-1 1
-1
1
I.2.1a
-1 1
-1
1
I.2.1b
-1 1
-1
1
I.2.1c
-1 1
-1
1
I.2.1d
-2 2
-2
2
I.2.1e
-1 1
-1
1
I.2.1f
20
-20 20
-20
20
I.2.1g
-1 1
-1
1
I.2.1h
21
I.2.2
Sketch the following sets
(a) [arg .[ < :,4 (c) [.[ = arg .
(b) 0 < arg (. 1 i) < :,3 (d) log [.[ = 2 arg .
Solution
a) Sector. b) Sector. c) Is a spiral curve starting at 0, spiraling to . (d) Is
a spiral curve, spiraling to 0, and to .
/4
/4
-5 5
-5
5
I.1.2a
/3
-5 5
-5
5
I.1.2b
(/2,0)
(,0)
(3/2,0)
(2,0)
I.2.2c
(1,0)
I.2.2d (-3 < x < 3) I.2.2d (-30 < x < 30) I.2.2d (-800 < x < 800)
22
I.2.3
For a xed complex number /, sketch the curve
_
c
i
+ /c
i
: 0 _ o _ 2:
_
.
Dierentiate between the cases [/[ < 1. [/[ = 1 and [/[ < 1.
Hint. First consider the case / 0, and then reduce the general
case to this case by a rotation.
Solution
For 0 < / < 1, an ellipse r
2
, (1 + /)
2
+
2
, (1 /)
2
= 1, traversed in positive
direction with increasing o. For / = 1, an interval [2. 2]. For 1 < / < +,
an ellipse traversed in negative direction. For / = jc
i'
, express equation as
c
i'=2
_
c
i('=2)
+ jc
i('=2)
_
to see that curve is rotate of ellipse or interval
by ,,2.
-3 -2 -1 1 2 3
-3
-2
-1
1
2
3
x
y
I.2.7 (b = 0.5)
-3 -2 -1 1 2 3
-3
-2
-1
1
2
3
x
y
I.2.7 (b = 1)
-3 -2 -1 1 2 3
-3
-2
-1
1
2
3
x
y
I.2.7 (b = 1.5)
23
I.2.4
For which : is i an :-th root of unity?
Solution.
i is an :
th
root of unity for i
n
= 1, : = 4/, / = 1. 2. 3. . . .
24
I.2.5
For : _ 1, show that
(a) 1 + . + .
2
+ + .
n
= (1 .
n+1
) , (1 .), . ,= 1
(b) 1 + cos o + cos 2o + + cos :o =
1
2
+
sin

n+
1
2

2 sin =2
.
Solution
(a)
Set
:
n
= 1 + . + .
2
+ + .
n
.
and multiply :
n
with . and have
.:
n
= . + .
2
+ .
3
+ + .
n+1
.
Now subtract .:
n
from the sum :
n
, and have
:
n
(1 .) = 1 .
n+1
.
If . ,= 1 we have after division by 1 ..
:
n
=
1 .
n+1
1 .
.
(b)
Apply (a) to . = c
i
and to . = c
i
.
1 + c
i
+ c
i2
+ + c
in
=
1 c
i(n+1)
1 c
i
.
1 + c
i
+ c
i2
+ + c
in
=
1 c
i(n+1)
1 c
i
.
Add the identities, and use the denitions of sine and cosine.
25
2 (1 + cos o + cos 2o + + cos :o) =
=
1 c
i(n+1)
1 c
i
+
1 c
i(n+1)
1 c
i
=
=
_
1 c
i(n+1)
_
c
i=2
c
i=2
c
i=2
+
_
1 c
i(n+1)
_
c
i=2
c
i=2
c
i=2
=
=
1
2i
c
i=2
_
1 c
i(n+1)
_
sin (o,2)
+
1
2i
c
i=2
_
1 c
i(n+1)
_
sin (o,2)
=
=
1
2i
_
c
i=2
c
i=2
+ c
i(n+
1
2
)
c
i(n+
1
2
)
sin (o,2)
_
=
sin (o,2) + sin
_
: +
1
2
_
o
sin (o,2)
=
= 1 +
sin
_
: +
1
2
_
o
sin (o,2)
.
Divide both sides with 2 and get
1 + cos o + cos 2o + + cos :o =
1
2
+
sin
_
: +
1
2
_
o
2 sin o,2
.
26
I.2.6
Fix : _ 1. Show that the : th roots of unity n
0
. . . . . n
n1
satisfy
(a) (. n
0
) (. n
1
) . . . (. n
n1
) = .
n
1.
(b) n
0
+ + n
n1
= 0 if : _ 2.
(c) n
0
n
n1
= (1)
n1
.
(d)
n1

j=0
n
k
j
=
_
0. 1 _ / _ : 1.
:. / = :.
Solution
(a)
Let n
0
. . . . . n
n1
be the : th roots of unity then n
j
= c
2ji=n
, and n
j
are
a roots of .
n
1 since
n
n
j
1 =
_
c
2ji=n
_
n
1 = c
2ji
1 = 0.
By the fundamental theorem of algebra, and since the root are simple and
the coecient for the .
n
term is 1 it follows that
.
n
1 = (. n
0
) (. n
1
) (. n
n1
) .
(b)
Using exercise (a) and multiply the factors in the product we get
.
n
1 = .
n
(n
0
+ n
1
+ n
n1
) .
n1
+c
2
.
n2
+ +(1)
n
n
0
n
1
n
n1
.
By identication of the coecients we see that
n
0
+ n
1
+ + n
n1
= 0 if : _ 2.
(c)
From (b) follows that
1 = (1)
n
n
0
n
1
n
n1
=n
0
n
1
n
n1
= (1)
n1
.
(d)
If 1 _ / _ : 1
n1

j=0
n
k
j
=
n1

j=0
_
c
2ji=n
_
k
=
n1

j=0
_
c
2ki=n
_
j
=
1
_
c
2ki=n
_
n
1 c
2ki=n
=
1 c
2ki
1 c
2ki=n
= 0.
27
If / = :
n1

j=0
n
k
j
=
n1

j=0
_
c
2ji=n
_
n
=
n1

j=0
c
2ji
=
n1

j=0
1 = :.
28
I.2.7
Fix 1 1 and : _ 1, : _ 0. Show that

.
m
.
n
+ 1

_
1
m
1
n
1
. [.[ = 1.
Sketch the set where equality holds. Hint. See (1.1) p.2.
Solution
a
b
a + b
I.2.7
-1 1
-1
1
I.2.7 (n = 1)
-1 1
-1
1
I.2.7 (n = 2)
-1 1
-1
1
I.2.7 (n = 3)
-1 1
-1
1
I.2.7 (n = 4)
-1 1
-1
1
I.2.7 (n = 5)
We use that [. n[ _ [.[ [n[ see (1.1) on page 2 in CA, and have that
[.
n
+ 1[ > [.
n
[ 1 = 1
n
1.
where the last equaltity is due to that [.[ = 1.
Some rearrangement gives
1
[.
n
+ 1[
_
1
(1
n
1)
.
29
and mulitplication by [.
m
[ = 1
m
gives

.
m
.
n
+ 1

_
1
m
1
n
1
. [.[ = 1.
For equality, we must have
[.
n
+ 1[ = 1
n
1.
because [.[
m
= 1
m
.
We rearrarange this equality to
[1[ +[.
n
+ 1[ = [.
n
[
and we conclude that 1 and .
n
+1 must lie on the same ray. (We have used
the fact that [c[+[/[ = [c + /[ implies that c. / lie on the same ray, see rst gure I.2.7)
If
. = 1c
i
then
.
n
= 1
n
c
in
.
Since 1 1, we require that c
in
= 1 thus
. = n
k
1c
i=n
.
where n
k
is an : th root of unity.
If
. = n
k
1c
i=n
.
then
.
n
+ 1 = 1
n
+ 1.
and

.
m
.
n
+ 1

=
1
m
1
n
1
.
30
I.2.8
Show that cos 2o = cos
2
o sin
2
o and sin 2o = 2 cos o sin o using de
Moivres formulae. Find formulae for cos 4o and sin 4o in terms of
cos o and sin o.
Solution
Let o R be given. Then by de Moivres formulae (on page 8 in CA) we
have for all : Z
cos :o + i sin :o = (cos o + i sin o)
n
Hence for : = 2 we get
cos 2o + i sin 2o =
= (cos o + i sin o)
2
= cos
2
o + i2 cos o sin o sin
2
o =
=
_
cos
2
o sin
2
o
_
+ i (2 cos o sin o) .
then, by setting the real and imaginary parts equal to each other we obtain
cos 2o = cos
2
o sin
2
o
sin 2o = 2 cos o sin o.
Similarly, applying de Moivres formulae for : = 4 we get (using the Binomial
Theorem)
cos 4o + i sin 4o = (cos o + i sin o)
4
=
= cos
4
o + i4 cos
3
o sin o 6 cos
2
o sin
2
o i4 cos o sin
3
o + sin
4
o =
= cos
4
o 6 cos
2
o sin
2
o + sin
4
o + i
_
4 cos
3
o sin o 4 cos o sin
3
o
_
.
then, by setting the real and imaginary parts equal to each other we obtain
cos 4o = cos
4
o 6 cos
2
o sin
2
o + sin
4
o
sin 4o = 4 cos
3
o sin o 4 cos o sin
3
o.
as required.
31
I.3.1
Sketch the image under the spherical projection of the following
sets on the sphere
(a) the lower hemisphere 2 _ 0
(b) the polar cap
3
4
_ 2 _ 1
(c) lines of lattitude A =
_
1 2
2
cos o, 1 =
_
1 2
2
sin o, for 2 xed
and 0 _ o _ 2:
(d) lines of longitude A =
_
1 2
2
cos o, 1 =
_
1 2
2
sin o, for o xed
and 1 _ 2 _ 1
(e) the spherical cap _ A _ 1, with center lying on the equator,
for xed . Separate into cases, according to various ranges of .
Solution
-2 -1 1 2
-2
-1
1
2
y
I.3.1a
-4 -2 2 4
-4
-2
2
4
y
I.3.1b
-2 -1 1 2
-2
-1
1
2
y
I.3.1c
-2 -1 1 2
-2
-1
1
2
y

I.3.1d
-6 -4 -2 2
-4
-2
2
4
y
I.3.1e (A = -1/2)
-2 -1 1 2
-2
-1
1
2
y
I.3.1e (A = 0)
32
-2 2 4 6
-4
-2
2
4
y
I.3.1e (A = 1/2)
(a)
Image is the unit disk.
(b)
Set 1 = 0 and 2 = 3,4 in A
2
+ 1
2
+ 2
2
= 1, we have A =
_
7,4. We have
the formula r = A, (1 2), thus r =
_
7. Image is the exterior of a disk,
with centre 0 and radius
_
7.
(c) Image is the disk, with centre 0 and radius
_
1 + .,
_
1 .. We have that
A =
_
1 2
2
cos o and 1 =
_
1 2
2
sin o, we can take the radius to image
for o = 0, thus : =
_
1Z
2
1Z
=
_
1+Z
_
1Z
.
(d) Image is lines of longitude o issuing from 0.
(e)
Case 1: 1 < < 0
Image is the exterior of the disk centered at 1, with radius
_
1
2
, [[.
Case 2: = 0
Image is the left halfplane.
Case 3: 0 < < 1
Image is a disk centered at 1, with radius
_
1
2
, [[.
Set 1 = 0 then 2 goes from
_
1
2
to
_
1
2
. We have the formula
r = A, (1 2) thus r goes from ,
_
1 +
_
1
2
_
=
_
1
_
1
2
_
, to
,
_
1
_
1
2
_
=
_
1 +
_
1
2
_
,.
33
I.3.2
If the point 1 on the sphere corresponds to . under the stereo-
graphic projection, show that the antipodal point 1 on the sphere
corresponds to 1, ..
Solution
For . = r + i the corresponding point on the sphere under transformation
given on p. 12 in CA is given by (A. 1. 2) where
A =
2r
[.[
2
+ 1
.
1 =
2
[.[
2
+ 1
.
2 =
[.[
2
1
[.[
2
+ 1
.
For . ,= 0, we have

1
.
=
1
r i
=
r + i
(r i) (r + i)
=
r + i
r
2
+
2
=
r
[.[
2
i

[.[
2
which corresponds to the point on the sphere given by (A
t
. 1
t
. 2
t
) where
A
t
=
2x
[z[
2

1
z

2
+ 1
=
2r
[.[
2
+ 1
= A.
1
t
=
2y
[z[
2

1
z

2
+ 1
=
2
[.[
2
+ 1
= 1.
2
t
=

1
z

2
1

1
z

2
+ 1
=
[.[
2
1
[.[
2
+ 1
= 2.
Hence, the map (A. 1. 2) (A. 1. 2) on the sphere corresponds to
. 1, . of C.
34
I.3.3
Show that as . travers a small circle in the complex plane in the
positive (counterclockwise) direction, the corresponding point 1
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 ` 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 A-axis corre-

sponds under stereographic projection to the inversion . 1,. of
C.
Solution
For . = r + i the corresponding point on the sphere under transformation
given on p. 12 in CA is given by (A. 1. 2) where
A =
2r
[.[
2
+ 1
.
1 =
2
[.[
2
+ 1
.
2 =
[.[
2
1
[.[
2
+ 1
.
For . ,= 0, we have
1
.
=
1
r + i
=
r i
(r + i) (r i)
=
r i
r
2
+
2
=
r
[.[
2
i

[.[
2
which corresponds to the point on the sphere given by (A
t
. 1
t
. 2
t
) where
A
t
=
2x
[z[
2

1
z

2
+ 1
=
2r
[.[
2
+ 1
= A.
1
t
=
2y
[z[
2

1
z

2
+ 1
=
2
[.[
2
+ 1
= 1.
2
t
=
1
[z[
2
1
1
[z[
2
+ 1
=
[.[
2
1
[.[
2
+ 1
= 2.
Hence, the map (A. 1. 2) (A. 1. 2) on the sphere corresponds to
inversion . 1,. of C. Moreover, the map (A. 1. 2) (A. 1. 2) is
given by rotation of the sphere by 180

about the A-axis.

36
I.3.5
Suppose (r. . 0) is the spherical projection of (A. 1. 2). Show that
the product of the distances from the north pole ` to (A. 1. 2) and
from ` to (r. . 0) is 2. What is the situation when (A. 1. 2) lies on
the equator on the sphere?
Solution
The distance from from the north pole ` = (0. 0. 1) to (A. 1. 2) is
_
A
2
+ 1
2
+ (2 1)
2
.
and the distance from the north pole ` = (0. 0. 1) to (r. . 0) =
_
X
1Z
.
Y
1Z
. 0
_
is
_
A
2
(1 2)
2
+
1
2
(1 2)
2
+ 1.
The product of distances is
_
A
2
+ 1
2
+ (2 1)
2
_
1=2
_
A
2
(1 2)
2
+
1
2
(1 2)
2
+ 1
_
1=2
=
A
2
+ 1
2
+ (2 1)
2
1 2
= 2.
When (A. 1. 2) lie on the equator, the product is simply the square of the
distance from ` to a point on the equator. By the Pythagorean Law, this is
1 + 1 = 2.
37
I.3.6
We dene the chordal distance d (.. n) between two points .. n C
+
to be the length of the straight line segment joining to the points
1 and Q on the unit sphere whose stereographic projections are .
and n, respectively.
(a) Show that the chordal distance is a metric, that is, it is sym-
metric, d (.. n) = d (n. .); satises the triangle inequality d (.. n) _
d (.. ) + d (. n); and d (.. n) = 0 if and only if . = n.
(b) Show that the chordal distance from . to n is given by
d (.. n) =
2 [. n[
_
1 +[.[
2
_
1 +[n[
2
. .. n C.
(c) What is d (.. )? Remark. The expression for d (.. n) shows
that innitesimal arc length corresponding to the chordal metric
is given by
do (.) =
2d:
1 +[.[
2
.
where d: = [d.[ is the usual Euclidean innitesimal arc length. The
innitesimal arc length do (.) determines another metric, the spher-
ical metric o (.. n), on the extended complex plane. See Section
IX.3.
Solution
(a)
Follows from fact that Euclidean distance in R
3
is a metric on the sphere.
(b)
Set . = r
1
+ i
1
. n = r
2
+ i
2
, we have
[. n[
2
=
= (. n) (. n) = (. n) (. n) =
= ...n .n + nn = [.[
2
+[n[
2
.n .n =
= [.[
2
+[n[
2
(r
1
+ i
1
) (r
2
i
2
) (r
1
i
1
) (r
2
+ i
2
) =
= [.[
2
+[n[
2
2r
1
r
2
2
1

2
. (1)
38
We take square of distance between 1 and Q
d (.. n)
2
= (A A
t
)
2
+ (1 1
t
)
2
+ (2 2
t
)
2
=
= A
2
+ 1
2
+ 2
2
+ A
t2
+ 1
t2
+ 2
t2
2 (AA
t
+ 1 1
t
+ 22
t
) =
= 2 2
_
4r
1
r
2
+ 4
1

2
+
_
[.[
2
1
_ _
[n[
2
1
_
_
[.[
2
+ 1
_ _
[n[
2
+ 1
_
_
=
= 2
_
_
[.[
2
+ 1
_ _
[n[
2
+ 1
_
4r
1
r
2
4
1

_
[.[
2
1
_ _
[n[
2
1
_
_
[.[
2
+ 1
_ _
[n[
2
+ 1
_
_
=
= 2
_
2[.[
2
+ 2 [n[
2
4r
1
r
2
4
1

2
_
[.[
2
+ 1
_ _
[n[
2
+ 1
_
_
(1)
=
4 [. n[
2
_
[.[
2
+ 1
_ _
[n[
2
+ 1
_.
Taking the positive square root we have
d (.. n) =
2 [. n[
_
1 +[.[
2
_
1 +[n[
2
. .. n C.
(c)
d (.. ) = lim
wo
2 [. n[
_
1 +[.[
2
_
1 +[n[
2
= lim
wo
2 [.,n 1[
_
1 +[.[
2
_
1, [n[
2
+ 1
=
2
_
1 +[.[
2
.
39
I.3.7
Consider the sphere of radius
1
2
in (A. 1. 2) space, resting on the
(A. 1. 0) plane, with south pole at the origin (0. 0. 0) and north pole
at (0. 0. 1). We dene a stereographic projection of the sphere onto
the complex plane as before, so that corresponding points (A. 1. 2)
and . ~ (r. . 0) lie on the same line through the north pole. Find
the equations for . = r + i in terms of A. 1. 2, and the equations
for A. 1. 2 in terms of .. What is the corresponding formula for the
chordal distance? Note. In this case, the equation of the sphere is
A
2
+ 1
2
+
_
2
1
2
_
2
=
1
4
.
Solution
The line through 1 = (A. 1. 2) and ` = (0. 0. 1) is on the form ` +
t (1 `), and it meats the r plane when
(r. . 0) = (0. 0. 1) + t ((A. 1. 2) (0. 0. 1)) = (tA. t1. 1 + t (2 1)) .
By simultaneous equations we have, and solve for A, 1 and 2
_
_
_
r = tA
= t1
1 + t (2 1) = 0
=
_
_
_
A =
x
t
1 =
y
t
2 = 1
1
t
We solve for the t that is a point on the sphere A
2
+1
2
+
_
2
1
2
_
2
=
1
4
we
have
r
2
t
2
+

2
t
2
+
_
1
2

1
t
_
2
=
1
4
=[.[
2
+
_
t
2
1
_
2
=
t
2
4
=
=[.[
2
t + 1 = 0 =t = [.[
2
+ 1
thus we have
_

_
A =
x
1+[z[
2
1 =
y
1+[z[
2
2 =
[z[
2
1+[z[
2
.
40
Set . = r
1
+ i
1
. n = r
2
+ i
2
, we have
[. n[
2
=
= (. n) (. n) = (. n) (. n) =
= ...n .n + nn = [.[
2
+[n[
2
.n .n =
= [.[
2
+[n[
2
(r
1
+ i
1
) (r
2
i
2
) (r
1
i
1
) (r
2
+ i
2
) =
= [.[
2
+[n[
2
2 (r
1
r
2
+
1

2
) . (1)
And the coordical distance
41
d (.. n)
2
= (A
1
A
2
)
2
+ (1
1
1
2
)
2
+ (2
1
2
2
)
2
=
=
_
r
1
1 +[.[
2

r
2
1 +[n[
2
_
2
+
_

1
1 +[.[
2

2
1 +[n[
2
_
2
+
_
[.
1
[
2
1 +[.[
2

[.
2
[
2
1 +[n[
2
_
2
=
=
_
r
1
_
1 +[n[
2
_
r
2
_
1 +[.[
2
__
2
+
_

1
_
1 +[n[
2
_

2
_
1 +[.[
2
__
2
+
_
[.[
2
[n[
2
_
2
_
1 +[.[
2
_
2
_
1 +[n[
2
_
2
=
=
(r
2
1
+
2
1
)
_
1 +[n[
2
_
2
+ (r
2
2
+
2
2
)
_
1 +[.[
2
_
2
+[.[
4
2 [.[
2
[n[
2
+[n[
4
_
1 +[.[
2
_
2
_
1 +[n[
2
_
2
+
+
2r
1
r
2
_
1 +[n[
2
_ _
1 +[.[
2
_
2
1

2
_
1 +[n[
2
_ _
1 +[.[
2
_
_
1 +[.[
2
_
2
_
1 +[n[
2
_
2
=
=
[.[
2
_
1 +[n[
2
_
2
+[n[
2
_
1 +[.[
2
_
2
+[.[
4
2 [.[
2
[n[
2
+[n[
4
_
1 +[.[
2
_
2
_
1 +[n[
2
_
2
+
+
2 (r
1
r
2
+
1

2
)
_
1 +[n[
2
_ _
1 +[.[
2
_
_
1 +[.[
2
_
2
_
1 +[n[
2
_
2
=
=
[.[
2
+ 2 [.[
2
[n[
2
+[.[
2
[n[
4
+[n[
2
+ 2 [n[
2
[.[
2
+[n[
2
[.[
4
+[.[
4
2 [.[
2
[n[
2
+[n[
4
_
1 +[.[
2
_
2
_
1 +[n[
2
_
2
+
+
2 (r
1
r
2
+
1

2
)
_
1 +[n[
2
_ _
1 +[.[
2
_
_
1 +[.[
2
_
2
_
1 +[n[
2
_
2
=
=
_
[.[
2
+[n[
2
_ _
1 +[n[
2
_ _
1 +[.[
2
_
2 (r
1
r
2
+
1

2
)
_
1 +[n[
2
_ _
1 +[.[
2
_
_
1 +[.[
2
_
2
_
1 +[n[
2
_
2
=
=
_
[.[
2
+[n[
2
2 (r
1
r
2
+
1

2
)
_ _
1 +[n[
2
_ _
1 +[.[
2
_
_
1 +[.[
2
_
2
_
1 +[n[
2
_
2
=
=
_
[.[
2
+[n[
2
2 (r
1
r
2
+
1

2
)
_
_
1 +[.[
2
_ _
1 +[n[
2
_
(1)
=
(1)
=
[. n[
2
_
[.[
2
+ 1
_ _
[n[
2
+ 1
_.
This gives
42
d (.. n) =
[. n[
_
_
[.[
2
+ 1
_
_
_
[n[
2
+ 1
_
.
43
I.4.1
Sketch each curve and its image under n = .
2
.
(a) [. 1[ = 1 (c) = 1 (e)
2
= r
2
1. r 0
(b) r = 1 (d) = r + 1 (f) = 1,r. r ,= 0
Solution
-4 -2 2 4
-4
-2
2
4
x
y
I.4.1a z-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.1b z-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.1c z-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.1a w-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.1b w-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.1c w-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.1d z-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.1e z-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.1f z-plane
44
-4 -2 2 4
-4
-2
2
4
x
y
I.4.1d w-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.1e w-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.1f w-plane
45
I.4.2
Sketch the image of each curve in the preceding problem under the
principal branch of n =
_
., and also sketch, on the same grid but
in a dierent color, the image of each curve under the other branch
of
_
..
Solution
n = .
1=2
=
_
[.[ c
i arg[z[
_
1=2
=
_
[.[ c
i Arg[z[+i2n
_
1=2
=
_
[.[c
i Arg z=2
-4 -2 2 4
-4
-2
2
4
x
y
I.4.2a z-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.2b z-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.2c z-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.2a w-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.2b w-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.2c w-plane
46
-4 -2 2 4
-4
-2
2
4
x
y
I.4.2d z-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.2e z-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.2f z-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.2d w-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.2e w-plane
-4 -2 2 4
-4
-2
2
4
x
y
I.4.2f w-plane
47
I.4.3
(a) Give a brief description of the function . n = .
3
, considered
as a mapping from the . plane to the n- plane. (Describe what
happens to n as . traverses a ray emanating from the origin, and as
. traverses a ray a circle centered at the origin.) (b) Make branch
cuts and dene explicitly three branches of the inverse mapping.
(c) Describe the construction of the Riemann surface of .
1=3
.
Solution
(a)
The function n = , (.) = .
3
. For . = :c
i
, we have .
3
= :
3
c
i3
. The radical
rays at angle o are mapped to rays at angle 3o, that is, arg n = 3 arg .. The
magnitude of a complex number is cubed, [n[ = [.[
3
. Circles, centered at the
origin of radius :, are mapped to cocentric circles with radius :
3
.
(b)
We make branch cuts at ( . 0],
. = n
1=3
=
_
[n[ c
i arg w
_
1=3
=
_
[n[ c
i Arg w+i2n
_
1=3
= c
i2n=3
_
[n[c
i Arg w=3
we choose
q (.) = .
1=3
= :
1=3
c
i=3
, : < o < :.
Sheet 1 : Take ,
1
(.) = q (.),
Sheet 2 : Take ,
2
(.) = c
2i=3
q (.),
Sheet 3 : Take ,
3
(.) = c
4i=3
q (.).
(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 , (.) is continuous on
surface
Spara
48

I.4.3a
-1 1
-1
1
I.2.1b
-1 1
-1
1

I.2.1c
Spara
49
I.4.4
Describe how to construct the Riemann surfaces for the following functions
(a) n = .
1=4
. (b) n =
_
. i. (c) n = (. 1)
2=5
.
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 dened 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 to 0.
(b) Use two sheets, can make branch cuts along horizontal line from +i
to i.
(c) Use ve sheets, can make branch cuts along real axis from to 1.
50
I.5.1
Calculate and plot for c
z
for the following points ..
(a) 0 (c) : (i 1) ,3 (e) :i,:. : = 1. 2. 3. . . .
(b) :i + 1 (d) 37:i (f) :(i 1) : = 1. 2. 3. . . .
Solution
-1 1
-1
1
I.5.1e
-0.4 -0.2 0.2
-0.2
0.2
I.5.1f
(a)
c
0
= 1.
(b)
c
i+1
= c
i
c
1
= c.
(c)
c
(i1)3
= c
=3
c
i=3
= c
=3
_
1
2
+
_
3
2
i
_
= 0.351 + 0.006i.
(d)
c
37i
= c
36i
c
i
= 1.
(e)
We take the limit for the sequence c
i=m
, : = 1. 2. 3. . . . as : , and
have
c
i=m
1.
51
as : . Because

c
i=m

= 1 the sequence approaches its limit along a

circle with radius 1.
(f)
We take the limit for the sequence c
m(i1)
= c
m
c
mi
, : = 1. 2. 3. . . . as
: , and have
c
m
c
mi
0.
as : . Because [c
m
c
mi
[ = c
m
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 n = c
z
. Indicate the images of horizontal and vertical
lines in your sketch.
(a) the vertical strip 0 < Re . < 1,
(b) the horizontal strip 5:,3 < Im. < 8:,3,
(c) the rectangle 0 < r < 1, 0 < < :,4,
(d) the disk [.[ _ :,2,
(e) the disk [.[ _ :,
(f) the disk [.[ _ 3:,2.
Solution
-4 -2 2 4
-4
-2
2
4
I.5.2a z-plane
-4 -2 0 2 4
2
4
6
8
I.5.2b z-plane
1 2 3
1
2
3
I.5.2c z-plane
-4 -2 2 4
-4
-2
2
4
I.5.2a w-plane
-4 -2 2 4
-4
-2
2
4
I.5.2b w-plane
1 2 3
1
2
3
I.5.2c w-plane
53
-4 -2 2 4
-4
-2
2
4
I..5.2d z-plane
-4 -2 2 4
-4
-2
2
4
I.5.2e z-plane
-4 -2 2 4
-4
-2
2
4
I.5.2f z-plane
-4 -2 2 4
-4
-2
2
4
I..5.2d w-plane
-10 10 20 30
-20
-10
10
20
I.5.2e w-plane
100
-100
100
I.5.2f w-plane
54
I.5.3
Show that c
z
= c
z
.
Solution
c
z
= c
xiy
= c
x
c
iy
= c
x
(cos () + i sin ()) = c
x
(cos i sin ) =
= c
x
(cos + i sin ) = c
x
(cos + i sin ) = c
x
c
iy
= c
x+iy
= c
z
.
55
I.5.4
Show that the only periods of c
z
are the integral multiples of 2:i,
that is, if c
z+
= c
z
for all ., then ` is an integer times 2:i.
Solution
Show that the only periods of c
z
are the integral muliples of 2:i that is, if
c
z+
= c
z
for all ., then ` is an integer times 2:i.
c
z
= c
z+
= c
z
c

=c

= 1 =` = 2::i
56
I.6.1
Find and plot log . for the following complex numbers .. Specify
the principal value.
(a) 2 (b) i (c) 1 + i (d)
_
1 + i
_
3
_
,2
Solution
(a)
Suppose that : = 0. 1. 2. . . .
log 2 = log [2[ + i Arg 2 + 2::i = log 2 + i2::.
(b)
Suppose that : = 0. 1. 2. . . .
log i = log [i[ + i Arg i + 2::i = i:,2 + i2::.
(c)
Suppose that : = 0. 1. 2. . . .
log (1 + i) =
= log [1 + i[ + i Arg (1 + i) + 2::i = log
_
2 + i:,4 + i2:: =
= log 2,2 + i:,4 + i2::.
(d)
Suppose that : = 0. 1. 2. . . .
log
_
1+i
_
3
2
_
=
= log

1+i
_
3
2

+ i Arg
_
1+i
_
3
2
_
+ 2::i =
= :i,3 + i2::.
57
-1 1 2
-20
-10
10
20
I.6.1a
-1 1 2
-20
-10
10
20
I.6.1b
-1 1 2
-20
-10
10
20
I.6.1c
-1 1 2
-20
-10
10
20
I.6.1d
58
I.6.2
Sketch the image under the map n = Log . of each of the following
gures.
(a) the right halv-plane Re . 0,
(b) the half-disk [.[ < 1, Re . 0,
(c) the unit circle [.[ = 1,
(d) the slit annulus
_
c < [.[ < c
2
, . , (c
2
.
_
c),
(e) the horizontal line = c,
(f) the vertical line r = c.
Solution
(b) We have a disk with radius less then 1, this means [.[ < 1, thus log [.[
< 0 and is unbounded, therefore, it goes from 0 to . Since Re (.) 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 . plane are stright lines in the n plane (n = log .).
Therefore, the image of this annulus under the log function is the rectangle,
bounded by r = log [
_
c[ =
1
2
and r = log [c
2
[ = 2.
-3 -2 -1 1 2 3
-3
-2
-1
1
2
3
I.6.2a z-plane
-2 2
-2
2
I.6.2b z-plane
-4 -2 2 4
-4
-2
2
4
I.6.2c z-plane
59
-3 -2 -1 1 2 3
-3
-2
-1
1
2
3
I.6.2a w-plane
/2
/2
I.6.2b w-plane

I.6.2c w-plane
-8 -6 -4 -2 2 4 6 8
-8
-6
-4
-2
2
4
6
8
I.6.2d z-plane
-4 -2 2 4
-4
-2
2
4
I.6.2e z-plane
-4 -2 2 4
-4
-2
2
4
I.6.2f z-plane
-4 -2 2 4
-4
-2
2
4
I.6.2d w-plane
-4 -2 2 4
-4
-2
2
4
I.6.2e w-plane
-4 -2 2 4
-4
-2
2
4
I.6.2f w-plane
60
I.6.3
Dene explicitly a continuous branch of log . in the complex plane
slit along the negative imaginary axis, C [0. i).
Solution
I.6.3
We have log . = log :c
i
= log : + io. To avoid the negative imaginary axis
we chose :,2 < o < 3:,2. We use the branch
,
_
:c
i
_
= log : + io. :,2 < o < 3:,2.
of log ..
61
I.6.4
How would you make a branch cut to dene a single-valued branch
of the function log (. + i 1)? How about log (. .
0
)?
Solution
Any straight line cut from .
0
to , in any direction, will do.
62
I.7.1
Find all values and plot them.
(a) (1 + i)
i
(b) (i)
1+i
(c) 2
1=2
(d)
_
1 + i
_
3
_
(1i)
Solution
(a)
Suppose that :. : = 0. 1. 2. . . .
(1 + i)
i
=
= c
i log(1+i)
= c
i(log[1+i[+i Arg(1+i)+i2m)
= c
i(log
_
2+i=4+i2m)
= c
2m
c
=4
c
i log
_
2
= [: = :] =
= c
2n
c
=4
c
i log
_
2
.
(b)
Suppose that :. : = 0. 1. 2. . . .
(i)
1+i
=
= c
(1+i) log(i)
= c
(1+i)(log[i[+i Arg(i)+i2m)
= c
(1+i)(i=2+i2m)
=
= c
2m
c
=2
c
i(=2+2m)
= [: = :] = c
2n
c
=2
c
i(=22n)
=
= ic
2n
c
=2
.
(c)
Suppose that : = 0. 1. 2. . . .
2
1=2
=
= c
log 2=2
= c
(log[2[+i Arg 2+i2m)=2
= c
(log 2+i2m)=2
= c
log 2=2
c
im
=
= 1,
_
2.
(d)
Suppose that : = 0. 1. 2. . . .
63
_
1 + i
_
3
_
1i
=
= c
(1i) log(1+i
_
3)
= c
(1i)(log[1+i
_
3[+i Arg(1+i
_
3)+i2n)
=
= c
(1i)(log 2+i=3+i2n)
= c
2n
c
log 2+=3
c
i( log 2+=3+2n)
=
= c
2n
c
log 2+=3
c
i( log 2+=3)
.
64
I.7.2
Compute and plot log
_
(1 + i)
2i
_
.
Solution
We rewrite the expression, suppose that /. :. : = 0. 1. 2. . . .
log
_
(1 + i)
2i
_
=
= log
_
_
c
log(1+i)
_
2i
_
= log
_
c
2i log(1+i)

= log
_
c
2i(log[1+i[+i Arg(1+i)+i2k)

=
= log
_
c
2i(log
_
2+i=4+i2k)
_
= log
_
c
4k
c
=2
c
i log 2

= [/ = :] = log
_
c
4m
c
=2
c
i log 2

=
= log

c
4m
c
=2
c
i log 2

+i Arg
_
c
4m
c
=2
c
i log 2
_
+i2:: = 4:::,2+i log 2+i2:: =
= :,2 + i log 2 + 4:: + 2:i:.
Thus the set is a square lattice.
-10 10 20
-20
-10
10
20
I.7.2
65
I.7.3
Sketch the image of the sector 0 < arg . < :,6 under the map n =
.
a
for
(a) c =
3
2
(b) c = i (c) c = i + 2
Use only the principal branch of .
a
.
Solution
-5 5
-5
5
I.7.3a z-plane
-5 5
-5
5
I.7.3b z-plane
-5 5
-5
5
I.7.3c z-plane
-5 5
-5
5
I.7.3a w-plane
-2 2
-2
2
I.7.3b w-plane
1
1
I.7.3c w-plane
Some points on the y-axis is marked, and the point (1. 0) too.
66
I.7.4
Show that (.n)
a
= .
a
n
a
, where on the right we take all possible
products.
Solution
Let ` (.n)
a
, and /. :. : = 0. 1. 2. . . .
` =
= c
a log(zw)
= c
a(log[zw[+i Arg(zw)+i2n)
= c
a(log[z[+log[w[+i Arg z+i Arg w+i2m+i2n)
=
= c
a(log[z[+i Arg z+i2(m+n))
c
a(log[w[+i Arg w)
.
a
n
a
.
Conversely, if ` .
a
n
a
, say
` =
= c
a log z
c
a log w
= c
a(log[z[+i Arg z+i2m)
c
a(log[w[+i Arg w+i2n)
=
= c
a(log[z[+log[w[+i Arg z+i Arg w+i2(m+n))
= c
a(log[zw[+i Arg(zw)+i2k+i2(m+n))
=
= c
a(log[zw[+i Arg(zw)+i2(k+m+n))
=
= c
a log(zw)
(.n)
a
We have that (.n)
a
= .
a
n
a
as sets
67
I.7.5
Find i
i
i
. Show that is does not coincide with i
ii
= i
1
.
Solution
We rewrite the expression i
i
, suppose that / = 0. 1. 2. . . .
i
i
=
= c
i log i
= c
i(log[i[+i Arg(i)+i2k)
= c
i(i=2+i2k)
=
= c
2k=2
.
Now we can nd i
i
i
we use that we have i
i
, suppose that /. : = 0. 1. 2. . . .
i
i
i
=
=
_
i
i
_
i
= c
i log(i
i
)
= c
i(log(e
2k=2
))
= c
i(log[e
2k=2
[+i Arg(e
2k=2
)+i2m)
=
= c
i(2k=2+i2m)
= c
2m
c
i(2k=2)
= ic
2m
=
= ic
2m
.
Which not coincide with i
ii
= i
1
= 1,i = i which is smaller than
ic
2m
. : = 0. 1. 2. . . ..
68
I.7.6
Determine the phase factors of the function .
a
(1 .)
b
at the branch
points . = 0 and . = 1. What conditions on c and / guarantee that
.
a
(1 .)
b
can be dened as a (continuous) single-valued function
on C [0. 1]?
Solution
Phase factors is c
2ia
at 0 and c
2ib
at 1. Require c
2ia
c
2ib
= 1, or c +/ = :,
where : Z , to have a continuous singlevalued determination of .
a
(1 .)
b
on C [0. 1].
69
I.7.7
Let r
1
< r
2
< < r
n
be : consecutive points on the real axis.
Describe the Riemann surface of
_
(. r
1
) (. r
n
). Show that
for : = 1 and : = 2 the surface is topologically a sphere with certain
punctures corresponding to the branch points and . What is it
when : = 3 or : = 4? Can you say anything for general :? (Any
compact Riemann surface is topologically a sphere with handles.
Thus a torus is topologically a sphere with one handle. For a given
:, how many handles are there, and where do they come from?)
Solution
For : = 1 and : = 2, the functions ,
1
=
_
. r
1
and ,
2
=
_
(. r
1
) (. r
2
),
and for a general : we have ,
n
(r) =
_
(. r
1
) (. r
2
) . . . (. r
n
). To con-
stuct the Riemann Suface we use two sheets with slits [r
1
. r
2
] . [r
3
. r
4
] . . . ..
If : is odd, we also need a slit [r
n
. +) . Identify top edge of slit on one
sheet with bottom edge of slit on other sheet. If : = 1 and : = 2 the surface
is topologically a sphere with certain punctures corresponding to the branch
points and . If : = 3 or : = 4, surface is a torus. For a general even
:, ,
n
(r) will be continuous on C ([r
1
. r
2
] ' [r
3
. r
4
] ' . . . ' [r
n1
. r
n
]), and
its Riemann surface will be a sphere with
n
2
1 holes. For a general odd :
though, ,
n
(r) will be continuous on C ([r
1
. r
2
] ' [r
3
. r
4
] ' . . . ' [r
n
. +]),
its Riemann surface will be a sphere with
n1
2
holes.
70
I.7.8
Show that
_
.
2
1,. can be dened as a (single-valued) continuous
function outside the unit disk, that is, for [.[ 1. Draw branch
cuts so that the function can be dened continuous o the branch
cuts. Describe the Riemann surface of the function.
Solution
The function is .
_
1 1,.
3
. If [.[ 1, can use the principal value of the
square root to dene a branch of the function. There are branch points at
. = 0 and .
3
= 1, that is at 0. 1. c
2i=3
and c
2i=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 identication of cuts and
with points at innity is a torus.
71
I.7.9
Consider the branch of the function
_
. (.
3
1) (. + 1)
3
that is posi-
tive at . = 2. Draw branch cuts so that this branch of the function
can be dened continuously o the branch cuts. Describe the Rie-
mann surface of the function. To what value at . = 2 does this
branch return if it is continued continuously once counterclockwise
around the circle [.[ = 2?
Solution
We have that the function is
(. + 1) .
2
_
.
_
(1 1,.
3
) (1 + 1,.).
If [.[ 1, the second square root can be dened to be single-valued for
[.[ 1. The value of the function at . = 2 returns to the negative of their
initial value then we travers the circle [.[ = 2 , because
_
. does. We can
construct a Riemann Surface by making cuts at 1. 0. 1. c
2i=3
. c
2i=3
and
. The function for the Riemann Surface is
(. + 1)
_
. (. 1) (. + 1) (. c
2i=3
) (. c
2i=3
).
72
I.7.10
Consider the branch of the function
_
. (.
3
1) (. + 1)
3
(. 1) that is
positive at . = 2. Draw branch cut so that this branch of the func-
tion can be dened continuously o the branch cuts. Describe the
Riemann surface of the function. To what value at . = 2 does this
branch return if it is continued continuously once counterclockwise
around the circle [.[ = 2?
Solution
We have that the function is
.
4
_
(1 1,.
3
) (1 + 1,.)
3
(1 1,.)
The branch that is positive for . = 2 is, it is dened continuously for [.[ 1.
Branch return to original value around circle [.[ = 2. We can construct a
Riemann Surface by making cuts at 0. 1. c
2i=3
and c
2i=3
.
The function for the Riemann Surface is
(. 1) (. + 1)
_
. (. + 1) (. c
2i=3
) (. c
2i=3
).
73
I.7.11
Find the branch points of
3
_
.
3
1 and describe the Riemann surface
of the function.
Solution
-1 1
-1
1
I.7.11a
-1 1
-1
1
I.7.11b
-1 1
-1
1
I.7.11c
We rewrite as follows
3
_
.
3
1 =
3
_
(. 1) (. c
2i=3
) (. c
2i=3
).
This equation will have 3 branch points, which are the cube root of unity.
So the phase factor is
c
2i=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 c
2i=3
, on each sheet. In this case the cuts share
common endpoint.
We nee three shets where ,
0
(.), ,
1
(.) = c
2i=3
,
0
(.) and ,
2
(.) = c
2i=3
,
0
(.)
Note: Can use Riemannformula 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 (. + n) = cos . cos n sin . sin n.
(b) sin (. + n) = sin . cos n + cos . sin n.
(c) cosh (. + n) = cosh . sinh n + sinh . sinh n.
(d) sinh (. + n) = sinh . cosh n + cosh . sinh n.
Solution
(a)
Using the denitions of sine and cosine functions given on page 29 in CA we
have,
cos (. + n) =
=
c
i(z+w)
+ c
i(z+w)
2
=
1
4
_
2c
i(z+w)
+ 2c
i(z+w)

=
=
1
4
__
c
i(z+w)
+ c
i(zw)
+ c
i(zw)
+ c
i(z+w)
_
+
+
_
c
i(z+w)
c
i(zw)
c
i(zw)
+ c
i(z+w)
_
=
=
c
iz
+ c
iz
2
c
iw
+ c
iw
2

c
iz
c
iz
2i
c
iw
c
iw
2i
=
= cos . cos n sin . sin n.
(b)
Using the addition formula (a), and using that sin . = cos (. :,2) and
cos . = sin (. :,2) we have,
sin (. + n) = cos
_
. + n
:
2
_
=
= cos . cos
_
n
:
2
_
sin . sin
_
n
:
2
_
=
= cos . sin n + sin . cos n = sin . cos n + cos . sin n.
(c)
Using the addition formula (a) and the formulas cos (i.) = cosh . and sin (i.) =
i sinh . given on page 30 in CA we have,
75
cosh (. + n) = cos (i (. + n)) = cos (i.) cos (in) sin (i.) sin (in) =
= cosh . cosh n i sinh . i sinh n = cosh . cosh n + sinh . sinh n.
(d)
Using the addition formula (b) and the formulas cos (i.) = cosh . and sin (i.) =
i sinh . given on page 30 in CA we have,
sinh (. + n) = i sin (i (. + n)) = i [sin (i.) cos (in) + cos (i.) sin (in)] =
= i [i sinh . cosh n + cosh . i sinh n] = sinh . cosh n + cosh . sinh n.
76
I.8.2
Show that [cos .[
2
= cos
2
r + sinh
2
, where . = r + i. Find all zeros
and periods of cos ..
Solution.
Use trigonometric formulas from page 29 and 30 in CA we have,
cos . = cos (r + i) = cos r cos (i)sin r sin (i) = cos r cosh i sin r sinh .
Now take the modulus squared, and use cosh
2
= 1 + sinh
2
,
[cos .[
2
= cos
2
r cosh
2
+ sin
2
r sinh
2
=
= cos
2
r
_
1 + sinh
2

_
+ sin
2
r sinh
2
=
= cos
2
r +
_
cos
2
r + sin
2
r
_
sinh
2
= cos
2
r + sinh
2
.
The identity for [cos .[
2
shows that the only zeros of cos . are the zeros of
cos . on the real axis, because
cos . = 0 =
_
cos r = 0 = r =

2
+ ::. : = 0. 1. 2. . . . .
sinh = 0 = = 0.
Translation by any period ` of cos . 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 . are 2::. < : < .
77
I.8.3
Find all zeros and periods of cosh . and sinh ..
Solution
We have that cosh . = cos (i.), thus the zeros of cosh . are at . = i:,2 +i::,
: = 0. 1. 2. . . ., and the periods of cosh . are 2::i, : = 0. 1. 2. . . ..
We have that sinh . = i sin (i.), thus the zeros of sinh . are at . = ::i,
: = 0. 1. 2. . . ., and the periods of cosh . are 2::i, : = 0. 1. 2. . . ..
78
I.8.4
Show that
tan
1
. =
1
2i
log
_
1 + i.
1 i.
_
.
where both sides of the identity are to be interpreted as subsets of
the complex plane. In other words, show that tan n = . if and only
if 2in is one of the values of the logarithm featured on the right.
Solution
Set . = tan n, we have that
. = tan n =
sin n
cos n
=
c
iw
c
iw
i (c
iw
+ c
iw
)
=
c
2iw
1
i (c
2iw
+ 1)
.
Solve for c
2iw
c
2iw
=
1 + i.
1 i.
.
and take logarithm and solve for n
n =
1
2i
log
_
1 + i.
1 i.
_
.
Because . = tan n then n tan
1
. and we have the identity
tan
1
. =
1
2i
log
_
1 + i.
1 i.
_
.
79
I.8.5
Let o denote the two slits along the imaginary axis in the complex
plane, one running from i to +i, the other from i to i. Show
that (1 + i.) , (1 i.) lies on the negative real axis (. 0] if and
only if . o. Show that the principal branch
Tan
1
. =
1
2i
Log
_
1 + i.
1 i.
_
maps the slit plane Co one-to-one onto the vertical strip [Re n[ < :,2.
Solution
i
-i
I.8.5 z-plane
-1
I.8.5

I.8.5
n
1
=
1+iz
1iz
n
2
= Log n
1
/2 /2
I.8.5 w-plane
n =
1
2i
n
2
=
1
2i
Log
_
1+iz
1iz
_
We begin to show that
1+iz
1iz
(. 0] if and only if . (i. i] '[i. i).
Set
1 + i.
1 i.
= n
1
and solve for ., thus
80
. =
1 n
1
1 + n
1
i
As n
1
goes from to 1,
1w
1
1+w
1
i goes from i to ialong iR, and when
n
1
goes from 1 to 0,
1w
1
1+w
1
i goes from i to i along iR.
Since we have that
1 + i.
1 i.
= n
1
=. =
1 n
1
1 + n
1
i
the map is one to one.
Remark that the map makes correspondence between the interval [i. i]
in . plane and [. 1] in n
1
plane, and between interval [i. i] in .
plane and [1. 0] in n
1
plane.
We have that n
2
= Log n
1
maps C(. 0) onto [Imn[ < :, thus n =
w
2
2i
maps [Imn
2
[ < : onto
_
[Re n[ <

2
_
, se gures.
We have that the function
Tan
1
. =
1
2i
Log
_
1 + i.
1 i.
_
maps Co onto
_
[Re n[ <

2
_
, where Tan
1
. is the principal branch for
tan
1
.. The other branches are given by ,
n
(.) = Tan
1
. +::, < : <
.
81
I.8.6
Describe the Riemann surface for tan
1
..
Solution
i
-i
+

+

+

+
+

+

+

+
_

_

_

_
_

_

_

_
I.8.6 z-plane
-1
+ + + + + +
_ _ _ _ _ _
I.8.6

+ + + + + + + + + + +
_ _ _ _ _ _ _ _ _ _ _
I.8.5
n
1
=
1+iz
1iz
n
2
= Log n
1
/2 /2
+

+

+

+

+

+

+

+

+

+

+
_

_

_

_

_

_

_

_

_

_

_
I.8.6 w-plane
n =
1
2i
n
2
=
1
2i
Log
_
1+iz
1iz
_
We have that
,
0
(.) = Tan
1
. =
1
2i
Log
_
1 + i.
1 i.
_
.
other branches of tan
1
. are ,
n
(.) = ,
0
(.) + ::, where < : < .
Use one copy of the double slit plane o for each integer :, and dene ,
n
(.) =
Tan
1
. + :: on the :
th
sheet to the (: + 1) th sheet along one of the cuts,
so that ,
n
(.) and ,
n+1
(.) have same value at the junction.
82
Make innite many copies of o and call themo
n
. Dene ,
n
(.) = Tan
1
.+::
on o
n
. Identify "+" side of cut on o
n
to "" side of cut on o
n+1
. Then
,
n
on o
n
continuous to ,
n+1
on o
n+1
, and ,
n
maps o
n
onto vertical strip
_
:
1
2
< Re . < : +
1
2
_
. Note that composite function is not dened at
i and (endpoints of slits), and its image omits the sequence
1
2
+ :,
< : < .
83
I.8.7
Set n = cos . and = c
iz
. Show that = n
_
n
2
1. Show that
cos
1
n = i log
_
n
_
n
2
1
_
.
where both sides of the identity are to be interpreted as subsets of
the complex plane.
Solution
Set n = cos . and = c
iz
, we have that
n = cos . =
c
iz
+ c
iz
2
=
+ 1,
2
.
Solve for
= n
_
n
2
1.
and set = c
iz
and take the logarithm and solve for .
. = i log
_
n
_
n
2
1
_
.
Because n = cos . then . cos
1
n we have the identity
cos
1
n = i log
_
n
_
n
2
1
_
.
84
I.8.8
Show that the vertical strip [Re (n)[ < :,2 is mapped by the function
. (n) = sin n one-to-one onto the complex . plane with two slits
(. 1] and [+1. +) on the real axis. Show that the inverse
function is the branch of sin
1
. = i Log
_
i. +
_
1 .
2
_
obtained by
taking the principal value of the square root. Hint. First show
that the function 1 .
2
on the slit plane omits the negative real
axis, so that the principal value of the square root is dened and
continuous on the slit plane, with argument in the open interval
between :,2 and :,2.
Solution
Set n = r + i, by formulas on page 30 in CA we have that
sin n = sin (r + i) = sin (r) cos (i) + cos (r) sin (i) =
= sin r cosh () + i cos (r) sinh .
If [Re n[ = [r[ < :,2 then sin n is mapped on C (. 1] ' [+1. +),
because
Im(sin n) = 0 = = 0 =sin n = sin (r)
and
1 < sin (r) < 1
for these r.
Let now . = sin n. The value of n is referred to as sin
1
., that is, the
complex number whose sine is .. Note that
. =
c
iw
c
iw
2i
. (1)
Now with = c
iw
and 1, = c
iw
in (1), we have
. =
1,
2i
.
Mulitplying the above by 2i and doing some rearranging, we nd that
85
2i. =
2
1 or
2
2i. 1 = 0.
With the quadratic formula, we solve this equation for and nd
= i. +
_
1 .
2
or c
iw
= i. +
_
1 .
2
We now take the logarithm of both sides of the last equation and divide the
result by i to obtain
n =
1
i
log
_
i. +
_
1 .
2
_
and, since n sin
1
.,
sin
1
. = i log
_
i.
_
1 .
2
_
.
86