VAN NOSTRAND MATHEMATICAL STUDIES

Editors
Paul R. Halmos, The University of Michigan
Frederick W. Gehring, The University of Michigan
Paul R. Halmos - LECTURES ON BOOLEAN ALGEBRAS
Shmuel Agmon-LECTURES ON ELLIPTIC BOUNDARY
VALUE PROBLEMS
Noel J. Hicks-NOTES ON DIFFERENTIAL GEOMETRY
Leopoldo Nachbin-TOPOLOGY AND ORDER
Sterling K. Berberian-NOTES ON SPECTRAL THEORY
Roger C. Lyndon-NOTES ON LOGIC
Robert R. Phelps-LECTURES ON CHOQUET'S THEOREM
L. E. Sigler-EXERCISES IN SET THEORY
George W. Mackey-LECTURES ON THE THEORY OF
FUNCTIONS OF A COMPLEX VARIABLE
Lars V. Ahlfors-LECTURES ON QUASICONFORMAL MAPPINGS
Additional titles will be listed and announced as issued.
Lectures on
QUASICONFORMAL MAPPINGS
by
LARS v: AHLFORS
Harvard University
Manuscript prepared with the assistance of
CLIFFORD J. EARLE Jr.
Cornell University
D. VAN NOSTRAND COMPANY, INC.
PRINCETON, NEW JERSEY
TORONTO NEW YORK LONDON
D. VAN NOSTRAND COMPANY, INC.
120 Alexander St., Princeton, New Jersey (Principal office)
24 West 40 Street, New York 18, New York
D. VAN NOSTRAND COMPANY, LTD.
358, Kensington High Street, London, W.14, England
D. VAN NOSTRAND COMPANY (CANADA), LTD.
25 Hollinger Road, Toronto 16, Canada
COPYRIGHT © 1966, BY
D. VAN NOSTRAND COMPANY, INC.
Published simultaneously in Canada by
D. VAN NOSTRAND COMPANY (Canada), LTD.
No reproduction in any form of this book, in whole or in
part (except for brief quotation in critical articles or reviews),
may be made without written authorization from the publishers.
PRINTED IN THE UNITED STATES OF AMERICA
ACKNOWLEDGMENTS
The manuscript was prepared by Dr. Clifford Earle from
rough long hand notes of the author. He has contributed many
essential corrections, checked computations and supplied many
of the bridges that connect one fragment of thought with the next.
Without his devoted help the manuscript would never have attain-
ed readable form.
In keeping with the informal character of this little vol-
ume there is no index and the references are very spotty, to say
the least. The experts will know that the history of the subject
is one of slow evolution in which the authorship of ideas cannot
always be pinpointed.
The typing was excellently done by Mrs. Caroline W.
Browne in Princeton, arid financed by Air Force Grant AFOSR-
393-63.
CONTENTS
CHAPTER I: DIFFERENTIABLE QUASICONFORMAL
MAPPINGS
Introduction . . . . . . . . . . . . . . . . . . . . . . . .. 1
A. The Problem and Definition of Grotzsch . . .. 2
B. Solution of Grotzsch's Problem. . . . . . . . .. 6
C. Composed Mappings. . . . . . . . . . . . . . . .. 8
D. Extremal Length " 10
E. A Symmetry Principle . . . . . . . . . . . . . . .. 16
F. Dirichlet Integrals. . . . . . . . . . . . . . . . .. 17
CHAPTER II: THE GENERAL DEFINITION
A. The Geometric Approach . . . . . . . . . . . . .. 21
B. A 1-q-c. map is Conformal. . . . . . . . . . . .. 23
CHAPTER III: EXTREMAL GEOMETRIC PROPERTIES
A. Three Extremal Problems. . . . . . . . . . . . .. 35
B. Elliptic and Modular Functions. . . . . . . . .. 40
C. Mori's Theorem. . . . . . . . . . . . . . . . . . .. 47
D. Quadruplets....................... 53
CHAPTER IV: BOUNDARY CORRESPONDENCE
A. The M-condition. . . . . . . . . . . . . . . . . . .. 63
B. The Sufficiency of the M-condition. . . . . . .. 69
C. Quasi-isometry . . . . . . . . . . . . . . . . . . .. 73
D. Quasiconformal Reflection. . . . . . . . . . . .. 74
E. The Reverse Inequality. . . . . . . . . . . . . .. 81
Quasiconformal Mappings
CHAPTER V: THE MAPPING THEOREM
A. Two Integral Operators. . . . . . . . . . . . . .. 85
B. Solution of the Mapping Problem. . . . . . . .. 90
C. Dependence on Parameters . . . . . . . . . . .. 100
D. The Calder6n-Zygmund Inequality. 106
CHAPTER VI: TEICHMULLER SPACES
A. Preliminaries . . . . . . . . . . . . . . . . . . . .. 117
B. Beltrami Differentials 120
C. !!J. is Open 128
D. The Infinitesimal Approach . . . . . . . . . . .. 137
CHAPTER I
DIFFERENTIABLE QUASICONFORMAL MAPPINGS
Introducti on
There are several reasons why quasiconformal mappings
have recently come to playa very active part in the theory of
analytic functions of a single complex variable.
1. The most superficial reason is that q.c. mappings are
a natural generalization of conformal mappings. If this were their
only claim they would soon have been forgotten.
2. It was noticed at an early stage that many theorems
on conformal mappings use only the quasiconformality. It is
therefore of some interest to determine when conformality is es-
sential and when it is not.
3. Q.c. mappings are less rigid than conformal mappings,
and are therefore much easier to use as a tool. This was typi-
cal of the utilitarian phase of the theory. For instance, it was
used to prove theorems about the conformal type of siinply con-
nected Riemann surfaces (now mostly forgotten).
1
2 Quasiconformal Mappings
4. Q.c. mappings play an important role in the study of
certain elliptic partial differential equations.
5. Extremal problems in q.c. mappings lead to analytic
functions connected with regions or Riemann surfaces. This
was a deep and unexpected discovery due to Teichmiiller.
6. The problem of moduli was solved with the help of
q.c. mappings. They also throw light on Fuchsian and Kleinian
groups.
7. Conformal mappings degenerate when generalized to
several variables, but q.c. mappings do not. This theory is
still in its infancy.
A. The Problem and Definition of Grotzsch
The notion of a quasiconformal mapping, but not the
name, was introduced by H. Grotzsch in 1928. If Q is a
square and R is a rectangle, not a square, there is no confor-
mal mapping of Q on R which maps vertices on vertices.
Instead, Grotzsch asks for the most nearly conformal mapping
of this kind. This calls for a measure of approximate confor-
mality, and in supplying such a measure Grotzsch took the
first step toward the creation of a theory of q.c. mappings.
All the work of Grotzsch was late to gain recognition,
and this particular idea was regarded as a curiosity and allowed
to remain dormant for several years. It reappears in 1935 in the
. work of Lavrentiev, but from the point of view of partial differ-
ential equations. In 1936 I included a reference to the q.c. case
Differentiable Quasiconformal Mappings
in my theory of covering surfaces. From then on the notion
became generally known, and in 1937 Teichmiiller began to
prove important theorems by use of q.c. mappings, and later
theorems about q.c. mappings.
3
We return to the definition of Gr6tzsch. Let w = f(z)
(z = x + iy, w = u + iv) be a C
1
homeomorphism from one re-
gion to another. At a point Zo it induces a linear mapping of
the differentials
(1)
du
dv
u dx + u dy
x y
v dx + v dy
x y
which we can also write in the complex form
(2)
with
dw = f dz + f- dz
?: Z
(3) f-=';Ii(f+if)
z x y
Geometrically, (1) represents an affine transformation
from the (dx, dy) to the (du, dv) plane. It maps circles a-
bout the origin into similar ellipses. We wish to compute the
ratio between the axes as well as their direction.
In classical notation one writes
(4)
with
G
The eigenvalues are determined from
4 Quasiconformal Mappings
(5)
and are
\
E- A P I
P G- A
o
(6)
\ \ _ E+G+[(E_G)2+4p
2
]'h
1\1' 1\2 - 2
(7)
The ratio a: b of the axes is
(
Al)lh = E+ G+[(E_G)2+4p
2
]'h
A
2
2(EG _ p
2
) 'h
The complex notation is much more convenient. Let us
first note that
(8)
This gives
f-
z
lh (u +v ) + j ~ (v - u )
x y x y
yiu - v ) + j ~ (v +u )
x y x y
(9)
I
f 1
2
- \ f_1
2
= u v - u v = ]
z z xy yx
which is the Jacobian. The Jacobian is positive for sense pre-
serving and negative for sense reversing mappings. For the
moment we shall consider only the sense preserving case.
Then 1f'Z I < 1f
z
1.
It now follows immediately from (2) that
where both limits can be attained. We conclude that the ratio
of the major to the minor axis is
(11)
If 1+ If-I
D - z z > 1
f - 1f zl - If'Z 1
Differential Quasiconformal Mappings
This is called the dilatation at the poiJ1t z. It is often more
convenient to consider
5
(12)
related to D l by
(13)
The mapping is conformal at z if and only if Dl
The maximum is attained when the ratio
1, d
l
= O.
f- clZ
z
f zdz
is positive, the minimum when it is negative. We introduce now
the complex dilatation
(14)
with Illll
(15)
f-
III = f z
z
d
l
. The maximum corresponds to the direction
arg dz = a = 'h arg fl
the minimum to the direction a ±"/2. In the dw-plane the di-
rection of the major axis is
(16)
where we have set
arg dw 'h arg v
(17)
The quantity v l may be called the second complex dilatation.
6
figure:
Quasiconformal Mappings
We will illustrate by the following self-explanatory
Observe that (3 - a = arg f z'
Definition 10 The mapping f is said to be quasiconfor-
mal if Df is bounded. It is K-quasiconformal if Df S K.
The condition Df SKis equivalent to d
f
S k =
(K -l)/(K+ 1). A l-quasiconformal mapping is conformal.
Let it be said at once that the restriction to C1-mappings
is most unnatural. One of our immediate aims is to get rid of
this restriction. For the moment, however, we prefer to push
this difficulty aside.
B. Solution of Grotzsch's Problem
We pass to Grotzsch's problem and give it a precise
meaning by saying that f is most nearly conformal if sup Df
is as small as possible.
Let R, R' be two rectangles with sides a, b and a ~
b '. We may assume that a: b sa': b ' (otherwise, interchange
a and b). The mapping f is supposed to take a-sides into
a-sides and b-sides into b-sides.
Differentiable Quasiconiormal Mappings 7
60
6' I
a
a'
The computation goes
a
a
,
S
f
Idi(x+ iY)1
f
(Iizl + liz I) dx
a
<
0 0
a b
a'b < f f (Iizl + l i ~ ) dx dy
o 0
lizl + lizl
lizl -lizl
a b
dx dy f f
o 0
a b
a' b ' f f Df dx dy
o 0
or
(1)
f f Dfdx dy
R
and in particular
~ . ~ < sup D
f
b' . b -
The minimum is attained for the affine mapping which
is given by
Hz)
8 Quasiconiormal Mappings
THEOREM 1. The affine mapping has the least maximal
and the least average dilatation.
The ratios m = al band m' = a' Ib' are called the
modhles of R and oR' (taken with an orientation). We have
proved that there exists a K-q.c. mapping of R on R' if and
only if
(2)
C. Composed Mappings
We shall determine the complex derivatives and complex
dilatations of a composed mapping g 0 i. There is the usual
trouble with the notation which is most easily resolved by in-
troducing an intermediate variable ,= f(z).
The usual rules are applicable and we find
(1)
(g 0 f) z = (g, 0 f) i z + (g, 0 f) 7z
( ~ 0 f)- = (g 0 f) i- + (g- 0 f) 7-
5 Z , Z , Z
When solved they give
(2)
where
1 -
- [(g 0 f) i-
1 z z
g- 0 i 1[(g 0 f)- i
, 1 z z
2 2
1 = Iizl - Iii I .
- (g 0 f) -7 ]
z z
- (g 0 f) i-]
z z
For g = i-I the formulas become
(3)
(i-I) 0 i = 7-11 ,
, z
Differ€f1tiable Quasicon£ormal Mappings 9
One derives, for instance,
(4) £
-1
/l -1 = -1/ f °
f
(5)
and, on passing to the absolute values,
d -1 = d
f
°£-1
f
In other words, inverse mappings have the same dilata-
tion at corresponding points.
From (2) we obtain
(6)
/l12 °£
£z /l12 0 f-/lf
lz 1 - 'ji.f/l12 0 f
If g is conformal, then /l12
(7)
o and we find
/l f .
If £ is conformal, /l f = 0 and
(8)
f' 2
/l o£ = ( ~
12 1£'[
which can also be written as
(9)
In any case, the dilatation is invariant with respect to all con-
formal transformations.
If we set g °£ h we find from (6)
(10)
£z /lh - /If
lz 1-'ji.f/lh
10
For the dilatation
Quasiconformal Mappings
(11)
and
d 0 f
h 0 r
1
(12)
log D -1 0 f == (Ilh' I l ~ ,
h 0 f
the non-euclidean distance (with respect to the metric ds
2 Idwl in Iwi < 1).
l-lwl
2
We can obviously use sup (Ilh' I l ~ as a distance be-
tween the mappings f and h (the Teichmiiller distance). It
is a metric provided one identifies mappings that differ by a
conformal transformation.
The composite of a K
1
-q.c. and a K
2
-q.c. mapping is
K
1
K
2
- q.c.
D. Extremal Length
Let r be a family of curves in the plane. Each y( r
shall be a countable union of open arcs, closed arcs or closed
curves, and every closed subarc shall be rectifiable. We shall
introduce a geometric quantity ,.\(r), called the extremal l€f1gth
of r, which is a sort of average minimal length. Its import-
ance for our topic lies in the fact that it is invariant under con-
formal mappings and quasi-invariant under q.c. mappings (the
latter means that it is multiplied by a bounded factor).
A function p, defined in the whole plane, will be called
allowable if it satisfies the following conditions:
Differentiable Quasiconformal Mappings 11
1. P? 0 and measurable.
2. A(p) = JJ p2 dx dy f, 0, 00 (the integral is over
the whole plane).
For such a p, set
Ly(p) fp Idzl
Y
if p is measurable on y*, L
y
(p) 00 otherwise. We intro-
duce
L (p) = inf L
y
(p)
y ~ - r
and
Defini tion:
L (p)2
sup --
p A(p)
for all allowable p.
We shall say that r 1 < r 2 if every y2 contains a y1
(the y2 are fewer and longer).
Remark. Observe that r 1 C r 2 implies r 2 < r 1 !
THEOREM 2. If r 1 < r 2' then h(r 1) :S h(r2) .
Proof. If Y1 S; Y2 then
L
Y1
(p) < L
Y2
(p)
inf L
Y1
(p) < inf L
Y2
(p)
and it follows at once that h (r1) :S h([' 2) .
*
(as a function of arc-length)
12 Quasiconformal Mappings
Example 1. [' is the set of all arcs in a closed rec-
tangle R which joins a pair of opposite sides.
For any p
a
f p(x + iy) dx > L (p)
o
f f p dx dy > b L (p)
R
b
2
L(p)2 < ab f f p2 dx dy < ab A(p)
R
L(p)2 < a
A(p) - b
This proves A(n ~ a/b.
On the other hand, take p = 1 in R, P = 0 outside.
Then L(p) = a, A(p) = ab, hence A(n ;:: a/b. We have
proved
a
b
Example 2. [' is the set of all arcs in an annulus
r 1 < Iz I ~ r2 which join the boundary circles.
Computation:
f f p dr dO > 217 L (p )
Differentiable Quasiconformal Mappings 13
f f p2 rdrde
Equality for p = IIr .
Example 3. The module of an annulus.
Let G be a doubly connected region in the finite plane
with C
1
the bounded, C
2
the unbounded component of the com-
plement. We say the closed curve y in G separates C1 and
C
2
if Y has non-zero winding number about the points of CI'
Let r be the family of closed curves in G which separate C1
and C
2
• The module M(G) A(r)-I, Consider, for example,
theannulus G = Ir
1
::; Izi :s r21,
217
L (p)
:s
f
0
217
L (p)
<
f
r
0
p de
r
f f pdrde
L (p) log (r 2 )
<
1
L(p )2
2 r
217 log (;:) f f / rdrd{) log (/)
<
1
L(p )2
217
<
A(p) log( r
2
/ r
1
)
14 Quasiconformal Mappings
so Up) = 1 and A (p)
M(G) il7 log(r
2
/r
1

Once again p = 1/217r gives equality. Indeed, for any y € 1
we have
1 < !n(y, 0)1 = .L [ Jdz 1 <.L f ~ -L (y)
- 217 Z - 217 IZ I - p ,
y y
= il7 log(r
2
/r1)' We conclude that
Suppose that all y € 1 are contained in a region n
and let ¢ be a K-quasiconformal mapping of n on n'. Let
l' be the image set of 1.
THEOREM 3. K-
1
.\(1) ~ .\(1') ~ K .\(1).
Proof. For a given p(z) define p' V;) = 0 outside
n' and
in n'. Then
fp'ldt:[ ?: fp [dz[
y y
f f
,2 dt:d = f f 2 [¢zl + [¢zl dxdy <_ KA(p) .
p S TJ n P 1¢z[ - 1¢zl
This proves .\' ?: 1\1.\, and the other inequality follows by
con!:!idering the inverse.
COROLLARY. '\(1) is a conformal invariant.
There are two important composition principles.
Differentiable Quasiconformal Mappings 15
*
I. 1
1
+ 1
2
= IY1 +Y21 Y1 f1
1
, Y2 f
1
2
1
II. 1
1
u r
2
THEOREM 4.
a) A(1
1
+ 1
2
) > A(1
1
) + A(1
2
) ;
b) A(1
1
U 1
2
)-1 2: A(1
1
)-1 + A(1
2
)-1
if 1
1
, 1
2
lie in disjoint measurable sets.
Proof of a). We may assume that 0 < A(1
1
), A(1
2
) < 00
for otherwise the inequality is trivial.
We may normalize so that
L 1(p 1) A (p 1)
L
2
(P2) = A (P2 )
Choose p
L(p) > L
1
(P1) + L
2
(P2)
A (p) < A (P1 ) + A (P2)
L(p)2
A = sup ATPT 2: A (P1 ) + A (P2 )
It follows that A 2: A1 + A2 .
+
Proof of b). If A = A(1
1
U 1
2
) = 0 there is no-
thing to prove. Consider an admissible p with L (p) > 0 and
set P1 = P on E
1
, P2 = P on E
2
, 0 outside (where E
1
and E
2
are complementary measurable sets with r 1 S E1 '
* Y1 + Y2 means "Y
1
followed by Y2'"
16 Quasiconformal Mappings
['2 S E/ Then L
I
(PI) 2: L(p),
A(p) = A(PI) + A(P2)' Thus
A (p) A (PI)
L(p)2 > L
I
(PI)2 +
and hence
q.e.d.
E. A Symmetry Principle.
For any y let y be its reflection in the real axis,
and let y+ be obtained by reflecting the part below the real
axis and retaining the part above it (y U y = y+ U (y+)-).
The notations rand ['+ are self-explanatory.
THEOREM 5. If [' = r then ,\([') = Yi ,\(['+) .
Proof.
1. For a given P set p(z)
Then
max (p (z), P(z)) .
and
This makes
L+(p)2 ~ 2 L(p)2 S 2,\([')
A(p) A(p)
and hence ,\ ([' +) S 2,\ ([' ) .
2. For given P set
Differentiable Quasiconformal Mappings 17
in upper halfplane
in lower halfplane
Then
L +(p+) = L + ( +)_ (p) L _ (p)
y y + y y+y
L (p) + L_(p) > 2L(p) .
y y
On the other hand
2A (p)
and hence
L(p)2
L +(p+)2
< Y2
y
<
Y2 ,\,(r+)
A (p) A (p+)
,\,(r)
<
Y2 ,\,(r+) .
q.e.d.
F. Dirichlet Integrals
z
Let ¢ be a K-q.c. mapping from n to n'. The
Dirichlet integral of a C
1
function u «() is
For the composite u 0 ¢ we have
18 Quasiconformal Mappings
(U 0 c/J)z (ul;, 0 c/J)c/Jz + (ul;, °c/J) ¢z
I(u °c/J) zl $ <iul;,l °c/J)<ic/Jzl + 1c/J:zP
and thus
(1)
D(u0c/J) S KD(u) .
Dirichlet integrals are
There is another formulation of this. We may consider
merely corresponding Jordan regions with boundaries y;' y , .
Let v on y' and v 0 c/J on y be corresponding boundary
values. There is a minimum Dirichlet integral Do(v) for func-
tions with boundary values v, attained for the harmonic func-
tion with these boundary values. Clearly,
(2)
One may go a step further and assume that v is given
only on part of the boundary. For instance, if v == 0 and
v == 1 on disjoint boundary arcs we get a new proof of the quasi-
invariance of the module.
In order to define the Dirichlet integral it is not neces-
sary to assume that u is of class C1. Suppose that u (z) is
continuous with compact support. Thus we can form the Fourier
transform
Differentiable Quasiconformal Mappings
Tf) = !! y)dxdy
n
and we know that
(UX)A = -
(I).) A = - iTfIi
It follows by the Plancherel formula that
D(u) =! !(e+Tf2)[liI2
and this can be taken as definition of D(u) .
19
CHAPTER II
THE GENERAL DEFINITION
A. The Geometric Approach
All mappings ¢ will be topological and sense preserv-
ing f rom a region 11 to a region 11'.
Definition A. ¢ is K-q.c. if the modules of quadrilat-
erals are K-quasi-invariant.
A quadrilateral is a Jordan region Q, Qc 11, together
with a pair of disjoint closed arcs on the boundary (the b-arcs).
Its module m(Q) = a/ b is determined by conformal mapping
on a rectangle
a
The conjugate Q* is the same Q with the complemen-
tary arcs (the a-arcs). Clearly, m(Q*) m(Q)-l.
21
22
Quasiconformal Mappings
The condition is m(Q ') ~ Km (Q ). It clearly implies
a double inequality,
l\lm(Q) ::; m(Q') S Km(Q) .
Trivial properties:
1. If ¢ is of class C
1
, the definition agrees with
the earlier.
2. ¢ and ¢-l are simultaneously K-q.c.
3. The class of K-q.c. mappings is invariant under
conformal mappings.
4. The composite of a K l-q.c. and a K
2
-q.c. map-
ping is K
1
K
2
-q.c.
K-quasiconformality is a local property:
THEOREM 1. If ¢ is K-q.c. in a neighborhood of
every point, then it is K-q.c. in n.
Proof.
Q
-
.---..
-
Q
We subdivide Q into vertical strips Qj and then each
image, Q.' into horizontal strips Q.. ' with respect to the ree-
l IJ
tangular structure of Q.'. Set m.. = m(Q .. ) etc. Then
1 IJ IJ
m = ~ m
j
.1. > ~ J ' _1_
m
i
- m
ij
~ . _1_,
J m..
IJ
The General Definition
23
If the subdivision is sufficiently fine we shall have m
I

J
• < Km . ~ .
- IJ
This gives m $ Km'.
LEMMA. m = m
1
+ m
2
only if the dividing line is
x = m
1
.
Proof. Let the conformal mapping functions be f
1
, f
2
.
Set p = If
1
'I in Q1' P = If
2
'I in Q2' P = 0 everywhere.
else. Then, integrating over Q
f f (p
2
- 1) dx dy 0
f f (p - 1) dx dy ::: 0
But fJ(p-1)2dxdy = ![[(p2-1)-2(p-1)]dxdy $ O. Hence
p = 1 a.e. and this is possible only if f 1 = f
2
= z.
THEOREM 2. A l-q.c. map is conformal.
Proof. We must have equality everywhere in the proof
of Theorem 1. This shows that the rectangular map is the iden-
tity.
B. The Analytic Definition
We shall say that a function u(x, y) is ACL (absolutely
24
Quasiconformal Mappings
continuous on lines) in the region n if for every closed rec-
tangle R c n with sides parallel to the x and y-axes,
u(x, y) is absolutely continuous on a.e. horizontal and a.e.
vertical line in R. Such a function has of course, partial de-
rivatives ux' u
y
a.e. in n.
The definition carries over to complex valued functions.
Definition B. A topological mapping ¢ of n ia K-q.c.
if
1) ¢ is ACL in n;
2) I¢zl ~ kJ¢zl a.e.
(
k = K-l)
K+l .
We ;;hall prove that this definition is equivalent to the
geometric definition. It follows from B that ¢ is sense pre-
serving.
We shall say that ¢ is differentiable at Zo (in the
sense of Darboux) if
¢(z)-¢(zo) = ¢z(zo)(z-zo)+¢z(zo)(z -zo)+ o(\z-zoP
The first lemma we shall prove is an amazing result due to Geh-
ring and Lehto.
LEMMA 1. If ¢ is topological* and has partial deriva-
tives a.e., then it is differentiable a.e.
By Egoroff's theorem the limits
¢(z+M-¢(z)
h
P(z + ik) - P(z )
k
'"
The result holds as soon as ¢ is an open mapping.
The General Definition 25
are taken on uniformly except on a set n - E of arbitrarily
small measure. It will be sufficient to prove that ¢ is differ-
entiable a.e. on E.
Remark. Usually Egoroff's theorem is formulated for
sequences. We obtain (1) if we apply it to
\
¢(z+h)-¢(z) I
siP -¢)z) .
0< Ih <lin h
The set E is measurable, and therefore it intersects
a.e. horizontal line in a measurable set. On such a line a.e.
point in E is a point of linear density 1. The same holds for
vertical lines. Therefore a.e. x
o
' Yo ( E is a point of linear
density 1 for the intersections of E with x = X
o
and y = Yo'
It will be sufficient to prove that ¢ is differentiable at such a
point Zo = X
o
+ iyo' and for simplicity we assume Zo = O.
Because of the uniformity ¢x and ¢y are continuous
on E. Given ( > 0 we can therefore find a [) > 0 such that
Ih I < [), [k I < [) and
(2)
as soon as
z ( E.
I¢)z) - ¢x(O)1 < (
[¢/z) - ¢y(O)1 < (
I¢ (z +h) - ¢ (z )
h
I¢ (z + ik) - ¢ (z)
k
Ix I < [), [y I < [),
We follow the argument which is used when one proves
that a function with continuous partial derivatives is differen-
tiable. It is based on the identity
26 Quasiconformal Mappings
¢(x+ iy) - ¢(Q) - x¢x(O) - Y¢y(O) =
[¢(x+ iy) - ¢(x) - y ¢y (x)] + [¢(x) - ¢(Q) - x ¢x (0)]
+ [y (¢y (x) - ¢y (0))]
If x f E we are able to derive by (2) that
The same reasoning can be repeated if y f E. Thus we have
proved what we wish to prove if either x f E or y f E.
We now want to make use of the fact that 0 is a point
of linear density 1. Let m
l
(x) be the measure of that part of
E which lies on the segment (-x, xL Then (ml(x))/2Ixl -> I
for x -> 0, and we can choose 8 so small that
m (x) > ~ I x l
1 I + f
for Ix [ < 8. For such an x the interval (-IX,x) cannot be
+f
free from points of E, for if it were we should have
Ix [ 2H
ml(x) ::; Ixl +I+f = IH Ix[
The same reasoning applies on the y-axis. If [z I < _8_ we
I +f
conclude that there are points xl' x
2
' Yl' Y2 f E with
We may then conclude that (3) holds on the perimeter of the
*
We are assuming for convenience that z lies in the first
quadrant.
The General Definition 27
rectangle ( ~ , ~ ) x (Y
1
' Y
2
).
To complete the reasoning we use the fact that ¢ sat-
isfies the maximum principle. There is a point z* = x* + iy*
on the perimeter such thai
I¢(x+ iy) - ¢(O) - x¢ (0) - y¢ (0)1
x y
$ 1¢(x*+iy*)-¢(O)-x¢x(O)-Y¢y(O)1
< 3f Iz* 1+ Ix-x* 1 I¢ (0)1 + Iy-y* 11¢ (0)1
x y
< 3dlH)lzl + fl¢ (O)llzl + fl¢ (O)llzl
x y
This is an estimate of the desired form, and the lemma is proved.
We can say a little more. If E is a Borel set in n we
define A (E) as the area of its image. This defines a locally
finite additive measure, and according to a theorem of Lebesgue
such a measure has a symmetric derivative a.e., that is,
f(z) lim .AlQ)
m(Q)
when Q is a square of center z whose side tends to zero.
Moreover,
ff(z)dxdy $ A(E)
E
(we cannot yet guarantee equality). But if ¢ is differentiable
at z it is immediate that f (z) is the Jacobian, and we have
proved that the Jacobian is locally integrable.
But f = I¢z 1
2
- I¢Z' 1
2
and if 2) is fulfilled we ob-
tain
28
Quasiconformal Mappings
We conclude that the partial derivatives are locally square in-
tegrable.
Moreover, if h is a test function (h ( C
l
with com-
pact support) we find at once by integration over horizontal or
vertical lines and subsequent use of Fubini's theorem
(4)
f f ¢x
h
dxdy
f f ¢yh dxdy
-f f ¢h
x
dxdy
-f f ¢hydxdy
In other words, ¢x and ¢y are distributional derivatives of ¢.
More important still is the converse:
LEMMA 2. If ¢ has locally integrable distributional
derivatives, then ¢ is ACL.
Proof. We assume the existence of integrable functions
¢l' ¢2
such that
f f¢l h dxdy -f f ¢ h
x
dxdy
(5)
f f ¢2 h dxdy - f f¢hydXdY
for all test functions.
Consider a rectangle RTJ = {O ::; x ::; a, 0 ::; y ::; TJI and
choose h = h (x) k (y) with support in RTJ' In
f f ¢l h (x )k (y) dx dy = - f f ¢ h' (x)k (y) dx dy
RTJ RTJ
we can first let k tend boundedly to 1. This gives
The General Definition
f f ¢I h(x)dxdy
RTJ
and hence
a
f ¢ 1(x, TJ)h (x )dx
o
f f ¢h'(x)dxdy
RTJ
a
-f ¢(x, TJ)h '(x )dx
o
for almost all TJ. Now let h "" h
n
test-functions such that °< h <
- n-
B- .! ). We obtain
n
run through a sequence of
1 and h "" 1 on (1.,
n n
¢(a, TJ) - ¢(o, TJ) (6)
a
f ¢I(x,TJ)dx
o
for almost all TJ. The exceptional set does depend on a, but
we may conclude that (6) holds a.e. for all rational a. By con-
tinuity it is then true for all a, and we have proved that
¢(x, TJ) is absolutely continuous for almost all TJ. Moreover,
we have ¢x "" ¢1., ¢y "" ~ a.e.
In other words, B is equivalent to B',
integrable distributional derivatives which satisfy
q.e.d.
¢ has locally
It is now easy to show that B is invariant under con-
formal mapping. We prove, a little more generally:
LEMMA 3. If w is a C
2
topological mapping and if
¢ has locally integrable distributional derivatives, so does
¢ 0 w , and they are given by
30 Quasiconformal Mappings
(6)
Proof. We note first that
Y) and x1J)
1J x 1J y Y1J
are reciprocal matrices (at correspon ding points). This makes
( x1J) = ! ( 1J
y

Y1J J -1J
x

with J = 1J - 1J ' the Jacobian of w.
x y y x
For any test function how we can put
II 0 + (¢1J 0 w)1J)(h 0 w)dxdy
I I 0 w) 7+ (¢1J 0 w) 1J
jHh
0 w) Jdxdy
I I (¢ Y1J - ¢1JY h d1J
I I ¢(- (hy1J) + 1J) d1J
I I +
I I (¢ ow)(-(h
c
0 w) - (h 0 w) 1J
x
) J dxdy
s J 1J J
= I I (¢ 0 0 w) - (h1J 0 w) 1J) dx dy
= - I I(¢ ow)(h oW)x dxdy
q.e.d.
The last lemma enables us to prove:
B => A .
The General Definition 31
Indeed, in view of the lemma we need only prove that a
rectangle with modulus m is mapped on a quadrilateral with
m' ~ Km, and this proof is exactly the same as in the differen-
tiable case.
We shall now concentrate on proving the converse:
A=>B.
First we prove: if ¢ is q.c. in the geometric sense,
then it is ACL.
Let A (77) be the image area under the mapping ¢ of
the rectangle a ~ x ~ f3, Yo ~ y ~ 77· Because A (77) is in-
creasing the derivative A '(77) exists a.e., and we shall assume
that A '(0) exists.
In the figure, let Qi be rectangles with height 77 and
base b.. Let b.' be the length of the image of b.. We wish
I I I
to show, first of all: If 77 is sufficiently small, then the length
of any curve in Qi' which joins the "vertical" sides is nearly
b. '.
I
To do so we first determine a polygon so that
n
b
' - !..
i 2
32 Quasiconformal Mappings
Next, take TJ so small that the variation of ¢ on vertical seg-
ments is < €I4n. Draw the lines through 'k that correspond
to verticals. Any transversal must intersect all these lines.
This gives a length
n
> b.' - (
1
1
On using the euclidean metric we have now, if ( <
min ~ b
i
',
b. ,2
> 1
4A.
1
b. ,2
1 <
4A
i
b.
K-l.
TJ
b. ,2
::; ~ _ 1 _
b.
1
~ b . < 4K A(TJ) • ( ~ b . )
1 - TJ 1
*
But A (TJ}/TJ'" A '(0) < 00. This shows that ~ b/ ... 0 with
~ bi' and hence ¢ is absolutely continuous.
q.e.d.
Finally, if ¢ is K-q.c. in the geometric sense, it is
easy to prove that the inequality
(k = K-l)
K+l
holds at all points where ¢ is differentiable. This proves
that A => B.
We have now proved the equivalence of the geometric
and analytic definitions, and we have two choices for the analy-
tic definition.
* The proof needs an obvious modification if b
i
' = 00 ••
The General Definition
33
COROLLARY 1. If the topological mapping ep satis-
fies epz = 0 a.e., and if ep is either ACL or has integrable
distributional derivatives, then ep is conformal.
Observe the close relationship with Weyl's lemma.
COROLLARY 2. If ep is q.c. and epz = 0 a.e., then
¢ is conformal.
Finally, we shall prove:
THEOREM 3. Under a q.c. mapping the image area is
an absolutely continuous set function. This means that null sets
are mapped on null sets, and that the image area can always be
represented by
A (E)
I IJ dxdy .
E
Proof. ep = u + iv can be approximated by C
2
func-
tions un + ivn in the sense thet un ~ u, vn ~ v and
Consider rectangles R such that u and v are abso-
lutely continuous on all sides.
If
R
[(u ) (v) - (u·) (v ) 1dx dy
mx ny my nx
f
u dv
m n
JR
34 Quasiconformal Mappings
As m, n .... 00 the double integral tends to ffRJ dx dy. For
m .... 00 the line integral tends to
ju dV
n
j v
n
du
aR aR
and for n .... 00 this tends to
j v du judv
aR aR
(all because u and v, and therefore UV, are absolutely con-
tinuous on aR).
We have proved
j j ] dxdy
R
for these R. It is not precisely trivial, but in any case fairly
easy to prove that the line integral does represent the image
area,'and that proves the theorem.
COROLLARY. ¢z f, 0 a.e.
Otherwise there would exist a set of positive measure
which is mapped on a nullset, and consideration of the inverse
mapping would lead to a contradiction.
Remark. It can now be concluded that Dirichlet integrals
are K-quasi-invariant under arbitrary K-q.c. mappings. It is
possible to prove the same for extremal lengths.
CHAPTER III
EXTREMAL GEOMETRIC PROPERTIES
A. Three Extremal Problems
Let G be a doubly connected region in the finite plane,
C
1
the bounded and C
z
the unbounded component of its comple-
ment. We want to find the largest value of the module M(G)
under one of the following conditions:
I. (Grotzsch) C1 is the unit disk (121 $ 1) and
C
z
contains the point R > 1.
II. (Teichmiiller) C
1
contains 0 and -1; C
z
con-
tains a point at distance P from the origin.
III. (Mori) diam(C
1
n Ilzl $ In ? A; C
z
con-
tains the origin.
We claim that the maximum of M(G) is obtained in the
following symmetric cases.
()
. ~
I.
35
R
Go
36 Quasiconformal Mappings
II.
-1 0
P
III.
A ~ .
Fig.!.
Case l. Let r be the family of closed curves that se-
parate C
1
and C
2
• We know that A(r) = M(G)-I.
Compare r with the family r of closed curves that lie
in the complement of C
1
U IR I, have zero winding number
about R and nonzero winding number about the origin. Evident-
ly, r c r, and hence A(r) ~ A(f). But r is a symmetric
family, and hence A(r) = 71 A(r +) by our symmetry principle.
SimilarlY, if r 0 is the family r in the alleged extremal case*,
then A(r0) = 71A(r0+) .
We show that r + = r 0+. Each curve y in r has
points PI' P2 on (-00, -1) and 0, R) respectively. Say they
divide y into arcs Yl and h so that y= Yl + Y2' Then
- + - + - + (- , - + -)+ H - + - +- b 1
y = y 1 + Y2 = Y1 + Y2 . ere y 1 + Y2 e ongs
to r 0' and we conclude that Y+ fro+. Thus t + C r 0+'
and the opposite inclusion is trivial.
We have proved A(r) ~ A(r) = A(r0)' hence M(G) <
M(Go) .
q.e.d.
* We enlarge r 0 slightly by allowing it to contain curves which
contain segments on the edges of the cut (R, (0). This is, of
course, harmless.
Extremal Geometric Properties
Case II. Let z = f(() map \([ < 1 conformallyonto C
1
U G
with [(0) = O. Koebe's one-quarter theorem gives jf'(0) I ~ 4P
with equality for G = G
1
.Suppose f(a) -1. The distortion
theorem gives
1 = If(a) [ < [at 1['(0)[
- (1- [a\)2
< 4Pla[
- (1- [a\)2
and equality is again attained when G = Gl' In other words
Ial ~ [a 11 where a
1
belongs to the symmetric case. The
module M(G) equals the module between the image of C1 and
the unit circle.
Finally, by inversion and application of Case I, if [al is
given the module is largest for a line segment, and it increases
when la[ decreases. Hence G
1
is extremal.
Case 1I/. Open up the plane by (= VZ. We get a figure
which is symmetric with respect to the origin
with two component images of C
1
and two of C
2
. The compo-
sition laws tells us that M(G) ~ YiM(6') where 6' is the re-
gion between C1- and C/. It is clear that equality holds in
the symmetric situation.
38
Quasiconformal Mappings
By assumption, C
1
contains points zl' z2' with
Iz 11 ::; 1, Iz 2I ::; 1, Iz 1- z 2I ~ '\. Let '1' '2 f C1+ ,
-'I' -,2 f C
1
- be the corresponding points in the '-plane.
The linear transformation
w =
'+ '1 '1 + '2
,- '1
'1 - '2
carries
(-'1' -'2)
into
(0, 1) and
'1
.... 00,
'2
.... W
o
where
w0= _ ( '2 + '1 )2
'2- '1
On setting u = ('2 + '1)/('2 -'1) one has
2('/ + '/)
'/ - '/
we get
lui
__1_
<
£ J4-,\2
lui
,\
lui <
2 + J4-->.!.
,\
Iwol
<
( 2 + ~ Y
One verifies that equality holds for the symmetric case, and by
use of Case II it follows again that M(G) is a maximum for the
case in Figure 1 .
Extremal Geometric Properties
39
In order to confonn with the notations in Kiinzi's book *
the extremal modules will be denoted
I. ~ log <1> (R)
277
II.
III.
~ log 'P(p)
277
~ log xC\.) .
277
There are simple relations between these functions. Ob-
viously, reflection of Go gives a twice as wide ring of type GI'
and one finds
(1)
Another relation is obtained by mapping the outside of the unit
circle on the outside of the segment (-1, 0). This gives
(2)
and together with (1), we find the identity
(3)
The previous computation in Case III gives, for instance,
(4)
*
x C\.) = <1> (V4+2X" + y4 - 2,\ )
,\
Hans P. Kiinzi, Quasikonforme Abbildungen, in Ergebnisse der
Mathematik,Springer Verlag, Berlin, 1960.
40
Quasiconformal Mappings
(1)
B. Elliptic and Modular Functions
The elliptic integral
z
w = f
o
dz
maps the upper half of the normal region G1 on a rectangle
I
I
I
:8
I
I
....._......J'O
with sides
p
a
f
dz
y(z+ Uz(p-z)
0
(2)
00
b
f
dz
y(z+ l)z(z-P)
P
Evidently,
(3)
1
2" log 'I' (p) =
This is an explicit expression, but it is not very convenient for
a study of asymptotic behavior. Anyway, we want to study the
connection with elliptic functions in much greater detail.
We recall that Weierstrass' is defined by
(4)
Extremal Geometric Properties
and satisfies the differential equation
where
41
(6)
(7)
It follows that the e
k
are distinct.
We set T = w
2
/ wI and consider only the halfplane
1m T > O. In that halfplane
e
3
-e
1
p(T) =
e
2
-e
1
is an analytic function .;, 0, 1. It is this function we wish to
study.
aT + b
CT + d
P is invariant under modular transformations
with (: ~ ) == ( ~ ~ ) mod 2, for g;J does not
change and the w
k
/2 change by full periods. The transforma-
tion T' = T+ 1 replaces p by 1/p, and T' = -1/T changes
pinto 1 - p. In other words
(8)
peT + 1)
p( - ~ )
and these relations determine the behavior of p(T) under the
whole modular group.
For purely imaginary T the mapping by g;J is as indi-
cated:
42
Quasiconformal Mappings
~ . _ $7////lII/l/1
o C4)1
-2-
The image is similar to the normal region with P p/l-p,
and we find
(9) rCp) = .1. log qt (l-.-L)
T7 -p
o < p < 1. It is evident from this formula that p varies mono-
tonically from 0 to 1 when T goes from 0 to 00 along the
imaginary axis.
It can be seen by direct computation that p" 1 as soon
as 1m T .... 00, uniformly in the whole halfplane. If we make
use of the relations (8) it is readily seen that the correspondence
between T and p is as follows:
-:1 o :1
Extremal Geometric Properties
....... _ -t... ",
- - I -
I
Fig. 2.
43
We shall let T(p) denote the branch of the inverse func-
tion determined by this choice of the fundamental region. Ob-
serve the symmetries.
(10)
Relation (8) gives
(11)
Also
They yield
T(-I) ;: l+i
2
T(
1-2i y'3 )
+ 1 + i
T (t) ;: T (p) ±. 1
T(l-p) ;: - -+.
npJ
rCp> ;: -T (p)
1 +iy'3
2
It is clear that e
TTiT
is analytic at p ;: 1 with a simple
zero. Therefore one can write
(12)
p _ 1 - a e
iTTT
, (a > 0)
but the determination of the constant requires better knowledge
of the function ..
It will be of some importance to know how Ip -11 varies
when 1m T is kept fixed_ In other words, if we write T;: S + it
44 Quasiconformal Mappings
we would like to know the sign of a/as log Ip -11. This har-
monic function is evidently zero on the lines s = 0, s = 1,
and a look at Figure 2 shows that it is positive on the half-
circle from 0 to 1. This proves
is log Ip-ll > 0
in the right half of the fundamental region, provided that the max-
imum principle is applicable. However, equation (12) shows
that a/as log Ip-ll .... 0 as r .... 00. Use of the relation (8)
shows that the same is true in the other corners, and therefore
the use of the maximum principle offers no difficulty.
We conclude that Ip -11 is smallest on the imaginary
axis and biggest on Re r = ± 1 .
The estimates of the function p (r) obtained by geomet-
ric comparison are not strong enough, but classical explicit de-
velopments are known. For the sake of completeness we shall
derive the most important formula.
The function
~ ( z ) - ~ ( u )
~ ( z ) - ~ ( v )
has zeros at z = ± u + mW
l
+ nW
2
and poles at z = ± v + mw
l
+ nW
2

A function with the same periods, zeros and poles is
00
II
n=-oo
F (z)
. (nw
2
±v-z)
2m W
1 _ e 1
(the product has one factor for each n and each ±). It is easy
to see that the product converges at both ends..
Extremal Geometric Properties 45
217i .!!.±..z
W
1 - e 1
217i y.±...z
W
1 - e 1
F (z)
It will be convenient to use the notation q = e
l7iT
e
l7iW2
/ WI. We separate the factor n = 0 and combine the
factors + n. This gives
- 217i !!.::...Z
W
1 _ e 1
217i Y.-z
W
1 - e 1
00
IT
n=l
+u+ Z
217i - W-
1- q2n e 1
with one factor for each combination of signs.
It is clear that
so(z) - SO(u)
SO(z) - SO( v)
F(z)
F(O)
In order to compute
1 - P =
e
2
- e
3
e
3
-e
1
we have to choose z = w
2
/2, U = (WI +(
2
)/2. v = w/2
which gives
= q,
217iu

e = - q,
Substitution gives
F(z) = 2
1+ q-l
00
IT
1
(1 +
(1+ q2n+l)2 (l+q2n-l)2
46 Quasiconformal Mappings
(14)
and
F CO) =
~
1+ q-l
(1 + q2n+l )2(1 + q2n-l)2
II
2 2
(1 + q2n)4
00
(1 + q2n-l y
1
n
4q
1 1 + q2n
Finally
00
( 1 + q'"
)'
(13)
1 - p = 16q II
1
1 + q2n-l
A similar computation gives
00 (1 _q2n-l )8
p = II
1 1 + q2n-l
«14) is obtained by the relation rCl/p) = rCp) ± 1 .)
We return to formula (9). It gives
(15) log 'PCP) = 17 1m T(I':P) = 17 1m T(1 + ~ )
or
'PCP)
qCp)-l
for
P
p
P+ 1
and, by (13),
8
I-p
1
16q
00 (l+q2n )
P+l
~ 1+ q2n-l
or
'PCP)
00
( 1 + q'" )'
(16) 16 II
P+l
1
1 + q2n-l
(17)
Extremal Geometric Properties
This gives the basic inequality
W(p) S 16(P+ 1)
47
(18)
and
(19)
Because lI> (R )
lI>(R)
-R-
W(R
2
- d
h
it follows further that
4 IT (_I_+_
q
,:..2_
n
__ ) 4
1 1+ q2n-l
lI>(R) < 4R
(20)
For X(A) we obtain
AX (A ) < 4( y'4+2X + y'4=2X)
AX(A) < 16 .
C. Mori's Theorem
Let , = ¢Az) be a K-q.c. mapping of Iz I < 1 onto
I( I < 1, normalized by ¢(O) = O.
Mori's theorem:
(1)
and 16 cannot be replaced by a smaller constant.
Remark. The theorem implies that ¢ satisfies a Holder
condition, and this was known earlier. It follows that ¢ has
a continuous extension to the closed disk Iz lSI. If we ap-
ply the theorem to the inverse mapping we see that the exten-
sion is a homeomorphism.
COROLLARY. Every q.c. mapping of a disk on a disk
can be extended to a homeomorphism of the closed disks.
48 Quasiconformal Mappings
In proving Mori's theorem we shall first assume that ep
has an extension to the closed disk.. It will be quite easy to re-
move this restriction.
For the proof we shall also need this lemma:
LEMMA. If ep (z) with ep (0) = 0 is K-q.c. and
homeomorphic on the boundary, then the extension obtained by
setting ep (1/ z) = 1/(¢;(z)) is a K-q.c. mapping.
Proof. It is clear that the extended mapping is K-q.c.
inside and outside the unit circle. It follows very easily that it
is ACL even on rectangles that overlap the unit circle. The K-
q.c. follows.
Proof of (1). There is nothing to prove if IZ
1
- z2
1
~ 1,
say. We assume that IZ
1
- z2
1
< 1.
Construct an annulus A whose inner circle has the seg-
ment zl' z 2 for diameter and whose outer circle is a concentric
circle of radius Y2.
Case (i). A lies inside the unit circle..
Consider the mapping
w =
Extremal Geometric Properties
We obtain
1
log
1
<
K
log ~ ( 1 1 1 ; - 2 ~ 1 1 ; I ; 1 2 1 )
217
Iz
2
-
z
1
1
217
<
K
log
8
217
1(2-1;1
1
49
This gives
Case (ii). A does not contain the origin. Now the image
of the inner continuum intersects 1(1 < 1 in a set with diam-
eter ~ 1I; 1 - I; 2
1
, and the outer continuum contains the origin.
Hence
and we obtain
To rid ourselves of the hypothesis that ¢ is continuous
on Iz 1= 1, consider the image region ~ r of Iz 1< r
--0
50 Quasiconformal Mappings
and map it conformally on !w! < 1 by t/J
r
with t/J/O) = 0,
t/J/(0) > O. The function t/J/¢ (rz)) is K-q.c. and continuous
on Iz I = 1. Therefore,
1t/J/¢(rz
2
))-t/J/¢(rz
1
))1 < 16!z2- z1[1
I
K
Now we can let r .... 1. It is a simple matter to show that t/Jr
tends to the identity, and we obtain
< 161z -z !l1
K
- 2 1
But this implies continuity on the boundary, and hence the strict
inequality is valid.
To show that 16 is the best constant we consider two
Mori regions (Case III) G(,\) and G('\2). If they are mapped
on annuli we can map them on each other by stretching the radii
in constant ratio
K
log X ('\1)
log X ('\2)
> 1
Because the unit circle is a line of symmetry it is clear that we
obtain a K-q.c. mapping which makes ,\ 1 correspond to '\2 .
We have now
/--......
/ f'
I I \
{ \
·,A1
1
I
\ I I
\ I /
" 1/
....... __ /
/--]'
/ "-
I \
I \
I . A
1
J
\ /
\ /
........... - '"
Extremal Geometric Properties
for small enough A
2
. Thus
A
2
? (A )1/K 16-f
1 16
1
/
K
51
and for large K the constant factor is arbitrarily close to 16.
There are of course strong consequences with regard to
normal and compact families of quasi conformal mappings.
THEOREM 1. The mappings of the
unit disk onto itself. normalized by ¢ (0) = 0, form a sequen-
tiaily compact family with respect to uniform convergence.
Proof. By Mori's theorem the functions ¢ are equicon-
tinuous. It follows by Ascoli's theorem that every infinite se-
quence contains a uniformly convergent subsequence: ¢n -> ¢.
Because Mori's theorem can be applied to the inverse ¢n-1 it
follows that ¢ is schlicht. It is rather obviously K-q.c.
For conformal mappings it is customary to normalize by
¢(O) = 0, I¢ '(0)[ = 1. For q.c. mappings this does not make
sense: we must normalize at two points.
For an arbitrary region n we normalize by ¢ (a1)
b
1
, ¢(a
2
) = b
2
where of course a
1
f. a
2
, b
1
f. b
2
·
THEOREM 2. With this fixed normalization the K-q.c.
mappings in n satisfy a uniform Holder condition
11K
1¢(z1)-¢(z2)1 M[z1 -z21
on every compact set. The family of such mappings is sequenti-
ally compact with respect to uniform convergence on compact sets.
52 Quasiconfotmal Mappings
Ptoof. For any z l' z2 f n there exists a simply con-
nected region G en, not the whole plane, which contains z l'
z2' al' a
2
· If A is a compact set in n, then A x A can be
covered by a finite number of G x G. Hence it is sufficient to
prove the existence of M for G.
We map G and G' = ep (G) conformally on unit disks
as follows:
The compact set may be represented by Is I :::; to < 1. Mori's
theorem gives
(1 - Is I) :::; 16(1 -Iw1)1/K
Hence Is I :::; to implies Iw I :::; Po (depending only on to)
and a similar reasoning gives f3 ~ fJ
o
.
We know that
If we can show that (w) and s (z) satisfy uniform Lipschitz
conditions
we are through. These are familiar consequences of the distor-
tion theorem.
Extremal Geometric Properties
For instance
53
Iz '(sl)1 > Colz'(O)1
la
z
- all < C'lz'(O)1
and hence
C'
The other inequality is easier.
It now follows that every sequence of mappings has a
convergent subsequence. To prove that the limit mapping is
schlicht we must reverse the inequalities. Although G' is now
a variable region this is possible because {3 < {3o' a number
that depends only on a (that is, on G, ai' a
z
).
D. Quadruplets
Let (ai' a
z
' a
3
, a
4
) and (b
l
, b
z
' b
3
, b
4
) be two
ordered quadruples of distinct complex numbers. There exists
a conformal mapping of the whole extended plane which takes
a
k
into b
k
if and only if the cross-ratios are equal. If they
are not equal it is natural to consider the following problem.
Problem 1. For what K does there exist a K-q.c.
mapping which transforms one quadruple into another.
54 Quasiconformal Mappings
It was Teichmiiller who first pointed out that the problem
becomes more natural if one puts topological side conditions on
the mapping.
To illustrate, consider the following figure:
a
4
aa
l.. 1..
a
1
at
There exists a topological mapping of the extended
plane which takes the line 1. into 1.' and £ into £'. Obvi-
ously, as a self-mapping of the sphere punctured at aI' a
2
, a
3
,
a
4
this mapping is quite different from the identity mapping (it
is not homotopic). Therefore, although the cross-ratios are
equal it makes sense to ask whether there exists a K-q.c.
mapping of this topological figure.
We shall have to make this much more precise. First of
all, there exist periods wI' w
2
such that, for T = w21wI,
P(T) =
Evidently, a linear transformation throws (aI' a
2
, a
3
, a
4
) into
(e
1
, e
2
, e
3
, (0), and we may as well assume that this is the
original quadruple.
Let n be the finite '-plane punctured at e
1
, e
2
, e
3
,
and let P be the z-plane with punctures at all points
Extremal Geometric Properties
55
m(w
1
/2) + n ( ~ /2). The projection «oJ defines P as a cover-
ing surface of n. In this situation we know that the fundamen-
tal group F of n has a normal subgroup G which is isomor-
phic to the fundamental group of P, and the group r of cover
transformations is isomorphic to F/G. We know F, G, and r
explicitly. F is a free group with three generators aI' a
2
, a
3
,
representing loops around e
1
, e
2
, e
3
. G is the least normal
subgroup that contains a
I
2
,a
2
2
,a
3
2
, (a
1
a
2
a
3
)2, and r is
the group of all transformations z .... ± z + row 1 + nw
2
. The
translation subgroup r 0 consists of all transformations z ....
z + mW
I
+ nw
2
. It is generated by ~ z = z + wI and A
2
z =
z + w
2
.
* * * Consider now a second quadruple (e
1
' e
2
' e
3
' (0)
and use corresponding notations n*, F *, etc. A topological
* * mapping ¢: n .... n induces an isomorphism of F on F
* which maps G on G . This means that the covering ¢ 0 «oJ:
* * * * P .... n and «oJ: P .... n correspond to the same subgroup
G* of F*, so there is a topological mapping t/J: p .... p*
such that ¢ 0 «oJ = «oJ * 0 t/J. Clearly, Al * = t/J 0 Al 0 t/J-l
and A
2
* = t/J 0 A
2
0 t/J-1 are generators of r 0*. We write
* * * * * *
Al Z = Z + wI' A
2
Z = Z + w
2
and call (wI' w
2
) the
base determined by t/J. Since ¢ does not uniquely determine
t/J, we must check the effect of a change of t/J on the base.
We may replace t/J by T* 0 t/J, T* I: r*, and we find that
the base (wI *, w
2
*) either stays fixed or goes into (-wI *,
-W
2
*). In any case, the ratio r* = w
2
*/ wI * is determined
*
by ¢, and we say that two homeomorphisms of n on n are
56 Quasiconformal Mappings
equivalent if they determine the same T*. By methods which
are beyond our scope here, one can show that two maps of 11
on 11* are equivalent if and only if they are homotopic.
Problem 2. For what K does there exist a K-q.c. map-
ping of 11 on 11* which is equivalent to a given CPo'
We need a preliminary calculation of extremal length. De-
note by (y1 I the family of closed curves in 11 which lift to
arcs in P with endpoints z and z + CU l' In general, (my1 T
n y2 I denotes the family of curves which lift to arcs with end-
points z and z + mcu
1
+ ncu
2
.
LEMMA. The extremal length of (Y11 IS 211m T,
(T cu/cu
1
).
Proof. We may assume that cu
1
sider the segment e in the figure:
'T
/ 7
-{ / ~ - ---/(---,I
T. Con-
Its projection is a curve Y f (Y1
1
. For a given p in 11 set
p = p(&O(z))I&O'(z)l. We obtain
f p dx 2: L (p) ,
e
Extremal Geometric Properties
and we conclude that
A _2_
$. 1m T
57
In the opposite direction, choose p so that p = 1. If
a curve y ( Iy11 is lifted to P its image y leads from a point
z to z + 1. Hence the length of y is ~ 1, and we have
L (p) = 1, A (p) = Y2 1m T •
This completes the proof.
Of course, w
1
may be replaced by any cW
2
+ dW
1
where (c, d) = 1. The result is that
(1)
2
1m (aT+ b )
cT+d
where ( ~ ~ ) is a unimodular matrix of integers.
We are now in a position to solve Problem 2. We lift the
* * mapping ¢: U .... U to t/J: P .... P and choose the base
(w/, w/) determined by t/J. It is clear that the image under
¢ of Iy1/ is the class Iy1*1 corresponding to w 1*. If ¢ is
K-q.c. it follows that
1\1 1m T $. 1m T* $. KIm T .
But !cY2 + dY11 is also mapped on ICY2* + dY1*1 and we
have also
1m aT* + b
c? + d
~ KIm aT+b
CT + d
for all unimodular transformations.
To interpret this result geometrically, let S be any uni-
modular transformation and let U be an auxiliary linear trans-
58 Quasiconformal Mappings
formation which maps the unit disk Iw I < 1 on the upper half-
planewith U(O) = r.
Mark the halfplanes bounded by horizontal lines through
Sr and K Sr.
We know that Sr* does not lie in the shaded part. Mapping
back by U-
1
S-1 we see that U-
1
r* does not lie in a
shaded circle
which is tangent to the unit circle at U-
1
S-1 (00) and whose
radius depends only on K. But the points S-1 (00) are dense
on the real axis, and hence the U-
1
S-1 (00) are dense on the
circle. This shows that r* is restricted to a smaller circle.
An invariant way of expressing this result is to say that the non-
euclidean distance between rand r* is at most equal to that
between ib and iKb, or •
* d[r, r 1 < log K
Extremal Geometric Properties 59
THEOREM 3. There exists a K-q.c. mapping equivalent
* to ¢o if and only if d[T, T 1 ~ log K .
We have not yet proved the existence. But this is im-
mediate, for we need only consider the affine mapping
(T* - 'T) Z + (T - T*) Z
T ."
which is such that ljJ (-z) = -;fJ (z), ljJ (z + 1) = ljJ (z ) + 1,
* .
and ljJ (z +T) = ljJ (z ) + T . As such, ljJ covers a mapping ¢
* of n on n which is equivalent to ¢o' and the dilatation is
precisely e
d
[T, T*1 .
What about Problem I? Here it is assumed that ¢o
maps e
l
, e
2
, e
3
on e
l
*, e
2
*, e
3
* in this order. When does
¢ have the same property? Assume that ¢o is covered by
ljJo: P .... p* which determines the base (wI *, w
2
*) and that
¢ lifts to ljJ determining (CW
2
* + dw
l
*, aw
2
* + bw
l
*). It
is found that ¢ will map e
l
, e
2
, e
3
on e
l
*, e
2
*, e
3
* if and
only if ( ~ ~ ) == (6 ~ ) mod 2 .
The set of linear transformations ~ ~ ~ @ such that
( ~ ~ ) is unimodular and congruent to the identity mod 2 is
the congruence subgroup (of level 2) of the modular group. The
solution to Problem 1 is therefore as follows:
THEOREM 4. There exists a K-q.c. of n on n*
which preserves the order of the points e
i
if and only if the
non-euclidean distance from T to the nearest equivalent point
of T* under the congruence subgroup is < log K.
60 Quasiconformal Mappings
In connection with the lemma it is of some interest to
determine the linear transformation ( ~ ~ ) == ( ~ ~ ) mod 2 for
which the extremal length (1) is a minimum. We claim that this
is the case for the point T that lies in the fundamental region
IT ± Y21 < Y2, IRe TI < 1
Then
Ic T + dl 2: Ic Re T + d I 2: Id I - Ic I
and
ICT + dl
> 7'21 c I - II d I - Y21 c II = Id I or Ic I - Id I
Under the parity conditions Id I 2: 1 and either Id I - Ic I 2: 1
or I c 1 - I d I 2:. 1. It follows that I c T + d 1
2
2:. 1 and thus A
is smallest for the identity transformation,
COROLLARY Let ¢ be any K-q.c, mapping of the
finite plane onto itself with K < V3. Then the vertices of any
equilateral triangle are mapped on the vertices of a triangle with
the same orientation.
Such a mapping can be approximated by piecewise affine
mappings. The triangle corres-
ponds to
l+iy'3
2
-1 + iva
and we know that the corresponding T = 2 ' The near-
t
' 'th I . -1 + i d h I'd
es pomt WI a rea p IS --2- , an t e non-euc 1 ean
Extremal Geometric Properties 61
* distance is log Y3. Hence K < y3 guarantees that Imp >
0, and this means that the orientation is preserved.
The remarkable feature of this corollary is that it re-
quires no normalization. The result is at the same time global
and local.
CHAPTER IV
BOUNDARY CORRESPONDENCE
A. The M-condition
We have shown that a q.c. mapping of a disk on itself
induces a topological mapping of the circumference. How regu-
lar is this mapping? Can it be characterized by some simple
condition? The surprising thing is that it can.
Things become slightly simpler if we map the upper
halfplane on itself and assume that 00 corresponds to 00. The
boundary correspondence is then given by a continuous increas-
ing real-valued function h(x) such that h(-oo) = - 00 and
h(+ 00) = + 00. What conditions must it satisfy?
We suppose first that there exists a K-q.c. mapping ¢
of the upper halfplane on itself with boundary values h (x ) .
It can immediately be extended by reflection to a K-q.c. mapping
of the whole plane, and we are therefore in a position to apply
the results of the precedin'g chapter. Namely, let e
1
< e
3
< e
2
63
64 Quasiconformal Mappings
be real points which are mapped on e
1
', e
3
', e
2
'. If
p =
p' =
and r, r' are the corresponding values on the imaginary axis
we have
K-
1
1m r ~ 1m r' ~ Kim r
We shall use only the very simplest case where e
1
, e ~ .
e
2
, are equidistant points x - t, x, x + t and consequently p =
Yi, which corresponds to r = i. In this case we have that
Equivalently, this means that
or
(1)
1 - p UK) ~ p , ~ p UK)
Actually, what we prefer are bounds for
1 - p'
-p
and from (1) we get
1 - p UK) h (x + t) - h (x) p UK)
pUK) < h(x) -h(x-t} < I-
p
UK)
Even better, let us recall that p (r+ 1) = 1/p (r). For
this reason the lower bound can be written as p (1 +iK) - 1 and
by our product formula (Chapter III, B, (13)) we find
K
00 (1 -2nTTK ).8
p(l+iK) - 1 = 16 e-
TT
II : ; , . + ~ e _
l_e-(2n-l)1TK
Boundary Correspondence
We have thus proved
)
( )
1 h(x + t) - h (x ) ( )
(2 MK- < h{x)-h(x-t) < MK
where
65
(3)
()
1 rrK 00 ( 1- e-{2n-l)rrK ) 8
MK =-e II
16 1 1 + e-
2nrrK
We call (2) an M-condition. Obviously (3) gives the best possi-
ble value, the upper bound
M(K) < 1 ~ err K
THEOREM 1. The boundary values of a K-q.c. mapping
satisfy the M(K) -condition (2).
It is important to study the consequences of an M-condi-
tion
(4)
M-1 < h(x+t)-h(x)
- h{x) - h{x + t)
< M
even quite apart from its importance for q.c. mappings. Let
H (M) denote the family of all such h. Observe that is is in-
variant under linear transformations S: x ... ax + b both of the
dependent and the independent variables. In other words, if
h I: H(M) then Sl 0 h 0 S2 I: H(M). We shall let Ho(M)
denote the subset of functions h that are normalized by h (0)
= 0,
h(1)
1.
For h I: Ho(M) we have immediately
(5)
1
<
h (Y2)
<
M
M+l M+l
66 Quasiconformal Mappings
and by induction this generalizes to
(6)
1
< <
This is true, with the inequality reversed, for negative n as
well. In other words we have for instance
which shows that the h f H0(M) are uniformly bounded on any
compact intervals (apply to h (x) and 1 - h (1-x)).
They are also equicontinuous. Indeed, for any fixed a
the function
h (a+ x) - h(a)
h (a + 1) - h (a )
is normalized. Hence 0 s;. x s;. 1/2
n
gives
h(a+x)-h(a) < (h(a+l)-h(a))(MM)n
+1
which proves the equicontinuity on compact sets. As a conse-
quence:
LEMMA 1. The space Ho(M) is compact (under uni-
form convergence on compact sets).
Indeed, the limit function of a sequence must satisfy (4),
and this immediately makes it strictly increasing.
Actually, the compactness characterizes the Ho(M) .
LEMMA 2. Let H
o
be a set of normalized homeomor-
phisms h which is compact and stable under composition with
linear mappings. Then H
o
C Ho(M) for some M.
Boundary Correspondence
67
Proof. Set a = infh(-l) , (3 = suph(-l) for h f H
o
'
There exists a sequence such that h (-1) .... a, and a subse-
n
quence which converges to a homeomorphism. Therefore a >
-00, and the same reasoning gives (3 < O.
·For any h f H
o
the mapping
k(x)_h(y+tx)-h(}) , t>O
- h(y+ t} - My
is in
H
o
'
Hence
<
h (y- t) - h (y)
< {3
a
h (y+ t) - h (y )
or
1
<
h (y+ t) - h (y)
<
1
a
h(y) - h(y-t} - ~
which is an M-condition.
We shall also need the following more specific informa-
tion:
LEMMA 3. If h f Ho(M) then
1
1
<
f
h (x) dx
<
M
M+1 M+1
0
Proof. Let us set F (x) = sup h (x ), h f H
o
(M) .
This is a curious function that seems very difficult to deter -
mine explicitly. However, some estimates are easy to come by.
We have already proved that F <'/2) ::; M/(M+ 1). Be-
cause
h (tx)
hftT
6B Quasiconformal Mappings
we obtain, for x = Yz,
h (1. )
_2_ < F(Yz)
h (t)
and hence
(7)
Similarly
for t > 0 .
h((l-t)x+ t) - h(t)
1 - h(t)
gives
h(1+t) _ h(t)
2
1 - h(t)
For t < 1 this gives
h(l+t) < F(Yz) + (1 -F(Yz))h(t)
2
and
(8) F(1+t) < F(Yz) + (l-F(Yz))F(t)
2 -
Adding (7) and (8)
(9) F ( ~ ) + F( l ~ t . ) < F(Yz) + F(t) .
Now
1 2
f F(t)dt = Yz f F ( ~ ) d t
o 0
1
= Yz f (F ( ~ ) + F ( ~ + ~ ) ) dt
o
1
< Yz F (Y2) + Yz ! F (t )dt
o
Boundary Correspondence 69
Hence
1
f F(t )dt :s F ('h)
o
and the assertion follows.
The opposite inequality follows on applying the result
to I-h(l-t).
Remark. From
M
< M+l
the weaker inequalities
1
1
:s Jh dt :s
o
are immediate, and since they serve the same purpose, Lemma
3 is a luxury.
B. The Sufficiency of the M-condition.
We shall prove the converse:
THEOREM 2. Every mapping h which satisfies an M-
condition is extendable to a K-q.c. mapping for a K that de-
pends only on M.
1
2y
vex, y)
The proof is by an explicit construction. We shall in-
deed set ¢(x, y) = u(x, y) + iv(x, y) where
y
u(x, y) = ~ y Jh(x+t)dt
-yy
f (h(x+t) -h(x-t))dt
o
(1)
70 Quasiconformal Mappings
It is clear that v (x, y) 2: 0 and tends to 0 for y .... O. More-
over, u(x,O) = hex), as desired.
The formulas may be rewritten
X+Y
u =
1
f
h (t )dt
2y
(1) ,
x ~ · y
x+y x
v
1
( f
h(t)dt -
f
h(t)dt)
2y
x X-Y
and in this fonn it is evident that the partial derivatives exist
and are
U
x
i
y
(h(x+y) -h(x-y))
1 jX+Y hdt+ 2
1
y (h(x+y) + h(x-y))
-2i
X--Y
1
- 2y2
X+Y
J
x
x
hdt - f
X-Y
h dt) + i
y
(h (x +y)
- h(x-y))
The following simplification is possible: if we replace
h (t) by hI(t) = h (at + b), a > 0, the M-condition remains
in force and ¢ (z) is replaced by ¢1(z) = ¢ (az + b). Thus
¢1 (i) = ¢ (ai + b), and since ai + b is arbitrary we need only
study the dilatation at the point i. Also, we are still free to
choose h(O) = 0, h(l) = 1.
The derivatives now become
Boundary Correspondence
71
U =
x
U =
y
'12(I-h (-1)),
1
-'12 f h dt+ '12(1+h(-I)) ,
-1
v =
x
v =
y
'12 (1+ h (-1))
1 0
-'12 f h dt - f h dt
o -1
The (small) dilatation is given by
d =
\
( U - v ) + i (V + U
y
)
x y x _
To simplify we are going to set
1
= 1- f h dt,
o
o
Tff3 = -h(-I) + f hdt
-1
Thus
U
x
'12(1 + (3)
U
y
'!2 - Tf(3)
giving
f3 -h(-I) ,
( > 0).
V
x
'12(1-{3)
v
y

d=
«I-e') + f3(I-Tf)) + - f3 (I+Tf))
«1+e') + + H(I-e'} -f3(I-Tf))
1 + + f32 (1 +Tf2) - 2f3 Tf )
1+ + f32 (I+Tf2) +
72 Quasiconformal Mappings
We have proved the estimates
.:s M,
1
M+ 1 < (
1 < ...JL
M+1 .:s." - M+1
(the last follows by symmetry).
This gives, for instance
1+d
2
< M(M+ 1)
1_d
2
D < 2M(M+1)
< M
M+1 '
As soon as we have this estimate we know that the
Jacobian is positive, from which it follows that the mapping ¢
is locally one to one.
It must further be shown that ¢ (z) .... 00 for z .... 00.
By (1) / we have
x x+y
U
1
(f
hdt +
f
hdt)
2y
x-y x
x+y x
v
2 ~ ( J
hdt -
J
h dt )
x x-y
x+y x
u
2
+ v
2
2; [( f
hdt Y+
(f
h dt)2 ]
x x-y
y
If x :::: 0, u
2
+ v
2
1 (
J
hdt Y > -
2y2
0
0
If x .:s 0,
u
2
+ v2 > 1 (
f
hdt Y
2?
-y
Boundary Correspondence 73
and both tend to 00 for y ~ 00. When y is bounded it is
equally clear that u
2
+ v
2
~ 00 with z.
Now ,= ¢ (z) defines the upper halfplane as a smooth
unlimited covering of itself. By the monodromy theorem it is a
homeomorphism.
1. Remark: Beurling and Ahlfors proved D < M
2
. To
do so they had to introduce an extra parameter in the definition
of ¢.
2. Remark: It may be asked how regular a function
h (t) is that satisfies an M-condition. For a long time it was
believed that the boundary correspondence would always be ab-
solutely continuous. But this is not so, for it is possible to con-
struct functions h that satisfy the M-condition without being
absolutely continuous.
C. Quasi-isometry.
For conformal mappings of a halfplane onto itself non-
euclidean distances are invariant. For q.c.-mappings, are they
quasi-invariant? Of course not. If noneuclidean distances are
multiplied by a bounded factor (in both directions) we shall say
that the mapping is quasi-isometric.
THEOREM 3. The mapping ¢ constructed in B 1S
quasi-isometric. Indeed, it satisfies
with a constant A that depends only on M.
74
(3)
Quasiconformal Mappings
It is sufficient to prove (2) infinitesimally; that is,
A-I ..hk.L < 1M..l < A..hk.L
y - v - y
Again, affine mappings of the z-plane are of no impor-
tance, and it is therefore sufficient to consider the point (0, 1).
We have
1 0
v (i ) Y:z f h dt - f h dt
o -1
and the estimates
M
"2
are immediate (using Lemma 3).
This is to be combined with
and
l<,bzl2 = J/s [(l+e) + f32(1+r?) +
One finds that (2) holds with
A = 4M
2
(M+ 1)
We omit the details.
D. Quasiconformal Reflection.
Consider now a K-q.c. mapping <,b of the whole plane
onto itself. The real axis is mapped on a simple curve L that
goes to 00 in both directions. Is it possible to characterize L
through geometric properties?*
* For instance, we know already that L has zero area.
Boundary Correspondence 75
First some general remarks. Suppose that L divides
the plane into nand n* corresponding to the upper halfplane
H and the lower halfplane H*. Let j denote the reflection
z .... z that interchanges Hand H*. Then ep °j °ep-l is a
sense-reversing K
2
_q.c. mapping which interchanges n, n*
and keeps L pointwise fixed. We say that L admits a K
2
_
q-c- reflection.
Conversely, suppose that L admits a K-q.c. reflection
w. Let [ be a conformal mapping from H to n. Define
H
in H,
in H*.
I
F = [
F=wo[oj
2
It is clear that F is K-q.c. So we see that L admits a re-
flection if and only if it is the image of a line under a q.c.
mapping of the whole plane. Moreover, we are free to choose
this mapping so that it is conformal in one of the half-planes.
We say that the conformal mapping [ admits a K ~ q . c . extension
to the whole plane.
. . * *
We can also consIder the conformal mappmg [ from H
to n*. The mapping j 0/*-1 °w °[ is a quasiconformal map-
ping of H on itself. Its restriction to the x-axis is h (x) =
[*-1 ° [, and we know that it must satisfy an M-condition. Ob-
serve that L determines h uniquely except for linear transfor-
mation (h(x) can be replaced by Ah(ax+b) + B).
(1)
76
Quasiconlormal Mappings
*
On the other hand, suppose that h is given and satis-
fies an M-condition. We know that there exists a q.c. mapping
with these boundary values: we make it a sense-reversing map-
* ping l from H to H . We cannot yet prove it, but there exists
a mapping ¢ of the whole plane upon itself which is conformal
* in H and such that ¢ 0 l 0 j is conformal in H . (This con-
dition determines J1 in the whole plane, and Theorem 3 of Chap-
ter V will assert that we can determine ¢ when J1¢ is given.)
Thus ¢ maps the real axis on a line L which in turn deter-
mines h.
How unique is L? Suppose L
1
and L 2 admit q.c. re-
flections WI and w
2
' and denote the corresponding conformal
* * mappings by 11> 11 , 1
2
, 12 . Assume that they determine the
same h = 1/-
1
0 11 = 1
2
*-1 0 1
2
, The mapping
g
= l/
2
0 1
1
-
1
in °
1
,
* I *-1 ...... *
1
2
OlIn u
1
'
*
is conformal in 0
1
U 01 and continuous on L
1
. Is it confor-
mal in the whole plane? To prove that this is so we show that
g is quasiconformal, for we know that a q.c. mapping which is
conformal a.e. is conformal.
Introduce F
1
and F2 as in (1). Form
G = F
2
-
1
0 I * 0 I *-1 0 F
2 1 1
in H*. It reduces to identity on the real axis, and we set G =
z in H. Then G is q.c. Hence F
2
0 G 0 F 1-1 is quasi-
conformal. It reduces to 1
2
* 0 11 *-1 in °
1
* and to 1
2
/0 1
1
-
1
in °
1
; that is, to g.
*
Boundary Correspondence n
We conclude that g is conformal. Hence 1
2
differs
from 11 only by a linear transformation, and L is essentially
unique.
There are two main problems:
Problem 1. To characterize L by geometric
properties.
* Problem 2. To characterize f (and f).
We shall solve Problem 1. I don't know how to solve Problem 2.
The characterization should be in analytic properties of the in-
variant f"j f'.
We begin by showing that a K-q.c. reflection retains
many characteristics of an ordinary reflection.
·z
~ o . i!
* We shall set ( == (U «() and suppose that (
¢C"Z), (0 = ¢(zo) when Zo is real.
All numerical functions of K. alone, not always the
same, will be denoted by C(K).
LEMMA 1.
< C(K)
78 Quasiconformal Mappings
Proof. We note that
p
Zo - Z
satisfies Ip I == 1. For any such p there is a corresponding
T situated on the lines Re T == ±%, 1m T 2: %.
We conclude that there is aT' corresponding to p
* ('0 - ')/('0 - ,) that is within n.e. distance log K of these
lines. This means that T' lies in a W-shaped region indicated
in our figure.
We note that the points ± 1 where p = "" are shielded
from the W-region. Also, p .... 1 as 1m T .... "" and p .... -1 at
the points ± '12, provided that they are approached within an
angle. Therefore p' is bounded by a constant C(K) and this
proves the lemma.
Remark. It is not necessary to investigate the behavior
of p (T) at T = ± %, for the true region to which T' is restrict-
ed does not reach these points.
Let 8 (,) be the shortest euclidean distance from 'to
L.
LEMMA 2.
C (K )-1
< C(K)
Boundary Correspondence
The proof is trivial.
79
* The regions nand n carry their own noneuclidean
metrics defined by
A
Idz I
y
for = /(z). The reflection w induces a K-q.c. mapping of
* H on H, and we know that it can be changed to a C (K) quasi-
isometric mapping. It follows that we can replace w by a re-
flection w' such that (at points t w
C(K)-l A s A*Idt I < C(K) A
But it is elementary to estimate in terms of
"--Y'
I \
I I
\ I
, ./
..... - ...
o
w
For this purpose map n conformally on Iw I < 1 with
= O. Schwarz's lemma gives
But the noneuclidean line element at the origin is 21 dw I. SO
= < _2_
-
In the other direction Koebe's distortion theorem gives
'I. 1

1
28
80 Quasiconformal Mappings
Combining the results with Lemma 2 we conclude
LEMMA 3. If there exists a K-q.c. reflection across L,
then there is also a C(K )-q. c. reflection which is differentiable
and changes euclidean lengths at most by a factor C(K) .
This is a surprising result, for a priori one would expect
the stretching to satisfy only a Holder con·dition.
I I
L
Now consider three. points on L such that ~ 3 lies be-
tween ~ 1 ' ~ 2 ' Now P = (zl - z3)/(zl - z2) is between 0
and 1 which means that T lies on the imaginary axis. There-
*.
fore T lies in an angle
1 * -1
2 arc tan K- :s arg T :s 17 - 2 arc tan K
It can also be chosen so that IRe T*1
ed to the following region
.,.-- ....
",
o
*
:s 1, so T is restrict-
Again it is obvious that Ip 1 has a maximum C (K) and
we have proved:
Boundary Correspondence 81
THEOREM 4. If ~ 1 ' ~ 3 ' ~ 2 are any three points on L
such that ~ 3 separates ~ 1 ' ~ 2 then
It is more symmetric to write
and in this form the best value of C (K) can be computed. It
corresponds to the point e
ia
and can be computed explicitly.
E. The Reverse Inequality.
We shall prove that the condition in the last theorem is
not only necessary but also sufficient. In other words:
THEOREM 5. A necessary and sufficient condition for
L to admit a q.c. reflection is the existence of a constant C
such that
(1)
<
C
for any three points on L such that ~ 3 is between ~ 1 and ~ 2 '
More precisely, if K is given C depends only on K,
and if C is given there is a K-q.c. reflection where K depends
only on C.
82 Quasiconformal Mappings
Introduce notations by this figure:
lSI
Let A
k
be the extremal distance * from a
k
to 13
k
in
0, and A/ the corresponding distance in 0*. Thus A
1
A
2
* *
= 1, A1 A2 = 1. Through the conformal mapping of 0,
let '1' '3' '2 correspond to x - t, x, x + t. This means that
Al = A
2
= 1. Through the conformal mapping of 0*, '1' '3'
'2 correspond to h (x - t), h (x), h(x+ t). If we can show
* that Al is bounded it follows at once that h satisfies an M-
condition, and hence that a q.c. reflection exists.
(2)
We show first that
c-
2
e-
2TT
::;
Al = 1 implies
I'2 -'11
I '3 -'21
Indeed, it follows from (1) that the points of {32 are at distance
~ C-
1
1'2 - '1 I from '2' while the points of a
2
have dis-
tance ::; C 1'3 - '2 I from '2· If the upper bound in (2) did
not hold, a
2
and /3
2
would be separated by a circular annu-
lus whose radii have the ratio e
2TT
• In such an annulus the ex-
*The extremal distance from a to {3 in E is the extremal length
of the family of arcs in E joining a to /3.
Boundary Correspondence 83
tremal distance between the circles is 1, and the comparison princi-
ple for extremal lengths would yield ,\ 2 > 1, contrary to hypothesis.
This proves the upper bound, and we get the lower bound by inter-
changing and °
Consider points ( a
2
, ( f3
2
° By repeated application
of (1)
and with the help of (2) we conclude that the shortest distance between
a
2
and f3
2
is :::: c4 e-2" - . To simplify notations write
Because of (1), all points on are within distance from .
Let r* be the family of all arcs in n* joining a
2
and f320
Then
where p is any allowable function (recall Chapter I, D). We choose
p = 1 in the disk { - M
1
+ M2}, p = 0 outside that
.) *
dISk. Then Ly(p M
2
for all curves y ( r , whether y stays
within the disk or not. We conclude that
1 ( M
2
)2
" M
1
+M
2
* *
Since \ '\2
is proved.
1, \ * has a finite upper bound, and the theorem
CHAPTER V
THE MAPPING THEOREM
A. Two Integral Operators.
Our aim is to prove the existence of q.c. mappings f with a
given complex dilatation Ill" In other words, we are looking for solu-
tions of the Beltrami equation
(1) f-Z == !J.fz '
where Il is a measurable function and III I ~ k < 1 a.e. The so-
lution f is to be topological, and f-Z' f z shall be locally integrable
distributional derivatives. We recall from Chapter II that they are then
also locally square integrable. It will turn out that they are in fact
locally LP for a p > 2.
The operator P acts on functions h f LP, P > 2, (with
respect to the whole plane) and it is defined by
(2) Ph«,) == _1 f f h(z)( ~ - 1) dxdy
" Z-s z
(All integrals are over the whole plane.)
LEMMA 1. Ph is continuous and satisfies a rmiforrn Holder
condition with exponent 1- 2/p.
85
86 Quasiconforma.l Mappings
The integral (2) is convergent because h £ LP and
t;/'(z(z-()) £ LP, 1/p + 1/q = 1. Indeed, 1 < q < 2, and for
such an exponent Iz(z-()I-
q
converges at 0, (;6 0) and 00.
The Holder inequality yields, for ( ;6 0,
IPh«()1 s 1;1 II
h
lip II z(z:,) II
q
Achange of variable shows that
and we find
(3)
with a constant K
p
that depends only on p. «3) is trivially fulfilled
if (= 0.)
We apply this result to hI(z) = h(z+(1) . Since
we obtain
which is the assertion of the lemma.
The Mapping Theorem
The second operator, T, is initially defined only for functions
h ( C
0
2
(C
2
with compact support), namely as the Cauchy princi-
pal value
(5)
We shall prove:
lim
( .... 0
1
17
LEMMA 2. For h ( C
o
2
, Th exists and is of class Cl.
Furthermore,
(6)
and
(7)
(Ph)z
(Ph)z
f f ITh 1
2
dxdy
h
Th
Proof. We begin by verifying (6) under the weaker assump-
tion h ( COl. This is clearly enough to guarantee that
( P h ) ~ = - ;; f f z ~ z ? ; dxdy
(8)
(Ph), = - ;; f f z ~ z ? ; dxdy .
Let y( be a circle wi th center , and radius (. By use of
Stokes' formula we find
88 Quasiconformal Mappings
1
17
II
h-
z_z, dxdy 1 II h
z
217i i="[ dz dZ
and
= lim
£ .... 0
1
217i
J
h dz
z-'
1
I I z h ~ ,
dxdy = 1 I I dh dz
17
217i ~
lim
[-
1
I
hdi" 1
f f ~ d Z d ~ ]
217i
z-'
+ 217i
£.... 0
I z ~ 1 >£( -,)

= Th(')
We have proved (6).
Observe that (8) can be written in the form
(9)
P(h
z
)
P(h)
h - h (0)
Th - Th(O) .
Under the assumption h £ C
0
2
we may apply (6) to hz, and
by use of the second equation (9) we find
(10)
(Th )z
(Th)z
These relations show that Th £ C1, Ph £ C
2
.
The Mapping Theorem 89
Because h has compact support it is immediate from the
definitions that Ph == 0(1) and Th == O(lz \-2) as z ~ c>o.
We have now sufficient information to justify all steps in the cal-
culation
== ii f f Ph(Ph)zz dzdi"
which proves the isometry.
ii f f(Ph )hz dz di"
f f Ihl
2
dxdy
q.e.d.
The functions of class C
o
2
are dense in L
2
. For this
reason the isometry permits us to extend the operator T to all
of L
2
, by continuity. Unfortunately, we cannot extend P in
the same way, for the integral becomes meaningless when h is
only in L
2
, and even if we use principal values the difficulties
are discouraging.
The key for solving the difficulty lies in a lemma of
Zygmund and Calderon, to the effect that the isometry relation
(7) can be replaced by
(11)
for any p > 1, with the additional information that Cp ~ 1 for
p ~ 2. Naturally, this enables us to extend T to LP, and for
p > 2 the P-transform is well defined.
90 Quasiconformal Mappings
The proof of the Zygmund-Calderon inequality is given in
section D of this chapter. At present we prove:
LEMMA 3. For h
( LP,
P > 2, the relations
(Ph)-Z h
(12)
(Ph)z Th
hold in the distributional sense.
We have to show that
f !(Ph)¢-Z
-! ! ¢h
(13)
! ! (Ph)¢z -! ! ¢Th
for all test functions ¢ ( Co 1. We know that the equations hold
when h ( C
0
2
. Suppose that we approximate h by h
n
( C
0
2
in the LP-sense. The right hand members have the right limits
since 11Th - Th
n
II P $ CP II h - h
n
II p' On the left hand side we
know by Lemma 1 that
IP(h-hn)1 $ K
p
Ilh-hnll
p
I
z
I
1
-
2
/
p
and since ¢ has compact support the result holds.
B. Solution of the Mapping Problem.
We are interested in solving the equation
(1)
where 111111"" $ k < 1. To begin with we treat the case where
11 has compact support so that f will be analytic at "".
The Mapping Theorem 91
We shall use a fixed exponent p > 2 such that kC
p
< 1.
THEOREM 1. If /l has compact support there exists a
unique solution I 01 (1) such that 1(0) = 0 and I
z
- 1 ( LP.
Prool. We begin by proving the uniqueness. This proof
will also suggest the existence proof.
Let I be a solution. Then I
z
= /lIz is of class LP,
and we can form PU
z
). The function
F = I-PU-)
z
satisfies F z = 0 in the distributional sense. Therefore F
is analytic (Weyl's lemma). The condition I
z
- 1 ( LP implies
F' - 1 (LP, and this is possible only for F' = 1, F = z+ a.
The normalization at the origin gives a = 0, and we have
(2)
It follows that
(3)
If g is another solution we get
I
z
- gz = T[/lU
z
- g)l
and by Zygmund-Calderon we would get
and since k CP < 1 we must have Iz = gz a.e. Because of
the Beltrami equation we have also I z = gz· Hence 1-g and
7 - i are both analytic. The difference must be a constant,
and the normalization shows that I = g.
92
(4)
Quasiconformal Mappings
To prove the existence we study the equation
h = T (/l h) + T /l .
The linear operator h ~ T (/l h) on LP has norm < k C < 1.
- P
Therefore the series
(5)
converges in LP. It is obviously a solution of (4).
If h is given by (5) we find that
(6)
is the desired solution of the Beltrami equation. In the first
place /l (h + 1) ( LP (this is where we use the fact that /l
has compact support) so that P [/l (h + 1)1 is well defined and
continuous. Secondly, we obtain
(7)
f-
z
/l(h+ 1)
T[/l(h+ 1)] + 1 h+l
and f z - 1 = h ( LP.
The function f will be called the normal solution of
(1).
We collect some estimates. From (4)
so
(8)
and by (7)
The Mapping Theorem 93
(9)
The Holder condition gives, from (3),
K
(10) 1£((1) - £((2)1 s 1 - ~ C II/lli
p
1(1 -(2
11
-2/
p
+ 1(1 -(21
p
Let v be another Beltrami coefficient, still bounded by
k, and denote the corresponding normal solution by g. We ob-
tain
f -g == T(lIf -vg )
z z r z z
and hence
II
f
z
-g
z
ll
p
s II
T
[v(fz-
g
z)lllp + lIT[(/l-v)fzll1p
< kCpllfz-gzllp + Cpll(/l-v)fzllp
We suppose that v .... /l a.e. (for instance through a se-
quence) and that the supports are uniformly bounded. The fol-
lowing conclusions can be made:
LEMMA 1. IIgz-fz lip .... 0 and g .... f, uniformly on
* compact sets.
More important, we want now to show that f has deriva-
tives if ./l does. For this purpose we need first a slight gener-
alization of Weyl's lemma:
* Since f-g is analytic at 00, the convergence is in fact uniform
in the entire plane.
94 Quasiconformal Mappings
LEMMA 2. If p and q are continuous and have locally
integrable distributional derivatives that satisfy Pz = qz'
then there exists a function f ( C
1
with f
z
p, f
F
= q.
It is sufficient to show that
f p dz + q dz 0
y
for any rectangle y. We use a smoothing operator. For ( > 0
define B/z) = 1/71(2 for [zl ::; (, B/z) = 0 for Izi > (.
The convolutions p * B( * B(, and g * B( * B(, are of
class C
2
and
Therefore
and the result follows on letting ( and (' tend to O.
We apply the lemma to prove:
LEMMA 3. If 11 has a distributional derivative
o
p > 2,
then f ( C
1
, and it is a topological mapping.
Proof. We try to determine ,\ so that the system
(11)
has a solution. The preceding lemma tells us that this will be
The Mapping Theorem
so if
95
(12)
or
,\-
z
(/1'\) z
(log '\)z
The equation
can be solved for q in LP, and we set
o = P (/1 q + /1z) + constant
in such a way that 0 ~ 0 for z ~ 00. Thus 0 is continuous
and
q
Hence ,\ = eO satisfies (12), and (11) can be solved with I
(C
1
. We may of course normalize by 1(0) = 0, and then I
is the normal solution since 0 ~ 0, ,\ ~ 1, I z ~ 1 at 00.
TheJacobian I/zI2-1/Z-12 = Cl-I/11
2
)e
20
is strictly
positive. Hence the mapping is locally one-one, and since f(z)
~ 00 for z ~ 00 it is a homeomorphism.
Remark. Doubts may be raised whether (eO) z = eO0 z
in the distributional sense. They are dissol ved by remarking
that we can approximate 0 by smooth functions on in the sense
that on ~ 0 a.e. and (on) z ~ 0 z in LP (locally). Then for
any test function 1>,
96 Quasicon£urmal Mappings
q.e.d.
Under the hypothesis of Lemma 3 the inverse function
£-1 is again K-q.c. with a complex dilatation ii 1 = f1-1
f
which satisfies lii
1
0 £1 = 1111.
Let us estimate II iiI lip . We have
f fl'j1
1
l
P
d ~ d T J = f fIIlIP(I£zI2 -1£zI2)dxdY
and thus
If we apply the estimate (10) to the inverse function we
find
(13)
The Mapping Theorem
It is now obvious how to prove
97
THEOREM 2 For any Il with compact support and
IIIl to ~ k < 1 the normal solution 01 the Beltrami equation is
a q.c. homeomorphism with III = Il·
We can find a sequence of functions Il
n
£. C
1
with
Il
n
.... Il a.e., Illnl ~ k and Il
n
= 0 outside a fixed circle.
The normal solutions In satisfy (13) with In in the place of
I and Il
n
in the place of Il· Because In.... I and II Il
n
\I p ....
1I1l11
p
' I satisfies (13), and hence I is one-one.
We recall the results from Chapter II. Because I is a
uniform limit of K-q.c. mappings In' it is itself K-q.c. As
such it has locally integrable partial derivatives, and these par-
tial derivatives are also distributional derivatives. We know
further that I
z
f 0 a.e. so that III = Iz/I
z
is defined a.e.
and coincides with Il.
(Incidentally, the fact that I maps null sets on null
sets can be proven more easily than before. Let e be an open
set of finite measure. Then
mes In<e) = f <ll
n
,zI2 -lln,z 1
2
) dx dy
e
e
1 2/ p
(mes e) -
98 Quasiconformal Mappings
and since II f II is bounded we conclude that f is indeed
n,z p
absolutely continuous in the sense of area.)
We shall now get rid of the assumption that /1 has com-
pact support.
THEOREM 3. For any measurable /1 with 11/111
00
< 1
there exists a unique normalized q. c. mapping f /1 with complex
dilatation /1 that leaves 0, 1, 00 fixed.
Proof. 1) If /1 has compact support we need only nor-
malize f.
2) Suppose /1 = 0 in a neighborhood of O. Set
1 z2
ii.(z) = /1 ('z) ~
z
Then ii has compact support. We claim that
Indeed,
f/1(z) =
1
f/1(11z )
f/1(z)
1 1
fii 01z)
Z
f ii(11z )2
7
Z
f!(z)
1 1
f!::. (II z)
Z
fj1(l/z )2
_2
z
z
Remark: The computation is legitimate because f ii is
differentiable a.e.
3) In the general case, set /1 = /11 + /12' /11 == 0
near 00, /12 = 0 near O. We want to find A so that
fA 0 /2 = f/1, fA == f/1 0 (1/12)-1 •
The Mapping Theorem
In Chapter I we showed that this is so for
and this ,\ has compact support. The problem is solved.
99
THEOREM 4. There exists a Il-conformal mapping of the upper
halfplane on itself with 0, 1, 00 as fixed points.
Extend the definition of Il by
;(z) = 1l(Z)
One verifies by the uniqueness that fll(i:) 71l(z). Therefore
the real axis is mapped on itself, and so is the upper halfplane.
SOROLLARY. Every q.c. mapping can be written as a
finite composite of q.c. mappings with dilatation arbitrarily close
to 1.
We may as well assume that f = fll is given as a q.c.
mapping of the whole plane. For any z, divide the noneuclidean
geodesic between 0 and Il (z) in n equal parts, the succes-
sive end points being Ilk (z). Set
Then
(
Ilk+1 - Ilk
1 - Ilk+1 Ilk
f
-1
o k
100 Quasiconformal Mappings
FIl/FIl(I) and FIl(I) ~ 1, the assertion
If fll is K-q.c. it is clear that gk = f
k
+
1
0 f
k
-1 is
K l/n_
q
.
c
., and
C. Dependence on Parameters.
In what follows fll will always denote the solution of
the Beltrami equation with fixpoints at 0, 1, 00. We shall need
the following lemma:
LEMMA. If k = 111111
00
~ 0 then I l f ~ - 111
1
,p ~ 0
for all p.
Note: II II
R
,p means the p-norm over Iz I ~ R:
Suppose first that 11 has compact support, and let F 11
be the normal solution obtained in Theorem 1. We know that
h = F 11 - 1 is obtained from
z
h = T (11 h) + Til
and this implies Ilh lip ~ c 1111 lip ~ O. Here p is arbitrary,
for the condition k Cp < 1 is fulfilled as soon as k is suffi-
ciently small.
Since f 11
follows for all 11 with compact support.
Let us write f(z) = 1/f ( ~ ). Then it is also true that
Ilt
ll
-11/
1
~ 0 when 11 has compact support. It is again suf-
z ,p
ficient to prove the corresponding statement for the normal solu-
tion F 11. One has
The Mapping Theorem 101
_ 1 \P dxdy
I
z
l
4
For Iz I > R, F
The part ff
l
< Izl < R can be written
f f
\
z2(Fz(z) - 1) +
F (z)2 F (z)2
1<lzl<R
and tends to zero because IIF -l11
R
.... O.
z ,P
may be assumed analytic, and F (z) .... z uniformly.
v
In the general situation we can write f = g 0 h where
J.Lh = J.Lt inside the unit disk and h is analytic outside. Then
J.Lh and J.L
g
are bounded by k and have compact support.
Since
v v v
f = (g 0 h)h + (g- 0 h)F = (g 0 h)h
z z z z z z z
we get
Ilfz-1111,p < 1I[(i
z
-1) 0 h]hzlll,p + Ilh
z
-
1
11
1
,p
For the first integral we obtain
f f l(gz-l) 0 hlPlhzl
P
dxdy
Izl<l
< _1_ f f Igz -liP 0 h-
1
I
P
-
2
dx dy
= 1-k2 _
hllzl<lI
1_l
k2
( !!
hllzl <11
. f f 0 .
Izi <I
102 Quasiconformal Mappings
One of these integrals is taken over a region slightly bigger than
the unit disk, but this is of no importance. The lemma is proved.
It is now assumed that /l depends on a real or complex
parameter t and that
/l(z, t) = tv(z) + te(z, t}
where v and E f L"" and II E (z, t )11"" ... 0 for t ... O. We
shall show that f/l = f(z, t) has a t-derivative at t = O.
For any I ~ I < 1 we can write
f ( ~ ) = 1 I f(z )dz
217i z - ~
Iz 1=1
- 1 II fz(z) dxdy
17 z - ~
Izi <1
(the Pompeiu formula). Replace z by 1/z in the line integral.
It becomes
1
217i
I
Izl =1
f(1jz)dz
z(l-z()
v
f(z )-1
1- z ~
dz
The convergence is guaranteed as soon as t is suffi-
...
ciently small to make K < 2. Indeed, the inverse of f satis-
fies a Holder condition with exponent 11K. For small Iz I we
v
have thus, If(z)! > mlzl
K
and hence
I
Izl=o
If(z)I-
1
jzlldzl
11 - z ~ 1
The Mapping Theorem 103
The constants can be recovered from the normalization
f(O) = 0, f(l) = 1. All told we get
(
2z z ) dx dy .
1- 1-z

v
_1 f f
17 f (z )2
Izl < 1
In the first integral, set

- - + - dxdy
z-l z
f- =
Il f z
= Il (fz-1)
+ Il
z
v
and use a corresponding expression for f- with
z
(zl Z)2 Il (II z ) . Because
Hf -111
1
-->0' and

--> v
z ,p
t
we find
= lim
HC) -C
t--> 0
t
_1
f f v(z)(

+

)
dxdy
17
z-I
-z-
Izl <1
ranges over a com-
-
z2
f f v(l/z)
[z[ <1
The convergence is clearly uniform when
)
1-z dx dy .
pact subset of < 1.
Finally, if 1/z is taken as integration variable in the
second integral, it transforms to the same integrand as the first,
104
and we have
where
Quasiconformal Mappings
i(I;) - kJJv(z)R(z, 1;) dxdy
R (z, !;)
1
-
I;
z-1
1;-1
+ -- =
z
By considering f(rz), one verifies easily that this formula is
valid for all 1;, and that the convergence is uniform on compact
sets.
For arbitrary to we assume
in the same sense as above, and we consider
where
and
Hz, t) = i 0 fila
= - 1 II ( v <to) 0 (lllo)-1 R (z, fila (I;))dx dy
IT I-lila1
2
f Z
= - kII v (t 0' z R (lila (z) , fila (1;)) dx dy
The Mapping Theorem
which is the general perturbation formula.
We may sum up our results as follows:
THEOREM S. Suppose
where
v(t), Il(t), e(s, t) (Co, 111l(t)11
00
< 1,
and II e(s, t)1100 -+ 0 as s -+ O.
Then
f Il(t+s) (,) = f J1(t) (,) + ti (" t) + o(s)
uniformly on compact sets, where
105
1(" t)
If v (t) depends continuously on" t (in the L00 sense)
then we can prove, moreover, that (a/at) fez, t) is a continuous
function of t. It is enough to prove this for t = 0, so we have
to show that
By inversion, the integral over the plane can again be written as
the sum of two integrals over Iz I ::;. 1, and they can be treated
in a similar manner. We c o n ~ i d e r only the first part.
The important thing is that the improper integral con-
verges uniformly; that is,
!! Iff(z)12 IR(fIl(z), fll(')1 dxdy < E
\
z-'i <0
zl <1
106 Quasiconformal Mappings
in a uniform manner. Indeed, this integral is comparable to
JJ IR(z, f
J1
«())1 dxdy .
fJ1[z: Iz-<I <81
The region of integration can be replaced by a disk with center
fJ1«() and radius < 28, say. The value of the integral is then
uniformly of order 0 (8).
For the remaining part of the plane (with a fixed 8) the
difference
tends rather trivially to 0, and it is also clear that
since II f J1 - 111 -+ 0 and the other factor is bounded.
p •
D. The Inequality.
We are going to prove that the operator
Th (0 = lim _.1 f f h (z ) dx dy
f -+ 0 17 (z _ ()2
Iz-(I >f
already defined for h f C
0
2
, can be extended to LP, p 2: 2,
so that
(1)
We prove first a one-dimensional analogue, due to Riesz.
The proof is taken from Zygmund, Trigonometric Series, first
edition (Warsaw, 1935).
LEMMA. For
The Mapping Theorem
f ( Co 1 on the real line, set
107

00
pr.v. 1 f J.i.!l dx .
17
-xl
Proof. Set
00
U + IV
f f(x) dx
..
, = + i.", ." > 0
The imaginary part V is the Poisson integral, and it is immedi-
ate that
=
The real part is
00
.,,)
1
f
x-t
f(x) dx
17
(x- + .,,2
-xl
00
1
f
f(g+x) - f(g.·.,x)
x
2
dx
17
x
x2 +.,,2
0
-+
We observe now that
lul
P

IF I
P
=
as .,,-+0.
p(p-l) lulp-2(u 2 + u 2)
x y
p(p-l)lvI
P
-
2
(u} + u/)
p2 IF I
P
-
2
(u} + u/)
It follows that
(IF I
P
- -p- lu I
P
)
p-l
p2(JFIp-2 - luI
P
-
2
)(u} + u/)
> 0 .
108
QuasiconformaI Mappings
We apply Stokes' formula to a large semicircle
"'--,
/ ,
I \
L J
and find easily
00
f
(IF «()I
P
- -p- lu«(W') d( ~ 0 .
p-l
It is easy to see that the integral goes to zero for Tf .... 00.
Hence
00 00
-""'l -""'l
for fixed Tf > O. We observe that
(J IF I
P
q'()2/
P
= IIu
2
+ v211p/2 < Il
u2
1l
p
/2 + Il
v2
1l
p
/2 .
Thus
Lettir.g Tf .... 0 we get the desired inequality.
The Mapping Theorem 109
We continue the proof of the Caldero'n-Zygmund inequal-
ity (1), following the method in Vekua, Generalized Analytic
Functions, Pergamon Press, 1962.
We define the operator
dxdy
zlzl
again as a principal value. On setting z
it can be written
re
i
() we see that
T* f(') =
1
f ( ~ 1
f('+re
i
()) - f('-re
i
())
dr )e-
i
() d()
2" r
o 0
This implies
00
H'+re
i
()) - H'-reie-,
~ l
IIT*fll
<
17
max
1
f
P
2
()
17
r
0
The norm on the right does not change if we replace , by 'e
i
(),
and then the integral becomes Hf()(') where f()(z) = f(ze
i
()).
The norm is of course a 2-dimensional norm. But it is clear
that our estimate of the I-dimensional norm can be used, for we
obtain
A
P
Ilf II P
P () P
110 QuasiconformaI Mappings
so that, finally, II r*f lip :::; ~ Apllf lip'
* We extend r to LP by continuity. The proof of (1)
will now be completed by showing that rf = - r*r*f for
f f C
0
2
. This, of course, allows us to extend r to LP also.
We have ..1- _1_ = __1_ . If f f C 1 we obtain
az Iz I 2zIz I 0
r*f(') - 1 IIHz+')..1- _1_ dxdy
17 az Iz I
1 I I f (z+') _1_ dxdy
17 z Iz I
(2)
1 ..1- 11Hz) dx dy
17 a, I z - ~
~ ~ Ilf(Z) ( l z ~ ' 1 - lh-) dxdy .
For any test function ¢ it follows that
This remains true for f f LP, because the integral on the
right is absolutely convergent (compare the proof for Pf).
This means that (2) is valid in the distributional sense.
The Mapping Theorem
We can now write
T*T*[(w)
111
(One should check the behavior for z = 0 and z = w. The
integral blows up logarithmically, and the boundary integral
over a small circle tends to zero.) We have to compute
...d... lim
az R ~ ""
The differentiation and limit can be interchanged, for it is quite
evident that
a
az
tends to zero, uniformly on compact sets. Moreover, we can re-
place our expression by
112 Quasiconformal Mappings
lim f f 1 1 dTf. J f f dedT/
Jz Iz-w! - Jz
< R. <R
for the difference, which is an integral over two slivers, tends
uniformly to zero.
By an obvious change of variable the first integral be-
comes
f f
<Rllz-wl
R/lz- wi 217
J f f drdO
Jzoo -I=l_-'=r-'-ei-O1--
_% R
. (z-w)lz-w!
and it is obvious that the limit is _ 17
z-w'
cond limit is - 171 z .
We now obtain
Similarly, the se-
T*T*t(w) = [1 f f t(z )(_1 - 1) dXdY]
Jw 17 z-w Z
= Pf(w) = -Tf(w)
Jw
This completes the proof of (1).
The fact that C 1 as p 2 is a consequence of
p
the
The Mapping Theorem
Riesz-Thorin convexity theorem:
The best constant C
p
-+is such that log C
p
function of 1/p .
113
is a convex
Proof. We consider
1\
1
P2 =
1
both
~ 2.
a
1
'
a
2
Assume
II Tfl11/al
<
C1l1fll1/al
II
Tf
Il
1
/
<
C211£111/ a
2
a
2
If a = (1- t )a
1
+ t a
2
the contention is that
II
Tf
11
1
/
a
:s; C
1
1
-
t
C/ II
f
11
1
/
a
<0 < t :s; 1).
Conjugate exponents will correspond to a and a' with
a + a' = 1. Similar meanings for a
1
', a
2
'. We note first that
I!
Tf
11
1
/
a
= s ~ p f Tf· g dxdy
where g ranges over all functions of norm one in L 1/a' Be-
cause simple functions (measurable functions with a finite number
of values) with compact support are dense in every LP it is no
restriction to assume that f and g are such functions.
For such fixed f, g set
1 = / Tf· g dx dy
The idea of the proof is to make f and g depend analytically
on a complex variable ,. 1 will be a particular value of an ana-
lytic function cI> <'), and we shall be able to estimate its modu-
lus by use of the maximum principle.
114 Quasiconformal Mappings
For any complex " define
ill1
F(') If I· a _f
~ ' If I
G(') Igi a -L
Ig I
where a(') = (1-')a
1
+ 'a
2
and a(')' = 1- a('). Ob-
serve that , enters as a parameter: F (,) and G(,) are func-
tions of z. We agree that F (,) = 0, G(,) = °whenever
f=O, g=O. Note that F(t) = f, G(t) = g. Weset
¢(,) =
F(,) is itself a simple function I. F
j
Xj which makes
TF (,) = I. F. TX.. Similarly G(,) = I. G. X. *, say, and we
II}}
find
<I> (,) = '"' F. G./!T X. . X.* dxdy .
L.Jl} 1 }
We see at once that it is an exponential- polynomial: <I> «()
I. aje \' with real \. Therefore <I> (,) is bounded if ~ =
Re ( stays bounded.
Consider now the special cases ~ = °and ~ = 1. For
~ = °we have Re a ( ~ ) = a
1
and hence
Ifl
a
/
a
Igl
a
1'1
a
'
It follows that
IIF (,) 11
1
/ a
1
II G(,) 11
1
/ a '
1
1 .
The Mapping Theorem
For simplicity we may assume that II f 11
1
1a
ly a normalization).
We now get
11S
1 (this is mere-
::; < C
1
.
1 1
A symmetric reasoning shows that
::; C
2
on == 1. We now conclude that
log - log C
1
- log C
2
< 0
on the boundary of the strip 0 ::; ::; 1. Since the lefthand num-
ber is subharmonic the inequality holds in the whole strip, and
for == t we obtain the desired result.
CHAPTER VI
..
TEICHMULLER SPACES
A. Preliminaries
Let S be a Riemann surface whose universal covering
-
S is conformallY isomorphic to the upper haIfplane H. The
-
cover transformations of S over S are represented by linear
transformations of H upon itself which form a discontinuous
subgroup r of the group n of all such transformations. We
can write
S r \ H (orbits)
and the canonical mapping
77: H -+ r\H
is a complex analytic projection of H on S.
Conjugate subgroups represent conformally equivalent
Riemann surfaces. Indeed, if r
o
== Bor B
o
-
1
, B
o
( n then
z -+ BOz maps orbits of r on orbits of r 0 (for B
o
Az
(B
o
ABo -I) Boz ). Therefore B
o
determines a one-one con-
formal mapping of So == r
o
\ H on S.
Conversely, let there be given a topological mapping
g: So -+ S .
-
It can be lifted to a topological mapping g: So -+ S which
117
118
obviously satisfies
QuasiconformaI Mappings
17 0 g g 0 17
0
and we hav.e g
·,I-
g
-l
So --.;:;..g • S
If g is conformal, so is g
1
0
= B
0
1 B
o
-1 .
The classes of conformally equivalent Riemann surfaces
correspond to classes of conjugate discontinuous subgroups of
n (without elliptic fixpoints).
But even if g is not conformal, it is still true that
A = g 0 A
o
0 g-1 ( 1
whenever A
o
( 1
0
, for
17 0 A
17 0 g 0 A
o
0 g-1
g 0 17
0
0 A
o
0 g-1
g 0 17
0
0 g-1
17
In other words, g defines an isomorphism () such that
() - --1
A
O
= g 0 A
O
0 g
It is not quite unique, for we may replace g by Bog 0 B
o
where B ( 1, B
o
( 1
0
, This changes () into ()' with
A
o
()' B 0 i 0 (B
o
A
o
B
o
-
1
) 0 g 0 B-
1
which means that we compose () with inner automorphisms of
1
0
and 1. We say that () and ()' are equivalent isomorphisms.
Teichmiiller 5paces 119
LEMMA. 8
1
and 8
2
determine equivalent isomorphisms
01 and 02 if and only if they are homotopic.
Proof. If 8l' 8
2
are homotopic they can be deformed
into each other via 8(t), say, which depends continuously on t.
We can then find g(t) so that it varies continuously with t,
and since
has values in a discrete set it must actually be constant.
Conversely, suppose 81' 8
2
determine. equivalent 01'
°
2
, By changing i
1
, i
2
we may suppose that 01 = 02' and
hence
Define g(t, z) as the point which divides the noneuclidean
line segment between i
1
(t, z) and i
2
(t, z) in the ratio
t:
(I-t).
Because
i
1
(Az)
Ail(z)
(A
A 0)
i
2
(Az) Ag
2
(z)
0
it follows that
i(t, Az) Ai(t, z)
)
-) 1
Hence 8(t = 17 0 g(t 0170- is a mapping from 50 to 5,
and we have shown that 8
1
and 8
2
are homotopic.
Definition of T(5
0
) (the Teichmiiller space):
Consider all pairs (5, t) where 5 is a Riemann surface
and f is a sense-preserving q.c. mapping of 50 onto 5. We
120 Quasiconformal Mappings
say that (Sl' f
1
) ...... (S2' f
2
) if f
2
0 1
1
-
1
is homotopic to a
conformal mapping of Sl on S2' The equivalence classes are
the points of T (So), and (SO' I) is called the initial point of
T (So),
-
Every f determines a q.c. mapping I of H on itself,
and thereby an isomorphism () of r
o
' Two isomorphisms corre-
spond to the same Teichmiiller point if and only if they differ by
an inner automorphism of n.
The space T (So) has a natural Teichmiiller metric: the
distance of (Sl' 11)' (S2' f
2
) is log K where K is the small-
est maximal dilatation of a mapping homotopic to f
2
0 f
1
-
1
Let us compare T(So) and T(Sl)' Let g be a q.c.
mapping of So on Sl' The mapping
(S, f) .... (S, fog)
induces a mapping of T(Sl) onto T(So)' Indeed, if (S, f) '"
(S', f '), then (S, fog) '" (S', f' 0 g). This mapping is clear-
ly isometric.
B. Beltrami Differentials.
-
A q.c. mapping f: So .... S induces a mapping f of H
on itself which satisfies
- -
(1) f 0 A
o
= A 0 f
for A ':. A
o
() Conversely, if f satisfies (1) it induces a map-
ping f.
* For typographical simplicity both mappings will henceforth be de-
noted by f.
Teichmiiller Spaces
121
From (1) we obtain
(A' 0 fH
z
(A' 0 f)f-
z
and thus the complex dilatation /If satisfies
/If = (/If 0 Ao)Ao 'I A
o
'
or
(2)
A measurable and essentially bounded function /l which
satisfies (2) for all A
o
(:ra is called a Beltrami differential
with respect to roo Another way to express the condition is to
say that
is invariant under r
o
'
Conversely, if /If does satisfy (2), then
/If 0 A = /If
o
and it follows that f 0 A
o
is an analytic function of f, or that
A = f 0 A
o
0 [I
is analytic, and hence a linear transformation.
The linear space of Beltrami differentials will be denoted
by B (r0) and its open unit ball with respect to the L00 norm
is denoted by B
1
<r
o
) .
For every I! ( B 1 (r
o
) we know that there exists a cor-
responding f/l which maps H on itself. We normalize it so
that it leaves 0, 1, 00 fixed. It is then unique.
We set
122 Quasiconiormal Mapping<s
and write r 11 for the corresponding group, 011 for the isomor-
phism. Since Oil represents a point in Teichmiiller space we
have actually defined a mapping
R
1
er
a
) .... T(Sa)
It is clearly continuous from L"" to the Teichmiiller metric.
There is an obvious equivalence relation: Ill"'" 112 if
and 0
112
are equivalent isomorphisms. It is very diffi-
cult to recognize this equivalence by direct comparison of III
and 112. Because we cannot solve the global problem we shall
be content to solve the local problem for infinitesimal deforma-
tions.
Before embarking on this road we shall discuss a differ-
ent approach which has several advantages. The mapping ill
was obtained by extending 11 to the lower halfplane by symme-
try. If instead we extend 11 to be identically zero in the lower
halfplane we get a new mapping that we shall call ill (again
normalized by fixpoints at 0, 1, ",,).
Clearly, ill gives a q.c. mapping of the upper halfplane
and a conformal mapping of the lower halfplane. The real axis
is mapped on a line L that permits a q.c. reflection.
J1 \ \ \
IIIIII \\\\\
L
Teichmiiller Spaces 123
It is again true that
A = f 0 A 0 f -1
Il Il a Il
is conformal, and hence a linear transformation (it is conformal
* in nand n , and it is q.c., hence conformal). We get a new
*
group r
ll
which is discontinuous on nUn . We call it a
Fuchsoid group. We also get a surface r" n = S and a q.c.
Il _ *
mapping So -> S as well as a conformal mapping So -> r \ n
_ Il
where So has the conjugate complex structure of So .
Observe that f Il and fIl are defined for all Il ( L 00
with 111111
00
< 1 (Il ( B1) even when the group r
o
reduces to
the trivial group. Our first result is
LEMMA 1. f Il = fV on the real axis if and only if
* fIl == f
v
on the real axis, and hence in H .
Proof 1). If f = f on the real axis if and only if
Il v
fIl (H) and f
v
(H) are the same and therefore
f 0 (l1l)-1 = f 0 (lV)-1 ,
Il v
for both are normalized conformal mappings of H on the same
region.
2) Suppose fll = fV on the real axis. The mapping
h = (lV)-1 0 fll reduces to the identity on the real axis, so
h can be extended to a q.c. mapping of the whole plane by put-
* ting h (z) = z in H . Consider the q.c. mapping A =
f" 0 h 0 (I )-1. In f (H\ A = f 0 (I )-1 is conformal.
v Il Il v Il
In 9H),
124 Quasiconformal Mappings
is conformal. Thus A is a linear transformation, and the nor-
malization makes it the identity. This means f
v
f
ll
in H*.
q.e.d.
We shall now make the assumption that r
o
is of the
first kind, which means that it is not discontinuous at any point
on the real axis. It is then known that the orbits on the real
axis are dense. In particular the fixpoints are dense. Under
these conditions we can prove
LEMMA 2. III and 112 in B <r
o
) determine the same
Teichmiiller point if and only if f1 = fll2 on the real axis.
Proof. If fill = fll2 on the real axis, then A
Il1
=
A
Il2
on the real axis and hence identically. Therefore (/1 =
(/2, and III and 112 determine the same Teichmiiller point.
Conversely, suppose (/1 and (J1l2 are equivalent iso-
morphisms. This means there is a linear transformation S f n
such that
A
Il2
0 S = S 0 A
Il1
for all A in r
o
.
We conclude that S maps the fixpoints of A
Il1
on the
fixpoints of A
Il2
(attractive on attractive). Since they corre-
spond to each other we see that
S 0 fill = fll2
on the real axis. By the normalization, S must be the identity.
q.e.d.
COROLLARY. If r
o
is of the first kind, III and 112
determine the same Teichmiiller point if and only if
f f In H*.
III 11
2
Teichmiiller Spaces 125
This means we may identify the Teichmiiller space T (So)
*
with the space of conformal mappings of H of the form fll'
11 ( B
1
(r
o
) .
An even better characterization is by consideration ofthe
Schwarzian derivative
f", (f ")2
It ,zl = _11_ _ ~ L
11 f' L. f'
11 11
Let us recall its properties with respect to composition. Con-
sider
and let primes mean differentiation. We get
F '(z)
F"
F'
F ",
IF, zl
For a better formulation, let us denote the Schwarzian by
[£1. The formula reads
[£ 0 g] = ([tJ 0 g )(g ')2 + [g]
There are two special cases. For f = A, a linear transforma-
tion,
[A 0 g] = [g]
and for g = A
[£ 0 A] = ([£] 0 A )(A') 2
126 Quasiconformal Mappings
On setting ¢J1 = [fJ11,
(¢ 0 A) A ' 2 [f 0
J1 J1
[fJ11
We see that ¢J1 satisfies
(¢J1 0 A )(A ')2 = ¢J1
which makes it a quadratic differential. (¢ dz
2
is invariant).
The following theorem is due to Nehari:
*
LEMMA 3. If f is schlicht in the halfplane H, then
\[fll ~ ~ y-2
b b
Proof. Suppose that F (<:) - <: + 1 + 2 + '" is
~ 7
schlicht for 1<:1 > 1. The integral ii II <:1 =r 'PdF measures
the area enclosed by the image of 1<: 1 = r, and is therefore po-
sitive. One computes
F' = 1 - ~ + ...
<:2
'PdF
2 ~ f
1<:1 =r
l- f (, + b
1
+ ... )( 1 - ~ - ... )d<:
21 ;: <:2
= 17 ( r
2
- ~ _ ... )
r 2
It follows that Ib 11 ~ 1. (More economically, Ib 1
12
+ 21 b 212
+ ... + n Ibnl2 + ... ~ 1, which is Bieberbach's FHichensatz.)
Note that
F"= 2b 1
<:3
6b
F'" = _ _ 1
t
+ ...
gives [F 1
Teichmiiller Spaces
6b
1
- - +... and hence
~ 4
127
lim 1,4[Fll $ 6
, -+ 00
Set ~ = Uz = (z-zo)/(z-zo), Zo x
o
+ iyo'
Yo < O. Consider F (,) = f(U-l ,). Then
[E] = ([F] 0 U)U,2
Here
-2iy
U' = __0_
(z-zo)2
2iyo
U ,.., --
z-zo
U,2 '" _ ~ U
4
(z -+ zo) .
4 Yo
On going to the limit we find
and then
q.e.d.
In view of the lemma it is natural to define a norm on
the quadratic differentials by
II¢> II = sup I¢> (z)1 y2
'128 Quasiconformal Mappings
C. ~ is Open.
We have defined a mapping /1 .... ¢/1 from the unit ball
B 1(r) to the space Q(r) of quadratic differentials with finite
norm. The image of B1(r) under this mapping will be denoted
by ~ (r). It is our aim to show that ~ (r) is an open sub-
set of Q(r).
The question makes sense even in the case where r is
the trivial group consisting only of the identity. In this case
the spaces will be denoted by B l' ~ , Q.
THEOREM. ~ is an open subset of Q.
Clearly, ~ consists of the Schwarzians [f] of functions
f which are schlicht holomorphic in the lower halfplane and have
a q.c. extension to the upper halfplane. We know already that
all ¢ in ~ satisfy II¢ II s 312 .
LEMMA 1. Every holomorphic ¢ with II¢ II < Y:z 1S m
~ .
Proof. We choose two linearly independent solutions of
the differential equation
(1) TI" = - Y:z ¢ TI
which we may normalize by T I ~ Tl2 - TI; Till. It is easy to
check that f = TI/
Tl
2 satisfies [[1 = ¢. Observe that the
solutions of (1) have at most simple zeros. Hence f has at most
simple poles, and at other points f' f O.
We want to show that f is schlicht and has a q.c. exten-
sion to the upper halfplane. To construct the extension we con-
sider
TeichmiiIIer Spaces 129
111(Z) + (Z-Z)11{(Z)
112(z) + (z - Z)11;(Z)
*
(z (. H )
We remark first that the numerator and denominator do not vanish
simultaneously (because 11{ 112 - 112 111 = 1). Therefore F is
defined everywhere, but could be 00.
(112 + (Z"- z) 11; )2
Y:z¢(z-z)2
F-
z
Simple computations yield
1
(2)
and thus
= Y:z¢(z-z)
2
The assumption implies \F
z
I ::; k \F
z
I for some k < 1.
Hence F is q.c. but sense-reversing.
z ( H
= { f(z)
F(z)
f (z)
The extension will be defined by
z (. H*
(3)
It must be shown that f is a q.c. 1-1 mapping.
This is easy if ¢ is very regular. Let us assume, spe-
cifically, that ¢ remains analytic on the real axis, and that it
has a zero of at least order 4 at 00. It is immediate that f and
F agree on the real axis, and it is also evident that ; is local-
ly schlicht. The assumption at 00 means that there are solu-
tions of (1) whose power series expansions at 00 begin with 1
and z respectively. We have thus
130
F( z) =
Quasiconformal Mappings
111 a
1
z + b
1
+ 0(\:1)
112 a
2
z + b
2
+ 0(1:1)
1. This gives
a
1
z + ~ + oqz 1-1)
a
2
z + b
2
+ O(lz 1-1)
which is also the limit of f.
Now the schlichtness of f follows by the monodromy
~
theorem. We may of course, compose f with a linear transfor-
mation to make it normalized.
To prove the general case we use approximation. Put
S z = (2nz-i)j(iz+2n). Then S H* CC H* and S z ~ Z
n n n
for n ~ 00. Set cPn(z) = cP (Snz) Sn '(z)2. We have
y
2
IcP
n
(z)1 = IcP(Sn
z
)IIS
n
'(z)1
2
y 2 < IcP(Snz )I(ImS
n
(z))2
and hence IIcP
n
ll ~ IIcP II· Now cP
n
has all the regularity pro-
~ ~ *
perties. We can find mappings f with [f 1 = -!.. in Hand
n n ~ n
uniformly bounded dilatations. By compactness there exists a
subsequence which converges to a solution f of the original
n
problem.
If cP
n
~ cP it is not hard to see that a normalized solu-
tion of 11" = - J.h. cP
n
1l converges to a normalized solution of
11" = - J.h. cP1I' Therefore, if we choose the same normalizations
- - *-
we may conclude that fa = f in H and in H . Hence f can
be extended by continuity to the real axis and is a solution of the
problem with
- 2
J1 = -2cPY
Teichmiiller Spaces
131
But if 1> l Q(r) one verifies that /1 l B (r ), and we
conclude
LEMMA 2. The origin of Q(r) is an interior point of
Suppose now that 1>0 l ~ and [fo1 = 1>0 where f
o
=
* * f/1
0
We assume that f 0 maps H on n, H on n . Then
the boundary curve L of n admits a q.c. reflection A. Ac-
cording to Lemma 3 of Chapter IV D, we can choose A so that
corresponding euclidean lengths have bounded ratio. This means
that A is C(K )-q.c. and
C(K)-1 < IAzl < C(K)
provided that f
o
is K-q.c.
If [f] 1> the composition rule for Schwarzians gives
1> - 1>0 = If, f
o
ll
o
,2
The noneuclidean metric in n* is such that
p(Old(' = ~
-y
Therefore, 111> - 1>0II S l shows that g
l[gl(OI S l p«()2
f 0 f
o
-1 satisfies
For sufficiently small l we have to show that g has a
q.c. extension.
We set t/J [g] and determine normalized solutions
71
1'
132 Quasiconformal Mappings
This time we construct
g«() =
17
1«()/172«(),
( ( 0*
i «()
17
1«(*) + «(-(*)
17
1'«(*)
(( 0 .
172«(*) + «(-(*)172 '«(*)
(Recall that
(*
= A«(L) Computation gives
/1 ;( ()
'h. «(- (*)2 t/J «(*) A,«()
((0
1+'h. «(- (*)2 t/J «(*) A(()
*
< Cp«(*)-1 .
But
IA(I < 1\::1 ~ C(K) and 1(-( I
Hence
1/1; 1~
( .
C(K)
< 1
(4)
1 - ( . C(K)
as soon as ( is small enough.
A
It must again be shown that g is continuous and schlicht.
There is no difficulty if L is analytic and t/J is analytic on L
with a zero of order four at 00.
The general case again requires an approximation argu-
ment. Let f = f 0 S where S is as in the proof of
nOn' n
Lemma 1, and let L
n
be the image of the real axis under In'
L
n
is an analytic curve that admits a K-q.c. reflection, and t/J
is analytic on Ln'
The Poincare density p of 0 * = f (H*) is > p,
n n n -
so that It/J I ~ (p2 implies It/J I ~ (pn
2
. Therefore we can
construct a sequence of normalized q.c. mappings i such that
A *
[gn1 = t/J in On and /1; satisfies the inequality (4). A
A n A
subsequence of the gn tends to a q.c. limit g which is equal
Teichmiiller Spaces 133
to g in n*. This proves Theorem 1. Since /l; satisfies (4)
we conclude:
COROLLARY. For every sequence of ¢n ( ~ converg-
¢o = [f/lo 1 ( ~ , there exist /l n .... /lo such that
= ¢n'
The proof is by writing ¢ = ~ 0 f 01 .
We come now to the most delicate part:
THEOREM 2. ~ (r) is an open subset of Q(r) .
.
Remark: This was first proved by Bers. The idea of the
proof that follows is due to Clifford Earle.
Given any /lo ( B, we construct a mapping
f3
0
: ~ .... ~
as follows: Given ¢ ( ~ there exists a /l ( B1 such that
¢ = ¢/l' With this /l we determine A by
(5) fA = f/l 0 ([ /l0)-1
and hence ¢A = ¢A .
1
It is evident that f3
0
is 1-1, and it carries
zero. Moreover, f3
0
is continuous. For if ¢n .... ¢
have just proved the existence of /In .... /l such that
The corresponding ¢A converge to ¢A'
n
and set f3
0
(¢) = ¢A' It is unique, for if ¢ = ¢ , then f /l
/l /l /l1
has the same boundary values as fl. Hence fA has the same
A
boundary values as f 1
134 Quasiconformal Mappings
LEMMA. (Earle). <P
1l
( Q(f') if and only if for every
A (f' there exists a linear transformation B such that
fll 0 A 0 ([11)-1 = B
on R.
Proof. <P
1l
( Q([') is equivalent to [ill 0 A] = [ill]'
and this is true if and only if f11 0 A 0 f 11-
1
= t, a linear
* transformation, in nil .
1) If B exists, then
C = f 0 A 0 f -1 = f 0 (f 11)-1 0 B 0 f 11 0 ([ )-1
11 11 11 11*
on f 11 (R). The first expression is holomorphic in nil' the
second in nil' Hence C is a linear transformation. *
2) If C exists, then
B = f 11 0 A 0 ([ 11)-1 = f 11 0 f -1 0 C 0 f 0 (f 11)-1
11 11
on R. But the last expression is a conformal mapping of H on
itself, hence a linear transformation. The lemma is proved.
Assume now that 110 ( B1(f'). We find that f3
0
maps
~ (f') on ~ (f'1l
0
) , for
fA 0 A
llo
0 (fA)-1 = fll 0 A 0 ([11)-1 = All
From the lemma we infer that Q(f') n ~ is mapped on
Q(f'1l
0
) n ~ . The origin has a neighborhood N in Q(f'1l
0
)
which is contained in ~ (f'1l
0
). Write N = Q(f'1l
0
) n No =
Q(f'1l
0
) n ~ n Nowhere No is a neighborhood in ~ . Then
f3-
1
(N) = Q([') n ~ n f3-
1
(N
o
) = Q(f') n f3-1(N
o
)
which is a neighborhood of <P in Q(f'). From N C ~ (f'1l
0
)
11
0
* Because it is quasiconformal and conformal a.e.
Teichmiiller Spaces
we get f3 -1 (N) c (r) and this proves that ¢ has a
/1
0
neighborhood in Q(r) which is contained in (r) .
135
q.e.d.
Conclusion:
If the group r of cover transformations of H over So
is of the first kind, the Teichmuller space T (S0) is identified
with (r), an open subset of Q(r). The Teichmiiller metric
defines the same topology as the norm in Q(r) .
Consider the case where So is a compact Riemann sur-
face of genus g > 1. Then Q(r) has complex dimension
3g- 3. Choose a basis ¢t' ... , Every ¢ f Q(r) is a
linear combination
¢ = 7
1
¢1 + ... + 7
3
_-
3

with complex coefficients. We find that (r) can be identi-
fied with a bounded open set in
We can show, moreover, that the parametrization by 7 =
(7
1
, ... , defines the Riemann surfaces of genus g as a
holomorphic family of Riemann surfaces.
According to Kodaira and Spencer a holomorphic family
may be described as follows:
There is given an (n+ I)-dimensional complex manifold
V and a holomorphic mapping 17: V .... M on an n-dimensional
complex manifold M. Each fiber 17-
1
(,), 7 f M is a Riemann
surface.
The complex structures of V and M are related in the
following way: There is an open covering lual of V so that
136 Quasiconiormal Mappings
for each a there is given a holomorphic homeomorphism h
a
:
Ua ~ C x M connected by ¢af3 = h
a
0 hf3-1 (for intersect-
int Ua' U(3)' For any T ( M the restriction of ¢af3 to
hf3 (Ua n Uf3 n 7T-1 (r)) shall be complex analytic (in fact, these
functions determine the complex structure of 7T-1(T).)
In our case, M will be the set ~ (r). Every T ( ~ (r)
* determines a ¢ = [i,), and ill is uniquely defined in H .
*
Hence Oil' Oil' and the group r Il are determined by T. To
emphasize the dependence on T we change the notations to ¢T'
* 0T' r
T
, etc. Since iT is determined in H by the differential
equation [iT1 = ¢T' we know that iT depends holomorphically
on the parameter T. For A ( r, the corresponding AT ( r T
*
is determined by the condition iT 0 A = AT 0 iT in H . This
means that AT depends holomorphically on T, a fact that will
be important.
The Riemann surfaces in our family will be S (T) = 0/rT
and V will be their union. Thus the points of V are orbits
r l with <: ( 0T and T ( ~ (r). The projection 7T: V ~ M is
defined so that 7T-
1
(r) = S(r).
Consider a point r T <:0 in V. We pick a fixed <:0 from
o
the orbit and d_etermine an open neighborhood N (<:0) such that
N is compact, contained in 0T • and does not meet its images
o
under r T' The neighborhood N (s <:0' TO) of r T <:0 will con-
o 0
sist of all r T<: such that II ¢T - ¢T II < E and <: ( N (<:0) .
_ 0
Here E shall be so small that N does not meet its images under
r T' This is possible because AT is near AT0 when ¢T is
Teichmiiller Spaces 137
near ¢T' As a consequence, there is only one (( N «(0) n
o
r T( and the mapping h: r l.... «(, T) is well-defined in
N (E, (0' TO) ,
The neighborhoods N (E, (0' TO) shall be a base for the
topology of V. We assert, in addition, that the parameter map-
pings h: r l.... «(, T) defined on the basic neighborhoods make
17: V .... M into a holomorphic family, Indeed, if two basic neigh-
borhoods U0 and U1 intersect, then in U0 n Ul' hO(r
T
() ==
«(, r) and h1(r
T
() == (AT (, T) for some AT ( r
T
• Hence the
mapping h
1
0 h
o
-1 is given by «(, r) .... (A
T
(, T), which we
know is holomorphic in both T and (. It is evident that the pa-
rameter mappings define a complex structure on V which agrees
with the conformal structure of the surfaces S (T) and makes the
mapping 17: V .... M holomorphic.
D. The Infinitesimal Approach.
We continue with a direct investigation of f 11, All that
does not make use of the mappings f
ll
,
For any function F(p.) and any y ( LC>O we set
lim
t .... 0
when the limit exists, and we omit the argument 11 when the de-
rivative is taken for 11 = O. We are assuming that t is real.
We have already derived the representation
i[vl«() == - ~ f f v(z)R(z, ()dxdy
138 Quasiconformal Mappings
where
R (z, ,) = 1 _ U - ~
z-' z z-1
We apply the formula to the symmetric case: v(Z") = i/ (z) ,
and we write more explicitly
(1) i[v1(') = - ; ffv(z)R(z, ')dxdy
H
-; f fi/(z)R(Z", ,) dxdy
H.
It is evident that f[v] is linear in the real sense, but
not in the complex sense. To obtain a complex linear functional
we form
(2)
and we find
(3) lI>[v](') = - ~ ffi/(z)R(Z", ')dxdy
H
It is holomorphic for 'l H. Its third derivative is
17
ff
i/(z) dxdy
(Z" _,)4
H
If v l B (r) one verifies that ¢ is a quadratic differential
(¢ l Q(r)).
(4)
From
fll(Az) = All fll(z)
with Il
(5)
t v we obtain after differentiation
i[v] 0 A = A[v] + A'f[v]
Teichmiiller Spaces 139
and hence A-
z
(the existence of A requires some, but not much, proof). Since
i [v lz = v differentiation of (5) gives
(v 0 A) A' = A- + A'v
z
o because v ( B (i). We conclude that the
. .
A are analytic functions. Moreover, AIA' is real on the real
axis and can hence be extended by symmetry to the whole plane.
The explicit formula for i[v 1 shows that it is o(lz 1
2
) at 00.
The apparent singularity of AIA' at A-loo is therefore re-
movable, and there is at most a double pole at 00. We conclude
that
(6)
where PAis a second degree polynomial. From
(A
l
A
2
)
(A
l
A
2
)'
we deduce that
(AI 0 A
2
) + (AI' 0 A
2
) A
2
(AI' 0 A
2
) A
2
'
(7)
A'
2
We shall say that v is trivial, and we write v ( N (i) ,
if all A[v] = O. There is a whole slew of equivalent condi-
tions:
LEMMA 1. The following conditions are all equivalent.
a) A[v] = 0 for all A ( i, c) i[v] = 0 on R,
b) P
A
= 0
for all A ( i, d) cI> [v] == 0,
e) ¢[v1 == 0,
f)
II
v¢ dxdy = 0
for all ¢ ( Q(['). *
i\H
140 Quasiconformal Mappings
Proof. a) < >b) by (6). c) > a) by (5). Con-
versely, if A '" 0 then since ;(0) = 0 it follows that ;(A 0)
= 0 for all A. .These points are dense on R (we are assuming
that the group is of the first kind) and hence, by continuity, f = 0
on R.
The definition of cI>, together with (5) and (6), gives
cI> ;, A - cI> = PA [v] + i PA [i v J
If cI> =0 0, it follows that P
A
[v] + i PA[i v] '" 0 on R. But
both polynomials are real on R, hence identically zero, so
d ) > b). Conversely, if f [v] = 0 on R, then cI> is pure-
ly imaginary on R and can be extended to be analytic in the
whole plane. Since cI> '" 0 (j z 1
2
) it is a first degree polynomial.
But it vanishes at 0 and 1. Hence cI> =0 0 and c) > d).
Since cI> '" '" ¢, d) > e). Conversely, if ¢ = 0,
cI> is a polynomial and we reason as above to conclude cI> =0 o.
Condition f) is the most important one. We prove it
only for the case that r\H is compact, and we represent it as
a compact fundamental polygon S with matched sides. If A =
.. .
0, then f 0 A = A' f. We know further that f z = v. By
Stokes' formula we find
f fv¢ dxdy
s
-ii f f;z ¢ dz dz
s
2 ~ f f ¢ dz
as
* For r\H non-compact, the condition shall hold for all ¢ f Q(r)
such that ffr\H
1
¢ I dxdy < 00.
Teichmiiller Spaces 141
But the condition means that
f ¢ dz =
f/dz
f
dz .
is invariant. Hence
is invariant, and the boundary integral vanishes.
To prove the opposite, consider a real so that (3)
takes the form
- !!v (z ) R (z, dx dy
H
We introduce a Poincare O-series
From the fact that I H IR (z, Idx dy < 00 it is quite easy to
deduce that the series converges. We obtain now
= - L: !f v(z)R(z,
A(S)
- L: J!v(Az)R(Az, '(z)1
2
dxdy
S
L: Jf v(z)R(Az,
= - ff v !/J dx dy
which is zero if f) holds. Hence <I> is identically zero. The
lemma is proved.
We need the following lemma.
142
Then
(8)
Quasiconformal Mappings
LEMMA 2. Suppose ¢ is analytic in Hand
sup I¢ ly2 < 00 •
12 II p(z)y2 dxdy .
17 (z
H
For the proof we note that
_ V. (z-z)2 = _ %[_1__ 2(z-C) +
(z (:z
_ 1
4
[ __1_ + z-( _ (z-<:yl
d:Z
Assume first that ¢ is still analytic on R. Then integration
by parts gives
12 I I ¢ i dx dy = _ L
17 (:z _ 217i
H R
It is easy to complete the proof by applying this formula to
¢(z+iE), E > 0, for the hypothesis guarantees absolute
convergence.
Compare formula (8) with (4). We have defined an anti-
linear mapping
A: v ¢[v1
from B (r) to Q(r). On the other hand we may define
A*: ¢ -¢ i
which is a mapping from Q(r) to B (r). Lemma 2 tells us
*
that AA is the identity.
By Lemma 1, e), v f N(r) if and only if Av = o.
From AA* = I we conclude that
TeichmrlIIer Spaces 143
* v - A Av £ N (r) .
In other words, v is equivalent modN(r) to -¢;[vli. Of
* course this is the only A ¢ which is equivalent to v, for if
- ¢ y2 £ N (r) then fJhH 1¢ (z ) 1
2
y2 dx dy = 0, hence ¢ =
o.
We conclude:
A establishes an isomorphism of B <r)/N (r) on Q(r) .
The inverse isomorphism of Q(i) on B(r)/N(r) is given by
A*.
A compact surface S, of genus g > Q, determines a
group i generated by linear transformations AI' ... , A
2el
which satisfy
(9) ~ A
2
A
l
-
l
A
2
-1 ... A
2el
-
l
A
2el
A
2el
-
2
-1 A
2el
-
l
= I.
We say that lA
l
, ..• , A
2el
l is a canonical set. If it belongs to
a surface, Al and A
2
have four distinct fixpoints. By passing
to a conjugate subgroup we can make Al have fixpoints at 0,
00 and A
2
to have a fixpoint at 1. When this is so we say that
the generating system is normalized.
Set
V set of normalized canonical systems,
T set of normalized canonical systems
that come from a surface of genus g.
It can be shown that V is a real analytic manifold of
dimension 6g - 6. We are going to prove that T is an open sub-
set of V, and that it carries a natural complex structure.
144 Quasiconformal Mappings
o all j
Let S determine r with normalized generators (A)
(A
1
, ... , We choose a basis v
1
' ... , of
B (r)lN (r) and set
vCT) = T
1
V
1
+ .,. + v
3
and T
k
= t
k
+ itk'. For small T we obtain a system (A)V(T)
( T. The points of the manifold V near (A) can be expressed
by local parameters u
1
' ... , and the mapping T (A )V(T)
takes the form
u
j
= h
j
(tl' t
1
', ... , t;g-3) .
We have to show
1) The h. are continuously differentiable,
}
2) The Jacobian is f. 0 at T = O.
1) has already been proved. The coefficients of the A
k
are differentiable functions of the uk' Therefore, if all Uk[v]
are 0, so are all A k [V] and hence all A[v]. Observe that
ah. ah
j
} = U. [Vk] - = u}.[iv
k
]
at
k
} at
k
'
If the Jacobian at the origin were zero there would exist real
numbers (k' TJk such that
ah. ah
j
(k + TJk at
k
'
This would mean
and hence
Hence «(k + iTJk) v
k
( N (r), and this is possible only if
Teichmiiller Spaces 145
A/1 .
all k' TJ k = o. We have proved that the Jacobian does not
vanish.
The proof shows that T is an open subset of V. It al-
so follows that if 11/111 is small enough, then there exist unique
complex numbers T
1
(/1), , T
3
g-
3
(/1) such that
AT1 (/1)v
1
+ + T
3
/1-3 (/1)v
3
/1-3
Set /1 = t
p
and differentiate with respect to t for t = o.
We obtain
+ ... + A[p]
This implies
+ ... + - P £. N(r) .
Replace p by i p. We can then eliminate the term p to obtain

L: + i £. N(r)
1
from which it follows that
k[i P1 = i k[P ]
In other words, the are complex linear functionals,
which means that the Tk(/1) are differentiable in the complex
sense at /1 = o.
From this we can prove that the coordinate mappings
(A/1) .... (T
1
(/1), ... , T
3
g-
3
(/1)) define a complex structure on T.
Indeed, we must show that on overlapping neighborhoods the
coordinates are analytic functions, and it suffices to show this
at the origin of a given coordinate system. Take /1 (T) =
l TjV
j
and /10 = /1(T
O
) near zero in B(r). Define A(r) in
146 Quasiconformal Mappings
B <r 11
0
) by f 11 (r) = f'\ (t)
C in Chapter I, we have
11 .
o f o. By the formulas of section
so that A depends analytically on T.
Now choose a basis A
1
, ... , for B(r
llo
)/ N(r
llo
) . .;
Near (A 11
0
) we have coordinate functions a
1
(A), ... ,
This means that for T near TO we can write uniquely
(AIl(r)) = (AIlO)A(r)) = (AIlO/a/A(r)) A)
Because a/A) is complex analytic at A ". 0, we conclude )
that ai is a complex analytic function of T at TO' This is
exactly what we required.
.J
VAN NOSTRAND MATHEMATICAL STUDIES
VAN NOSTRAND MATHEMATICAL STUDIES are paperbacks
focusing on the living and growing aspects of mathematics. They
are not reprints, but original publications. They are intended to
provide a setting for experimental, heuristic and informal writing
in mathematics that may be research or may be exposition.
Under the editorship of Paul R. Halmos and Frederick W. Gehring,
lecture notes, trial manuscripts, and other informal mathematical
studies will be published in inexpensive, paperback format.
PAUL R. HALMOS graduated from the University of Illinois in
1934. He took his master's degree in 1935 and his doctorate in 1938,
and stayed on for another year to teach at the University. He then
spent three years at the Institute for Advanced Study, two of them
as assistant to John von Neumann. He has returned to the Institute
on two subsequent occasions, in 1947-48 as a Guggenheim fellow
and in 1957-58 on sabbatical leave. After teaching at Syracuse
University for three years he joined the faculty of the University
of Chicago in 1946; he is now Professor of Mathematics at the
University of Michigan.
FREDERICK W. GEHRING attended The University of Michi-
gan from 1943 to 1949, and received his doctorate from the Univer-
sity of Cambridge in 1952 after three years' study in England. He
spent the next three years as a Benjamin Pierce Instructor at
Harvard University, and returned to The University of Michigan
in 1955. From 1958-59 Professor Gehring was a Guggenheim and
Fulbright fellow at the University of Helsinki, Finland, and the
following year was an N.S.F. fellow at the Eidgenossische Tech-
nische Hochshule in Zurich, Switzerland. He is now Professor of
Mathematics at the University of Michigan.
A VAN NOSTRAND MATHEMATICAL STUDY
under the general editorship of
PAUL R. HALMOS and FREDERICK W. GEHRING
The University of Michigan

About this book:
In essentially self-contained exposition, this book presents the first
treatment in English of a subject rapidly growing in its interest and
importance to mathematicians: German works in the same field
overlap only partly. Intended for advanced graduate study, this
volume derives from lectures delivered by Professor Ahlfors at
Harvard University during the Spring Term of 1964.

The Author:
LARS V. AHLFORS is William Caspan Graustein Professor of
Mathematics at Harvard University. As well as serving as chairman
of Harvard's department of mathematics from 1948 to 1950, the
author has held professorships at the University of Helsingfors
and the University of Zurich. A renowned mathematician, Pro-
fessor Ahlfors was awarded the Field's medal for mathematical
research at the 1936 International Congress of Mathematicians; he
is a member of numerous distinguished societies, including the
National Academy of Science. Professor Ahlfors has contributed
much to the development of conformal mapping,. Riemann sur-
faces, and other branches of the theory of functions of a complex
variable: he is the author of Complex Analysis and, with L. Sario, of
Riemann Surfaces.
D. VAN NOSTRAND COMPANY, INC.
120 Alexander Street Princeton, New Jersey 08540

VAN NOSTRAND MATHEMATICAL STUDIES

Editors Paul R. Halmos, The University of Michigan Frederick W. Gehring, The University of Michigan

Paul R. Halmos - LECTURES ON BOOLEAN ALGEBRAS Shmuel Agmon-LECTURES ON ELLIPTIC BOUNDARY VALUE PROBLEMS Noel J. Hicks-NOTES ON DIFFERENTIAL GEOMETRY Leopoldo Nachbin-TOPOLOGY AND ORDER Sterling K. Berberian-NOTES ON SPECTRAL THEORY Roger C. Lyndon-NOTES ON LOGIC Robert R. Phelps-LECTURES ON CHOQUET'S THEOREM L. E. Sigler-EXERCISES IN SET THEORY George W. Mackey-LECTURES ON THE THEORY OF FUNCTIONS OF A COMPLEX VARIABLE Lars V. Ahlfors-LECTURES ON QUASICONFORMAL MAPPINGS

Additional titles will be listed and announced as issued.

Lectures on

QUASICONFORMAL MAPPINGS

by
LARS

v: AHLFORS

Harvard University

Manuscript prepared with the assistance of

CLIFFORD

J. EARLE Jr.

Cornell University

D. VAN NOSTRAND COMPANY, INC.
PRINCETON, NEW JERSEY TORONTO NEW YORK LONDON

England D. Kensington High Street. BY D. VAN NOSTRAND COMPANY (Canada). Toronto 16.D. 120 Alexander St. INC. VAN NOSTRAND COMPANY. New York 18. London. Princeton. VAN NOSTRAND COMPANY (CANADA). in whole or in part (except for brief quotation in critical articles or reviews). New York D.. Canada COPYRIGHT © 1966. 358. PRINTED IN THE UNITED STATES OF AMERICA .14. INC. VAN NOSTRAND COMPANY. New Jersey (Principal office) 24 West 40 Street. LTD. No reproduction in any form of this book. LTD. Published simultaneously in Canada by D. may be made without written authorization from the publishers. VAN NOSTRAND COMPANY. W. 25 Hollinger Road. LTD.

checked computations and supplied many of the bridges that connect one fragment of thought with the next. Clifford Earle from rough long hand notes of the author. . He has contributed many essential corrections. arid financed by Air Force Grant AFOSR- 393-63. to say the least. Without his devoted help the manuscript would never have attained readable form. In keeping with the informal character of this little volume there is no index and the references are very spotty. The typing was excellently done by Mrs. Caroline W. Browne in Princeton. The experts will know that the history of the subject is one of slow evolution in which the authorship of ideas cannot always be pinpointed.ACKNOWLEDGMENTS The manuscript was prepared by Dr.

. ... . B. Quadruplets. . . . . . . . The Problem and Definition of Grotzsch B. . . .. . . . Quasi-isometry . .. . . C. . . . . . . map is Conformal. . .. . . .. D. . . . . . . .CONTENTS CHAPTER I: DIFFERENTIABLE QUASICONFORMAL MAPPINGS Introduction . . . . D. .. . Extremal Length E. . . . . A 1-q-c. " . E. . . . . .. . Dirichlet Integrals. Three Extremal Problems. .. . . .. . . . .. D. . . ..... . . . .. . .. . F. . . Mori's Theorem. . . . . .. . . .. Quasiconformal Reflection. . 1 2 6 8 10 16 17 21 A. . . . Solution of Grotzsch's Problem. . The Reverse Inequality... . . . . . . . . . . . . . . . .. . . . . 40 47 53 63 B. . . .. .. Composed Mappings. . . . . . . . .. . . . . . . C. . . . . . . . . . . . A Symmetry Principle . . . . . .. . . CHAPTER IV: BOUNDARY CORRESPONDENCE A. A. . . . . .. The Geometric Approach . . . . . .. B. . The Sufficiency of the M-condition. . .. . . . . . . . . The M-condition. .. CHAPTER II: THE GENERAL DEFINITION .. . . . . . . . . . C.. . . . . . . 23 35 CHAPTER III: EXTREMAL GEOMETRIC PROPERTIES A. . . . . Elliptic and Modular Functions.. . .. . . . . . . .. 69 73 74 81 ... . . . . . . .. . . . . . .

.. . B. . . . is Open The Infinitesimal Approach . D. 85 B. . . . . D. Beltrami Differentials !!J. . . Dependence on Parameters . . Two Integral Operators. . . . . . . . .Quasiconformal Mappings CHAPTER V: THE MAPPING THEOREM A. . .. . C. . . . .. . .. . . . . . CHAPTER VI: TEICHMULLER SPACES 90 100 106 A. Preliminaries . . The Calder6n-Zygmund Inequality. . . . . . . .. . . . . . . . . 117 120 128 137 . . . . . . . Solution of the Mapping Problem. . . C. .

2. If this were their only claim they would soon have been forgotten. It was noticed at an early stage that many theorems on conformal mappings use only the quasiconformality.c. mappings are a natural generalization of conformal mappings. It is therefore of some interest to determine when conformality is essential and when it is not. mappings are less rigid than conformal mappings. it was used to prove theorems about the conformal type of siinply connected Riemann surfaces (now mostly forgotten). 1.CHAPTER I DIFFERENTIABLE QUASICONFORMAL MAPPINGS Introduction There are several reasons why quasiconformal mappings have recently come to playa very active part in the theory of analytic functions of a single complex variable.c. 3. This was typical of the utilitarian phase of the theory. and are therefore much easier to use as a tool. The most superficial reason is that q. For instance. Q. 1 .

Extremal problems in q. was introduced by H. It reappears in 1935 in the . This was a deep and unexpected discovery due to Teichmiiller.c. case . and in supplying such a measure Grotzsch took the first step toward the creation of a theory of q. mappings lead to analytic functions connected with regions or Riemann surfaces.c. Grotzsch in 1928. 6. Conformal mappings degenerate when generalized to several variables. mappings play an important role in the study of certain elliptic partial differential equations. This theory is still in its infancy. 5. This calls for a measure of approximate conformality. mappings. 7. Grotzsch asks for the most nearly conformal mapping of this kind.c. and this particular idea was regarded as a curiosity and allowed to remain dormant for several years. The Problem and Definition of Grotzsch The notion of a quasiconformal mapping. Q. In 1936 I included a reference to the q. Instead. but not the name. If Q is a square and R is a rectangle. but q.c. but from the point of view of partial differential equations. All the work of Grotzsch was late to gain recognition. work of Lavrentiev.c. there is no conformal mapping of Q on R which maps vertices on vertices. The problem of moduli was solved with the help of q.2 Quasiconformal Mappings 4.c. A. mappings. not a square. mappings do not. They also throw light on Fuchsian and Kleinian groups.

dy) to the (du. w = u + iv) be a C 1 homeomorphism from one re- gion to another.Differentiable Quasiconformal Mappings 3 in my theory of covering surfaces. We wish to compute the ratio between the axes as well as their direction.Ii(f+if) z x y dw = f ?: dz + f. From then on the notion became generally known. and later theorems about q. (1) represents an affine transformation from the (dx. dv) plane. We return to the definition of Gr6tzsch. At a point Zo it induces a linear mapping of the differentials du (1) u x dx + u y dy vx dx + v y dy dv which we can also write in the complex form (2) with (3) f-='.dz Z Geometrically. and in 1937 Teichmiiller began to prove important theorems by use of q. mappings. It maps circles about the origin into similar ellipses. mappings. Let w = f(z) (z = x + iy.c.c. In classical notation one writes (4) with G The eigenvalues are determined from .

It now follows immediately from (2) that where both limits can be attained.u ) x y fz yiu x . Then 1f'Z I < 1fz 1 .u y x = ] v v which is the Jacobian. The Jacobian is positive for sense preserving and negative for sense reversing mappings.vy ) + j~ (v x + uy ) Ifz 12 . We conclude that the ratio of the major to the minor axis is (11) D f - Ifz 1+ If-I z 1 zl .A I 2 o _ E+G+[(E_G)2+4p 2]'h 1\1' 1\2 - The ratio a: b of the axes is (7) ( Al)lh A2 = E+ G+[(E_G)2+4p 2]'h 2(EG _ p 2 ) 'h The complex notation is much more convenient.A P P \ G.4 (5) and are (6) \ \ Quasiconformal Mappings E.\ f_1 2 = z u xy . Let us first note that lh (u x + vy ) + (8) This gives (9) j~ (v .I f'Z 1 f > 1 . For the moment we shall consider only the sense preserving case.

is positive.clZ z f zdz 1. . dl = O. We introduce now the complex dilatation (14) fIII = f z z with (15) Illll dl . the minimum when it is negative. In the dw-plane the di- arg dw 'h arg v where we have set (17) The quantity v l may be called the second complex dilatation.Differential Quasiconformal Mappings This is called the dilatation at the poiJ1t z. The maximum corresponds to the direction arg dz = a = 'h arg fl the minimum to the direction a rection of the major axis is (16) ± "/2. It is often more convenient to consider (12) 5 related to D l by (13) The mapping is conformal at z if and only if D l The maximum is attained when the ratio f.

Solution of Grotzsch's Problem We pass to Grotzsch's problem and give it a precise meaning by saying that f is most nearly conformal if is as small as possible. The mapping f is supposed to take a-sides into a-sides and b-sides into b-sides. A l-quasiconformal mapping is conformal. Let R. b and a ~ b '. We may assume that a: b sa': b ' sup D f (otherwise. For the moment. interchange a and b). It is K -quasiconformal if D f S K.6 Quasiconformal Mappings We will illustrate by the following self-explanatory figure: Observe that (3 . . One of our immediate aims is to get rid of this restriction. R' be two rectangles with sides a. The condition D f SKis equivalent to d f S k = (K -l)/(K + 1). Let it be said at once that the restriction to C1-mappings is most unnatural.a = arg f z' Definition 10 The mapping f is said to be quasiconfor- mal if D f is bounded. however. B. we prefer to push this difficulty aside.

Differentiable Quasiconiormal Mappings 7 60 6' a The computation goes a I a' a a . S a'b < fI f f (Iizl 0 a b di(x+ iY)1 < f 0 (Iizl + liz I) dx + li~) dx dy dx dy o 0 lizl + lizl lizl -lizl a' b ' or (1) o ff 0 a b ff o 0 a b D f dx dy ff R Dfdx dy and in particular ~. ~ < sup D f b' . b - The minimum is attained for the affine mapping which is given by Hz) .

c. The affine mapping has the least maximal and the least average dilatation.+ (gZ .iz z (g (g 0 f) -7 z ] z f) z i-] z g- . z .i z z - 0 where 1 = Iizl .Iii I . 0 f) i z + (g. = (g . 0 i 1 1 [(g 2 0 f). We have proved that there exists a K-q. i-I the formulas become i = 7-11 . mapping of R on R' if and only if (2) C. f) i.8 THEOREM Quasiconiormal Mappings 1. trouble with the notation which is most easily resolved by introducing an intermediate variable . = 0 2 For g (3) (i-I) . 0 f) 7z f) 7Z (~ 5 0 0 0 When solved they give 1 (2) 1 . The ratios m = al band m' = a' I b' are called the modhles of R and oR' (taken with an orientation). There is the usual f(z).[(g 0 f) .= The usual rules are applicable and we find (g (1) 0 f) z f)Z = (g. Composed Mappings We shall determine the complex derivatives and complex dilatations of a composed mapping g 0 i.

If £ is conformal. From (2) we obtain (6) /l12 ° £ £z /l12 0 f-/lf lz 1 .'ji./If 1-'ji.f/lh lz . (4) /l -1 f = -1/ f ° £-1 and.f/l12 0 f If g is conformal. /l f (8) /l 12 = 0 and f' 2 o£ =(~ 1£'[ which can also be written as (9) In any case.Differ€f1tiable Quasicon£ormal Mappings 9 One derives. (5) d -1 = d f f ° £-1 In other words. then /l12 (7) o and we find /l f . for instance. inverse mappings have the same dilatation at corresponding points. If we set g ° £ h we find from (6) (10) £z /lh . on passing to the absolute values. the dilatation is invariant with respect to all conformal transformations.

\(r).c. We shall introduce a geometric quantity .c. A function p. will be called allowable if it satisfies the following conditions: . Extremal Length Let r be a family of curves in the plane.c. called the extremal l€f1gth of r.10 For the dilatation (11) and (12) Quasiconformal Mappings d h 0 r 1 0 f log D h 0 f -1 0 f == (Ilh' Il~ . 1 2 D. Its import- ance for our topic lies in the fact that it is invariant under conformal mappings and quasi-invariant under q. and every closed subarc shall be rectifiable. which is a sort of average minimal length.. closed arcs or closed curves. Each y( r shall be a countable union of open arcs. is a metric provided one identifies mappings that differ by a conformal transformation.c. mapping is K K . defined in the whole plane. The composite of a K 1-q. and a K 2 -q. mappings (the latter means that it is multiplied by a bounded factor).q. sup (Ilh' Il~ as a distance beIt We can obviously use tween the mappings f and h (the Teichmiiller distance). the non-euclidean distance (with respect to the metric ds 2 Idwl l-lwl 2 in Iwi < 1).

If r 1 < r 2' (p) (p) h(r 1) h(r 2) . We intro- inf y~-r L (p) y L (p)2 sup . For such a p. A (p) = JJ p2 dx dy f. Y2 then L Y1 Y1 < L Y2 (p) (p) inf L < inf L Y2 and it follows at once that h (r 1) :S h([' 2) . We shall say that r1 < r2 C if every y2 contains a y1 (the y 2 are fewer and longer). Remark.p A(p) for all allowable p. Observe that THEOREM r1 r2 then implies r2 <r1 :S ! 2. 2. L y (p) duce L (p) and Defini tion: otherwise. If Y1 S.Differentiable Quasiconformal Mappings 11 1. 0. Proof. set Ly(p) f Y = p Idzl 00 if p is measurable on y*. P? 0 and measurable. 00 (the integral is over the whole plane). * (as a function of arc-length) .

12 Quasiconformal Mappings Example 1. P = 0 outside. On the other hand. hence A(n . [' is the set of all arcs in a closed rec- tangle R which joins a pair of opposite sides.:: a/b. A(p) = ab. take p = 1 in R. We have a b Example 2. [' is the set of all arcs in an annulus r1 < Iz I ~ r2 which join the boundary circles. Computation: ff p dr dO > 217 L (p ) . For any p a f o R p(x + iy) dx > L (p) ff p dx dy > b L (p) b 2 L(p)2 < ab < a b ff R p2 dx dy < ab A(p) L(p)2 A(p) - This proves A (n ~ a/b. Then L(p) proved = a.

C 2 the unbounded component of the complement. We say the closed curve y in G separates C 1 and C 2 if Y has non-zero winding number about the points of C I' Let r be the family of closed curves in G which separate C 1 A(r)-I. and C 2 • The module M(G) theannulus G = Ir 1 ::. Let G be a doubly connected region in the finite plane with C 1 the bounded. 217 0 f 217 0 L (p) r < f p de r L (p) log (r 2 ) 1 < ff pdrde 2 r L(p )2 log ( / ) 1 < 217 log (.:) < 217 log( r 2 / r ) 1 ff / rdrd{) L(p )2 A(p) . The module of an annulus. for example. Izi L (p) :s :s r21.Differentiable Quasiconformal Mappings 13 ff Equality for p = p2 rdrde II r . Consider. Example 3.

1 . For a given p (z) define p' V. K.\. Proof.14 Quasiconformal Mappings Once again p = 1/217r gives equality. COROLLARY. .2 dt:d = p S TJ f nf 2 P [¢zl + [¢zl 1 z[ . and the other inequality follows by con!:!idering the inverse. Indeed.L 217 f~ I I Z L (y) p . There are two important composition principles.\(1) ~ . so Up) M(G) = € 1 0)1 = .\(1).\(1') ~ K .1 ¢ ¢zl dxdy <_ KA(p) . for any y we have 1 < !n(y.\' ?: 1\1. This proves . l ' be the image set of 1. THEOREM 3.L [ 217 J y = dz Z 1 <. 1 and A (p) log(r2 /r 1)· il7 y log(r2 /r 1)' We conclude that il7 Suppose that all y€ 1 are contained in a region n Let and let ¢ be a K -quasiconformal mapping of n on n'. '\(1) is a conformal invariant. Then f y p'ldt:[ ?: f y p [dz[ ff .) = 0 outside n' in and n'.

0 outside (where E 1 E 2 are complementary measurable sets with r 1 S E 1 ' =P * Y1 + Y2 means "Y 1 followed by Y2'" . We may assume that 0 < A(11 ). 11 u r2 4. A(12 ) < 00 for otherwise the inequality is trivial. Y2 f 1 2 1* THEOREM a) b) if 1 1 . 11 + 12 = II. We may normalize so that L 1 (p 1 ) A (p 1 ) L 2 (P2) = A (P2 ) Choose p L(p) > L 1 (P1) + L 2(P2) A (p) < A (P1 ) + A (P2) 2: A (P1 ) + A (P2 ) + A = sup L(p)2 ATPT It follows that A 2: A 1 + A 2 . P2 = P on E 2 . U 1 2 ) = 0 there is nothing to prove. IY1 +Y21 Y1 f11 . Consider an admissible p with L (p) > 0 and 1 Proof of b). If A = A(1 set P1 and on E 1 . 1 2 A(1 + 1 2 ) > A(1 ) + A(12 ) .Differentiable Quasiconformal Mappings 15 I. 1 1 A(1 U 1 )-1 2: A(1 )-1 + A(1 )-1 1 2 1 2 lie in disjoint measurable sets. Proof of a).

A Symmetry Principle. P (z)) . and This makes L+(p)2 A(p) ~ 2 L(p)2 A(p) S 2. and let y+ be obtained by reflecting the part below the real axis and retaining the part above it (y U y = y+ U (y+)-). A (p) L(p)2 A (PI) A(p) = A(PI) + A(P2)' Thus > L (PI)2 + I and hence q. If [' = r then .e.d.16 ['2 Quasiconformal Mappings S E/ Then LI (PI) 2: L(p). E. For a given P set p(z) Then max (p (z).\([') and hence .\([') = Yi . 1.\(['+) . Proof. THEOREM 5. The notations rand ['+ are self-explanatory. 2. For given P set .\ ([' +) S 2.\ ([' ) . For any y let y be its reflection in the real axis.

e.(r+) < Y2 . mapping from Dirichlet integral of a C1 n to n'.d.(r+) . F.\. > 2L(p) On the other hand 2A (p) and hence L(p)2 A (p) .\.Differentiable Quasiconformal Mappings 17 in upper halfplane in lower halfplane Then L +(p +) = L + ( +)_ (p) y y + y L (p) + L_(p) y y L _ (p) y+y .(r) L +(p+)2 y Y2 A (p+) < < Y2 . Dirichlet Integrals z Let ¢ be a K-q.c. q.\. The function u «() is For the composite u 0 ¢ we have .

In order to define the Dirichlet integral it is not necessary to assume that u is of class C 1. (2) One may go a step further and assume that v is given only on part of the boundary. There is another formulation of this... . Clearly.18 (U 0 Quasiconformal Mappings c/J)z $ (ul. Thus we can form the Fourier transform . We may consider merely corresponding Jordan regions with boundaries y.l ° c/J)<ic/Jzl 1c/J:zP and thus (1) D(u0c/J) S KD(u) . Dirichlet integrals are quasi~invariant. Suppose that u (z) is continuous with compact support. Let v on y' and v 0 c/J on y be corresponding boundary values. There is a minimum Dirichlet integral Do(v) for functions with boundary values v.. 0 c/J)c/Jz + (ul. attained for the harmonic function with these boundary values. For instance. if v == 0 and v == 1 on disjoint boundary arcs we get a new proof of the quasiinvariance of the module.' y . ° c/J) ¢z + I(u ° c/J) zl <iul.

Tf) = 2~ !! ei(X~+Y1])u(x. n (UX)A (I).Differentiable Quasiconformal Mappings 19 Ii(~. .i~1i = iTfIi It follows by the Plancherel formula that D(u) =! !(e+Tf2)[liI2 d~dTf and this can be taken as definition of D (u) .) A y)dxdy and we know that = .

Clearly. ¢ is K-q. Definition A. = a/ b is determined by conformal mapping a The conjugate Q* is the same Q with the complementary arcs (the a-arcs). A quadrilateral is a Jordan region Q. The Geometric Approach All mappings ¢ will be topological and sense preserving f rom a region 11 to a region 11'. Its module m (Q) on a rectangle Qc 11.CHAPTER II THE GENERAL DEFINITION A. 21 . if the modules of quadrilaterals are K -quasi-invariant. m(Q*) m(Q)-l.c. together with a pair of disjoint closed arcs on the boundary (the b-arcs).

---. and a K 2 -q.) Q We subdivide Q into vertical strips Qj and then each image. 2. The composite of a K l-q. > ~J' m i J _1_ mij ~. Set mIJ . m .22 Quasiconformal Mappings The condition is m (Q ') ~ Km (Q ).. in a neighborhood of every point.' into horizontal strips Q IJ ' with respect to the ree.c. _1_. in Proof. Q.. mapping is K 1 K 2 -q. Q .c.'. l tangular structure of Q. then it is K-q. K-quasiconformality is a local property: THEOREM 1.c.. Then .. - = m(Q IJ etc.c. mappings is invariant under conformal mappings. 3. 1 m = ~ mj .c. If ¢ is K-q.c.1. If ¢ is of class C 1 . IJ . 1. m(Q') Trivial properties: S Km(Q) . n.c. ¢ and ¢-l are simultaneously K-q. It clearly implies a double inequality. The class of K-q. l\lm(Q) ::. the definition agrees with the earlier. 4..

Then. We must have equality everywhere in the proof of Theorem 1. y) is ACL (absolutely .c. m = m + m 1 2 only if the dividing line is x = m . f 2 . and this is possible only if f 1 = f 2 = z. A l-q.~.1) dx dy 0 (p . THEOREM 2.e. 1 Proof. else. Let the conformal mapping functions be f 1 . Proof. This shows that the rectangular map is the identity. The Analytic Definition We shall say that a function u(x. map is conformal. integrating over Q ff ff But fJ(p-1)2dxdy p = 1 (p 2 . IJ This gives m $ Km'. Set p = If1 'I in Q1' P = If2 'I in Q2' P = 0 everywhere. B. Hence = a. LEMMA.The General Definition 23 J - If the subdivision is sufficiently fine we shall have m I• • < Km .1) dx dy ::: 0 ![[(p2-1)-2(p-1)]dxdy $ O.

By Egoroff's theorem the limits ¢(z+M-¢(z) h P(z + ik) .e. Such a function has of course.e. . If ¢ is topological* and has partial deriva- tives a. The definition carries over to complex valued functions. u(x. in n. (k = 2) I¢zl ~ kJ¢zl K-l) K+l .e. vertical line in R. We . a.e.e.P(z ) k '" The result holds as soon as ¢ is an open mapping. ¢ is ACL in n.e.. partial derivatives ux' u y a.24 Quasiconformal Mappings continuous on lines) in the region tangle R c n if for every closed rec- n with sides parallel to the x and y-axes. It follows from B serving. then it is differentiable a. horizontal and a.c. y) is absolutely continuous on a. LEMMA 1. We shall say that ¢ is differentiable at sense of Darboux) if ¢(z)-¢(zo) = that ¢ is sense preZo (in the ¢z(zo)(z-zo)+¢z(zo)(z -zo)+ o(\z-zoP The first lemma we shall prove is an amazing result due to Gehring and Lehto.. Definition B. A topological mapping ¢ of if 1) n ia K-q.hall prove that this definition is equivalent to the geometric definition.

It is based on the identity .¢y(O)1 < ( ¢ (z ) ¢ (z) [k h (2) I¢ (z + h) I¢ (z + ik) k as soon as z ( E. Therefore a. Ih I < [). Ix I < [). On such a line a.e. Given ( > 0 we can therefore find a [) > 0 such that I¢)z) . Because of the uniformity ¢x and ¢y are continuous = on E. The same holds for vertical lines. on E.¢)z) I . It will be sufficient to prove that ¢ is differ- Remark. We obtain (1) if we apply it to h The set E is measurable. and therefore it intersects siP 0< Ih < lin ¢(z+h)-¢(z) \ . horizontal line in a measurable set.The General Definition are taken on uniformly except on a set entiable a. xo' Yo ( E is a point of linear density 1 for the intersections of E with x = Xo and y = Yo' It will be sufficient to prove that ¢ is differentiable at such a point Zo Xo + iyo' and for simplicity we assume Zo = O.e. I < [) and We follow the argument which is used when one proves that a function with continuous partial derivatives is differentiable. a. [y I < [).¢x(O)1 < ( [¢/z) .e.e. Usually Egoroff's theorem is formulated for sequences. point in E is a point of linear density 1. 25 of arbitrarily n-E small measure.

¢y (0))] If x f E we are able to derive by (2) that The same reasoning can be repeated if y proved what we wish to prove if either x f E. Ixl +I+f = IH Ix[ The same reasoning applies on the y-axis. If [z I conclude that there are points xl' x 2 ' Yl' Y2 f < _8_ we I +f E with We may then conclude that (3) holds on the perimeter of the * We are assuming for convenience that z lies in the first quadrant. We now want to make use of the fact that 0 is a point of linear density 1.¢(Q) . xL Then (ml(x))/2Ixl -> I for x -> 0. .x ¢x (0)] + [y (¢y (x) . and we can choose 8 so small that m (x) > 1 ~Ixl I + f for Ix [ < 8. For such an x the interval (-IX.26 Quasiconformal Mappings ¢(x+ iy) .¢(x) .¢(Q) .x) cannot be +f free from points of E.y ¢y (x)] + [¢(x) .x¢x(O) . Thus we have E or y f f E.Y¢y(O) = [¢(x+ iy) . for if it were we should have Ix [ 2H ml(x) ::. Let ml (x) be the measure of that part of E which lies on the segment (-x.

x¢ (0) . 27 To complete the reasoning we use the fact that ¢ satisfies the maximum principle.e. But if ¢ is differentiable at z it is immediate that f (z) is the Jacobian. and according to a theorem of Lebesgue such a measure has a symmetric derivative a.y¢ (0)1 x y $ 1¢(x*+iy*)-¢(O)-x¢x(O)-Y¢y(O)1 < 3f Iz* 1+ Ix-x* 1 I¢ x (0)1 + Iy-y* 1 1¢y (0)1 < 3dlH)lzl + fl¢x (O)llzl + fl¢y (O)llzl This is an estimate of the desired form.The General Definition rectangle (~. that is. Moreover.AlQ) m(Q) when Q is a square of center z whose side tends to zero. This defines a locally finite additive measure.I¢Z' 1 and if 2) is fulfilled we ob- . But tain f = 2 2 I¢z 1 . f(z) lim .¢(O) . f E f(z)dxdy $ A(E) (we cannot yet guarantee equality). If E is a Borel set in n we define A (E) as the area of its image. and the lemma is proved.. There is a point z* on the perimeter such thai = x* + iy* I¢(x+ iy) . ~) x (Y1 ' Y2 ). We can say a little more. and we have proved that the Jacobian is locally integrable.

If ¢ has locally integrable distributional derivatives.f ¢ hx dxdy ¢2 h dxdy f¢hydXdY for all test functions. We assume the existence of integrable functions ¢l' ¢2 such that (5) ff ff ff RTJ ¢l h dxdy -f f . ¢x and ¢y are distributional derivatives of ¢. 0 ::. This gives . y ::. if h is a test function (h ( C l with com- pact support) we find at once by integration over horizontal or vertical lines and subsequent use of Fubini's theorem ff (4) ¢x h dxdy ¢yh dxdy f f -f f -f f ¢hx dxdy ¢hydxdy In other words. x ::. a. Consider a rectangle RTJ = {O ::. More important still is the converse: LEMMA 2. then ¢ is ACL. Proof. Moreover.28 Quasiconformal Mappings We conclude that the partial derivatives are locally square integrable. TJI and choose h = h (x) k (y) with support in RTJ' In ¢l h (x )k (y) dx dy = - ff RTJ ¢ h' (x)k (y) dx dy we can first let k tend boundedly to 1.

q. The exceptional set does depend on a. but we may conclude that (6) holds a. TJ) (6) f o a ¢I(x. and they are given by . for all rational a.hn - ° B- .The General Definition ff and hence ¢I h(x)dxdy ff RTJ ¢h'(x)dxdy RTJ f o a ¢ 1 (x..d.e. TJ) . In other words.¢(o. By continuity it is then true for all a. we have ¢x "" ¢1. TJ)h '(x )dx o for almost all TJ. so does 0 w . We prove.TJ)dx for almost all TJ. integrable distributional derivatives which satisfy ¢ has locally It is now easy to show that B is invariant under con- formal mapping.! ). n . B is equivalent to B'. If w is a C 2 topological mapping and if ¢ ¢ has locally integrable distributional derivatives.. TJ) is absolutely continuous for almost all TJ. and we have proved that ¢(x. We obtain n ¢(a.e. TJ)h (x )dx -f a ¢(x. Now let h "" hn run through a sequence of test-functions such that < < 1 and hn "" 1 on (1. Moreover. a little more generally: LEMMA 3.e. ¢y "" ~ a.

¢1JY~) h d~ d1J I I ¢(.(h1J w) 1J) dx dy I I(¢ ow)(h oW)x dxdy q. For any test function how we can put II [(¢~ 0 w)~x + (¢1J 0 w)1J)(h I I [(¢~ 0 w) 7 0 0 w)dxdy 0 + (¢1J w) 1JjHh 0 w) Jdxdy I I (¢ ~ Y1J . .30 Quasiconformal Mappings (6) Proof.(hy1J) ~ + (hY~) 1J) d~ d1J I I ¢(-h~Y1J + h1JY~) d~d1J I I (¢ ow)(-(h c s w) ~x J - (h 0 1J 0 w) 1J x ) J dxdy J =I =- I (¢ 0 w)(-(h~ 0 w) ~x .e. We note first that (~x ~Y) 1J x 1J y and (X~ y~ x1J) Y1J are reciprocal matrices (at correspon ding points).~y 1J x ' the Jacobian of w. This makes ( Y~ x~ x1J) Y1J = ! ( J 1J -1J x y -~y) ~x with J = ~x 1J y . The last lemma enables us to prove: B => A .d.

in view of the lemma we need only prove that a rectangle with modulus m is mapped on a quadrilateral with m' ~ Km.The General Definition 31 Indeed.. I To do so we first determine a polygon so that n b ' . First we prove: if ¢ is q..!. Yo ~ y ~ 77· Because A (77) is in- creasing the derivative A '(77) exists a.c. Let A (77) be the image area under the mapping ¢ of the rectangle a ~ x ~ that A '(0) exists. let Q i be rectangles with height 77 and base b. We wish I I I to show. and this proof is exactly the same as in the differen- tiable case. in the geometric sense. '..' be the length of the image of b. first of all: If 77 is sufficiently small. f3. then it is ACL. i 2 .e. We shall now concentrate on proving the converse: A=>B. and we shall assume In the figure.. Let b. then the length of any curve in Qi ' which joins the "vertical" sides is nearly b.

TJ • b.2 4A.2 ::.e.d. This shows that ~ b/ .c. if ¢ is K-q. This gives a length n > b. and we have two choices for the analytic definition.) 1 * But A (TJ}/TJ'" A '(0) ~ < 00. > b. take TJ so small that the variation of ¢ on vertical segments is < €I 4n. ~_1_ b. . bi' and hence ¢ is absolutely continuous. 1 ~b. in the geometric sense. Draw the lines through 'k that correspond to verticals.2 b. b. .32 Quasiconformal Mappings Next. This proves (k = that A = > B. * The proof needs an obvious modification if b i ' = 00 •• .' . if ( min ~ < bi '.. it is easy to prove that the inequality K-l) K+l holds at all points where ¢ is differentiable. .( 1 1 On using the euclidean metric we have now. 1 1 4A i 1 < K-l. < 1 - 4K A(TJ) TJ (~b. Finally. Any transversal must intersect all these lines. 0 with q.. We have now proved the equivalence of the geometric and analytic definitions.

If the topological mapping 33 ep satis- fies epz = 0 a. ep = u + iv can be approximated by C 2 func~ tions un + ivn in the sense thet un u.c.e. If ep is q. then ep is conformal.c.. This means that null sets are mapped on null sets. Observe the close relationship with Weyl's lemma.(u·) (v nx 1 dx dy ) ) ny my JR f um dv n . mapping the image area is an absolutely continuous set function. we shall prove: THEOREM 3. and that the image area can always be represented by A (E) II E J dxdy . Finally. If R [(u mx (v) . v n ~ v and Consider rectangles R such that u and v are absolutely continuous on all sides. and epz = 0 a.. COROLLARY 2. and if ep is either ACL or has integrable distributional derivatives.The General Definition COROLLARY 1. Proof. Under a q. then ¢ is conformal.e.

and therefore are absolutely con- j j R ] dxdy for these R. For the line integral tends to aR and for n . m . n . We have proved aR UV. COROLLARY... Remark.c.. mappings. (all because u and v.... 0 a. It is not precisely trivial. Otherwise there would exist a set of positive measure which is mapped on a nullset. 00 ju dV n j v n du aR judv this tends to j v du aR tinuous on aR)... possible to prove the same for extremal lengths.. 00 00 Quasiconformal Mappings the double integral tends to ffRJ dx dy. It is .'and that proves the theorem. and consideration of the inverse mapping would lead to a contradiction. It can now be concluded that Dirichlet integrals are K-quasi-invariant under arbitrary K-q. ¢z f.e.34 As m. but in any case fairly easy to prove that the line integral does represent the image area.

We want to find the largest value of the module M (G) under one of the following conditions: I. C z con- tains the origin. (Teichmiiller) C 1 contains 0 and -1. C z contains a point at distance P from the origin. I. C 1 the bounded and C z the unbounded component of its complement. (Mori) diam(C 1 n Ilzl $ In ? A. We claim that the maximum of M (G) is obtained in the following symmetric cases. Three Extremal Problems Let G be a doubly connected region in the finite plane. III. () ~ . (Grotzsch) C 1 is the unit disk (121 $ 1) and C z contains the point R > 1. II. 35 R Go .CHAPTER III EXTREMAL GEOMETRIC PROPERTIES A.

36 II. This is. Each curve points PI' P 2 on (-00. 71A(r 0+) . (0). r be the family of closed curves that se- parate C 1 and C 2 • We know that A(r) = M(G)-I. Evidently.1 + Y2+ -)+ .be 1ongs . Say they divide y into arcs Yl and so that y = Yl + Y2' Then . SimilarlY. and hence r0 = A(r) ~ A(f). -1) and y in r has 0. Compare r with the family r of closed curves that lie in the complement of C 1 U IR I. of course. if then A(r 0) is the family r in the alleged extremal case*. harmless. have zero winding number about R and nonzero winding number about the origin. < * We enlarge r0 slightly by allowing it to contain curves which contain segments on the edges of the cut (R. Fig. y Y to r 0' and we conclude that h Y+ fro +. We show that r + = r 0 +.d. and hence A(r) = 71 A(r +) by our symmetry principle. We have proved A(r) ~ A(r) M(Go) . Thus t + C r 0+' and the opposite inclusion is trivial. Let Case l.e. -1 Quasiconformal Mappings 0 P A~. III. = A(r 0)' hence M (G) q.+ = y 1+ + Y2+ = (. . R) respectively. But r is a symmetric family. r c r. H ere y 1+ + Y2+ .!.

Koebe's one-quarter theorem gives jf'(0) I ~ 4P with equality for G theorem gives = G 1 . Open up the plane by (= VZ. Hence G 1 is extremal.[a\)2 and equality is again attained when G = Gl' In other words Ial ~ [a 11 where a1 belongs to the symmetric case. the symmetric situation. The module M (G) equals the module between the image of C 1 and the unit circle. and it increases Case 1I/. Finally. Let z = f(() map \([ < 1 conformallyonto C 1 U G with [(0) = O. It is clear that equality holds in .Extremal Geometric Properties Case II. by inversion and application of Case I. The distortion 1 = If(a) [ < [at 1['(0)[ (1. We get a figure which is symmetric with respect to the origin with two component images of C 1 and two of C 2 .Suppose f(a) -1.[a\)2 - < 4Pla[ (1. if when la[ decreases. The composition laws tells us that M(G) ~ YiM(6') where 6' is the region between C 1.and C / . [al is given the module is largest for a line segment.

W o where 0 = _ ( On setting u = ('2 + '1)/('2 -'1) one has '2.'1 '2 '1 )2 + 2('/ + ' / ) '/ . carries (-'1' -'2) into (0. 00.\ (2+~Y One verifies that equality holds for the symmetric case. f Iz 2I ::. Let '1' '2 f C 1 . .\2 2+ lui < Iwol < J4-->. 1... + be the corresponding points in the '-plane.z 2I ~ '\. C 1- -'I' -..'2 '1 ... 1) and w '2 .. 2 1. The linear transformation w = '+ '1 .'/ we get lui __ 1_ lui £ < .!.. C 1 contains points zl' z2' with Iz 11 ::.\ J4-.38 Quasiconformal Mappings By assumption. .'1 '1 + '2 '1 . Iz 1. and by use of Case II it follows again that M (G) is a maximum for the case in Figure 1 .

\ ) Hans P. Berlin.\ y4 . 277 There are simple relations between these functions. (4) x C\. Kiinzi.Springer Verlag.2. 1960.) = <1> (V4+2X" + . Quasikonforme Abbildungen. in Ergebnisse der Mathematik. * ~ log 277 277 <1> (R) ~ log 'P(p) ~ log xC\.Extremal Geometric Properties 39 In order to confonn with the notations in Kiinzi's book the extremal modules will be denoted I.) . we find the identity (3) The previous computation in Case III gives. This gives (2) and together with (1). Ob- viously. for instance. 0). III. * . II. reflection of Go gives a twice as wide ring of type GI' and one finds (1) Another relation is obtained by mapping the outside of the unit circle on the outside of the segment (-1.

Elliptic and Modular Functions The elliptic integral (1) w = f o z dz maps the upper half of the normal region G 1 on a rectangle I I I I I :8 .J'O with sides a p f 0 00 dz y(z+ U z(p-z) (2) b f P dz y(z+ l)z(z-P) Evidently..... connection with elliptic functions in much greater detail. but it is not very convenient for a study of asymptotic behavior. we want to study the This is an explicit expression.._.. We recall that Weierstrass' (4) ~-function is defined by . (3) 2" log 'I' (p) 1 = 2~ Anyway...40 Quasiconformal Mappings B..

Extremal Geometric Properties and satisfies the differential equation 41 where (6) It follows that the e k are distinct. In that halfplane p(T) = e 3 -e 1 e 2 -e 1 is an analytic function . and T' = -1/T changes pinto 1 .~ ) and these relations determine the behavior of p(T) whole modular group. for g. It is this function we wish to study.. 1. We set 1m (7) T T = w 2 / wI and consider only the halfplane > O..J does not change and the w k change by full periods.p. In other words peT + 1) (8) p( . The transforma- tion T' = T + 1 replaces p by 1/p. For purely imaginary cated: T under the the mapping by g.J is as indi- . 0. P is invariant under modular transformations aT CT +b +d with ( : ~) /2 == (~ ~) mod 2.

42

Quasiconformal Mappings

~. _
o
-2C4)1

$7///lII/l/1 /
p/l-p, rCp)
=

The image is similar to the normal region with P and we find
(9)

.1. log
T7

qt (l-.-L) -p

o< p

< 1. It is evident from this formula that p varies monoT

tonically from 0 to 1 when imaginary axis.

goes from 0 to

00

along the

It can be seen by direct computation that p" 1 as soon

as

1m

T ....

00,

uniformly in the whole halfplane. If we make

use of the relations (8) it is readily seen that the correspondence between
T

and p is as follows:

-:1

o

:1

Extremal Geometric Properties

43

.......
-

_ -t...
I
I

",
-

Fig. 2. We shall let T(p) denote the branch of the inverse function determined by this choice of the fundamental region. Observe the symmetries. They yield
(10)

+1+ i

T(-I) ;: l+i 2 T( 1-2i y'3 )

1 +iy'3

2

Relation (8) gives
T

(t) ;:

T

(p)

±. 1

(11)
T(l-p) ;: -

-+. npJ

Also

rCp> ;: - T (p)
It is clear that e TTiT is analytic at p ;: 1 with a simple

zero. Therefore one can write
(12)
p _ 1 - a e iTTT
,

(a >

0)

but the determination of the constant requires better knowledge of the function .. It will be of some importance to know how when 1m
T

Ip -11

varies

is kept fixed_ In other words, if we write

T;: S

+ it

44

Quasiconformal Mappings

we would like to know the sign of

a/as log Ip -11.
=

This har-

monic function is evidently zero on the lines s circle from 0 to 1. This proves

0, s = 1,

and a look at Figure 2 shows that it is positive on the half-

is

log

Ip-ll >

0

in the right half of the fundamental region, provided that the maximum principle is applicable. However, equation (12) shows that

a/as log Ip-ll . .

0 as r ....

00.

Use of the relation (8)

shows that the same is true in the other corners, and therefore the use of the maximum principle offers no difficulty. We conclude that axis and biggest on Re r

Ip -11
=

is smallest on the imaginary

±1.

The estimates of the function p (r) obtained by geometric comparison are not strong enough, but classical explicit developments are known. For the sake of completeness we shall derive the most important formula. The function
~(z) - ~(u) ~(z) - ~(v)

has zeros at and poles at

z z

=±u+

mW

l

+ nW 2

= ± v + mwl + nW 2 •

A function with the same periods, zeros and poles is

00

F (z)

II
n=-oo

1_ e

2m

. (nw 2 ±v-z)
W

1

(the product has one factor for each n and each ±). It is easy to see that the product converges at both ends ..

Z W 1 217i Y.z F (z) 1_ e 1.z W 1 1 00 217i - +u+ W- Z IT n=l 1.e 3 e 3 -e 1 = (WI + ( 2 )/2.q2n e 1 with one factor for each combination of signs.±.e 1. 00 Substitution gives F(z) = 2 1+ q-l IT 1 (1 + ~n+2)(I+ ~n)2(I+q2n-2) (1+ q2n+l)2 (l+q2n-l)2 .. This gives 217 i !!.SO(u) F(z) SO(z) .Extremal Geometric Properties It will be convenient to use the notation q = e l7iT 45 el7iW2 / WI.-z W 2 17 i .P = we have to choose z = w 2 /2.::.±.e W 1 217i y.. which gives e~ 217iu U e2 . It is clear that so(z) .e 1.. = .q. We separate the factor n = 0 and combine the factors + n..!!.SO( v) F(O) In order to compute 1.. v = w/2 = q.

qCp)-l for p P P+ 1 8 I-p 1 P+l 16q ~ 00 (l+q2n ) 1+ q2n-l or (16) 'PCP) P+l 16 II ( 1 + q'" 1 + q2n-l 1 00 )' .46 and F CO) Quasiconformal Mappings = ~ 2 1+ q-l 2 II (1 + q2n+l )2(1 + q2n-l)2 (1 + q2n)4 1 4q 00 n 1 (1 + q2n-l 1 + q2n y )' )8 ~) Finally 00 (13) 1 . by (13).p = 16q II ( 1 + 1 q'" 1 + q2n-l A similar computation gives (14) p = II 1 00 (1 _ = 17 q2n-l 1 + q2n-l «14) is obtained by the relation rCl/ p) = rCp) ± 1 . It gives (15) or log 'PCP) = 17 1m T(I':P) 1m T (1 + 'PCP) and.) We return to formula (9).

If we ap- ply the theorem to the inverse mapping we see that the exten- Every q.:. ¢ has Iz lSI. and this was known earlier. and 16 cannot be replaced by a smaller constant. mapping of a disk on a disk can be extended to a homeomorphism of the closed disks. Mori's Theorem Let . mapping of = Iz I < 1 onto I( I < 1. normalized by ¢(O) Mori's theorem: (1) O. AX(A) C. COROLLARY. . = ¢Az) be a K-q.2_ 1 1+ q2n-l lI>(R) < 4R For X (A) we obtain (20) AX (A ) < 4( y'4+2X + y'4=2X) < 16 .c.c. Remark.Extremal Geometric Properties This gives the basic inequality (17) 47 W(p) S 16(P+ 1) Because lI> (R ) W(R 2 - dh it follows further that (18) and (19) lI>(R) -R- 4 IT (_I_+_q n_ _ ) 4 . It follows that a continuous extension to the closed disk sion is a homeomorphism. The theorem implies that ¢ satisfies a Holder condition..

mapping. A lies inside the unit circle. Construct an annulus A whose inner circle has the segment zl' z 2 for diameter and whose outer circle is a concentric circle of radius Y2. follows. Proof. For the proof we shall also need this lemma: LEMMA..48 Quasiconformal Mappings In proving Mori's theorem we shall first assume that ep has an extension to the closed disk . If ep (z) with ep (0) = 0 is K-q.c. It is clear that the extended mapping is K-q. Consider the mapping w = ..c. Proof of (1). We assume that IZ 1 - z2 1 ~ 1. and homeomorphic on the boundary.c. IZ 1 - z2 1 < 1. inside and outside the unit circle. (z)) is a K-q. Case (i). The Kq. It follows very easily that it is ACL even on rectangles that overlap the unit circle. It will be quite easy to remove this restriction. There is nothing to prove if say. then the extension obtained by setting ep (1/ z) = 1/(¢.c.

21 .z1 1 1 K < 217 K < 217 log ~ (111.1 1 This gives Case (ii). consider the image region ~r of Iz 1 < r --0 . Now the image of the inner continuum intersects 1(1 < 1 in a set with diameter ~ Hence 1 I. and the outer continuum contains the origin.I.121 ) 8 log 1(2-1.-2~11. and we obtain To rid ourselves of the hypothesis that ¢ is continuous on Iz 1 = 1.Extremal Geometric Properties We obtain 49 217 1 log Iz2.I. 1 . A does not contain the origin.

. It is a simple matter to show that tends to the identity. To show that 16 is the best constant we consider two Mori regions (Case III) G (. If they are mapped on annuli we can map them on each other by stretching the radii in constant ratio K log X ('\1) log X ('\2) > 1 Because the unit circle is a line of symmetry it is clear that we obtain a K-q.. I \ f' ·... and hence the strict inequality is valid.c.. t/J /(0) > O.A1 1 " ..... Therefore.50 Quasiconformal Mappings and map it conformally on !w! < 1 by t/J r with t/J/O) = 0. We have now I { \ \ / /-.. and we obtain t/J r < 161 z 2 -z 1 !l1 K But this implies continuity on the boundary. __ I / I I \ I / I / 1/ I \ \ I /--]' . mapping which makes . 1t/J/¢(rz 2))-t/J/¢(rz 1 ))1 < 16!z2....z 1[1 I K Now we can let r ..... - '" / / . The function t/J (rz)) is K-q.. 1.\) and G ('\2)...\ 1 correspond to '\2 .c.... and continuous /¢ on Iz I = 1.. "- \ A1 \ J ..

c. Because Mori's theorem can be applied to the inverse ¢n -1 it ¢(O) = 0.Extremal Geometric Properties for small enough A2 . There are of course strong consequences with regard to normal and compact families of quasi conformal mappings. THEOREM 1. normalized by ¢ (0) = 0. 51 Thus 2 A ? (A )1/K 16-f 1 16 1 / K and for large K the constant factor is arbitrarily close to 16. b 1 f. With this fixed normalization the K-q. a 2 . I¢ '(0)[ = 1. The family of such mappings is sequentially compact with respect to uniform convergence on compact sets. It follows by Ascoli's theorem that every infinite sequence contains a uniformly convergent subsequence: ¢n follows that ¢ is schlicht. By Mori's theorem the functions ¢ are equicontinuous. For conformal mappings it is customary to normalize by -> ¢. The K~quasiconformalmappings of the unit disk onto itself. . form a sequen- tiaily compact family with respect to uniform convergence. For q.c. n we normalize by ¢ (a 1) b 1 . It is rather obviously K-q.c. mappings this does not make For an arbitrary region sense: we must normalize at two points. b2 · THEOREM 2. ¢(a 2 ) = b2 where of course a 1 f. 11K mappings in n satisfy a uniform Holder condition 1¢(z1)-¢(z2)1 ~ M[z1 -z21 on every compact set. Proof.

to implies Is I : :. Hence it is sufficient to prove the existence of M for G.52 Quasiconfotmal Mappings Ptoof. . to < 1. Po (depending only on a similar reasoning gives f3 ~ fJ o . We map G and G' as follows: = ep (G) conformally on unit disks The compact set may be represented by theorem gives (1 - Is I : :. not the whole plane. We know that If we can show that (w) and s (z) satisfy uniform Lipschitz conditions we are through. then A x A can be covered by a finite number of G x G. These are familiar consequences of the distortion theorem. Mori's Is I) :::. For any z l' z2 f n there exists a simply con- nected region G en. which contains z l' z2' al' a 2 · If A is a compact set in n. 16(1 -Iw 1)1/K to) Hence and Iw I : :.

mapping which transforms one quadruple into another. There exists a conformal mapping of the whole extended plane which takes if and only if the cross-ratios are equal.c. ai' a z). Problem 1. b4 ) be two ordered quadruples of distinct complex numbers. It now follows that every sequence of mappings has a convergent subsequence. If they ak into bk are not equal it is natural to consider the following problem. . Although G' is now a variable region this is possible because {3 < {3o' a number that depends only on a (that is. a 4 ) and (b l . To prove that the limit mapping is schlicht we must reverse the inequalities. For what K does there exist a K-q.Extremal Geometric Properties For instance 53 Iz '(sl)1 > Colz'(O)1 la z and hence all < C' C'lz'(O)1 The other inequality is easier. b z' b 3 . Quadruplets Let (ai' a z' a 3 . on G. D.

. a linear transformation throws (aI' a 2 .54 Quasiconformal Mappings It was Teichmiiller who first pointed out that the problem becomes more natural if one puts topological side conditions on the mapping. a 4 ) into (e . First of all. into There exists a topological mapping of the extended plane which takes the line 1. e 2 . mapping of this topological figure. as a self-mapping of the sphere punctured at aI' a 2 . e 3 .' and £ into £'. although the cross-ratios are equal it makes sense to ask whether there exists a K-q. consider the following figure: a4 aa l. (0). Obvi- ously. Therefore.. e3 . a 3 . and we may as well assume that this is the 1 2 original quadruple. for T = w21wI. e . To illustrate. a 4 this mapping is quite different from the identity mapping (it is not homotopic). a1 1. We shall have to make this much more precise. there exist periods wI' w 2 such that. Let be the finite '-plane punctured at e 1 . and let P be the z-plane with punctures at all points n . P (T) = Evidently.c. at 1. a 3 .

A topological n . n * correspond to the same subgroup ¢: 0 t/J: p . We may replace t/J by T* 0 t/J. and we say that two homeomorphisms of n on n* are .. The translation subgroup r0 consists of all transformations z . p* such that ¢ 0 «oJ = «oJ * 0 t/J. etc... ± z + row 1 + nw 2 . e 3 .. 55 The projection «oJ defines P as a cover- ing surface of n. so there is a topological mapping the base (wI *. ~ z + mW I + nw 2 . and we find that G* of F*.a3 2 . (a 1a 2 a 3 )2.. a3 . n * and «oJ: P * ... F *.a2 2 . T* I: r*.. and use corresponding notations mapping z = z + wI and A 2 z = Consider now a second quadruple (e 1 *' e 2 *' e 3 *' (0) n*. and r is I the group of all transformations z . we must check the effect of a change of t/J on the base. It is generated by z + w2 . . -W *).. We write * A 2 *Z = Z + w 2 * and call (wI' w 2 *) the * Al * Z = Z + wI' base determined by t/J. This means that the covering ¢ «oJ: * P . We know F. and r explicitly. Since ¢ does not uniquely determine t/J. the ratio r* = w */ wI * is determined 2 2 by ¢. representing loops around e 1 .. G is the least normal subgroup that contains a 2 . F is a free group with three generators aI' a2 . In this situation we know that the fundamenhas a normal subgroup G which is isomor- tal group F of n phic to the fundamental group of P... n* induces an isomorphism of F on F * which maps G on G *.Extremal Geometric Properties m(w1 /2) + n (~ /2). G... In any case. and the group r of cover transformations is isomorphic to F /G. Clearly. e 2 . w 2 *) either stays fixed or goes into (-wI *.. Al * = t/J 0 Al 0 t/J-l and A 2 * = t/J 0 A 2 0 t/J-1 are generators of r 0*.

Proof.56 Quasiconformal Mappings equivalent if they determine the same T*.I7 Its projection is a curve Y f (Y 1 1 For a given p in 11 set . Denote by (y 1 I the family of closed curves in 11 which lift to arcs in P with endpoints z and z + n y2 CU l' In general. p = p(&O(z))I&O'(z)l. e . (T cu/cu 1 ). map- ping of 11 on 11* which is equivalent to a given CPo' We need a preliminary calculation of extremal length. By methods which are beyond our scope here. The extremal length of (Y11 IS 211m T. For what K does there exist a K-q. (my 1 T I denotes the family of curves which lift to arcs with end- points z and z + mcu 1 + ncu 2 . LEMMA. We obtain f p dx 2: L (p) . one can show that two maps of 11 on 11* are equivalent if and only if they are homotopic. We may assume that cu 1 T. Con- sider the segment e in the figure: 'T / -{ /~----/(---.c. Problem 2.

Hence the length of p= 1. We are now in a position to solve Problem 2. it follows that t/J: P .Extremal Geometric Properties and we conclude that A $. This completes the proof.. 57 _2_ 1m T In the opposite direction. let S be any unimodular transformation and let U be an auxiliary linear trans- .. KIm T . A (p) = Y2 1m Of course. P * and choose the base by t/J. U * to (w/.c. But !cY2 + dY11 is also mapped on ICY2* + dY1*1 and we have also 1m aT* + b c? + d ~ KIm aT+b CT + d for all unimodular transformations. If y leads from a point y is ~ 1. choose p so that a curve y ( Iy 11 is lifted to P its image z to z + 1.. w 1 may be replaced by any cW 2 + dW 1 where (c. 1\1 1m T $. The result is that (1) 2 1m (aT+ b ) cT+d where (~ ~) is a unimodular matrix of integers.. It is clear that the image under Iy 1 *1 corresponding to w 1 *. We lift the mapping ¢: U . and we have T • L (p) = 1. If ¢ is 1m T* $. w/) determined ¢ of Iy 1/ is the class K-q.. d) = 1.. To interpret this result geometrically.

r *1 < log K .58 Quasiconformal Mappings formation which maps the unit disk planewith U(O) Sr and K Sr. or • d[r.1 r* does not lie in a which is tangent to the unit circle at U. This shows that r* is restricted to a smaller circle.1 radius depends only on K. But the points on the real axis. Mark the halfplanes bounded by horizontal lines through We know that Sr * does not lie in the shaded part. Iw I < 1 on the upper half- = r. Mapping back by U. An invariant way of expressing this result is to say that the noneuclidean distance between rand r* is at most equal to that between ib and iKb. and hence the U1 S-1 S-1 S-1 (00) and whose (00) are dense (00) are dense on the circle.1 shaded circle S-1 we see that U.

. There exists a K-q.T*) Z .. ljJ covers a mapping ¢ . It is found that ¢ will map e l . mapping equivalent to ¢o if and only if d[T. The solution to Problem 1 is therefore as follows: THEOREM 4. e 2 . w 2 *) and that ¢ lifts to ljJ determining (CW 2* + dw l *. e 3 * in this order. We have not yet proved the existence. e 3 * if and only if (~ ~) == ~) mod 2 . aw 2* + bw l *).'T) which is such that ljJ (-z) and ljJ (z + T) of = Z + (T . ljJ (z ) + T* . . ljJ (z + 1) = ljJ (z ) + 1. There exists a K-q. e 2 *. for we need only consider the affine mapping (T* . T*1 ~ log K . (6 The set of linear transformations ~ ~~@ such that (~ ~) is unimodular and congruent to the identity mod 2 is the congruence subgroup (of level 2) of the modular group.. T*1 What about Problem I? Here it is assumed that ¢o maps e l .Extremal Geometric Properties THEOREM 59 3. e 2 *. e 3 on e l *. e 3 on e l *.c. of n on n* which preserves the order of the points e i if and only if the non-euclidean distance from T to the nearest equivalent point of T* under the congruence subgroup is < log K. e 2 . When does ¢ have the same property? Assume that ¢o is covered by ljJo: P . n on n* ed which is equivalent to ¢o' and the dilatation is precisely [T. But this is immediate.fJ (z).c." T = -.. p* which determines the base (wI *. As such.

c. ponds to The triangle corres- l+iy'3 2 -1 + and we know that the corresponding T = 2 ' The near'th . COROLLARY Let ¢ be any K-q.Ic I 2: 1 I 1 I I c T + d 1 2 2:..II d I . Then the vertices of any equilateral triangle are mapped on the vertices of a triangle with Such a mapping can be approximated by piecewise affine mappings. 1 and thus A is smallest for the identity transformation. an d t he non-euc I'd ean 1 iva .60 Quasiconformal Mappings In connection with the lemma it is of some interest to determine the linear transformation (~ ~) == (~ ~) mod 2 for which the extremal length (1) is a minimum. mapping of the finite plane onto itself with K < the same orientation. V3. It follows that = Id I or Ic I - Id I Under the parity conditions Id I 2: 1 and either Id I . Then TI < 1 Ic T + dl 2: and Ic Re T + d I 2: Id I .Y21 c II or c d I 2:.Ic I ICT + dl > 7'21 c I . We claim that this is the case for the point T that lies in the fundamental region IRe IT ± Y21 < Y2. -1 + i ' es t pomt WI a rea I p IS --2. 1.

and this means that the orientation is preserved.Extremal Geometric Properties distance is log 61 Y3. Hence K < y3 guarantees that Imp * > 0. The result is at the same time global and local. . The remarkable feature of this corollary is that it requires no normalization.

What conditions must it satisfy? We suppose first that there exists a K-q. Things become slightly simpler if we map the upper halfplane on itself and assume that 00 corresponds to 00.c. The boundary correspondence is then given by a continuous increasing real-valued function h(x) such that h(-oo) = 00 and h (+ 00) = + 00. The M-condition We have shown that a q. mapping ¢ of the upper halfplane on itself with boundary values h (x ) . let e 1 < e 3 < e 2 63 .c. How regular is this mapping? Can it be characterized by some simple condition? The surprising thing is that it can.c. Namely.CHAPTER IV BOUNDARY CORRESPONDENCE A. mapping of the whole plane. It can immediately be extended by reflection to a K-q. and we are therefore in a position to apply the results of the precedin'g chapter. mapping of a disk on itself induces a topological mapping of the circumference.

what we prefer are bounds for -p and from (1) we get 1 .+~e -2nTTK _ ). x. are equidistant points x . If p = p' = and r.t. this means that or (1) 1 .1 and by our product formula (Chapter III.64 Quasiconformal Mappings be real points which are mapped on e 1 '.p UK) pUK) h (x + t) . B. e 2 '. x + t and consequently p = Yi. let us recall that p (r+ 1) = 1/p (r). For this reason the lower bound can be written as p (1 + iK) .p' < h(x) -h(x-t} < p UK) I. ~ p UK) Actually. (13)) we find p(l+iK) .8 l_e-(2n-l)1TK . e~. e 3 '.p UK) ~ p. r' are the corresponding values on the imaginary axis we have K.1 1m r ~ 1m r' ~ Kim r We shall use only the very simplest case where e 1 .1 = 16 e-TTK II 00 (1 :... In this case we have that Equivalently. e 2 .h (x) 1 . which corresponds to r = i.p UK) Even better.

e II ( 1. the upper bound M(K) THEOREM < 1~ err K 1. ax + b both of the dependent and the independent variables. We shall let Ho(M) denote the subset of functions h that are normalized by h (0) = 0. (5) h(1) 1. Obviously (3) gives the best possible value.h (x ) h{x)-h(x-t) < ( ) MK where (3) 1 rrK 00 () MK = . mappings.c. mapping satisfy the M(K) -condition (2).e-{2n-l)rrK ) 8 16 1 1+ e-2nrrK We call (2) an M-condition.c.. In other words. Observe that is is in- variant under linear transformations S: x .Boundary Correspondence 65 We have thus proved (2 ) ( ) MK1 < h(x + t) .. The boundary values of a K-q.h{x + t) < M even quite apart from its importance for q. It is important to study the consequences of an M-condi- tion (4) M-1 < h(x+t)-h(x) h{x) . I: For h Ho(M) we have immediately 1 M+l < h (Y2) M < M+l . if h I: H(M) then Sl 0 h 0 S2 I: H(M). Let H (M) denote the family of all such h.

1/2n gives +1 which proves the equicontinuity on compact sets. for some M. Let H o be a set of normalized homeomorThen H o C Ho(M) phisms h which is compact and stable under composition with linear mappings. .h (1-x)). the limit function of a sequence must satisfy (4). Indeed. x s. Hence 0 h(a+x)-h(a) s. for negative n as well. They are also equicontinuous. with the inequality reversed. As a consequence: LEMMA < (h(a+l)-h(a))(MM)n 1. for any fixed a the function h (a+ x) .h(a) h (a + 1) . In other words we have for instance which shows that the h f H 0 (M) are uniformly bounded on any compact intervals (apply to h (x) and 1 .h (a ) is normalized.. The space Ho(M) is compact (under uni- form convergence on compact sets).. Indeed.66 Quasiconformal Mappings and by induction this generalizes to (6) 1 < < This is true. the compactness characterizes the H o(M) . Actually. LEMMA 2. and this immediately makes it strictly increasing.

Boundary Correspondence Proof. Be- hftT .h (y) < < {3 1 or 1 a -~ which is an M-condition. If h f Ho(M) 1 then h (x) dx 1 M+1 < f 0 = < M M+1 H (M) . M/(M + 1). and the same reasoning gives (3 < O. and a subse- = infh(-l) . We have already proved that F cause h (tx) <'/2) ::... Let us set F (x) sup h (x ). We shall also need the following more specific information: LEMMA 3. Set a 67 h f H ' o There exists a sequence such that h n (-1) . ·For any h f H o the mapping .h (y ) < h(y) .h(y-t} h (y+ t) .h(y+ t} . Therefore a > -00.h (y) h (y.t) . h mine explicitly.My is in H o' Hence a < h (y + t) . (3 = suph(-l) for quence which converges to a homeomorphism. t>O k(x)_h(y+tx)-h(}) . However. o This is a curious function that seems very difficult to deter f Proof. some estimates are easy to come by. a..

h(t) gives h(1+t) _ h(t) 2 1 .6B Quasiconformal Mappings we obtain. for x = Yz. ) _2_ h (t) and hence < F(Yz) (7) Similarly h((l-t)x+ t) .h(t) for t > 0 . 1 . h (1.) 1 2 < F(Yz) + F(t) .h(t) For t < 1 this gives h(l+t) 2 < F(Yz) + (1 -F(Yz))h(t) and (8) F(1+t) 2 < F(Yz) + (l-F(Yz))F(t) - Adding (7) and (8) (9) F(~) + F( l~t. Now f o F(t)dt = Yz f 0 F(~)dt = Yz f o 1 (F (~) + F (~ + ~)) dt < Yz F (Y2) + Yz ! o 1 F (t )dt .

y) (1) ~y 1 2y -yy vex. The Sufficiency of the M-condition.c. y) J f o y h(x+t)dt (h(x+t) -h(x-t))dt . Lemma 3 is a luxury. < M+l the weaker inequalities M 1 :s J o 1 h dt :s are immediate. Every mapping h which satisfies an M- condition is extendable to a K-q. The proof is by an explicit construction. mapping for a K that depends only on M. y) = u(x. From 1 F(t )dt :s F ('h) and the assertion follows. The opposite inequality follows on applying the result to I-h(l-t). and since they serve the same purpose. y) where u(x. We shall prove the converse: THEOREM 2. B. We shall indeed set ¢(x.Boundary Correspondence Hence 69 f o Remark. = y) + iv(x.

70

Quasiconformal Mappings

It is clear that v (x, y) 2: 0 and tends to 0 for y .... O. More-

over, u(x,O)

=

hex), as desired.

(1) ,

The formulas may be rewritten X+Y 1 h (t )dt u = 2y

v

1 2y

f (f
x

x~·y

x+y h(t)dt -

x

f

h(t)dt)

X-Y

and in this fonn it is evident that the partial derivatives exist and are
U

x

i

y

(h(x+y) -h(x-y))

-2i

1

jX+Y X--Y

hdt+ 21 (h(x+y) + h(x-y)) y

X+Y

1
2y2

J
x

hdt -

f

x

h dt) +

i

y

(h (x + y)

X-Y

- h(x-y))

The following simplification is possible: if we replace
h (t) by hI(t)
=

h (at + b),

a > 0,

the M-condition remains
=

in force and ¢ (z) is replaced by ¢1 (z)

¢ (az + b). Thus

¢1 (i)

= ¢ (ai + =

b), and since ai + b is arbitrary we need only

study the dilatation at the point i. Also, we are still free to choose h(O) 0, h(l)

= 1.

The derivatives now become

Boundary Correspondence
Ux = U
y

71

'12(I-h (-1)), -'12

=

f
-1 -1

1

h dt+ '12(1+h(-I)) ,

vx vy

=

'12 (1+ h (-1))
-'12

=

f
o
\

1

h dt -

f
y )

0

h dt

The (small) dilatation is given by

d =

(U

x

-

v

+ i (Vx +

U )

_ y

To simplify we are going to set

~

=

1-

f
o

1

h dt,

f3

-h(-I) ,

Tff3 = -h(-I) +

f
-1

o
hdt

( > 0).

Thus
U U

x
y

'12(1 + (3)

V

x

'12(1-{3)
'12(~+Tf{3)

'!2 (~ -

Tf(3)

v y

giving

d=

«I-e') + f3(I-Tf)) + H(I+~) - f3 (I+Tf)) «1+e') + f3(I+~)) + H(I-e'} -f3(I-Tf)) 1 + ~2 + f32 (1 + Tf2) - 2f3 (~+ Tf ) 1+ ~2 + f32 (I+Tf2) + 2f3(~+Tf)

72

Quasiconformal Mappings

We have proved the estimates

.:s
1

M,

M+ 1

1

< (

<

M+1

M

'

M+1

< ...JL .:s." - M+1

(the last follows by symmetry). This gives, for instance 2 1+d < M(M+ 1) 2 1_d
D

< 2M(M+1)

As soon as we have this estimate we know that the Jacobian is positive, from which it follows that the mapping ¢ is locally one to one.
It must further be shown that

¢ (z)

....

00

for z ....

00.

By (1) / we have
U

1

2y

(f
x

x

x+y hdt +

f
x x

hdt)

x-y

v

2~ ( J hdt x+y

x+y

u2 + v2

2; [(

f
x

hdt

If x :::: 0,

u2 + v 2

> -1 ( 2y2

J Y (f J Y
x-y
x

h dt )

+

h dt)2 ]

y

x-y

hdt

0

0

If x

.:s

0,

u 2 + v2

>

2?

1 (

-y

f

hdt

Y

Indeed. THEOREM 3. are they quasi-invariant? Of course not. for it is possible to construct functions h that satisfy the M-condition without being absolutely continuous.c. For conformal mappings of a halfplane onto itself noneuclidean distances are invariant. C. Remark: Beurling and Ahlfors proved D < M2 . For q.-mappings. For a long time it was believed that the boundary correspondence would always be absolutely continuous. . By the monodromy theorem it is a homeomorphism. To do so they had to introduce an extra parameter in the definition of ¢. 1. But this is not so. 2.Boundary Correspondence and both tend to 00 73 for y ~ 00. Quasi-isometry. When y is bounded it is with z. The mapping ¢ constructed in B 1S quasi-isometric. If noneuclidean distances are multiplied by a bounded factor (in both directions) we shall say that the mapping is quasi-isometric. Remark: It may be asked how regular a function h (t) is that satisfies an M-condition. it satisfies with a constant A that depends only on M. equally clear that u 2 + v 2 Now .= ~ 00 ¢ (z) defines the upper halfplane as a smooth unlimited covering of itself.

bzl2 = J/s [(l+e) + f32(1+r?) + 2f3(~+1])1 One finds that (2) holds with A = 4M 2(M+ 1) We omit the details.b of the whole plane onto itself. D.74 Quasiconformal Mappings It is sufficient to prove (2) infinitesimally. 1).. The real axis is mapped on a simple curve L that goes to 00 in both directions.l -< A. that is. This is to be combined with M and l<. we know already that L has zero area. and it is therefore sufficient to consider the point (0.L -< 1M. (3) A-I .. affine mappings of the z-plane are of no importance.L y v y Again. . We have v (i ) and the estimates Y:z f o 1 h dt - f -1 0 h dt "2 are immediate (using Lemma 3).c. Quasiconformal Reflection..hk. mapping <. Consider now a K-q. Is it possible to characterize L through geometric properties?* * For instance.hk.

Let j denote the reflection z . . z that interchanges Hand H*. Ob- serve that L determines h uniquely except for linear transformation (h(x) can be replaced by Ah(ax+b) + B).c.c. We can also consIder the conformal mappmg [ * from H * to n*. Then ep ° j ° ep-l is a sense-reversing K 2 _q. and we know that it must satisfy an M-condition. K~q.c. we are free to choose this mapping so that it is conformal in one of the half-planes.. mapping which interchanges n. n* and keeps L pointwise fixed. Conversely. suppose that L admits a K-q. So we see that L admits a reflection if and only if it is the image of a line under a q.. Suppose that L divides the plane into nand n* corresponding to the upper halfplane H and the lower halfplane H*. extension . F=wo[oj It is clear that F is K-q. Its restriction to the x-axis is h (x) [*-1 = ° [. reflection w.c.Boundary Correspondence 75 First some general remarks.. in H*.c. Define H (1) I F 2 = [ in H. Moreover.reflection. The mapping j 0/*-1 ° w ° [ is a quasiconformal map- ping of H on itself. Let [ be a conformal mapping from H to n. We say that L admits a K2 _ q-c. We say that the conformal mapping [ admits a to the whole plane. . mapping of the whole plane.

to g. We know that there exists a q.c.c. 0 G ° 0 11 *-1 in °* 1 0 F 1-1 is quasi- and to 12 /0 11. is conformal. reflections WI and w 2 ' and denote the corresponding conformal 0 mappings by 11> 11 *. It reduces to 1 2* in 1 . Form 2 0 G = F -1 I * 2 0 I *-1 1 0 F 1 * = in H*. for we know that a q. Is it conformal in the whole plane? To prove that this is so we show that g is quasiconformal. 12 O l I n u 1 * ° ' 01 * and continuous on L 1 . (This con- dition determines J1 in the whole plane. suppose that h is given and satis- fies an M-condition. Assume that they determine the same h = 1/. Introduce F 1 and F 2 as in (1).. How unique is L? Suppose L 1 and L 2 admit q. mapping with these boundary values: we make it a sense-reversing mapping l from H to H *. that is. 12 *. The mapping 1 g is conformal in 0 1 = U l/ * 11in 1 .. 12. I *-1 . and Theorem 3 of Chapter V will assert that we can determine ¢ when J1¢ is given. Then G is q. mapping which is conformal a.76 Quasiconlormal Mappings On the other hand.1 * .e.1 11 = 1 2*-1 2 0 0 12 .. It reduces to identity on the real axis.) Thus ¢ maps the real axis on a line L which in turn determines h. We cannot yet prove it.. and we set G z in H. but there exists a mapping ¢ of the whole plane upon itself which is conformal in H and such that ¢ 0 l 0 j is conformal in H * .. Hence F2 conformal.c.c.

* We shall solve Problem 1. Problem 2. I don't know how to solve Problem 2. will be denoted by C (K). alone. < C(K) . All numerical functions of K. (0 = (U «() and suppose that ( ¢(zo) when Zo is real.Boundary Correspondence n We conclude that g is conformal. ·z ~o . not always the same. There are two main problems: Problem 1. To characterize L by geometric properties. LEMMA 1. The characterization should be in analytic properties of the invariant f"j f'. We begin by showing that a K-q. and L is essentially unique. Hence 12 differs from 11 only by a linear transformation. i! We shall set ( * == ¢C"Z).c. reflection retains many characteristics of an ordinary reflection. To characterize f (and f).

) in our figure. T . 1m T 2: %. Also.78 Proof. "" and p .. p . Remark. that is within n.) be the shortest euclidean distance from 'to L. Quasiconformal Mappings We note that p Zo Z satisfies Ip T I == 1.... provided that they are approached within an and this Therefore p' is bounded by a constant C (K) proves the lemma. Let 8 (. For any such p there is a corresponding T situated on the lines Re == ± %.. 1 as 1m the points angle. LEMMA 2.. -1 at ± '12....e.. for the true region to which T' is restrict- ed does not reach these points. distance log K of these T' lines. It is not necessary to investigate the behavior of p (T) at T = ± %.')/('0 . This means that lies in a W-shaped region indicated We note that the points ± 1 where p = "" are shielded from the W-region. C (K )-1 < C(K) . We conclude that there is aT' corresponding to p * ('0 .

I I \ . SO A(~o) = 2Iw'(~o)1 < _2_ - In the other direction Koebe's distortion theorem gives 'I. and we know that it can be changed to a C (K) quasiisometric mapping."--Y' .. Schwarz's lemma gives But the noneuclidean line element at the origin is 21 dw I./ . The regions nand n metrics defined by 79 * carry their own noneuclidean A Id~1 for ~ Idz I y = /(z).. It follows that we can replace w by a reflection w' such that (at points C(K)-l A ld~1 t < w '(~)) C(K) A Id~1 s A*Idt I But it is elementary to estimate A(~) in terms of 8(~)...Boundary Correspondence The proof is trivial.. • Iw'(~o)1 1 1 28 (~o) .c. For this purpose map n w(~o) conform ally on o 8(~o) w < 1 with Iw I = O. ~o. The reflection w induces a K-q.. mapping of * H on H.. \ I I .

c.-- . c. This is a surprising result. lies on the imaginary axis. I I L Now consider three. o Again it is obvious that we have proved: Ip 1 has a maximum C (K) and . reflection across L..1 :s arg It can also be chosen so that * :s 17 ... so T* is T restrict- ed to the following region "..2 arc tan K -1 I Re T*1 :s 1. for a priori one would expect the stretching to satisfy only a Holder con·dition. If there exists a K-q. points on L such that ~3 lies between ~1' ~2' Now P fore T = (zl .z3)/(zl . reflection which is differentiable and changes euclidean lengths at most by a factor C (K) . ..z2) T is between 0 and 1 which means that *.80 Quasiconformal Mappings Combining the results with Lemma 2 we conclude LEMMA 3. There- lies in an angle 2 arc tan K . then there is also a C (K )-q.

if K is given C depends only on K.c. . and if C is given there is a K-q. A necessary and sufficient condition for L to admit a q. E.Boundary Correspondence THEOREM 81 4. It corresponds to the point e ia and can be computed explicitly. reflection where K depends only on C. In other words: THEOREM 5.c. The Reverse Inequality. We shall prove that the condition in the last theorem is not only necessary but also sufficient. If ~1' ~3' ~2 are any three points on L such that ~3 separates ~1' ~2 then It is more symmetric to write and in this form the best value of C (K) can be computed. reflection is the existence of a constant C such that (1) < C for any three points on L such that ~ 3 is between ~1 and ~2' More precisely.

c. correspond to x . and hence that a q. reflection exists. let '1' '3' '2 A 1* A 2* the corresponding distance in 0*. '1' '3' correspond to h (x . h(x + t). not hold. This means that Al = A2 = 1. .t. x + t. Thus A1 A 2 = 1.'2 I from '2· If the upper bound in (2) did a 2 and /3 2 would be separated by a circular annu- lus whose radii have the ratio e 2TT • In such an annulus the ex- * The extremal distance from a to {3 in E is the extremal length of the family of arcs in E joining a to /3.82 Quasiconformal Mappings Introduce notations by this figure: lSI Let Ak be the extremal distance * from 0.t). Through the conformal mapping of 0*. h (x). Through the conformal mapping of 0. and A/ a k to 13 k in = 1.'1 I from '2' while the points of a 2 have disC 1'3 . x. I I '2 -'11 '3 -'21 Indeed. We show first that Al (2) = '2 1 implies c-2 e-2TT ::.1 1'2 tance ::. it follows from (1) that the points of {32 are at distance ~ C. . If we can show that Al * is bounded it follows at once that h satisfies an Mcondition.

We choose p = 1 in the disk { I~ . Then Ly(p ~ M2 for all curves y ( r *.~21 . of (1) ~/ ( f3 2 ° By repeated application and with the help of (2) we conclude that the shortest distance between a2 and f3 2 is :::: c4 e-2" 1~2 .~31 . and we get the lower bound by interchanging ~l and ~3 ° Consider points ~ ( a2 . all points on ~ are within distance ~ from ~ . and the theorem is proved. \ * has a finite upper bound. ) dISk. . D).\ 2 > 1. We conclude that '\2*~ Since \ 1 " ( M M 1 +M 2 2 )2 * '\2 * 1.Boundary Correspondence 83 tremal distance between the circles is 1. To simplify notations write Because of (1). This proves the upper bound. p = 0 outside that within the disk or not. Let Then r* be the family of all arcs in n* joining a2 and f320 where p is any allowable function (recall Chapter I. and the comparison principle for extremal lengths would yield . whether y stays ~ M1 + M 2}. contrary to hypothesis.

CHAPTER V

THE MAPPING THEOREM

A. Two Integral Operators.
Our aim is to prove the existence of q.c. mappings f with a given complex dilatation Ill" In other words, we are looking for solutions of the Beltrami equation
(1)

f-Z == !J.fz

'

where Il is a measurable function and

III I

~ k

<

1 a.e. The so-

lution f is to be topological, and f-Z' f z shall be locally integrable distributional derivatives. We recall from Chapter II that they are then also locally square integrable. It will turn out that they are in fact locally
LP

for a

p > 2.
f

The operator P acts on functions h

LP,

P > 2, (with

respect to the whole plane) and it is defined by

(2)

Ph«,) ==

_1
"

ff

h(z)(

Z-s

~

-

1) dxdy
z

(All integrals are over the whole plane.)
LEMMA 1.

Ph is continuous and satisfies a rmiforrn Holder

condition with exponent 1- 2/p.

85

86

Quasiconforma.l Mappings

The integral (2) is convergent because h

£

LP

and

t;/'(z(z-())

£

LP,

1/p + 1/q
q

=

1. Indeed, 1 < q < 2, and for
00.

such an exponent Iz(z-()I-

converges at 0, (;6 0) and

The Holder inequality yields, for ( ;6 0,

IPh«()1

s 1;1 I h lip II z(z:,) II
q

A change of variable shows that

and we find (3) with a constant Kp that depends only on p. «3) is trivially fulfilled if (= 0.) We apply this result to hI (z)
=

h(z+ (1) . Since

we obtain

which is the assertion of the lemma.

The Mapping Theorem
The second operator, T, is initially defined only for functions
h ( C0 2 (C 2 with compact support), namely as the Cauchy princi-

pal value (5)

( .... 0

lim

1
17

We shall prove:
LEMMA 2.

For h ( C o2 , Th exists and is of class Cl. (Ph)z
(Ph)z

Furthermore,
(6) h

Th

and
(7)

ff

ITh

2
1

dxdy

Proof. We begin by verifying (6) under the weaker assumption h ( COl. This is clearly enough to guarantee that

(Ph)~
(8)

= -

;;

f f z~z?; f f z~z?;

dxdy

(Ph),

= - ;;

dxdy

.

Let y( be a circle wi th center , Stokes' formula we find

and radius (. By use of

dxdy 1 217i II 1 217i h i="[ z dz dZ = £ .) Th(') We have proved (6). lim £.. 0 = dxdy = 1 217i II [.. Observe that (8) can be written in the form (9) P(h z ) P(h) £ h .h (0) Th .... .I 1 217i hdi" 1 z-' + 217i y£ ff Iz~1 >£( -.88 Quasiconformal Mappings 1 17 II hz_z. lim 0 J dh dz ~ h dz z-' ~dZd~ ] and 1 17 I I zh~. Ph £ C2 . and by use of the second equation (9) we find (Th )z (10) (Th)z These relations show that Th £ C 1.. Under the assumption h we may apply (6) to hz.Th(O) C02 .

Unfortunately. 2 dxdy q. with the additional information that Cp p ~ ~ 1 for 2. For this reason the isometry permits us to extend the operator T to all of L 2 . we cannot extend P in the same way. by continuity. for the integral becomes meaningless when h is only in L 2 . and even if we use principal values the difficulties are discouraging. The key for solving the difficulty lies in a lemma of Zygmund and Calderon. to the effect that the isometry relation (7) can be replaced by (11) for any p > 1. The functions of class C o2 are dense in L 2 .e. Naturally.d. and for p > 2 the P-transform is well defined. We have now sufficient information to justify all steps in the calculation == ii f f Ph(Ph)zz dzdi" ii f f (Ph )h z dz di" f f Ihl which proves the isometry. this enables us to extend T to LP. .The Mapping Theorem 89 z ~ Because h has compact support it is immediate from the definitions that Ph == 0(1) and Th == O(lz \-2) as c>o.

Th n II P $ know by Lemma 1 that CP II h . We have to show that (13) f !! !(Ph)¢-Z (Ph)¢z -! ! -! ! ¢h ¢Th for all test functions ¢ ( Co 1. Solution of the Mapping Problem. the relations h Th (Ph)-Z (12) (Ph)z hold in the distributional sense. We are interested in solving the equation (1) where 111111"" $ k < 1. At present we prove: LEMMA 3.h n II p' On the left hand side we IP(h-hn)1 $ K p Ilh-hnll p Iz I1. To begin with we treat the case where 11 has compact support so that f will be analytic at "". We know that the equations hold when h ( C 0 2 . .90 Quasiconformal Mappings The proof of the Zygmund-Calderon inequality is given in section D of this chapter. The right hand members have the right limits since 11Th . For h ( LP.2 / p and since ¢ has compact support the result holds. B. P > 2. Suppose that we approximate h by h n ( C 0 2 in the LP-sense.

F = z+ a. and we have It follows that (3) If g is another solution we get I z .1 ( LP. and this is possible only for F' = 1. Therefore is analytic (Weyl's lemma). The condition I z .gz = T[/lU z - g)l and by Zygmund-Calderon we would get and since k C P < 1 we must have I z = g z a. Prool. = I-PU-) z F satisfies F z = 0 in the distributional sense. The function F Let I be a solution.e. We begin by proving the uniqueness. The normalization at the origin gives a (2) = 0. The difference must be a constant.1 ( LP implies F' .1 (LP.The Mapping Theorem We shall use a fixed exponent p > 2 such that kC p THEOREM 91 < 1. Then I z = /lIz is of class LP. = Because of the Beltrami equation we have also I z g z· Hence 1. If /l has compact support there exists a unique solution I 01 (1) such that 1(0) = 0 and I z . .g and 7- i are both analytic. This proof will also suggest the existence proof. 1. = and the normalization shows that I g. and we can form PU z ).

we obtain fz (7) and f z . The function f will be called the normal solution of (1). Secondly.1 /l(h+ 1) T[/l(h+ 1)] + 1 h+l = h ( LP.92 Quasiconformal Mappings To prove the existence we study the equation (4) Therefore the series (5) h = T (/l h) + T /l . It is obviously a solution of (4). We collect some estimates. If h is given by (5) we find that (6) is the desired solution of the Beltrami equation. The linear operator h ~ T (/l h) on LP has norm < - k C P < 1. In the first place /l (h + 1) ( LP (this is where we use the fact that /l is well defined and has compact support) so that P [/l (h + 1)1 continuous. converges in LP. From (4) so (8) and by (7) .

from (3).e. .. 0 and g . and denote the corresponding normal solution by g. (for instance through a sequence) and that the supports are uniformly bounded. The following conclusions can be made: LEMMA 1.The Mapping Theorem 93 (9) The Holder condition gives. uniformly on compact sets. (10) 1£((1) ..£((2)1 s 1-~C p K II/lli p 1(1 -(2 11 -2/ p + 1(1 -(21 Let v be another Beltrami coefficient.g z)lllp + lIT[(/l-v)fzll1p < kCpllfz-gzllp + Cpll(/l-v)fzllp We suppose that v . the convergence is in fact uniform in the entire plane.. IIgz-fz lip . For this purpose we need first a slight generalization of Weyl's lemma: * Since f-g is analytic at 00../l does... f.. We obtain f z -g z == T(lIf z -vg z ) r and hence II f z -g z ll p s II T [v(fz.. still bounded by k.. * More important. we want now to show that f has derivatives if . /l a.

\ so that the system (11) has a solution. Proof. It is sufficient to show that f y define B/z) p dz + q dz 0 for any rectangle y. and it is a topological mapping. If 11 has a distributional derivative p > 2. are of Therefore o and the result follows on letting ( and (' tend to O. B(. We try to determine . then f ( C1 . If p and q are continuous and have locally integrable distributional derivatives that satisfy then there exists a function f ( C 1 Pz = p. (. We use a smoothing operator. For ( > 0 = 1/71(2 for B( [zl ::.94 LEMMA Quasiconformal Mappings 2. The convolutions p class C 2 and * * * B( * B(. The preceding lemma tells us that this will be . We apply the lemma to prove: LEMMA 3. f F qz' with f z = q. and g B/z) = 0 for Izi > (.

= eO satisfies (12). Iz ~ 1 at 00.\ ~ 1. and (11) can be solved with I 0 We may of course normalize by 1(0) = 0. = Remark. and (on) z ~ z in LP (locally). They are dis sol ved by remarking that we can approximate that on ~ 0 0 by smooth functions on in the sense 0 a. and we set o in such a way that and q 0 = ~ P (/1 q + /1z) + constant 0 for z ~ 00.e. Doubts may be raised whether (eO) z eO0 z in the distributional sense. TheJacobian I/zI2-1/Z-12 ~ = Cl-I/11 2)e 20 is strictly positive. . Then for any test function 1>. Thus 0 is continuous Hence . Hence the mapping is locally one-one.\ (C 1 . and since f(z) 00 for z ~ 00 it is a homeomorphism. and then I ~ is the normal solution since 0.The Mapping Theorem so if (12) or (log '\)z The equation . .\z 95 (/1'\) z can be solved for q in LP.

We have f fl'j1 Pd~dTJ f fIIlIP(I£zI2 l = -1£zI2)dxdY and thus If we apply the estimate (10) to the inverse function we find (13) .96 Quasicon£urmal Mappings q. Under the hypothesis of Lemma 3 the inverse function £-1 is again K-q. Let us estimate 1 II iiI lip . with a complex dilatation ii 1 = f1-1 f which satisfies lii 1 0 £1 = 1111.e.d.c.

e. Because I is a uniform limit of K-q. II Iln \I p .. We recall the results from Chapter II.z 1 2 ) dx dy (mes e) - 1 2/ p .c. the fact that I maps null sets on null sets can be proven more easily than before.... Then mes In<e) = f <ll e e n .. and these partial derivatives are also distributional derivatives.. so that and coincides with Il..zI2 -lln. The normal solutions In satisfy (13) with In in the place of I and Il n in the place of Il· Because In. and hence I is one-one. We know further that I z f 0 a.The Mapping Theorem It is now obvious how to prove THEOREM 97 2 For any Il with compact support and III = Il· IIIl to ~ k < 1 the normal solution 01 the Beltrami equation is a q. mappings In' it is itself K-q. Il a.. Illnl ~ k and Iln = 0 outside a fixed circle.. C 1 with Il n . I and 1I1l11 p ' I satisfies (13).c.c. homeomorphism with We can find a sequence of functions Il n £. III = Iz/I z is defined a.. As such it has locally integrable partial derivatives. Let e be an open set of finite measure.e.e. (Incidentally.

For any measurable /1 with 11/111 00 < 1 00 there exists a unique normalized q. 1) If /1 has compact support we need only nor- malize f. Set ii. We want to find A so that 0 /2 = f/1. fA == f/1 0 (1/12)-1 • .) We shall now get rid of the assumption that /1 has compact support. 3) In the general case.(z) = /1 ('z) ~ 1 z2 z Then ii has compact support. f/1(z) Z 1 f ii(11 z )2 1 7 1 _2 fii 01 z) Z f!(z) Z 1 fj1(l/z )2 f!::.z I p is bounded we conclude that f is indeed absolutely continuous in the sense of area. We claim that f/1(z) = 1 f/1(11 z ) Indeed. fixed. = /11 + /12' /11 == 0 /12 fA = 0 near O. set /1 near 00. THEOREM 3. c.98 and since Quasiconformal Mappings II f n.e. Proof. 1. (II z) z z Remark: The computation is legitimate because f ii is differentiable a. mapping f /1 with complex dilatation /1 that leaves 0. 2) Suppose /1 = 0 in a neighborhood of O.

There exists a Il-conformal mapping of the upper halfplane on itself with 0. We may as well assume that f = fll is given as a q. Every q. mapping of the whole plane. THEOREM 4.c.\ has compact support. The problem is solved. Set Then Ilk+1 . Extend the definition of Il by . the successive end points being Ilk (z). SOROLLARY. 00 as fixed points. and so is the upper halfplane.The Mapping Theorem 99 In Chapter I we showed that this is so for and this . divide the noneuclidean geodesic between 0 and Il (z) in n equal parts. mappings with dilatation arbitrarily close to 1.Ilk ( o f k -1 1 . For any z. Therefore One verifies by the uniqueness that fll(i:) the real axis is mapped on itself.Ilk+1 Ilk .c.(z) = 1l(Z) 71l(z).c. mapping can be written as a finite composite of q. 1.

the following lemma: LEMMA.1 is obtained from z h = ~ T (11 h) + Til ~ O. One has . Note: I II R . It is again suf- ficient to prove the corresponding statement for the normal solution F 11. and this implies ciently small.p ~ 0 when 11 has compact support.100 Quasiconformal Mappings If fll is K-q. Let us write f(z) = 1/ f ( ~ ).p means the p-norm over Iz I Suppose first that 11 has compact support. 00.c. Here p is arbitrary. it is clear that gk K l/n_ q . Dependence on Parameters. We know that h = F 11 . and let F 11 be the normal solution obtained in Theorem 1. and = f k+ 1 0 f k -1 is C. the assertion follows for all 11 with compact support.. 1. c .p ~ 0 ~ R: for all p. We shall need If k = 111111 00 ~ 0 then Ilf~ - 111 1. Since f 11 Ilh lip c 1111 lip for the condition k C p < 1 is fulfilled as soon as k is suffi- FIl/FIl(I) and FIl(I) ~ 1. Then it is also true that Iltll z -11/ 1 . In what follows fll will always denote the solution of the Beltrami equation with fixpoints at 0.

2 dx dy ~ 1_lk2 ( !! hllzl <11 1~_112P . .Lh and J. For Iz I > R.P . and F (z) .p < 1I[(iz -1) 0 h]hzlll. Since f z = (g z v 0 h)h z + (gz v 0 h)Fz = (g z v 0 h)h z we get Ilfz-1111.1 1 1.. z uniformly.1 IP.1) + F (z)2 F (z)2 1<lzl<R and tends to zero because ff \ _ 1 \P dxdy Iz l4 IIF z -l11 R .Lt inside the unit disk and h is analytic outside.The Mapping Theorem 101 The part ffl < Izl < R can be written z2(F z(z) ...Lh = J..f f Izi < I IhzI2P-4)~ 0 .p + Ilh z .p For the first integral we obtain ff Izl<l = l(gz-l) 0 hlPlhzl P dxdy < _1_ 1-k2 ff hllzl<lI Igz -liP Ih~ _ 0 h.. In the general situation we can write f = g v 0 h where J.L g are bounded by k and have compact support. O.. F may be assumed analytic. Then J.

.~ . For any I~ I < 1 we can write f(~) = 1 217i I Iz 1=1 f(z )dz z . It is now assumed that /l depends on a real or complex parameter t and that /l(z. the inverse of f satisfies a Holder condition with exponent v . We shall show that f/l = f(z. If(z)! > mlzl K and hence Izl=o I If(z)I. O. Replace z by It becomes 1/ z in the line integral. 11K. For small Iz I we have thus.. t) has a t-derivative at t = O.1 17 II Izi < 1 fz(z) z .102 Quasiconformal Mappings One of these integrals is taken over a region slightly bigger than the unit disk. t} where v and E f L"" and II E (z.. t )11"" . The lemma is proved. 0 for t . Indeed. 1 217i Izl = 1 I v f(1jz)dz f(z )-1 z(l-z() 1.z~ dz The convergence is guaranteed as soon as t is sufficiently small to make K < 2... but this is of no importance.1 jzlldzl z~1 11 - . t) = tv(z) + te(z..~ dxdy (the Pompeiu formula).

p we find f(~) = t--> -->0' and ~ t --> v lim HC) -C t 0 _1 17 ff Izl ff [z[ v(z)( <1 z:~ z2 z-I ~ + -z- ~-1 ) dxdy ) -~ v(l/z) . .with ~(z) z (zl Z)2 Il (II z ) .z _1 17 f f ~z(z) v Izl < 1 fz f (z )2 ( ~ 2z 1. In the first integral.~ f(~) ~-1) dxdy z-l + . set = Il f z = Il (f z -1) + Il v and use a corresponding expression for f. if 1/z is taken as integration variable in the second integral. All told we get . Because Hfz-111 1 . f(l) = 1.The Mapping Theorem 103 The constants can be recovered from the normalization f(O) = 0. The convergence is clearly uniform when pact subset of I~I ranges over a com- Finally. it transforms to the same integrand as the first. <1 < 1..z~ 1-z ~z ) dx dy .~(~ 1-z~ ~ 1-z ~z dx dy .

) dxdy I. and we consider where and a~ Hz.104 and we have i(I.1 IT =- II ( kII 2 I-lila 1 ~S:) fZ 0 (lllo)-1 R (z. one verifies easily that this formula is valid for all 1.) ~ 1 - + -z 1.) Quasiconformal Mappings - kJJv(z)R(z.)) dx dy .. fila (I. fila (1. z-1 where R (z. 1.-1 = By considering f(rz). t) = i 0 fila v <to) =. and that the convergence is uniform on compact sets. !. For arbitrary to we assume in the same sense as above.))dx dy v (t 0' z )(I~Or R (lila (z) .

that is. and they can be treated con~ider only the first part. Il(t). t) (Co. and II e(s. fll(')1 dxdy < E . that (a/at) fez.) f J1(t) (. The important thing is that the improper integral converges uniformly. We may sum up our results as follows: THEOREM 105 S. moreover. t)1100 -+ 0 as = s -+ O. the integral over the plane can again be written as the sum of two integrals over in a similar manner.. t) is a continuous = function of t. We Iz I ::. Then f Il(t+s) (. 111l(t)11 00 < 1. 1. e(s.) + ti (" t) + o(s) uniformly on compact sets. where 1(" t) If v (t) depends continuously on" t (in the L00 sense) then we can prove. !! z-'i <0 \zl < 1 Iff(z)12 IR(fIl(z).The Mapping Theorem which is the general perturbation formula. It is enough to prove this for t to show that 0. so we have By inversion. Suppose where v(t).

due to Riesz. The proof is taken from Zygmund. D. We prove first a one-dimensional analogue. say. Indeed. first edition (Warsaw.106 Quasiconformal Mappings in a uniform manner. can be extended to LP. .1 f -+ 0 f 17 ff Iz-(I >f h (z ) (z _ ()2 dx dy already defined for h so that (1) C 0 2 . 1935). f J1 «())1 dxdy . this integral is comparable to JJ fJ1[z: IR(z. and it is also clear that since I f J1 . Iz-<I <81 The region of integration can be replaced by a disk with center fJ1«() and radius < 28. We are going to prove that the operator Th (0 = lim _. p 2: 2. For the remaining part of the plane (with a fixed 8) the difference tends rather trivially to 0. Trigonometric Series. The Calder~n-ZygmundInequality. The value of the integral is then uniformly of order 0 (8).111 p -+ • 0 and the other factor is bounded.

2 f(x) dx 00 f(g+x) . .v....!l 00 x-~ dx .i..2 )(u} > 0 .) 00 1 17 -xl 1 17 f f 0 x-t (x.·.x) x x2 dx x2 +. f 00 ~ f(x) dx . lul P p(p-l) lulp-2(u 2 + u 2) x ~lvlP p(p-l)lvI P . 1 17 f J. Set U + IV ~ .f(g.2 (u} + = u/) + u/) y ~ It follows that IF IP lu IP ) p2 IF IP.The Mapping Theorem LEMMA."..." > 0 is the Poisson integral.. = The imaginary part ate that V ~ + i.2 -+ Hf(~) We observe now that ~ as . 107 For Hf(~) f ( Co 1 on the real line. -xl Proof. and it is immediV(~) = f(~) The real part is u(~. . set pr.. .~)2 + .-+0...2 (u} + u/) ~ (IF IP - -p- p-l p2(JFIp-2 - luI P.

108 QuasiconformaI Mappings We apply Stokes' formula to a large semicircle / "'--. ... 00. J \ L and find easily 00 I f Hence 00 -""'l (IF «()I P - -p- p-l lu«(W') d( ~ 0 . 00 -""'l for fixed Tf > O... Lettir. 0 we get the desired inequality..g Tf . We observe that (J IF I Thus P q'( )2/ P= IIu2+ v211p/2 < Il u2 1l p/2 + Il v2 1l p/2 . .. It is easy to see that the integral goes to zero for Tf .

The Mapping Theorem 109 We continue the proof of the Caldero'n-Zygmund inequality (1). We define the operator dxdy zlzl again as a principal value. Pergamon Press. following the method in Vekua. 1962. for we obtain H'.f('-re i()) r dr )e.~l 'e reie-. Generalized Analytic Functions. But it is clear that our estimate of the I-dimensional norm can be used. On setting it can be written z re i () we see that T* f(') = 2" 1 f (~ 1 o 0 00 f('+re i()) .i() d() This implies IIT*fll P < 17 max 2 () 1 17 0 f H' i +re ()) r The norm on the right does not change if we replace . by i (). The norm is of course a 2-dimensional norm. and then the integral becomes Hf()(') where f()(z) = f(ze i()). AP P Ilf() I P P .

~ Apllf lip' r* to LP by continuity.110 so that. If f 2z Iz I C 1 we obtain r*f(') . of course.1- _1_ = _ _ 1_ .1az f z _1_ dxdy Iz I I I (z+') Iz I _1_ dxdy (2) 17 1 . finally.r*r*f r f for f f C 0 2 . The proof of (1) We extend will now be completed by showing that rf = .. . QuasiconformaI Mappings I r*f lip : :.. 11Hz) dx dy Iz-~ ~ ~ Ilf(Z) (lz~'1 For any test function ¢ it follows that lh-) dxdy . 0 az Iz I 17 . because the integral on the right is absolutely convergent (compare the proof for Pf).. allows us to extend We have to LP also. This remains true for f f LP. This. This means that (2) is valid in the distributional sense.1 1 17 IIHz+').1- a.

The integral blows up logarithmically. Moreover.The Mapping Theorem We can now write 111 T*T*[(w) (One should check the behavior for z = 0 and z = w. for it is quite evident that a az tends to zero.) We have to compute . we can replace our expression by .. lim az R ~ "" The differentiation and limit can be interchanged.d. uniformly on compact sets.... and the boundary integral over a small circle tends to zero.

which is an integral over two slivers.Jz 1~llz-~ I~-w! < R . J de dT/ Iz-w! I~-wl . By an obvious change of variable the first integral becomes J~ ff !~I < Rllz-wl J Jz o o f R/lz. (z-w)lz-w! and it is obvious that the limit is _ cond limit is . We now obtain T*T*t(w) z-w' 17 Similarly.171 z .112 Quasiconformal Mappings 1 1 dTf. the se- = Jw ~ [1 f f t(z )(_1 . The fact that Cp the ~ 1 as p ~ 2 is a consequence of .1) dXdY] z-w 17 Z = -~ Jw Pf(w) = -Tf(w) This completes the proof of (1). !~! <R for the difference. tends lim ~ Jz ff d~ ff uniformly to zero.wi f -I 217 =l_-'=r-'-ei-O1-- drdO _% R .

1).The Mapping Theorem Riesz-Thorin convexity theorem: The best constant C function of p 113 -+is such that log Cp is a convex 1/ p .t )a 1 + t a 2 the contention is that 1 t I Tf 11 1/ a :s. 1 will be a particular value of an analytic function cI> <'). g set 1 = / Tf· g dx dy The idea of the proof is to make f and g depend analytically on a complex variable . . I! Tf 11 1 / a = s~p f Tf· g where g ranges over all functions of norm one in L 1/a' Be- cause simple functions (measurable functions with a finite number of values) with compact support are dense in every LP it is no restriction to assume that f and g are such functions.. a2 '. and we shall be able to estimate its modu- lus by use of the maximum principle. Similar meanings for a 1'. For such fixed f. a + a' = C1 - C/ I f 11 1/ a <0 < t :s. We consider Proof. 1\ a1' 1 Assume 1 P2 = a both 2 ~ 2. I Tfl11/al < I Tf Il 1/ a 2 < If a = C1l1fll1/al C211£111/ a 2 (1. We note first that dxdy Conjugate exponents will correspond to a and a' with 1.

a('). and we I I } } <I> (. '"' F. Weset ¢(.) = f=O. G.) 11 1 / a ' 1 1 .) = I. F j X j which makes TF (. F. X. = (1-')a 1 + 'a 2 = 1.) are func- tions of z. aje \ ' with real \ .) is bounded if Re ( stays bounded.) and G (. X. say. Similarly G (.* dxdy ..1 . } } <I> «() We see at once that it is an exponential.) = 0. For and hence Ifl a / a Igl a 1'1 a ' It follows that IIF (. .) 11 1/ a 1 II G (.) = I.polynomial: I. We agree that F (./!T X. J lG. Ob- enters as a parameter: F (. *. TX. Therefore <I> (.) is itself a simple function I.114 Quasiconformal Mappings For any complex " F(') define G(') ill1 If I· a _f ~' If I Igi a -L Ig I and a(')' where a(') serve that . Consider now the special cases ~ we have Re a(~) = a 1 ~ = ~ = ° = ° and ~ = 1.) find = L . Note that F(t) ° whenever = f. G(t) = g. g=O. G (.) = F (.

. 1 IIG(~)1I1/a' 1 < C1 . 1. and for ~ == t we obtain the desired result.The Mapping Theorem For simplicity we may assume that ly a normalization). A symmetric reasoning shows that ::.~ log C 2 < 0 on the boundary of the strip 0 ::. C2 on ~ == 1. We now conclude that log I¢(~)I .(1-~) log C 1 . Since the lefthand number is subharmonic the inequality holds in the whole strip. ~ ::. We now get I¢(~)I IITF(~)1I1/a I¢(~)I 11S 1 (this is mere- II f 11 11a ::.

The cover transformations of S over S are represented by linear transformations of H upon itself which form a discontinuous subgroup can write S and the canonical mapping Let S be a Riemann surface whose universal covering r of the group n of all such transformations. o Bo ( n then on orbits of r0 (for Bo Az (Bo ABo -I) Boz ). We r\ -+ H (orbits) 77: H r\H is a complex analytic projection of H on S. Conversely. TEICHMULLER SPACES A. let there be given a topological mapping g: So -+ S . It can be lifted to a topological mapping g: So - -+ S which 117 .1 . Therefore Bo determines a one-one conH formal mapping of So == on S. Preliminaries S is conformallY isomorphic to the upper haIfplane H.CHAPTER VI .. if z -+ B Oz maps orbits of r r o\ ro == Bor B . Indeed. Conjugate subgroups represent conformally equivalent Riemann surfaces.

118
obviously satisfies

QuasiconformaI Mappings

17 0

g

g

0

17

0

·,I--l
g
So --.;:;..g

•S

If g is conformal, so is g

and we hav.e g

1

0

= B 0 1 Bo-1

.

The classes of conform ally equivalent Riemann surfaces correspond to classes of conjugate discontinuous subgroups of

n

(without elliptic fixpoints). But even if g is not conformal, it is still true that

A
whenever A o

=

g

0

A

o
0 0 0

0

g -1
0 0 0

(

1

(

1 0 , for
17
0

A

17

g
17 17

A

g g
17

0

0

g-1 A o g-1 g-1
o
0 0

In other words, g defines an isomorphism

() such that

AO
where B ( 1,
A B

()

=

g

-

0

AO

0

g

--1
0

It is not quite unique, for we may replace g by Bog

B

o

o(
B

10 ,
0

This changes () into ()' with
0

o()'

i

(B

1 o A o B o- )

0

g

0

B- 1

which means that we compose () with inner automorphisms of 1 0 and 1. We say that () and ()' are equivalent isomorphisms.

Teichmiiller 5paces
LEMMA.

119

8 1 and 82 determine equivalent isomorphisms

01

and 02

if and only if they are homotopic.

Proof. If 8 l' 8 2 are homotopic they can be deformed into each other via 8 (t), say, which depends continuously on t.
We can then find and since

g(t)

so that it varies continuously with

t,

has values in a discrete set it must actually be constant.

°

Conversely, suppose 81' 8 2 determine. equivalent 01' 2 , By changing 1, i 2 we may suppose that 01 = 02' and hence

i

Define

g (t,

z) as the point which divides the noneuclidean

line segment between

i 1 (t,
i
1

z)

and

i 2 (t,

z)

in the ratio

t:

(I-t). Because

(Az)

Ail (z) Ag (z)
2

(A

A 0)
0

i 2 (Az)
it follows that

i(t, Az)
Hence
8(t )

Ai(t, z)
is a mapping from 50 to 5,

=

-) 17 0 g(t 0170- 1

and we have shown that 8 1 and 82 are homotopic.

Definition of

T(5 0 )

(the Teichmiiller space):

Consider all pairs (5, t) where 5 is a Riemann surface and f is a sense-preserving q.c. mapping of 50 onto
5. We

120

Quasiconformal Mappings

say that (Sl' f 1) ...... (S2' f 2 ) if f 2 0 11- 1 is homotopic to a conformal mapping of Sl on S2' The equivalence classes are the points of T (So), and (SO' I) is called the initial point of
T (So),

Every f determines a q.c. mapping I of H on itself, and thereby an isomorphism () of

-

ro'

Two isomorphisms corre-

spond to the same Teichmiiller point if and only if they differ by an inner automorphism of

n.

The space T (So) has a natural Teichmiiller metric: the distance of (Sl' 11)' (S2' f 2 ) is log K where K is the smallest maximal dilatation of a mapping homotopic to f 2 0 f 1- 1 Let us compare T(So) and T(Sl)' mapping of So on Sl' The mapping
(S, f) .... (S, fog)

Let g be a q.c.

induces a mapping of T(Sl) onto T(So)' Indeed, if (S, f) '"
(S', f '),

then (S, fog) '" (S', f'

0

g). This mapping is clear-

ly isometric.

B. Beltrami Differentials.
A q.c. mapping on itself which satisfies
(1)
f f: So .... S

induces a mapping f of H
A
f

-

-

0

A

o

=

0

-

for A ':. A o() ping f.

Conversely, if f satisfies (1) it induces a map-

*

For typographical simplicity both mappings will henceforth be de-

noted by f.

For every I! ( B 1 (ro ) we know that there exists a corresponding f/l which maps H on itself. It is then unique. if /If does satisfy (2). 1. The linear space of Beltrami differentials will be denoted by B (r 0) and its open unit ball with respect to the L 00 norm is denoted by B1 <ro ) . then and it follows that f 0 Ao A = o is an analytic function of f. We set 00 fixed. We normalize it so that it leaves 0. or that f 0 A = /If A o 0 [I is analytic. . and hence a linear transformation.Teichmiiller Spaces From (1) we obtain (A' (A' 0 0 121 fH z f)fz and thus the complex dilatation /If satisfies /If = (/If 0 Ao)Ao 'I A o ' or (2) A measurable and essentially bounded function /l which satisfies (2) for all A o with respect to say that (:ra is called a Beltrami differential roo Another way to express the condition is to is invariant under ro ' /If 0 Conversely.

Since Oil represents a point in Teichmiiller space we have actually defined a mapping R1 era ) .. ". Because we cannot solve the global problem we shall be content to solve the local problem for infinitesimal deformations. If instead we extend 11 to be identically zero in the lower halfplane we get a new mapping that we shall call ill normalized by fixpoints at 0..). There is an obvious equivalence relation: Ill"'" 112 if and 0112 are equivalent isomorphisms. The mapping ill was obtained by extending 11 to the lower halfplane by symmetry.122 and write Quasiconiormal Mapping<s r 11 for the corresponding group.c. The real axis is mapped on a line L that permits a q. It is very difficult to recognize this equivalence by direct comparison of III and 112. reflection.c. T(Sa) It is clearly continuous from L"" to the Teichmiiller metric. Before embarking on this road we shall discuss a different approach which has several advantages. (again J1 \ \ \ I I I I I I \\\\\ L . ill gives a q. mapping of the upper halfplane and a conformal mapping of the lower halfplane. 011 for the isomor- phism.. Clearly. 1..

2) Suppose fll h = fV on the real axis. so h can be extended to a q. and hence in H *.c. If f = f on the real axis if and only if Il v fIl (H) and f v (H) are the same and therefore f 0 (l1l)-1 = f 0 (lV)-1 . Consider the q. Our first result is LEMMA 1. f Il = fV on the real axis if and only if f Il == f v on the real axis. We get a new group r ll which is discontinuous on nUn *. The mapping = (lV)-1 0 fll reduces to the identity on the real axis. We also get a surface mapping So _ where So -> r" n Il = S and a q.Teichmiiller Spaces 123 It is again true that A Il = f 0 Il A a 0 f -1 Il is conformal.c. Proof 1). and f Il are defined for all Il ( L 00 Observe that f Il with 111111 00 < 1 (Il ( B 1) even when the group ro reduces to the trivial group.c. and hence a linear transformation (it is conformal in nand n *.. mapping of the whole plane by put- ting h (z) = z in H *. hence conformal). and it is q. mapping A f" v = 0 h 0 (I )-1.c. In 9H). We call it a Fuchsoid group. Il In f (H\ Il A = f v 0 (I )-1 Il is conformal. _ -> S as well as a conformal mapping So r Il\ n * has the conjugate complex structure of So . Il v for both are normalized conformal mappings of H on the same region. .

d. morphisms.e. This means f v f ll in H*. We shall now make the assumption that ro is of the first kind. By the normalization. Under these conditions we can prove LEMMA 2. It is then known that the orbits on the real axis are dense.d. Since they correspond to each other we see that S 0 fill = fll2 on the real axis. Therefore (/1 = Conversely. In particular the fixpoints are dense. . COROLLARY.e. then A Il1 (/2. We conclude that S maps the fixpoints of A Il1 on the fixpoints of A Il2 (attractive on attractive). III and 112 determine the same Teichmiiller point if and only if f III f 11 2 In H*. This means there is a linear transformation S such that f n A Il2 0 S = S 0 A Il1 for all A in ro . suppose (/1 and (J1l2 are equivalent iso- on the real axis. q. A Il2 = on the real axis and hence identically. Thus A is a linear transformation. S must be the identity. and III and 112 determine the same Teichmiiller point.124 Quasiconformal Mappings is conformal. which means that it is not discontinuous at any point on the real axis. q. If ro is of the first kind. If fill = fll2 = fll2 on the real axis. III and 112 in B <ro ) determine the same f1 Teichmiiller point if and only if Proof. and the normalization makes it the identity.

zl For a better formulation. a linear transform a- There are two special cases. let us denote the Schwarzian by [£1. We get F '(z) F" F' F ". For f tion. 11 _11_ _ f' ~ L. The formula reads [£ 0 g] = ([tJ 0 g )(g ')2 = + [g] A. (f ") L f' 11 2 Let us recall its properties with respect to composition.zl 11 = f". [A and for g 0 g] = [g] 0 =A [£ 0 A] = ([£] A )(A') 2 .Teichmiiller Spaces 125 This means we may identify the Teichmiiller space T (So) with the space of conformal mappings of H * of the form fll' 11 ( B1 (r o) . IF. An even better characterization is by consideration ofthe Schwarzian derivative It . Consider and let primes mean differentiation.

: <:2 _ . + n Ibnl2 + . and is therefore po<: sitive. ) .~ . F'" = _ _ F"= 2b 1 + .126 Quasiconformal Mappings On setting ¢J1 = [fJ11.) Note that F' = 1 - ~ <:2 + ... (More economically.. One computes 2~ 1<:1 =r = 17 ( f 'PdF l21 f (. )( 1 . The following theorem is due to Nehari: LEMMA * 3. ~ 1... <:3 6b 1 t . If f is schlicht in the halfplane H. (¢ dz 2 is invariant).. Ib 112 + 21 b 212 + .. (¢ 0 J1 A) A ' 2 [f 0 J1 [fJ11 We see that ¢J1 satisfies (¢J1 0 A )(A ')2 = ¢J1 which makes it a quadratic differential.... Suppose that F (<:) - b schlicht for 1<:1 > 1. r2 r 2 + b 1 + .. The integral ii ~ 1 + b 7 2 + '" is II <:1 = r 'PdF measures the area enclosed by the image of 1 1 = r.... )d <: ~ It follows that Ib 11 ~ 1. then <: + \[fll ~ ~ y-2 Proof. which is Bieberbach's FHichensatz..

. -+ 00 and hence 1..z-zo U. Then 0 [E] = ([F] Here U)U.4[Fll = $ 6 x o + iyo' Set ~ Yo = Uz = (z-zo)/(z-zo). Consider f(U-l .).2 '" _ ~ U 4 Yo 4 (z -+ zo) .. ..Teichmiiller Spaces gives [F 1 127 . Zo F (.- 6b 1 ~4 +. On going to the limit we find and then q..d. lim .2 -2iy U' = __0_ (z-zo)2 2iyo U . In view of the lemma it is natural to define a norm on the quadratic differentials by II¢> II = sup I¢> (z)1 y2 .e.) < O.

It is easy to check that f = TI/ Tl 2 satisfies [[1 = ¢. is an open subset of Q.. We want to show that f is schlicht and has a q..c. Proof. Clearly. The image of B 1 (r) under this mapping will be denoted by ~ (r). extension to the upper halfplane. We know already that all ¢ in ~ satisfy LEMMA II¢ I s 312 . We have defined a mapping /1 . ¢/1 from the unit ball B 1(r) to the space Q (r) of quadratic differentials with finite norm. ~ Quasiconformal Mappings is Open.. ~ consists of the Schwarzians [f] of functions f which are schlicht holomorphic in the lower halfplane and have a q.TI. r is the trivial group consisting only of the identity. T i l l . We choose two linearly independent solutions of the differential equation (1) TI" = - Y:z ¢ TI which we may normalize by TI~ Tl2 .c. Every holomorphic II¢ II < Y:z 1S m ~. and at other points f' f O.'128 C. ~ Q. To construct the extension we consider . It is our aim to show that ~ (r) is an open sub- set of Q(r). Hence f has at most simple poles. ¢ with 1. In this case the spaces will be denoted by B l' THEOREM. extension to the upper halfplane. Observe that the solutions of (1) have at most simple zeros. The question makes sense even in the case where ~.

TeichmiiIIer Spaces 111(Z) + (Z-Z)11{(Z) 112(z) + (z . = 1). k \F zI for some k < 1. and that it has a zero of at least order 4 at 00. 1-1 mapping. but sense-reversing. is local- ly schlicht. The extension will be defined by (3) f (z) = { f(z) F(z) z (. This is easy if ¢ is very regular. but could be Simple computations yield 00. and it is also evident that . that ¢ remains analytic on the real axis. The assumption at 00 means that there are solu00 tions of (1) whose power series expansions at and z respectively. Therefore F is Fz (2) 1 (112 + (Z". It is immediate that f and F agree on the real axis.Z)11.c.c. We have thus begin with 1 . specifically.(Z) 129 (z (. )2 Y:z¢(z-z)2 and thus = Y:z¢(z-z) 2 The assumption implies \Fz I ::. H ) * We remark first that the numerator and denominator do not vanish simultaneously (because 11{ 112 .z) 11.112 111 defined everywhere. Hence F is q. H* z ( H It must be shown that f is a q. Let us assume.

By compactness there exists a subsequence which converges to a solution problem.130 Quasiconformal Mappings 111 112 a 1z + b1 + a 2z + b 2 + 0(\:1) 0(1:1) 1. for n ~ Then Sn H* CC H* and S n z ~ Set cPn(z) = cP (Snz) Sn '(z)2.J. cP n 1l con verges to a normalized solution of 11" = . fn of the original If cP n ~ cP it is not hard to see that a normalized solu- tion of 11" = . Hence f be extended by continuity to the real axis and is a solution of the problem with J1 = -2cPY 2 .we may conclude that fa = f in H and in H * . We may of course. 00. uniformly bounded dilatations. We have Z y IcP n (z)1 = IcP(Sn z )IIS n '(z)1 2y 2 < IcP(Sn z )I(ImS n (z))2 and hence IIcP n ll ~ IIcP II· Now cP n has all the regularity pro~ ~ * perties. To prove the general case we use approximation. compose f with a linear transformation to make it normalized.h. This gives F( z) = a1 z + ~ + 2 oqz 1-1) a2 z + b which is also the limit of f. + O(lz 1-1) Now the schlichtness of f follows by the monodromy ~ theorem..J. Put Sn z 2 = (2nz-i)j(iz+2n).h.can . We can find mappings f n with [fn1 = ~n in Hand -!. if we choose the same normalizations . cP1I' Therefore.

c.1>0 II S l ~ -y f 0 shows that g l f -1 satisfies o l[gl(OI S For sufficiently small q. reflection A.2 is such that p(Old(' = Therefore. and C(K)-1 < IAzl < C(K) provided that f o is K-q. extension.c.c. The origin of Q (r) is an interior point of Suppose now that 1>0 f /1 0 l ~ and [fo 1 = We assume that f 0 maps H on n. Then the boundary curve L of n admits a q.1>0 = The noneuclidean metric in n* If. We set l p«()2 we have to show that g has a t/J [g] and determine normalized solutions 71 1' . we can choose A so that corresponding euclidean lengths have bounded ratio. If [f] 1> the composition rule for Schwarzians gives 1> .Teichmiiller Spaces But if 1> conclude LEMMA l 131 l Q (r) one verifies that /1 B (r ).c. This means that A is C (K )-q. 111> . f ollo . H * 1>0 where f o on = n*. and we 2. Ac- cording to Lemma 3 of Chapter IV D.

There is no difficulty if L is analytic and with a zero of order four at 00.( .(*)2 t/J «(*) A .c.c. and let L n be the image of the real axis under In' = f 0 S where S is as in the proof of L n is an analytic curve that admits a K-q. A A subsequence of the gn tends to a q. A It must again be shown that g is continuous and schlicht. Hence < 1 1 .c.(*)2 t/J «(*) A(() < 1\::1 ~ C(K) and ( 1/1.«() ((0 1 + 'h. limit g which is equal .p. 2 implies It/J I ~ (p n . «(. (Recall that (* = A«(L) Computation gives /1 . t/J is analytic on L The general case again requires an approximation argument. n satisfies the inequality (4).( () But IA(I (4) 'h. and is analytic on Ln' The Poincare density so that [gn 1 A t/J It/J I in ~ (p2 On * > n of 0 n * = f n (H*) is . 1(-(*I C(K) < Cp«(*)-1 . mappings i such that = t/J and A /1. reflection. Therefore we can p construct a sequence of normalized q. «(. C(K) as soon as ( is small enough. ( ( 0* i «() 17 1«(*) + «(-(*) 17 1 '«(*) 172«(*) + «(-(*)172 '«(*) (( 0 .132 Quasiconformal Mappings This time we construct g«() = 17 1«()/172«(). 1 ~ . Let f nOn' n Lemma 1.

~ (r) is an open subset of Q(r) .. f3 0 is continuous. there exist /l n = ¢n' /lo such that The proof is by writing ¢ = ~ 0 f 01 .. and it carries zero. This proves Theorem 1. We come now to the most delicate part: THEOREM 2.. Since /l. /l such that The corresponding ¢A n converge to ¢A' . Given any /lo ( B. ¢o = [f/lo 1 ( ~. we construct a mapping f3 0 : ~ . ~ as follows: Given ¢ ( ~ there exists a /l ( B 1 such that ¢ = ¢/l' With this /l we determine A by (5) fA = f/l 0 ([ /l0)-1 and set /l has the same boundary values as f l . then f /l Hence fA has the same boundary values as f A 1 and hence ¢A = ¢A It is evident that f3 0 is 1-1.... Remark: This was first proved by Bers. For if ¢n . f3 0 (¢) = ¢A' It is unique.... we conclude: COROLLARY. Moreover. 1 ¢ /l1 . ¢ have just proved the existence of /In .Teichmiiller Spaces 133 satisfies (4) to g in n*. The idea of the . for if ¢ /l = . For every sequence of ¢n ( ~ converg. proof that follows is due to Clifford Earle....

then C = f 11 0 A f -1 11 = f 11 0 (f 11)-1 0 B 0 f 11 0 ([ )-1 on f 11 (R).1(N) Q([') n ~ n f3. hence a linear transformation. The origin has a neighborhood N in Q(f'1l 0 ) which is contained in Q(f'1l 0 ) n~ n = ~ (f'1l 0 ).e. From N C ~ (f'1l 0 ) * Because it is quasiconformal and conformal a. then = A 0 ([ 11)-1 = f 11 0 f -1 11 0 C 0 f 0 11 (f 11)-1 on R. Quasiconformal Mappings ( Q(f') if and only if for every 1l A (f' there exists a linear transformation B such that (Earle).134 LEMMA. * f 11 0 2) If C exists.1(N o) Q(f') n f3-1(N o) which is a neighborhood of <P 11 0 in Q (f'). Write N Nowhere No is a = Q(f'1l 0 ) n No = neighborhood in ~. a linear transformation. for fA 0 A llo 0 (fA)-1 = fll 0 A 0 ([11)-1 ~ = All From the lemma we infer that Q(f') n is mapped on Q(f'1l 0 ) n ~. Then = f3. <P fll Proof. But the last expression is a conformal mapping of H on itself. . The first expression is holomorphic in second in B nil' the 11* nil' Hence C is a linear transformation. in nil *. Assume now that 110 ( B 1 (f'). The lemma is proved. <P 0 A 0 ([11)-1 = B on R. 0 1) If B exists. We find that f3 0 maps ~ (f') on ~ (f'1l 0 ) . 0 ( Q([') is equivalent to [ill and this is true if and only if f 11 0 A 0 f 11-1 = 1l A] = [ill]' t.

.Teichmiiller Spaces we get 135 has a /1 0 neighborhood in Q(r) which is contained in ~ (r) .. the Teichmuller space T (S 0) is identified with ~ (r).. . Choose a basis ¢t' . 73~-3) 7 = defines the Riemann surfaces of genus g as a holomorphic family of Riemann surfaces. an open subset of Q(r). Conclusion: If the group r of cover transformations of H over So is of the first kind. moreover. .). We find that ~ (r) can be identified with a bounded open set in C3~-3.. that the parametrization by (71 . f3 -1 (N) c ~ (r) and this proves that ¢ q. We can show.e. According to Kodaira and Spencer a holomorphic family may be described as follows: There is given an (n+ I)-dimensional complex manifold V and a holomorphic mapping 17: V ..d. M on an n-dimensional 17.3. Each fiber surface. complex manifold M.. The Teichmiiller metric defines the same topology as the norm in Q(r) . . + 73 _. 1 7 f M is a Riemann The complex structures of V and M are related in the following way: There is an open covering lu al of V so that .(. Consider the case where So is a compact Riemann surface of genus g > 1.3 ¢3~-3 with complex coefficients.. Then Q(r) has complex dimension 3g. ¢3~-3' Every ¢ linear combination ¢ f Q(r) is a = 71 ¢1 + ...

a fact that will be important. etc. in V. * Hence Oil' Oil' and the group 0T' r Il are determined by T. Thus the points of V are orbits r l with <: ( 0T 7T- and T ( ~ (r). contained in 0T • and does not meet its images o under r T' The neighborhood N (s <:0' TO) of r T <:0 will cono 0 sist of all r T<: such that II ¢T . M will be the set ~ (r). The Riemann surfaces in our family will be S (T) = 0/r T and V will be their union.) In our case. We pick a fixed <:0 from o the orbit and d_etermine an open neighborhood N (<:0) such that Consider a point r T <:0 N is compact.).136 Quasiconiormal Mappings ha: for each a there is given a holomorphic homeomorphism Ua ~ C x M connected by ¢af3 = h a 0 hf3 -1 (for intersectint Ua' U(3)' For any T ( M the restriction of ¢af3 to hf3 (U a n Uf3 n 7T-1 (r)) shall be complex analytic (in fact. Since iT is determined in H * by the differential equation [iT 1 = ¢T' we know that iT depends holomorphically on the parameter T. these ~ (r) functions determine the complex structure of 7T-1(T). For A ( r. _ 0 Here E shall be so small that N does not meet its images under r T' This is possible because AT is near AT 0 when ¢T is . This means that AT depends holomorphically on T. 0 the corresponding AT ( rT is determined by the condition iT A = AT 0 iT in H * . The projection 7T: V ~ M is 1 (r) defined so that = S(r). Every determines a ¢ = T ( [i. To emphasize the dependence on T we change the notations to ¢T' r T.¢T II < E and <: ( N (<:0) . and ill is uniquely defined in H *.

. ll For any function F(p. T) defined on the basic neighborhoods make 17: V . if two basic neigh- borhoods U 0 and U 1 intersect... ()dxdy .. (0' TO) .. N (E.. T). M into a holomorphic family.. r) and h 1 (rT ( ) == (AT (. «(. 0 lim when the limit exists. in addition. which we T know is holomorphic in both T and (. that the parameter map- r l... and we omit the argument 11 when the derivative is taken for 11 = O. Indeed..Teichmiiller Spaces near ¢T' As a consequence. We continue with a direct investigation of f 11. The neighborhoods N (E.. T) is well-defined in 137 n l. pings h: shall be a base for the We assert.. (A (.) and any y ( L C>O we set t . We have already derived the representation i[vl«() == - ~ ff v(z)R(z... then in U 0 «(.. (0' TO) topology of V.. T) mapping h 1 0 for some n Ul' AT ( r T • hO(rT ( ) == Hence the is given by «(. there is only one (( N «(0) o r T( and the mapping h: r «(. We are assuming that t is real.. All that does not make use of the mappings f .. It is evident that the pah o-1 rameter mappings define a complex structure on V which agrees with the conformal structure of the surfaces S (T) and makes the mapping 17: V . r) . D. M holomorphic. The Infinitesimal Approach.

138 where Quasiconformal Mappings R (z. .. It is evident that f[v] is linear in the real sense. From fll(Az) = All fll(z) with Il (5) t v we obtain after differentiation i[v] 0 A = A[v] + A'f[v] .) dxdy H. ff -. Its third derivative is (4) 17 ff H i/(z) (Z" _. but not in the complex sense. ')dxdy i/(z)R(Z". ')dxdy It is holomorphic for 'l H. To obtain a complex linear functional we form (2) and we find (3) lI>[v](') = - ~ ff H i/(z)R(Z".)4 dxdy If v l B (r) one verifies that ¢ is a quadratic differential (¢ l Q(r)). . f f H v(z)R(z.) = 1 z-' _U z - ~ z-1 We apply the formula to the symmetric case: v(Z") and we write more explicitly (1) i[v 1(') = i/ (z) . = .

for all A ( i. Since i [v lz = v differentiation of (5) gives (v 0 A) A' = Az + A' v . is real on the real axis and can hence be extended by symmetry to the whole plane. and there is at most a double pole at that (6) We conclude where PAis a second degree polynomial. proof). The following conditions are all equivalent. = 0 cI> [v] == 0. if all A[v] tions: LEMMA = O. A and hence Az o because v ( B (i). There is a whole slew of equivalent condi- 1. e) i[v] = 0 on R. ¢[v 1 == 0. We conclude that the are analytic functions. but not much.Teichmiiller Spaces 139 (the existence of A requires some. The explicit formula for i[v 1 shows that it is o(lz 12 ) at The apparent singularity of AIA' at 00. movable. Moreover. c) d) a) b) A[v] PA = 0 for all A ( i. From (A A ) l l 2 2 (AI 0 0 A ) + (AI' 2 0 A )A 2 2 (A A )' (AI' A )A ' 2 2 we deduce that (7) A' 2 We shall say that v is trivial. f) II v¢ dxdy = 0 for all ¢ ( Q(['). and we write v ( N (i) . AIA' . * i\H . A-loo is therefore re00.

it follows that PA [v] + i PA[i v] '" 0 on R.. 0 (j z 12 ) it is a first degree polynomial. together with (5) and (6). and we represent it as a compact fundamental polygon S with matched sides. so > b). Hence cI> d) 0 and c) > d). = z = v. hence identically zero. Conversely. =0 cI> is a polynomial and we reason as above to conclude cI> o. If A 0. Quasiconformal Mappings a) < >b) by (6). the condition shall hold for all ¢ f Q(r) such that ffr\H 1 ¢ < 00. if f [v] = 0 on R.These points are dense on R (we are assuming that the group is of the first kind) and hence. The definition of cI>. > e). by continuity. 0 A = A' f. . gives cI> .cI> = PA [v] + i PA [i v J 0.(0) = 0 it follows that . Conversely.(A 0) = 0 for all A. then cI> is pure- ly imaginary on R and can be extended to be analytic in the whole plane. then f . Con- versely. By Stokes' formula we find f fv¢ s dxdy -ii f f. Since cI> '" Since cI> '" '" ¢. if A '" 0 then since . But both polynomials are real on R. f = 0 on R. We prove it only for the case that r\H is compact. z ¢ s I dxdy dz dz 2~ as f f ¢ dz * For r\H non-compact. if ¢ = 0. c) > a) by (5). . We know further that f . =0 But it vanishes at 0 and 1. A If cI> d) =0 .140 Proof. Condition f) is the most important one. .

and the boundary integral vanishes. ~)dxdy ~)IA '(z)1 2 dxdy v(Az)R(Az. The lemma is proved. is invariant. Hence ¢ dz = f dz . . - ~) Idx dy < 00 it is quite easy to deduce that the series converges. We obtain now ~ = = - !f . Hence <I> is identically zero. ~)A'(z)2dxdy v dx dy which is zero if f) holds. ~) dx dy We introduce a Poincare O-series From the fact that I H IR (z. consider a real ~ so that (3) takes the form ~ -~ !! H v (z ) R (z. To prove the opposite. S v(z)R(Az.Teichmiiller Spaces But the condition means that f/dz f 141 is invariant.~ L: J! -~ L: J f ~ ff !/J ~ L: A(S) v(z)R(z. We need the following lemma.

Suppose ¢ is analytic in Hand sup I¢ ly 2 < 00 • Then (8) 12 17 II H = _ (z _~)4 p(z)y2 dxdy . I we conclude that . (z-z)2 (z _~)4 _ 1 %[_1__ ~z _~)2 (z_~)2 _ 2(z-C) (:z _~)3 + 1_ + z-( 4 ~ [ __ d:Z (:z-~) ~ (z-<:yl (z-~)3J Assume first that ¢ is still analytic on R.142 LEMMA Quasiconformal Mappings 2. Then integration by parts gives 12 17 217i H R It is easy to complete the proof by applying this formula to ¢(z+iE). e). convergence. For the proof we note that _ V. Lemma 2 tells us v f N(r) if and only if A v = o. Compare formula (8) with (4). for the hypothesis guarantees absolute v ~ ¢[v 1 from B (r) to Q (r). On the other hand we may define ¢ ~ -¢ i which is a mapping from Q(r) to B (r). We have defined an antilinear mapping A: A*: that AA * is the identity. By Lemma 1. From AA * = II ¢ (:z _ ~)4 i dx dy = _ L E > 0.

Al and A 2 have four distinct fixpoints. set of normalized canonical systems that come from a surface of genus g. determines a group i generated by linear transformations AI' . £ 143 N (r) . 00 and A 2 to have a fixpoint at 1. When this is so we say that the generating system is normalized. of genus g > Q.. We are going to prove that T is an open subset of V.[vli. v is equivalent modN(r) to -¢. and that it carries a natural complex structure... The inverse isomorphism of Q(i) on B(r)/N(r) is given by A*.-1 2el 2 A 2el . We say that lA l .¢ y2 £ N (r) then fJhH 1¢ (z ) 12 y2 dx dy = 0. hence ¢ = o. A 2 l 2 2el l A 2el A . . A compact surface S.l A -1 . for if . T It can be shown that V is a real analytic manifold of dimension 6g . We conclude: A establishes an isomorphism of B <r)/N (r) on Q(r) . . Of course this is the only A *¢ which is equivalent to v.A *Av In other words. Set V set of normalized canonical systems. A l is a canonical set. By passing to a conjugate subgroup we can make Al have fixpoints at 0..TeichmrlIIer Spaces v . If it belongs to 2el a surface. A 2el which satisfy (9) ~ A A . ..6 .l = I.• .

1) has already been proved. if all Uk[v] are 0.. . . + T3~3 v 3 and Tk = t k + itk'. 0 at T = O. so are all A k [V] and hence all A[v]. u6~6' and the mapping T ~ (A )V(T) takes the form uj = h j (tl' t 1 '. and this is possible only if .144 Quasiconformal Mappings Let S determine (A 1..g-3) . Observe that ah. t3~-3' t... .. The coefficients of the A k are differentiable functions of the uk' Therefore. .. are continuously differentiable. v3~-3 of B (r)lN (r) and set vCT) = T1V1 + . [Vk] } k ah j atk ' = u}... For small T we obtain a system (A)V(T) ( T.[ivk ] If the Jacobian at the origin were zero there would exist real numbers (k' TJk such that ~ This would mean (k at~ ah. . ah j + TJk at ' k o all j and hence Hence ~ «(k + iTJk) v k ( N (r). A2~)' r with normalized generators (A) We choose a basis v 1' . . The points of the manifold V near (A) can be expressed by local parameters u 1' .. We have to show 1) The h. at } = U. } 2) The Jacobian is f..

Define A(r) in . (T 1(/1).Teichmiiller Spaces all ~k' TJ k vanish..P 3~-3 £.3(/1)) define a complex structure on T.. We have proved that the Jacobian does not The proof shows that T is an open subset of V. + ~3~-3[p]v3g-31 This implies A[p] ~1[plv1 + . AT1 (/1)v 1 + A/1 . T g. and it suffices to show this at the origin of a given coordinate system.. From this we can prove that the coordinate mappings (A/1) . We can then eliminate the term p to obtain L: 1 (~k[P] + i ~k[ip])vk £.. It also follows that if 11/111 is small enough. . are differentiable in the complex /1 = o. the ~k are complex linear functionals.. 3 Indeed. Set /1 = tp and differentiate with respect to t for t = o. We obtain A[~1[plv1 + . N(r) from which it follows that ~k[i P 1 which means that the sense at Tk(/1) = i ~k[P ] In other words... T3g-3 (/1) such that + T3 /1-3 (/1)v 3 /1-3 complex numbers T1(/1). .. + ~3~-3[plv3~-3 .. N(r) . we must show that on overlapping neighborhoods the coordinates are analytic functions. Replace p by i p. then there exist unique . 145 = o. Take /1 (T) = l TjV j and /10 = /1(T O) near zero in B(r).

a3~_3(A). . .. we conclude that a i is a complex analytic function of exactly what we required. . f o.146 Quasiconformal Mappings = B <r 11 0 ) by f 11 (r) f'\ (t) o 11 . This means that for T near TO we can write uniquely (AIl(r)) A3~-3 for B(r llo )/ N(r llo ) .. = (AIlO)A(r)) = (AIlO/a/A(r)) A) ) This is Because a/A) is complex analytic at A "... .. T at TO' .J . . By the formulas of section C in Chapter I. Near (A 11 0 ) we have coordinate functions a 1(A). we have so that A depends analytically on Now choose a basis A1 . T. 0.

He then spent three years at the Institute for Advanced Study. He took his master's degree in 1935 and his doctorate in 1938. trial manuscripts. Switzerland.F. After teaching at Syracuse University for three years he joined the faculty of the University of Chicago in 1946. and returned to The University of Michigan in 1955. They are not reprints. Gehring. Finland. Under the editorship of Paul R. HALMOS graduated from the University of Illinois in 1934. He has returned to the Institute on two subsequent occasions. he is now Professor of Mathematics at the University of Michigan. From 1958-59 Professor Gehring was a Guggenheim and Fulbright fellow at the University of Helsinki. fellow at the Eidgenossische Technische Hochshule in Zurich. FREDERICK W. lecture notes. Halmos and Frederick W. in 1947-48 as a Guggenheim fellow and in 1957-58 on sabbatical leave. They are intended to provide a setting for experimental. but original publications. . and stayed on for another year to teach at the University. GEHRING attended The University of Michigan from 1943 to 1949. and other informal mathematical studies will be published in inexpensive. paperback format. and the following year was an N. He is now Professor of Mathematics at the University of Michigan. heuristic and informal writing in mathematics that may be research or may be exposition. PAUL R.VAN NOSTRAND MATHEMATICAL STUDIES VAN NOSTRAND MATHEMATICAL STUDIES are paperbacks focusing on the living and growing aspects of mathematics. He spent the next three years as a Benjamin Pierce Instructor at Harvard University. two of them as assistant to John von Neumann.S. and received his doctorate from the University of Cambridge in 1952 after three years' study in England.

GEHRING The University of Michigan About this book: • In essentially self-contained exposition. VAN NOSTRAND COMPANY. 120 Alexander Street Princeton. INC.. of Riemann Surfaces. Professor Ahlfors was awarded the Field's medal for mathematical research at the 1936 International Congress of Mathematicians. including the National Academy of Science. the author has held professorships at the University of Helsingfors and the University of Zurich. As well as serving as chairman of Harvard's department of mathematics from 1948 to 1950. with L. HALMOS and FREDERICK W. AHLFORS is William Caspan Graustein Professor of Mathematics at Harvard University. A renowned mathematician. Intended for advanced graduate study. this volume derives from lectures delivered by Professor Ahlfors at Harvard University during the Spring Term of 1964. Riemann surfaces. and other branches of the theory of functions of a complex variable: he is the author of Complex Analysis and. New Jersey 08540 .A VAN NOSTRAND MATHEMATICAL STUDY under the general editorship of PAUL R. Sario. he is a member of numerous distinguished societies. Professor Ahlfors has contributed much to the development of conformal mapping. The Author: • LARS V. this book presents the first treatment in English of a subject rapidly growing in its interest and importance to mathematicians: German works in the same field overlap only partly. D.