Schramm Lwner Evolutions

Manjunath Krishnapur
Indian Institute of Science

A BSTRACT. Lecture notes from SLE seminar in Summer and Fall of 2010 at IISc,
Bangalore. Mainly, we follow (some chapters of) Wendelin Werners St. Flour lecture
notes on SLE, but filling in most proofs. Much of this latter is in Greg Lawlers book
from which we borrow a lot of things. Some proofs are a little different. However, we
want to keep it streamlined to a minimal introduction to SLE. In particular, we stick
to chordal versions of all theorems.

Contents
Chapter 1. Complex analysis background
1.1. Boundary behaviour of conformal maps
1.2. The standard setting
1.3. Brownian motion
1.4. Harmonic measure
1.5. Beurlings theorem
1.6. Schwarzs lemma and its many variants
1.7. Half-plane capacity
1.8. A heuristic derivation of Lwners equation
1.9. Closeness of conformal maps
1.10. Chordal version of Lwners differential equation
1.11. Continuously growing hulls
1.12. Lwner evolution in reverse

1
1
2
3
4
5
7
8
9
11
13
14
15

CHAPTER 1

Complex analysis background

1.1. Boundary behaviour of conformal maps
When does a conformal map on a region extends continuously to the boundary
point? We present two results - Schwarzs reflection principle which applies to general holomorphic function, and Theorem 1.1.3 which applies only to conformal maps.
Theorem 1.1.1 (Schwarzs reflection principle). Let f : + be analytic and suppose that Im f ( z) 0 as Im z 0. Then, f extends as an analytic function to and
the extension satisfies f ( z) = f ( z).
Needless to say, one can also perform reflection along other curves, for example if
I is a circular arc and f is an analytic function on such that Im f ( z) 0 (or if
| f ( z)| 1) as z I , then f extends as an analytic function across I by reflection. This
can be deduced from Theorem 1.1.3 by mapping I to a line segment by a conformal
w
map. For example, if I , then the map w i 11+
w works.
Definition 1.1.2. Let be a region. We say that z n converge to along a
curve if there exists a curve : [0, 1] such that ( t) if t < 1 and (1) = and
( t n ) = z n for some t 1 < t 2 < . . ..
Declare ( z n ) (wn ) if the interlaced sequence z1 , w1 , z2 , w2 , . . . converges to a
boundary point along a curve. We call the equivalence classes as prime ends. (The
usual definition is more generally applicable, but we need only this much. This is
because boundary points that we consider will have at least one prime end corresponding to them).
Theorem 1.1.3 (Rudin 14.18). Let be a bounded s.c. region. Let f : be any
Riemann map (one-one onto conformal map).
(1) If z n along a curve, then f ( z n ) has a limit that belongs to = . The
limit is the same for all sequences in the same prime end.
(2) The limits thus obtained are different for distinct prime ends.
Remark 1.1.4. The proof of Theorem 1.1.3 given in Rudins book depends on a result
that a bounded holomorphic function f on has radial limits at a.e point of the
circle. This is a result of interest in itself and is proved using maximal functions (see
chapter 11 of Rudin). For the purpose of proving Theorem 1.1.3, it clearly suffices to
show that for a.e. point on the circle, there is some curve converging to it from inside
the disk, along which the limit of f exists. This fact will fall out of other probabilistic
considerations later.
Riemann mapping theorem asserts that given a s.c. , , there does exist a
one-one onto conformal map f : . The space of all such conformal maps is
three-dimensional, since the group of conformal automorphisms of the disk is. For
example, to get uniqueness, specify interior points z0 and 0 and ask f ( z0 ) = 0
1

1. COMPLEX ANALYSIS BACKGROUND

and f 0 ( z0 ) > 0. Any three constraints should do as well, and as a consequence of

Theorem 1.1.3, these constraints can be on the boundary.
Corollary 1.1.5. Let be a bounded s.c. region. Let f : be any Riemann map
(one-one onto conformal map). Then the conclusions of Theorem 1.1.3 are valid, with
S
= R {} in place of .
i
P ROOF. ( z) := zz+
i maps onto conformally and is a continuous bijection
from onto . Apply Theorem 1.1.3 to f and tranfer results back by 1 .

1.2. The standard setting

Definition 1.2.1. A bounded set K is said to be a hull if K = K and \K is
s.c. Let r(K ) = sup{| z| : z K }.
S
Example 1.2.2. + , (0, i ] and (1, 1 + i ] (1, 1 + 2 i ] etc, are hulls while {1 + i y : y > 0}
or 2 i + or { z : x 1, x y 1} etc., are not.
Theorem 1.2.3 (Riemann mapping theorem). Let K be a hull. Then, there exists a
unique conformal map : \K such that ( z) = z + az + O ( z2 ) as z . The
number a is non-negative.
P ROOF. By Riemann mapping theorem, there is a conformal map : \K .
S
Corollary 1.1.5 asserts that t := () is a well-defined number in R {}. Postcompose by z t1zz1 which maps conformally onto itself to get a conformal :
\K such that () = .
Since K is bounded, for all real x with | x| > x0 = r(K ), as z x from above,
we get f ( z) R from above. Schwarzs reflection principle applies and extends
S
S
analytically to \(K K ) and the extension maps (, x0 ] [ x0 , ) into R. We want
to take the power series expansion of at . More precisely, set F (w) = 1/(1/w)
for |w| < 1/r(K ) and write F (w) = b 1 w + b 2 w2 + . . . as F (0) = 0. Since F maps reals to
reals, b i are all real numbers.
Thus, for | z| > r(K ), we can write ( z) = F (1/ z)1 = a 1 z + a 0 + a 1 z1 + a 2 z2 + . . .
where a i are real, which is what we mean by the power series of at . Moreover
a 1 > 0 or else would map parts of upper half plane into the lower half plane. As
a 1 > 0 and a 0 is real, w (wa 0 )/a 1 maps conformally onto itself. Compose with
to get the conformal map ( z) = z + az + O ( z2 ) where a = a 1 /a 1 . The uniqueness
is also clear.
It remains to show that a 0. This is really a boundary version of Schwarzs
lemma. For z0 , let z0 ( z) := ( z z0 )/( z z0 ) is a Riemann map from onto the
1
1
unit disk taking z0 to 0. Let w0 := ( z0 ). Then if g := w0
maps
z0 , then g
1 0
into with g(0) = 0. Hence by Schwarzs lemma, |( g ) (0)| 1 with equality if and
only if g( z) = z. Thus,
1 | g0 (0)| = |0w0 (w0 )||0 ( z0 )||0z0 ( z0 )| =

2
3
(Im z0 ) |0 ( z0 )| 1 + a y0 + O ( y0 )
=
Im ( z0 )
1 a y02 + O ( y03 )

where in the end we chose z0 = i y0 . Letting y0 , we see that a must be nonnegative. Further, a = 0 if and only if K is empty.

Remark 1.2.4. Schwarzs lemma (or Schwarz-Pick lemma) has the following meaning. Consider the disk with the hyperbolic metric (1 | z|2 )2 ( dx2 + d y2 ). If f :

is analytic, the push-forward of the metric is

| f 0 ( z )| 2
( dx2 + d y2 ).
(1| f ( z)|2 )2

Schwarz-Pick

1.3. BROWNIAN MOTION

lemma is the statement that any analytic function f : is a contraction in the

hyperbolic metric, and if it is not a strict contraction at some z, then f is a Mbius
transformation of the disk onto itself, which is an isometry of the hyperbolic metric.
dx2 + d y2

It is easy to check that with the metric

is isomorphic to the the disk
y2
with its hyperbolic metric, and hence Schwarz-Pick lemma gives that for any holo| g0 ( z)|
morphic g : , we have Im g( z) Im1 z which is what we used above.
Example
1.2.5. Suppose K = [ x, x + it] where x is real
p
p and tp> 0. Then K ( z) =
2
2
( z x) + t where the principal brach of square root re i = re i/2 for (0, 2).
2
We can write the expansion K ( z) = z + t z/2 + O ( z2 ). For K = r + := { z : | z| r }
2

we can see that K ( z) = z + rz . A not so trivial exercise is to find K for the oblique
slit [0, te i ] for some 0 < < and t > 0.
1.3. Brownian motion
What is Brownian motion? We give a quick definition and a convenient way to
visualize it. It can be shown that Brownian motion does exist.
Definition 1.3.1. Brownian motion in the plane is W = (Wt ) t0 is a collection of
complex-valued random variables on a probability space such that (a) W0 = 0 w.p.1.
(b) For any t 1 < t 2 < . . . < t k the random variables Wt i Wt i1 are independent, and
for any s < t the real and imaginary parts of Wt Ws are independent N (0, t s)
distributed. (c) The function t Wt is a continuous function w.p.1.
A convenient way to visualize Brownian motion is topfix h > 0, imagine a particle
which picks a random direction and moves a distance 2 h p
in a time duration of h,
then picks a random direction again and moves a distance 2 h etc. As h 0, this
random trajectory converges to Brownian motion in a precise sense.
What we really use in the sequel are the following basic properties of Brownian
motion.
d
d p
(1) Symmetries: e i W () = W () (rotation invariance). W ( r ) = rW () (scale ind

variance). W = W (reflection invariance).

(2) Strong Markov property: Let be a stopping time. Then, conditional on
(Wt ) t0 , the future path W ( + ) W () is a standard Brownian motion.
Our interest in Brownian motion for now is because of its usefulness as a tool to
prove things about harmonic functions and related things! Here is the basic result
that connects Brownian motion to harmonic functions and the Dirichlet problem.
Theorem 1.3.2. Let be a region such that c has a connected component
containing more than one point1. Let := inf{ t : Wt }. Let f : R be a bounded
measurable function.
(1) P z ( < ) = 1 for any z and u( z) := E [ f (W )] is well defined and harmonic in .
(2) Assume that f is continuous at a point and that there is a connected
set K c that contains and has more than one point. Then u( z) f ()
as z .
1Here and elsewhere, we think of as a subset of the sphere S{}, so is included in c .

1. COMPLEX ANALYSIS BACKGROUND

(3) Assume that for every , there is a set K as in (2). Then, for any
f C (), the function u is the unique solution to the
Dirichlet problem: Find v C () such that v = 0 in , and v = f on .
P ROOF. (1) We leave the part about P z ( < ) = 1 to Exercise 1.3.3. Fix z and
r such that z + r . Define the stopping

time T := inf{ t : |Wt z| = r }. By strong

Markov property of W we get E u( z) WT = u(WT ). Take expectations again to get
R 2
u( z) = E [ u(WT )] = 0 u( z + re i ) d2 by rotation invariance of W . Thus u has meanvalue property. Therefore u = 0 in .
(2) Fix > 0
and pick < dia(K ) so that | f () f ()| < for any B(, ). If
z , then P z W exits B(, ) before hitting K 0 [Why?]. On the complement of
this event W B(, ) and hence | f (W ) f ()| < . Taking expectations, we get

E [| f (W ) f ()|] + 2k f kP z W exits B(, ) before hitting K 0

as z . That is u( z) f (), hence u C () and u = f on .
(3) u is a solution by part (2). Uniqueness is by maximum principle.

Exercise 1.3.3. Let be a region such that c has a connected component

containing more than one point and let := inf{ t : Wt }. Then, P z ( < ) = 1 for
any z . [Hint: By symmetry, there is a positive probability that W started at a
point inside ( z, r ) exits the disk through a given arc I ( z, r ). Try to construct
a sequence of disks and arcs so that if W exits those disks in those arcs, then it hits
c ].
Remark 1.3.4. Martingales are another way to link harmonic functions to Brownian
motion! Suppose u is a harmonic function on . Then, using the mean value property of u, one can check that u(Wt ) is a (bounded and continuous) martingale. By the
a.s.
martingale convergence theorem there is a random variable X such that u(Wt ) X .
This means that for a.e Brownian trajectory starting at 0, the limit of u along the
trajectory exists. In particular, for a.e point on , there is at least one trajectory
converging to it, along which the limit of u exists. This (applied to Re f and Im f )
is precisely what we needed to complete the proof of Theorem 1.1.3 (see the remark
following that theorem)!
1.4. Harmonic measure
Definition 1.4.1. Let be a region such that c has a connected component
containing more than one point. Then := inf{ t : B t } is finite w.p.1. Define the
z ( A ) = P z (W A ) for Borel set A . We call z the harmonic measure on in as
seen from z. It is a probability measure supported on .
An alternative definition is to define for each z the linear functional T z on
C b () by T z f = u( z) where f C b () and u is the unique solution to the Dirichlet
problem with boundary value f (the solution is unique for every bounded continuous
f , by part (3) of Theorem 1.3.2). Then T z is a positive linear functional with T z 1 = 1,
whence by Riesz representationR theorem,2 there is a unique Borel probability measure z on such that T z f = f d z for all f C b (). That this definition agrees
with the earlier one follows immediately from Theorem 1.3.2.
2By the footnote on the previous page, is a compact subset of the sphere S{} and hence any
continuous function f on is also bounded. Of course, then f must be continuous at also to start
with.

1.5. BEURLINGS THEOREM

Example 1.4.2. On , z ( dt) = P ( z, e it ) dt where P ( z, e it ) =

kernel for the unit disk. On , z ( dt) =
find z ([1, 1])?

y
dt
( x t )2 + y2

1 1| z|2
2 | z e it |2

is the Poisson

where z = x + i y. On + can you

These examples are atypical. Usually we cannot find z exactly but only estimate it. The next section will introduce one such estimate which will be of great use
to us.
1.5. Beurlings theorem
If K is a compact subset of the plane we let K := {| z| : z K } be its radial projection.
Theorem 1.5.1 (Beurlings projection theorem and estimate). Let K be compact
and let V = \K and let V = \K .

(1) For any z we have Vz (K ) V

| z| (K ).

(2) If K is connected, 0 K and K , ;, then Vz () c 0

p
| z |.

P ROOF.
(1) By rotating we may assume that z = | z|. Let L = { xe i : x
R}. Apply Lemma 1.5.2 with the lines L /2 , L /4 , L /8 , . . . successively to
get sets K 1 = R /2 (K ), K 2 = R /4 R /4 (K 1 ), K 3 = R /8 R /8 (K 2 ) etc. with
Vj
z (K j ) increasing in j . Then, K j is confined to the sector {2 j arg z
Vj

2 j }. z (K j ) being the probability for a BM started at z to hit K j before

exiting , it can be proved that it converges to Vz (K ).

(2) Clearly K = [0, 1]. Although part (a) applies to compact subsets of the

disk, in the form Vz () Vz () it applies equally to K . Thus Vz (K )

p
[0,1]
[0,1]
r ([0, 1]) and one can explicitly compute that r () = 4 arctan r
p
c0 r.

Lemma 1.5.2. Let z and let K be a compact subset of . Let L be a line passing
through 0 and let R L denote reflection across L. Let K + be the part of K on the same
S
side as z and let K be the part of K on the other side. Let K := K R L (K + ). Set

V = \K and V = \K . Then, V (K ) V (K ).
z

P ROOF. [Bernt ksendal] Without loss of generality, we assume Im z > 0 so that

S
K + = K + and K = K . We just write A 0 for R L ( A ). If H := K + (K \K +0 ),
then H K while H = K . Thus, proving the lemma for H implies the lemma
for K . Thus we may assume that K +0 K = ;. In addition, one can show from
properties of Brownian motion that if K n K , then P z (W hits K n before ) converges
to P z (W hits K before ) and similarly for decreasing limits. Thus we may assume
that K + and K are compact and that the mutual distances between K +0 , K and
[1, 1] are strictly positive.
S
S
Let A = K 0 = K +0 K 0 and B = K 0 = K +0 K 0 and let I = [1, 1]. Start a BM at z
S
and run it till it hits K and successively record which of the sets A, I, K are hit.
For example, AI AK denotes the set of all paths that hit A first, then I , then A and
then K . Then,
(1.5.1)

V
z (K ) = P z (K ) + P z ( AK ) + P z ( IK ) + P z ( AIK ) + P z ( I AK ) + P z ( AI AK ) + . . .

The sum is over all strings in which A and I alternate a number of times and is

terminated by K . Similarly, we can write Vz (K ) by recording which of the sets B, I

1. COMPLEX ANALYSIS BACKGROUND

and K are hit. But now, B can only be followed by I , so we get

V
z (K ) = P z ( I K ) + P z (BI K ) + P z ( IBI K ) + P z (BIBI K ) + . . .

(1.5.2)

We show that each term here is equal to one of the terms in (1.5.1). More precisely,
Claim :

P z ( I K ) = P z ( IK ), P z (BI K ) = P z ( AIK ), P z ( IBI K ) = P z ( I AIK ), etc.

For clarity of explanation, we first show that P z ( I K ) = P z ( IK ). Indeed, write IK =

F
F
IK IK + and I K = IK IK +0 with the obvious meaning for the new symbols. For
any path IK + , let = inf{ t : t I } and define a new path ( t) := ( t)1 t +
0
R (( t))1 t> . Then, IK +
. Thus, P z ( IK + ) = P z ( IK +0 ), by the strong Markov property. That P z ( IK ) = P z ( I K ) follows.
More generally, say to show that P z ( IBI K ) = P z ( I AIK ), let V = \K and U =
V \ I and write
Z Z Z Z
U
V
U
V
P z ( AI AIK ) =
U
z ( dw1 )w1 ( dt 1 ) t 1 ( dw2 )w2 ( dt 2 ) t 2 (K )
A I

A I

and similarly
P z (BIBI K ) =

Z Z Z Z
B I B I

U
V
U
V
U
z ( dw1 )w1 ( dt 1 ) t 1 ( dw2 )w2 ( dt 2 ) t 2 (K ).

V
By considering reflection as before, V
t 2 (K ) = t 2 (K ) for any t 2 I . Further, as B =
R ( A ), the integrals are also equal (make a change of variables from w i to R (w i )).

Beurlings estimate is very useful to us later. Here is one example that we use
later.
Corollary 1.5.3. Let be a s.c. domain (other than the whole plane). Then, for any
z , if d = dist( z, c ), then for any x > 0 we have

c0
P z (W travels at least distance xd away from z before exiting ) p .
x
P ROOF. Let with | z | = d . Consider B(, xd ) with K = . By assumption that is s.c., it follows that K is connected and connects to B(, xd ) (if xd is
small enough, which is all we need to consider). Apply Beurlings estimate.

A standard application of harmonic measure in complex analysis is to PhragmenLindelf type theorems. This may be omitted and is included only for general interest.
Corollary 1.5.4. Let f be an analytic function on a neighbourhood of = {0 Re z
1} and suppose that | f ( z)| M R := exp{exp{CR 1 }} for some > 0. If | f ( z)| M on
{Re z = 0 or 1} and Then, | f ( z)| M for all z .
P ROOF. For R > 0, let S R := [0, 1] [R, R ] and let u( z) be a harmonic function
on S R whose boundary values are equal to log | f |. Since log | f | is subharmonic, we
get log | f ( z)| u( z) for any z S R . Fix z and let z (R ) denote the harmonic measure of the top and bottom edges of S R , as seen from z. Thus, log | f ( z)| u( z)
z (R ) log M R + log M , by the Brownian motion solution to the Dirichlet problem. We
need an estimate for z (R ). Divide S R into b2R c squares of side 1 and argue that
z (R ) e cR for some c > 0. Letting R we get | f ( z)| M .

1.6. SCHWARZS LEMMA AND ITS MANY VARIANTS

1.6. Schwarzs lemma and its many variants

Schwarzs lemma is the well-known statement that a holomorphic f :
that fixes 0 satisfies | f 0 (0)| 1 and | f ( z)| | z| for all z and strict inequalities
hold everywhere unless f is one-one and onto the disk. In which case it is a linear
fractional transformation f ( z) = (az + b)/( cz + d ) for some a, b, c, d with |a|2 | b|2 =
1. Pick generalized this statement to say that for any holomorphic f : , we
have
(1.6.1)

| f ( z) f (w)|
|1 f ( z) f (w)|

| f 0 ( z)|
| z w|
1
and

for any z, w
|1 zw|
1 | f ( z)|2 1 | z|2

with strict inequalities everywhere unless f is one-one and onto the disk. These
inequalities follow from Schwarzs statement by pre-composing (and post-composing)
f with linear fractional transformation that takes 0 to z (respectively, f ( z) to 0).
In the form given by Pick, Schwarzs lemma attains a deep geometric meaning.
On the disk , define a Riemannian metric (1| z|2 )2 ( dx2 + d y2 ) called the hyperbolic
R1
metric. This means that an curve : [0, 1] has length `() = 0 (1 | t |2 )1 | t | dt.
The length minimizing curves are called geodesics and they may be seen to be exactly
arcs of circles that intersect at right angles (this includes diameters of which
are arcs of circles of infinite radius).
If we call each z a point and each geodesic a line, then this system of points
and lines satisfy the axioms of the hyperbolic geometry discovered by Bolyai, Gauss
and Lobachevsky. The axioms are the same as that of Euclidean geometry, except for
the parallel postulate, which is replaced by a statement that given a line and a point
not on it, there exist at least two lines through the point that do not intersect the given
line. We say that the disk with the hyperbolic metric forms a model for hyperbolic
geometry.
Exercise 1.6.1. Show that with the metric y2 ( dx2 + d y2 ) is isometric to the
hyperbolic metric on . What are the geodesics?
Returning to Schwarz-Pick lemma, the inequalities 1.6.1 state precisely that any
holomorphic map from the disk to itself is a contraction of the hyperbolic metric. And
if it is not strictly contracting even at one point, then it must be an isometry of the
disk, in other words a linear fractional transformation that maps the disk injectively
onto itself. Here are some variants of Schwarzs lemma that we shall need.
Lemma 1.6.2.

(1) Let f : be holomorphic. Then,

f ( z ) f ( w) | z w|
| f 0 ( z)|
1

and

f ( z ) f ( w) | z w|
Im f ( z) Im z

with strict inequalities unless f ( z) = (az + b)/( cz + d ) for some a, b, c, d R

with ad bc = 1.
(2) Let f : be any holomorphic function that extends analytically to, and
fixes two distinct points a, b . Then, f 0 (a) f 0 ( b) 1 with equality if and
only if f is a hyperbolic isometry of .
(3) Let f : be holomorphic function that extends analytically to, and fixes
. Suppose f ( z) = z + O (1) as z . If f extends analytically to some x R
and f ( x) R, then | f 0 ( x)| .
P ROOF.

(1) Let ( z) =

z i
z+ i

and apply Schwarz-Pick to f 1 .

1. COMPLEX ANALYSIS BACKGROUND

(2) Derive this one from part (3) using a linear fractional transformation that
maps a and b to 0 and . We have not omitted absolute values by mistake.
Both f 0 (a) and f 0 ( b) must be positive as f maps (a, ) and ( b, )
into .
(3) must be positive as f maps into . Without loss of generality take
x = 0 = f ( x) and write f ( z) = z + O ( z2 ) as z 0 so that f 0 (0) = which is
also positive as f maps into near 0. Now take z = i y and w = iv and
apply part (1) to get

i y i v + O (1) | i y iv + O (1)|

.
i y + i v + O (1) | i y + iv + O (1)|
y

The left hand side is 1 2 v + O ( y2 /v2 ) while the right hand side is 1 2 v +
O ( y2 /v2 ) from which it follows that as claimed.

1.7. Half-plane capacity

Definition 1.7.1. For a hull K , Theorem 1.2.3 yields a unique conformal map K :
H \K such that K ( z) = z + a(zK ) + O ( z2 ) as z . The number a(K ) is called
the half-plane capacity of K as seen from infinity.
2

Example 1.7.2. If K = r + , then K ( z) = z + rz + O ( z2 ), hence a(K ) = r 2 . If K =

p
2
[0, it], then K ( z) = z2 + t2 = z + 2t z + O ( z2 ) whence a(K ) = t2 /2.
Lemma 1.7.3. We have the following properties of half-plane capacity.
(1) If K 1 K 2 are hulls, then a(K 1 ) a(K 2 ) with equality if and only if K 1 = K 2 .
(2) a( rK + t) = r 2 a(K ) if r 0 and t R.

S
(3) Let K := inf{ t : Wt K R}. Then, a(K ) = lim yE i y Im WK .
y

P ROOF.
(1) We saw that a(K ) 0 with equality if and only if K = ;. Now,
let K 3 := K 1 (K 2 \K 1 ). Then, K 2 = K 3 K 1 from which it follows that
a(K 2 ) = a(K 1 ) + a(K 3 ) a(K 1 ) with equality if and only if K 3 = ; which is
equivalent to K 1 = K 2 .

2
)
2
(2) True, since rK + t ( z) = r K zr t + t = ( z t) + r za(K
t + O (( z t) ) + t = z +
r 2 a( K )
+ O ( z2 ).
z
Im{K ( z) z} is

a bounded harmonic function

on \K and equal to Im z
S
for z K R. By Theorem 1.3.2 we have E z Im WK = Im z Im K ( z). Set

z = i y, multiply by y to get yE i y Im WK = y2 y y a(yK ) + O ( y2 ) a(K )

as y .

R
Exercise 1.7.4. If K is a hull with r (K ) < 1, then a(K ) = 2 E e it [Im B ] sin t dt.
(3)

With the probabilistic interpretation, we can estimate a(K ) using Brownian motion
hitting probabilities which in turn are estimated by Beurlings estimate. Thus we
have the following fact.
Theorem 1.7.5. Let K 1 , K 2 be two hulls.
S
(1) If K 1 K 2 and every point of (\K 2 ) (which may be smallerpthan K 2 R)
S
is within distance of K 1 R. Then a(K 1 ) a(K 2 ) a(K 1 ) + c 0 r (K 2 )3/2 .
(2) K 1 , K 2 general with r (K i ) r . If every point of (p\K 2 ) is within distance
S
of K 1 R and vice versa, then |a(K 1 ) a(K 2 )| c 0 r 3/2 .

1.8. A HEURISTIC DERIVATION OF LWNERS EQUATION

P ROOF.
(1) We know that a(K 1 ) a(K 2 ). To show the second inequality,
start Brownian motion W at i y and let 2 1 be the hitting times of
S
S
S
(2 r 2 ) R, of K 2 R and of K 1 R, respectively.
Let A := {|W | = 2 r 2 }. If A c occurs, then W R and 1 = 2 = and
hence Im W2 Im W1 = 0. On the event A := {|W | = 2 r 2 }, we may deduce
p

from corollary 1.5.3 that P Im W2 Im W1 > x| A c 0 / x for x < r 2 and

equals 0 for x > r 2 . This is because W2 is within distance of some point
S
in K 1 R.
Further, P i y ( A ) r 2y as y , hence using the probabilistic interpretation of half-plane capacity, a(K 2 ) a(K 1 ) is the limit as y of

yE i y Im W2 yE i y Im W1
= yP i y ( A ) E i y Im W2 Im W1 A
Z r2 p

c0 r 2
p dx = c 0 1/2 r 3/2
2 .
x
0
S
(2) Immediate upon applying the first part to the pairs K 1 K 1 K 2 and K 2
S
K1 K2.

Example 1.7.6. Let K t = { e i : 0 < t} for t < 1 and K 1 = + := { z : | z| 1}.

Then, a(K t ) is continuous in t, including t = 1. Had our condition in part (1) of
S
Theorem 1.7.5 said K 2 R instead of (\K 2 ), continuity at t = 1 would not have
followed!
Suppose is a simple curve in with 0 R. Then, \[0, t] has a unique unbounded component whose complement we denote by K t . Then K t is an increasing
family of hulls. Again, a(K t ) is continuous in t. This too follows easily from Theorem 1.7.5 because (\K t ) [0, t] and is continuous. A special case is when is a
simple curve with t R for any t > 0 and 0 R. In this case, K t = [0, t].
Remark 1.7.7. There is a well-known quantity called logarithmic capacity. There
are many ways to define it, but for a connected compact set K , its capacity c(K )
turns out to be equal to |0 ()| where is a Riemann map from \K onto := \
with () = . Here 0 () is the reciprocal of the coefficient a in the expansion
( z) = az + b + cz1 + . . . near .
S
If K is a hull in the upper half-plane, is a(K ) related to c(K ) or perhaps c(K K )?
S
If maps \K onto \+ , then it extends to a map from \(K K ) onto . The
expansion at may be written as K ( z) = az + b + cz1 + . . .. Then K ( z) = a1 (( z) +
( z)1 ) b/a. This has the expansion K ( z) = z + a1 ( c + a1 ) z1 +O ( z2 ) which shows
S
that a(K ) = a1 c+a2 . Of course c(K K ) = a1 . There does not seem to be any simple
S
relationship between a(K ) and c(K K ).
1.8. A heuristic derivation of Lwners equation
Let t be a simple curve with (0) R and ( t) if t > 0. Let g t be conformal
maps from \[0, t] . We want to understand how the maps g t vary with t, the
ultimate goal being to derive Lwners differential equation

g t ( z) =

a t
g t ( z) U t

where a t = a([0, t]), U t = g t ( t ) and a t , g t denote time ( t) derivatives. Among several issues raised by this statement, we have addressed a few. For example U t is
well-defined because t corresponds to only one prime end in \[0, t]. What about

10

1. COMPLEX ANALYSIS BACKGROUND

differentiability of a t ? Theorem 1.7.3 shows that a t is strictly increasing and Theorem 1.7.5 shows that a t is continuous. Therefore, can always be reparameterized
so that a t is smooth. In particular, if we set s = a1 (2s) , defined for s < a(), then
a([0, s]) = 2 s. is called the natural parameterization of .
Assume now that (0) is in its natural parameterization, that is a t = 2 t. Fix t
and let s = g t ( t+s ). If G s = [0,s] , then for any h > 0we have g t+h = G h g t . From
this, it is clear that is also a simple curve in its natural parametrization (see the
proof of the firstppart of Lemma 1.7.3). If we heuristically replace [0, h] by a vertical
h] having the same half-plane capacity 2 h, then by Example 1.2.5,
slit [U t ,U t + i 2 p
we get g t+h ( z) ( g t ( z) U t )2 + 4 h. Differentiate by h and set h = 0 to get
2
g t ( z) =
g t ( z) U t
which is exactly Lwners equation! If you have been able to calculate the conformal
maps for oblique slits asked for in Example 1.2.5, then check that replacing [0, h]
by any oblique slit with the same half-plane capacity 2 h will also result in the same
differential equation.
The gaps in this heuristic are that we only calculated the right derivative of g t
above and that we did not justify how to replace [0, h] by a slit. Both these issues
are addressed by proving several estimates in the next section about how conformal
maps K differ if the hulls K differ only a little. Then we put together the proof in
section 1.10.

1.9. CLOSENESS OF CONFORMAL MAPS

11

1.9. Closeness of conformal maps

We develop some required estimates in this section. The first theorem makes
precise in what sense the approximations K ( z) z or K ( z) z + a(K ) z1 hold. The
last one gets an estimate for how much a set near K can be blown up by K .
Theorem 1.9.1. Let K be a hull and K as before. Then, |K ( z) z| 3 r (K ) for all
z \K .
P ROOF. By scaling it suffices to assume that r (K ) < 1 and prove that |K ( z) z|
3 for all z \K . For any fixed x 1 (or x 1) we claim that K ( x) is increasing
(respectively, decreasing) in K , as long as r (K ) < 1. Assuming this claim, apply it to
the hulls ; K + to get

x K ( x) x +

1
1
for x 1 and x K ( x) x + for x 1.
x
x

In particular, |K ( x) x| 3 for | x| 1. For z + \K , we must have K ( z)

[K (1), K (1)] [2, 2]. Thus | zK ( z)| 3. Thus we have shown that lim sup |K ( z)
z| 3 whenever z (\K ) (note that the limit value is 0 as z ). By maximum
modulus principle, |K ( z) z| 3 for any z \K .
It remains to prove the claim that K ( x) L ( x) for x 1 if K L with r (L) < 1.
If J = K (L\K ), then L = J K . Hence it suffices to show that K ( x) x for
a single hull K . The map K ( z) K ( x) + x fixes x and and has derivative 1 at
1
. Applying part (3) of Lemma 1.6.2 to
we get 0K ( x) 1. For large y, K ( y) =
K
y + a(K ) y1 + O ( y2 ) > y. Therefore K ( x) > x for all x > 1.

Theorem 1.9.2. If K is a hull, then K ( z) z a(zK ) c 0 a(K| z)|r2(K ) whenever | z| >

16 r (K ).
The proof we give is exactly as in Lawler. Later we describe an attempt at a
slightly different reasoning that did not quite work. Do let me know if you see how
to fix it.
P ROOF. Scale and assume that r (K ) = 1. Let g( z) = K ( z) z Ka z and let v =
Im g. By harmonicity of v and Theorem 1.3.2, we write

v( z) = E z [Im{K (W ) W }] a(K ) Im(1/ z) = E z [Im W ] a(K ) Im(1/ z)

S
where is the hitting time of K R by the Brownian motion W . If denote the hitting
S
time of R, then by the strong Markov property, we can write

v( z) = E z E e it [Im W ] a(K ) Im(1/ z)

Z
2
=
E e it [Im W ] p( z, e it ) sin t dt

where p( z, e it ) is the density of W on and by writing a(K ) as in Exercise 1.7.4. In

Lemma lem:hittingofcircle we calculate p( z, e it ) explicitly and show that p( z, e it ) =
Im(1/ z) 2 sin t (1 + O (1/| z|)). Hence, we get
|v( z)| c 0 Im(1/ z)

1
| z|

Z2
2
Im( z)
E e it [Im W ] sin tdt c 0 a(K )
.

| z |3
0

12

1. COMPLEX ANALYSIS BACKGROUND

Now, consider |w| 2. Then the disk (w, |w|/2) is contained in {| z| > 1} and
hence, we can use Poisson integral formula to write for |w0 w| < R := |w|/2,
Z2
R 2 |w0 |2 dt
v(w ) = v(w + R e it ) 0
|w R e it |2 2
0

from which we differentiate with respect to x (or y) and set w0 = w and use the bounds
on v (that holds for all t since |w + R e it | |w|/2 1). We get

1 Im w
1
max |v x (w)|, |v y (w)| c 0 a(K )
c0
.
3
R | w|
| w |3
Since v = Im g, we get | g0 (w)| = |v y (w) + iv x (w)| c 0 a(K ) |w1|3 . Note that g() = 0.
Therefore, integrating g0 along a path from z to we get
Z
Z
a( K )
| z + it|3
.
t c0
| g( z)| | g0 ( z + it)| dt c 0 a(K )
d
| z |2
0

This completes the proof.

We used the following lemma in the proof.

S
Lemma 1.9.3. Let | z| > 1 and := inf{ t : Wt R}. Let p( z, e it ) be the density of W
conditional on W . Then, p( z, e it ) = Im(1/ z) 2 sin t (1 + O (1/| z|)) for all t [0, ].

P ROOF. Let | z| > 1. For any arc I , the harmonic measure z ( I ) is the value at
S
S
z of the harmonic function which has boundary values 1 on I and 0 on R {} (\ I ).
By conformal invariance of harmonic functions, we can map it forward by z z + 1/ z
which takes to [2, 2]. By knowledge of the Poisson kernel on , the hitting density
w
of W , started from w = z + 1/ z is given by p( s) = |Im
for s R. Pulling it back to
w s |2
the original problem, we find that
Im( z + z1 )
2 sin t
p( z, e it ) =

z + z1 2 cos t 2
which can easily be seen to be equal to Im(1/ z) 2 sin t (1 + O (1/| z|)).

Remark 1.9.4. Taking z = i y in Lemma1.9.3 and letting y % , in conjunction with

part (3) of Lemma 1.7.3 solves Exercise 1.7.4
S
Theorem 1.9.5. Let K be a hull and let : [0, 1] be a curve such that (0) K R
and ( t) \K for t (0, 1]. Let d be the diameter of and let ` := sup
t Im t .
Assume that d 10` . Then, there exists a constant c 0 such that dia K ()
p
c 0 d ` .
The main idea in the proof is to replace diameter by a conformally invariant
quantity that is comparable to the diameter of a set. The conformally invariant
quantity we use will be the harmonic measure. For instance, the length of an interval
on the line can be estimated by the hitting probability by a Brownian motion started
far off, that is, if = inf{ t : Wt R}, then lim yP i y (W [a, b]) = 1 ( b a). The following
y

exercise provides the needed generalization of this.

Exercise 1.9.6. Then there exists c 1 , c 2 such that for any connected hull K we have

c 1 dia(K ) lim yP i y (W [a, b]) c 2 dia(K )

y
S
where is the hitting time of K R by W .

1.10. CHORDAL VERSION OF LWNERS DIFFERENTIAL EQUATION

13

P ROOF. [Proof of Theorem 1.9.5] The idea is as follows. Let W be a Brownian

S
motion started at i y for a large y and let be the hitting time of K R. We use an argument (analogous to what is needed for the upper bound in Exercise 1.9.6) to bound
the probability that W hits before time . This probability is the harmonic measure
of as seen from i y in \K . By conformal invariance of harmonic functions, it is
the same as the harmonic measure of K () as seen from K ( i y) in . Then using
the lower bound in Exercise 1.9.6, we get the desired bound for dia(K ()).
S
S
Now for details. Let L = K ( 0 , 10` ) and let be the hitting time of L R
by W . It s easy to prove that P i y (W 0 + ` ) c 0 y1 ` (left as exercise). If
W 0 + ` , then there is no hope for W to hit before time . On the other hand,
p
if W = 0 + ` then we claim that P (W hits before time ) c 0 d /` . This
follows by Beurlings estimate but with a small twist. See Exercise 1.9.7.p
Puting everything together, we get yP i y(W hits before time ) c 0 d ` . We
get the same bound for yPK ( i y) (W hits K () before time ) by conformal invarince
1
of harmonic functions. Since p
K ( i y) = y + O ( y ), using the lower bound in Exercise 1.9.6 we get dia(K ()) c 0 d ` .

Exercise 1.9.7. Let K be a connected set in the plane connecting 0 to . Then for
p
any , we have P (W hits before hitting K ) c 0 .
1.10. Chordal version of Lwners differential equation
Let be a simple curve in with 0 = 0 and t for t > 0. Let g t : \[0, t]
be the unique conformal map such that g t ( z) = z + azt + O ( z2 ) as z . Here

a t = a([0, t]) is the half-plane capacity. Then, by Theorem 1.7.5 a t continuous.

Lemma 1.7.3 asserts that it is also strictly increasing, as is simple. Thus, there
is a unique way to reparameterize so that a t = 2 t for all t. We call this the standard parameterization.

Theorem 1.10.1 (Lwner). Let be a simple curve in standard parameterization and

g t the associated conformal maps normalized hydrodynamically. Then, U t := g t ( t )
is continuous in t and t g t ( z) = g t ( z2)Ut for z \[0, t].
P ROOF. Let K t = [0, t]. Fix T > 0 and consider t < s [0, T ] throughout. Let
h t,s ( z) = g s gt 1 . In earlier notation g t = K t and h t,s = K t,s where K t,s := g t (K s \K t )
(we translate the hull so that it is close to the origin). Then,

k g t g s ksup\K s = sup | h t,s ( z) z| : z K t,s

3 r (K t,s U t )
(by Theorem 1.9.2 applied to K t,s U t )
p
C ,T dia([ t, s])
(by Theorem 1.9.5).

From this it immediately follows that t U t is continuous (see Remark 1.10.2).

Next, we observe that if z \K s then

g s ( z) g t ( z) 2( s t) = h t,s (w) w 2( s t)
(where w = g t ( z))
g t ( z) U t
w Ut
( s t)dia(K t,s )
(1.10.1)
C ,T
| g t ( z ) U t |2
where the last inequality is implied by Theorem 1.9.2 if | g t ( z) U t | > 2 r (K t,s ).
If we fix and z \K s and let t s, then the last condition holds for t close
enough to s. Divide (1.10.1) by t s and let t s to get g s ( z) = g s ( z2)Us . Here we used

14

1. COMPLEX ANALYSIS BACKGROUND

the continuity of g t ( z) and U t in t. Similarly, we can fix z \K t , divide by s t and

let s t in (1.10.1). The condition | g t ( z) U t | > 2 r (K t,s ) holds again for s close enough
to t and we get g t+ ( z) = g ( z2)U . Thus we have proved Lwners differential equation

g t ( z) =

2
g t ( z)U t .

Remark 1.10.2. In the coursep

of the proof we claimed that the continuity of U t in
t follows from k g t g s k C ,T dia([ t, s]). As a matter of fact, this inequality can
be used to prove the existence of U t := lim g t ( z) as z t in \K t without having
to invoke the theorem on boundary values of conformal maps. We leave this as an
exercise or to look up the proof of Lemma 4.2 in Lawlers book.
Remark 1.10.3. Suppose K t is a one-parameter family of hulls such that K t ( K s
for t < s and let g t be the hydrodynamically normalized conformal maps from \K t
T
onto . Assume in addition that >0 K t,t+ = {U t } for every t. We then say that the
family of hulls is right continuous. It is clear that a right continuous family of hulls
can be reparameterized so that a t := a(K t ) = 2 t. Then the proof of Theorem 1.10.1
goes through exactly as stated above. In particular, U t is continuous in t and g t ( z) =
2
g t ( z)U t for any z \K t and any t.
U is called the driving function.
1.11. Continuously growing hulls
Suppose K t is a one-parameter family of hulls such that K t ( K s for t < s and let
g t be the hydrodynamically normalized conformal maps from \K t onto . Define
K t,s as in the proof above, and assume that for every t there is a point U t R such
T
that >0 K t,t+ = {U t }. We then say that the family of hulls is right continuous with
driving function U .
From a(K s ) = a(K t ) + a(K t,s ) and Theorem 1.7.5 we see that a(K t ) is continuous.
Hence a right continuous family of hulls can be reparameterized so that a t := a(K t ) =
2 t.
In this case, one half of the proof of Theorem 1.10.1 goes through exactly as
stated above.
Proposition 1.11.1. Let K be right-continuous with driving function U . Assume
that the K is in its natural parameterization. Then U t is right continuous in t and
g t+ ( z) = g t ( z2)Ut for any z \K t and any t.
P ROOF. As usual it suffices to check this at t = 0 with U0 = 0. For h > 0 and z
K 2h we have | g h ( z)| | g h ( z) z| + | z| which is bounded by 3 r (K h ) + r (K 2h ) which goes
to zero as h 0. Since U0 = 0 and Uh K h,2h = g h (K 2h ), it follows that Uh U0 0
as h 0. This shows right continuity of U . The right derivative of g t can be found
exactly as before.

Exercise 1.11.2. Let be a curve in and let K t be the hull generated by [0, t],
that is \K t is the unique unbounded component of \[0, t]. Assume that K t , K s
for any t < s.3 Then K t is strictly increasing and right continuous.
The left continuity of U and the left derivative of g t dont follow automatically.
For example, if as in the above exercise crosses itself, the U jumps. To be specific,
3If K is an increasing family of hulls, it can be reparameterized so that it is strictly increasing. But
t
if K is also generated b a curve , then it may not be possible to reparameterize so that K t is strictly
increasing in t.

1.12. LWNER EVOLUTION IN REVERSE

15

take t = 2 it for t 1 and t = ie i( t1) for 1 t 1 + 2. At t 0 = 1 + , the curve crosses

itself. There are two prime ends corresponding to the boundary point 1 of \K 1+
and U t0 and U t0 are the limits of g t0 ( z) as z 1 along these two prime ends.
Definition 1.11.3. If K is right continuous with driving function U , and t U t is
continuous, then we say that K is a continuously growing family of hulls.
Since this definition puts in exactly what is needed to make the proof of Theorem 1.10.1, we get
Proposition 1.11.4. Let K be a continuously growing family of hulls with driving
function U . Assume that the K is in its natural parameterization. Then g t ( z) =
2
g t ( z)U t for any z \K t and any t.
1.12. Lwner evolution in reverse
We start with a converse to Proposition 1.11.4.
Theorem 1.12.1. Let U t be a continuous real valued function with U0 = 0. Then
there exists a unique continuously increasing family of hulls K that is naturally
parameterized and whose driving function is U .
P ROOF. Consider the differential equation x t = 2/( x t U t ) in the complex plane.
For each z, let T z be the maximal time for which the solution exists, starting with
the initial condition x0 = z. This unique solution we denote by g t ( z). Thus g 0 ( z) = z
and t g t ( z) = g t ( z2)Ut for t [0, T z ). Clearly, T z = inf{ t : g t ( z) = U t }.
Let K t := { z : T z t}. By continuity of solutions in the initial conditions, K t
is closed in . If kU k t = max{Us : s t}, then the bound on the speed | g t ( z)| = 2/kU k t
implies that K t is bounded. Further, if z K t , then g s ( z) K t for any s < t. Since
S
g t ( z) = U t , this shows that K t R is connected. Hence \K t is simply connected.
Therefore K t are hulls, and obviously (weakly) increasing.
Now we claim that g t is a bijection from \K t onto . By definition for any
z \K t , the solutions exist for [0, t], hence g t is well-defined on \K t . The timeIm y
reversed differential equation is y t = Ut 1 yt . For this, Im y t = |U yt |2 which shows that
t
t
if y0 , the the solution exists for all time and yt . This provides the inverse of
g t on .
Next we show that g t is analytic. We just sketch the idea. Let x t = F ( x t , t) be a
differential equation with F holomorphic in the first variable and continuous in the
second. Recall the method of Picards iteration to get a solution (locally). There we
start with X 0 ( t, x) = x for x and define
Z t
X n+1 ( t, x) = x +
F ( X n ( s, x), s) ds.
0

All this will be done locally. Inductively it is clear that each X n is analytic in x.
Picards iterates converge locally uniformly to the solution of the ODE. Therefore
the solution is also holomorphic.
Thus we have shown that g t are conformal maps from \K t onto . For fixed t
and large z, we can argue (How?) that g t ( z) = 2/ g t ( z) + O ( g t ( z)2 ) and hence g t ( z) =
z + 2zt + O ( z2 ). Thus K is in its natural parameterization.

Now we are ready to define SLE!

16

1. COMPLEX ANALYSIS BACKGROUND

Definition 1.12.2. Let B be a standard one-dimensional Brownian motion. For

p
0, let K t be the increasing family of hulls with driving function U t := B t as assured
by Theorem 1.12.1. The family of random hulls K is called SLE().
It is a non-trivial theorem of Rohde and Schramm that SLE()is generated by
a random curve as in Exercise 1.11.2. Usually that curve is called SLE(). For
many results, it suffices to think of SLE()as a family of increasing hulls, but we
shall anyway assume the result of Rohde and Schramm without proof and simply
talk of SLE()curves. Sometimes, the end result, K (or ) is what we care about.
It is a hull that connects 0 to (if K , then K would be bounded and hence
a t = 2 t would be bounded above!).