MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I'm having a brain freeze.

Let $B$ be the space of complex valued measurable functions on the unit disk in the complex plane with essential supremum less than 1. Then, the universal Teichmuller space $T$ can be though of as a quotient space of $B$.

I vaguely recall reading somewhere that for every point $p \in T$ has a unique representative of the form

$\mu(z) = k \frac{\bar{q}}{q}$, where $q$ is a holomorphic function on the unit disk.

First of all, am I crazy and am remembering something that's completely off?

Secondly, if there is a statement like that, where should I be looking?

share|cite|improve this question
  1. Strebel discovered (in 1962) that Teichmüller's existence theorem fails for quasiconformal maps of the unit disk. Uniqueness fails too for points in $T(D^2)$ which do not have Teichmuller representatives. See Strebel's examples in the book by Gardiner and Lakic, "Quasiconformal Teichmüller Theory," p. 177-178.

  2. On the other hand, the uniqueness theorem in certain sense does hold (Theorem 5, Chapter 4 of the same book). More precisely, if a quasiconformal map $f: D^2\to D^2$ is a Teichmüller map, i.e., it has Beltrami differential of the Teichmüller form $t \bar{\phi}/|\phi|$ (where $\phi$ is holomorphic), then $f$ is the unique extremal quasiconformal map in its equivalence class $[f]\in T(D^2)$. In particular, $[f]$ contains no other Teichmüller maps.

  3. On third hand, Strebel proved in 1976 ("On the existence of extremal Teichmueller mappings") that for $[f]\in T(D^2)$ given by, say, smooth boundary values, the Teichmüller existence theorem does hold.

share|cite|improve this answer
Thank you for the reference. I'm actually looking at a beltrami differential of a very particular form: mu(z) = k \bar{q}/q, where q is the Koebe function precomposed with the map exp(2pi i z). (So it is a beltrami differential on a punctured disk.) Theorem 5 doesn't really apply in my case, but Chapter 9 of the reference you gave me (which contains the pages you gave) seems to be the right place to start. Again, thank you very much for your answer. – BrainDead Apr 3 '12 at 2:32
@BrainDead: You are welcome. Judging by your description, your $q$ is holomorphic on the upper half-plane (composition of two holomorphic functions), so Theorem 5 would apply. Maybe the problem is that your quadratic differential is unbounded: I forgot if Theorem 5 has boundedness requirement on $q$. – Misha Apr 3 '12 at 3:44
Right, by punctured disk, I meant the strip [0,1] x [0,infinity] in the upperhalf plane. And the problem is that $\phi = q^2$ does not have a finite $L^1$ norm. Theorem 3 in Chapter 9.3 (pg.183) seems to be exactly what I need. – BrainDead Apr 3 '12 at 4:59

I think you probably referred to this fact: you can find for your point $p \in T$ a representative in the form of a Beltrami differential $\mu$ expressed as $\mu=k\frac{\bar{\phi}}{\vert \phi \vert}$, with $\phi$ a quadratic differential. But I doubt about the uniqueness, as you can always multiply $\phi$ by some positive constant...

Hubbard's book on Teichm\üller theory contains certainly all of that. Also, I think that Lehto's book has an entire chapter on Universal Teichm\üller space.

share|cite|improve this answer
@Sylvain: The question is about uniqueness of $\mu$ rather than uniqueness of $\phi$. Once you know $\mu$, the holomorphic quadratic differential $\phi$ is determined uniquely up to a constant positive multiple. This is the usual ambiguity in Teichmüller theory which is usually remedied by assuming that $\phi$ has unit norm (for some choice of the norm on the space of quadratic differentials). – Misha Mar 31 '12 at 1:22
Thanks for the comment! together with your answers, the whole thing is much clearer now... – Sylvain Bonnot Apr 1 '12 at 16:05

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.