It is known that each complex structures on $\mathbb{R}^2$ is biholomorphic to either $\mathbb{C}$ or the open unit disk $\Delta$.

One can continuously deform one complex structure to the other as is for example done in Winkelmann - Deformations of Riemann surfaces (page 3).

My question is

Can this deformation be taken to be holomorphic on the deformation parameter? That is, does there exist a non-trivial complex analytic family $M \to D$ where $D \subset \mathbb{C}$ is a small disk, the central fiber is biholomorphic to $\mathbb{C}$, and the generic fiber is biholomorphic to $\Delta$?

Note that all the theorems that assure complex analytic triviality of deformations when $H^1(X,TX)$ vanishes, use the hypothesis that $X$ is compact.

  • 5
    $\begingroup$ How about $M=\{|z w|<1 \}\subset D\times \mathbb{C}$ with projection to the first coordinate? $\endgroup$ – user_1789 Nov 29 '19 at 9:13
  • $\begingroup$ Beautiful! I think it works. If you write it in an answer with a small argument it will be the accepted solution $\endgroup$ – Paul Nov 29 '19 at 13:25

The answer is no. This follows from the so-called $\lambda$-lemma of Sullivan, Mane Sad and Lyubich: Let $D$ be a disk, $C$ the complex plane and $A$ any set in C. Let $f:D\times A\to C$ be a function with the following properties:

  1. $\lambda\mapsto f(\lambda,z)$ is holomorphic for every $z\in A$.

  2. $z\mapsto f(\lambda,z)$ is injective for every $\lambda\in D$,

  3. $z\mapsto f(0,z)=z$ for every $z\in A$.

Then for every $\lambda\in D$ the map $z\mapsto f(\lambda,z)$ is quasisymmetric (=quasiconformal if $A$ is open)

Since the plane is not conformally equivalent to a disk, this implies that the answer to your question is negative.

Ref. MR0732343 Mañé, R.; Sad, P.; Sullivan, D., On the dynamics of rational maps, Ann. Sci. École Norm. Sup. (4) 16 (1983), no. 2, 193–217.

| cite | improve this answer | |
  • 1
    $\begingroup$ I am sorry but I don't know how to apply your answer to my situation. You are looking at my $D$ to be embedded in your C? And then my family $M$ would be your product? Observe that my central fiber is not a disk by hypothesis. Also the conclusion of the $\lambda$-lemma is that the map admits a quasiconformal extension $\endgroup$ – Paul Nov 29 '19 at 5:01
  • 1
    $\begingroup$ Your map $f$ is my family $M \to D$ ? $\endgroup$ – Paul Nov 29 '19 at 5:18
  • 1
    $\begingroup$ $1$ is not satisfied in my situation. $\endgroup$ – Paul Nov 29 '19 at 5:40
  • 1
    $\begingroup$ I probably misunderstood what is a "complex analytic family". Can you define it? On my opinion "a deformation depending on a complex analytic parameter" is exactly what satisfies 1,2,3. $\endgroup$ – Alexandre Eremenko Nov 29 '19 at 13:39
  • 1
    $\begingroup$ What does that mean? That it was my fault that your answer is wrong because my question was too easy? I don’t mind admitting that my question was “trivial” (whatever that means) or that I didn’t know something (that I learned today) or that I’m more ignorant than a lot of people. I asked this question because I am no expert and I’m learning deformation theory on my own with no experts around me. $\endgroup$ – Paul Nov 29 '19 at 19:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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