2
$\begingroup$

Assume that $D_n\subset D$, where $D$ is the unit disk is an increasing sequence of Jordan domains with smooth boundaries such that $\cup_{n=1}^\infty D_n=D$ and let $f_n: D \to D_n$ be conformal mapping such that $f(0)=0$ and $\arg(f_n'(0))=0$. Is $f_n$ uniformly convergent to $Id$.

$\endgroup$
2
  • 1
    $\begingroup$ You mean $f_n(0)=0$ and $\mathrm{arg}f_n'(0)=0$ don't you? $\endgroup$ Nov 7, 2013 at 21:51
  • 1
    $\begingroup$ @PietroMajer : Surely this is what is meant. I assumed it in my answer. $\endgroup$ Nov 7, 2013 at 21:51

1 Answer 1

5
$\begingroup$

The answer is no. Consider the following sequence $D_n$. Take the disc $|z|<1-1/n=r_n$, and remove from it the arc $\{ z:|z|=r_{n-1},|\arg z|<\pi-1/n\}$ and the interval $(r_{n-1},r_n)$. The result is a simply connected region $D_n$. This $D_n$ contains $|z|<r_{n-1}$, so the union of $D_n$ is the whole unit disc, and they are nested. Small deformation will make these regions Jordan, smooth, etc.

Evidently the convergence of the conformal maps $f:D\to D_n$ is uniform on compact subsets of $D$ (this follows from Caratheodory thm), but not uniform in $D$, because near the point $z=-1$ the function $f_n(z)$ takes values close to every point of the unit circle. (A small neighborhood of $-1$ blows up under $f_n$ to almost the whole ring $r_{n-1}<|z|<r_n$.)

$\endgroup$
2
  • $\begingroup$ Yes you right, I understand the point. Thanks. $\endgroup$
    – user36162
    Nov 8, 2013 at 7:40
  • $\begingroup$ You're right of course, I was confused and my answer was wrong. Thank you for pointing it out. Great answer, +1 ! $\endgroup$ Nov 8, 2013 at 12:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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