Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

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

In Dynamical Systems and Ergodic Theory by Pollicott and Yuri, there is an easy, one dimensional, fixed point theorem:

If $T$ is a continuous map on a closed interval $J$ so that $T(J)\supseteq J$, then $T$ has a fixed point.

Here the space is mapped onto itself, in contrast to the "usual" fixed point theorems where it is mapped into itself.

Out of idle curiosity, I wonder if it's true in higher dimensions. To be concrete, let $D=\{(x,y): x^2+y^2\leq 1\}$ be the closed unit disk in $\mathbb{R}^2$. Suppose that $T$ is continuous on $D$ and $T(D)\supseteq D$, does $T$ have a fixed point?

If $T$ is one-to-one, for instance, this follows by applying Kakutani's fixed point theorem to the set valued map $x\mapsto T^{-1}(\{x\})$ on $D$. But I'm not sure whether the result holds in general.

share|cite|improve this question
Upvote for "out of idle curiosity"! – Tim Mar 1 '12 at 20:25
Isn't the proof the same as for Brouwer's fixed point theorem? Project $z \in D$ to its "shadow" on $\partial D$ if you put a light source at $f(z)$. This would give a continuous map $D \rightarrow \partial D$ that fixes $\partial D$ if $f$ has no fixed points. – WimC Mar 1 '12 at 20:29 that doesn't work. It doesn't have to be continuous at $\partial D$. – WimC Mar 1 '12 at 20:36
Could we perhaps map $z \in D$ to both intersection points of the line connecting $z$ and $f(z)$ on $\partial D$. That would give a map $D \rightarrow \partial D \times \partial D$ and $\partial D$ would map to a non-contractible curve on this torus. – WimC Mar 1 '12 at 20:42
@WimC Thanks very much for your input. I am slower than you are, and I'm still processing the "no that doesn't work" comment. If you come up with a solution, please post it as an answer. Thanks! – Byron Schmuland Mar 1 '12 at 20:49
up vote 6 down vote accepted

Unfortunately, this is not true, as I convinced myself using a cork coaster not unlike these. A counterexample is given by a map that maps the upper third of the disk to the lower half, the lower third to the upper half, and the middle third to a band that connects the two outside the disk.

share|cite|improve this answer
You just beat me to it! – TonyK Mar 1 '12 at 22:07
Very nice! Coasters are indispensible math tools. – WimC Mar 1 '12 at 22:11
@joriki Thanks very much for this. I never would have thought of this on my own. – Byron Schmuland Mar 1 '12 at 22:50

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.