Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. 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

Given a Riemannian surface with nonnegative Gaussian curvature, the area of a ball of radius $r$ around any point has area at most $\pi r^2$. I have a simple proof of this in the Euclidean cone case (a surface which is flat except at a discrete set of cone points), which shows that all points in the ball can be reached with geodesics that do not pass through a cone point.

My question: who would have proved this first, and does anyone know a good reference for such? All of my searches find much more complicated results involving sectional curvature in higher dimensional manifolds.

share|cite|improve this question
For completeness, here's the link to our paper with a proof of the Euclidean cone case: – Geoffrey Irving Dec 5 '12 at 18:46

Spivak credits Diquet (1848) with the result that

$$K(p) = \lim_{r\to 0} 12 \frac{\pi r^2-A(r)}{\pi r^4}.$$

This comes from an elementary argument with the Taylor expansion for the metric in geodesic normal coordinates. (See the Appendix for Chapter 3 of Volume II of Spivak's magnum opus Differential Geometry. Also look at a brief entry on Wikipedia: "Bertrand-Diquet-Puiseux Theorem.")

share|cite|improve this answer
Out of curiosity, do you know if authors around this time ever considered the relationship of $A(r)$ to $K$ when $r$ is large? (P.S. welcome to math.SE!) – treble Dec 2 '12 at 23:00

Surely Gauss knew about this. Consider $S \subset \mathbb R^3$, a two dimensional oriented embedded surface. Recall that the Gauss map $G: S \to S^2$ is defined by $$G(p) = \mathbb n(p)$$ where $\mathbb n(p)$ is the unit normal of $S$ at $p$, translated to the origin (so that $\mathbb n \in S^2$, the unit sphere). Then the Gaussian curvature at $p$ effectively measures the signed area distortion of a small rectangle near $p$ under the map $G$. Thus the Gauss curvature $K$ is defined to be $$\textrm{det}(DG)$$ where $DG$ is the derivative of $G$, say, written with respect to a local parametrization of $S$ in some neighborhood of $p$. The hypothesis that $K := \textrm{det}(DG) \geq 0$ is thus saying something about area deformation. So it is not a long walk from here to the proof that you want (most proofs also involve one other concept: normal (or "geodesic") coordinates, also invented by Gauss). As far as a reference is concerned, consider looking at Ted Shifrin's great book Differential Geometry. I don't know if he proved the volume comparison result you wanted per se, but the "pieces" are certainly there.

Edit: After a little bit more thought, I guess it is more difficult than I thought to outright prove this from first principles based on what I said above. I don't know whether or not I've ever seen it worked out completely just for, say, surfaces in $\mathbb R^3$ (the more general results imply this case of course, but that is not what your question is about).

share|cite|improve this answer

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.