Let $X$ be a Riemann surface with analytic boundary. Assume that $X$ has negative Euler characteristic. Then there exists a conformal hyperbolic metric $X$ such that $\partial X$ consists of geodesics (conformal in the sense of being compatible with the given complex structure on $X$).

Does anyone has a reference (or even better, a quick proof) of this result?

Edit: Let me state a stronger version. Suppose $X$ is embedded in an open Riemann surface $Y$. Then there exists a conformal hyperbolic metric on a neighborhood of $X$ in $Y$ such that $\partial X$ consists of geodesics. Is this true?

  • $\begingroup$ The 1992 book: P. Buser, Geometry and Spectra of Compact Riemann Surfaces? $\endgroup$
    – YCor
    May 10, 2023 at 15:07
  • $\begingroup$ The stronger version also holds: actually a neighborhood N of X (in the doubling of X) also embeds into Y. Indeed by the Schwarz reflection principle the embedding of X into Y extends to a holomorphic map from a neighborhood N into Y. This extension is an embedding, if N is sufficiently small. $\endgroup$
    – user_1789
    May 10, 2023 at 18:45

2 Answers 2


A good reference is

W. Abikoff, The real analytic theory of Teichmuller space, Springer, 1980. (Chap. II section 1).

The idea is that you construct the double: it is the result of gluing of your surface with its mirror image. This is a compact surface, it has a hyperbolic metric, and the restriction of this metric on the original surface is the hyperbolic metric with geodesic boundary.


See Theorem 1(b) in the following paper; it is enough that the boundary is smooth.

Osgood, B.; Phillips, R.; Sarnak, P. Extremals of determinants of Laplacians. J. Funct. Anal. 80 (1988), no. 1, 148--211.

  • $\begingroup$ But analyticity of the boundary is necessary. Consider, e.g., a pair of pants. If every pair of pants admits such a metric, then they would be isomorphic to a standard hyperbolic pair of pants, which already has analytic boundary. I guess a more precise way of stating what I'm looking for is that the hyperbolic metric should extend to a hyperbolic metric in a neighborhood of $X$ (after embedding $X$ in an open Riemann surface). $\endgroup$
    – Yuxiao Xie
    May 10, 2023 at 7:23
  • $\begingroup$ @YuxiaoXie: You are wrong (confusing different categories: Your question was in differential geometry category and now you are switching to a related, but different category of bordered Riemann surfaces). $\endgroup$ May 10, 2023 at 15:23
  • $\begingroup$ @MoisheKohan I don't understand why my question was in the differential category. I work on a given Riemann surface and I only consider conformal metrics on it. $\endgroup$
    – Yuxiao Xie
    May 10, 2023 at 16:30

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.