MathOverflow is a question and answer site for professional mathematicians. 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

I'm curious about the following question:

Given $(M^n,g)$ a closed Riemannian manifold, is there always a $C^\infty$ isometric embedding $F:(M^n,g) \to (\mathbb{S}^{m},g_{std})$ for $m$ large enough?


  • If we replace $(\mathbb{S}^{m},g_{std})$ by $(\mathbb{R}^m,\delta)$ for $m$ large enough the answer is "yes" by the $C^\infty$-Nash Embedding Theorem.

  • Alternatively, if you replace $C^\infty$ by $C^1$, I think the answer is also "yes" and one can simply use the same h-principle proof as for the Nash--Kupier theorem (this is claimed somewhere in Eliashberg and Mišačev's book on the h-principle, but I don't remember the exact place).

Closely related questions that I'd be curious to know about are (arranged in increasing order of difficulty, with the original question between 1 and 2):

  1. Is this true if we relax the target to be a sphere of any radius?

  2. Is there an integer $N = N(n)$ and constant $D > 0$ so that if $diam(M^n,g) \leq 1$ then for any $N(n)$-dimensional manifold of diameter at least $D$, $(V^N,g_V)$ we can always find an isometric embedding of $M$ into $V$?

  3. What happens to the last question if we drop the diameter restrictions?

share|cite|improve this question
I think this post already answers your questions:… – Matthias Ludewig Jun 13 '13 at 22:10
Thanks! I guess I'll change my question to ask for a reference then, as the answer in the linked question did not give one. – Otis Chodosh Jun 13 '13 at 22:35

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.