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

A Morse function $f: M \rightarrow \mathbb R$ on a connected closed manifold $M$ is called $\mathit{perfect}$ with respect to the field $\mathbb F$ if all of the Morse inequalities are equalities, i.e. the number of critical points of $f$ with index $k$ coincides with the $k$-th Betti number of $M$ with respect to $\mathbb F$-coefficients for all $k$.

Now assume that $f: M \rightarrow \mathbb R$ is a perfect Morse function on a closed connected Riemannian manifold and that $N \subset M$ is a closed submanifold, such that $N$ contains all the critical points of $f$ and the restriction $f|_N$ is a Morse function on $N$. If in addition the gradient $\nabla f$ is tangent to $N$ along $N$, i.e. $\nabla f|_N \in \Gamma(TN)$, then $N$ is a union of flow lines and the critical points of $f|_N$ are exactly the critical points of $f$.

Now my question is: Is it possible to deduce perfectness of $f|_N$ from perfectness of $f$ in this setting, or are there additional conditions under which this is possible?

share|cite|improve this question
up vote 8 down vote accepted

Isn't there a counterexample with $(M,N)=(\mathbb CP^2,\mathbb RP^2)$?

share|cite|improve this answer
Ok, I see. Thank you. The problem is that the attaching map restricts to a map of degree 2. Is that a problem of orientability? Can one avoid problems like that by using $\mathbb Z/2\mathbb Z$-coefficients, for instance? – Stephan Wiesendorf Jan 10 '11 at 20:54
The fact that $\mathbb RP^2$ is not orientable is irrelevant, since I could just as well take $(\mathbb CP^3,\mathbb RP^3)$. – Tom Goodwillie Jan 11 '11 at 4:17
For that matter, I am pretty sure there is an example of a metric and perfect Morse function on $M=S^1\times S^3$ such some submanifold $N$ diffeomorphic to $S^1$ consists of the four critical points and four trajectories of the gradient flow. – Tom Goodwillie Jan 11 '11 at 4:21

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.