In Klingenberg's Notes, he makes the following definition:
$\Lambda M$ will be said to satisfy the condition (C) of Palais-Smale if:
Given a sequence $\{c_m\}$ on $\Lambda M$ satisfying:
(i) The sequence $\{E(c_m)\}$ is bounded
(ii) The sequence $\{||\text{grad} ~E(c_m)||_1\}$ tends to zero
Then $\{c_m\}$ has limit points and any limit point is a critical point of $E$
He then proceeds to prove $\Lambda M$ satisfies the condition (C). (OBS: $\Lambda M$ is the free loop space of a compact riemannian manifold)
But I think something is wrong... is it really limit point that he wants?
I think he means that $c_m$ has some convergent subsequence... and every subsequence limit is a critical point. Is that it?
He means that $c_m$ has some convergent subsequence.