Let $G$ be a topological group (or a Lie group). If $G^0$ is the identity component, we know that $G^0$ is a normal subgroup of $G$. But is it true that if $C$ is any connected component of $G$, there is $a\in G$ such that $C=aG^0$?
I tried to come up with some counter-example, to no avail (I'm low on imagination, I guess), and I couldn't prove it either.
Initially I thought about the problem for topological groups, but if the situation is somehow easier for Lie groups you can discuss this case. Help?