0
$\begingroup$

How do I prove the following assertion:

Let $\nabla$ be a connection on a riemannian manifold. $\nabla$ is compatible with the metric if and only if for all vector fields $X,Y,Z$ we must have:

$X\langle Y,Z \rangle = \langle\nabla_X^Y,Z\rangle+\langle Y,\nabla_X^Z\rangle$

$\endgroup$
  • $\begingroup$ What's your definition of "compatibility"? $\endgroup$ – Neal May 12 '12 at 15:24
  • $\begingroup$ For every $X,Y$ parallel vector fields along a curve we have: $\langle X,Y \rangle$ = constant, but suppose you know that the product rule can be applied to $\frac{d\langle X,Y \rangle}{dt}$ $\endgroup$ – Jr. May 12 '12 at 16:35
  • $\begingroup$ $\frac{d\langle X,Y \rangle}{dt}=\langle \frac{DX}{dt},Y\rangle + \langle X,\frac{DY}{dt}\rangle$ $\endgroup$ – Jr. May 12 '12 at 16:40
1
$\begingroup$

Look at the Proposition 3.2 and Corollary 3.3

$\endgroup$
  • $\begingroup$ actualy I was trying to understand that corollary but stuck, but after some time I understand :) $\endgroup$ – Jr. May 12 '12 at 19:06
  • $\begingroup$ Dear Jr. Knowing what have tried out precisely, would be helpful to those who want to answer your questions. $\endgroup$ – Ehsan M. Kermani May 12 '12 at 19:09
  • $\begingroup$ Do Carmo says at the beginning of the proof of Proposition 3.2: "It is obvious that the equation $$\frac d{dt} \langle X,Y \rangle = \langle \frac{DX}{dt},Y\rangle+\langle X,\frac{DY}{dt}\rangle, \quad t\in I$$ implies that $\nabla$ is compatible with $\langle \, , \, \rangle$." How exactly does this equation imply that $\frac d{dt}\langle X,Y\rangle=0$, so that the desired conclusion follows? (I'm self-studying this topic right now.) $\endgroup$ – New day rising Jan 24 '17 at 8:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.