In the derivation of the BGV theorem paper:


There is the following relation (numbered (8)):

$\dfrac{d}{d s}\left[a^2(t) \dfrac{d \mathbf{x}}{d s}\right]=0$

It's very likely dumb but i don't see how they get that start from the FLWR metric and geodesic equation...

If anyone can help.


1 Answer 1


This can be seen from the variational principle $$ \delta \int\sqrt{-g_{\mu\nu}dx^\mu dx^\nu}=0 $$ and eq.(7) in the paper, i.e. $u_\mu u^\mu=-1$. From the Lagrange equations $$ \frac{d}{ds}\frac{\partial L}{\partial \dot x_\mu}=\frac{\partial L}{\partial x_\mu}, $$ where the dot means derivative with respect to $s$, it is seen immediately that $$ \frac{d}{ds}\left(\frac{1}{\sqrt{\dot x_0^2-a^2(t)|\dot{\bf x}|^2}}a^2(t)\dot{\bf x}\right)=0, $$ giving the expected result.

  • $\begingroup$ Thanks. I will try to fill-in the gaps in the next three weeks and understand why you have a square root in th LHS of the final result that doesn't appear on the paper. I will come back after these 3 weeks if i fail. $\endgroup$ Mar 28 at 13:28
  • $\begingroup$ But the square root is 1 due to $u_\mu u^\mu=-1$ so, it is exactly the result in the paper. $\endgroup$
    – Jon
    Mar 28 at 13:40
  • $\begingroup$ Indeed the square root is 1. Try however to give me the time to check all LoL.... You are faster than your shadow and i'm slow as a snail... ;-) $\endgroup$ Mar 28 at 13:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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