Recent content by pgf

You're right this seems to be the crux of my problem. I have trouble picturing light being reemitted with a new velocity without also being reemitted with a new phase. But clearly that's not what happens. Thanks for your patience!

Why would I think that? Ignorance, no doubt. But here goes: If the only difference between the divergent light paths is expected be their velocity, then extinction in the final light path, which will presumably equalize their velocities, would seem to eradicate what is hoped to be measured.

here's a start on wikipedia: http://en.wikipedia.org/wiki/Extinction_theorem_of_Ewald_and_Oseen and this chapter intro to "Classical Optics and Its Applications": http://books.google.com/books?id=Nrkfmf4RizIC&lpg=PP1&pg=PA168#v=onepage&q&f=false

What about the final, shared path to the detector or eyepiece? It feels like extinction there should mask any differences created in the divergent paths. But my "feeling" about this isn't worth much -- I realize that. :-)

I think I need more explanation than that. If the light paths being measured for anisotropy are extinguished and reemitted (at a different velocity) by the air surrounding the device, then how can the results of the experiment not be affected?

I just did some more searching of the forum, and this note from pmb_phy in 2008 summarizes my thoughts on this. The original is here: * https://www.physicsforums.com/showpost.php?p=1784283&postcount=67 but I'll quote the relevant part:

It's well understood that extinction effects must be considered when conducting experiments to measure the velocity of light from moving sources. Fox showed that all such experiments prior to the early 60s were flawed on that basis. I'd like to understand why early M-M type experiments...