Special relativity is one of the most tested theories in physics. A central postulated of SR is that the speed of light is the same for all observers in inertial reference frames, and this leads to time dilation and the other effects that have been tested with a high degree of precision. We know, however, that the earth is rotating and revolving around the sun, the sun is revolving around the galactic center etc. So at any given time there is a net acceleration of any earthbound reference frame and it is therefore not strictly inertial. Is this such a small effect as to be totally negligible for testing of SR, at least to the precision of testing to date? I'm thinking in particular of the Michelson-Morley experiment. If the lab frame of reference has a net acceleration in some direction, wouldn't the speed of light be different parallel and perpendicular to that direction? Is this effect totally negligible?