A pulsed ruby optical maser has been constructed which allows the beams from opposite ends to be superimposed. Interference fringes have been observed which can be interpreted to show that the coherence predicted by theory is, in fact, obtained, and also that the relative spatial distribution of the light over the ruby face is essentially constant even though the light is emitted in short bursts.
P. Kisliuk and D. J. Walsh, "The Interference between Beams from the Opposite Ends of a Ruby Optical Maser," Appl. Opt. 1, 45-46 (1962)