Abstract
Einstein's derivation of the dependence of sound velocity on frequency in a mixture of chemically reacting ideal gases of negligible absorption has been extended to mixtures of real absorbing gases. The results show that deviations of experiment from the earlier theory are probably due to departure of the gases from ideal properties. They also show that absorption is probably strong enough to render reaction rate determination by velocity measurements impossible, but that direct absorption measurements may be of some use. It is shown that extraneous phenomena may hide those considered.