The velocity of sound in pure ethylene has been determined at various pressures, temperatures and frequencies. It is concluded that equilibrium is attained between the various states of the vibrational energy on every effective collision, but that a considerable activation energy of collision is required for the conversion of translational into vibrational energy. Collisions with argon, helium, and nitrogen molecules have no appreciable effect on the vibrational energy of the ethylene molecule. Collisions with hydrogen molecules are about ten times as effective as ethylene-ethylene collisions in producing transitions in the lower vibrational energy states of ethylene. Hydrogen shows a dispersive region at ultrasonic frequencies which is apparently due to the failure of the rotational energy transitions to follow the acoustical cycle. It has been found necessary to suppose that ethylene-hydrogen collisions are about twenty times as effective as hydrogen-hydrogen collisions in exciting the rotational energy of hydrogen molecules.
