Since the initial development of the "passage A" mouse leukemia virus in 1957, this virus has been propagated in our laboratory by serial passage in newborn C3H(f) mice. At the present time, 10(-2)-10(-3) dilutions in physiological saline solution of this mouse-passaged virus induce lymphatic leukemia in practically all inoculated mice after a latency of 3-5 months. On the other hand, when the same virus was propagated on NIH 3T3 mouse embryo cells in tissue culture for more than 10 years, its leukemogenic potency became considerably reduced. Recent bioassay experiments carried out in our laboratory demonstrated that after such prolonged propagation in tissue culture this virus now induced leukemia in less than 15% of the inoculated suckling C3H(f) mice; only undiluted or 10% dilutions of the tissue culture fluid (very occasionally 10(-2) or 10(-3) dilutions) induced leukemia after a prolonged latency varying from 5.5 to 18 months. The passaged and the tissue-culture-grown virus strains are identical immunologically and indistinguishable in their morphology when examined by electron microscopy. The tissue-culture-grown virus, attenuated in its leukemogenic potency, does not, however, confer immunity against a challenge with the mouse-passaged virus.