Evidence is given for the first minutes of interaction between Sendai virus and red blood cells (RBC) to be determining for the process of hemolysis, the latter proceeding further independently on the presence of virus. This period was found to correspond approximately to the time interval needed for the total loss of intracellular potassium from cells predestined to hemolysis. Identical quantitative relationships for both hemoglobin and potassium were found between the concentration of virus, that of RBC and the amount of material released from cells, in consistence with the idea that hemolysis is an "all-or-none" process. Monovalent cations and anions affect hemolysis in a manner similar to that found for hypotonic malonamide solutions. Bivalent cations, especially Ba++ inhibit hemolysis probably by reacting with acidic groups on the cell surface which are involved in the "hemolytic reaction." Adenosine triphosphate (diphosphate) and organophosphorous compounds affect hemolysis in an opposite manner, probably due to their different action on the "contractile system" of the cell membrane. These findings are compatible with a colloid-osmotic mechansim of hemolysis.