This is the report of a four year study to determine whether sulfanilamide (para - amino - benzene - sulfonamide) administered prophylactically prevents recrudescences of acute rheumatic fever. A preliminary description of the first two years of our work appeared in January 1939.1 As Coburn2and others have shown, beta hemolytic streptococcus infections usually precede episodes of rheumatic fever, and the antistreptolysin titer of the patient's blood is elevated during the attack.3While the beta hemolytic streptococcus is not considered to be the direct and sole cause of rheumatic fever, its close association with periods of rheumatic activity makes it seem probable that the organism plays a significant role in initiating the disease process. If the factor of infection with the beta hemolytic streptococcus could be removed or sufficiently attenuated, it is possible that the active rheumatic state might not reappear in a susceptible person. The Tréfouëls, Nitti and Bovet