Thirty-eight cases of human rabies occurred in the United States and its territories from 1960 to 1979. The major source of exposure to rabies has changed from indigenous dogs and cats in the 1940s and 1950s to wild carnivores and bats (11 of the 27 cases with known exposures); unusual exposures (3 cases) and exposures in a foreign country (7 cases) have also become more important. No exposure could be identified for 6 of the 38 cases. Two patients received optimal prophylaxis, 14 suboptimal, and 22 no prophylaxis after exposure. Some cases might have been prevented by an increased awareness of the risks and treatment for exposure to rabies, and use of the new rabies vaccines. The diagnosis was often made late in the clinical course including after death in 8 cases. This delay, in part, resulted from the diversity in the clinical presentation. Rabies should be considered in any case of encephalitis or myelitis. Laboratory confirmation of the diagnosis was often delayed. Testing for serum antibodies was the most reliable test in unvaccinated patients, and isolation of virus was the test most likely to be positive early in the illness.