For 1950-74, 360 cases of malignant breast tumors were identified among the 63,000 females of the Radiation Effects Research Foundation's (Hiroshima and Nagasaki) Extended Life-Span Study sample of survivors of the 1945 atomic bombings of Hiroshima and Nagasaki; 288 of these females were residing in one of these two cities at the time of bombing (ATB). Two-thirds of all cases were classified as breast cancers on the basis of microscopic review of slides, and 108 cases received an estimated breast tissue dose of at least 10 rads. The number of cases of radiogenic breast cancer could be well estimated by a linear function of radiation dose for tissue doses below 200 rads. Excess risk estimates, based on this function, for women 10-19, 20-29, 30-39, and 50 years old or older ATB were 7.3, 4.2, 2.6, and 4.7 cases per million women per year per rad, respectively. Women irradiated in their forties showed no dose effect. Among all women who received at least 10 rads, those irradiated before age 20 years will have experienced the highest rates of breast cancer throughout their lifetimes. Separate excess risk estimates for Hiroshima and Nagasaki did not differ significantly, which indicates that for radiogenic breast cancer the effects of neutrons (emitted only in the Hiroshima explosion) and gamma radiation were about equal. Radiation did not reduce the latency period for the development of breast cancer, which was at least 10 years. The distribution of histologic types of cancers did not vary significantly with radiation dose. The data suggested that irradiation prior to menarche conferred a greater risk than irradiation after menarche.