Cancer risk is studied among Japanese A-bomb survivors, patients medically irradiated, workers in radiation occupations, and persons exposed to naturally occurring radiations. Biochemical approaches are incorporated into studies to help define basic mechanisms. Program members serve on committees advising the government and international agencies. Results of studies suggest that (1) indoor exposure to radon appears less hazardous than previously believed; (2) radiotherapy for breast cancer increases the risk of cancer in the opposite breast slightly, but not among women treated over age 45; (3) the risk of radiation-induced breast cancer declines with increasing age at exposure, and is lower among women whose first pregnancy occurred before age 20 than among nulliparous women or women with a late age at first pregnancy; (4) radiotherapy for breast cancer increases the risk of leukemia, especially in combination with alkylating agents; (5) low-dose radiotherapy to treat uterine bleeding induces many more leukemias than high-dose radiotherapy to treat cancers of the cervix or uterus; (6) high-dose radioactive iodine treatments do not appreciably increase the risk of leukemia, suggesting that protraction of dose reduces risk; (7) radiotherapy for childhood cancers greatly increases the risk of thyroid cancer; (8) radiotherapy for bilateral retinoblastoma greatly increases the risk of osteosarcoma, indicating the importance of a genetic susceptibility in radiation risk; (9) in both A-bomb survivors and uranium miners, radiation-induced lung cancers tend to be of the small-cell subtype.