The mortality experience of 13,385 tuberculosis patients treated between 1925 and 1954 in Massachusetts was determined through August 1986. Among 6,285 patients examined by X-ray fluoroscopy an average of 77 times during lung collapse therapy and followed for up to 50 yr (average = 25 yr), no increase in the total number of cancer deaths occurred [standardized mortality ratio (SMR) = 1.05, n = 424]. In contrast, the 7,100 patients treated by other means were at significant risk of dying from cancer (SMR = 1.3), especially of sites linked to cigarette smoking and alcohol use. Among the irradiated patients, estimates of mean radiation doses to the breast, lung, esophagus, and active bone marrow were 0.75, 0.84, 0.80, and 0.09 Gy, respectively. Cancers of the breast (SMR = 1.4, n = 62) and esophagus (SMR = 2.1, n = 14) were significantly increased. The risk of esophageal cancer, however, decreased with time since exposure. Lung cancer (SMR = 0.8, n = 69) and leukemia (SMR = 1.2, n = 17) were not elevated. Despite a wide range of doses to the lung, reaching over 8 Gy, there was no evidence of a dose response. Lung cancer risk also did not vary by time since exposure or age at exposure. Adjustment for smoking and the amount of lung tissue at risk did not appreciably modify these findings. These data suggest that frequent exposures to low doses of radiation over a period of several years increase the occurrence of cancer of the breast. When compared with studies of atomic bomb survivors, however, the fractionated exposures experienced by this cohort appear less effective in causing lung cancer than single exposures of the same total dose.