To evaluate further the relationship between high-dose radiotherapy and leukemia incidence, a nested case-control study was conducted in a cohort of 22,753 women who were 18-month survivors of invasive breast cancer diagnosed from 1935 to 1972. Women treated for breast cancer after 1973 were excluded to minimize the possible confounding influence of treatment with chemotherapeutic agents. The cases had histologically confirmed leukemia reported to the Connecticut Tumor Registry (CTR) between 1935 and 1984. A total of 48 cases of leukemia following breast cancer were included in the study. Two controls were individually matched to each leukemia case on the basis of age, calendar year when diagnosed with breast cancer, and survival time. Leukemia diagnoses were verified by one hematologist. Radiation dose to active bone marrow was estimated by medical physicists on the basis of the original radiotherapy records of study subjects. Local radiation doses to each of the 16 bone marrow components for each patient were reconstructed; the dose averaged over the entire body was 530 rad (5.3 Gy). Based on this dosage and assuming a linear relationship between dose and affect, a relative risk (RR) in excess of 10 would have been expected. However, there was little evidence that radiotherapy increased the overall risk of leukemia (RR = 1.16; 90% confidence interval [CI], 0.6 to 2.1). The risk of chronic lymphocytic leukemia, one of the few malignancies without evidence for an association with ionizing radiation, was not significantly increased (RR = 1.8; n = 10); nor was the risk for all other forms of leukemia (RR = 1.0; n = 38). There was no indication that risk varied over categories of radiation dose. These data exclude an association between leukemia and radiotherapy for breast cancer of 2.2-fold with 90% confidence, and provide further evidence that cell death predominates over cell transformation when high radiation doses are delivered to limited volumes of tissue.