The risk of breast cancer and endometrial cancer in 1483 menopausal and postmenopausal women who had received estrogen therapy for other than contraceptive use was examined. The incidence and mortality rates of the study cohort (cohort A) were compared with the age-specific cancer incidence and mortality rates of the female populations of Ontario (cohort B) and of Saskatchewan (cohort C). A significant difference was found in each of the two comparisons between the observed and expected survival curves for breast cancer incidence and breast cancer mortality. However, the incidence rate of endometrial cancer in the study cohort (cohort A) did not differ significantly from the rates of the reference populations (cohorts B and C). The results suggest that the women in the study may have received some protective effect against breast cancer by taking estrogens, but the estrogens did not protect them against endometrial cancer and may indeed have predisposed them to this cancer.