In order to determine the role of postmenopausal estrogen use in survival of patients with endometrial cancer, we identified, through a population-based tumor registry serving the area, all white women in King County, Washington, aged 50 to 74, who developed endometrial cancer between January, 1975, and April, 1976 (N = 363). Estrogen use prior to diagnosis and survival status were ascertained through personal interview, medical records of private physicians, and registry follow-up data. The 4-year relative survival ratio in estrogen users was 1.05 compared to 0.898 in nonusers. Overall, after adjustment for age, the survival in estrogen users was significantly better than in nonusers (chi 2 = 30.5; p much less than 0.0001). While estrogen use may predispose to an increased incidence of endometrial cancer, it appears that such use does not lead to increased mortality from this condition.