We conducted a case-control study to evaluate the effect of replacement estrogen use alone, and the combined effect of estrogen and progesterone use, on the risk for endometrial cancer. We studied women age 50–64 years at Group Health Cooperative of Puget Sound. We identified 172 incident cases of endometrial cancer diagnosed during the years 1979–1989, and controls of similar age and duration of membership in the plan. Women who had used conjugated estrogens alone for 5 or more years were at increased risk for developing endometrial cancer compared with nonusers [adjusted rate ratio (RR) = 22.0; 95% confidence interval (CI) = 6.5–74.1]. Users of unopposed estrogens of 3–4 years' duration had a relative risk of 1.9 (95% CI = 0.4–8.7). Women who had used medroxyprogesterone acetate in combination with the estrogen therapy for 3 or more years had a risk near that of nonusers (adjusted RR = 1.3; 95% CI = 0.5–3.4) and that of users of unopposed estrogens for less than 5 years. There were insufficient women who used estrogen and progesterone together for 5 or more years to derive a separate risk estimate for these women.