All British Columbia (Canada) women under 75 years of age who were diagnosed with breast cancer during 1988-89 were asked to complete a postal questionnaire which included detailed information on menopausal estrogen use. Controls were drawn from the Provincial Voters List, matched by five-year age category to the cases. The present analysis consists of 699 cases and 685 controls who were postmenopausal due to natural causes or to a hysterectomy. There was no overall increase in risk of breast cancer associated with ever-use of unopposed estrogen (odds ratio [OR] = 1.0, 95 percent confidence interval [CI] = 0.8-1.3). For estrogen use of 10 years or longer, the relative risk [RR] was 1.6 (CI = 1.1-2.5). The risk estimate for current users was somewhat elevated (OR = 1.4, CI = 1.0-2.0). Compared with women who never used hormone preparations, women who had used estrogen plus progestogen had an RR of 1.2 (CI = 0.6-2.2). Our results suggest that ever-use of estrogen, with or without progestogen, does not appreciably increase the risk of breast cancer. However, long-term and recent use of unopposed estrogen may be associated with a moderately increased risk.