A clinical trial was carried out at The Christie Hospital, Manchester, England, between November 1976 and June 1982. Following surgery, patients with clinical stage T1-T3a, N1-N2b, M0 tumors were randomly allocated in the following way: premenopausal women to either tamoxifen 20 mg/day for 1 year or to an irradiation-induced menopause; postmenopausal women to either tamoxifen for 1 year or no further treatment (controls). In the node-negative subgroup, an analysis at a median follow-up of 10 years shows no statistically significant difference in overall or disease-free survival between tamoxifen-treated patients and irradiation-induced menopause and control patients. Side effects were few, and compliance was excellent. There was no significant increase of second primary tumors in target endocrine organs in the tamoxifen-treated patients. For the whole series of 961 patients, a log-rank analysis of events, allowing for node status, still shows a statistically significant benefit for tamoxifen-treated patients (P = .04).