Altogether 117 patients with advanced breast cancer were treated with either tamoxifen 10 mg by mouth twice daily or aminoglutethimide 250 mg by mouth four times daily with hydrocortisone 20 mg twice daily in a randomised cross-over trial in which patients who failed to respond to the first treatment or relapsed while receiving it were switched to the other. Eighteen (30%) out of 60 patients initially treated with tamoxifen achieved an objective response and 11 (18%) showed stable disease. Seventeen (30%) out of 57 patients treated initially with aminoglutethimide achieved an objective response and 13 (23%) achieved stable disease. Objective responses in bone metastases were achieved more commonly with aminoglutethimide (11 patients (35%)) than with tamoxifen (five (17%)). The predicted median duration of response for tamoxifen was 15 months and for aminoglutethimide over 15 months (no significant difference). Five (15%) out of 34 patients who failed to respond to tamoxifen and four out of six patients who relapsed after responding to tamoxifen subsequently responded to aminoglutethimide. In contrast, only two (6%) out of 31 patients who failed to respond to aminoglutethimide and none out of four patients who relapsed while receiving aminoglutethimide subsequently responded to tamoxifen. The main side effects occurring in the 97 patients who received aminoglutethimide as first- or second-line treatment were lethargy and drowsiness (36 patients) and rash (29); seven patients had to stop treatment because of side effects. In contrast, side effects were rare and mild with tamoxifen and no patient had to stop treatment because of them. Both tamoxifen and aminoglutethimide appeared from this study to be equally effective in the medical endocrine treatment of advanced breast cancer.