Seventy-nine patients with an histologically proven disseminated breast cancer, never treated before with additive hormonal therapy, entered into a randomized trial between june 1981 and december 1982. In the first group 44 patients were given continually a daily dose of tamoxifen (TAM) of 20 mg/m2. In the 2nd group 35 patients were given a daily dose of TAM of 20 mg/m2 for 15 days and then an oral daily dose of medroxyprogesterone acetate of 350 mg/m2 for the next 15 days. In both groups I and II, the treatment was stopped at the first manifestation of progression of the disease. The hormonal receptor status was determined in 30 patients of the group I and 23 patients of the group II. An objective response to treatment was observed in 48 per cent of the patients of the group I and 60 per cent of the group II. This difference is not significant (X2 = 1,05). However, the mean duration of therapeutic response is significantly higher in the group II (p = 0,01).