In a trial that began in 1978, 1312 evaluable patients under 80 years of age who either had negative axillary nodes or were postmenopausal with positive axillary nodes were randomised to receive adjuvant tamoxifen 20 mg daily for 5 years, or tamoxifen for the treatment of first relapse. Estimates of oestrogen receptor (ER) content of primary tumour specimens were made in 57%. There has been a highly significant delay in relapse in the adjuvant arm of the trial. This benefit supersedes that from tamoxifen given as treatment for recurrent disease in control-arm patients (93% received this) so that benefit from adjuvant tamoxifen was maintained in the overall survival comparisons. This improvement seems to be independent of nodal and menopausal status. It does not differ significantly with ER level, although the greatest benefit in disease-free survival is in patients with levels of 100 fmol/mg protein or more.