Twenty-five symptomatic postmenopausal women with an intact uterus were assigned in random double-blind fashion to receive 0.625 mg of conjugated estrogens on either a cyclic (3 weeks on, 1 week off) or continuous (daily) basis. The incidence of endometrial hyperplasia as demonstrated by screening biopsies at 6 and 12 months of therapy was 4.5 per 100 woman-months in the cyclic group and 3.7 per 100 woman-months in the continuous group (a difference not statistically significant). Thus, in this study, cyclic therapy was found to offer no advantage over continuous therapy. In our opinion, the rate of hyperplasia development in both groups unacceptably high, and efforts must be directed at reducing its incidence primarily.