In a randomized, double-blind, endoscopically controlled clinical trial, 40 patients with duodenal ulcer were treated with either ranitidine (320 mg/day) or cimetidine (800 mg/day) for 28 days. The rate of healing of the ulcers (ranitidine, 78%, cimetidine, 45%) did not differ significantly in the two treatment groups. In a subsequent 'open' study, ranitidine (150 mg twice daily) healed the ulcers of 79% of 19 patients treated for 28 days. Treatment with ranitidine was not associated with significant adverse reactions and had no lasting effect on gastric secretory capacity. Whereas basal levels of serum testosterone rose significantly during treatment with cimetidine, no such increase was observed during treatment with ranitidine. We conclude that ranitidine can effectively and safely heal duodenal ulcers.