The aim of this work was to assess the usefulness of somatostatin in acute peptic ulcer bleeding. Sixty-four patients with endoscopic evidence of duodenal ulcer bleeding completed a double-blind trial to compare the effectiveness of somatostain (6 mg i.v. per day) with that of placebo; incidence of treatment failure (i.e., rebleeding or persistent haemorrhage) was not different in the somatostatin and in the placebo groups (respectively 24% and 23%). Because the efficacy of cimetidine had previously been demonstrated in gastric ulcer bleeding, somatostatin was tested against cimetidine (1.6 g per day) in 50 patients with gastric ulcer haemorrhage. The treatment failed to control bleeding in 15% of somatostatin and in 17% of cimetidine treated subjects (n.s.). Both treatments were well tolerated and blood requirements were not different in the trial groups. These results show that the efficacy of somatostain in peptic ulcer bleeding is not different from placebo in duodenal ulcers and not different from cimetidine in gastric ulcers.