SUMMARY Seventy-seven patients with gastroduodenal ulcer were treated with two methyl-prostaglandin E2 analogues, m-PGE2, in a double-blind clinical trial. Each of three groups was given 15 S-15 methyl PGE2 methyl ester, 15 R-15 methyl PGE2 methyl ester, and placebo, respectively. Both forms of m-PGE2 analogues appeared to reduce gastric acid secretion, to shorten ulcer healing, and also to produce some side-effects, form 'S' being the more potent. Prompt healing of the ulcer with these agents did not prevent the recurrence of the disease. As the serum gastrin response to a meal after m-PGE2 administration was not reduced, this agent seems directly to affect oxynthic cells.