Sixty out-patients with duodenal ulcers that were healed at the end of a 4-week treatment with pirenzepine, cimetidine or placebo were admitted to a double-blind placebo-controlled trial to study the effectiveness of pirenzepine (100 mg/daily) in preventing recurrence of ulcers. Six patients did not complete the trial. After 12 months of treatment 15 of the 26 patients had recurrences in the pirenzepine-treated group and 27 of the 28 in the placebo group. The difference is highly significant (X2 = 9.570, P less than 0.01). The tolerability of pirenzepine was good.