The effect of an antacid tablet regimen (total acid-neutralizing capacity, 280 mmol/day) and placebo was studied in 75 patients with duodenal ulcer in a double-blind 4-week trial. The ulcer healed in 30 out of 37 (81%) patients treated with antacids as compared with 9 out of 38 (24%) patients treated with placebo (p less than 0.001). Ulcer symptoms in the antacid- and the placebo-treated groups did not differ significantly until the last week of treatment (p less than 0.01). Both constipation and diarrhoea were slightly more common in antacid- than in placebo-treated patients (NS). Serum concentration of aluminium increased significantly (p = 0.01), whereas serum concentrations of calcium, phosphorus, magnesium, iron, and total iron-binding capacity did not change during treatment. No other side effects of antacids were recorded. Thirty-eight patients with unhealed ulcer after cessation of the antacid/placebo treatment were treated openly with 150 mg ranitidine twice daily. The ulcer healed in 31 out of 37 (83%) patients (one drop-out) after 4 weeks' treatment, and only one patient remained with unhealed ulcer after 6 weeks' ranitidine treatment. No side effects due to ranitidine were recorded.