Ninety-one patients with reflux oesophagitis were randomly allocated to treatment with one chewable antacid tablet (acid-neutralizing capacity, 30 mmol) four times daily, 400 mg cimetidine twice daily, or placebo. The study was double-blind, with a double-dummy technique. Endoscopy was performed before inclusion and after 8 weeks' treatment. Symptoms were recorded on diary cards and on visual analogue scales. Statistically significant healing of oesophagitis was achieved in all three treatment groups, but none of the active regimens were significantly superior to placebo. Symptoms were significantly reduced with both cimetidine and antacids compared with placebo. Patients taking antacids consumed significantly less extra antacids for pain relief and had significantly better global assessment score than patients taking cimetidine during the first and second half of the study, respectively. In conclusion, neither cimetidine nor antacids were significantly superior to placebo in healing of reflux oesophagitis. Both the active regimens were superior to placebo for symptomatic relief.