The secretin injection test is considered a useful adjunct to the diagnosis of gastrinoma, although it may lack specificity. This study determined whether the release of gastrin in response to secretin was different in duodenal ulcer and control patients. Tests were performed on 10 duodenal ulcer patients, 10 normal control subjects, 20 patients asymptomatic after ulcer surgery, of whom 13 had a vagotomy and drainage, 4 a highly selective vagotomy and 3 a vagotomy and antrectomy. The secretin test was also performed in 49 patients with endoscopically proven recurrent ulcers. The surgery performed in this latter had consisted of a vagotomy and drainage in 36, a highly selective vagotomy in 7 and a vagotomy and antrectomy in 6 patients. The basal plasma gastrin level was similar in normal controls, duodenal ulcer patients and patients with vagotomy and antrectomy, both with and without recurrent ulcers. The level was elevated in all the other groups of patients with vagotomy both with and without recurrent ulcer. The plasma gastrin did not change significantly after secretin injection in the normal control or asymptomatic ulcer surgery patients, but rose in the duodenal ulcer patients and all the groups of patients with recurrent ulcer. Most of these increases were validated statistically as were the differences in response between the ulcer and control patients. These results indicate that there are differences in the plasma gastrin response to intravenous secretin between active duodenal ulcer and control patients. This findings may aid our understanding of the pathophysiology of peptic ulcer disease and explains the lack of specificity of the secretin test.