The effect of sham feeding upon gastric acid secretion and pancreatic polypeptide release was investigated in 28 patients with duodenal ulcer in order to evaluate whether high basal vagal activity is the cause of basal acid hypersecretion in patients with duodenal ulcer and basal secretion higher than 30% of their peak acid output. The patients were divided into two groups based on the ratio of basal/pentagastrin stimulated peak acid output (BAO/PAO) was higher or lower than 0.30: group A n = 19 (BAO/PAO less than or equal to 0.30) and group B n = 9 (BAO/PAO greater than 0.30). Gastric acid response to sham feeding (SAO) was significantly higher than basal level in group A (SAO: 11.4 mEq/h (2.5-20.1) vs BAO: 5.2 mEq/h (0.8-22.9), p less than 0.01, median (range)) while in group B the acid secretion did not increase with sham feeding (SAO: 9.6 mEq/h (4.5-13.6) vs BAO: 8.8 mEq/h (6.3-13.8) ns, median (range)). A negative correlation (r= -0.6118226, p less than 0.01) was found between acid increase expressed as basal subtracted sham feeding response (SAO-BAO) and BAO/PAO ratio of the entire group of duodenal ulcer patients (n = 28) suggesting that the greater is basal acid secretory capacity the smaller is acid increase in response to residual vagal activation. Pancreatic polypeptide response to sham feeding was higher in group A than in group B but no correlation (r = 0.20, n = 28) nor individual covariation was found between acid and pancreatic polypeptide secretions during vagal stimulation. sham feeding did not change serum gastrin. It is concluded that an increased vagal stimulation seems to be the cause of basal hypersecretion in a subgroup of patients with duodenal ulcer. The lact of correlation between the pancreatic polypeptide and acid responses to vagal stimulation interferes with the reliability of pancreatic polypeptide as indicator of vagal tone on gastric parietal cells.