SUMMARY1. The effect of i.v. administration of prostaglandin (PG) E2 (10-40 jug kg-' h-'), 16,16-dimethyl PGE2 (0-1-0-5 jag kg-h-'), PGE, (16)(17)(18)(19)(20) jg kg-h-1), PGA, (5-11 #ug kg-' h-') and PGF2a (40 , 3. At the dose tested, PGF2a had little effect on acid output, and did not alter the relationship between [H+] and gastric flow.4. There is a linear relationship between acid output and gastric flow and this relationship is similar during stimulation of gastric secretion by pentagastrin, histamine or insulin. Gastric acid inhibitory doses of cimetidine, atropine and somatostatin did not alter this relationship. In contrast the A-and E-type prostaglandins displaced this relationship to the right of the normal line observed with the acid stimulants alone. A-and E-type prostaglandins reduced the slope of the line relating acid output and gastric flow from -150-170 ,uequiv/ml-1 to 100-120 #equiv ml-', this being taken as evidence of dilution of the parietal H+ secretion with a non-parietal secretion.5. The volume of non-parietal gastric secretion was calculated as the gastric flow at zero acid output by extrapolation of linear plots of acid output versus gastric flow. Unstimulated gastric flow measured directly was 0 75 ml 15 min'. The calculated non-parietal flow was in the range 0-52-0 90 ml 15 min' during stimulation of gastric secretion with pentagastrin, histamine and insulin, and inhibition of pentagastrinstimulated acid secretion with cimetidine, atropine and somatostatin. PGE2 (1-51 ml 15 min-') and 16,16-dimethyl PGE2 (1-20 ml 15 min') nearly doubled the calculated non-parietal flow.