The effects of intravenous infusion of amino acids, fat, or glucose on unstimulated pancreatic secretion were studied in 31 healthy male volunteers. Each subject was studied twice on two separate days. On both days pancreatic outputs were measured during a 4-hr basal period that was followed by a 4-hr test period. During the test period either one of the digestive end products (100 ml/hr 10% amino acids in 13 subjects; 100 ml/hr 10% fat in 8 subjects; 150 ml/hr 10% glucose in 10 subjects) or saline was intravenously infused. The infusion of the digestive end products or saline on the two separate days was done in a randomized order. Six of the 10 subjects that had already been studied for glucose received a higher glucose dose (100 ml 20% glucose as a bolus followed by 300 ml/hr 10% glucose) in a third additional experiment. Intravenous infusion of amino acids significantly stimulated pancreatic outputs of trypsin and chymotrypsin, but left the outputs of amylase, lipase, bicarbonate, and volume unaffected. The low-glucose dose, as well as the fat infusion did not alter any of the pancreatic outputs. To analyze the relationships between different enzymes during the infusion of digestive end products, regression lines were calculated from the scatter of all individual pairs of enzyme measurements. Significantly different regression functions found for each condition (NaCl control, amino acids, fat, glucose) when trypsin and chymotrypsin were plotted versus lipase and amylase: amino acids shifted the function to a trypsin- and chymotrypsin-dominant pattern of secretion, glucose to an amylase-dominant pattern, and fat to a lipase-dominant pattern.(ABSTRACT TRUNCATED AT 250 WORDS)