Secretin stimulates insulin release directly and augments insulin responses when administered prior to an insulinogenic stimulus. The magnitude and specificity of this augmentation effect of secretin pretreatment was evaluated in normal subjects to pulses of the following stimuli: 5 g glucose (n − 6), 1 g arginine (n − 8), 2 μg isoproterenol (n − 6), 0.5 g tolbutamide (n − 5) and 0.5 mg glucagon (n − 6). Secretin was administered either as a 15 U pulse with a 3 U/min infusion for 25 min or as 4–150 U pulses. A period of 30 min elapsed after either the cessation of the infusion or the final 150 U pulse. In all studies, the acute insulin responses immediately following secretin were observed and had returned to baseline levels prior to the administration of any insulinogenic stimulus. Secretin only augmented the acute insulin responses to 5 g glucose pulses (presecretin glucose: 28 ± 20 μU/ml, mean ± SD; postsecretin glucose pulse: 45 ± 37 μU/ml, P <.05), an increase of 56 ± 21% (P < .005). A highly linear relationship was noted (r ∗.98, P < .01) between the acute insulin response to glucose before and fter secretin suggesting that the insulin response to the control glucose pulse is a major determinant of the glucose stimulated response after secretin. Secretin pretreatment did not augment the insulin responses to arginine, isoproterenol, tolbutamide and glucagon. The specificity of secretin for augmenting glucose stimulated insulin release suggests a possible role for secretin in carbohydrate metabolism.