The interrelations of the insulin secretagogues, glucose, arginine (Arg), and gastric inhibitory polypeptide (GIP) were quantified in six normal young men in five sets of experiments with the hyperglycemic clamp technique (125 mg/dl above basal glucose levels for 2 h). After 60 min of intravenous glucose alone, one of the following was added: A) oral glucose (OG) (40 g/m2); B) 15 g.m-2.h-1 Arg infusion; C) 15 g.m-2.h-2 Arg infusion and OG; D) 7.5 g.m-2.h-1 Arg; E) 7.5 g.m-2.h-1 Arg and OG. The clearance rate of Arg was similar for B, C, D, and E. In all experiments, plasma GIP levels were unchanged from the basal level during the 1st h. The increases in plasma GIP levels in experiments C and E were similar to the increase when OG alone was ingested (A). When the stimulatory effect of the secretagogue(s) alone on insulin (IRI) is computed, the increase due to OG (A) and to 7.5 g.m-2.h-1 Arg (D) were similar and additive (A + D approximately equal to E). However, the stimulatory effect of 15 g.m-2.h-1 Arg + OG (C) on IRI was not significantly greater than 15 g.m-2.h-1 alone (B). The 15 and 7.5 g.m-2.h-1 Arg infusion produced different patterns of insulin and glucagon secretions. At the lower dose, the response of both hormones to Arg decreased with time. Arg and GIP act through a similar and possibly common mechanism on the beta-cell. However, only Arg was found to be alpha-cytotropic. GIP does not appear to influence the metabolic clearance of Arg. The dose-response relationships to Arg of the beta- and alpha-cell appear similar.