We have determined the effect of insulin infused at 1 and 5 mU/kg/min on gluconeogenesis from alanine in 48-hr fasted men. The conversion of alanine to glucose was measured by the arterial-hepatic venous catheterization technique combined with the infusion of 14C-alanine. During insulin infusion, euglycemia was maintained by variable glucose infusion. When insulin was infused at 1 mU/kg/min the net splanchnic production of 14C-glucose was suppressed by 80% but glucagon infused at the end of the study resulted in substantial release of 14C-glucose from the liver suggesting marked accumulation of labeled glucose in glycogen. When insulin was infused at 5 mU/kg/min the splanchnic release of 14C-glucose was also markedly suppressed but in contrast to the lower insulin dose very little labeled glucose accumulated in glycogen. Neither the high nor the low dose insulin infusion had any effect on net splanchnic alanine uptake and plasma glucagon levels fell by 35% in both protocols. These data demonstrate that in 48-hr fasted man, (1) a small increment in insulin concentration will suppress glucose production but mostly by diverting the newly formed glucose into glycogen; (2) at higher concentrations, insulin will inhibit glucose production mainly by suppressing glucoeogenesis; and (3) this insulin-induced suppression of gluconeogenesis is due to an intrahepatic effect rather than an effect on the splanchnic extraction of alanine.