Cardiovascular actions of insulin were studied by intravenous infusions of insulin (4 and 8 mU/kg per min) in normal conscious dogs. This resulted in increases in cardiac output, heart rate, and left ventricular derivative of pressure with respect to time (dP/dt) and dP/dt/P, as blood glucose was reduced. The inotropic and chronotropic effects of insulin were not related to hypoglycemia, as they persisted even when blood glucose was restored to control values or when it was prevented from falling by a simultaneous infusion of glucose. These cardiac effects were accompanied by increases in plasma catecholamines, and were abolished by propranolol pretreatment. Both plasma epinephrine and norepinephrine increased during insulin hypoglycemia, but only norepinephrine increased during insulin infusion when euglycemia was maintained. Mean arterial blood pressure did not change significantly during insulin hypoglycemia, but rose if euglycemia was maintained, probably due to the selective increase in norepinephrine in the latter condition. A pressor response also occurred in propranolol-pretreated dogs during insulin hypoglycemia, but was abolished when the animals also had been pretreated with phentolamine, indicating that the vasoconstrictor action of insulin was mediated via alpha adrenergic receptors. Insulin infusion increased left ventricular work and myocardial blood flow in dogs with and without hypoglycemia. Myocardial blood flow, however, did not change significantly during insulin infusion in dogs pretreated with propranolol. As propranolol also diminished the inotropic response, it appears that the increase in myocardial blood flow caused by insulin in the normal dog is causally related to the increased myocardial metabolic demand. Insulin also produced vasomotor effects on other vascular beds. In skeletal muscle, blood flow was increased under all study conditions, except during insulin hypoglycemia after propranolol-pretreatment when unopposed alpha-mediated vasoconstriction was present. The persistent increase in flow during both alpha and beta adrenergic blockade suggests that insulin has a direct dilator effect on skeletal muscle vasculature. In the adrenal gland, flow was increased except during euglycemia, when no rise in plasma epinephrine was observed, suggesting coupling between adrenal flow and catecholamine release. In the splanchnic bed, flow was decreased during euglycemia, when plasma norepinephrine rose, and during beta adrenergic blockade with propranolol, when unopposed alpha-mediated vasoconstriction also predominated. A similar pattern was found in the kidney, except that renal blood flow also fell after combined alpha and beta adrenergic blockade. The results show that the vasomotor effects on regional flows are mediated both via adrenergic mechanisms, and in the case of skeletal muscle and kidney, via mechanisms unrelated to sympathetic stimulation.