The glucose clamp technique allows quantification of tissue sensitivity to insulin without modulation by the effects of hypoglycemia. The amount of glucose infused to maintain euglycemia during an insulin infusion gives a direct measure of tissue sensitivity. We have used this technique to compare the metabolic effects of highly purified porcine, highly purified bovine, and conventional insulins in diabetic subjects: (A) a group normally taking conventional bovine insulin, and (B) diabetics who had originally been treated with conventional bovine insulin and had changed to highly purified porcine insulin. Subjects were studied while attached to Biostator with blood glucose values clamped at the basal level during a 2-h insulin infusion (0.01 U/kg/h). both groups of diabetics received significantly more glucose in the first hour of the monocomponent porcine insulin infusion than during the same period either of conventional or highly purified bovine insulin infusion. Highly purified porcine insulin infusion was also associated with a more rapid fall in blood ketone body concentrations. Highly purified porcine insulin infusion was accompanied by a greater and earlier increase in free insulin concentrations. All diabetics had antibodies to insulin but no differential antibody binding of beef and pork insulin was found in the two groups of patients. Thus, use of highly purified porcine insulin is associated with a more rapid onset of action and a greater initial increment in free insulin values.