HISTORICALLY, the study of the pathogenesis of diabetes mellitus was in the traditional pattern of endocrinology: removal of the pancreas led to experimental diabetes, and administration of insulin, a pancreatic isolate, ameliorated the diabetic symptoms (1, 2). These observations led to the widely held belief that human diabetes was primarily a disease of the pancreas, characterized by the inability of the B cell to secrete sufficient insulin to control glycemia. After insulin became available, evidence emerged suggesting that human diabetes mellitus has a multifactorial etiology. Early investigators identified an unexpected variability among diabetics in the ability of injected insulin to ameliorate hyperglycemia (3–5). Larger doses of insulin were required to normalize the blood sugar in patients with the milder nonketotic form of the disease common in the older population, whereas smaller doses were adequate for younger, ketosis-prone diabetics.