The term diabetes mellitus has in the past implied a single cause. Ever since the days of von Mering and Minkowski this disease has been attributed to a pancreatic abnormality. However, within the last few years the work of a number of investigators has made it clear that this conception is far too restricted. Houssay,1for example, has shown that the rapidly fatal diabetes produced by pancreatectomy in animals may be strikingly ameliorated by subsequent hypophysectomy. In the same sense Mann2has produced fatal hypoglycemia in depancreatized dogs by hepatectomy. A number of workers have produced hyperglycemia and glycosuria by injection of pituitary extracts, and Young3reported that he obtained permanent "diabetes" in dogs solely as the result of prolonged injection of anterior pituitary extracts. It is therefore clear that clinical investigators must now attempt to discover the specific cause of the hyperglycemia in all patients who