The relationship of neurasthenia and fatigue to carbohydrate metabolism has long been of interest to investigators in psychiatry and internal medicine. Weir Mitchell1treated his neurasthenic patients with forced feeding among other measures and obtained good results in some instances. In 1929 Gillespie2studied cases of fatigue and exhaustion and found that about one third of them showed a decreased glucose tolerance in the oral test. No other observers have found hyperglycemia in fatigue, although it has been described in some of the psychoses. In 1943 Harrison and Finks3reported 31 cases in which hypoglycemia was a factor in the production of symptoms referable to the cardiovascular system. These symptoms appeared two or more hours after meals, were relieved by glucose, were reproduced by insulin and were usually accompanied by a low normal blood sugar. Glucose tolerance was studied by the oral method, but, as the authors