Several investigators 1 " 7 have reported that glucose is a stress agent capable of stimulating adrenal cortical activity. A depression of the lymphocyte count served as the criterion of corticosteroid discharge though Drury 5 and Jordan, Last, Pitesky and Bond 3 described glucose induced eosinopenia in laboratory animals and Hungerland and Raming 6 and Artunkel 7 reported the latter phenomenon in humans.Recant et al 8 , however, failed to produce a significant fall in circulating eosinophils in eight normal subjects after the intravenous injection of glucose. Steeples and Jensen 9 reported that in the white rat glucose inhibited the release of adrenal cortical hormones and Skelton 10 found no adrenal cholesterol or ascorbic acid depletion after glucose though the response to corticotropin was enhanced.The possible influence of hyperglycemia on adrenal function and the known intimate relationship of the 11-oxygenated adrenal steroids to carbohydrate metabolism has led to investigation of the physiologic status of the adrenal cortex in diabetes mellitus. Lazarus and his co-workers 11 found that n of 22 diabetic patients failed to exhibit lymphocytopenia after intravenous glucose though the remaining 11 and all of the 18 normal controls did show a significant lymphocyte depression. Wilson et al 12 suggested a compensatory hypofunctional state of the adrenals and Field and Marble 13 found reduced adrenal cortical reserve in diabetic patients. In a paper by Talbot and his associates 14 urinary corticoid values in four controlled diabetic subjects were in the same range as hypopituitary subjects and lower than in Addisonian patients; Forbes and her co-workers 15 as well as Miller and Mason 16 reported 17-ketosteroid values in diabetic patients somewhat lower than normal.The question of the adrenocortical stimulating prop-