The accepted picture of the nephrotic syndrome is that of a grossly oedematous patient with massive albuminuria but having no hypertension or azotaemia. The amount of albumin circulating in the plasma is greatly reduced, resulting in a fall in the total protein content and a reversal of the albumin/ globulin ratio. The cholesterol content of the blood is markedly raised. Little is known of the aetiology of nephrosis itself, or of the spontaneous alterations which may occur in its manifestations. The oedema and albuminuria, for example, may vary dramatically and unpredictably, and occasionally vanish completely to result in the cure of the patient. As regards the oedema, it is suggested that this is caused by a retention of sodium which may in turn be produced by overactivity of the salt-retaining hormones of the adrenal glands. That such a mechanism will produce oedema is known, and is exemplified by the oedema occurring when an excess of desoxycorticosterone acetate (D.O.C.A.) is administered to normal subjects or even to patients with Addison's disease where dehydration and an excessive sodium output characterize the untreated case. Levitt and Bader (1951) have shown that in the extracellular fluids both cortisone (17-hydroxyI 1-dehydrocorticosterone) and A.C.T.H. (adrenocorticotrophic hormone) have a sodium-retaining action most pronounced in the first few days of therapy. Furthermore, experimental albuminuria in animals is increased by either cortisone or A.C.T.H. and diminished by adrenalectomy (Addis, Marmorston, Goodman, Sellers and Smith, 1950). It may therefore be accepted that an excess of adrenal hormones contribute to oedema and albuminuria. It has been firmly established that, following an intensive course of either cortisone or A.C.T.H., there is a transient depression of endogenous cortisone production by the patient (Luetscher and Deming, 1950; Kendall, 1951; McIntosh and Holmes, 1951). Presuming that a similar response occurred in nephrotic patients following sudden withdrawal of cortisone or A.C.T.H. it seemed possible that a diuresis from loss of sodium and a lessening of the albuminuria might result. In order to test this hypothesis. cortisone was administered to nephrotic patients. and observations made on the effects of its sudden withdrawal. Cortisone was preferred to A.C.T.H. because contamination of the latter with antidiuretic hormone of the posterior pituitary frequently occurs and would make alterations arising. in the water balance of the patients difficult to interpret.