Methods Twenty-eight patients were selected from a group of 39 with various degrees of renal failure on the basis of normal serum potassium (4.0 to 5.0 mmoles/liter) and plasma renin activity (PRA)(200 to 700 fmoles/liter/sec). The patients did not use any medication except sodium bicarbonate, aluminum hydroxide, calcium carbonate, and vitamin D supplements. The diagnoses are listed in Table 1. The patients were assigned to three groups according to the degree of renal failure; that is, group 2 had creatinine clearances between 21 and 70 mI/mm; group 3, between 11 and 20 mI/mm; and group 4, between 3 and 10 mI/mm. Twenty-two healthy volunteers serving as controls (group 1) were kept on normal diet. Blood samples were collected at 9 AM after the subjects had been in the upright position for at least half an hour. Twenty-four-hour urine for analysis was collected during the day before investigation. Potassium and sodium were measured by flame photometry. Renal function was assessed by calculating the 24-hour creati-nine clearance. PRA (fmoles/liter/sec) was determined radioim-munologically according to a modification of the method described by Haber (incubation of undiluted plasma for 1 hr at 37 C and at a pH of 5.6, followed by deproteinization of the plasma with 4 N ammonia/acetone (1: 9)[7]. After 2.0 ml of plasma was extracted with dichloromethane and purified on Sephadex LH 20 columns (1.0>< 50 cm), plasma aldosterone (PALDO, pmoles/liter) was assayed by radioimmunoassay with the radioassay buffer as the mobile phase, as described by Nowaczynski et al [8].