The overnight urine electrolyte and plasma potassium responses to spironolactone (25, 50, 100, and 200 mg daily) at steady state were investigated in 11 healthy subjects after fludrocortisone challenge. For sodium excretion and urine log10 10 Na/K, log dose-response relationships were defined (P less than 0.001 in each case), but there was little evidence of dose-related influences on potassium excretion or plasma potassium. The findings confirm the limitations of the fludrocortisone model in evaluating the potassium-sparing properties of aldosterone antagonists at steady state and of the urine sodium; potassium ratio as an overall index of renal antimineralocorticoid activity.