Furosemide, 40 mg, was given to eight healthy volunteers as an i.v. dose and as oral doses (tablet and solution) with and without food intake. The urine and plasma were sampled frequently and analyzed on their content of furosemide (high-performance liquid chromatography). The urine flow and chloride excretion rate were used as measures of the effect. In spite of a 3-fold difference (28 vs. 9 mg/8 hr, P less than .001) in the cumulative urinary excretion of furosemide between i.v. and postprandial oral administration, no significant difference in the diuretic effect was found (2-2.2 liters/8 hr). The drug excretion-response curves showed parallel shifts depending on mode of administration of furosemide. Clockwise hysteresis, indicating acute tolerance development to the diuretic effect, was seen after the oral doses after food intake. This within-dose development of tolerance was modeled with an extended Hill equation. The tolerance development seems to have a near relationship to acute volume depletion (inadequate substitution of urine losses), probably activating some intrarenal mechanism for extracellular fluid volume preservation. Thus, the time course of furosemide excretion, as well as the degree of renal compensation, determine the renal sensitivity to furosemide. This has important implications for the proper design and interpretation of studies of the excretion-response relationship of diuretics.