To investigate the quantitative importance of sodium balance to arterial pressure changes produced by changes in sodium intake, we studied normotensive white and black subjects and hypertensive black patients with two protocols. Protocol 1 used a 3-day control period with a 150 mEq sodium intake/day followed by 4 days of salt depletion (SD) with a diet providing 9 mEq/day of sodium and furosemide, 1 mg/kg, given the first day and then 3 days of salt loading (SL), during which 25 mL/kg of isotonic sodium chloride solution was given intravenously each day (3.88 mEq sodium/kg/day). In protocol 2, the sequence of sodium intake changes was reversed. For both protocols, sodium balance was calculated by subtracting urinary sodium excretion from sodium intake and expressed in mEq/kg, either positive or negative. In protocol 1, the hypertensives had statistically significant changes in arterial pressure with changes in salt intake, and they also lost more sodium than normotensives during SD. In protocol 2, blacks, both normotensives and hypertensives, had statistically significant pressure changes with both SL and SD, and black hypertensives retained less sodium during SL than either normotensive group. Spearman correlations showed no relationship between sodium balance and mean arterial pressure, suggesting that salt-sensitive hypertension results not from the magnitude of sodium retention, but from the pressor mechanisms evoked.