This study tested the hypothesis that breathing at the upper end of the normal range of end tidal CO2 decreases renal sodium excretion. Normotensive human subjects learned to self-regulate end tidal CO2 using a respiratory gas monitor and feedback procedure. Urine flow rates were increased by a standardized water drinking regimen. Urinary volume and sodium and potassium excretion were decreased during 30 minutes of inhibited (i.e. high normal end tidal CO2) breathing, compared with levels preceding and after task performance. Blood pressure, but not heart rate, increased during task performance. Plasma volume increase under these conditions is indicated by the observation that urinary excretion of an endogenous digoxin-like factor was increased. The physiological mechanism by which inhibited breathing elicits renal sodium retention remains to be determined. This breathing pattern could mediate the role of behavioral stress in some forms of hypertension.