The effect of surgical denervation on the antinatriuretic response to morphine was assessed in conscious rats prepared with arterial and venous catheters and bladder cannulas. In sham-operated animals, morphine sulfate (4 mg/kg i.v. plus 2 mg/kg X hr) caused a marked reduction in sodium excretion (650 +/- 218 vs. 2394 +/- 265 nEq/100 g X min in vehicle-treated animals; P less than .05). Although glomerular filtration rate was reduced, fractional excretion of sodium, expressed as a percentage of the filtered load, was also markedly decreased by morphine (0.62 +/- 0.18 vs. 1.51 +/- 0.15% in the vehicle group; P less than .05). The changes in sodium excretion were readily reversed by naloxone. In rats subjected to the renal denervation procedure, morphine had no effect on sodium excretion (1614 +/- 261 vs. 1926 +/- 520 nEq/100 g X min) or on fractional sodium excretion (1.92 +/- 0.50 vs. 1.41 +/- 0.37%). However, glomerular filtration rate was reduced by morphine as in the sham-denervated rats. Blood pressure and heart rate were not significantly affected by morphine in either group. The results suggest that morphine enhances renal tubular sodium reabsorption, at least in part, by a mechanism dependent on intact renal nerves. Because there was no evidence of generalized sympathetic activation (as judged by blood pressure and heart rate changes), the effect may be selective for the renal innervation.