Endogenous opioids are released during stress and appear to contribute to the cardiovascular suppression seen in shock. When the opiate receptor antagonist naloxone is administered intravenously to anesthetized dogs subjected to hemorrhage, mean arterial pressure, maximal left ventricular dp/dt, and cardiac output increase. This study tests the hypothesis that naloxone acts by potentiating the effects of neurally and adrenally released catecholamines. If this hypothesis is correct, then blockade of endogenous catecholamine release should attenuate the response to naloxone, and administration of exogenous adrenergic agonists prior to naloxone treatment should restore the response. Catecholamine release was attenuated by a combination of surgical adrenal denervation and pharmacological ganglionic blockade. Adrenal denervation or chlorisondamine alone attenuated and, in combination, blocked the response to naloxone in hemorrhaged dogs. Infusion of alpha- and beta-adrenergic agonists at a constant rate prior to treatment restored the response to naloxone. Naloxone appears to improve cardiovascular function in hemorrhagic shock by potentiating the effect of released catecholamines and not by increasing sympathoadrenal discharge.