The opioid antagonist, naltrexone, reduces intake of, and operant responding for, ethanol, but reports of how the opioid agonist morphine alters these effects are conflicting. We examined the discrimination and plasma levels of ethanol with naltrexone and morphine pretreatments. Rats were trained to discriminate ethanol (1.5 g/kg; i.g.) from water, under a two-lever, food-reinforced procedure. Ethanol and pentobarbital, but not amphetamine, substituted for ethanol in a dose-related manner. Naltrexone reduced ethanol-induced, ethanol-appropriate responding to about 35%, but the peripherally-acting antagonist, naltrexone methobromide, was without effect. Morphine neither substituted for nor enhanced ethanol-appropriate responding. Rather, ethanol-induced, ethanol-appropriate responding was attenuated in a dose-related manner by morphine administration. Neither naltrexone nor morphine altered ethanol-appropriate responding to the substitution with pentobarbital. In another group of rats, both naltrexone and morphine decreased plasma ethanol levels, and delayed the time of peak concentrations, suggesting that opiates alter the behavioral effects of ethanol through both pharmacokinetic and pharmacodynamic mechanisms. The similarities between an opioid agonist and an antagonist suggest that either naltrexone has opioid agonist-like effects, or that these effects occur through non-opioidergic mechanisms.