Recent studies have demonstrated that ethanol blocks N-methyl-D-aspartate (NMDA) responses in vitro. In the present study, evidence is provided that ethanol, when administered by the systemic route to rats, also inhibits NMDA-evoked electrophysiological activity in vivo at behaviorally relevant doses. Ethanol, at doses in rats ranging from those producing minimal changes in spontaneous behavioral activity (0.75 g/kg) to those producing marked suppression of behavioral activity (2.5 g/kg), produced a dose-dependent inhibition of the ability of NMDA, when iontophoresed onto neurons of the medial septum (MS), to activate MS neurons. However, at all doses of ethanol tested, a proportion of MS neurons responded to ethanol with essentially complete inhibition of NMDA-evoked activity, whereas other MS neurons responded to ethanol with little or no inhibition of NMDA-evoked activity. By way of comparison, MK-801, a non-competitive NMDA antagonist, antagonized NMDA-evoked activity in all MS neurons tested. In contrast to the actions of ethanol, MK-801 increased, rather than decreased, behavioral activity even at doses that completely inhibited NMDA-evoked activity in all MS neurons tested. These latter findings provide evidence that inhibition of NMDA-evoked activity cannot account for all of the behavioral effects of ethanol. In conclusion, while the present results demonstrate for the first time that ethanol can inhibit NMDA-evoked neuronal activity in vivo, they also indicate that additional neural actions must contribute to ethanol's pharmacological profile.