Intracellular recordings were made in pontine slice preparations of the rat brain containing the locus coeruleus (LC). Ethanol at 100 mM, but not at 10 or 30 mM inhibited depolarizing responses to pressure-applied N-methyl-D-aspartate (NMDA) in LC neurons of ethanol-naive rats. Ethanol (100 mM) had a similar effect in LC neurons of ethanol-naive rats, of rats treated with ethanol for 14 days (3 g/kg daily, i.p.) and of rats treated with equicaloric amounts of saccharose (5 g/kg daily, i.p.). The blood concentration of ethanol was markedly decreased at 4 h, and was below the detection limit at 24 h after the last injection. Behavioral measurements in the open-field system demonstrated the development of tolerance in rats receiving ethanol for 14 days. Moreover, an anxiety-related reaction was shown to develop when the acute effect of the last ethanol injection vanished. Therefore, in subsequent in vitro experiments, ethanol (10 mM) was continuously present in the superfusion medium in order to mimic a steady blood concentration and to prevent a withdrawal-like situation. Under these conditions, ethanol (100 mM) still continued to inhibit the NMDA-induced depolarization in slices of untreated rats, but became ineffective in slices of ethanol-treated rats at 4 h after the last injection. By contrast, a supersensitivity to ethanol developed in brain slices at 24 h after the last ethanol injection. In conclusion, in vitro tolerance between systemically and locally applied ethanol at LC neurons could only be demonstrated when a low concentration of ethanol was added to the superfusion medium to simulate the blood concentration of this compound.