Ethanol-induced accumbal dopamine elevations have been linked to ethanol consumption. It is unclear, however, where along the mesolimbic dopamine system this effect is initiated and why the ethanol-induced dopamine elevations are transient, returning to pre-drug baseline before brain and blood ethanol levels decline. Using in vivo microdialysis, Experiment 1 investigated the effect of local ethanol application in the nucleus accumbens, the ventral tegmental area and the nucleus accumbens+the ventral tegmental area, on accumbal dopamine. Experiment 2 examined whether the rapid withdrawal of dopamine response to ethanol involves activation of GABA(A)-receptors, by analyzing the effect of accumbal co-perfusion of picrotoxin and ethanol. In Experiment 1, ethanol perfusion into the ventral tegmental area alone did not affect accumbal dopamine. Ethanol co-perfusion of one of the tested doses into the ventral tegmental+the nucleus accumbens produced higher dopamine levels than ethanol perfusion into the nucleus accumbens alone during 120-160 min following perfusion onset. In Experiment 2, accumbal ethanol perfusion caused a transient increase in nucleus accumbens dopamine. Co-perfusion of ethanol and picrotoxin produced a sustained dopamine elevation. These data support the hypothesis that the primary effect of ethanol on accumbal dopamine is in the nucleus accumbens, but that a secondary effect of nucleus accumbens ethanol perfusion, such as release of acetylcholine in the ventral tegmental area, enables ethanol to act as a nicotinic acetylcholine receptor co-agonist in this area. Moreover, recruitment of GABA(A)-receptor activity appears responsible for the second, declining phase with respect to dopamine levels following ethanol administration.