Anxiolytic effects of ethanol have been proposed to be important factors in the initiation of ethanol consumption. To examine this hypothesis, drug-naive Wistar rats were tested in the elevated plusmaze to determine their initial level of anxiety. Based on their response, we separated the animals into anxious and non-anxious groups. After that, animals went through an oral ethanol self-administration procedure. Rats that were initially classified as anxious showed a significantly (P<0.01) higher intake and preference for ethanol during the initiation phase of the voluntary drinking procedure than non-anxious animals. In another experiment, intraperitoneal (IP) injections of ethanol (0.5–1.5 g/kg) produced dose-dependent anxiolytic effects in rats when tested in the elevated plus-maze procedure. Blood ethanol levels following IP injections during the plus-maze test were similar to those reached during the oral ethanol self-administration procedure, which shows that the rats indeed drank sufficient amounts of ethanol to experience its anxiolytic effects. These findings indicate that the basal level of anxiety plays an important role in vulnerability to alcohol drinking.