Many clinical studies based on retrospective self-reports indicate a relationship between anxiety and increased alcohol consumption or relapse in individuals with alcohol abuse or dependence. However, by these retrospective studies it cannot be definitely concluded whether the alcohol abuse or the anxiety was first. In the present study, alcohol-consuming behaviour was determined in three rat strains showing different anxiety-related behaviour but being not genetically selected for high or low alcohol consumption. The innate anxiety of the three rat strains (Harlan-Fischer, Wistar-BgVV and Wistar-Harlan) was measured by the elevated plus maze test. Thereafter voluntary ethanol intake was measured for 3 months followed by a progressive ratio paradigm, in which the number of responses required to obtain alcohol was successively increased during session. The point at which rats ceased to respond (breaking point) was taken as a measure of their motivation to obtain ethanol. The study revealed that Harlan-Fischer rats showing most anxiety-related behaviour in the elevated plus maze test displayed the lowest ethanol intake [g/kg/d b.w.] and the lowest breaking points in the progressive ratio paradigm. The Wistar-Harlan rats with least anxiety-related behaviour and the Wistar-BgVV rats with medium anxiety-related behaviour drank more alcohol and showed higher breaking points than the Harlan-Fischer rats. Thus, in the present study, a distinct relationship between innate anxiety and alcohol-consuming behaviour in rat strains not genetically selected for high and low ethanol intake could not be shown.