This paper reports the effects of a diazepam treatment on voluntary ethanol intake in rats included in an animal model of alcoholism. In a first dose-seeking experiment, rats had a choice between 10% (w/v) ethanol and water for 24 h each week. Single doses of diazepam between 2 and 20 mg/kg injected i.p. prior to the 24-h choice caused a dose-dependent decrease in voluntary ethanol intake from 3.2 +/- 0.4 g/kg/day down to 2.3 +/- 0.3 g/kg/day (P < 0.01) after a dose of 20 mg/kg. In a second experiment, psychological dependence was induced by a 1-year intermittent exposure to ethanol (a choice between 10% ethanol and water for 24 h each week, followed by an i.p. injection of 2.0 g/kg of ethanol). After this year, the rats were given a continuous choice between ethanol and water. A 3-week treatment with diazepam (20 mg/kg/day, i.p.) was started in week 68, during which period a choice of 10% (w/v) ethanol was available only on the first and the last days of treatment. On the first day of the diazepam treatment, ethanol intake was decreased from a pre-experimental value of 2.7 +/- 0.3 g/kg/day to 1.2 +/- 0.1 g/kg/day (P < 0.001). On the last day of the treatment, voluntary intake was higher than before the treatment (3.8 +/- 0.27 g/kg/day, P < 0.01). Ethanol intake remained elevated during the week after the end of the diazepam treatment (P < 0.05). When single doses of diazepam (20 mg/kg) were re-tested 10 and 19 weeks after the treatment, there was no decrease in ethanol intake, indicating that the initial effect had not been re-established.