We have examined the anxiolytic activity of acute and chronic antidepressant treatment in an animal model of anxiety involving novelty-suppressed feeding. Rats were food deprived for 48 h, placed into a novel environment containing food, and the latency to begin eating was recorded. Chronic (21 days), but not acute injections of desipramine (DMI; 10 mg/kg) and amitriptyline (AMI; 10 mg/kg) significantly reduced the latency to begin eating compared to controls, but the percentage decrease was not as great as that seen with either acute or chronic treatment with diazepam (2 mg/kg) or adinazolam (20 mg/kg). A time course study indicated that at least 2 weeks of treatment was necessary to observe a significant anxiolytic effect of antidepressants. The anxiolytic effect of the antidepressants was specific to the novel environment, as 2 weeks of treatment with either diazepam or DMI did not influence the latency to begin eating in the home cage. Finally, a single dose of the central benzodiazepine receptor antagonist, Ro15-1788 (20 mg/kg), given 15 min prior to testing, did not block the anxiolytic effects of chronic DMI, while it completely eliminated the effect of chronic diazepam treatment. These data suggest that antidepressants acquire anxiolytic properties following chronic administration and that this effect appears to be independent of the benzodiazepine receptor system.