Antidepressant drugs are effective in the treatment of panic in human panic disorder patients. One hypothesis is that the anti-panic effects of antidepressant drugs are mediated by an increase in the activity of serotonergic neurons within dorsal raphe nucleus (DRN) leading to an increase in serotonin-mediated inhibition of the dorsal periaqueductal gray (DPAG). In order to test this hypothesis, we investigated the effects of reversible inhibition of the DRN on behavior in the elevated T-maze (ETM) in control rats and rats chronically treated with imipramine. Rats were injected daily with imipramine (15 mg/kg i.p.) or saline for 24 days. A guide cannula was implanted in the DRN on day 14. Lidocaine (4%, 0.2 microL) or saline was injected into the DRN 10 min before the test in the ETM followed immediately by the open-field test (day 21). Three days later, the infusions were crossed-over, rats microinjected into the DRN with saline in the first trial received lidocaine and vice versa, and the behavioral tests were repeated (day 24). Chronic saline plus lidocaine in the DRN and chronic imipramine (plus saline or lidocaine in the DRN) impaired inhibitory avoidance, indicating an anxiolytic effect. In the one-way escape, lidocaine facilitated it, suggesting a panicogenic effect, while chronic imipramine impaired it, which is indicative of a panicolytic effect. Moreover, lidocaine blocked the facilitatory effect of chronic imipramine. The locomotor activity in the open field was not changed by any treatment compared to the control group. These effects were congruent with the hypothesis that the DRN has a dual effect on anxiety: increasing learned anxiety and decreasing innate anxiety. Moreover, they suggest that the DRN exerts a crucial role in the antipanic-like effect of chronic imipramine in the ETM.