Biochemical and electrophysiological approaches were used to assess possible changes in 5-HT1A receptors in the rat brain after long-term treatment with an anxiolytic benzodiazepine. Rats were treated with diazepam (2 mg/kg i.p. daily) during 14 days and then untreated for 1 day (protocol A) or 5 days (protocol C) until they were killed for in vitro investigations on 5-HT1A receptors. In addition, other rats (protocol B) received the same 14-day treatment with diazepam, followed by 1 mg/kg of the drug on days 15 and 16, and 0.5 mg/kg on days 17 and 18, and were killed 24 h after the last injection. In vitro binding and quantitative autoradiographic experiments with [3H]8-hydroxy-2-(di-n-propylamino)tetralin ([3H]8-OH-DPAT) showed that the characteristics of 5-HT1A receptor binding sites in the hippocampus and the dorsal raphe nucleus were not significantly altered by the administration of diazepam under the treatment protocols A, B and C. Furthermore, in vitro electrophysiological recordings of serotoninergic neurons in the dorsal raphe nucleus of brain stem slices revealed no modification in the sensitivity of somatodendritic 5-HT1A autoreceptors in rats treated with diazepam according to the protocols A and B. However, under the conditions of protocol C, the potency of 8-OH-DPAT to depress the firing rate of serotoninergic neurons was significantly enhanced, as expected of a hypersensitivity of somatodendritic 5-HT1A autoreceptors. These data support the hypothesis that some functional changes in these receptors could occur during benzodiazepine withdrawal. However, they do not support the idea of a reduced anxiolytic efficacy of 5-HT1A receptor agonists as a result of prior treatment with a benzodiazepine.