Using microdialysis, we examined the effects of the antidepressant drug fluoxetine on 5-hydroxytryptamine (5-HT) output in rat brain. Fluoxetine (1, 3 and 10 mg/kg i.p.) dose dependently increased 5-HT output in the dorsal and median raphe nuclei and four forebrain areas. Maximal elevations were noted in the raphe nuclei. At 1 and 3 mg/kg, fluoxetine elicited minor or no increases of 5-HT output in the forebrain. When citalopram was present in the perfusion fluid, fluoxetine (10 mg/kg) reduced 5-HT output, an effect reversed by the administration of the selective 5-HT1A receptor antagonist ¿N-[2-(4-(2-methoxyphenyl)-1-piperazinyl) ethyl]-N-(2-pyridyl) cyclohexane carboxamide.3HCl¿ (WAY 100635). This reduction was more marked in the frontal cortex than in the dorsal hippocampus. Consistent with this, WAY 100635 potentiated the effect of 3 and 10 mg/kg fluoxetine more in the frontal cortex than in the dorsal hippocampus. The administration of a combination of WAY 100635 (0.3 mg/kg s.c.) and the 5-HT1B/1D receptor antagonist ¿N-[4-methoxy-3-(4-methyl-1-piperazinyl)phenyl]-2'-methyl-4'-(5-methyl-1 ,2,4-oxadiazol-3-yl),[1,1-biphenyl]-4-carboxiamide¿ (GR 127935; 5 mg/kg s.c.) potentiated the effect of 3 mg/kg fluoxetine to an extent similar to that of WAY 100635 alone in both areas. These results suggest that somatodendritic 5-HT1A receptors offset the effect of fluoxetine in the frontal cortex but not (or to a lesser extent) in the dorsal hippocampus. GR 127935 may have a partial antagonistic action at terminal 5-HT autoreceptors in vivo.