The characteristics of the serotonin (5-HT) output in the dorsal and median raphe nuclei of the rat were studies using in vivo microdialysis. The basal output of 5-HT increased after KCl was added to the perfusion fluid. In contrast, neither the omission of calcium ions nor the addition of 0.5 microM tetrodotoxin affected dialysate 5-HT or 5-hydroxyindoleacetic acid (5-HIAA). Reserpine did not decrease the output of 5-HT and 5-HIAA 24 h later and p-chloroamphetamine increased 5-HT in both vehicle- and reserpine-treated rats severalfold. 8-Hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT), at 1 or 10 microM, perfused into the raphe did not change the outputs of 5-HT or 5-HIAA. Higher doses (0.1, 1, and 10 mM) increased extracellular 5-HT in the raphe, probably via an inhibition of uptake. In animals bearing two probes (raphe nuclei and ventral hippocampus), only the 10 mM dose of 8-OH-DPAT perfused into the raphe decreased the hippocampal output of 5-HT and 5-HIAA. The systemic injection of 0.1 mg/kg 8-OH-DPAT decreased dialysate 5-HT and 5-HIAA in the raphe and hippocampus. These results suggest that extracellular 5-HT in raphe nuclei originates from a cytoplasmic pool and is not dependent on either nerve impulse of 5-HT neurons or local activation of 5-HT1A receptors.