We used whole cell current- and voltage-clamp recording in neonatal rat brain stem slices to characterize firing properties and effects of serotonin (5-HT) on neurons (n = 225) in raphe pallidus (RPa) and raphe obscurus (ROb). Of a sample of 51 Lucifer yellow-filled neurons recovered after immunohistochemical processing to detect tryptophan hydroxylase (TPH), 34 were found to be TPH immunoreactive (i.e., serotonergic). Serotonergic neurons had long-duration action potentials and fired spontaneously at low frequency (approximately 1 Hz) in a pattern that was often irregular; at higher firing frequencies the discharge became more regular. These neurons displayed spike frequency adaptation, with maximal steady-state firing rates of < 4 Hz. The overwhelming majority of identified serotonergic neurons was hyperpolarized by bath-applied 5-HT (94%; n = 32 of 34); conversely, most cells in this sample that were hyperpolarized by 5-HT were serotonergic (78%; n= 32 of 41). TPH-immunonegative neurons were separated into two populations. One group had properties that were indistinguishable from those of serotonergic caudal raphe neurons. The other group was truly distinct; those neurons had more hyperpolarized resting membrane potentials, were not spontaneously active, had shorter-duration action potentials, and were depolarized by 5-HT. Caudal raphe neurons responded to 5-HT (1-5 microM) with membrane hyperpolarization in current clamp (-13.4 +/- 1.1 mV, mean +/- SE) or with outward current in voltage clamp (16.0 +/- 1.4 pA). The current induced by 5-HT was inwardly rectifying and associated with an increase in peak conductance and was highly selective for K+. It was completely blocked by 0.2 mM Ba2+ but not by glibenclamide, an inhibitor of ATP-sensitive K+ channels. Effects of 5-HT were dose dependent, with an EC50 of 0.1-0.3 microM. The 5-HT1A agonist 8-OH-DPAT mimicked, and the 5-HT1A antagonists (+)WAY 10,0135 and NAN 190 blocked, effects of 5-HT. The 5-HT2A/C antagonist ketanserin did not inhibit the effects of 5-HT. Fewer 5-HT-responsive neurons were encountered in slices exposed acutely to pertussis toxin (approximately 13%) than in adjacent control slices not exposed to pertussis toxin (approximately 85%). In addition, in neurons recorded with pipettes containing GTP gamma S (0.1 mM), 5-HT induced an inwardly rectifying current that did not reverse on washing. In many cells recorded with GTP gamma S, a current developed in the absence of agonist that had properties identical to those of the 5-HT-sensitive current; when followed for extended periods, the agonist-independent GTP gamma S-induced conductance desensitized, returning toward control levels with a time constant of approximately 18 min. Together these results indicate that serotonergic neurons of ROb and RPa are spontaneously active in a neonatal rat brain stem slice preparation and that hyperpolarization of those neurons by 5-HT1A receptor stimulation is due to pertussis toxin-sensitive G protein-mediated activation of an inwardly rectifying K+ conductance. In addition, we identified a group of nonserotonergic medullary raphe neurons that had distinct electrophysiological properties and that was depolarized by 5-HT.