This study examined whether the antinociception produced by electrical stimulation of medullary raphe nuclei is mediated by activation of monoaminergic neurons projecting to the spinal cord. Ninety-four sites distributed about the midline of the medulla were stimulated and their ability to produce antinociception was determined using two different analgesiometric tests. Stimulation of sites in the nuclei raphe pallidus and raphe obscurus did not produce antinociception, but rather, produced tremor and, on occasion, extensor rigidity. In contrast, stimulation at sites located in the nucleus raphe magnus and the adjacent nucleus reticularis paragigantocellularis produced antinociception, as indicated by increased tail flick latencies and decreased responsiveness to noxious pinch applied to the extremities. Intrathecal administration of saline prior to electrical stimulation of these sites did not attenuate either the elevation of tail flick latencies or the decreased responsiveness to pinch. However, intrathecal administration of 30 micrograms of either methysergide or phentolamine prior to stimulation at these same sites significantly attenuated the increase in tail flick latency and restored responsiveness to pinch. Naloxone (1 mg/kg, s.c.) did not attenuate the stimulation-produced antinociception evoked from any of these sites. These data support the postulate that the antinociceptive effect of electrical stimulation of the nucleus raphe magnus and the nucleus reticularis paragigantocellularis is mediated by activation of serotonergic and noradrenergic neurons projecting to the spinal cord. The inability of either methysergide or phentolamine alone to completely antagonize the elevation of tail flick latency further suggests that the serotonergic and noradrenergic bulbospinal systems are coactivated by electrical stimulation of these sites.