1. In the anaesthetized cat, electrical stimulation of the bulbar reticular formation produced a short latency (2 1 + 0 3 ms) positive potential in the cord dorsum. In contrast, stimulation of the nucleus raphe magnus with strengths below 50 ,uA evoked a slow negative potential with a mean latency of 5-5 + 0-6 ms that persisted after sectioning the contralateral pyramid and was abolished by sectioning the ipsilateral dorsolateral funiculus. 2. The field potentials evoked by stimulation of the bulbar reticular formation and of the nucleus raphe magnus had a different intraspinal distribution, suggesting activation of different sets of segmental interneurones. 3. Stimulation of these two supraspinal nuclei produced primary afferent depolarization (PAD) in single Ib fibres and inhibited the PAD elicited by group I volleys in single Ia fibres. The inhibition of the PAD of Ia fibres produced by reticulospinal and raphespinal inputs appears to be exerted on different interneurones along the PAD pathway. 4. It is concluded that, although reticulospinal and raphespinal pathways have similar inhibitory effects on PAD of Ia fibres, and similar excitatory effects on the PAD of Ib fibres, their actions are conveyed by partly independent pathways. This would allow their separate involvement in the control of posture and movement.Studies on the possible role of the nucleus raphe magnus (NRM) and of the adjacent brainstem reticular formation (RF) in the control of sensory information, including nociception, have established that electrical stimulation of the RF and NRM reduces the spontaneous and evoked activity of dorsal horn and spinothalamic neurones that receive nociceptive and non-nociceptive inputs (McCreery & Bloedel, 1975;Mokha, McMillan & Iggo, 1985; see Gebhart & Randich, 1990). On the basis of the similarity of the effects exerted by the stimulation of the NRM and of the RF, it has been suggested that both nuclei are part of a common functional system that regulates nociceptive and non-nociceptive information (Anderson, 1984;Hammond & Yaksh, 1984;Willis, 1984). Activation of the RF and of the NRM also affects motor activity (Peterson, 1984). It produces mono-and polysynaptic excitatory and inhibitory potentials in spinal motoneurones (Jankowska, Lund, Lundberg & Pompeiano, 1968;Peterson, 1984;Fung & Barnes, 1989;Takakusaki, Ohta & Mori, 1989), as well as dorsal root potentials (DRPs), the latter suggesting presynaptic inhibitory actions (Lundberg & Vyklicky, 1966;Proudfit & Anderson, 1974;Jimenez, Rudomin & Solodkin, 1987). At the time when this research was started, the available evidence indicated that electrical stimulation of the bulbar RF produced primary afferent depolarization (PAD) of tendon organ and of cutaneous afferents, and inhibited the PAD generated in muscle spindle afferents (Rudomin, Jimenez, Solodkin & Dueiias, 1983;Rudomin, Solodkin & Jimenez, 1986;Jimenez et al. 1987;Jimenez, Rudomin & Solodkin, 1988;Rudomin, 1991 City, Mexico), diluted with isotonic saline (1:10) or dextran (10 %) was infused...