1. The effects of electrical stimulation of cervical vagal afferent fibers on the nociceptive tail-flick (TF) reflex and responses of spinal dorsal horn neurons to noxious cutaneous stimulation were studied in adult rats treated as neonates with either capsaicin or vehicle. 2. Vagal afferent stimulation (VAS) produced biphasic, intensity-dependent effects on the TF reflex in vehicle-treated and untreated control rats. The TF reflex was facilitated in both groups of rats at lesser intensities of VAS (2.5-50 μA) and fully inhibited at greater intensities of VAS (50-100 μA). In contrast, biphasic effects of VAS on the TF reflex generally were not produced in rats treated as neonates with capsaicin. Facilitation of the TF reflex was produced in these rats by lesser intensities of VAS as well as by typically 'inhibitory' intensities of VAS; the TF reflex was never inhibited in 6/12 rats, even at the greatest intensity of VAS tested (1,000 μA). When the TF reflex was inhibited by VAS in capsaicin-treated rats, the intensities required were significantly greater than those required in vehicle-treated or untreated rats. 3. In electrophysiological experiments, 77 neurons were recorded in the lumbar spinal dorsal horn of pentobarbital sodium-anesthetized, paralyzed rats treated as neonates with either vehicle or capsaicin. The neurons had receptive fields on the glabrous skin of the plantar surface of the ipsilateral hind foot, and all responded to mechanical stimuli of both nonnoxious and noxious intensities; 16/77 neurons also responded to noxious thermal stimulation. In vehicle-treated rats, nociceptive responses of 50% of 30 units studied were biphasically modulated by VAS, 33% were only inhibited, and 17% were only facilitated by VAS at the intensities tested (5-500 μA). In capsaicin-treated rats, nociceptive responses of 32% of 47 units studied were biphasically modulated by VAS, 15% were only inhibited, and 34% were only facilitated by VAS at the intensities tested (5-500 μA). In addition, nociceptive responses of neurons facilitated at lesser intensities of VAS and not affected at greater intensities of VAS were observed in capsaicin- treated rats (19% of the 47-unit sample). Overall, the proportion of the neuronal sample inhibited by VAS was less, and the proportion of the sample facilitated by VAS was greater in capsaicin-treated rats compared with vehicle-treated rats. 4. The efficacy of the capsaicin treatment was evaluated immunocytochemically. In vehicle-treated rats, substance P (SP) and calcitonin gene-related peptide (CGRP)-like immunoreactivity was localized in axons and terminals densely distributed in the superficial dorsal horn of the lumbar spinal cord and Lissauer's tract. SP- and CGRP-like immunoreactivity was also present in the nucleus tractus solitarius and the spinal trigeminal tract. Neonatal treatment with capsaicin clearly reduced both SP- and CGRP- like immunoreactivity in the spinal cord, spinal trigeminal tract, and the nucleus tractus solitarius. 5. The present results confirm the powerful modulatory effects of cervical VAS on spinal nociceptive reflexes and spinal nociceptive transmission. VAS-produced inhibition of these response measures was selectively attenuated, whereas VAS-produced facilitation was more readily produced in rats treated as neonates with capsaicin. Thus VAS- produced inhibitory effects on spinal nociception likely depend on capsaicin- sensitive, C-fibers in the vagus nerve, whereas VAS-produced facilitatory effects on nociception are independent of these afferents.