Single unit electrical activity has been recorded from dorsal horn neurones in the lumbar spinal cord of adult rats which had been treated at birth with either capsaicin (50 mg kg-1) or with the solvent-vehicle only. The responses of these neurones to electrical stimulation of A- and C-fibres in the sural nerve and to natural stimulation of their cutaneous receptive fields have been studied. In vehicle-injected rats, 54% of the units driven by electrical stimulation of the A-fibres in the sural nerve could also be driven by stimulation of the C-fibres in this nerve. In capsaicin-treated animals, only 30% of such units had a C-fibre input from the sural nerve. In vehicle-injected rats, 51.5% of the neurones with a C-fibre input showed a 'wind-up' effect on repetitive C-fibre stimulation of the sural nerve at 1 Hz. A similar proportion of neurones (55%) displayed this effect in capsaicin-treated rats. There were fewer neurones with very intense 'wind-up' in capsaicin-treated compared to vehicle-treated rats. In capsaicin-treated animals, greater proportions of neurones with 'wind-up' were superficially located in the dorsal horn, had small receptive fields and were driven only by cutaneous nociceptors. The proportions of neurones driven by innocuous mechanical stimulation of the skin, by noxious mechanical stimulation or by both forms of stimulation were similar in vehicle-injected and capsaicin-treated animals. In capsaicin-treated rats, more neurones had 'medium-sized' receptive fields than in vehicle-injected rats. In capsaicin-treated rats, more neurones had receptive fields in the foot and ankle than in vehicle-injected animals, where receptive fields in the toes were predominant. Some neurones showed expanded receptive fields after repetitive electrical stimulation of C-fibres at 1 Hz. This expansion occurred more often in neurones recorded from capsaicin-treated animals than in those of vehicle-injected rats. These results are discussed in relation to the role of afferent C-fibres in sensory mechanisms.