The present experiments were designed to investigate the neuronal mechanisms, at the level of the primary somatosensory cortex, which underlie the observation that somatosensory cortical potentials evoked by air puff stimuli directed at the forearm are decreased, in a nonspecific and widespread manner, during voluntary movements about the elbow. Unitary discharge was recorded from 131 cells receiving cutaneous input from the hairy skin of the forearm or hand (areas 3b and 1) of two monkeys trained to perform rapid movements of the contralateral arm (elbow flexion or extension). Evoked unitary responses to air puff stimuli applied to the centre of the cell's receptive field, at various delays before and after the onset of movement, were recorded. Movement produced a significant decrease in the short latency excitatory response to the air puff in 89% of the cells (117/131); the remaining 11% were not modulated by movement. This movement-related "gating" of cutaneous inputs occurred regardless of the response pattern of the cells to movement alone, being observed in 91% of the cells with no movement-related discharge, and 89% of those with movement-related discharge. The air puff responses of cells with inputs from the forearm and the dorsum of the hand were all similarly modulated by movement and the modulation was clearly present prior to the onset of movement (mean onset, -66 ms). Variation in the depth of modulation as a function of the direction of the movement, flexion or extension, was observed in only a very small proportion of the modulated units (16/117); most showed no relationship to direction. It is suggested that, in this experimental situation, much of the modulation appears to occur at a pre-cortical level since there was no relationship between the pattern of discharge of cells in relation to movement alone and the pattern of movement-related gating of their responses to the air puff. Effects which might be consistent with a cortical origin for the modulation were only infrequently observed. The present results are strikingly similar to those obtained using the evoked potential method, and thus support the hypothesis that in this task of rapid elbow movements, movement modulates the transmission of cutaneous signals from the hairy skin of the distal forelimb to primary somatosensory cortex in a nonspecific and widespread fashion.