SUMMARY1. A method is described which permitted stable extracellular recordings to be made from 115 neurones in nucleus interpositus of the cerebellum in unanaesthetized free-to-move cats. At least 95 % of the neurones were cerebellar efferent cells since they were antidromically invaded following electrical stimulation of the brachium conjunctivum in the region of the contralateral red nucleus.2. In cats in a state of quiet wakefulness the majority of interpositus neurones were tonically active at rates ranging from 12 to 77 impulses/sec (over-all mean 34/sec). The remaining neurones were silent or discharged only a few impulses throughout observation periods of a few minutes.3. Cutaneous afferent volleys elicited by single shocks to the superficial radial nerves in the forearm at intensities too weak to evoke a flexion reflex or behavioural arousal produced changes in firing frequency in 62 % of eighty-one cells tested. Response patterns varied widely but in 86 % of the responding cells the earliest change was a short latency (6-20 ms) increase in discharge probability which from post-stimulus time histograms was found usually to average around one impulse per stimulus. In only four cells (8 %) the earliest response was a depression of the tonic firing. However, in many cells the initial acceleration was followed by a reduction in firing frequency which lasted between 10 and 85 ms.4. In 56 % of the responding cells a longer latency (25-80 ms) acceleration was present. Such accelerations varied widely in duration (from 55 to 550 ms) but most commonly lasted 100-200 ms. These responses were usually the most prominent feature in the response pattern: in the majority of neurones between two and five impulses were added per stimulus.5. Considering the whole time course of the responses, the net effect of nerve volleys was to produce an increase in nuclear cell output.6. These neurones which were influenced by nerve stimulation also discharged in response to taps to the forepaws.7. The responses to nerve stimulation are compared with those encountered in previous studies using cats anaesthetized with chloralose or barbiturates and with the responses ofPurkinje (P) Thach, 1968Thach, , 1970Mortimer, 1975) and cats (McElligott, 1976;Burton & Onoda, 1977) and these studies have revealed brisk tonic activity in stationary animals plus phasic alterations in the discharge of many units during conditioned movements of the limbs. Burton & Onoda (1977) have also presented evidence that activity patterns during forelimb movement in the cat are strongly influenced by input from rate sensitive movement detectors in the moving limb. This last is in good accord with the abundant evidence (see Bloedel, 1973; Oscarsson, 1973;Armstrong, 1974) that there are powerful inputs from deep (and cutaneous) receptors in the limbs to the 'intermediate' zone of the cerebellar cortex, the P cells of which project to and inhibit the ip. neurones (Voogd, 1964;Armstrong & Schild, 1978;Ito, Yoshida, Obata, Kawai & Udo, 1970).So far, studies in fr...