We recorded the motor evoked potentials (MEPs) from the abductor pollicis brevis muscle, after supramaximal electrical transcranial stimulation, and studied the effect of paired transcranial shocks with varying interstimulus time intervals, in 10 normal subjects, 4 patients with median nerve neuropathy and 2 patients with motoneurone disease. In relaxed muscles the amplitude of the MEP evoked by a single shock averaged 30% of the M wave. With intervals from 1 to 2.5 msec 2 shocks evoked one MEP far larger in size than the control MEP (70% of the M wave). With intervals of 10 msec and longer, the 2 shocks evoked 2 independent MEPs; the size of the MEP following the second shock (test) was inversely correlated with the size of the control MEP: the more the control MEP approached the size of the M wave, the smaller the test MEP. Single motor unit records showed that, in the normal subjects and patients with peripheral neuropathy, the same motor unit was activated either by the first or the second shock, whereas in the patients with motoneurone disease it fired twice. In active muscles, the control MEP averaged 70% of the M wave. With intervals of 10 msec and longer the test MEP was markedly suppressed; with 100 msec intervals it fully recovered. In relaxed muscles, by delivering a double shock at a 1.5 msec interval, thus evoking a large MEP, followed by a second double-shock, the test MEP was completely suppressed for a period of 20 msec; it began to recover at 50 msec intervals and fully recovered after 150 msec. These results indicate that: (1) high-threshold spinal motoneurones can profit from temporal summation if double-shocks are delivered at short time intervals; (2) the synchronous excitation of the motoneuronal pool produced by transcranial stimulation is followed by a 20 msec period of absolute inhibition, possibly through a massive activation of the Renshaw system; (3) during voluntary contraction, only a portion of the motoneuronal pool remains refractory, possibly because of the enhanced spinal excitability.