1. Using two magnetic stimulators, we investigated the effect of a conditioning magnetic stimulus over the motor cortex of one hemisphere on the size of EMG responses evoked in the first dorsal interosseous (FDI) muscle by a magnetic test stimulus given over the opposite hemisphere. 2. A single conditioning shock to one hemisphere produced inhibition of the test response evoked from the opposite hemisphere when the conditioning-test interval was 5-6 ms or longer. We shall refer to this as interhemispheric inhibition. However, the minimum latency of inhibition observed using surface EMG responses may have underestimated the true interhemispheric conduction time. Single motor unit studies suggested values 4-7 ms longer than the minimum interval observed with surface EMG. 3. Interhemispheric inhibition was seen when the test muscle was active or relaxed. Increasing the intensity of the conditioning stimulus increased the duration of inhibition: increasing the intensity of the test stimulus reduced the depth of inhibition. 4. The conditioning coil had to be placed on the appropriate area of scalp for inhibition to occur. The effect of the conditioning stimulus was maximal when it was applied over the hand area of motor cortex, and decreased when the stimulus was moved medial or lateral to that point. 5. The inhibitory effect on the test stimulus probably occurred at the level of the cerebral cortex. In contrast to the inhibition of test responses evoked by magnetic test stimuli, test responses evoked in active FDI by a small anodal electric shock were not significantly inhibited by a contralateral magnetic conditioning stimulus. Similarly, H reflexes in relaxed forearm flexor muscles were unaffected by conditioning stimuli to the ipsilateral hemisphere. However, inhibition was observed if the experiment was repeated with the muscles active.