Reciprocal inhibition is the automatic antagonist alpha motor neurone inhibition which is evoked by contraction of the agonist muscle. This so-called natural reciprocal inhibition is a ubiquitous and pronounced phenomenon in man and must be suspected of playing a major role in the control of voluntary movements. The spinal pathways underlying this inhibitory phenomenon were studied. The disynaptic reciprocal Ia inhibitory pathway between the tibial anterior muscle and the soleus alpha motor neurones was identified and described in man. It was shown that the inhibition can be evoked in most healthy subjects at rest, but the degree of inhibition varies considerably from one subject to another. It was concluded that it corresponds to the disynaptic reciprocal Ia inhibitory pathway which has been extensively described in animal experiments. The disynaptic reciprocal inhibition was shown to increase during the dynamic phase of a dorsiflexion movement of the foot, but not during the tonic phase. However, when the peripheral afferent feedback from the contracting muscle was blocked by ischaemia, an increase of the inhibition was revealed also during the tonic phase of the dorsiflexion. The concealment of this increase during unrestrained peripheral feedback from the muscle was thought to be due to the post-activation depression mechanism; a mechanism which was described further and which probably involves reduced transmitter release at Ia afferent terminals as a result of previous activation of these afferent fibers. Hence the hypothesis was supported that alpha motor neurones and the corresponding inhibitory interneurones, which project reciprocal inhibition to the antagonist motor neurones, are activated in parallel during voluntary contraction of agonist muscles. An additional reciprocal inhibitory mechanism, the long latency reciprocal inhibition, was described between the tibial anterior muscle and the soleus alpha motor neurones. It was shown to be evoked by group I afferent activity and it was seen at a conditioning-test interval of 3-6 msec. It was initiated by supraspinal pathways and it was seen only during dorsiflexion of the foot. The pathway underlying this inhibitory mechanism is unknown, but it was suggested that it was mediated by a propriospinal pathway. The possible contribution to reciprocal inhibition of presynaptic inhibition of soleus Ia afferent fibers was studied by an indirect method. It was concluded that this presynaptic inhibition was increased during even slight dorsiflexion of the foot and that the increase was mainly dependent upon the peripheral afferent input from the contracting muscle.(ABSTRACT TRUNCATED AT 400 WORDS)