Presynaptic inhibition of Ia terminals and postactivation depression at the Ia fibre-motor neuron (MN) synapses were compared in the upper and lower limbs of both sides in subjects from different populations: 49 spastic patients with hemiplegia [mainly with a lesion in the middle cerebral artery (MCA) area], two tetraplegics and 35 healthy subjects. Presynaptic inhibition was assessed using D1 inhibition of the soleus and the flexor carpi radialis (FCR) H reflexes elicited by electrical stimuli applied to the nerve supplying antagonistic muscles, and postactivation depression was explored by varying the time interval between two consecutive H reflexes. In normal subjects no right-left asymmetry was found in the amount of presynaptic Ia inhibition, homosynaptic depression or the H(max)/M(max) ratio. In the hemiplegic side of patients with MCA area lesions, the H(max)/M(max) ratio was significantly increased in the soleus but not in the FCR. Presynaptic inhibition of Ia terminals, which was significantly reduced at the cervical level on the hemiplegic side (and also, but to a lesser extent, on the unaffected side), was unchanged at the lumbar level. Homosynaptic depression was similarly reduced at the cervical and lumbar levels on the hemiplegic side but not modified on the unaffected side. It is argued that the decrease in presynaptic inhibition of Ia terminals is more a correlate of spasticity than a mechanism underlying it. The decrease in postactivation depression, which very probably contributes to the exaggeration of the stretch reflex characterizing spasticity, might be a consequence of the changes in the pattern of activation of Ia afferents and MNs following the motor impairment.

SUMMARY1. I a projections and recurrent inhibition from quadriceps to tibialis anterior and soleus motoneurones were investigated in man.2. Changes in the firing probability of individual voluntarily activated motor units were studied following electrical stimulation of the femoral nerve or quadriceps tendon tap.3. Femoral nerve stimulation evoked an early increase in the firing probability of tibialis anterior units. This excitation was also evoked by a tendon tap, had a low threshold and its central delay was estimated to be the same as that of the homonymous monosynaptic Ia excitation. These findings strongly suggest that the femoral nerve-induced excitation is I a in origin and mediated through a monosynaptic pathway.4. The frequency of heteronymous Ia excitation from quadriceps was about the same to both ankle flexor and extensor units (79 and 70 % respectively).5. In 80 % of both tibialis anterior and soleus units the I a excitation was followed by a decrease in firing probability. This inhibition had a short latency and a long duration (up to 40 ms); it always appeared with the quadriceps reflex discharge and increased with it. These findings suggest that this decrease in firing probability is due to the Renshaw inhibition evoked by the quadriceps motoneurone discharge.6. Both the Ia excitation and the following inhibition of tibialis anterior and soleus units were also observed when the stimulation was applied to the nerve of the vastus lateralis (a pure knee extensor).7. The functional significance of these identical projections from quadriceps to both ankle flexor and extensor motoneurones is discussed with regard to the requirements of bipedal stance and gait.

SUMMARY1. Reciprocal inhibition between elbow flexor and extensor muscles (biceps and triceps brachii) has been investigated in nine healthy subjects. Two techniques were used to assess changes in motoneurone excitability after stimulation of antagonist muscle afferents: (1) monosynaptic reflexes elicited by a mechanical stimulation of the distal muscle tendon (tendon tap); (2) post-stimulus time histograms (PSTH) of voluntarily activated motor units.2. Electrical stimulation of the antagonist muscle nerve produced a short-latency and short-lasting inhibition of the flexor and extensor motoneurones. The amount of this inhibition was found to be similar in both motor nuclei.3. The inhibition could be evoked with conditioning electrical stimuli as low as 0 7 x motor threshold (MT) or by very weak tendon taps applied to the antagonist tendon. In the former case the threshold of this inhibition was found to be consistently increased after raising the threshold of Ia afferent fibres by a longlasting muscle vibration. Since a contribution from cutaneous afferent fibres was ruled out, it is concluded that this inhibition was Ia in origin.4. Post-stimulus time histograms of voluntarily activated triceps and biceps motor units were made following electrical stimulation of homonymous and antagonist muscle afferents. This enabled an estimate of the central synaptic delay of the inhibitory process. An average central delay of 0-94 ms in excess of that of monosynaptic facilitation was found, thus suggesting that the inhibitory process could be mediated by only one interneurone. 5. A conditioning reflex discharge elicited in the antagonist muscle by a tendon tap depressed or suppressed this inhibition. This depression was maximal when the reflex discharge was elicited 10-20 ms before the conditioning stimulus for the inhibition and never lasted more than 30 ms. It is argued that the only mechanism compatible with such a depression is the inhibitory activity of Renshaw cells acting on the pathway mediating reciprocal inhibition.6. We conclude that group Ia afferent fibres from elbow extensor and flexor

The distribution of homonymous and heteronymous recurrent inhibition among the motor nuclei innervating the main muscles of the human upper limb has been investigated in 25 healthy subjects. Homonymous recurrent inhibition was studied with a specially designed electrophysiological method with paired H-reflexes, previously described by Bussel & Pierrot-Deseilligny (1977), combined with a pharmacological study using a cholinergic agonist, the L-acetylcarnitine (Rossi & Mazzocchio 1991). These methods were used to investigate the Flexor Carpi Radialis (FCR), Extensor Carpi Radialis (ECR), Opponens Pollicis (OP) and Abductor Digiti Minimi (ADM) motor nuclei. In the Deltoid, Triceps, Flexor Carpi Ulnaris (FCU) and Extensor Carpi Ulnaris (ECU) motor nuclei in which it was impossible to evoke clearly distinguishable H-reflexes, homonymous recurrent inhibition was studied with the PSTH technique: homonymous recurrent inhibition was found in Deltoid, Triceps, FCR, FCU, ECR, ECU motor nuclei but not in OP and ADM motor nuclei. Heteronymous recurrent inhibition was studied with the PSTH technique in the Deltoid, Biceps, Triceps, FCR, ECR, FCU, ECU, Flexor Digitorum Communis (FDC), Extensor Digitorum Communis (EDC) motor nuclei as well as those innervating the intrinsic muscles of the hand. The following results were obtained: (1) motor neurones innervating muscles acting at digits do not receive any heteronymous recurrent inhibition; (2) motor neurones innervating muscles acting at the wrist give recurrent inhibition to motor nuclei of proximal muscles but do not receive any recurrent projections from the latter; and (3) motor neurones innervating proximal muscles (acting at shoulder or elbow) are interconnected by recurrent inhibition and receive heteronymous recurrent projections from some wrist muscles but not from intrinsic hand muscles.

1. Interneurones mediating reciprocal inhibition between wrist flexors and extensors in man are characterized using both Renshaw cells and transarticular group I afferent activation. 2. Renshaw cells were activated by reflex discharges evoked by a tendon tap. The tendon tap was applied to the tendon of the muscles from which the Ia fibres responsible for the reciprocal inhibition originated. Contrary to what was observed both in the cat hindlimb and in human elbow muscles, this Renshaw cell activation never resulted in a long depression of the reciprocal inhibition between wrist flexors and extensors. 3. Convergence from group I elbow muscle afferents and antagonistic group I afferents onto interneurones mediating reciprocal inhibition between wrist muscles was revealed in post‐stimulus time histogram (PSTH) experiments using the technique of spatial facilitation. 4. The characteristics of the interneurones mediating reciprocal inhibition between wrist flexors and extensors could therefore be summarized as follows: (a) they are fed by antagonistic group I afferents and group I afferents originating from both flexor and extensor elbow muscles; (b) they are not inhibited by Renshaw cells; (c) they are not excited by low threshold cutaneous afferents; and (d) they are probably interposed in a disynaptic pathway. 5. It is therefore concluded that interneurones mediating reciprocal inhibition between wrist flexors and extensors in man differ both from Ia interneurones and from interneurones interposed in the Ib reflex pathways and these characteristics are related to the complex circumduction movements developed in the wrist.