The Vestibular Nuclei and Their Connections, Anatomy and Functional Correlations

Abstract

NATURE 97 dromically by stimulation of the lumbar cord were located in the caudal half of the nucleus. A discontinuity in the rising phase of the response of some cells could be seen with frequencies of stimulation of 50-200isec, while approximately half the units showed such a discontinuity at 400/sec. Only infrequently was it possible to observe in isolation a component resembling the A spike of a motoneurone. In 10 cells attempts were made to separate the spike into two components by stimulation with paired shocks at different intervals. It was regularly found that at intervals of 1·5-2·5 msec the second spike became somewhat broader and smaller, the A-B discontinuity was emphru;ized, and the latency was slightly lengthened. It was never possible to obtain an A spike without a B spike and at intervals of 0·5-1·3 msec the second spike failed completely. It can therefore be concluded that the response of lateral vestibular neurones consists of two components but that, although it is sometimes possible to separate these two components with double shocks in cells that have been penetrated by a microelectrode 5 , there is a high safety factor for conduction between the two regions, as there is in cells of the reticular formation•. The effect of single shocks, delivered to various ipsi-lateral leg nerves at 1/sec, on the activity of lateral vestibular cells has been observed for 32 neurones. The effectiveness of the stimuli was usually determined by counting the number of spikes occurring during a sweep, using counting circuits previously described 7 • Of the 32 cells 19 were found to b e excit ed while 1 was inhibited. Most of the unaffected cells were found in preparations in which no cells were affected and in which there was little spontaneous activity, suggesting that the condition of the animal is an important factor. In general, stimuli to mixed or cutaneous nerves were usually effective, while stimuli to muscle nerves increased the activity very little and often had no influence at all. Thus stimuli to the superficial peroneal nerve were effective 11/ 12 times; to the plantar nerve 14/ 15 times; to the sural nerve 14116 times; and the stimulus often raised the number of spikes occurring in a 100-or 200-msec period following the stimulus by 50-100 per cent. Stimuli to nerves, such as biceps anterior, biceps posterior-semitendinosus, and gastrocnemius-soleus, were often without effect and rarely increased the number of spikes by more than 10-20 per cent. Stimulation of contralateral nerves was also effective, but the action was usually somewhat weaker than that of ipsilateral nerves. In the three cases in which it has been tried, stimulation of the superficial radial nerve of the ipsilateral fore-limb also proved effective. Usually a stimulus to a leg n erve increased the number of spikes simply by causing spikes to occur more often than they did during spontaneous firing. In several cells, however , such as that shown in Fig. 1B and 10, the stimulus resulted in a repetitive discharge of the cell and the pattern of this discharge was quite different from the pattern of spontaneous activity. In such cases the latency of the first spike from the arrival of the afferent volley at the cord was quite long compared to the latency of anti-dromic firing, usually 10-35 msec. Activation of lateral vestibular cells by stimuli to cutaneous or mixed nerves has been obtained with shocks as weak as 1·2-1·4 times the threshold of the largest fibres in the nerve (1·2-1:4T). The intensity of the effect then grew as the shock was strengthened to 15-20T. The threshold for contralateral activation has been of the same order of magnitude. In two cases where it has been possible to measure the threshold of the effect from muscle n erves, shocks greater than 8T were r equired. These experiments demonstrate that it is possible to modify the activity of lateral vestibular cells located in the 'hind-limb area' of the nucleus by stimulation of p eripheral nerves in animals in which the cerebellum has been removed. In these preparations the effect is usually excitatory, as it is in decerebrate cats with the cerebellum intact; and the degree of convergence as well as the cutan-eous threshold of activation appear similar in the two types of preparation 8 • The action could be subserved in part by spinovestibular fibres 9 • As these fibres course in the dorsolateral quadrant of the cord, it would be expected that their inputs would be mainly ipsilateral' 0 , although some fibres with bilateral input are present in this regionn. The bilateral nature of the effect that we have observed suggests that ventral or ventrolateral pathways, the input of which can be expected to b e bilateral' 0 , may be of importance. Such pathways could involve spino-reticular and spino-olivary fibres 12 • Some afferent fibres from the r eticular formation to the vestibular nuclei have b een described 13 as have collaterals from oli vo-cerebellar fibres 14 • It may be noted that although parts of the reticular formation may be involved, the thresholds for vestibular activation by cutaneous n erves observed in our experiments are lower than those associated with the arousal r eaction 15 •