Bidirectional neurones in the cervical enlargement of the cat spinal cord with axons descending to sacral segments and ascending to the cerebellum and the lateral reticular nucleus

Neurones located in cervical segments (C6/C7) of the spinal cord were investigated electrophysiologically in cats deeply anaesthetized with alpha-chloralose. Extracellular recordings of antidromic action potentials were performed in order to establish whether long descending propriospinal neurones projecting to sacral segments could have collateral axonal branches ascending to supraspinal centres. The effects of stimulation of the lateral reticular nucleus (LRN) and the inferior cerebellar peduncle (restiform body, RB), as well as the thirteenth thoracic (Th13) and sacral (S1/S2) segments of the spinal cord were tested in 93 cells. Two main groups of cells were identified: 54 % of the total sample were classified as purely propriospinal and 46 % as bidirectional neurones. Various patterns of projections, as well as the ipsi-, contra- or bilateral courses of axons in the lateral funiculi of the spinal cord, enabled several types of neurones to be distinguished within the above groups. Comparison between particular types showed no significant difference with respect to location in the grey matter (predominantly Rexed's laminae VII-VIII) and the conduction velocities of descending axons. However, the mean axonal conduction velocities of branches ascending to LRN and/or RB were significantly lower in comparison to those measured for spinal collaterals. The hypothetical function of the neurones examined is discussed. Since the same information can be conveyed simultaneously by these branching neurones to lower spinal segments and supraspinal centres, an integrative role in the system of motor control is suggested.     

Distribution of the sacral neurones of origin of the ascending spinal tracts with axons passing through the lateral funiculi of the lowermost thoracic segments: an experimental HRP study in the cat.

The locations of 249 cell bodies of the ascending tract neurones in the grey matter of S1-S3 segments of the spinal cord were reconstructed by histochemical staining, after their axons (or axonal collaterals) at the level of the Thl3 segment were injected with horseradish peroxidase (HRP). In three cats in which the injections of HRP were restricted to the lateral part of the lateral funiculi (llf), about 84% of 159 retrogradely labelled cells were found on the contralateral side, while about 16% were located ipsilaterally. They were the most numerous in S2, S3 and S1 segments, respectively, and the neurones were distributed mainly in the lateral laminae I-VII, medial laminae V, VI and lamina VIII. In three other animals in which the injections of the marker were limited to the dorsal part of the lateral funiculi (dlf), 84 of the 90 ascending tract neurones were found to be distributed in the S2 and S3 segments both ipsi- (lateral laminae III-V) and contralaterally, (lateral laminae IV and V as well as the medial laminae VII and VIII) in similar numbers. The remaining six of the 90 cells with only contralateral projections at the dorsolateral funiculus at Thl3 were scattered within the S1 segment. These data are consistent with the results of studies on sacral spinocerebellar, spinothalamic and spinoreticular projections, as well as the localization of sacral spinocervical and priopriospinal neurones. They may also imply the importance of the bilateral fiber course of the neurones of origin of ascending tracts in the S2 and S3 segments within the dorsolateral funiculus.     

The lateral reticular nucleus in the cat

Summary Intracellular recording from neurones in the lateral reticular nucleus (LRN) demonstrated that, in addition to the previously identified excitatory ipsilateral forelimb tract (iF tract) (Clendenin et al. 1974c) there is an inhibitory tract mediating information from the ipsilateral forelimb to the LRN. The excitatory and inhibitory tracts were similarly organized. The tract neurones were monosynaptically activated by affcrents in the ipsilateral forelimb and projected to the same area of the LRN. They will be considered as excitatory and inhibitory components of the iF tract and denoted the excitatory and inhibitory iF tract (EiF and IiF tracts). Stimulation of the descending ipsilateral dorsolateral funiculus (iDLF) in the C3 segment evoked disynaptic EPSPs and IPSPs in LRN neurones contacted by the EiF and IiF tracts. The responses in individual LRN neurones evoked from the iDLF were similar to the responses evoked from the forelimb nerves suggesting that the EiF and IiF tracts are monosynaptically activated by fibres in the iDLF. The dorsal portion of the magnocellular part of the LRN constituted the main termination area of both the EiF and IiF tracts. Neurones in this area have previously been shown to project ipsilaterally to lobule V in the pars intermedia of the cerebellar anterior lobe and to the paramedian lobule (Clendenin et al. 1974a). IPSPs evoked from the IiF tract in LRN neurones outside the main termination area had smaller amplitudes and longer latencies. This finding suggests that these responses were generated by thin axon collaterals given off from dorsally located stem axons.     

The lateral reticular nucleus in the cat

Intracellular recordings were obtained from 204 neurones in the lateral reticular nucleus (LRN). LRN neurones contacted by the bVFRT were identified by the responses evoked on stimulation of descending fibres in the contralateral ventral quadrant of the spinal cord (cVQ) at cervical (C5cVQ) and lumbar (L2cVQ) levels. Stimulation of the cVQ evoked excitatory or inhibitory responses in 124 of the 204 LRN neurones. EPSPs were evoked in 45, IPSPs in 52 and both EPSPs and IPSPs in 27 LRN neurones. The shortest latencies of the responses evoked from the cVQ indicated that both EPSPs and IPSPs were disynaptic. This finding was confirmed by direct stimulation of the ascending fibres in the ipsilateral ventrolateral funiculus at C3 (C3iVLF) or L1 (L1iVLF). In most LRN neurones activated or inhibited from the cVQ, stimulation of the iVLF evoked similar responses at a monosynaptic latency. These results indicate that the bVFRT consists of roughly equally large groups of excitatory and inhibitory neurones monosynaptically connected with the LRN. Excitatory and inhibitory bVFRT neurones had similar peripheral receptive fields and termination areas in the LRN. LRN neurones were divided into those contacted by cervical bVFRT neurones and lumbar bVFRT neurones. The former group consisted of LRN neurones responding to C5cVQ stimulation at latencies below 5 ms, whereas the latter group contained LRN neurones responding to stimulation of the L2cVQ. Cervical bVFRT neurones projected to most parts of the LRN whereas the projection of lumbar bVFRT neurones were confined to the ventrolateral part of the nucleus. Excitatory and inhibitory vVFRT neurones of each group had similar termination areas.(ABSTRACT TRUNCATED AT 250 WORDS)     

The lateral reticular nucleus in the cat

The afferent paths from the spinal cord and from trigeminal afferents to the lateral reticular nucleus (LRN) were investigated by intracellular recording from 204 LRN neurones in preparations with a spinal cord lesion at C3 that spared only the ipsilateral ventral quadrant. Stimulation of nerves in the limbs evoked EPSPs and JPSPs in 201 of 204 tested LRN neurones. The strongest input was from the ipsilateral forelimb (iF) which evoked EPSPs in 49% and IPSPs in 73% of the LRN neurones. Each of the other limbs evoked EPSPs in approximately 20% and IPSPs in approximately 25% of the neurones. Stimulation of the ipsilateral trigeminal nerve (iTrig) evoked EPSPs in 32% and IPSPs in 46% of the neurones. The shortest latencies of the EPSPs and IPSPs indicated a disynaptic connection between primary afferents in the iF and iTrig and the LRN. The most direct pathways for excitatory and inhibitory responses from the other limbs were trisynaptic. Stimulation of the ventral part of the ipsilateral funiculus (iVLF) at C3 (C3iVLF) evoked monosynaptic responses in 189 of 201 tested LRN neurones. Monosynaptic EPSPs were recorded in 104 neurones and monosynaptic IPSPs in 126 neurones. Monosynaptic EPSPs and IPSPs were encountered in all parts of the LRN. Stimulation of the iVLF at L1 (L1iVLF) evoked monosynaptic EPSPs and IPSPs in the ventrolateral part of the LRN. The termination areas of excitatory and inhibitory fibres appeared to be the same. LRN neurones without monosynaptic EPSPs or IPSPs from the L1iVLF were located mainly in the dorsal part of the magnocellular division. Stimulation of the dorsal funiculi (DF) at C2 and the ipsilateral trigeminal nerve (iTrig) evoked excitatory and inhibitory responses in the LRN. The shortest latencies of EPSPs and IPSPs indicated disynaptic connections.     

Evidence that dopamine-beta-hydroxylase immunoreactive neurons in the lateral reticular nucleus project to the spinal cord in the rat.

The existence of noradrenergic projections from the lateral reticular nucleus (LRt) to the dorsal quadrant of cervical, thoracic, or lumbar spinal cord was investigated using a combined method of WGA-apo-HRP-gold retrograde tracing and dopamine-beta-hydroxylase (DBH) immunocytochemistry. Preliminary retrograde tracing studies indicated that LRt neurons projecting to cervical, thoracic, or lumbar spinal cord were characteristically located near the perimeter of the LRt. Double-labeling experiments demonstrated that a portion of these peripherally-located, spinal-projecting neurons were DBH-immunoreactive. Double-labeled neurons were also located at the parvocellular division of the contralateral LRt in the thoracic injection cases. Double-labeled neurons were not observed at the subtrigeminal division in cervical, thoracic, or lumbar injection case. The results suggest the possibility that the noradrenergic LRt-spinal pathway might be involved in a variety of pain processing and cardiovascular regulatory functions in the rat.     

The lateral reticular nucleus in the cat—I. An experimental anatomical study of its spinal and supraspinal afferent connections

The localization of the neurons from which the main ascending and descending projections to the lateral reticular nucleus originate has been studied in nine cats, using the retrograde axonal transport of horseradish peroxidase injected within that nucleus. The ascending spinoreticular neurons are widely distributed from the cervical to the sacral segments of the spinal cord. These neurons, which are of different sizes, are mainly located within Rexed's laminae VI, VII and VIII, but they spread both dorsally to laminae II–VI as well as ventrally to lamina IX. Labeled neurons of a size similar to motoneurons are particularly found within lamina IX, intermingled with the motoneurons. The spinal projection to the lateral reticular nucleus is crossed and uncrossed, with the ipsilateral spinoreticular neurons being located more dorsally within the grey matter of the spinal cord than the contralateral spinoreticular neurons. Moreover, while neurons with a crossed ascending projection are almost equally distributed along the whole rostro-caudal extension of the spinal cord, those with an uncrossed projection are predominantly located within the cervical segments of the spinal cord. Additional evidence indicates that the spinoreticular projection to the lateral reticular nucleus is somatotopically organized.In addition to the spinoreticular projection, the lateral reticular nucleus receives a crossed rubroreticular projection and a crossed fastigioreticular projection originating from the rostro-ventralmost part of the nucleus. A few neurons in the interposite nuclei also project to the lateral reticular nucleus.     

The lateral cervical nucleus in the cat III. An electron microscopical study after transection of spinal afferents

Summary The ultrastructure of terminal degeneration within the lateral cervical nucleus (LCN) after transection of its spinal afferent fibers 2 days–2 years earlier is described. The degeneration after 2 days was of both the neurofilamentous and dense type. The highest number of degenerating terminals, about 15%, was found after 4–5 days. Then most of the degenerating boutons were of the dense type. The degenerating terminals had synaptic contact with cell bodies and dendrites of LCN-neurons. Removal of the degenerating boutons seemed to be effected by a phagocytic cell present in increased number compared to the normal LCN. In cases with long survival times an increase in the number of astroglial filaments was observed. In an animal where the spinal afferents to the LCN had been cut 2 years earlier a decrease in medium size of the neurons was observed. The amount of dendritic spines was also considerably smaller than normally.     

An electrophysiological study of thalamo-caudate neurones in the cat

Summary Thalamo-caudate projection neurones were identified in cats by antidromic activation from the caudate nucleus, under chloralose anaesthesia or in cerveau isolé preparations.Units in nuclei centralis lateralis, medialis dorsalis, and centrum medianum-parafascicularis responded antidromically to one or more electrodes of a caudate nucleus array at latencies ranging from 0.4 to 16 ms. The responses did not appear to result from stimulus spread to the internal capsule.A separate population of medial thalamic units responded post-synaptically to caudate stimulation, at modal latency exceeding modal antidromic latency by 2 ms. Comparison of apparent conduction velocities for each thalamic nucleus indicated that the postsynaptic responses were activated by collaterals of thalamo-caudate neurones.A proportion of thalamo-caudate neurones responded to somatic stimulation; some of these were inhibited by caudate stimulation, and a few discharged on substantia nigra stimulation. These results demonstrate the possibility of somatic afferent and nigrofugal inputs to the caudate nucleus mediated by neurones of the medial thalamus.     

Projections of reticular neurones to dorsal regions of the spinal cord in the cat

Reticulospinal neurones in the cat were identified by extracellular recording and antidromic stimulation of their axons in the cord. Approximately 34% of reticulospinal neurones in the medulla, and 28% in the pons, were found to project to dorsal regions of the cord, between T9 and L2. Most of these neurones had one branch situated dorsally and another in the ventral or ventrolateral funiculus. Some branches travelled for short distances in the dorsal columns. Microstimulation techniques demonstrated the presence of branches of reticulospinal fibres in the dorsal horn. The results may provide an anatomical basis for the widespread effects of stimulation of the reticular formation on afferent transmission in the spinal cord.     

Neurones in the cervical enlargement of the cat spinal cord antidromically activated from sacral segments and the inferior cerebellar peduncle

The electrophysiological investigation of neurones located in the cervical enlargement of the spinal cord was performed in eight-chloralose anaesthetized cats. Neurones were recorded intracellularly or extracellularly and identified by antidromic stimulation. The main purpose of the study was to test whether these neurones give off collateral branches ascending to the inferior cerebellar peduncle and descending to the sacral segments (S1/S2). Recordings were made from 78 neurones located in medial and central parts of Rexed's laminae VII and VIII of C6/C7 segments. Four subpopulations could be distinguished from their patterns of propriospinal or supraspinal projections: (a) ascending/descending neurones with axons ascending to RB and descending to S1/S2 (23%); (b) ascending/descending neurones projecting to RB and the level of Th13 (14%); (c) propriospinal neurones descending to Th13 (15%); (d) propriospinal neurones descending to S1/S2 (48%). Within these groups, ipsilateral, contralateral and bilateral descending projections were observed. The mean axonal conduction velocities for descending and ascending collaterals of bidirectional neurones were 59 and 39 m/s, respectively. Results suggest that parallel transmission of information to supraspinal and spinal centres plays an important role in the process of movement coordination.     

Fluorescent double-label study of lateral reticular nucleus projections to the spinal cord and periaqueductal gray in the rat.

Following injections of WGA-HRP into either the spinal cord or periaqueductal gray, labeled neurons were observed bilaterally along the periphery of the lateral reticular nucleus (LRN) magnocellular division. The possibility that some of these neurons in the LRN provide collateral axonal branches to both the periaqueductal gray and the spinal cord was investigated in rats using a retrograde double-labeling method employing two different fluorescent tracers, True Blue and Nuclear Yellow. Following sequential injection of the two fluorescent axonal tracers into the spinal cord and periaqueductal gray in the same animal, a modest number of double-labeled neurons were observed bilaterally near the medial and dorsal perimeter of the magnocellular division of the LRN. The labeled neurons were distinctly multipolar in shape and measured approximately 15-18 mu in their greatest transverse diameter. No double-labeled neurons were observed in the parvocellular or subtrigeminal divisions of the LRN. Based upon these observations, it is suggested that collaterals of the LRN-spinal pathway provide feedback information to the periaqueductal gray that might then be used to modulate the participation of the latter cell group in a variety of pain processing and cardiovascular regulatory functions.     

Properties of vestibular neurones projecting to neck segments of the cat spinal cord

1. Vestibular neurones projecting to the upper cervical grey matter (vestibulocollic neurones) were identified by localized microstimulation in the C3 segment of the cat spinal cord.

2. The neurones were found in the lateral (Deiters'), medial and descending nuclei bilaterally and projected to the spinal cord in the lateral and medial vestibulospinal tracts (LVST and MVST). Ipsilateral axons of Deiters' neurones were mostly in the LVST, axons of medial and descending neurones in the MVST; a few Deiters' neurones had axons in the MVST; some descending neurones had axons in the LVST. Most axons of contralateral neurones were in the MVST.

3. The axons of 62% of ipsilateral vestibulocollic Deiters' neurones not only gave off a collateral to C3, but also extended as far as the cervical enlargement (`branching'); some of these neurones projected as far as the upper thoracic cord, almost none to the lumbar cord. Ipsilateral descending nucleus neurones branch in the same fashion, but there is no branching in the relatively small medial nucleus population.

4. A large majority of vestibulocollic neurones receive monosynaptic excitation from the ipsilateral labyrinth and a number are inhibited by stimulation of the contralateral labyrinth (commissural inhibition). It is possible that commissural inhibition acts on a broad population of vestibular neurones involved in the control of eye, head and trunk movement.

5. Vestibulocollic neurones do not make up a homogeneous population acting only on the neck. Instead it is likely that subpopulations, for example branching and non-branching neurones, have different functions.

     

An ipsilateral projection from the red nucleus to the lateral reticular nucleus in the cat

Summary Injections of the wheat germ agglutinin — horse-radish peroxidase complex into the lateral reticular nucleus reveal that in addition to the well known contralateral rubroreticular connection, there is also a small but clear cut ipsilateral projection. Cells of various sizes participate in this ipsilateral pathway, and the retrogradely labelled neurons lie dispersed throughout the entire red nucleus.     

Electrophysiological investigation of spino-olivary projections originating from sacral segments of the cat spinal cord

Ascending projections of sacral spinal cord neurones (S1-S2) to the dorsal accessory olivary nucleus (DAO) were electrophysiologically investigated in 3 adult cats under deep alpha-chloralose anaesthesia. Antidromic action potentials were recorded extracellularly from 19 cells following stimulation of their axons in both the contralateral dorsal accessory olivary nucleus (coDAO) and the contralateral lateral funiculus at the level of lower thoracic segments (Th13). Two groups of neurones were identified in the gray matter of S1-S2 segments: one distributed in the medial part of Rexed's laminae VI and VII (n = 5), the other located in the ventromedial part of lamina VIII (n = 14). Axonal conduction velocities of neurones investigated were comprised in the range 32-55 m/s. A significant decrease of conduction velocity was observed in each case when distal and proximal parts of the axon were compared. Our research confirmed anatomical data concerning spino-olivary neurones originating from sacral segments. However, we suggest that axons of this pathway give off collaterals to other spinal or supraspinal centres.     

Arrangement of neurons in the medullary reticular formation and raphe nuclei projecting to thoracic, lumbar and sacral segments of the spinal cord in the cat

Summary The distribution of neurons in the medullary reticular formation and raphe nuclei projecting to thoracic, lumbar and sacral spinal segments was studied, using the technique of retrograde transport of horseradish peroxidase (HRP), alone or in combination with nuclear yellow (NY). Retrogradely labeled cells were observed in the lateral tegmental field (FTL), paramedian reticular nucleus, magnocellular reticular nucleus (Mc), in the gigantocellular nucleus (Gc), lateral reticular nucleus (LR), lateral paragigantocellular nucleus (PGL), rostral ventrolateral medullary reticular formation (RVR), as well as in the medullary raphe nuclei following the injection of the tracer substance(s) into various levels of the spinal cord. The FTL, the ventral portion of the paramedian reticular nucleus (PRv), Mc, LR, PGL and the raphe nuclei were found to project to thoracic, lumbar and sacral spinal segments. This projection was bilateral; the contralaterally projecting fibers crossed the midline at or near their termination site. The dorsal portion of the paramedian reticular nucleus (PRd), Gc and the RVR projected mainly to thoracic segments. This projection was unilateral. Experiments in which the HRP-injection was combined with lesion of the spinal cord showed that some descending raphe-spinal axons coursed presumably alongside the central canal. Experiments with two tracer substances suggested that some reticulo and raphe-spinal neurons had axon collaterals terminating both in thoracic and sacral spinal segments.Abbreviations CC Central Canal - FTL Lateral Tegmental Field - Gc Gigantocellular Nucleus - IO Inferior Olive - LR Lateral Reticular Nucleus - Mc Magnocellular Reticular Nucleus - Nc Cunetae Nucleus - Ng Gracile Nucleus - P Pyramidal Tract - PGL Lateral Paragigantocellular Nucleus - PRd Paramedian Reticular Nucleus,dorsal portion - PRv Paramedian Reticular Nucleus, ventral portion - RB Restiform Body - Ro Nucleus Raphe Obscurus - Rm Nucleus Raphe Magnus - Rpa Nucleus Raphe Pallidus - RVR Rostral Ventrolateral Medullary Reticular Formation - TSp5 Tractus Spinalis Nervi Trigemini - V4 Fourth Ventricle - 12N Hypoglossal Nerve - A B C D E and F correspond to levels Fr 16.0 Fr 14.7 Fr 12.7 Fr 11.6 Fr 10.0 and Fr 9.2 posterior to the frontal zero     

Ascending tract neurones processing information from group II muscle afferents in sacral segments of the feline spinal cord.

1. Ascending tract neurones located in the dorsal horn of sacral segments of the spinal cord have been investigated by extracellular and intracellular recording in the anaesthetized cat. The aim was to determine whether information from group II afferents that terminate within the sacral segments is conveyed to supraspinal structures and which types of neurones are involved. 2. A considerable proportion of ascending tract neurones found in the dorsal horn in the same segments as the pudendal (Onuf's) motor nucleus were excited by group II muscle afferents. The great majority (93%) of these neurones had axons ascending in ipsilateral funiculi. Spinocervical tract neurones constituted the largest proportion (82%) of such neurones, while very few spinocerebellar tract and propriospinal neurones and no postsynaptic dorsal column neurones were found among them. 3. In addition to activation by group II muscle afferents all of the neurones were strongly excited by cutaneous afferents. The most potent excitation was evoked by afferents of the posterior biceps-semitendinosus and gastrocnemius muscle nerves and by afferents of the cutaneous femoris, sural and pudendal nerves. The latencies of intracellularly recorded excitatory potentials were indicative of a high incidence of monosynaptic coupling between the afferents and ascending tract neurones. 4. The highly effective monosynaptic excitation of spinocervical tract neurones in the sacral segments by group II afferents is in contrast to the weak disynaptically mediated actions of group II afferents on such neurones in the L6-L7 segments but comparable to the actions of group II afferents on ascending tract neurones in the midlumbar segments. 5. Both the patterns of peripheral input and the latencies of synaptic actions in ascending tract neurones were similar to those in interneurones at the same locations (accompanying report). Similar information is therefore likely to be processed by both categories of neurones. 6. The role of sacral spinocervical tract neurones as a system for transmitting information from group II muscle afferents to supraspinal centres and the potential contribution of this system to the perception of limb position are discussed.     

Visceral nociceptive input to the area of the medullary lateral reticular nucleus ascends in the lateral spinal cord

In halothane-anesthetized rats, neurons stereotaxically located in the region of the medullary lateral reticular nucleus (LRN) and responsive to urinary bladder distension (UBD) were characterized using extracellular electrodes. Most neurons excited by UBD were also excited by noxious stimuli applied to bilateral receptive fields comprising at least half of the body surface. These bilateral nociceptive specific (bNS) neurons exhibited graded responses to graded intensities of UBD. Neuronal responses to noxious UBD were highly positively correlated with responses to noxious colorectal distension, suggesting a convergence of visceral sensory information in the area of LRN. Bilateral lateral mid-cervical spinal cord lesions virtually abolished activity of bNS neurons evoked by noxious UBD, while dorsal midline lesions had no significant effect. These data support a role for neurons in the region of the LRN in visceral nociception and implicate traditional lateral spinal cord pain pathways in the transmission of visceral information to caudal ventrolateral medullary structures.     

Corticofugal axons in the lateral geniculate nucleus of the cat

Summary Injections of horseradish peroxidase (HRP) were made into the optic radiations just above the lateral geniculate nucleus of four cats to trace the anterograde filling of corticofugal axons terminating in the perigeniculate and lateral geniculate nuclei. The different types of axons were classified according to axonal diameter and the morphology of the terminal appendages. Judging from their morphological organization we suggest that the corticofugal axons are, in the main, slowly conducting and that they have a restricted terminal distribution which extends, however, to a multiplicity of levels in both perigeniculate and lateral geniculate nuclei. These morphological characteristics may have physiological implications in determining the role of the corticofugal pathway.     

An ultrastructural study of the lateral reticular nucleus in the rat

Summary A systematic study of the normal synaptic patterns within the lateral reticular nucleus (LRN) of the rat revealed various synaptic relationships. Two types of axon terminals were identified according to the morphology of the synaptic vesicles contained within them. Axon terminals with round vesicles established asymmetrical synaptic contacts with the somata and all areas of the dendritic trees including somatic and dendritic appendages. Pleomorphic-vesicle terminals established symmetrical synaptic contacts on somata and their appendages and on all sizes of dendrites and their appendages. Both round and pleomorphicvesicle terminals were infrequently seen to synapse upon the somata and proximal dendrites. The round-vesicle terminals outnumbered the pleomorphic-vesicle terminals on the dendritic trees. Terminals of the en passant type were also common throughout the LRN. Both round and pleomorphic-vesicle terminals were observed simultaneously contacting the soma and one or more dendritic profiles, or two different dendritic profiles. Synaptic configurations (glomeruli) were also observed in all three divisions of the nucleus. They consisted of a large, central, round-vesicle terminal contacting a number of small-calibre dendritic processes. This arrangement was surrounded by one or more sheets of glial lamellae. Puncta adherentia were observed on the apposed membranes of adjacent cells, adjacent dendrites and adjacent axon terminals.