Thick Axons Research Articles

The neuronal organization of a unique cerebellar specialization: The valvula cerebelli of a mormyrid fish

AbstractPhotomicrograph of a Golgi cell from the valvular ridge of the cerebellum of a mormyrid fish (Gnathonemus petersii). The cell was filled with neurobiotin in a transverse slice of the valvular ridges and the neurobiotin was revealed with red fluorescent dye. Nissl counterstaining is shown in green. The cell was photographed in a slice of 200 μm using a confocal laser scanning microscope. The cell body is located in the basal region of the ganglionic layer. Several smooth primary dendrites arise from the molecular side of the soma and give off secondary and tertiary branches irregularly. The dendrites become thinner and smoother as they extend further into the molecular layer. The thick axon arises from the granular layer side of the soma and projects into the granular layer where it branches extensively. En passant synaptic boutons are visible on the axonal branches. The large ball‐like ending of the main axon is probably a cut branch of the axon at the edge of the slice. J. Comp. Neurol. 509: 449‐473, 2008. © 2008 Wiley‐Liss, Inc.

The neuronal organization of a unique cerebellar specialization: The valvula cerebelli of a mormyrid fish

The distal valvula cerebelli is the most prominent part of the mormyrid cerebellum. It is organized in ridges of ganglionic and molecular layers, oriented perpendicular to the granular layer. We have combined intracellular recording and labeling techniques to reveal the cellular morphology of the valvula ridges in slice preparations. We have also locally ejected tracer in slices and in intact animals to examine its input fibers. The palisade dendrites and fine axon arbors of Purkinje cells are oriented in the horizontal plane of the ridge. The dendrites of basal efferent cells and large central cells are confined to the molecular layer but are not planar. Basal efferent cell axons are thick and join the basal bundle leaving the cerebellum. Large central cell axons are also thick, and they traverse long distances in the transverse plane, with local collaterals in the ganglionic layer. Vertical cells and small central cells also have thick axons with local collaterals. The dendrites of Golgi cells are confined to the molecular layer, but their axon arbors are either confined to the granular layer or proliferate in both the granular and ganglionic layers. Dendrites of deep stellate cells are distributed in the molecular layer, with fine axon arbors in the ganglionic layer. Granule cell axons enter the molecular layer as parallel fibers without bifurcating. Climbing fibers run in the horizontal plane and terminate exclusively in the ganglionic layer. Our results confirm and extend previous studies and suggest a new concept of the circuitry of the mormyrid valvula cerebelli.

Neurobiology of the basal platyhelminth Macrostomum lignano: map and digital 3D model of the juvenile brain neuropile

We have analyzed brain structure in Macrostomum lignano, a representative of the basal platyhelminth taxon Macrostomida. Using confocal microscopy and digital 3D modeling software on specimens labeled with general markers for neurons (tyrTub), muscles (phalloidin), and nuclei (Sytox), an atlas and digital model of the juvenile Macrostomum brain was generated. The brain forms a ganglion with a central neuropile surrounded by a cortex of neuronal cell bodies. The neuropile contains a stereotypical array of compact axon bundles, as well as branched terminal axons and dendrites. Muscle fibers penetrate the flatworm brain horizontally and vertically at invariant positions. Beside the invariant pattern of neurite bundles, these "cerebral muscles" represent a convenient system of landmarks that help define discrete compartments in the juvenile brain. Commissural axon bundles define a dorsal and ventro-medial neuropile compartment, respectively. Longitudinal axons that enter the neuropile through an invariant set of anterior and posterior nerve roots define a ventro-basal and a central medial compartment in the neuropile. Flanking these "fibrous" compartments are neuropile domains that lack thick axon bundles and are composed of short collaterals and terminal arborizations of neurites. Two populations of neurons, visualized by antibodies against FMRFamide and serotonin, respectively, were mapped relative to compartment boundaries. This study will aid in the documentation and interpretation of patterns of gene expression, as well as functional studies, in the developing Macrostomum brain.

Stochastic Simulations on the Reliability of Action Potential Propagation in Thin Axons

It is generally assumed that axons use action potentials (APs) to transmit information fast and reliably to synapses. Yet, the reliability of transmission along fibers below 0.5 μm diameter, such as cortical and cerebellar axons, is unknown. Using detailed models of rodent cortical and squid axons and stochastic simulations, we show how conduction along such thin axons is affected by the probabilistic nature of voltage-gated ion channels (channel noise). We identify four distinct effects that corrupt propagating spike trains in thin axons: spikes were added, deleted, jittered, or split into groups depending upon the temporal pattern of spikes. Additional APs may appear spontaneously; however, APs in general seldom fail (<1%). Spike timing is jittered on the order of milliseconds over distances of millimeters, as conduction velocity fluctuates in two ways. First, variability in the number of Na channels opening in the early rising phase of the AP cause propagation speed to fluctuate gradually. Second, a novel mode of AP propagation (stochastic microsaltatory conduction), where the AP leaps ahead toward spontaneously formed clusters of open Na channels, produces random discrete jumps in spike time reliability. The combined effect of these two mechanisms depends on the pattern of spikes. Our results show that axonal variability is a general problem and should be taken into account when considering both neural coding and the reliability of synaptic transmission in densely connected cortical networks, where small synapses are typically innervated by thin axons. In contrast we find that thicker axons above 0.5 μm diameter are reliable.

Projection patterns of lamina VIII commissural neurons in the lumbar spinal cord of the adult cat: An anterograde neural tracing study

This study was designed to characterize the morphology of commissural axons, with the goal of revealing some of the organizing principles of their projections in the lumbosacral cord. Axons were labeled anterogradely with biotinylated-dextran amine which was injected in the left lamina VIII and the adjoining parts of lamina VII in the lumbar segments L5–L6 in seven cats. After 3–4 weeks, commissural axons were well labeled throughout lumbosacral segments L1–S2. After crossing the midline at the injection level, labeled axons traveled rostrally and/or caudally in the contralateral ventral and lateral funiculi giving off multiple axon collaterals. The trajectories of 34 single axons were traced in their entirety from their points of origin to their distal ends. Most of these axons were thin (proximal diameter <3.5 μm) and short (<30 mm), and gave off 6 to 32 axon collaterals at short intercollateral distances (mean <2 mm) in the lumbosacral enlargement. Some thicker axons (diameter >3.5 μm) ascended as far as the thoracic level; these supplied only four to six collaterals at long intercollateral intervals (∼6.5 mm). All of the axons except one projected unilaterally. The axons as a whole terminated throughout the contralateral ventral horn. However, axons that traveled in different parts of the white matter had different characteristic terminal arborizations. The collaterals of axons that traveled in the ventral funiculus terminated preferentially in laminae VII–VIII, while those in the lateral funiculus terminated in lamina IX. Although the collateral branching patterns differed from one axon to another, collaterals arising from a particular axon usually exhibited similar patterns at different rostrocaudal levels. These uniform collateral termination patterns indicate that the morphology of each neuron might be specifically related to its function. This may allow future studies to identify different functional types of commissural neurons on the basis of much less extensive reconstructions.

Ommatidial type‐specific interphotoreceptor connections in the lamina of the swallowtail butterfly, Papilio xuthus

The eye of the butterfly Papilio xuthus contains a random array of three types of ommatidia (types I-III), each bearing nine photoreceptors, R1-R9. Of the six spectral classes of photoreceptors identified, types I, II, and III ommatidia contain four, three, and two classes, respectively: the ommatidia are thus spectrally heterogeneous. The photoreceptors send their axons to the lamina where, together with some large monopolar cells (LMCs), the nine from a single ommatidium contribute to a module called a lamina cartridge. We recently reported that among different photoreceptor axon terminals visualized by confocal microscopy, the number and length of axon collaterals differ for different spectral receptors, suggesting that lamina circuits are specific for each ommatidial type. Here we studied the distribution of synapse-like structures in the cartridges, first characterizing a photoreceptor by measuring its spectral sensitivity and then injecting Lucifer yellow (LY). We subsequently histologically identified the type of ommatidium to which the injected photoreceptor belonged, cut serial ultrathin sections of the entire lamina, labeled these with anti-LY immunocytochemistry, and then localized synapse-like structures. We found numerous interphotoreceptor contacts both within and between cartridges, the combination of which was again specific for the ommatidial type. R3 and R4, which are green-sensitive photoreceptors in all ommatidia, have thick axons lacking collaterals. We found that these cells exclusively make contacts with LMCs and not with photoreceptors. We therefore presume that R3 and R4 construct a system for motion vision, whereas other randomly distributed spectral types provide inputs for color vision.

Morphological identification of ganglion cells expressing the α subunit of type II calmodulin‐dependent protein kinase in the macaque retina

Expression of the alpha subunit of type II calmodulin-dependent protein kinase (alphaCamKII) distinguishes the koniocellular neurons of the primate lateral geniculate nucleus (LGN) from the primary parvo- and magnocellular neurons, but whether the same neurochemical signature distinguishes the retinal ganglion cells providing them input is not known. We find that, in the retina, alphaCamKII expression also differentiates two primary groups of ganglion cell, both characterized by broad, sparsely branching dendritic trees and cell bodies intermediate in size between the parvo- and magnocellular-projecting ganglion cells. Cells in the first group have three or four primary dendrites, a thick axon, and a rounded cell body and likely are made up of multiple types. In contrast, ganglion cells in the second group demonstrate a highly regular morphology, with strictly two primary dendrites and a thinner axon emanating from a smaller, elliptical cell body. This cell resembles the "large sparse" ganglion cell identified by others in retrograde labeling from the LGN and represents about 2% of all ganglion cells. In the optic nerve, alphaCamKII+ axons are also intermediate in size and form a bimodal distribution, correlating with the axonal sizes of the two groups of ganglion cell. For the LGN, we describe a group of alphaCamKII+ axon terminals with morphology consistent with terminals from retinal ganglion cells. These terminals form long, filamentous contacts with alphaCamKII+ relay cells and increase in frequency from the dorsal to the ventral koniocellular regions. Our results indicate that ganglion cells expressing alphaCamKII represent multiple projections to the brain, at least one of which provides input to one or more koniocellular regions of the LGN.

Morphology of early developing oligodendrocytes in the ventrolateral spinal cord of the chicken.

The oligodendroglial population includes Type I and II cells related to several thin axons, Type III cells with a few processes in relation to relatively thick axons and Type IV cells related to a single thick axon. This structural diversity of oligodendrocytes is accompanied by a molecular heterogeneity. In the chicken spinal cord, oligodendrocytes have begun to contact axons at embryonic day (E)10 and compact sheaths have appeared by E12. At the latter stage, most sheath-forming oligodendrocytes contact more than one axon. At E15, however, each sheath-forming cell seems to have developed a Schwann cell-like anatomy, being related to a single axon. Based on these findings, the present study examines more thoroughly the anatomy of early developing oligodendrocytes in the chicken spinal cord. Examination of slices immunostained with antibodies against the oligodendroglial marker O4 showed that a few positive cells are present at E6, after which the occurrence increases with age. At E12 most immunostained cells have two or more processes. At E15 however, dye-injected oligodendrocytes have developed a Type IV structure. Between E12 and E15, mean sheath length increases about 4x, from 50 microm to over 200 microm, while the length of the spinal cord increases 36% only. This indicates that early oligodendrocytes in chicken white matter develop a Type IV anatomy between E12 and E15 through an elimination of sheaths.

Role of Cajal-Retzius and subplate neurons in cerebral cortical development

A synaptic network is already formed in the marginal zone of the early telencephalon before the arrival of the first wave of radial migration of neuroblasts from the subventricular zone to form the cortical plate. Cells and fibers forming the marginal zone are mainly the Cajal-Retzius (C-R) neurons and their processes. The origin of these cells is not yet proved but is likely either the median ganglionic eminence or the mesencephalic neuromere. The bipolar or multipolar C-R neurons populate the molecular layer of the fetal cortical plate and are sparse in the adult. Their thick axon emits collaterals for synaptic contact with pyramidal neurons initially in layer 6 and later with in all layers. C-R neurons produce GABA, possibly ACh, several calcium-binding proteins (eg, calmodulin, parvalbumin, calretinin) and several neuropeptides; they are rich in ribosomes. Subplate neurons, beneath the cortical plate, emit pioneer axons in the incipient formation of the internal capsule and also commissural fibers of the early hippocampus. C-R cells express products of the genes RELN, LIS1, and DS-CAM, which mediate radial neuroblast migration and lamination of the cortical plate and important in the pathogenesis of lissencephaly. A subpopulation of C-R neurons also expresses a p53 product implicated in cell survival and apoptosis. In addition to forming the first intrinsic synaptic circuits of the cortical plate and its first afferent and efferent connections with subcortical structures, they may play additional roles in the formation of ocular dominance columns, in regulating neuronogenesis, and in cortical repair. They do not disappear by apoptosis at the completion of cell migration, as was previously thought, but their functional role in the mature brain remains unknown.

Axon collaterals of Meynert cells diverge over large portions of area V1 in the macaque monkey.

Patchy intrinsic connections, originating mainly from horizontal collaterals of pyramidal neurons, have been demonstrated in area V1 and many other cortical areas. In this article, we identify a network of intrinsic connections concentrated in layer 6 of area V1. These are visualized by extracellular injections of anterograde tracers in V1, which label small clusters of large terminal boutons in layer 6, in conjunction with thick axon segments. These segments can be traced back to infragranular Meynert cells (n = 10), which are retrogradely labeled from the injections. By using serial section analysis, we identified the following features of this distinctive system of Meynert cell collaterals: (1) terminal clusters are relatively small (<100 microm); (2) each cluster has a small number of rather large boutons (up to 3.0 microm); (3) there is typically a termination-free zone in the immediate vicinity (0.5-2.0 mm) of the cell body; (4) a single neuron has multiple branches that can extend up to 8.0 mm from the soma; and (5) the collaterals are concentrated in layer 6. These features are different from those of horizontal intrinsic connections in the supragranular layers of area V1. They are consistent with fast dynamics and a possible role in wide-field motion processing, such as has been associated with Meynert cells from other studies.

Distribution of the calcium-binding proteins parvalbumin, calbindin D-28k and calretinin in the retina of two teleosts

Using monoclonal antibodies against parvalbumin (PV) and calbindin (CB), and a polyclonal antiserum against calretinin (CR), the expression patterns of these proteins in the retina of the tench and rainbow trout were studied at light microscopic level in in toto preparations and radial sections. Parvalbumin was present in subpopulations of small amacrine cells in both species, but these cells were more abundant and had a clear centre-periphery gradient distribution in the tench. Using the McAB 300 monoclonal antibody against CB, glial cells such as Müller cells, astrocytes in the nerve fibre layer, and sparse large cells close to the entrance of the optic nerve were observed in both species. Moreover, this antibody strongly labelled H1 horizontal cells and their thick axon terminals in the tench retina, whereas only a small population of amacrine cells was stained in the trout. Calretinin was expressed in different types of ganglion cells and numerous neurones located in the inner plexiform layer in both species, but was more abundant and more strongly stained in the trout retina, where some bipolar cells were easily distinguishable. A comparison to current results in other vertebrate species is offered.

Local current spread in electrically compact neurons of the fly

Analyses of active and passive membrane properties predict an asymmetry in the spread of electrical current through a neuron. Simulated current injection into a large-diameter compartment of a biophysically realistic model neuron causes a local potential shift that can spread throughout the cell. In contrast, causing the same local potential shift in the dendritic tip of the same neuron results in only minimal changes in electrical potential in the rest of the cell. Using calcium as a reporter of electrical activity in neurons in the fly's lobula plate we find that current injected into the thick axon caused depolarization throughout the cell, whereas activation of a dendritic region remained local. These results have important implications for the ability of integrating neurons to perform local computations of synaptic input without additional hardware.

Myelination of prospective large fibres in chicken ventral funiculus.

In mammals, the oligodendrocyte population includes morphological and molecular varieties. We reported previously that an antiserum against the T4-O molecule labels a subgroup of oligodendrocytes related to large myelinated axons in adult chicken white matter. We also reported that T4-O immunoreactive cells first appear in the developing ventral funiculus (VF) at embryonic day (E)15, subsequently increasing rapidly in number. Relevant fine structural data for comparison are not available in the literature. This prompted the present morphological analysis of developing and mature VF white matter in the chicken. The first axon-oligodendrocyte connections were seen at E10 and formation of compact myelin had started at E12. Between E12 and E15 the first myelinating oligodendrocytes attained a Schwann cell-like morphology. At hatching (E21) 60% of all VF axons were myelinated and in the adult this proportion had increased to 85%. The semilunar or polygonal oligodendrocytes associated with adult myelinated axons contained many organelles indicating a vivid metabolic activity. Domeshaped outbulgings with gap junction-like connections to astrocytic profiles were frequent. Oligodendrocytes surrounded by large myelinated axons and those surrounded by small myelinated axons were cytologically similar. But, thick and thin myelin sheaths had dissimilar periodicities and Marchi-positive myelinoid bodies occurred preferentially in relation to large myelinated axons. We conclude that early oligodendrocytes contact axons and form myelin well before the first expression of T4-O and that emergence of a T4-O immunoreactivity coincides in time with development of a Type IV phenotype. Our data also show that oligodendrocytes associated with thick axons are cytologically similar to cells related to thin axons. In addition, the development of chicken VF white matter was found to be similar to the development of mammalian white matter, except for the rapid time course.

Ultrastructural Characterizations of Olfactory Pathway Neurons in Layer II of the Entorhinal Cortex in Monkey

The somatic size, shape, dendritic and axonal morphology, and synaptology of representative neurons in layer II of the primate entorhinal cortex (EC) were analyzed. Layer II ?islands? contained large spinous multipolar cells with triangular somata and local circuit axons in addition to multipolar neurons with large, radially arrayed, aspinous, primary processes and thick tapering axons. Small pyramidal neurons with a single, spinous, apical primary segment that bifurcated a short distance from the somata were also found in layer II. Subsequent spinous segments of these neurons with long terminal segments exhibited a paucity of branching in addition to having thick axons tapering into subjacent layers. The importance of providing these additional axonal, dendritic, and synaptic characterizations lies in the contextual role these neurons play in the connectional patterns of the EC with regard to olfaction, olfactory memory, and pathological variations.

Neural circuitry and neutrotransmitters that mediate the acoustic startle reflex in the rat

Because the acoustic startle reflex has such a short latency (e.g., 5 ms measured electromyographically in the neck, 8 ms in the hindleg), it must be mediated by a relatively simple neural pathway. We believe that the acoustic startle pathway is mediated by three synapses. These include auditory nerve inputs onto cochlear root neurons, which are cells embedded in the auditory nerve that receive the first afferent information in the auditory system. They send exceedingly thick axons (e.g., 7 μm in diameter) to cells in the nucleus reticularis pontis caudalis which in turn project directly to motoneurons in the facial motor nucleus (pinna reflex) or spinal cord (whole body startle). Chemical lesions of cochlear root neurons or of cells in the ventral lateral nucleus reticularis pontis caudalis eliminate startle. Local infusion of either NMDA or non-NMDA antagonists into either the nucleus reticularis pontis caudalis or the spinal cord, via intrathecal infusion, markedly depress acoustic startle, suggesting that both of these glutamate receptor subtypes may mediate the acoustic startle reflex.

3rd workshop of the European CMT consortium: 54th ENMC International Workshop on genotype/phenotype correlations in Charcot-Marie-Tooth type 1 and hereditary neuropathy with liability to pressure palsies 28-30 November 1997, Naarden, The Netherlands.

3rd workshop of the European CMT consortium: 54th ENMC International Workshop on genotype/phenotype correlations in Charcot-Marie-Tooth type 1 and hereditary neuropathy with liability to pressure palsies 28-30 November 1997, Naarden, The Netherlands.

Functional architecture of primate cone and rod axons.

The cone axon is nearly four times thicker than the rod axon (1.6 vs 0.45 microns diameter). To assess how signal transfer and integration at the terminal depend on cable dimensions, a transducer (cone = ohmic conductance, rod = current source) coupled via passive cable to a sphere with a chloride conductance (representing GABAA receptor) was modelled. For a small signal in peripheral cone with a short axon, steady photosignal transfers independently of axon diameter despite a significant chloride conductance at the cone terminal. Temporally varying photosignal also transfers independently of axon diameter up to 20 Hz and is attenuated only 20% at 50 Hz. Thus, to accomplish the basic electrical functions of a peripheral cone, a thin axon would suffice. For a foveal cone with a long axon steady photosignal transfers independently of axon diameter, but temporally varying photosignal is attenuated 5-fold at 50 Hz for a thick axon and 10-fold for a thin axon. This might contribute to the lower sensitivity of central retina to high temporal frequencies. The cone axon contains 14-fold more microtubules than the rod axon, and its terminal contains at least 20-fold more ribbon synapses than the rod's. Since ribbon synapses sustain high rates of exocytosis, the additional microtubules (which require a thicker axon) may be needed to support a greater flux of synaptic vesicle components.

Morphology of basal optic tract terminals in the turtle, Pseudemys scripta elegans.

The morphologies of axon terminals of retinal ganglion cells projecting to the basal optic nucleus (BON) via the basal optic tract (BOT) were studied in the red-eared turtle. The BOT was visualized on the ventral surface of the brainstem in vitro, and either biotinylated dextran amine was injected extracellularly or neurobiotin was injected into physiologically identified axons during intracellular recordings. Up to 16 hours after tracer injection, the brains were fixed, sectioned parasagittally, and stained for biotin and Nissl substance. The diameters and depths of extracellularly filled axons were measured at three BON sites. Fourteen axons were reconstructed from serial sections with the aid of appropriate computer software. Analysis of extracellularly filled retinal axons revealed that about three times more axons were present just inside the rostral border of the BON compared with its caudal border. Thick (2-4 microm) axons were located within 100 microm from the ventral border, whereas thin (<2 microm) axons were found throughout the nucleus. Only the thinnest axons (<1 microm) extended caudally from the nucleus, indicating that some extracellularly labelled fibers passed through the BON. The intracellularly filled axons were more similar to the thick axons filled extracellularly and arborized entirely within the BON. All of the intracellularly filled axons had thick ventral trunks from which many thin branches extended dorsally and obliquely within the BON. The thin branches bifurcated repeatedly to form bead-like varicosities or boutons that often formed clusters within regions of 150 microm3 or less. These clusters may reflect areas of focused synaptic contact on BON cells with specific direction preferences.

Classes of axons and their distribution in the optic nerve of the tree shrew (Tupaia belangeri).

It has been suggested that retinal ganglion cells (RGCs) of Tupaia can be subdivided into three classes that correspond to the X, Y, and W classes in the cat. Estimates of these classes as determined by electrophysiological experiments and by histological studies of the retina are at variance. Because the RGC classes differ in axon diameter, this parameter could serve as a reliable criterion for the evaluation of RGC classes and their proportions. An electronmicroscopic analysis of four optic nerves was carried out. The density of axons and their diameters was recorded from cross sections in the middle of the nerves. Based on a theoretical model, axons were classified according to the three known RGC classes. In addition, we investigated how axons of different size are distributed within the nerve. On average, the total number of axons is 570,000. More than 99% of axons are myelinated. Axon diameters can be 0.2-3.6 microns, and fiber diameters can be 0.3-4.6 microns. The frequency distributions of axon and fiber diameter are positively skewed and multimodal. Our analysis revealed three distinct axon diameter groups with the following mean axon diameters and proportions: the small or S group: 0.55 micron, 70%; the medium-sized or M group: 0.88 micron, 20%; the large of L group: 1.39 microns, 10%. Density and axon diameter plots produced consistent patterns, according to which axons of different sizes were distributed in the optic nerve. Thick axons are located dorsotemporally and centrally in the nerve. The average diameter value decreases toward the periphery. The smallest values are found at the ventronasal border of the nerve. The observed axon diameter groups probably arise from functionally distinct RGC types. There is evidence that these groups correspond to functional W, X, and Y RGC classes. Our study also provides the first evidence of the existence of a topographic order of fibers within the nerve of the tree shrew.

Ventral nucleus of the lateral lemniscus in guinea pigs: cytoarchitecture and inputs from the cochlear nucleus.

Cytoarchitectonic criteria were used to distinguish three subdivisions of the ventral nucleus of the lateral lemniscus in guinea pigs. Axonal tracing techniques were used to examine the projections from the cochlear nucleus to each subdivision. Based on the cell types they contain and their patterns of input, we distinguished ventral, dorsal, and anterior subdivisions of the ventral nucleus of the lateral lemniscus. All three subdivisions receive bilateral inputs from the cochlear nucleus, with contralateral inputs greatly outnumbering ipsilateral inputs. However, the relative density of the inputs varies: the ventral subdivision receives the densest projection, whereas the anterior subdivision receives the sparsest projection. Further differences are apparent in the morphology of the afferent axons. Following an injection of Phaseolus vulgaris-leucoagglutinin into the ventral cochlear nucleus, most of the axons on the contralateral side and all of the axons on the ipsilateral side are thin. Thick axons are present only in the ventral subdivision contralateral to the injection site. The evidence from both anterograde and retrograde tracing studies suggests that the thick axons originate from octopus cells, whereas the thin axons arise from multipolar cells and spherical bushy cells. The differences in constituent cell types and in patterns of inputs suggest that each of the three subdivisions of the ventral nucleus of the lateral lemniscus makes a distinct contribution to the analysis of acoustic signals.