Role of GABAergic inhibition in the coding of interaural time differences of low-frequency sounds in the inferior colliculus.

A major cue for the localization of sound in space is the interaural time difference (ITD). We examined the role of inhibition in the shaping of ITD responses in the inferior colliculus (IC) by iontophoretically ejecting gamma-aminobutyric acid (GABA) antagonists and GABA itself using a multibarrel pipette. The GABA antagonists block inhibition, whereas the applied GABA provides a constant level of inhibition. The effects on ITD responses were evaluated before, during and after the application of the drugs. If GABA-mediated inhibition is involved in shaping ITD tuning in IC neurons, then applying additional amounts of this inhibitory transmitter should alter ITD tuning. Indeed, for almost all neurons tested, applying GABA reduced the firing rate and consequently sharpened ITD tuning. Conversely, blocking GABA-mediated inhibition increased the activity of IC neurons, often reduced the signal-to-noise ratio and often broadened ITD tuning. Blocking GABA could also alter the shape of the ITD function and shift its peak suggesting that the role of inhibition is multifaceted. These effects indicate that GABAergic inhibition at the level of the IC is important for ITD coding.

Effects of amplitude modulation on the coding of interaural time differences of low-frequency sounds in the inferior colliculus. I. Response properties.

Most sounds in the natural environment are amplitude-modulated (AM). To determine if AM alters the neuronal sensitivity to interaural time differences (ITDs) in low-frequency sounds, we tested neuronal responses to a binaural beat stimulus with and without modulation. We recorded from single units in the inferior colliculus of the unanesthetized rabbit. We primarily used low frequency ( approximately 25 Hz) modulation that was identical at both ears. We found that modulation could enhance, suppress, or not affect the discharge rate. In extreme cases, a neuron that showed no response to the unmodulated binaural beat did so when modulation was added to both ears. At the other extreme, a neuron that showed sensitivity to the unmodulated binaural beat ceased firing with modulation. Modulation could also affect the frequency range of ITD sensitivity, best ITD, and ITD tuning width. Despite these changes in individual neurons, averaging across all neurons, the peak and width of the population ITD function remained unchanged. Because ITD-sensitive neurons also time-locked to the modulation frequency, the location and sound attributes are processed simultaneously by these neurons.

Effects of amplitude modulation on the coding of interaural time differences of low-frequency sounds in the inferior colliculus. II. Neural mechanisms.

In our companion paper, we reported on interaural time difference (ITD)-sensitive neurons that enhanced, suppressed, or did not change their response when identical AM was added to both ears. Here, we first examined physical factors such as the difference in the interaural correlation, spectrum, or energy between the modulated and unmodulated signals. These were insufficient to explain the observed enhancement and suppression. We then examined neural mechanisms by selectively modulating the signal to each ear, varying modulation depth, and adding background noise to the unmodulated signal. These experiments implicated excitatory and inhibitory monaural inputs to the inferior colliculus (IC). These monaural inputs are postulated to adapt to an unmodulated signal and adapt less to a modulated signal. Thus enhancement or suppression is created by the convergence of these excitatory or inhibitory inputs with the inputs from the binaural comparators. Under modulation, the role of the monaural input is to shift the threshold of the IC neuron. Consistent with this role, background noise mimicked the effect of modulation. Functionally, enhancement and suppression may serve in detecting the degree of modulation in a sound source while preserving ITD information.

Direct innervation of identified tectothalamic neurons in the inferior colliculus by axons from the cochlear nucleus.

The present study sought to identify tectothalamic neurons in the rat inferior colliculus that receive their innervation directly from the cochlear nuclei and to identify the axons that provide the innervation. A direct projection would bypass the binaural interactions of the superior olivary complex and provide the quickest route to the neocortex. Axons, primarily from the dorsal cochlear nucleus, were labeled with anterograde transport of dextran and terminated in the central nucleus of the inferior colliculus in a laminar pattern. Most labeled axons were thin and simply branched. Other axons were thicker, gnarly, less frequently observed and probably originated from the ventral cochlear nucleus. None had concentrated endbulbs or a nest of endings. Both types of axons terminated primarily in the central nucleus and layer 3 of the external cortex. This pattern suggests that the combination of these subdivisions in the rat are equivalent to the central nucleus as defined in other species. Tectothalamic neurons in the inferior colliculus in the same animals were identified by retrograde transport from the medial geniculate body and intracellular injection of Lucifer Yellow. A number of different cell types act as tectothalamic neurons and receive contacts from cochlear nucleus axons. These include flat cells (disc-shaped), less-flat cells and stellate cells. Two innervation patterns were seen: a combination of axosomatic and axodendritic contacts, and predominantly axodendritic contacts. Both patterns were seen in the central nucleus, but axosomatic contacts were seen less often in the other subdivisions. This is the first study to show direct connections between cochlear nuclear axons and identified tectothalamic neurons. The layers of axons from cochlear nuclei may provide convergent inputs to neurons in the inferior colliculus rather than the heavy inputs from single axons typical of lower auditory nuclei. Excitatory synapses made by axons from the cochlear nuclei on tectothalamic neurons may provide a substrate for rapid transmission of monaural information to the medial geniculate body.

Spatial organization of the reciprocal connections between the cat dorsal and anteroventral cochlear nuclei.

We are studying the interconnections between the anteroventral cochlear nucleus (AVCN) and the dorsal cochlear nucleus (DCN). Biotinylated dextran was injected into the DCN, where the best frequency of responses was also recorded. Ventrotubercular neurons in AVCN were labeled, along with cochlear nerve fibers and the axons of cells in DCN. In AVCN, a central band of labeled cochlear nerve axons and large endbulbs was labeled. Bordering this band was a 'fringe' of smaller tuberculoventral axonal endings forming pericellular nests. Most AVCN neurons projecting to DCN were stellate, elongate, or giant cells, located in the posterior division of AVCN, regardless of the DCN injection site. About 75% of the labeled AVCN cells lay within the bands of labeled cochlear nerve fibers. Another 15% were in the outer fringes on either side of these bands, while 10% were outside the bands and the fringes. These findings suggest that most AVCN neurons projecting to the DCN conform to the tonotopic map. A significant portion of the ventrotubercular neurons occupy side-bands in AVCN. Reciprocally, the tuberculoventral tract forms a robust fringe of axonal endings flanking the central bands. The neuronal and axonal bands and side-bands may underlie excitatory and inhibitory signal transformations.

GABA- and glycine-immunoreactive projections from the superior olivary complex to the cochlear nucleus in guinea pig.

Retrograde transport of horseradish peroxidase was combined with immunocytochemistry to identify the origins of potential gamma-aminobutyric acid (GABA) -ergic and glycinergic inputs to different subdivisions of the cochlear nucleus. Projection neurons in the inferior colliculus, superior olivary complex, and contralateral cochlear nucleus were examined, but only those from the superior olivary complex contained significant numbers of GABA- or glycine-immunoreactive neurons. The majority of these were in periolivary nuclei ipsilaterally, with a sizeable contribution from the contralateral ventral nucleus of the trapezoid body. Overall, 80% of olivary neurons projecting to the cochlear nucleus were immunoreactive for GABA, glycine, or both. Most glycine-immunoreactive projection neurons were located ipsilaterally, in the lateral and ventral nuclei of the trapezoid body and the dorsal periolivary nucleus. This suggests that glycine is the predominant neurotransmitter used by ipsilateral olivary projections. Most GABA-immunoreactive cells were located bilaterally in the ventral nuclei of the trapezoid body. The contralateral olivary projection was primarily GABA-immunoreactive and provided almost half the GABA-immunoreactive projections to the cochlear nucleus. This suggests that GABA is the predominant neurotransmitter used by contralateral olivary projections. The present results suggest that the superior olivary complex is the most important extrinsic source of inhibitory inputs to the cochlear nucleus. Individual periolivary nuclei differ in the strength and the transmitter content of their projections to the cochlear nucleus and may perform different roles in acoustic processing in the cochlear nucleus.

A physiological and structural study of neuron types in the cochlear nucleus. I. Intracellular responses to acoustic stimulation and current injection.

Neurons in the cochlear nucleus differ in their discharge patterns when stimulated by tones. They also differ in their responses to depolarizing current injection in vitro. We made intracellular recordings from neurons in the cochlear nucleus of gerbils and chinchillas. The responses to tones and to depolarizing current were compared for the same neurons. Three categories of response patterns to tones were observed: chopper, primary-like, and onset. Chopper neurons responded with regularly spaced action potentials to stimulation with tones and to injections of depolarizing current. Their response rate rose with increasing levels of current to a maximum, which was comparable to that evoked by suprathreshold tones. These observations suggest that the regularity and maximal firing rate of these neurons are determined by voltage-dependent membrane properties. Primary-like neurons responded with irregularly spaced action potentials to tones. Injection of depolarizing current into these neurons produced a single action potential at current onset, which could be followed by a few irregularly spaced action potentials. The response rate showed little relation to current level. These data suggest that the membrane characteristics of primary-like neurons are different from those of chopper neurons. Onset neurons produced action potentials only at the beginning of the stimulus for both tones and depolarizing current, even though there was a sustained depolarization throughout the duration of the tone. The findings suggest that cochlear nucleus neurons have different membrane properties and that these properties may play a critical role in a neuron's temporal response pattern to acoustic stimulation.

A physiological and structural study of neuron types in the cochlear nucleus. II. Neuron types and their structural correlation with response properties.

The present study examined the morphological cell types of neurons labeled with intracellular horseradish peroxidase injections, many of them following electrophysiological recordings in the cochlear nucleus of gerbils and chinchillas. Most of the subdivisions and neuronal types previously described in the cat were identified in the present material, including spherical and globular bushy cells, stellate, bushy multipolar, elongate, octopus, and giant cells in the ventral cochlear nucleus, and a cartwheel cell in the dorsal cochlear nucleus. In many cases these structurally distinct neurons were correlated with their characteristic responses to stimulation by sound or intracellular injection of depolarizing current. The dendritic terminals of the elongate, antenniform, and clavate cells of the posteroventral cochlear nucleus link each of these cell types with neighboring structures in distinct patterns, which may provide a basis for differences in synaptic organization. These cell types differ from each other and from the stellate cells of the anteroventral cochlear nucleus. Despite their heterogeneous morphology, most of these neurons had a regular discharge in response to stimulation (choppers). Irregularly firing neurons (primary-like) had very different structures, e.g., the spherical and globular bushy cells and the bushy multipolar neuron. They, too, represent a heterogeneous population. An onset neuron was identified as an octopus cell. This paper compares the morphological observations with the electrophysiological properties of different cell types reported in a companion paper (Feng et al. [1994] J. Comp. Neurol.). Together, these findings imply that response properties may be partially independent of neuronal structure. Morphologically distinct neurons can generate similar temporal patterns in response to simple acoustic stimuli. Nevertheless, the synaptic organization of these different neuron types, including their connections, would be expected to affect or alter the cells' responses to appropriate stimuli. The possibility is raised that membrane properties and synaptic organization complement and interact with each other.

Connectivity of neurons in identified auditory circuits studied with transport of dextran and microspheres plus intracellular injection of lucifer yellow.

The components of a neural circuit are usually distinguished in separate experiments to identify long connections, presynaptic, and postsynaptic components. We describe a procedure to visualize these components in the same experiment. Neurons in the inferior colliculus the axons of which project to the medial geniculate body were identified by retrograde transport of latex microspheres, while their innervation from the cochlear nucleus was simultaneously visualized by anterograde transport of dextrans. In aldehyde-fixed slices, the microsphere-labeled neurons near dextran-labeled axons were injected with biotinylated Lucifer Yellow. Subsequent avidin-biotin histochemistry allowed permanent visualization. The specific neurons involved in this circuit and the axonal contacts they received were easily visualized with the light microscope. This method allows the study of complex innervation patterns in the mammalian central nervous system.

Synaptic organization of globular bushy cells in the ventral cochlear nucleus of the cat: a quantitative study.

The synaptic organization of globular bushy cells of the anteroventral cochlear nucleus was quantitatively analyzed in order to understand better their functional attributes. A method was devised to estimate the concentrations and relative proportions of synapses on the entire postsynaptic surface of Golgi-impregnated neurons, by sampling with limited series of sections for electron microscopy. This provided a characteristic synaptic profile which was homogeneous for the population measured. The total concentration of synaptic endings decreases with distance from the soma. The cochlear, presumably glutamatergic and excitatory, endings with large spherical vesicles (LS) account for most of this decrease. Of the noncochlear inputs, the putative glycinergic endings with flattened vesicles (FL) decrease slightly, and the presumed GABAergic terminals with pleomorphic vesicles (PL) maintain a relatively constant concentration, while endings with small spherical vesicles (SS) increase on the distal dendrites. LS endings have the largest proportion of synapses near the soma, while FL synapses maintain a constant proportion in all cell regions, and PL and SS proportions increase on higher-order dendrites. Excitatory and inhibitory synapses have significant inputs to the axon hillock and initial segment, as well as to the distal dendrites, where dual synapses may provide a way to sample the activity of surrounding neurons. These features must be considered in explanations of physiological properties, such as the synaptic security, level of spontaneous activity, and well-timed, rapid onset responses, as well as their potential for normalizing and synchronizing an important inhibitory pathway involved in binaural signal processing. Synaptic profile analysis should be useful for experimental studies and for developing realistic computational models.

Glycine immunoreactive projections from the dorsal to the anteroventral cochlear nucleus.

The aim of the present study was to investigate whether projections from the dorsal cochlear nucleus (DCN) to the anteroventral cochlear nucleus (AVCN) use either of two inhibitory transmitters, glycine or GABA. Retrograde HRP labeling of DCN-to-AVCN projection neurons was combined with postembedding immunocytochemistry in the DCN of guinea pigs. Following injections of HRP in the anterior or posterior divisions of AVCN, large numbers of neurons were labeled in the DCN. All of these were located in the deep layer, except for a few granule cells. Nearly all (96%) of the projection neurons were immunoreactive for glycine and most had dendritic and somatic morphologies corresponding to those of elongate neurons (so-called 'corn' cells); only a few resembled small stellate neurons. Few (3%) retrogradely labeled neurons were immunoreactive for GABA. The results suggest that projections from the deep DCN to the AVCN are formed primarily by glycinergic elongate neurons. These projections could have a substantial inhibitory influence on the output of neurons in the AVCN.

Uptake and retrograde transport of [3H]GABA from the cochlear nucleus to the superior olive in the guinea pig.

The purpose of the present study is to determine which descending projections to the cochlear nucleus may use gamma-aminobutyric acid (GABA) as a neurotransmitter. [3H]GABA (120 microM) was injected into the cochlear nucleus of albino and pigmented guinea pigs. After survival times between 0.25 and 16 h, the brain stems were prepared for light microscopic autoradiography. After 2 h survival there was a pulse of label, which progressed through the fibres from the cochlear nucleus to the ipsilateral superior olive. After 5 h, retrogradely labelled neuronal cell bodies and fibres were located in the superior olivary complex bilaterally. In the trapezoid body, clusters of labelled cells were seen in the lateral nucleus, ipsilaterally, and in the ventral nucleus, bilaterally. Also there were labelled cells in the ipsilateral dorsal and anterolateral periolivary nucleus. Large and small cells of several types were labelled. Survival times of 10 h or more resulted in very light, diffuse labelling. Projections to the cochlear nucleus labelled by retrograde transport of horseradish peroxidase that did not take up [3H]GABA included the inferior colliculus, bilaterally, and the cochlear nucleus and periolivary nuclei (other than ventral trapezoid nucleus), contralaterally. The selective labelling of cell groups in the superior olive with the moderately low concentration of [3H]GABA used is consistent with the high-affinity uptake of [3H]GABA by synaptic endings in the cochlear nucleus and its retrograde by transport GABA-ergic neurons. This provides evidence for a descending projection system for inhibitory feedback from the superior olive to the cochlear nucleus.

Glycine-immunoreactive projection of the cat lateral superior olive: possible role in midbrain ear dominance.

Neurons in the lateral superior olive are optimally excited by stimulation of the ipsilateral ear, as are those in the inferior colliculus by stimulation of the contralateral ear. This reversal of ear dominance may result, in part, from distinct crossed excitatory and uncrossed inhibitory pathways ascending from the lateral superior olive. To explore this possibility, immunoreactivity for two putative inhibitory neurotransmitters, glycine and GABA, was examined in projection neurons that retrogradely transported horseradish peroxidase from the cat inferior colliculus. The results suggest that the projection from the lateral superior olive can be segregated, immunocytochemically, into three components: 1) a crossed, glycine-negative (-) projection; 2) an uncrossed, glycine-positive (+) projection; and 3) an uncrossed, glycine(-) projection. Additional evidence suggests that the terminal fields of the two uncrossed projections may distribute differently within the inferior colliculus. Glycine(+) or glycine(-) projection neurons, crossed or uncrossed, do not differ in the size, shape, or location of their somata. However, most glycine(-) neurons are heavily encrusted with glycine(+) endings; glycine(+) neurons have 40-60% fewer of these endings. Glycine(-) neurons located in the lateral limb have fewer glycine (+) perisomatic endings than those in the medial limb. Few projection neurons are GABA(+), and GABA(+) perisomatic endings are rare in the lateral superior olive. Thus, there is a heavy uncrossed projection from the cat lateral superior olive to the inferior colliculus that may be glycinergic and inhibitory. Furthermore, there is a bilateral projection that is not glycinergic or GABAergic, which may be excitatory. The potential contribution of these pathways to contralateral ear dominance in the inferior colliculus is discussed.

A degenerative disorder of the central auditory system of the gerbil.

A previously unidentified disorder which affects primarily the cochlear nucleus was observed in two species of gerbils, Meriones unguiculatus and M. libycus. Unusual lesions were observed in the cochlear nucleus bilaterally in all animals examined. In light and electron microscopic specimens these lesions were characterized by the presence of microcysts and vacuolar neuronal degeneration. The microcysts resembled large holes, containing trabeculae, organelles and cellular remnants. Also observed were light and dark degeneration of neuronal perikarya and degenerated axons, dendrites, and synapses, accompanied by phagocytosis. Astrocytosis was not conspicuous. In the one cochlea examined, no microcysts were observed. In young animals the microcysts were prevalent in the cochlear nerve root region and the posteroventral cochlear nucleus. In older animals the microcysts increased in number and area. In the oldest animals, the microcysts had spread to other central auditory structures, including the superior olivary complex, the nuclei of the lateral lemniscus, and the inferior colliculus. Other regions of the brain were largely free of microcysts. The etiology and behavioral manifestations of this disorder are unknown, although it is clearly neurodegenerative and perhaps genetically determined.

Distribution of cells projecting to thalamus vs. those projecting to cerebellum in subdivisions of the dorsal column nuclei in raccoons.

To learn the distribution of cells projecting to the thalamus, as opposed to the cerebellum, in the mechanosensory nuclei of the dorsal medulla of raccoons, we analyzed the retrograde transport of horseradish peroxidase from the ventrobasal complex of the thalamus and from the cerebellum. We found six nuclear regions projecting heavily to the thalamus with very small projections to the cerebellum: Bischoff's, central cuneate, central gracile, rostral cuneate, rostral gracile nuclei, and cell group z. Two regions showed heavy projections to the cerebellum with no projections to the thalamus: the lateral portion of the external cuneate nucleus and the compact portion of cell group x. Four regions showed more equivalent projections to both target regions: basal cuneate, medial portion of the external cuneate nucleus, medial tongue extension of the external cuneate nucleus, and reticular portion of cell group x. Three more ventral regions were labeled: lateral cervical nucleus from thalamic injections but not from cerebellar injections; central cervical nucleus from cerebellar injections, which crossed the midline, but not from thalamic injections; and lateral reticular nucleus from both target regions. In most medullary regions, most cells project to one target and very few project to the other; we suggest that the cells projecting to the minor target convey samples of the information going to the major target.

Medullary sources of projections to the kinesthetic thalamus in raccoons: external and basal cuneate nuclei and cell groups x and z.

In raccoons and other mammals, a pathway for kinesthetic sensation (from muscles, fascia, tendons, and joints) reaches the anterodorsal cap of the ventrobasal thalamus and the anteriormost part of the somatic sensory cerebral cortex. To find the medullary component of this kinesthetic pathway in raccoons, small injections of horseradish peroxidase were made in the thalamus under guidance of simultaneous electrophysiological recording from kinesthetic projections. As determined by retrograde labeling following these injections, kinesthetic thalamic subregions receive projections as follows: caudomedial from cells in the external cuneate nucleus and its medial tongue, rostromedial from cells in basal cuneate nucleus, and rostrolateral from cells in cell group z and the reticular division of cell group x. Electrophysiological recording showed kinesthetic representations in each of these medullary regions. Labeled cells were also observed in the infratrigeminal subnucleus of the lateral reticular nucleus. Cats have kinesthetic projections to the thalamus from the basal cuneate and cell group z; raccoons (and monkeys) have these plus projections from the external cuneate and cell group x. This suggests that the kinesthetic projection system in raccoons and monkeys is expanded in correlation with their more dextrous use of the hand.

Demarcations of the mechanosensory projection zones in the raccoon thalamus, shown by cytochrome oxidase, acetylcholinesterase, and Nissl stains.

To determine anatomically the boundaries and internal organization of the kinesthetic and cutaneous mechanosensory regions of the ventrobasal thalamus, alternate section series from electrophysiologically mapped tissues from 14 raccoons were stained for cytochrome oxidase, myelinated fibers, acetylcholinesterase, and Nissl substance. Microelectrode tracks, along with electrolytic lesions placed as tissue markers, reveal that the mechanoreceptor projection zones have higher cytochrome oxidase and lower acetylcholinesterase staining than some neighboring regions. Both these enzymatic stains reveal particularly sharp boundaries separating the mechanoresponsive region, from the lateral posterior nucleus dorsally and from the ventroposterior inferior nucleus ventrally. The kinesthetic projection zone is often separated from other mechanoreceptor projections by bundles as well as laminae of myelinated fibers, similar to those separating cutaneous projections from distinct body parts. These subdivisions are particularly well marked by the cytochrome oxidase stain. The combination, in neighboring sections, of the use of the several stains adds considerably to the visible delineation of these functionally distinct regions, beyond what can be seen in Nissl-stained sections.

Organization of postcranial kinesthetic projections to the ventrobasal thalamus in raccoons.

To determine the presence and organization of kinesthetic, as compared with other mechanosensory projection zones in the thalamus of raccoons, unit-cluster responses to mechanical stimulation of the postcranial body were mapped electrophysiologically in the thalami of 14 raccoons anesthetized with Dial-urethane. A distinct zone of kinesthetic projections (from receptive fields in muscles, tendons, and joints) was found in the rostral and dorsal aspects of the mechanosensory projection zone. These projections are somatotopically organized: those from axial structures lie dorsalmost and those from successively more distal limb regions are successively more caudoventral. The kinesthetic forelimb representation is large and lies rostrodorsal to a large central core of cutaneous projections from the forepaw digits. A few scattered kinesthetic projections were found at the caudal edge of the sensory thalamic region. The large, spatially and somatotopically distinct kinesthetic projection zone in the thalamus parallels those seen in the cortex and medulla of raccoons. Similar findings in monkeys, and suggestions from data in cats and humans support the hypothesis of a distinct pathway to the cortex for kinesthetic information in all mammals.

Mechanosensory projections to cuneate, gracile, and external cuneate nuclei in a tree squirrel (fox squirrel, Sciurus niger).

The distribution and organization of mechanosensory projections to the cuneate, gracile and external cuneate nuclei were mapped in tree squirrels anesthetized with ketamine and urethane. Tungsten microelectrodes were used to record unit and unit cluster responses to mechanical stimulation. Most responses in the gracile nucleus were to stimulation of skin and hairs of the tail, hind foot and leg, and trunk, in that order going from dorsal to ventral in the nucleus. A similar distribution of responses was found in the cuneate nucleus to stimulation of forelimb, neck and pinna going from dorsomedial to ventrolateral in the nucleus. Some responses to stimulation or movement of subcutaneous tissue were found in the cuneate and gracile nuclei. Responses in the external cuneate were to stimulation of deep-lying tissues in the hand medially, the lower arm centrally and the shoulder and trunk laterally. The pattern of projections in the tree squirrel differed most strikingly from that seen in raccoons and opossums in the relatively small extent of projections from the glabrous skin of the forepaws which were concentrated in a small region near the obex in squirrels. In contrast, there was a large representation of hairy receptive fields located on the forelimb in all regions of the squirrel cuneate nucleus. Otherwise, the somatotopic organization of mechanosensory projections to these dorsal column nuclei in tree squirrels was similar to that reported in other mammals.

Contralateral as well as ipsilateral projections to raccoon cerebellum from external cuneate nuclei and cell groups f and x.

To assess the locations and densities of cells in the dorsal medulla giving rise to ipsilateral versus contralateral projections to the mechanosensory regions of cerebellar cortex in the anterior lobe and paramedian lobule, these cortical regions were injected unilaterally with horseradish peroxidase in each of 5 raccoons. To show injection sites and retrogradely labelled cells in the medulla, sections through the medulla and the cerebellum, in at least two different planes for each, were reacted with tetramethylbenzidine; alternate sections were reacted with cobalt-enhanced diaminobenzidine. Labelled were 60-80% of cells in the ipsilateral and 15-25% of cells in the contralateral external cuneate nuclei, as well as 25-50% of cells in the ipsilateral and the contralateral cell groups f and x near the descending vestibular nucleus. Substantial contralateral, as well as ipsilateral cerebellar projections from external cuneate nuclei and cell group x may be related to development of forelimb dexterity in raccoons, since these nuclei mediate forelimb muscular sensibility. The numerical complementarity of ipsilaterally versus contralaterally projecting cells suggests that they represent two separate populations.