The inferior olive of a prosimian primate, Galago senegalensis. II. Olivocerebellar projections to the vestibulocerebellum

Relationship between complex and simple spike activity in macaque caudal vermis during three-dimensional vestibular stimulation

Lobules 10 and 9 in the caudal posterior vermis [also known as nodulus and uvula (NU)] are thought important for spatial orientation and balance. Here, we characterize complex spike (CS) and simple spike (SS) activity in response to three-dimensional vestibular stimulation. The strongest modulation was seen during translation (CS: 12.8 +/- 1.5, SS: 287.0 +/- 23.2 spikes/s/G, 0.5 Hz). Preferred directions tended to cluster along the cardinal axes (lateral, fore-aft, vertical) for CSs and along the semicircular canal axes for SSs. Most notably, the preferred directions for CS/SS pairs arising from the same Purkinje cells were rarely aligned. During 0.5 Hz pitch/roll tilt, only about a third of CSs had significant modulation. Thus, most CSs correlated best with inertial rather than net linear acceleration. By comparison, all SSs were selective for translation and ignored changes in spatial orientation relative to gravity. Like SSs, tilt modulation of CSs increased at lower frequencies. CSs and SSs had similar response dynamics, responding to linear velocity during translation and angular position during tilt. The most salient finding is that CSs did not always modulate out-of-phase with SSs. The CS/SS phase difference varied broadly among Purkinje cells, yet for each cell it was precisely matched for the otolith-driven and canal-driven components of the response. These findings illustrate a spatiotemporal mismatch between CS/SS pairs and provide the first comprehensive description of the macaque NU, an important step toward understanding how CSs and SSs interact during complex movements and spatial disorientation.

Oculomotor cerebellum

The anatomical, physiological, and behavioral evidence for the involvement of three regions of the cerebellum in oculomotor behavior is reviewed here: (1) the oculomotor vermis and paravermis of lobules V, IV, and VII; (2) the uvula and nodulus; (3) flocculus and ventral paraflocculus. No region of the cerebellum controls eye movements exclusively, but each receives sensory information relevant for the control of multiple systems. An analysis of the microcircuitry suggests how sagittal climbing fiber zones bring visual information to the oculomotor vermis; convey vestibular information to the uvula and nodulus, while optokinetic space is represented in the flocculus. The mossy fiber projections are more heterogeneous. The importance of the inferior olive in modulating Purkinje cell responses is discussed.

Vestibular signals in the parasolitary nucleus

Vestibular primary afferents project to secondary vestibular neurons located in the vestibular complex. Vestibular primary afferents also project to the uvula-nodulus of the cerebellum where they terminate on granule cells. In this report we describe the physiological properties of neurons in a "new" vestibular nucleus, the parasolitary nucleus (Psol). This nucleus consists of 2,300 GABAergic neurons that project onto the ipsilateral inferior olive (beta-nucleus and dorsomedial cell column) as well as the nucleus reticularis gigantocellularis. These olivary neurons are the exclusive source of vestibularly modulated climbing fiber inputs to the cerebellum. We recorded the activity of Psol neurons during natural vestibular stimulation in anesthetized rabbits. The rabbits were placed in a three-axis rate table at the center of a large sphere, permitting vestibular and optokinetic stimulation. We recorded from 74 neurons in the Psol and from 23 neurons in the regions bordering Psol. The activity of 72/74 Psol neurons and 4/23 non-Psol neurons was modulated by vestibular stimulation in either the pitch or roll planes but not the horizontal plane. Psol neurons responded in phase with ipsilateral side-down head position or velocity during sinusoidal stimulation. Approximately 80% of the recorded Psol neurons responded to static roll-tilt. The optimal response planes of evoked vestibular responses were inferred from measurement of null planes. Optimal response planes usually were aligned with the anatomical orientation of one of the two ipsilateral vertical semicircular canals. The frequency dependence of null plane measurements indicated a convergence of vestibular information from otoliths and semicircular canals. None of the recorded neurons evinced optokinetic sensitivity. These results are consistent with the view that Psol neurons provide the vestibular signals to the inferior olive that eventually reached the cerebellum in the form of modulated climbing fiber discharges. These signals provide information about spatial orientation about the longitudinal axis.

A review of otolith pathways to brainstem and cerebellum

Our knowledge of otolith pathways is developing rapidly, but is still far from complete. Primary afferents from the sacculus and utricle terminate mainly in the lateral, inferior and caudal superior vestibular nuclei, and the ventral cerebellum, in particular the nodulus. Otolith signals descend via reticulo- and vestibulospinal pathways in the spinal cord to influence neck motoneurons and ascending proprioceptive afferents. Utricular information can reach the extraocular eye muscles via mono-, di-, and multisynaptic pathways, but saccular afferents probably only by multisynaptic pathways. The otolith signals are relayed from the vestibular nuclei, medullary reticular formation, inferior olive, and lateral reticular nucleus to sagittal zones in the caudal cerebellar vermis (nodulus and uvula), and influence the deep cerebellar nuclei. The graviceptive information could be channeled by the cerebellar efferents back to the vestibular and inferior olive complex, or fed into ascending pathways that would innervate the mescencephalon, the thalamus, and cerebral cortex.

Zonal organization of the climbing fiber projection to the flocculus and nodulus of the rabbit: a combined axonal tracing and acetylcholinesterase histochemical study

The localization and termination of olivocerebellar fibers in the flocculus and nodulus of the rabbit were studied with anterograde axonal transport methods [wheatgerm agglutinin-horseradish peroxidase (WGA-HRP) and tritiated leucine] and correlated with the compartments in the white matter of these lobules delineated with acetylcholinesterase histochemistry (Tan et al. J. Comp. Neurol., 1995, this issue). Olivocerebellar fibers originating from the caudal dorsal cap travel through floccular compartments FC2 and FC4 to terminate as climbing fibers in floccular zones FZII and FZIV. Fibers from the rostral dorsal cap and the ventrolateral outgrowth traverse compartments FC1 and FC3, which are interleaved with compartments FC2 and FC4, and terminate in zones FZI and FZIII. Fibers from the rostral pole of the medial accessory olive traverse the C2 compartment and terminate in the C2 zone. FZI-III extend into the adjoining folium (folium p) of the ventral paraflocculus. The C2 zone continues across folium p into other folia of the ventral paraflocculus and into the dorsal paraflocculus. Four compartments and five zones were distinguished in the nodulus. Medial compartment XC1 contains olivocerebellar fibers from the caudal dorsal cap and subnucleus beta that terminate in the XZI zone. Olivocerebellar fibers from the rostral dorsal cap and the ventrolateral outgrowth occupy XC2 and terminate in XZII. The XC4 compartment contains fibers from both the caudal dorsal cap and from the rostral dorsal cap and the ventrolateral outgrowth. The latter terminate in a central portion of the XZIV zone. The dorsomedial cell column projects to the XZIII zone, which is present only in the dorsal part of the nodulus. The rostral medial accessory olive projects to the XZV zone, which occupies the lateral border of the nodulus. These results confirm and extend the conclusions of Katayama and Nisimaru ([1988] Neurosci. Res. 5:424-438) on the zonal pattern in the olivo-nodular projection in the rabbit. Additional observations were made on the presence of a lateral A zone (Buisseret-Delmas [1988] Neurosci. Res. 5:475-493) in the hemisphere of lobules VI and VII. Retrograde labeling of the nucleo-olivary tract of Legendre and Courville ([1987] Neuroscience 21:877-891) was observed after WGA-HRP injections into the inferior olive including the rostral dorsal cap and the ventrolateral outgrowth. The anatomical and functional implications of these observations are discussed.

Distribution of corticotropin-releasing factor and calcitonin gene-related peptide in the developing mouse cerebellum

Corticotropin-releasing factor (CRF)-like immunoreactive (IR) fibers were investigated ontogenically in the mouse cerebellum. CRF-IR was detected in the climbing fiber and mossy fibers as in other species. In addition, CRF-IR dense fiber plexuses were detected from postnatal day (PD) 2 to 9, in the developing Purkinje cell layer of the vermal lobules, paraflocculus, flocculus and crus 1 ansiform lobule, gradually forming a pericellular nest around the Purkinje cell somata. Immunoelectron-microscopical analysis showed that dense fibers made synaptic contacts with the Purkinje cell somata on PD 7. In the lobules mentioned above, CRF-IR dense fibers showed parasagittal banded patterns. Calcitonin gene-related peptide (CGRP)-IR showed similar fiber bands at these stages. Interestingly, these two patterns of peptidergic fiber bands were complementary in distribution. From around PD 9, CRF-IR fibers lost the immunoreactive dots in the Purkinje cell layer. Immunoreactivity at this stage was observed in the axons projecting to the molecular layer, and thin CRF-IR fibers began to appear in the neighboring area. Numerous typical climbing fiber-like CRF-IR fibers were found throughout the cerebellar cortex from PD 16 to adult. The inferior olivary complex (the origin of climbing fibers) appears to be the origin of these dense fiber plexuses as CRF-IR cells were already present from PD 2 in the dorsal cap nucleus, beta subnucleus and caudomedial part of the accessory olivary nucleus. No neurons containing both CRF and CGRP immunoreactivities were observed. These results suggest that CGRP- and CRF-IR developing climbing fibers innervate different compartments of Purkinje cells, especially in the vestibular cerebellar cortex in mice. Furthermore, CRF-IR fibers gradually changed to become typical climbing fibers, while CGRP-IR disappeared altogether.

Zonal organization of climbing fiber projections to the nodulus in the cat

Climbing fiber projections from the inferior olive to the nodulus of the cerebellum were studied by means of the retrograde transport of horseradish peroxidase in the cat. Following large and small injections into various parts of the nodulus, the distribution of labeled cells in the inferior olive was investigated. The findings indicate the existence of five longitudinal zones extending throughout the dorsal and ventral nodulus: (1) the caudal part of the nucleus beta projects to a most medially located zone (caudal beta zone) with a width of about 0.6 mm; (2) the rostral part of the nucleus beta projects to a zone located at 0.6-1.2 mm from the midline (rostral beta zone); (3) the caudal part of the dorsal cap (dc) projects to a zone located in the intermediate part of the nodulus at 1.2-1.8 mm from the midline (caudal dc zone); (4) the ventrolateral outgrowth (vlo) of the dc and the rostral part of the dc project to a zone at 0.3-0.9 mm from the lateral edge of the nodulus (rostral dc and vlo zone); and (5) finally the dorsomedial cell column (dmcc) projects to the most lateral zone (dmcc zone).

Olivary morphology and olivocerebellar topography in adult lurcher mutant mice

In adult lurcher mice, in which virtually all cerebellar Purkinje cells have degenerated as a direct consequence of mutant gene action, the inferior olivary complex suffers a severe retrograde transneuronal atrophy. Our analysis indicates a 63% cell loss in the lurcher inferior olive, homogeneously distributed between the medial and dorsal accessory, and principal olivary subdivisions. Olivary neurons are reduced in cross-sectional area by 30% in lurcher mice, compared to normal controls. All olivary subdivisions morphologically identifiable in normal mice are also found in the lurcher inferior olive. Analysis of olivocerebellar topography by retrograde transport of lectin-conjugated horseradish peroxidase and fluorogold, in both single and double labeling paradigms, reveals no abnormalities in the general organization of this highly ordered projection. This stability may be based on the initial establishment of the topographic pattern in late embryogenesis or early postnatal periods, prior to the onset of lurcher Purkinje cell degeneration, or, alternatively, the lurcher gene may not alter critical afferent and target characteristics at stages when the topographic relationship is being established. Once established, the olivocerebellar projection is apparently not dependent on the Purkinje cell for long-term maintenance of its general topographic organization.

Evidence of an x zone in lobule V of the squirrel monkey (Saimiri sciureus) cerebellum: the distribution of corticonuclear fibers

The distribution of corticonuclear fibers to medial-most parts of the posterior interposed nucleus (NIP) from lateral areas of the vermis was studied in the squirrel monkey (Saimiri sciureus), using a silver impregnation method. The origin and course of degenerated fibers were studied in serial sections. The distribution pattern of corticonuclear fibers from a series of small well localized lesions placed in the vermis and paravermal cortex of lobule V is compatible with the interpretation that an x zone is present in Saimiri. A comparison of the positions of lesions and the trajectory of fibers arising therein suggests that corticonuclear input to medial-most parts of the NIP originated from a narrow cortical area (about 0.5-0.7 mm wide) located between a cortical area projecting into the medial cerebellar nucleus (the A zone) and a laterally adjacent area (the B zone) which related to the lateral vestibular nucleus. This NIP-projecting cortical area, located about 1.7 mm to 2.5 mm off the midline in lobule V, is interpreted as the x zone in this primate; it extends from lobule IV into lobule VI in squirrel monkey. Corticonuclear fibers of zone x in this primate form a comparatively small terminal field in the medial-most portions of NIP. This contrasts with the distribution of corticonuclear fibers of the C2 zone which consistently distribute to terminal fields that are shifted into more central areas of NIP. There appears to be no overlap of the corticonuclear terminal fields in the NIP for zone x versus the C2 zone. These results were correlated with data from the literature on the distribution of olivocerebellar fibers to the x zone and the C2 zone and the arrangement of cerebellar nucleoolivary projections into the inferior olive from the NIP. The x zone and the C2 zone both receive input from the contralateral medial accessory olive (MAO), both zones project into the NIP, and the NIP projects into those regions of the MAO which, in turn, project to these respective cortical zones and into the NIP. This suggest that the x zone is a component of the NIP-MAO circuit. Furthermore the proposed function of the x zone would support the view that this sagittal strip may have a more extensive rostrocaudal distribution in primates as compared to the cat.

Columnar organisation of the inferior olive projection to the posterior lobe of the rat cerebellum

The organisation of the olivocerebellar projection to lobules VI, VIII, and IX of the posterior lobe of the rat cerebellum was investigated in detail by using the retrograde tracer wheat germ agglutinin-horseradish peroxidase. Small, well-defined rostro-caudally orientated columns of olive cells were found to project to different parasagittal areas in the posterior lobe. A column of olive cells about 2,000 microns in rostro-caudal length in subnucleus "c" and nucleus beta of the caudal medial accessory olive (MAO) provides climbing fibre input to the most medial part of lobules VI and IX, but this projection is displaced laterally in lobule VIII by a projection from a column of cells about 600 microns in rostro-caudal length in lateral caudal MAO (subnucleus "a"). It is possible that each of these columns of olivary neurones may be further subdivided in the rostro-caudal axis so that different sections project to different medio-lateral parts of the cortex. A fine-grain 'sublobular' localisation may also exist: the projection to midline lobule VIc arises at caudal levels of the olive from a band of cells in the transition region between subnucleus "c" and nucleus beta, whilst by comparison the projection from caudal levels of the olive to lobules VIa and VIb arises from cells located more ventrally in nucleus beta. Evidence is also presented to confirm that the posterior lobe vermis in the rat extends further laterally than in other mammals and that part of it receives a projection from a column of olive cells, 1,000 microns in rostro-caudal length, in a newly defined region of caudal MAO, termed subnucleus "b1."

Afferents to the cerebellar flocculus in cat with special reference to pathways conveying vestibular, visual (optokinetic) and oculomotor signals

Horseradish peroxidase (HRP) was injected into the cerebellar flocculus of 20 cats to determine: (a) the proportions of afferents from the various brain stem nuclei; (b) possible projections from the basilar pontine nuclei; and (c) sources of saccadic eye movement signals recorded from flocculus Purkinje cells. Results confirm earlier findings that the flocculus receives large numbers of mossy fibre afferents from the vestibular and perihypoglossal nuclei, bilaterally, and climbing fibres from the contralateral inferior olive (dorsal cap, ventrolateral outgrowth, medial accessory olive, ventral bend of principal olive). In addition, large numbers of HRP-labeled neurons have been identified within: (i) the basilar pontine nuclei, bilaterally, where they are distributed in columns in the dorsolateral, lateral, ventral medial and dorsomedial nuclei; (ii) the nucleus reticularis tegmenti pontis; (iii) several of the cranial motor nuclei, VI, VII, X (retrofacial n.), XI (n. ambiguus), and XII; (iv) the raphe magnus, pontis and obscurus; (v) the lateral reticular nucleus, pars subtrigeminalis. Finally, new information is presented which shows that large numbers of flocculus projecting neurons are located within the medial longitudinal fasciculus at two locations; one just rostral to the hypoglossal nucleus and another group extends 2-3 mm rostral to the abducens nucleus. These groups are bilateral, and have been termed, respectively, the caudal and intermediate interstitial nucleus of the medial longitudinal fasciculus. Both groups correspond in location to physiologically identified neurons in cat which fire in relation to saccadic eye movements. Their projection to the flocculus, in part, explains the saccadic discharge of Purkinje cells in the flocculus.

Zonal organization of climbing fiber projections to the uvula in the cat

Climbing fiber projections from the inferior olive to the uvula of the cerebellum were studied in the cat by using retrograde axonal transport of horseradish peroxidase. Following large and small injections into various parts of the uvula, the distribution of labeled cells in the inferior olive was investigated. The findings indicate six longitudinal zones extending throughout the dorsal and ventral uvula: the caudal part of the nucleus beta projects to a most medially located zone (caudal beta zone) with a width of about 0.4 mm; the rostral part of the nucleus beta projects to a zone located at about 0.6 mm from the midline (rostral beta zone); the caudal part of the medial accessory olive (MAO) projects to a zone (caudal MAO zone) located lateral to the rostral beta zone; the dorsomedial cell column projects to a zone (dorsomedial cell column zone) located in the intermediate part of the uvula at about 1.2 mm from the lateral edge of the uvula; the ventral lamella of the principal olive (PO) projects to a zone (ventral lamella of PO zone) about 0.7 mm from the lateral edge of the uvula; finally, the rostral part of the MAO projects to the most lateral zone (rostral MAO zone). These conclusions are in general agreement with those of earlier studies and also provide a more detailed zonal configuration of climbing fiber projections to the uvula.

Parasagittal zonal pattern of olivo-nodular projections in rabbit cerebellum

Injection of horseradish peroxidase (HRP) into the nodulus of the rabbit retrogradely labeled cells in 5 subdivisions of the contralateral inferior olive. Localized HRP injections at each of the medial, intermediate and lateral parts of the nodulus revealed 6 longitudinal zones, two zones in each part, projected differentially from the 5 olivary subdivisions: (i) the medial-most zone projected from the beta nucleus, (ii) the lateral zone of the medial part and (iii) the lateral zone of the intermediate part both from the dorsal cap, (iv) the medial zone of the intermediate part from the ventrolateral outgrowth, (v) the medial zone of the lateral part from the dorsomedial cell column, and (vi) the lateral-most zone from the rostrolateral part of the medial accessory olive. A complication is that the zones (v) and (vi) seemed to cover the dorsal lamina of the nodulus, but not the ventral lamina. The lateral part of the ventral nodulus seemed to be projected from the dorsal cap and may therefore be a lateral extension of zone (iv). There was an indication that the rostral-most area of the medial accessory olive also projects to the nodulus, but a specific receptive zone for this projection was unclear. The present results suggest that the small lateral area of the nodulus previously found to be involved in cardiovascular control is projected from the rostrolateral part of the medial accessory olive and/or the dorsomedial cell column.

The olivonodular projection: a re-examination based on folial cerebellar implants

In 5 cats implants of crystalline wheat germ agglutinin-horseradish peroxidase complex were placed in one or two folia of the nodulus with no contamination of adjacent cerebellar lobules. In the inferior olive only the dorsal cap with the adjacent ventrolateral outgrowth and the caudalmost part of nucleus beta were found to project to the nodulus. Negative findings were consistently made in all other parts of the inferior olive, some of which have previously been described to contribute to the olivonodular projection.

Topographical distribution of Purkinje cells in the uvula and the nodulus projecting to the vestibular nuclei in cats

The localization of the Purkinje cells in the uvula and nodulus projecting to the vestibular nuclei and the prepositus hypoglossal nucleus (PH) was studied by means of retrograde axonal transport of horseradish peroxidase in cats. Findings indicate a zonal organization in the uvula and nodulus projecting to the vestibular nuclei as follows; the Purkinje cells located in the medial half of the uvula except for the area along the posterolateral fissure project to the middle part of the inferior vestibular nucleus (IV) (middle IV zone); those in the lateral half of the uvula other than the laterocaudal part project to the caudal part of the IV (caudal IV zone); those in the mediorostral part of the nodulus and the middle part of the nodulus project to the middle part of the medial vestibular nucleus (MV) (middle MV zone); those in the lateral part of the nodulus project to the caudal part of the MV (caudal MV zone); those in the medial part of the uvula and nodulus along the posterolateral fissure project to the dorsal peripheral part of the superior vestibular nucleus (SV) (SV zone). There is no specific projection zone in the uvula and nodulus projecting to the lateral vestibular nucleus, the ventral peripheral and the central part of the SV, the rostral part of the MV, the rostral part and the caudal pole of the IV, the caudal one-third of the group f, the group x and the PH.

The inferior olive of Saimiri sciureus: olivocerebellar projections to the anterior lobe

Olivocerebellar projections to lobules IV and V of the anterior lobe of the squirrel monkey (Saimiri sciureus) cerebellum were studied using a wheat germ agglutinin-horseradish peroxidase conjugate (WGA-HRP). Placements of WGA-HRP in the cerebellar cortex, labeled neurons in different subgroups of the contralateral inferior olive depending on the position of the injection site. Injections in medial aspects of lobules IV and V labeled cells in subgroups a and b of the caudal medial accessory olive (MAO) and in the lateral dorsal accessory olive. Rostral parts of subgroup a of MAO and medial dorsal accessory olive contained labeled neurons when injections were made in intermediate areas of the cerebellar cortex. Subsequent to involvement of more lateral parts of lobules IV and V, labeled cells were found in the dorsal lamella and lateral bend of the principal olive. When the position and size of injection sites were compared to distribution of retrogradely labeled olivary somata, it was clear that zones A, B, C1, C2, C3 and D were present in lobules IV and V of squirrel monkey. The general pattern of olivary labeling in this study is similar to that reported for other species.

Zonal organization of olivocerebellar projections to the uvula in rabbits

Olivocerebellar projections to the uvula were studied by means of retrograde axonal transport of horseradish peroxidase (HRP) in pigmented rabbits. The distribution pattern of labeled cells in the inferior olive was compared among cases following large- and microinjections of HRP into the uvula. Findings indicate topographically organized projections to longitudinally oriented zones. There are at least 6 zones in the rabbit's uvula. The caudal part of the nucleus beta projects contralaterally to a most medially located zone (caudal beta zone). The rostral part of the nucleus beta projects to a little more laterally located zone (rostral beta zone) at a distance of about 1 mm from the midline of the uvula. The caudolateral part of the MAO projects to a zone (caudolateral MAO zone) located laterally to the rostral beta zone. The dorsomedial cell column projects to a zone (dorsomedial cell column zone) located in the intermediate part of the uvula at about 2 mm from the midline. The rostrolateral part of the MAO projects to the most lateral zone (rostrolateral MAO zone) of the uvula. Finally, the ventral lamella of the PO projects to a zone (ventral lamella of PO zone) located between the rostrolateral MAO zone and the dorsomedial cell column zone.

Afferents to the flocculus of the cerebellum in the rhesus macaque as revealed by retrograde transport of horseradish peroxidase

To investigate the afferent projections to the flocculus in a nonhuman primate, we injected horseradish peroxidase into one flocculus of six rhesus macaques (Macaca mulatta) and processed their brains according to the tetramethylbenzidine protocol to reveal retrogradely labeled neurons. Labeled neurons were found in a large set of nuclei within the rostral medulla and the pons. The greatest numbers of labeled neurons were in the vestibular complex and the nucleus prepositus hypoglossi. There were neurons labeled bilaterally throughout all the vestibular nuclei except the lateral vestibular nucleus, but most of the labeled neurons were in the caudal parts of the medial and inferior vestibular nuclei and in the central part of the superior vestibular nucleus; the nucleus prepositus was also labeled bilaterally, primarily caudally. Modest numbers of labeled neurons were found in the y-group, most ipsilaterally, and many neurons were labeled in the interstitial nucleus of the vestibular nerve. No labeled neurons were found in the vestibular ganglion following a large injection into the flocculus. A second large source of afferents to the flocculus was the medial, paramedial, and raphe reticular formation. Dense aggregates of labeled neurons were located in several pararaphe nuclei of the rostral medulla and the rostral pons and in the nucleus reticularis paramedianus of the medulla and several component nuclei of the nucleus reticularis tegmenti pontis bilaterally. Several groups of cells within and abutting upon the medial and rostral aspects of the abducens nucleus were labeled bilaterally. There was a modest projection from two parts of the pontine nuclei. Both a dorsal midline nucleus ventral to the nucleus reticularis tegmenti pontis and a collection of nuclei in a laminar region adjacent to the contralateral middle cerebellar peduncle contained labeled neurons whose numbers, while modest, were large compared to the projections to the flocculus in other animals. This generic difference may be due to the greater development of the smooth pursuit system in monkeys and the consequent need for a more substantial input from the cerebral cortex. As in other genera, the inferior olive projected to the flocculus via the dorsal cap of Kooy and the contiguous ventrolateral outgrowth. The projection was completely crossed and large injections labeled virtually every neuron in the dorsal cap, suggesting that the dorsal cap is the principal source of climbing fiber afferents.(ABSTRACT TRUNCATED AT 400 WORDS)