THE JOURNAL OF COMPARATIVE NEUROLOGY 306:221-244 (1991)

Organization of the Cerebellum in the Pigeon (CoZumba Zivia):I. Corticonuclear and Corticovestibular Connections J.J.A. ARENDS AND H. PHILIP ZEIGLER Biopsychology Program, Hunter College, City University of New York, New York, New York 10021

ABSTRACT The projections of the cerebellar cortex upon the cerebellar nuclei and the vestibular complex of the pigeon have been delineated using WGA-HRP as an anterograde and retrograde tracer. Injections into individual cortical lobules (11-1%) produce a pattern of ipsilateral terminal labeling of both the cerebellar and vestibular nuclei. The pattern of corticonuclear projections indicates both a rostrocaudal and a mediolateral organization with respect to the lobules and is consistent with a division of the cerebellar nuclei into a medial (CbM) and a lateral (CbL) nucleus. The retrograde experiments indicate that these nuclei receive projections, respectively, from Purkinje cells within medial (A) and lateral (C) longitudinal zones, which alternate with longitudinal zones (B, E ) projecting upon the vestibular complex. Purkinje cells in (vestibulocerebellar) lobules Ixb-X show only limited projections upon the cerebellar nuclei, but do project extensively upon the cerebellovestibular process (PCV), as well as upon the medial, superior, and descending vestibular nuclei. As the injection site shifts from medial to lateral, there is a corresponding shift in focus of the projection within PCV from areas bordering CbM to those abutting CbL. The topographic organization of corticovestibular projections is less clear-cut than those of the corticonuclear projections. Lobules 11-X project upon the lateral vestibular nucleus (anterior lobe) or the dorsolateral vestibular nucleus (posterior lobe). These projections originate from either side of the lateral (C) zone. Projections originating from the medialmost (B) zone are interrupted in lobules VI and VII. The anterior and posterior portions of the lateralmost (E) zone overlap along lobules VI and VII. In addition, the E zone of the anterior lobe is the source of projections upon the medial, the descending, and the superior vestibular nuclei. Projections from the auricle and adjacent lateral unfoliated cortex (F zone) focus upon the infracerebellar nucleus, the medial tangential nucleus, and the medial division of the superior vestibular nucleus. The data suggest that the cerebellar cortex of the pigeon, like that of mammals, may be subdivided into a mediolaterally oriented series of longitudinal zones, with Purkinje cells in each zone projecting ipsilaterally to specific cerebellar nuclei or vestibular regions. For cortical regions exclusive of the auricle and lateral unfoliated cortex, three such zones (A, B, and C) are defined that are comparable in their efferent targets with the A, B, and C zones of mammals. There does not appear to be a D zone in the pigeon. The results are discussed in relation to comparative data on amphibians, reptiles, and mammals. Key words: birds, comparative, horseradish peroxidase, parasagittal zonation

The cerebellum is well developed in birds (Larsell, '48, '67; Pearson, '72). Although lacking the hemispheric specialization of mammals, the avian cerebellum, like that of other amniotes, is divisible into a more superficial cortical region and a group of deep cerebellar nuclei. Indeed, in midsagittal section the avian and mammalian cerebella are remarkably similar in appearance. The avian cerebellar Cortex is divisible into granular, molecular, and Purkinje layers, and the

o 1991 WILEY-LISS, INC.

organization of its neuronal elements appears similar to that of mammals (Pearson, '72). Like its mammalian Accepted December 19' Address reprint requests to H.P. Zeigler, Hunter College, Dept. of Psychology, 695 park Ave,, NewYork, N y 1o021, J.J.A. .&-ends is now at the Dept. of Anatomy and Neurobiology, St. Louis University School of Medicine, 1402 S. Grand Blvd., St Louis, MO 63104.

J.J.A. ARENDS AND H.P. ZEIGLER

222 counterpart, it receives afferents from a wide variety of CNS regions including the spinal cord, inferior olive, reticular formation, trigeminal and vestibular nuclei, pontine nuclei, and a number of mesencephalic and diencephalic structures (e.g., Brodal et al., '50; Whitlock, '52; Gross, '70; Karten and Finger, '76; Clarke, '77; Arends et al., '84; Arends and Zeigler, '89). As in mammals, these inputs are conveyed to the cerebellar cortex via mossy fiber andlor climbing fiber systems (Feirabend, '83). Purkinje cells function as the output elements of the cerebellar cortex projecting upon the cerebellar and specific vestibular nuclei (Wold, '81; Arends, '85; Arends and Voogd, '89). The organization of the avian cerebellum is of both anatomical and functional interest. Despite the absence of any structure unequivocally homologous with mammalian motor cortex, birds exhibit precise control of axial and distal musculature during manipulative and locomotor behaviors. For example, studies of jaw motor control in the pigeon have shown that its grasping and manipulation behaviors have many functional similarities with the prehensile behavior of primates and involve motor control strategies similar to those employed by humans in analogous motor tasks (Klein et al., '85; Bermejo and Zeigler, '89; Bermejo et al., '891. Analysis of cerebellar organization in the pigeon should provide data relevant to the general problem of brain evolution as well as laying the foundation for neurobehavioral studies of motor control mechanisms. Previous reports described the organization of trigeminocerebellar and olivocerebellar connections (Arends and Voogd, '89; Arends and Zeigler, '89). The present study delineates the pattern of corticonuclear and corticovestibular connections. The two companion studies examine the cerebellar efferent projections (Arends and Zeigler, '91) and the organization of corticovestibular connections with eye and neck premotor areas (Arends et al., '91). Some of the data on which these studies are based have been presented in preliminary form (Arends et al. '83; Arends, '85).

MATERIALS AND METHODS Male and female White Carneaux or Silver King pigeons were anesthetized with loco-Equithesin (2 mlkg, im) and placed in a Kopf stereotaxic instrument with a pigeon head

holder (Karten and Hodos, '67). Injections of the more dorsal lobules, the lateral unfoliated cortex and the auricle, were made under visual control. Injections in other cerebellar and brainstem targets were guided by stereotaxic coordinates (Karten and Hodos, '67). Pressure injections of 2-10% WGA-HRP in 0.1 M Tris Buffer, pH 7.8 were made through a 1 pL Hamilton syringe with a micropipette (tip diameter 20-40 pm) glued to the syringe. Iontophoretic injections of WGA-HRP ( 2 4 % in 0.01 M Tris Buffer, pH 8.0) were made using micropipettes (tip diameters 10-60 pm) and a Midgard CS-3 current source. Continuous currents ranged from 2-4 pA for periods of 5-15 minutes. At the end of each injection, the electrode was disconnected but left in place for 10 minutes. Following survival times of 24-48 hours, birds were anesthetized and perfused by bilateral carotid catheterization with a warm (40°C) solution of 4% dextran in 0.9% saline, followed by a cold (4°C) solution of 4% glutaraldehyde and 5% sucrose in 0.1 M phosphate buffer (pH 7.4). Brains were stored overnight in a mixture of 4% glutaraldehyde and 30% sucrose in 0.1 M phosphate buffer, cut in 40-50-pm coronal sections on a freezing microtome, and collected in 0.1 M acetate buffer. Sections were reacted for the presence of HRP with TMB using the protocol of Mesulam ('781, but total incubation duration was increased to 1.5 to 2 hours by rinsing sections briefly in 0.01 M acetate buffer (pH 3.3) and transferring them to fresh, complete reaction medium every 30 minutes (personal communication, R.M.L. Faull). Selected sections through the injection site were treated again with the DAB-heavy metal intensification method of Adams ('81)and used to estimate the size of the injection sites. Alternate sections were left unstained or were counterstained with the rapid thionin procedure of Adams ('80). Sections were examined under both brightfield and darkfield illumination and tracings of labeled cells, axons, and terminal fields were made on projection drawings of Nissl-stained sections using a drawing tube or a microprojector. To aid data presentation, cortical injection sites were reconstructed and projected upon a midsagittal view of the cerebellum. The data on which this work is based were obtained in experiments with 95 pigeons, some with bilateral injections. Delineation of cortical efferent projections is based

Abbreviations An Aur BC BCP CbLlL CbMiM

ic im in vm CCD

ccv

CJR el CR CTrX FLM I-x Infli La Li

nucleus angularis cochlearis auricle brachium conjunctivum brachium cerebellopetale nucleus cerebellaris lateralis nucleus cerebellaris medialis pars intercalata pars intermedialis pars interna pars ventromedialis commissura cerebellaris dorsalis commissura cerebellaris ventralis corpus juxtarestiforme lateral (unfoliated) cortex corpus restiforme crossed cochlear tract fasciculus longitudinalis medialis lobules I-X nucleus infracerebellaris nucleus laminaris cochlearis lingula

Mc mVI PC PCVIp PrV PHiph RIX RVII RVIIIc RVIIIv Tdt Tec Tel WTTD VDLldl VeDid VeLdild VeLv VeMlm VeSis v4lvi4

nucleus magnocellularis cochlearis nucleus nervi abducentis cerebellar peduncle processus cerebellovestibularis nucleus sensorius principalis nervi trigemini prehypoglossal area radix nervi glossopharyngei radix nervi facialis radix nervi octavus, pars cochlearis radix nervi octavus, pars vestibularis nucleus vestibularis tangentialis optic tectum telencephalon (nucleus)tractus descendens newi trigemini nucleus vestibularis dorsolateralis nucleus vestibularis descendens nucleus vestibularis lateralis, pars dorsalis nucleus vestibularis lateralis, pars ventralis nucleus vestibularis medialis nucleus vestibularis superior IVth ventricle

PIGEON CEREBELLAR CORTICAL EFFERENTS

Fig. 1. External morphology of the cerebellum in the pigeon. (a,b,c) In situ dorsal, lateral, and caudal views; (d) ventral view; (e) lateral view illustrating the location and nomenclature of the individual

223

lobules. Asterisks, auricle; arrows in (d) indicate lateral recess of the fourth ventricle; arrows in ( c ) and (e) indicate the cerebellar peduncles.

224

J.J.A. ARENDS AND H.P. ZEIGLER

Fig. 2. Photomicrographs of serial cross sections (30 pm; Nissl stain) through the cerebellar nuclei of the pigeon. Every fourth section was selected except those in (e) through (h),which represent every second section. Explanatory drawings are provided in Figure 3. Scale bar equals 1mm.

PIGEON CEREBELLAR CORTICAL EFFERENTS

Fig. 3. Outline drawings of structures shown in Figure 2. The shaded area in the drawings outlines the location of the cerebellovestibular process (PCV); striations indicate the location of the cerebellopetal brachium (BCP). (i)through (n)indicate the location of the infracerebellar nucleus; (i) through (1) that of the dorsolateral vestibular nucleus (VDL).

225

J.J.A. ARENDS AND H.P. ZEIGLER

226

I+,

moderate to heavy; - , minimal to light;

-. none; ?, not scored because of passing fiber confounds; inj, not scored because of proximity to injection site

from the surrounding white matter at most levels, its borders with parts of the subjacent structures (e.g., the cerebellovestibular process: PCV) are sometimes indistinct. Figures 2 and 3 present a series of photomicrographs and drawings of transverse sections through the cerebellar RESULTS nuclei and adjacent areas in the pigeon. Two main cerebelNormal morphology of the cerebellar cortex lar nuclear masses are apparent; a more dorsally situated and the central cerebellar nuclei in the pigeon medial (CbM) and a more ventrally situated lateral (CbL) division. Within CbM, several subnuclei may be distinFigure la-' the location and grossm o r p h o l o ~ of the cerebellum of the pigeon. The most striking features guished on the basis of their cyto-and fibroarchitecture, ofthis structure include a division into a number of lobules, including internal (CbMin), intermediate (CbMim), and or folia (b,e), a region of unfoliated cortex (e) and (d) the intercalate (CbMic) divisions (see also Pearson, '72; Wold auricle: a rostrolateral protuberance of lobules IXb and X. '81). The figures also redefine two structures indicated in Lobules and sublobules are labeled according to Larsell the atlas (PCV and VDL; Karten and Hodos, '67) and ('48, '671, except for lobule IX where we recognize only introduce a third: the infracerebellar nucleus (Inf). The (IX)a and (1X)bsubdivisions (Fig. le). In our nomenclature, delineation of these nuclei is based on both cytoarchitecture lobules IXb and x the vestibu~ocerebe~~um in the and hodoloa; their status in relation to homotypical nuclei pigeon (Schwarz and Schwarz, '83). The ventral view in (d) in mammals is discussed in this and the two companion also indicates the unique morphology of lobules I and X, the fourth ventricle with its lateral recesses, and the cerebellar Anterograde tracing experiments peduncles, which connect the cerebellum with the underlving brainstem (c). To determine the organization of corticonuclear/cortiThe cerebellar nuclei of the pigeon are located in the wall covestibular projections, injections were made into each of of the fourth ventricle, extending rostrocaudally from the the 10 cerebellar lobules, including the lateral unfoliated level of the principal sensory nucleus of the trigeminus (A. cortex and the auricle but excluding the distal portion of the 0.60) to the caudal tip of the laminar cochlear nucleus (ca. lingula (lobule I). These injections varied in their size, in P.1.10). Whereas the nuclear mass may be distinguished their mediolateral extent, and in their proximodistal locaupon anterograde studies employing 66 birds; 29 birds were used in retrograde studies defining the zonal organization of the cortex.

~~

PIGEON CEREBELLAR CORTICAL EFFERENTS

227

b

Yr 1 ' Fig. 4. Six sets (a-f)of serial cross sections through the cerebellar nuclei of the pigecn, illustrating the distribution of corticonuclear labeling (shading) after injections of WGA-HRP into each of the cerebellar lobules I-V. The top two rows indicate the location of the cortical injection sites in a standardized midsagittal projection (top row), and an actual representative cross section (second row). In some

p c

cases the extent of the injection site was estimated from both DABprocessed (black) and TMB-processed (shaded) material. In other cases this distinction was either irrelevant (complete lobular injections) or nonexistent (small iontophoretic injections). Case numbers: (a) 8336; (b) 8375; (c) 8546; (d) 8444; (el 8429; (0 8344.

228

.

J.J.A. ARENDS AND H.P. ZEIGLER

Js

Q

Fig. 5. Six sets (g-1) of serial cross sections through the cerebellar nuclei of the pigeon, illustrating the distribution of corticonuclear labeling (shading) after injections of WGA-HRP into each of the cerebellar lobules V-VI. Conventions as in Figure 4. Case numbers: (g)8427; (h) 8441; (i) 8443; (j) 8575; (k) 8337; (1) 8302.

PIGEON CEREBELLAR CORTICAL EFFERENTS

229

r Vlll

Vlb

V

1x0

Vlll

I"

1x0

IXb I

IXh

X

U

?;

Y

& P C

Fig. 6. Six sets (m-r) of serial cross sections through the cerebellar nuclei of the pigeon, illustrating the distribution of corticonuclear labeling (shading) after injections of WGA-HRP into each of the cerebellar IobulesW-IX. Conventions as in Figure 4. Case numbers: (m) 8315; (n) 8608; (a) 8306; (p) 8550; (9) 8611; (r) 84107.

J.J.A. ARENDS AND H.P. ZEIGLER

230

tion within the lobule. For reasons of clarity, the experimental data are presented in two sets, one involving projections primarily upon the cerebellar nuclei, the other projections upon the vestibular nuclei. Table 1 provides a representative sample of cases from the experimental series, noting for each case the injection site and the location of terminal labelling within cerebellar and/or vestibular structures. Where the injection was unilateral, the labeling was invariably ipsilateral. Injections in lobules I-X. Figures 4-7, 8a-d illustrate terminal labeling in the cerebellar nuclei after a series of injections into lobules I through X. The data of two eases with injections into widely separated lobules-case 8546 (lobule 111, Fig. 4c) and case 8550 (lobule VIII, Fig. 6p)illustrate the experimental strategy. The injections, which covered the entire mediolateral extent of the lobule on one side but spilled across the midline, produced terminal labeling throughout both the cerebellar nuclei and the vestibular complex. Anterograde labeling in the medial nucleus produced by the lobule I11 injection was restricted to rostroventral portions, whereas after injection in lobule VIII, labeling was largely restricted to the caudal parts. This rostrocaudal difference, though present, was less distinct for the lateral nucleus. The injections also labeled axons within the lateral vestibular nucleus (VeL). For the lobule I11 the terminal labeling was dense in both the ventral and dorsal parts of the nucleus, whereas after the lobule VIII injection it was faint and covered only the dorsal part. After the lobule VIII injection, terminal labeling was present throughout the dorsolateral vestibular nucleus (VDL) but was not seen after the lobule I11 injection. In both cases, labeling was present throughout the descending vestibular nucleus and, to a lesser degree, in the adjacent part of the medial vestibular nucleus. However, the VeD/ VeM projection from lobule I11 was denser than that from lobule VIII . Topographic organization of corticonuclear projections. A subset of the data in Figures 4-7 was derived from cases with small injections differing in their mediolateral location and extent within a given lobule. In contrast to the larger injections, some of these cases produced labeling restricted either t o the medial (Figs. 4a,b, 5i) or the lateral (Figs. 4e, n

n

Fig. 7. Three sets (s-u) of serial cross sections through the cerebellar nuclei of the pigeon, illustrating the distribution of corticonuclear labeling (shading) after injections of WGA-HRP into each of the cerebellar lobules IX-X. Conventions as in Figure 4. Case numbers: (s) 8606; (t)8420; (u) 8351.

Fig. 8. Photomicrographs of representative cross sections through selected levels of cerebellar andior vestibular nuclei, to illustrate anterograde labeling after injections into various cortical lobules. (a) Case 8336: Label in the ventromedial internal subdivision of CbM after an injection into lobules I and 11, medial. (b)Case 8306: Label in internal and intercalate subdivisions of CbM after an injection in lobules V-VII. ( c ) Case 8550: Label in the intermediate subdivision of CbM after an injection in lobule VIII. (d)Case 8306: Label in ventral CbL after an injection in lobule V-VII. ( e ) Case 8645: Label in the infracerebellar nucleus after an injection in the auricle. (f Case 8554: Label in CbL, infracerebellar nucleus and various vestibular nuclei after an injection in the lateral portion of lobule IXb. Asterisk indicates the caudal pole of the tangential vestibular nucleus. Note the juxtarestiform body funneling between the laminar and angular cochlear nuclei. Arrow indicates fiber and terminal labeling in the prehypoglossal region. The brightness of portions of the ventricular border is an artifact. (g) Case 8483: Label in rostral PCVafter an injection in lobule I W , medial. (h) Case 8351: Label in intermediate PCV after an injection in lobule IXb and X, medial. Note that at this level CbL is completely outlined by the PCV label. (i) Case 8554: Label in caudal PCV after an injection in the lateral portion of lobule IXb. Note that at this level CbL is again surrounded by the label in PCV, but now termination is present within the lateral margin of the nucleus. Arrowheads indicate retrogradely labeled cells in PCV. The brightness of the ventricular lining is an artifact. Scale bars = 500 ym, except c and f: 1mm.

Figure 8

Fig. 9. Photomicrographs of representative cross sections through selected levels of vestibular and/or cerebellar nuclei to illustrate anterograde labeling after injections into various cortical lobules. (a) Labelingin VeLd after an injection in lobule 111(case 8546). Asterisks in (a+) indicate the lateral recess of the fourth ventricle. (b) Labeling in the infracerebellar nucleus and in VDL after an injection (top) in the lateral unfoliated cortex (case 8652). ( c )Labeling in M L and CbL after an injection in lobule VIII (case 8550). Arrows in (b) and (c) indicate an area labeled by the lateral (b) but not by the more medial (c) injection. (d)Label in VeLv after an injection in lobule 111 (case 8546). Arrows

indicate corticovestibular fibers descending lateral to VeS. (e 1 Labeling in central and medial portions of VeS after an injection in lateral lobule IXb, extending into the auricle (case 8554). Note the retrogradely labeled cells in the lateral portion of VeS. The brightness of the ventricular border of VeM is an artifact. ( f ) Labeling in the medial tangential nucleus following an injection in the auricle (case 8645). (g) light label in central and lateral portion of VeS after an injection in medial X (case 8483). Asterisks in (el and (g) mark the same (medial VeS) area. In all panels, the midline is to the left. Scale bars = 500 pm.

PIGEON CEREBELLAR CORTICAL EFFERENTS 5g,l, 6n,q) cerebellar nucleus. In all these cases, regardless of the lobule involved, terminal labeling restricted to CbM was seen only after injections in a region adjacent to the midline, whereas terminal labeling restricted to CbL was seen only in cases with injections lateral to that region. Figures 4-7 illustrate the shifts in the location of terminal labeling within the cerebellar nuclei as the injection site shifts from lobule to lobule. Thus for CbM injections restricted to lobules 1-11 (Fig. 4a,b), the focus of labeling was the medial part of the internal division of the nucleus (CbMin), including its ventromedial portion. After injections in lobule 111-V (Figs. 4c-f, 5gJ) the focus shifted to more lateral parts of CbMin and included rostral portions of CbMim. A further shift of the injection site from lobule V to lobule VI (Figs. 5h-1; 60) produced terminal labeling in the intercalate division (CbMic) as well as in more caudal parts of CbMim. The focus of labeling within CbMim shifted still more caudally as the location of the injection site shifted from lobule VII to lobule IXa (Figs. 6; 7s) and there was increasing involvement of more lateral portions of caudal CbMim. Comparable, though less marked shifts were seen in CbL. Injections in lobules I and I1 did not label CbL, but after an injection in lobule I11 (Fig. 4c) the focus of terminal labeling was located rostrally and dorsally within the nucleus. As the injection site shifted from lobules IV to VI, the labeling shifted in a ventral direction and extended more caudally (Figs. 4d-f, 5 , 60); extending still more laterally and caudally (Figs. 6, 7s) after injections in lobules VII-IXa. However, an injection in lateral lobules V-VIa (case 8676, not illustrated) produced a column of terminal labeling throughout most of the AP extent of CbL. Organization o f corticovestibular projections. Cortical projections to the vestibular nuclear regions may be organized in three main groups, arising, respectively, from (1) lobules I-IXa, (2) lobules IXb and X- the vestibulocerebellum- and (3) the auricle and adjacent lateral unfoliated cortex (Table 1;Figs. 8,9, 10.) The injections of group 1 may be subdivided into those lying medial or lateral to cortical regions projecting to the lateral cerebellar nucleus. Injections in the medial (corticovestibular) region of the anterior lobe (lobules I-VIa) produced terminal labeling in both dorsal and ventral portions of the lateral vestibular nucleus (Deiters: VeLd, VeLv; Fig. 9a,d). After comparable injections in the posterior lobe (lobules VII-X), terminal label was seen in the dorsolateral vestibular nucleus (VDL; Fig. 9c). Injections covering the proximolateral portion of the lobules of the anterior lobe produced dense terminal labeling in the border region between the superior and medial vestibular nuclei, which extended caudally into the descending vestibular nucleus. Massive labeling was also seen in the dorsal portion of the lateral vestibular nucleus, and some label was present in PCV. The results of an injection in the lateral portion of lobule I11 are shown in Figure 10. After injections into the proximolateral portions of lobules of the posterior lobe (often involving the lateral unfoliated cortex) terminal labeling was present in VDL (Fig. 9B). Injections into lobules ZXb-X. After injections into these regions, corticonuclear label, if present, was limited to CbL. When the injection was restricted to the medial portion of lobule IXb (Fig. 7t: case 84201, terminal labeling was found predominantly in the dorsal portion of PCV and, more caudally, between CbL and the IVth ventricle. After a larger injection into the medial portions of lobules IXb and

233 X (case 8351; Fig. 7u),the bulk of the terminal labeling was found diffusely throughout PCV, extending into the lateral part of the subjacent superior vestibular nucleus (VeS). There was a lesser projection to the descending vestibular nucleus, which extended medially and medioventrally into the medial vestibular nucleus and the subjacent prehypoglossal region. No labeling was found either in the lateral or the tangential vestibular nuclei. An injection covering the lateral part of lobule IXb and extending into the caudal part of the auricle (case 8554; Fig. 11)labelled PCV as well as a medial subdivision of VeS. Caudolateral PCV contained numerous labeled fibers coursing through the lateral margin of CbL, where indications of termination were seen, continuing through the infracerebellar nucleus to terminate in VDL and more caudoventrally in VeD, VeM, and the prehypoglossal region (see Fig. 8f). There was also a distinct area of termination within medial parts of the tangential nucleus. Znjections into the auricle and the unfoliated cortex. After an injection into the auricle (Fig. 12; case 8338), substantial labeling was seen in the infracerebellar nucleus and in the medial tangential nucleus, extending caudally into the descending vestibular nucleus. Label was also present in medial VeS. Labeled fiber fragments were present in VDL, VeM, VeD, and the prehypoglossal region, but no evidence of terminal labeling was seen in PCV. An injection of the lateral unfoliated cortex (case 8652) produced terminal labeling in VDL and in the infracerebellar nucleus (Fig. 9b) as well as in the medial vestibular nucleus, with labeled fiber fragments evident in VeD. In several additional cases, the injection sites straddled the border between the unfoliated and adjacent proximal foliated cortex. In the case with the most caudal injection (8614: lobule VII), terminal label was virtually restricted to VDL, with only labeled fiber fragments in CbL and VeD. Injections covering lateral portions of the anterior lobe (cf. 8672: Fig. 10) produced massive terminal label in VeLd and some in VDL. A somewhat larger injection covering lobules VI and VII (case 8670), produced substantial labeling in VeLd, some label in CbL and VDL, and labeled fiber fragments in VeD. Photomicrographs of terminal labeling in selected target areas are presented in Figures 8 and 9. Retrograde labeling within the cerebellar and vestibular nuclei aRer injections in the cerebellar cortex. In addition to the terminal labeling described above, cortical injections produced retrograde labeling of neurons both within the cerebellar nuclei and the vestibular complex. Within the cerebellar nuclei, retrograde labeling tended to be located within the area of terminal labeling. Contralatera1 labeling was very rare. Within the vestibular complex, the coincidence between anterograde (corticovestibular) and retrograde labeling was seen only for PCV and VeD and VeS. For VeD, the overlap was limited, with many retrogradely labeled cells lying outside areas of corticovestibular labeling. In VeS, relatively few labeled cells were found

Fig. 10 (see page 234). Brightfield photomicrographs of equidistant serial cross sections through the vestibular nuclei of the pigeon (case 8672) after an injection of WGA-HRP into lobule I11 [inset in (a)] illustrating (1)the cytoarchitecture of the vestibular nuclei and (2) the corticovestibular projections. Arrows in (a)and (h) indicate, respectively, the most rostral and caudal locations of anterograde label. Asterisks in (e) through (h) mark the ventrolateral division of the medial vestibular nucleus. Scale bar = 500 ( ~ m .

Figure 10

PIGEON CEREBELLAR CORTICAL EFFERENTS

235

e

b

IV

I

X

Fig. 11. Equidistant serial cross sections through the cerebellar and vestibular nuclei of the pigeon, after an injection of WGA-HRP into lobule IXb illustrating corticonuclearand corticovestibularprojections (shaded). The injection site (bottom) is illustrated in a representative cross section and ifi a standardized midsaggital projection (case 8554).

within the termination area after an injection in lobule IXb (case 8554), but many more were labeled after an injection covering the auricle (case 8645). Retrogradely labeled cells in VeS were also found in the contralateral homotopic region, after injections in either lobule IXb or the auricle. No retrograde label was seen in VeL or VDL, the infracerebellar nucleus or the tangential nucleus after cortical injections.

Retrograde tracing experiments Injections into the cerebellar and vestibular nuclei. Injections covering the medial cerebellar nuclear group (Figs. 13A, 16a) invariably produced a longitudinal zone of Purkinje cell labeling within the medial portion of the cortex, extending throughout all the lobules with the exception of the distal portion of the lingula. Figure 13B

shows the results of an injection largely limited to the intercalate subdivision of the medial cerebellar nuclear group, with minor involvement of the caudally adjacent medial portion of CbL. Labeling coincided with the lateral part of the zone labeled by the injection shown in Figure 13A but extended only from lobules IV through VIII, with substantial label only in lobule VI. Figures 13C, 16b, and 17a show the labeling produced by an injection covering both the lateral cerebellar nucleus and the bulk of the subjacent vestibular nuclei. The Purkinje cell label, seen throughout the entire lateral half of the cortex, was complementary to the cortical labeling produced by an injection in the medial cerebellar nucleus (Fig. 13A), with the exception of the vestibulocerebellar lobules IXb-X, where Purkinje cell labeling was observed throughout the entire width of the lobules. An injection centered on

236

J.J.A. ARENDS AND H.P. ZEIGLER

I

I

B

I

I

I

I

I

I

I

a

I b

C

d

1

I

I

I

I

I

I

I

I

Figure 13

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I Figure 14

I

I

I

I

I

I

I

PIGEON CEREBELLAR CORTICAL EFFERENTS

239

Fig. 15. Section keys indicating the location of the sections depicted in Figures 13 and 14 upon individually reconstructed cerebellar midsagittal projections.

the dorsal part of VeL, extending into VDL and the overlying infracerebellar and lateral cerebellar nuclei (Fig. 13D, see also 16d), labeled a narrow but prominent strip of Purkinje cells immediately lateral to the zone labeled in Figure 13A. It is most conspicuous within the anterior lobe, barely evident in lobules VI and VII, and becomes visible again in lobules VIII through IXa, continuing well into the vestibulo-cerebellum.In this latter area this zone is somewhat obscured by the surrounding label. In this and comparable cases, many labeled Purkinje cells were also seen in the most lateral portion of the foliated cortex, in the auricle, and in the adjacent unfoliated cortex. In contrast to the massive labeling seen in the preceding case (Fig. 13C), many cells in this lateralmost cortical region remained unlabeled. In the anterior lobules, labeled Purkinje cells

Fig. 13 (see page 237). Distribution of retrogradely labeled Purkinje cells (black dots) in the cerebellar cortex of four cases (A-D) with WGA-HRP injections (shaded areas) into cerebellar and vestibular nuclei. The midline is indicated with vertical marks above and below each section. The extent of the injection sites was estimated from TMB-processed material. For each case the series of sections is arranged from caudal to rostral (left to right). A: Case 8540:Injection in CbM. B: Case 8446:Injection in CbMic. C: Case 8552:Injection in CbL and vestibular nuclear complex. D: Case 8484:Injection centered on VeLd. Fig. 14 (opposite). Distribution of retrogradely labeled Purkinje cells (black dots) in the cerebellar cortex of four cases (E-€I) with WGA-HRP injections (shaded areas) into cerebellar and vestibular nuclei. Conventions as in Figure 13.E: Case 8445:Injection in CbL and the infracerebellar nucleus. F: Case 8433:Injection in VeD and Ta. G: Case 8724:Injection in VeS. H: Case 8605:Injection centered on the ventrornedial vestibular nuclear complex.

were seen in a region lying between the two labeled zones described above, a region retrogradely labeled after CbL injections (Fig. 13Df). After an injection into the lateral cerebellar nucleus, including the infracerebellar nucleus but with minimal vestibular involvement (Figs. 14E, 16c,f), Purkinje cell labeling was seen immediately lateral to that produced by the injection in VeLd (Fig. 13D) but did not extend into the lingula. Labeling in the auricle and the lateral unfoliated cortex, extending caudally into the lateral vestibulocerebellum, consisted of two to three strips with intervening unlabeled areas (Figs. 14Ed, 17b). The ventralmost unlabeled region was particularly prominent immediately lateral to the lateral recess of the IVth ventricle. Labeling was produced in this same region by an injection covering the tangential vestibular nucleus (Fig. 14F, see also 17c). An injection centered on the superior vestibular nucleus (Fig. 14G) produced a pattern of labeled Purkinje cells similar to that seen after an injection in tangential and adjoining descending vestibular nuclei (Fig. 14F). The injection labeled two strips of Purkinje cells across the auricle and lateral unfoliated cortex in a pattern complementary to that seen after an infracerebellar injection. However, in the VeS case, labeling of the lateralmost foliated cortex of the anterior lobe was only a fraction of that seen after injections in the tangentialidescending vestibular nuclei (Fig. 14F).The labeling seen after the injection in the infracerebelhdlateral cerebellar nuclei (Fig. 14E) was even less than that seen in the present case (Fig. 14G). After a VeS injection, labeling was also found throughout the vestibulocerebellum, but was more discontinuous in its medial division (Fig. 14Gc,d). After injections covering the rostral portion of VeD, medial Ta, and the lateral vestibular nucleus at the junction

240

J.J.A. ARENDS AND H.P. ZEIGLER

Fig. 16. Brightfield photomicrographs of representative sections through the anterior lobe of the cerebellum, illustrating retrogradely labelled Purkinje cell zones in the cerebellar cortex of several of the cases shown in Figures 13 and 14. Solid line = midline; pairs of interrupted lines in each panel outline the labelled zones. (a)Case 8540, A zone. (b)Case 8552, B, C, and E zones. ( c )Case 8445, C zone. (d)Case

8603, B zone. (e) Case 8605, E zone. (f)Case 8445, C zone at the level of the lateral unfoliated cortex. Arrowheads indicate labeled fragments of F zone caused by spread of the injection into the infracerebellar nucleus. Arrows indicate the boundaries of C zone Purlunje cell axons. Scale bar = 1mm.

of its ventral and dorsal portions (Figs. 14H, 16e), subsequent labeling was largely restricted to the lateralmost foliated cortex of anterior lobules 11-VI, with some labeling of the auricle and lateral unfoliated cortex. Labeling of more medial cortex was found primarily in the vestibulocerebellum.

DISCUSSION The cerebellar cortex in amniotes projects upon both the cerebellar and the vestibular nuclear complexes (e.g., Bangma et al.,'83; Bangma and Ten Donkelaar, '84). In mammals, the corticonuclear (and corticovestibular) projec-

PIGEON CEREBELLAR CORTICAL EFFERENTS

241

Fig. 17. Brightfield photomicrographs of representative sections through the lateral unfoliated cortex of the cerebellum, illustrating retrogradely labeled Purkinje cell zones in the cerebellar cortex. (a) Case 8552; The injection in CbL and the subjacent vestibular nuclear complex labels virtually all cells in the lateral unfoliated cortex. (b)Case

8815; An injection in CbL and the infracerebellar nucleus labels lateral unfoliated cortex exclusive of two strips (arrowheads). ( c ) Case 8433; Strip of labeled Purkinje cells in the rostral auricle after an injection in Ta. Scale bars = (a, b) 1mm; (c) 500 pm.

tions exhibit both a rostral-to-caudal and medial-to-lateral organization, which is hodologically related to that of the cortex. The most medial (fastigial) nucleus receives its input mainly from medial (vermal) cortex, whereas projections upon the interposed and lateral (dentate) nuclei arise from progressively more lateral cortical regions (Brodal, '81; Haines et al., '82). These observations were consistent with a conceptualization of the cerebellar cortex as organized in longitudinal zones, with each cortical zone projecting upon a specific cerebellar or vestibular nucleus (Jansen and Brodal, '40).Using myeloarchitectural criteria, Voogd ('64) identified four major cortical divisions, excluding the vestibulocerebellum. These are, from medial to lateral, an A zone, projecting upon the fastigial nucleus, a B zone projecting upon the lateral vestibular nucleus, a C zone projecting upon an interposed nucleus, and a D zone projecting upon the dentate nucleus. The present data confirm and delineate such a mediolateral (zonal) organization for the pigeon cerebellar cortex, relate it to the rostrocaudal organization of the lobules, and support a division of the cerebellar nuclei into a medial (CbM) and a lateral (CbL) nucleus.

medial portions of the cerebellar cortex). In contrast, the location of terminal labeling within VeD and VeM did not appear to vary with the injection site. The organization of projections from lobules IXb-X and the auricle is somewhat anomalous by comparison with that of other cortical regions. Those medial cortical regions, which in other lobules would project to the medial cerebellar nucleus, proved to project to the cerebellovestibular process, whereas those that in other lobules project upon CbL, here project upon both CbL and PCV. All such cortical injections produced labeling throughout PCV, but as the injection sites shifted mediolaterally, there was a corresponding shift in the focus of the projection from areas abutting CbM to those bordering CbL. These observations account for the presence of retrogradely labeled Purkinje cells throughout the width of lobules IXb-X after either CbM or CbL injections. Moreover, the observed pattern of cortical afferents suggest that PCV may function, hodologically, with respect to lobules IXb and X (the vestibulocerebellum) as do the cerebellar nuclei for lobules I-IXa. The location, cytoarchitecture, and hodology of PCV are reminiscent of the (ventral) parvocellular portions of the mammalian cerebellar nuclei. In mammals, input to these parvocellular regions is dominated by either (1) primary and secondary vestibulocerebellar projections, or (2) corticonuclear projections from cerebellar cortex in receipt of vestibular input, primarily the vestibulocerebellum (Haines, '77). The pigeon's PCV also appears to receive both primary and secondary vestibulo-cerebellar fibers (Arends, unpublished observations; Schwarz and Wood, '77; Correia et al., '83; Eden, pers. comm; Schwarz and Schwarz, '86) as well as cortical efferents from the vestibulo-cerebellum. The efferent connections of PCV include large portions of the

Organization of the corticonuclear and corticovestibular projections For the corticonuclear projection from lobules 11-IXa, there is demonstrable topographic organization in both the mediolateral and rostrocaudal planes. For the corticovestibular projection from these lobules, there is a systematic relation between the location of the injection site and the presence and/or location of terminal labeling in the lateral (VeL) and dorsolateral (VDL) vestibular nuclei (the only vestibular nuclei receiving inputs from both the lateral and

242

J.J.A. ARENDS AND H.P. ZEIGLER

Fig. 18. Diagrammatic representation of cerebellar cortical zones and their efferent nuclear projections in the pigeon. The cortical zones

(A,B,C,E,F: see central figure) and their efferent targets are differentially shaded (see abbreviations list).

vestibular nuclear complex (Arends and Zeigler, ’911, but not the extraocular motor nuclei.’ The corticovestibular projection of lobules IXb and X, excluding the auricle, is focused largely upon VeS. As compared with the (mediolateral) topography of the projection upon PCV, a reversed (lateral-medial) topography is evident in the projection to VeS. Although the auricle is contiguous with the lateral portions of lobules IXb and X, its projections are restricted to the infracerebellar nucleus, the medialkentral VeS, and a medial portion of the tangential nucleus.

contiguous target nuclei, and the presence in one target nucleus of projecting fibers en route to another. Nevertheless, careful comparison of the anterograde data with the pattern of retrograde labeling confirms the existence of at least four longitudinal cortical zones. These include medial and lateral areas, labeled after injections into the medial and lateral cerebellar nuclei, respectively, and a cortical area lying between the medial and lateral zones, which was labeled by injections in dorsolateral or lateral vestibular nuclei. The lateralmost cortex, including the auricle, its surrounding lateral unfoliated cortex, and the proximolateral parts of most lobular cortex, was retrogradely labeled after vestibular injections. The existence of additional zones or subzones may be inferred from the pattern of labeling. A subdivision of the medialmost zone is suggested by the observation that its lateral portion may be retrogradely labeled without simultaneously labeling the more medial portion. Such a pattern of labeling is seen after an injection abutting on the ventricle and covering a sizeable portion of the intercalate subnucleus of CbM (Fig. 13B). This subzone is widest in lobule VI, tapering off symmetrically in lobules V andVII with a only a few cells labeled in IV and VIII. A comparable subdivision of the lateralmost cortex into “auricular/central lateral unfoli-

Zonal organization of the cerebellar cortex The accurate delineation of cerebellar cortical zones is confounded by a number of methodological problems, including the apparent interdigitation of the zones, the difficulty of restricting injections to a single one of a series of ’It has been reported that (caudolventral parvocellular portions of the mammalian dentate nucleus project upon oculomotor regions (Haines, ’77). These reports are based on anterograde experiments, which may have compromised other preoculomotor structures, e.g., dorsal cell group Y or the infracerebellar nucleus. The possible homologues of the avian infracerebellar nucleus and dorsolateral vestibular nucleus (VDL) are discussed in a subsequent work (Arends et al., ’91).

PIGEON CEREBELLAR CORTICAL EFFERENTS ated cortex" and "peripheral lateral unfoliatedJproximolateral lobular cortex" is indicated by the anterograde label seen after an injection restricted to the auricle. For the auricle and the immediately surrounding unfoliated cortex, a unique set of three different but partially complementary patterns of Purkinje cell labeling may be distinguished. Each pattern is related to a specific nucleus (infracerebellar, tangential, superior vestibular) and consists of dorsoventrally complementary strips running in a rostrocaudal direction. Together, the strips cover the auricle and immediately adjacent unfoliated cortex, whereas the rostralmost part of the unfoliated cortex abutting the anterior lobules 11-VI contains fewer labeled Purkinje cells. The difference between the three patterns is clearest at the level of the lateral recess of the IVth ventricle (Figs. 13Dd, 14Fd,Ge).

Nucleocortical connections Although no systematic analysis of these connections was carried out, it is clear that nucleocortical connections in the pigeon are largely ipsilateral and reciprocal. A small number of nuclear cells maintain ipsilateral nonreciprocal connections, whereas contralateral projections are very scarce.

Comparative considerations The cerebellar cortex of nonmammalian vertebrates is characterized by a mediolaterally organized pattern of longitudinal cortical zones, projecting, alternately, upon cerebellar and vestibular nuclei: amphibians (Grover, '83), reptiles (Bangma, '83; Bangma et al., '83; Bangma and Ten Donkelaar, '84) and birds (Wood, '79; Feirabend, '83; Feirabend and Voogd, '86; Arends, '85; Arends and Voogd, '89). The relative stability of this organization across classes suggests that it reflects a basic vertebrate scheme. The cerebellar cortex of the pigeon shows both an elaboration of the basic vertebrate pattern and many comrnonalities with the organization observed in mammals. However, in contrast to mammals, the pigeon has only three main cortical zones if the auricle and the adjacent lateral cortex are excluded. In both classes the two medial zones project to the medial cerebellar and lateral vestibular nuclei, respectively. Thus the connections of the pigeon's most medial zone correspond with those of the A1 zone of Voogd ('69) and the connections of the laterally adjacent region corresponds to those of his B zone. (The pigeon lacks a region equivaient to the mammalian A2 zone, projecting to vestibular targets other than the dorsal part of the lateral vestibular nucleus, i.e., to the magnocellular portion of the medial vestibular nucleus, VeMmc.) Since theVeMmc in mammals is also referred to as the ventral portion of the lateral vestibular nucleus, or part of it, the situation in birds and mammals may, in fact, be similar. In mammals no posterior lobe B zone has been defined, although projections from lobules VIII-X to the lateral vestibular nucleus have been observed (see, e.g., Walberg and Jansen, '61; van Rossum, '69; Henkel and Martin, '77; Bigare, '80; Balaban, '84; Klinkhachorn et al., '84; Umetani et al., '86; Bernard, '89; Umetani and Tabuchi, '88; Tabuchi et al., '89). In the pigeon, as in mammals, few corticovestibular projections arise from the medial portions of lobules VI and VII. In birds, the third (C) zone projects to the lateral cerebellar nucleus and, in contrast to mammals, does not seem to be further subdivided. The cortex lateral to the C zone projects to a variety of vestibular nuclei and, based upon its efferent projections, may be further subdivided.

243 We propose to label these cortical areas E and F zone, respectively (see below). The E zone includes the proximolateral lobular cortex, which projects upon VeLdNDL and an area encompassing portions of VeS, VeM, and VeD (Arends and Voogd, '89; Arends et al., '91). The F zone includes the auricular and adjacent unfoliated cortex. Its vestibular targets appear to be exclusively preoculomotor (Arends et al., '91), suggesting that these cortical regions correspond to the mammalian flocculus and ventral paraflocculus. In mammals, the compartmentalization of the cerebellar white matter parallels the (olivocerebellar) afferent and (corticonuclear) efferent organization (Voogd, '64, '69; Voogd and Bigare, '80). Feirabend and Voogd ('86) have distinguished a number of fiber compartments in the chicken cerebellum and associated several of them with specific cortical zones, but no such data are available for the pigeon. Figure 18 presents a schematic diagram illustrating the zonal organization of the cerebellar cortex in the pigeon. On the basis of comparative and functional considerations, Bangma and Ten Donkelaar ('84) have proposed that the ground plan of vertebrate cerebellar organization involves a medial and a lateral cerebellar nucleus corresponding, respectively, to the fastigial and interposed nuclei of mammals. The pigeon's medial cerebellar nucleus corresponds in its connections to the fastigial nucleus of mammals. Nothing in the present data provides criteria that would allow us to identify the pigeon's lateral nucleus, unambiguously, with either the interposed or dentate nucleus of mammals. However, a study of cerebellar nuclear efferents in the pigeon (Arends and Zeigler, '91b) indicates that the projections of the pigeon's lateral cerebellar nucleus closely resemble those of the mammalian (anterior) interposed nucleus. We therefore suggest that the pigeon's cerebellar cortex lacks a D zone and that the cortical zone projecting to the lateral nucleus in the pigeon is homologous to all or part of the C zone of mammals.

ACKNOWLEDGMENTS This work and the two companion studies (Arends and Zeigler, '91; Arends et al., '91) were supported by NSF Grant BNS 85-07374, NIMH Grant MH-08366, and Research Scientist Award MH-00320, and by the Biopsychology Program, Hunter College, CUNY. We thank Dr. R.L.M. Faull (Department of Anatomy, University of Auckland Medical School, Auckland, New Zealand) for his help at an early stage of the project, and Dr. John Cabot (SUNY, Stonybrook, NY) for making available cases with HRP injections of the cervical spinal cord. We are indebted to Dr. J.M. Wild and Professor J.L. Dubbeldam for comments on earlier drafts. The technical contributions of Mrs. Agnes Reilly are gratefully acknowledged.

LITERATURE CITED Adams, J.C. (1980) Stabilizing and rapid thionin staining of TMB-based HRP reaction product. Neurosci. Lett. 17:7-9. Adams, J.C. (1981) Heavy metal intensification of DAB-based HRP reaction product. J. Histochem. Cytochem. 29:775. Arends, J.J.A. (1985) Organization of corticonuclear and olivocerebellar connections in the pigeon. SOCNeurosci. Abstr. 111692. Arends, J.J.A., and J. Voogd (1989) Topographical aspects of the olivocerebellar system in the pigeon. Exp. Brain Res. (Suppl.) 17:52-57. Arends, J.J.A., and H.P. Zeigler (1989) Cerebellar connections of the trigeminal system in the pigeon. Brain Res. 487:69-78.

244 Arends, J.J.A., and H.P. Zeigler (1991) Organization of the cerebellum in the pigeon (Colurnba liuia):11. Projections of the cerebellar nuclei. J. Comp. Neurol. 306:245-272. Arends, J.J.A., R.W. Allan, and H.P. Zeigler (1991) Organization of the cerebellum in the pigeon (Columba liuia):111. Corticovestibular connections with eye and neck premotor areas. J. Comp. Neurol. 306:273-289. Arends, J.J.A., A. Woelders-Blok, and J.L. Dubbeldam (1984) The efferent connections of the nuclei of the descending trigeminal tract in the mallard (Anusplatyrhynchos L.).Neuroscience 13~797-817. Balaban, C.D. (1984) Olivo-vestibular and cerebello-vestibular connections in albino rabbits. Neuroscience 12:129-149. Bangma, G.C. (1983) Cerebellar connections in some reptiles. Ph.D. thesis, Nijmegen, pp. 1-148. Bangma, G.C., and H.J. ten Donkelaar (1984) Cerebellar efferents in the lizard Varanus exanthematicus I. Corticonuclear projections. J. Comp. Neurol. 225t447459. Bangma, G.C., H.J. ten Donkelaar, and A. Pellegrino (1983) Cerebellar corticonuclear projections in the red-eared turtle Pseudemys scripta elegans. J. Comp. Neurol. 215258-274. Bermejo, R., and H.P. Zeigler (1989) Prehension in the pigeon. 11. Kinematic analysis. Exp. Brain Res. 75t577-585. Bermejo, R., R.W. Allan, J.D. Deich, D. Houben, and H.P. Zeigler (1989) Prehension in the pigeon I: Descriptive analysis. Exp. Brain Res. 75:569-576. Bernard, J-F. (1987) Topographical organization of olivocerebellar and corticonuclear connections in the rat-A WGA-HRP study: I. Lobules IX, X and the flocculus. J. Comp. Neurol. 263:241-258. Bigare, F. (1980) De efferente verbindingen van de cerebellaire schors van de kat. (Een onderzoek naar de corticonucleaire en corticovestibulaire projekties door middel van retrograad axonaal transport van mierikswortel peroxidase (HRP)). Ph.D. thesis, Leiden. Brodal, A. (1981) Neurological Anatomy in Relation to Clinical Medicine. The Cerebellum. Oxford Oxford University Press, pp. 294-393. Brodal, A., K. Kristiansen, and J. Jansen (1950) Experimental demonstration of apontine homologue in birds. J. Comp. Neurol. 9223-69. Clarke, P.G.H. (1977) Some visual and other connections to the cerebellum ofthe pigeon. J. Comp. Neurol. 174532-552. Correia, M.J., A.R. Eden, K.N. Westlund, and J.D. Coulter (1983) A study of some of the ascending and descending vestibular pathways in the pigeon (Colurnba liuia)using anterograde transneuronal autoradiography. Brain Res. 278t53-61. Feirabend, H.K.P. (1983) Anatomy and development of longitudinal patterns in the architecture of the cerebellum of the White Leghorn (Gallus domesticus). Ph.D. thesis, Leiden. Feirabend, H.K.P., and J. Voogd (1986) Myeloarchitecture of the cerebellum of the chicken (Gullus domesticus): An atlas of the compartmental subdivision of the cerebellar white matter. J. Comp. Neurol. 251:44-66. Gross, N.B. (1970) Sensory representation within the cerebellum of the pigeon. Brain Res. 21:280-283. Grover, B.G. (1983)Topographic organization of cerebellar efferents in the frog (Rana esculenta) as revealed by retrograde transport of wheat germ agglutinin conjugated horseradish peroxidase. Neurosci. Lett. (Suppl.) 14:S146. Haines, D.E. (1977) A proposed functional significance of parvicellular regions of the lateral and medial cerebellar nuclei. Brain, Behav. Evol. 14:328-340. Haines, D.E., G.W. Patrick, and P. Satrulee (1982) Organization of cerebellar corticonuclear fiber systems. In S.L. Palay and V. Chan-Palay (eds): The Cerebellum, New Vistas. Exp. Brain Res. (Suppl.), 6t320-371. Henkel, C.K., and G.F. Martin (1977) The vestibular complex of the American opossum Didelphis uirginiana. 11. Afferent and efferent connections. J. Comp. Neurol. 172t321-348. Jansen, J., and A. Brodal(1940) Experimental studies on the intrinsic fibers

J.J.A. ARENDS AND H.P. ZEIGLER of the cerebellum. I1 The corticonuclear projection. J. Comp. Neurol. 73267-321. Karten, H., and T.E. Finger (1976) A direct thalamo-cerebellar pathway in pigeon and catfish. Brain Res. 102335-338, Karten, H.J., and W. Hodos (1967) A Stereotaxic Atlas of the Brain of the Pigeon (Colurnba liuia). Baltimore: Johns Hopkins University Press. Klein, B.G., J.R. Deich, and H.P. Zeigler (1985) Grasping in the pigeon: Final common path mechanisms. Behav. Brain Res. 18:201-213. Klinkhachorn, P.S., D.E. Haines, and J.L. Culbertson (1984) Cerebellar cortical efferent fibers in the North American opossum Didelphis uirginiana. 11. The posterior vermis. J. Comp. Neurol. 227:439-451. Larsell, 0. (1948) The development and subdivisions of the cerebellum of birds. J. Comp. Neurol. 89:123-190. Larsell, 0. (1967) The Comparative Anatomy and Histology of the Cerebellum from Myxinoids through Birds. Minneapolis: Univ. of Minnesota Press. Llinas, R., and J.I. Simpson (1981) Cerebellar Control of Movement. In A. Towe and E.S. Luschei (eds): Handbook of Behavioral Neurobiology: Motor Coordination. New York: Plenum Press, pp. 231-321. Mesulam, M.M. (1978) Tetramethyl benzidine for horseradish peroxidase neurochemistry. A non-carcinogenic blue reaction product with superior sensitivity for visualizing neural afferents and efferents. J. Histochem. Cytochem. 26: 106-11 7. Pearson, R. (1972) The Avian Brain. London: Academic Press. Schwarz, D.W.F., and I.E. Schwarz (1986) Projections of afferents from individual vestibular sense organs to the vestibular nuclei in the pigeon. Acta Otolaryngol. 102r463-473. Schwarz, D.W.F., and S. Wood (1977) Microgeometry ofcerebellar corticonuclear projection in the pigeon. SOC. Neurosci. Abstr. 3t182. Schwarz, I.E., and D.W.F. Schwarz (1983) The primary vestibular projections to the cerebellar cortex in the pigeon (Columba liusa). J. Comp. Neurol. 216:438444. Tabuchi, T., T. Umetani, and T. Yamadori (1989) Corticonuclear and corticovestibular projections from the uvula in the albino rat: Differential projections from the sublobuli of the uvula. Brain Res. 492176-186. Umetani, T., and T. Tabuchi (1988) Topographic organization of the corticonuclear and corticovestibular projections from the pyramis and copula pyramidis in the albino rat: An autoradiographic orthograde tracing study. Brain Behav Evol. 32:160-168. Umetani, T., T. Tabuchi, and R. Ichimura (1986) Cerebellar corticonuclear and corticovestibular fibers from the posterior lobe of the albino rat, with comments on zones. Brain Behav. Evol. 29t54-67. van Rossum, J. (1969) Corticonuclear and corticovestibular projections of the cerebellum. Ph.D. thesis, Leiden; van Gorcum, Assen. Voogd, J. (1969) The importance of fibre connections in the comparative anatomy of the mammalian cerebellum. In R. Llinas (ed): Neurobiology of Cerebellar Evolution and Development. Chicago: Am. Med. Assoc., pp. 493-514. Voogd, J., and F. Bigare (1980) The topographic distribution of olivary and corticonuclear fibers in the cerebellum. A review. In J. Courville (ed): The Inferior Olivary Nucleus, Chap. 10. New York: Raven Press. Walberg, F., and J. Jansen (1961) Cerebellar corticovestibular fibres in the cat. Exp. Neurol. 3t32-52. Whitlock, D.G. (1952) A neurohistological and neurophysiological study of afferent fiber tracts and receptive areas of the avian cerebellum. J. Comp. Neurol. 97:567-623. Wold, J.E. (1981) The vestibular nuclei in the domestic hen (Gallus domesticus): VI. M e r e n t s from the cerebellum. J. Comp. Neurol. 2Ok319-34 1. Wood, S. (1979) Microgeometry of cerebellar corticonuclear projection in the pigeon. Master’s thesis, University of Toronto.