The Vestibular Complex of the American Opossum Didelphis virginiana 11.

AFFERENT AND EFFERENT CONNECTIONS C. K. HENKEL AND G . F. MARTIN Department of Anatomy, The Ohio State University, College of Medicine, Columbus, Ohio 4321 0

ABSTRACT We have demonstrated the connectivity of the opossum’s vestibular nuclei using degeneration, autoradiographic and horseradish peroxidase techniques and have found it to be generally comparable to that reported for the cat. Apart from the primary input described in Part I of our study, the cerebellum provides the major source of afferent connections to the vestibular complex. Axons from the cerebellar cortex distribute mainly to vestibular areas which receive no primary afferent projections, e.g., the dorsal part of the lateral vestibular nucleus, the dorsolateral margin of the inferior vestibular nucleus as well as cell groups comparable to “f‘and “x.” In contrast, fastigial fibers show considerable overlap with primary vestibular input, particularly in the ventral part of the lateral nucleus, the central part of the inferior nucleus and the medial nucleus. Axons of fastigial origin also distribute to the superior vestibular nucleus, to subnuclei “f ’and “x” and to the parasolitary region. Although spinal fibers are diffuse within the main vestibular nuclei, they ramify quite densely within subnucleus “x.” Most of the spinovestibular projection appears to arise in the cervical spinal cord of the opossum. Ipsilateral connections from the interstitial nucleus of Cajal and surrounding areas end predominantly, but not exclusively, in the medial vestibular nucleus. A crossed midbrain projection has been verified from the red nucleus to cell group “x” and the lateral part of the inferior nucleus, as well as to an area possibly comparable to cell group “z,” as described for the cat. In Part I of our study we have shown that the major targets of primary vestibular fibers are the central part of the superior nucleus, a portion of the parabrachial complex possibly comparable to subnucleus “y,” the ventral part of the lateral nucleus and the medial nucleus. All of these primary zones give rise to fibers supplying extraocular nuclei and surrounding areas (present study). The ascending midbrain fibers from the superior nucleus end mainly ipsilaterally, whereas those from the putative subnucleus “y” and the medial vestibular nucleus distribute contralaterally for the most part. Although the dorsal part of the lateral vestibular nucleus has no primary vestibular input, it does receive a major projection from the cerebellar cortex. This same region of the lateral nucleus projects to the spinal cord, but not to extraocular nuclei. The ventral part of the lateral nucleus, and perhaps the medial nucleus, also relay to the spinal cord. Additional projections from all vestibular nuclei to the reticular formation provide indirect routes through which the vestibular nuclei can potentially affect multiple systems, including those innervating the spinal cord. Finally, commissural vestibular connections of the opossum are shown to arise within all four major nuclei. J. COMP. NEUR., 172: 321-348.

32 1

322

C. K. HENKEL AND G . F. MARTIN

Although the vestibular nuclei are char- survival times employed many axon termiacterized in part by their input from laby- nals are engulfed by glial cells when rinthine and otolithic receptors, they viewed with the electron microscope, such receive additional connections which can times were the most useful for evaluating modify their activity and subsequently the overall trajectory and distribution of effect many levels of the nervous system. the fibers in question. Due to the interruption of fibers en Extensive anatomical studies of vestibular connectivity have been reported in the cat passage in many of the degeneration prepa(see Brodal and Pompeiano, ’72, for cur- rations, we also employed the autoradiorent reviews) and such studies provide the graphic method (Cowan e t al., ’72). All basis for most clinical and physiological cases were prepared by stereotaxically concepts. However, similarly detailed de- placing 12-40pCi of tritiated leucine (New scriptions are not available for other mam- England Nuclear) in the site to be studied mals, making it difficult to evaluate the using a Hamilton syringe attached to a mipossibility of species differences. As an crodrive system. The animals survived extension of our study on the conformation either 24 hours for fast labelling or 10 to 13 and cytology of the opossum’s vestibular days to augment labelling of slowly transcomplex, we have undertaken an evalua- ported marker. They were sacrificed by intion of its many inputs as well as its projec- tracardiac perfusion of warm saline foltions to other cell groups. To our knowl- lowed by a 3% paraformaldehyde-1% gluedge, this report is the first to provide such taraldehyde fixative. Sections were cut on a freezing microtome and serially mounted data for any marsupial. before being defatted and subsequently MATERIALS AND METHODS dipped in NTB-2 nuclear track emulsion. Over 300 opossum brains are available in After two to six weeks exposure at 4°C the our laboratory with lesions in different sections were developed in D-19 Kodak parts of the brain and spinal cord. The developer and counterstained with cresyl cases relevant to the present study in- violet. In order to visualize the origin of certain cluded those with lesions of the cerebellar cortex, the cerebellar deep nuclei, various projections several cases were prepared by areas of the midbrain, the spinal cord, the the horseradish peroxidase technique reticular formation and the vestibular nu- (LaVail et al., ’73). A 30-60% solution of clei themselves. Most of them have been the maker was injected (20-60 minutes/inused for previous studies and were proc- jection) into the appropriate target using essed by either the Fink-Heimer technique the same microdrive system employed for (‘67) or one of its modifications after sur- the autoradiographic experiments. After vival times from 5 to 14 days. In most in- 20 to 48 hours, the animals were sacrificed stances every fifth section was stained for by intracardiac perfusion of saline followed Nissl substance in order to facilitate identi- by either buffered formalin or a 3% parafication of nuclear divisions. Many such formaldehyde-1% glutaraldehyde fixative. cases, as well as others prepared spe- The brains were removed, blocked and cifically for this study, provided data relaFig. 1 In this drawing degeneration is plotted in tive to opossum vestibular connections and the lateral vestibular nucleus in three separate cases were examined in light of the nuclear with cerebellar cortical lesions. The sections at the organization reported in Part I. Because of reader’s left are in each case from the rostra1 part of the lateral nucleus, and the ones to the reader’s right the relatively long survival times used in are from the caudal part. The cortical lesions are most experiments, our definition of termi- sketched in the box at the middle of each pair of line nal degeneration refers to either frag- drawings. An attempt has been made in this and all mented axons which have left the major subsequent drawings to represent degenerating axons with short lines and stippling similar to their actual pathway and appear thin and randomly appearance. The coarser markings indicate larger deoriented or to finely particulate debris. Al- generating axonal fragments, while the very small though the authors recognize that at the series of dots represent the finer arborizing fibers. ~

CONNECTIONS OF THE OPOSSUM VESTIBULAR COMPLEX

Figure 1

323

324

C. K. HENKEL AND G. F. MARTIN

\ [

TrMo contra-a

TrMo

i

ipsi-a

\

Fastigial lesio

Fig. 2 In this drawing degenerating fibers are plotted as they appear in rostral parts of the vestibular complex following a large fastigial lesion (fig. 4). The contralateral (left) and ipsilateral (right) Vestibular nuclei are shown and the sections of the vestibular complex are arranged rostral to caudal (a-c). The more caudal sections are illustrated in figure 3.

325

CONNECTIONS OF THE OPOSSUM VESTIBULAR COMPLEX

/

\

Cc D

Cc D

vst I

L,f

vs; I con t r a - e

vst I

con r r a - f

contra

i>

ipsiif

-

Fig. 3 Degeneration in caudal parts of the vestibular complex after a large fastigial lesion (fig.4). The contralateral distribution is shown on the reader’s left, and the ipsilateral sections to the right. The line drawings are in rostra1 to caudal sequence (d-g).

326

C. K. HENKEL AND G . F. MARTIN

CONNECTIONS OF THE OPOSSUM VESTIBULAR COMPLEX

kept in fixative at 4°C overnight before being transferred to a 2530% sucrose-0.1 M phosphate buffer (pH-7.4) rinse. Frozen sections were collected in TRIS buffer (0.5 M, pH 7.6) and subsequently incubated at 37" in a TRIS solution of 3-3'-diaminobenzidine tetrahydrochloride and hydrogen peroxide according to the Graham and Karnovsky method ('66). The sections were rinsed, mounted and counterstained with cresyl violet. Each case was then studied to determine both the spread of the enzyme at the injection site and the location of neurons with brown granules of reaction product. Several problems associated with using the autoradiographic and HRP techniques and interpreting the results obtained are discussed in a previous communication (Martin et al., '76b). OBSERVATIONS

Afferent projections to the vestibular complex Vestibular afferent8 from the cerebellar cortex In order to determine the total cerebellar cortical input to the vestibular complex, we first examined cases with nearly complete cerebellar cortical ablation. However, the nature of cerebellar blood supply Fig. 4 Nissl stained section through the fastigial lesion in P 224. The degeneration elicited by this lesion is plotted in figures 2 and 3. The fastigial (Fast) and interpositus (InP) nuclei are labelled. Fig. 5 Nissl stained section through the largest part of the lesion in P 284. The ipsilateral degeneration produced by this lesion is shown in figure 9. The superior colliculus (CS), nucleus of Darkschewitsch (Dk), interstitial nucleus of Cajal (Iflm) and red nucleus (Rb) are labelled. Fig. 6 Nissl stained section through the lesion site of P 24. In this case the lesion impinged upon the medial nucleus and undercut many of the fascicles issuing from it. The medial vestibular (VstM) and facial (Fac) nuclei are labelled. Fig. 7 Nissl stained section through the lesion in P 177 which destroyed almost all of the superior and lateral vestibular nuclei. The motor trigeminal (TrMo), su erior olivary and part of the superior vestibular &stS) nuclei are labelled. Fig. 8 Section through the heaviest 3H leucine labelling of the superior vestibular nucleus (VstS) in P 386. The motor trigeminal nucleus (TrMo) is indicated for reference.

327

makes it difficult to produce such lesions without concomitant damage to the underlying corpus medullare and deep nuclei. Since brains with lesions restricted to one or more cerebellar lobules show little or no undercutting of such structures, we will describe the results obtained from several selected cases which demonstrate best the observed organization of cerebellar cortico-vestibular fibers. Although these cases do not necessarily provide data on input from all lobules, comparison with the results obtained from brains with large lesions suggests that the sum of their input is comparable to the whole. Following superficial lesions of the anterior vermis (shown in figs. la,b) degenerating axons can be traced posteriorly to their major destination within the ipsilatera1 nucleus fastigius and the interpositus nuclei. The degenerating fibers which are destined for the vestibular complex turn anteriorly and either course through the anterior interpositus and lateral nuclei or pass among the lateral cell groups of the nucleus fastigius. Many fragmented axons arch over the lateral ventricular recess and turn back into the vestibular complex via the juxtarestiform body as far rostrally as the superior nucleus. Still others course directly into the lateral vestibular nucleus (fig. la). Some degeneration is present in the contralateral cerebellar nuclei and vestibular complex, but it can be attributed to either direct spread of the lesion across the midline or secondary vascular involvement. Although there is evidence that anterior vermal fibers end in the superior vestibular nucleus, most of them distribute to the lateral nucleus. In cases such as P-279 there is scattered degeneration within almost the entire dorsal extent of the lateral nucleus in rostra1 sections (fig.la) as well as in the area adjacent to the cochlear border more caudally (fig. lb). In P-294 (figs. lc,d) the lesion was placed so as to include the paravermal zone of the anterior lobe and the heaviest degeneration is restricted to the most dorsal cell groups of the lateral nucleus both rostrally and

328

C. K. HENKEL AND C. F. MARTIN

Fig. 9 A series of sections arranged from rostral to caudal, a-d, indicating the extent of ipsilateral midbrain degeneration in representative levels of the medial and inferior vestibular nuclei (P 284). The arrow in frame d indicates degenerating axons, some of which reach the medial and inferior nuclei.

caudally. In both cases a minimal amount of deeper and more lateral position within the degeneration extends into the inferior nu- interposed and fastigial nuclei than those cleus and laterally into the vestibular root. from the anterior lobe. Although most of Axonal debris is commonly found in close the lesions cross the midline resulting in apposition to the somata and proximal bilateral degeneration, the most extensive dendrites of cells of all sizes, as well as in debris is found on the ipsilateral side. Once the surrounding neuropil. degenerating fibers reach the vestibular Several brains are available with lesions complex (primarily in the jwtarestiform of the posterior uermis (figs. le,f) although body or in more rostral fascicles) they turn none show involvement of the flocculonod- caudally and end extensively throughout ular lobe. In each of them the degenerating the inferior as well as the lateral nucleus. axons course rostrally in a somewhat In P-266 only a few degenerating fibers are

CONNECTIONS OF THE OPOSSUM VESTIBULAR COMPLEX

seen laterally within the lateral vestibular nucleus (fig. le), although some terminal debris is present in its medial dendritic fields (Part I). Likewise, in contrast to the degeneration produced by anterior lobe lesions, that present in the caudal part of the lateral nucleus is more centrally dispersed and does not end as rofusely along the cochlear border (fig. I f . Many degenerating axons continue to the inferior nucleus where most of them terminate in a very distinct band along its dorsal and lateral surface and in cell groups “f’ and “x” (fig. 10). Finally, a few fragmented axons course centrally to the caudal end of the inferior nucleus. Although some degeneration is present within the vestibular nuclei after lateral hemisphere damage, problems such as direct paravermal spread of the lesion, fiber undercutting and secondary vascular involvement militate against making positive statements as to the origin of the injured axons. It should be noted, however, that no case contained degeneration beyond the areas described in the above accounts. Vestibular afferents from the fastigial complex Several cases were examined in which electrolytic lesions had been placed in the nucleus fastigius. In one particularly good brain, most of the fastigial nucleus is damaged except for a small medial area of large cells (P-244: fig. 4) producing what we feel is a representative degeneration display. The degenerating axons are plotted in figures 2 and 3 and it can be seen that they are present bilaterally. Although the degenerating fibers in the ipsilateral superior nucleus are more abundant and comprise more distinct fascicles than those on the opposite side, the preterminal debris is about equal bilaterally (fig. 2a). Degenerating fascicles traverse the lateral nucleus, but distribute only modestly to its ventral regions. Virtually none end in its dorsal subdivision (fig. 2: contra-b). In contrast, on the side of the lesion dense terminal degeneration is dispersed throughout the lateral nucleus especially among its

P

329

dorsal cell groups (fig. 2: ipsi-b). The ipsilateral degeneration is so dense in the dorsal part of the lateral nucleus that it completely packs the neuropil. Degenerating axons also end bilaterally in a band adjacent to the ventricular surface in the rostra1 part of the medial nucleus (figs. 2c,d). Although degeneration products are also located in more central portions of the comparable nuclei, they are sparse, particularly contralaterally and caudal to the acoustic striae (compare ipsi and contra in figs. 3d,e). On both sides, degenerating fibers sweep over the nucleus solitarious from the inferior nucleus (fig. 3f) and distribute to the most caudal part of the medial nucleus. As might be expected they are most numerous on the side of the lesion. The degenerating axons that course through the lateral nuclei continue caudally to their inferior counterparts. Although many of them leave the main bundles and distribute diffusely within the inferior nuclei, terminal debris is most extensive in the dorsal, lateral rim of small cells described in Part I. Degeneration in the latter area is dense on both sides and appears to be continuous with that within groups “f’ and “x” caudally (figs. 3e, 11).It should be noted, however, that not all areas of “f’ and “x” contain degeneration and that axonal debris is not present in either subnucleus when the lesion is limited to the caudal part of the nucleus fastigius. After large lesions (e.g., P-224) degeneration is present bilaterally within the caudal part of the inferior nucleus, where it finally distributes within the nucleus parasolitarius (fig. 3g). Since axons of passage are interrupted after fastigial lesions, much of the degeneration present in the ipsilateral vestibular complex is necessarily a result of undercutting fibers from either the cerebellar cortex or the contralateral nucleus fastigius. For that reason we injected 3H-leucineinto the fastigial complex of three animals. Unfortunately, two of the cases are contaminated because of spillover into the underlying medial vestibular nucleus and reticu-

330

C. K. HENKEL AND G . F. MARTIN

CONNECTIONS OF THE OPOSSUM VESTIBULAR COMPLEX

33 1

lar formation and the other is not com- pons and inferior olive), no evidence of pletely satisfactory because of technical transport to the vestibular complex was difficulties. In spite of such problems, how- seen (1 to 10-day survival). ever, we are able to see evidence for most of the contralateral projections indicated Vestibular afferents from the midbrain in the degeneration material. Although In those midbrain cases with large there is some suggestion that ipsilateral faslesions involving the periaqueductal grey, tigiobulbar fibers exist, there is no indicathe nucleus of Darkschewitsch, the intertion of a projection to the dorsal part of the of Cajal and/or their desstitial nucleus lateral vestibular nucleus of either side. cendingaxons (e.g., fig. S),evidence is presSuch results strongly indicate that the deent for an ipsilateral projection to the generation present in the latter area after medial and inferior vestibular nuclei (see fastigial damage is a result of undercutting 9 for the distribution of the degenerafig. fibers from the cerebellar cortex. Concerntion produced by the lesion in fig. 5). Nuing the question of laterality, it is of merous fragmented fibers leave the main interest to note that after HRP is injected pathway and follow the hypoglossal nerve into the medial vestibular nucleus labelled neurons are found bilaterally within the dorsally through the paramedian reticular formation (fig. 9d). Some end within the nucleus fastigius. perihypoglossal cell groups, while others Numerous cases with lesions of the intersweep over the dorsal motor nucleus of the positus-dentate complex were also examvagus and into the medial vestibular nuined for vestibular degeneration. In most of them the fastigial and cerebellar cortical cleus (arrow: fig. 9d). Evidence for a comaxons were undercut producing a picture parable projection is present in two brains of ipsilateral degeneration similar to that with injections of radioactive leucine limpresent on the side of fastigial lesions. The ited mainly to the interstitial nucleus of conclusion that such degeneration is Cajal (24-hour survival). Subsequent to large lesions which damfastigial and cortical in origin is reinforced age either the rostral two-thirds of the red by data obtained from ten brains in which nucleus or its efferent fibers, degenerating tritiated leucine was placed in the interpositus and/or dentate nuclei. Although axons can be traced from the crossed rubsilver grains are well above back round in robulbar tract into the vestibular complex. many areas of the neuraxis t.g., the Upon diverging from their main course opposite thalamus, red nucleus, basilar such axons distribute densely to that part of the rostral cuneate-gracile complex sugFig. 10 Fink-Heimer impregnated section at the gested in Part I to be cell group “2” and border of the inferior vestibular nucleus (VstI) and the spinal trigeminal tract (trs).Sparse degeneration is more rostrally to cell group “x.” Still others seen here in cell group “X’ after a posterior vermal enter the inferior vestibular nucleus, and lesion (P 266: 6g. 1). The spinal trigeminal tract (trs) preterminal debris is scattered there, as is labelled. well as in the medial nucleus. Occasionally Fig. 11 Fink-Heimer preparation. A photomicrosingle degenerating axons are seen within graph of cell group “ X ’ illustrating degenerating more rostral divisions of the vestibular fastigial fibers (P 224). Fig. 12 High power photomicrograph of labelled complex, but these are always random and rubral (Rb) neurons at the injection site in P 418. The very sparse. Small rubral lesions involving open block arrow designates a heavily labelled large- only the caudal half of the red nucleus did medium sized neuron whereas the solid block arrow not produce obvious degeneration within points to a similarly labelled giant neuron. Fig. 13 Low power photomicrograph (semi-dark the vestibular complex. field) of the caudal portion of cell group “ X ’ subseBecause rubral lesions necessarily interquent to the tritiated leucine injection of the red nu- rupt axons arising in other areas, tritiated cleus shown in figure 12. Silver grains clearly mark “X’in its relation to the lateral cuneate nucleus leucine was injected into the red nucleus of (CuL).The spinal trigeminal tract (trs) is labelled for two animals which were sacrificed after a 10-day survival time. Neurons were lareference.

332

C. K. HENKEL AND G. F. MARTIN

C.2 Hemisection

P 320 Figure 14

f

CONNECTIONS OF THE OPOSSUM VESTIBULAR COMPLEX

333

belled within the red nucleus (fig. 12), as topographic organization. One high cerviwell as within the closely adjacent tegmentum, but no obvious labelling was present within either the nucleus of Darkschewitsch or the interstitial nucleus of Cajal. When the developed sections were examined, labelling could be traced to cell groups “x” and “z” contralaterally (fig. 13). Although present, silver grains over the inferior vestibular nucleus are less apparent. No ipsilateral labelling in the perihypoglossal nucleus or medial vestibular nucleus is noticeable. An attempt was made to mark the midbrain neurons which roject to the medial vestibular nucleus their main ipsilateral vestibular target) by injecting the latter nucleus with HRP. Although there is considerable s illover of the enzyme at the injection site e.g., fig. 18e) and the medial vestibular nucleus is black with reaction product only a very few midbrain neurons show evidence of retrograde transport. Lightly labelled neurons are present in the ipsilateral interstitial nucleus of Cajal (fig. 18a) as well as in a paramidline area referred to by Oswaldo-Cruz and RochaMiranda (’68) as the rostral oculomotor complex (asterisk in fig. 18a). No labelled neurons are present in the red nucleus, although due to its very light projection to the medial nucleus they might not be expected.

e

P

Vestibular afferents from the spinal cord A series of experimental cases with spinal cord lesions was studied to determine the distribution of spinal fibers to the vestibular complex and the possibility of a Fig. 14 A series of stacked sections (rostral to caudal, a-0 to illustrate the distribution of spinovestibular fibers after a C-2 spinal hemisection. In frame f the arrow indicates dense degeneration in that part of the rostral gracile-cuneate complex which may be comparable to cell group “z” in the cat. The arrow in frame d indicates a fascicle of degenerating fibers that continues from the dorsal column area and distributes to cell group “x” (frame e) and the inferior vestibular nucleus. Injured axons are noted (arrow, frame b) leaving the main course of the dorsal spinocerebellar tract and distributingto the lateral vestibular nucleus. The asterisks in frame c demarcate the olivocochlear bundle.

cal lesion (C-2) severed the posterior columns, the entire right lateral funiculus and the associated grey matter, partially cutting the ventrolateral white matter. Gliosis and axonal swelling in the ventral funiculus, however, suggest even further involvement. As expected, degeneration fills both the rostral end of the (medial) cuneate nucleus and the lateral cuneate nucleus. Many degenerating axons continue rostrally, however, and end in the indistinct zone of darkly stained cells indicated by the arrow in figure 14f as well as within cell group “x” (fig. 14e). In progressively more rostral sections a few degenerating fascicles follow the cell strands of group “x” and distribute along the lateral extreme of the spinal trigeminal tract as well as in the caudal art of the inferior vestibular nucleus figs. 14c,d). Although most of these fibers can be traced from the dorsal column area, some appear to travel with the dorsal spinocerebellar tract. Degenerating axons within the main vestibular nuclei are present but scattered. Most of the degeneration in the caudal part of the inferior vestibular nucleus appears to be from dorsal column and dorsal spinocerebellar sources, whereas that within the caudal third of the medial nucleus (figs. 14d,e) is seen to diverge from the main spinobulbar tracts. The latter fibers spread dorsally through the reticular formation, ending finally along the ventricle in the perihypoglossal and medial vestibular nuclei. Axonal debris increases in the rostral half of the inferior vestibular nucleus (fig. 14d). At that level fragmented axons leave the dorsal spinocerebellar tract and course medially through the inferior nucleus, ending there as well as within the adjacent medial nucleus (fig. 14d). Other fibers continue to enter the medial nucleus from the reticular formation, and in one brain cut in the sagittal plane, it was apparent that they turn rostrally before terminating. Axons from the dorsal spinocerebellar tract leave the main course and distribute to both the lateral (arrow: figs. 14b,c) and superior nuclei (fig. 14a). Such fibers are most numer-

P

334

C. K. HENKEL AND G . F. MARTIN

/

H-LeucineInjection

Figure 15

CONNECTIONS OF THE OPOSSUM VESTIBULAR COMPLEX

ous within peripheral areas of the latter nucleus. Another animal’s s inal cord was hemisected at C-5 (P-321! leaving intact most neck afferents to the brainstem. As in the previous case degeneration in the dorsal column area distributes to the rostral end of the gracile-cuneate complex. However, the degeneration in group “x” is obviously diminished and very little extends further than the lateral edge of the inferior vestibular nucleus. Likewise, little or no degeneration can be traced from the ventral spinobulbar tracts to the medial nucleus. The contribution of spinocerebellar fibers to these caudal areas is also lacking as evidenced by the paucity of axonal debris in the inferior nucleus rostral to the level of group “x”. Only scattered debris can be traced from the dorsal spinocerebellar bundles to the lateral and superior vestibular nuclei in rostral sections. We have examined cases with thoracic, lumbar and sacral lesions. While some degeneration from dorsal column and spinocerebellar pathways distributes to group “x” in each, it is considerably less than after cervical lesions, and only a few single fragmented axons are found in the main nuclear divisions.

Efferent projections from the vestibular complex Ascending projections from the vestibular complex In order to study ascending vestibular projections we first examined cases in which the vestibular nuclei or their efferent fibers had been damaged. In one case a Fig. 15 A series of drawings illustrating only the periaqueductal region of the midbrain (rostral to caudal, a-d) indicating the distribution of silver grains over oculomotor and accessory ocular regions after the 3H leucine placement shown in the insert and illustrated in figure 8. Contralateral (contra) and ipsilateral (ipsi) sides are marked in the first section. The cerebral aqueduct (aq), dorsal raphe nucleus (RaD),lateral vestibular nucleus (VstL),medial longitudinal fasciculus (Iflm), medial vestibular nucleus (VstM), nucleus of Darkschewitsch (Dk), oculomotor nucleus (OcM),superior vestibular nucleus (VstS)and trochlear nucleus (Tro) are labelled.

335

large lesion destroyed most of the superior and lateral vestibular nuclei (P-177:fig. 7) as well as part of the rostral medial nucleus and cerebellum. Degenerating axons issue from the damaged area and distribute bilaterally to the abducens nuclei as well as to the vestibular complex of the opposite side (commissural fibers). The degenerating axons which ascend to midbrain centers do so primarily in the ipsilateral medial longitudinal fasciculus, although a few are present in the same area contralaterally. In addition, injured axons can be followed from the brachium conjunctivum (cerebellar damage) to the contralateral oculomotor complex. Most of the degeneration within the trochlear and oculomotor nuclei is located on the side of the lesion. Although subgroups of the opossum oculomotor complex are not obvious in Nissl preparations, we noticed that at some levels terminal debris is not uniformly dispersed and may therefore be revealing otherwise indistinguishable subnuclei. A few injured fibers also extend into the periaqueductal grey and a large number continue rostrally into the nucleus of Darkschewitsch and the interstitial nucleus of Cajal. Those few degenerating axons which ascend in the contralateral medial longitudinal fasciculus also terminate within ocular areas. Although only a few of them enter the contralateral trochlear nucleus, there is a very distinct puff of terminal debris at intermediate levels of the oculomotor complex. Because of the cerebellar contamination, however, and the fact that degenerating axons may be traced to that region from the brachium conjunctivum, the origin of the degenerating axons is not certain. A much smaller lesion was placed in the superior vestibular nucleus of another animal, damaging only its ventromedial border and interrupting efferent fibers coursing to the medial longitudinal fasciculus. As in P-177 axonal fragments can be followed from the medial longitudinal fasciculus to both the trochlear and oculomotor nuclei, but only on the side of the

336

C. K. HENKEL AND G. F. MARTIN

,

CONNECTIONS OF THE OPOSSUM VESTIBULAR COMPLEX

lesion. Some axonal debris continues rostrally and is distributed within the nucleus of Darkschewitsch, the interstitial nucleus of Cajal and the central grey. The results obtained from degeneration experiments were substantiated by autoradiography. In P-386 (fig. 15) tritiated leucine was injected into the superior and lateral nuclei also labelling a few cells in the rostra1 medial nucleus and the deep cerebellar nuclei. A photomicrograph through part of the injection site is shown in figure 8. Although silver grains are visually above background over the contralateral oculomotor nucleus and some of the accessory ocular cell groups, most of them are found on the same side. Label is present over the trochlear nucleus, the oculomotor nucleus, the nucleus of Darkschewitsch, the interstitial nucleus of Cajal and the central grey (fig. 15). Another animal (P-148: not illustrated) was subjected to a lesion which included portions of the medial and inferior vestibular nuclei as well as the dorsal cochlear nucleus. Although a few degenerating axons can be traced to the abducens nucleus on the side of the lesion, most of those in the medial longitudinal fasciculus distribute to the opposite side of the midbrain. Such fibers terminate within the trochlear and oculomotor nuclei, as well as in the central grey and accessory ocular nuclei. Reinforcement of these findin s was obtained from still another brain P-24: fig. 6) in which the fibers issuing from the caudal part of the vestibular complex were undercut by the lesion. While the lesion in the latter two cases potentially interrupted axons from both the medial and inferior nu-

7

Fig. 16 A series of drawings, indicating the presence of silver grains in the midbrain following an injection of tritiated leucine into the medial and inferior vestibular nuclei (insert). The grains are distributed primarily on the contralateral side (marked on reader’s left in section a). In the insert the blackened area indicates the needle tract and surrounding necrosis, whereas heavily labelled neurons are present throughout the shaded area. The dorsal cochlear nucleus (Cc), facial nucleus (Fac) and restiform body (cr) are labelled in addition to those areas labelled in figure 15.

337

clei, a more lateral lesion in brain P-115 damaged only the inferior nucleus (also some of the spinal trigeminal complex) and no degeneration can be seen in midbrain ocular centers. Autoradiographic studies confirm the existence of a predominantly contralateral projection to the midbrain from caudal vestibular areas. In P-382 (fig. 16) the inferior and medial vestibular nuclei are labelled with tritiated leucine, as well as the caudal part of the lateral nucleus. Silver grains are present contralaterally over each of the extraocular nuclei and the accessory groups, as well as over a restricted area of the ipsilateral oculomotor nucleus (fig. 16b). The retrograde transport of horseradish peroxidase was employed to label the neurons in the vestibular complex which project to the midbrain ocular nuclei. In three successful cases, the injected marker was limited to the oculomotor nucleus and surrounding areas and the vestibular complex was surveyed in order to identify neurons whose cytoplasm contained reaction product. On the side of the placement numerous medium-size neurons are labelled in generally central areas of the superior nucleus (fig. 17b) and a few neurons of comparable size are positive for the reaction product in the ventral part of the lateral nucleus (fig. 17c). On the opposite side, however, labelled neurons are present mainly in the medial vestibular nucleus (figs. 17c,d) although a few are also seen within the superior nucleus (fig. 17b). Again, such neurons fall within the medium size range. The number of HRP-positive neurons in nuclei other than the ipsilateral superior nucleus is comparatively low, however, amounting to only two or three cells per section. In addition, small to medium sized neurons are positively labelled in the contralateral parabrachial complex (terminology of Oswaldo-Cruz and Rocha-Miranda, ’68). The latter area also receives primary vestibular fibers, as described in Part I of our study, and we have suggested that it may be comparable to cell group “y” of the cat. Another

338

C. K. HENKEL AND G . F. MARTIN

CONNECTIONS OF THE OPOSSUM VESTIBULAR COMPLEX

339

One brain was examined in which the medial vestibular nucleus was injected. However, the needle tract passed through nucleus fastigius and its neurons were also Vestibular projections to the reticular covered with silver grains. Although label formation is above background in many reticular In those brains with vestibular damage areas on both sides, those areas also corredegenerating axons are numerous within spond to the targets of fastigial fibers (Marthe reticular formation. However, since tin et al., ’74b). cerebellar cortical and fastigial fibers In light of the above results, we injected course through the vestibular complex and horseradish peroxidase into the medulla are necessarily interrupted by the lesions, oblongata of several animals, heavily lainterpretation of these cases was difficult belling the nucleus paramedian, the nu(the reader is referred to Martin et al., cleus interfascicularis hypoglossi, the sub’74b; Sreesai, ’75; Haines et al., ’76, for nuclei ventralis of the nucleus medulla obdescriptions of cerebellar cortico- and fas- longata centralis and, in some cases, the tigio-reticular projections in the opossum). inferior olivary nucleus. Although positiveIn order to more precisely evaluate vestib- ly reacting neurons are scattered throughular input to the reticular formation, par- out most of the ipsilateral vestibular comticularly the areas which also receive pro- plex, they are especially numerous in the jections from the cerebellum, we utilized superior and inferior nuclei. Most of the lathe autoradiographic material referred to belled neurons are medium size although a above. few in the inferior and medial nuclei fall In cases with labelling of both the within the small size range. superior and lateral vestibular nuclei, silver Because of the positive evidence for grains are above background (mainly ipsi- vestibular input to the reticular formation laterally) in the nucleus medulla oblongata we examined several cases in which triticentralis, the nucleus parvocellularis, the ated leucine had been injected into the renucleus lateralis reticularis, the nucleus ticular formation. In one case with heavy pontis centralis oralis and caudalis, the nuc- labelling of nucleus pontis centralis cauleus gigantocellularis and paragigantocel- dalis neurons and in another with comparalularis dorsalis, the nucleus paramedian bly heavy labelling of the nucleus giganand the nucleus interfascicularis hypoglos- tocellularis, the magnocellular pontine si. Evidence for a small vestibular input to raphe and a small part of pontis centralis the ipsilateral nucleus reticulotegmenti caudalis, there is definite evidence for a pontis is also present. projection to the vestibular complex. TerFig. 17 A series of representative sections, minal labelling is most obvious in the stacked rostra1 to caudal (a-d),illustratingthe location ventral part of the lateral vestibular nuof HRP-positive neurons (large black dots) in the cleus and it is bilateral.

interesting group of labelled cells is present within the contralateral abducens nucleus (fig. 17c).

vestibular complex after the placement shown in the insert. The concentration of reaction product at the injection site is indicated by the relative shading in the insert and the side of the placement is on the viewer’s left in each of the sections.The arrow in fig.a points to the region designated as comparable to subnucleus “y,” The abducens nucleus (Ab), brachium conjunctivum (bc), dorsal cochlear nuclei (CcD), facial (Fac), medial parabrachial nuclei (PBrm), medial vestibular nucleus (VstM), motor trigeminal nucleus (TrMo), restiform body (cr), sensory trigeminal root (rv), s inal trigeminal tract (trs), superior olivary nucleus superior vestibular nucleus (VstS),ventral (CcV) cochlear nuclei and vestibular nerve (n VIIIv) are labelled.

bs),

Commissural vestibular projections Evidence for commissural vestibular connections is present in most of the available degeneration and autoradiographic material. In cases with lesions including both the superior and medial nuclei, numerous axonal fragments can be followed beneath the ventricle to the contralateral vestibular complex. Although such fibers end most profusely in the superior nucleus, considerable degeneration can also be seen

NILHVN ’d ‘3 CINV 1 3 8 N 3 H ‘X ‘3

OPE

CONNECTIONS OF THE OPOSSUM VESTIBULAR COMPLEX

in the medial nucleus, the ventral region of the lateral nucleus and in the contralateral inferior nucleus. When lesions are restricted to the medial vestibular nucleus, the greatest crossed vestibular degeneration is found in its contralateral counterpart, although debris can also be traced to the other vestibular nuclei. The data obtained from brains prepared for autoradiography confirm the results gained by degeneration methods. In order to visualize the neurons giving rise to at least part of the commissural connections, we employed the case referred to previously in which horseradish peroxidase was injected into the medial vestibular nucleus filling its caudal half and extending into both the reticular formation and the inferior vestibular nucleus (sections d,e: fig. 18). Although distinctly labelled neurons are most numerous in the opposite medial nucleus (figs. 21,22), they are present in some part all four of the main vestibular nuclei as we have defined them (fig. 18). Most of the labelled neurons are medium sized, although a few in the inferior and medial nuclei measure in the small size range. In addition to the neurons found in the main vestibular nuclei contralaterally, small cells are labelled in cell group "y" (figs. 18-20).No other vestibular subgroups show labelling with the reaction product. DISCUSSION

In the following account we will consider our data on the major afferent vestibular connections first and attempt to compare our findings with those reported for other species. Secondly, we will discuss the vestibular projections to various areas of the brainstem including those considered to be commissural. The reader is referred to Martin et al. ('75) for information concerning vestibulospinal connections in the same species.

Aflerent projections to the vestibular complex Cerebellar cortico-uestibular connections have been studied in the opossum by

341

Sreesai ('75), who reported that they distribute bilaterally. In contrast, our findings suggest that they are mainly, if not exclusively, unilateral as described for placental species (see Van Rossum, '69; Haines, '75, for recent reviews). Although our material provides evidence for only a small cerebellar cortical projection to the superior vestibular nucleus (perhaps reflecting the sparing of the flocculonodular lobe by most lesions, see Angaut and Brodal, '67) comparable input to the lateral and inferior nuclei is extensive. Similar atterns have been described in the cat Walberg and Jansen, '61; Angaut and Brodal, '67), the albino rat (Goodman, et al., '63; Achenback and Goodman, '68), the rabbit (Van Rossum, '69), the galago (Haines, '75) and the rhesus monkey (Eager, '67). Projections to the medial vestibular nucleus from the posterior lobe have been reported for other species although such a connection cannot be verified from our material. We have observed degeneration in the medial vestibular nucleus after cortical lesions, but in such cases we were not able to exclude fastigial contamination. The opossum is similar to the cat (Walberg and Jansen, '61) in that the heaviest input to the lateral vestibular nucleus is from the anterior lobe rather than the posterior vermis. When considered in their totality, anterior vermal fibers in the opossum end throughout the dorsal part of the lateral nucleus although evidence for a topographical organization is present (see the somatotopic organization described for the cat; Walberg and Jansen, '61). The more lateral anterior lobe lesions (paravermis) produced particularly extensive degeneration in the most dorsal part of the lateral nucleus both rostrally and caudally and it has been suggested by Jansen and Brodal ('40), and more recently by Voogd ('64) in the cat, Van Rossum ('69) in the rabbit, and Haines (personal communication) in the galago, that special longitudinal strips of the paravermis are the source of this projection. It is of interest that labelled

P

342

C. K. HENKEL AND G . F. MARTIN

CONNECTIONS OF THE OPOSSUM VESTIBULAR COMPLEX

neurons are present in most of the vestibular nuclei except the dorsal part of the lateral nucleus after HRP is placed in the anterior lobe of the cerebellum (unpublished results). Such neurons are particularly numerous within the inferior nucleus and subgroups “f” and “x.” A few anterior lobe axons continue beyond the lateral nucleus to the rostral part of the inferior vestibular nucleus. However, degeneration within the caudal part of the inferior nucleus and its associated cell groups is found only after posterior vermal damage, a finding also reported for the rat (Achenbach and Goodman, ’68), the cat (Angaut and Brodal, ’67) and galago (Haines, ’75). The posterior vermal input to the inferior nucleus (primarily from the pyramis and uvula) is largely distributed to its dorsal and lateral margin bordering on the cochlear complex. Only a few fibers are located centrally. Axonal degeneration in this rim of the inferior nucleus extends into subgroup “f’and to the most rostral part of group “x.” Similar connections have been reported for the cat (Angaut and Brodal, ’67) and galago (Haines, ’75). Fastigiovestibular pathways have been previously described in the opossum by Foltz and Matzke (‘60)and Martin et al. ~~~

~

~

Fig. 19 Low power photomicrograph of the parabrachial cell group (arrow) referred to herein as “y.” This section was taken from P 448, illustrated in figure 18. The enclosed area is shown in a higher power dark field hotomicrograph in figure 20. The brachium pontis pbp) and motor trigeminal nucleus (trMo) are indicated. Fig. 20 The small open block arrow points to a blood vessel in subnucleus “y” which corresponds to that similarly indicated in the insert. The small, faintly labelled commissural neurons of group “y” (small arrowheads) can best be visualized in the insert provided. These neurons are labelled contralaterallyfollowing the medial nucleus HRP injection shown in figure 18. Fig. 21 Low power photomicrograph of the medial nucleus from P 448, figure 18. The (open and solid block) arrows indicate two of the several commissural neurons labelled after HRP-injectionin the opposite medial nucleus. These same neurons are shown under higher power in figure 22. Fig. 22 High power dark field photomicrograph of the neurons indicated in figure 21.

343

(‘74b). Our study has attempted to expand these findings in light of the cytoarchitectural detail described in Part I. Opossum fastigiovestibular projections appear to be similar to those of the angolin (Bautista and Foltz, ’68), the cat Walberg et al., ’62), and several primates (Bautista and Matzke, ’66; Carpenter, ’59) in that the crossed fibers end in some portion of all the main vestibular nuclei. Crossed fastigial axons end extensively in the dorsal rim of small cells which separates the inferior vestibular nucleus from the cochlear complex. The degeneration in this area is continuous with that found within cell groups “f’ and “x” as we have defined them. It should be noted, however, that degeneration is not present in either of the latter areas after lesions which are restricted to the caudal nucleus fastigius, a finding similar to that reported for other species (Rasmussen,’33;Jansen and Jansen, ’55; Cohen et al., ’58; Walberg et al., ’62; Bautista and Matzke, ’66). Fastigial axons continue caudally through the vestibular complex and end in an area we have designated as the nucleus parasolitarius following the terminolog employed in the cat (Walberg et al., ’627. It is necessary to consider interrupted fibers of passage from the cerebellar cortex and contralateral nucleus fastigius in order to interpret the ipsilateral degeneration present after fastigial lesions. Since no autoradiographic evidence is present for a projection from the nucleus fastigius to the dorsal part of the lateral vestibular nucleus, we are forced to assume that the ipsilateral degeneration present in the latter area after fastigial lesions results from undercutting cerebellar cortical fibers. Unfortunately, the autoradiographic results related to the possibility of ipsilateral fastigial projections to other vestibular areas is not so clear. However, the presence of labelled neurons in the ipsilateral fastigial nucleus after HRP injections into the medial vestibular nucleus suggests that an ipsilateral projection exists. Foltz and Matzke (‘60) suggested that the nucleus interpositus contributes to the

7

344

C. K. HENKEL AND G . F. MARTIN

uncinate fasciculus in the opossum and Roy and Courville (‘74) have reported comparable findings in the cat. Since interpositus and dentate lesions often interrupt fastigial and/or cortical fibers in the opossum, we have examined autoradiographic cases with tritiated leucine placements in both the nucleus interpositus and dentatus, and have seen no evidence of vestibular projections. It is possible, of course, that our negative evidence simply reflects failure to label the appropriate neurons and/or the difficulty in distinguishing the border between the fastigial and osterior interpositus nuclei in the opossum Martin et al., ’74b). We have described an ipsilateral midbrain projection to the medial vestibular nucleus with only minimal input to the inferior nucleus. Our data is consistent with that described in most early reports (see Pompeiano and Walberg, ’57, for review) suggesting that the fibers in question arise mainly from the so-called interstitial nucleus of Cajal. However, there is an indication that some of them also arise from paramedian neurons closely related to the oculomotor complex. Crossed rubro - uestibular projections have been described previously in the opossum (Martin et al., ’74a) and the cat (Edwards, ’72),and in the present study we have attempted to clarify their targets according to the vestibular divisions described in Part I. Such fibers leave the crossed rubrobulbar tract, skirt the edge of the spinal trigeminal nucleus and end in the transitional regions between the vestibular complex and dorsal column nuclei. This small complement of fibers reaches the rostral and medial edge of the lateral cuneate nucleus, ending there and in the region referred to as group “x.” This distribution is comparable to that reported in the cat (Edwards, ’72). It was noted that rubral fibers do not extend into the most rostra1 area of nucleus “x,” e.g., along the spinal trigeminal tract, or in the lateral angle of the inferior vestibular nucleus. In addition, rubral axons end in relation to neurons tentatively iden-

P

tified as group “z.”Although fibers to nucleus “x” and possibly “z” constitute the heaviest portion of this modest projection, occasionally rubral axons can be followed into the inferior nucleus, and to an even lesser degree, into the medial vestibular nucleus. Little evidence of rubral input to subgroup “f” of the inferior nucleus was observed in contrast to the findings of Edwards (’72) in the cat. Mehler (‘69) and Hazlett et al. (’72) described spinobulbar projections in the opossum and in their reports spinouestibuZur connections were briefly mentioned. As reported in other mammals, i.e., the hedgehog (Jane and Schroeder, ’71), the rat (Mehler, ’69),the cat (Pompeiano and Brodal, ’57), the domestic pig (Braezile and Kitchell, ’68), the sheep (Rao et al., ’69), the tree shrew (Schroeder and Jane, ’71) and certain primates (see Mehler, ’69, for review), the projection to the main vestibular nuclei is modest. The heaviest distribution of spinovestibular fibers is to subnucleus “x.” By examining cases with lesions at different spinal levels we have sought to determine the distribution and possible somatotopic organization of spinovestibular fibers in the opossum. As might be expected, high (C-2) cervical hemisections produce the heaviest vestibular degeneration and the amount of degeneration in the vestibular complex is considerably diminished, but similarly distributed, after C-5 lesions. By contrast thoracic, lumbar and sacral spinal lesions elicit only limited and very sparse vestibular degeneration. This is particularly apparent with respect to the lateral part of the inferior vestibular nucleus and cell roup “x.” Although Pompeiano and Brodal ’57) suggested that in the cat the largest complement of spinovestibular fibers arises within the lumbosacral spinal cord, our material leads us to the conclusion that they arise most extensively from cervical levels in the opossum. The existence of degeneration in the oral pole of the gracile and cuneate nuclei after thoracic, lumbar or sacral lesions reinforces our suggestion that those

7

CONNECTIONS OF THE OPOSSUM VESTIBULAR COMPLEX

345

areas contain neurons comparable to group motor HRP injections. They are most nu“z” in the cat (Pompeiano and Brodal, ’57; merous by far on the side opposite the lesion and apparently contribute to the Rustioni, ’73). contralateral projections to the oculomotor Eferent projectionsfrom the vestibular nucleus demonstrated by orthograde techcomplex niques. The existence of primary vestibular Both degeneration and autoradiographic input to the parabrachial area (Part I), studies of ascending vestibular projections together with its projection to ocular nuclei reveal that crossed and uncrossed fibers suggests that it has an important role in vescourse in the medial longitudinal fasciculus tibulo-ocular reflexes. Anatomical studies of the opossum as reported for several pri- in the cat indicate that cell group “y” mates (Tarlov, ’69), the cat (Tarlov, ’70) receives saccular fibers from the eighth and the rabbit (Highstein, ’73a,b). Most of nerve (Gacek, ’69) and projects contralatthe fibers from the superior vestibular nu- erally to the oculomotor complex (Graybiel cleus end ipsilaterally in the trochlear and and Hartwieg, ’74). The parabrachial cell oculomotor nuclei, the periaqueductal group described herein for the opossum grey, the nucleus Darkschewitsch and the may be comparable to cell group “y” of the interstitial nucleus of Cajal, although a few cat. Our retrograde results also confirm the end in generally comparable areas of the recent report by Graybiel and Hartwieg opposite side. It is also clear that the medial (‘74) that neurons within the abducens nuvestibular nucleus projects mainly to cleus project to the contralateral oculomidbrain ocular and accessory ocular nu- motor nucleus. clei of the o posite side (see Tarlov, ’72, for We have examined a number of cases a discussion . It should be noted, however, which generated data concerning vestihthat there was evidence for an ipsilateral lo-reticular projections in the opossum. Alprojection from the medial nucleus to the though it appears that such connections dorsal part of the oculomotor complex. arise from all four major divisions of the In order to localize and characterize the vestibular complex, they originate most neurons that project to the oculomotor nu- extensively from the superior and inferior cleus and adjacent areas, we have em- nuclei. The vestibular nuclei in the ployed the horseradish peroxidase tech- opossum project to numerous areas of the nique. When the oculomotor nucleus and reticular formation, as suggested for the cat surrounding regions were black with reac- (Ladpli and Brodal, ’68), and it should be tion product, labelled neurons were pres- emphasized that many of those areas ent ipsilaterally in the superior vestibular receive fastigial connections (Martin et al., nucleus and the ventral part of the lateral ’74b) and project to the spinal cord (Martin nucleus. In both areas the positively react- and Dom, ’71; Martin et al., ’76a). It is of ing cells are medium-sized and it is clear interest that some of the areas receiving that they are most numerous in the central vestibular and fastigial connections project part of the superior nucleus. As would be back to the vestibular complex. expected from the orthograde results, HRP Evidence from both degeneration and labelled neurons of comparable size are autoradiographic studies indicates that vespresent in the contralateral medial vestibu- tibular commissural pathways are present lar nucleus (as well as in the superior nu- in the opossum and, while there is no doubt cleus), but not in great numbers. It should that they arise from all four major nuclei, be emphasized, of course, that negative re- they appear to issue most extensively from sults are by no means conclusive. the superior and medial nuclei as suggested Neurons also are labelled in the para- in the cat by Brodal (‘72a). Horseradish brachial area (terminology of Oswaldo- peroxidase experiments revealed crossed Cruz and Rocha-Miranda, ’68) after oculo- connections to the medial vestibular nu-

P

346

C. K. HENKEL AND G. F. MARTIN

cleus from all four major vestibular nuclei cleus distribute contralaterally for the most as well as from the putative cell group “y.” part. Although the dorsal part of the lateral SUMMARY AND CONCLUSIONS vestibular nucleus has no primary vestibuApart from the primary input described lar input, it does receive a major projection in Part I of our study, the cerebellum is from the cerebellar cortex. This same the major source of afferent fibers to the vestibular region projects strongly to the vestibular complex. Cerebellar cortical spinal cord, but not the extraocular nuclei. axons end mainly outside of the area of pri- The ventral part of the lateral nucleus and mary afferent distribution, e.g., in the dor- perhaps the medial nucleus also relay to sal part of the lateral vestibular nucleus, the spinal cord. Additional projectionsfrom the dorsolateral margin of the inferior all vestibular nuclei to the reticular formavestibular nucleus and in cell groups “x” tion provide an indirect route through and “f.” In contrast, fastigial fibers overlap which vestibular discharge can potentially extensively with vestibular input; primarily affect multiple systems, including those in the ventral part of the lateral nucleus, innervating the spinal cord. Finally, comthe central part of the inferior nucleus and missural connections arise within all four of the medial nucleus. Axons of fastigial origin the main vestibular nuclei. This and previous studies from our laboalso distribute sparsely to the superior vestibular nucleus and heavily to subnuclei ratory have provided data concerning brainstem supraspinal motor centers in the “f,” “x” and the parasolitary region. Spinal fibers constitute a lesser input, but opossum and indicate that they are distribute diffusely throughout the main generally comparable in conformation and vestibular nuclei. The densest distribution connectivity to those of more commonly of spinal fibers is found within subnucleus used placental mammals. Such studies pro“x.” In contrast to what has been reported vide the necessary baseline in the adult for the cat, most spinovestibular fibers from which to interpret the results of developmental studies of these same centers. appear to arise in the cervical cord. Ipsilateral connections from the inter- The accessibility of such information, the stitial nucleus of Cajal end predominantly ready availability of the opossum as an in the medial vestibular nucleus with only a experimental animal and its unique marvery small contribution to its inferior coun- supial embryology make it a potentially terpart. A second midbrain projection has excellent model for such studies. been substantiated from the red nucleus to ACKNOWLEDGMENTS the cell group “x,” the lateral part of the inferior nucleus, as well as to an area possiThis investigation was supported by the bly comparable to cell group “z.” United States Public Health Service Grant In Part I of our study we have shown that NS-07410. The authors wish to thank Docthe major targets of primary vestibular tors David Clark and James King for their fibers are the central part of the superior help, Ms. Malinda Amspaugh for typing the nucleus, a portion of the parabrachial com- manuscript and Mr. Gabriel Palkuti for plex possibly comparable to subnucleus photographic assistance. “y,” the ventral part of the lateral nucleus LITERATURE CITED and the medial nucleus. These areas project most heavily to the extraocular nuclei Achenbach, K. E., and D. C.Goodman 1968 Cerebellar projections to pons, medulla and spinal cord and surrounding areas. The ascending albino rat. Brain, Behav. and Evol., 1: 43-57. midbrain fibers from the superior nucleus h gina uthe t , P., and A. Brodal 1967 The projection of the end mainly ipsilaterally, whereas those “vestibulocerebellum” onto the vestibular nuclei in from the area probably comparable to subthe cat. Arch. Jtal. Biol., 105: 441-479. nucleus “y” and the medial vestibular nu- Bautista, N. S., and F. M. Foltz 1968 The efferent

CONNECTIONS OF THE OPOSSUM VESTIBULAR COMPLEX

connections of the cerebellar nuclei in the pangolin. J. Comp. Neur., 132: 213-226. Bautista, N. S., and H. A. Matzke 1966 A degeneration study of cerebellofugal fibers in monkey. J. f i Hirnforsch., 8: 283-299. Braezile, J. E., and R. L. Kitchell 1968 Ventrolateral spinal cord afferents to the brain stem in the domestic pig. J. a m p . Neur., 133: 363-372. Brodal, A. 1972 Organization of the commissural connections: Anatomy. In: Progress in Brain Research. Vol. 37. Basic Aspects of Central Vestibular Mechanisms. A. Brodal and 0. Pompeiano, eds. Elsevier, Amsterdam, p. 167. Brodal, A., and 0. Pompeiano 1972 Basic aspects of central vestibular mechanisms. In: Progress in Brain Research. Vol. 37 Elsevier, Amsterdam. Carpenter, M. B. 1959 Lesions of the fastigial nucleus in the rhesus monkey. Am. J. Anat., 104: 1-33. Cohen, D., W. W. Chambers and J. M. Sprague 1958 Experimental study of the efferent projections from the cerebellar nuclei to the brainstem of the cat. J. Comp. Neur., 109: 233-259. Cowan, W. M., D. I. Gottlieb, A. E. Hendrickson, J. L. Price and T. A. Woolsey 1972 The autoradiographic demonstration of axonal connections in the CNS. Brain Res., 37: 21-51. Eager, R. 1967 Long cerebellar corticofugal fibers in the monkey. In: Third Symposium on the Role of the Vestibular Organ in Space Exploration. NASASP 152, pp. 225-237. Edwards, S. B. 1972 The ascending and descending projections of the red nucleus in the cat: an experimental study using an autoradiographic tracing method. Brain Res., 48: 45-63. Fink, R. P., and L. Heimer 1967 Two methods for selective silver impregnation of degenerating axons and their synaptic endings in the central nervous system. Brain Res., 4: 369-374. Foltz, F. M., and H. A. Matzke 1960 An experimental study on the origin, course and termination of the cerebellar fibers in the opossum, J. Comp. Neur., 114: 107-125. Gracek, R. R. 1969 The course and central termination of first order neurons supplying vestibular endorgans in the cat. Acta Oto-laryng. (Stockholm), (SUPPI.),254: 1-66. Goodman, D. C., R. E. Hallett and R. B. Welch 1963 Patterns of localization in the cerebellar corticonuclear projections of the albino rat. J. Comp. Neur., 121: 51-67. Graham, R. C., Jr., and M. J. Karnovsky 1966 The early stages of absorption of injected horseradish peroxidase in the proximal tubules of mouse kidney: Ultrastructural cytochemistry by a new technique. J. Histochem. Cytochem., 14: 291-302. Graybiel, A. M. 1974 Visuo-cerebellar and cerebello-visual connections involving the ventral lateral geniculate nucleus. Exptl. Brain Res., 20: 303-306. Graybiel, A. M., and E. A. Hartwieg 1974 Some afferent connections of the oculomotor complex in

347

the cat: An experimental study with tracer techniques. Brain Res., 81: 543-551. Haines, D. E. 1975 Cerebellar corticovestibular fibers of the osterior lobe in a prosimian primate, the lesser bugbaby (Galago senegalensis). J. Comp. Neur., 160: 363-398. Haines, D. E., J. L. Culberson and G. F. Martin 1976 Laterality and topography of cerebellar cortical efferents in the opossum (Didelphis marsupialis uirginiana). Brain Res., 106: 152-158. Hazlett, J. C., R. Dom and G. F. Martin 1972 Spinobulbar, spino-thalamic and medial lemniscal connections-in the American opossum, Didelphis marsupialis oirginiana. J. Comp. Neur., 146: 95-118. Highstein, S. M. 1973 The organization of the vestibulo-oculomotor and trochlear reflex pathways in the rabbit. Exptl. Brain Res., 17: 285-300. Jane, J. A., and D. M. Schroeder 1971 A comparison of dorsal column nuclei and spinal afferents in the European hedgehog (Erinaceus europeus). Exptl. Neurol., 30: 1-17. Jansen, J., and A. Brodal 1940 Experimental studies on the intrinsic fibers of the cerebellum. 11. The cortico-nuclear projection. J. Comp. Neur., 73: 267321. Jansen, J., and J. Jansen, Jr. 1955 On the efferent fibers of the cerebellar nuclei in the cat. J. Comp. Neur., 102: 607-632. Ladpli, R., and A. Brodal 1968 Experimental studies of commissural and reticular formation projections from the vestibular nuclei in the cat. Brain Res., 8: 65-96. avail, J. H., K. R. Winston and A. Tish 1973 A method based on retrograde intra-axonal transport of proteins for identification of cell bodies of origin of axons terminating within the CNS. Brain Res., 58: 470-477. lartin, G. F., M. S. Beattie, J. C. Bresnahan, C. K. Henkel and H. C. Hughes 1975 Cortical and brainstem projections to the spinal cord of the American opossum, Didelphis uirginiana. Brain, Behav. and Evol., 12: 270-310. Martin, G. F., and R. Dom 1971 Reticulospinal fibers of the opossum, Didelphis uirginiana. 11. Course, caudal extent and distribution. J. Comp. Neur., 141: 467-484. Martin, G. F., R. Dom, S. Katz and J. S. King 1974a The organization of projection neurons in the opossum red nucleus. Brain Res., 78: 17-34. Martin, G. F., C. K. Henkel and J. S. King 197613 Cerebello-olivary fibers: Their origin, course and distribution in the North American opossum. Exptl. Brain Res., 24: 219-236. Martin, G. F., J. S. King and R. Dom 1974b The projections of the deep cerebellar nuclei of the i i opossum, Didelphis marsupialis oirginiana. J. f Hirnforsch., 15: 545-573. Mehler, W. R. 1969 Some neurological species differences-a posteriori. Annals of the N.Y. Academy of Sciences, 167: 424-468. Oswaldo-Cruz, E., and C. E. Rocha-Miranda 1968

348

C. K. HENKEL AND G . F. MARTIN

The Brain of the Opossum (Didelphis marsupialis). A Cytoarchitectonic Atlas in Stereotaxic Coordinates. Instituto de Biofisica, Rio de Janeiro, Brasil. Pompeiano, O., and A. Brodal 1957 Spino-vestibular fibers in the cat. An experimental study. J. Comp. Neur., 108: 353-381. Pompeiano, O., and F. Walberg 1957 Descending connections to the vestibular nuclei. An experimental study in the cat. J. Comp. Neur., 108: 465-503. Rao, G . S., J. E. Braezile and R. L. Kitchell 1969 Distribution and termination of spinoreticular afferents in the brain stem of sheep. J. Comp. Neur., 137: 185-196. Rasmussen, A. T. 1933 Origin and course of the fasciculus uncinatus (Russell) in the cat, with observations on other fibre tracts arising from the cerebellar nuclei. J. Comp. Neur., 57: 165-197. Rossum, J. Van 1969 Corticonuclear and Corticovestibular Projections of the Cerebellum. Van Gorcum and Co., Assen. Roy, R., and J. Courville 1974 A projection from the posterior interposed nucleus of the cerebellum. Anat. Rec., 178: 453-454. Rustioni, A. 1973 Non-primary afferents to the nucleus gracilis from the lumbar cord of the cat. Brain Res., 51: 81-95.

Schroeder, D. M., and J. A. Jane 1971 Projection of dorsal column nuclei and spinal cord to brainstem and thalamus in the tree shrew, Tupaia glis. J. Comp. Neur., 142: 309-350. Sreesai, M. 1974 Cerebellar cortical projections of the opossum (Didelphis marsupialis ufrginiana).J. fur Hirnforschung., 15: 529-544. Tarlov, E. 1969 The rostra1 projections of the primate vestibular nuclei: An experimental study in macaque, baboon and chimpanzee. J. Comp. Neur., 135: 27-56. 1970 Organization of vestibulo-oculomotor projections in the cat. Brain Res., 20: 159-179. 1972 Anatomy of two vestibulo-oculomotor projection systems. In: Progress in Brain Research. Vol. 37. Basic Aspects of Central Vestibular Mechanisms. A. Brodal and 0. Pompeiano, eds. Elsevier, Amsterdam, pp. 471-491. Voogd, J. 1964 The Cerebellum of the Cat. Structure and Fibre Connexions. Van Gorcum, Assen. Walberg, F., and J. Jansen 1961 Cerebellar corticovestibular fibers in the cat. Exptl. Neurol., 3: 32-52. Walberg, F., 0. Pompeiano, A. Brodal and J. Jansen 1962 The fastigiovestibular projection in the cat. An experimental study with silver impregnation methods. J. Comp. Neur., 118: 49-76.