THE JOURNAL OF COMPARATIVE NEUROLOGY 368:608-619 (1996)

Projections From the Cerebellar Interposed and Dorsal Column Nuclei to the Thalamus

in the Rat: A Double Anterograde Labelling Study

T.D. AUMANN, J.A. RAWSON, C. PICHITPORNCHAI, AND M.K. HORNE Departments of Anatomy (T.D.A., M.K.H.) and Physiology (C.P., J.A.R.), Monash University,

and Department of Neurology, Monash Medical Centre (M.K.H.), Clayton, Victoria, Australia, 3168

ABSTRACT It is generally agreed that cerebellar and lemniscal pathways project to largely separate

areas of the thalamus and influence different functional areas of the cerebral cortex. Cerebellar afferents arise from neurones in the deep cerebellar nuclei and terminate in the ventral lateral group of thalamic nuclei or the “motor thalamus,” whereas lemniscal afferents arise from the dorsal column nuclei and terminate in the adjacent ventral posterior group of thalamic nuclei or “sensory thalamus.” However, it remains unclear whether or not these pathways converge onto thalamic neurones in the border zone between motor and sensory thalamus. The aim of this study was to compare directly the locations of cerebellar interposed and dorsal column nuclei terminals in the rat thalamus by using a double anterograde labelling technique. Microinjections of dextran-tetramethylrhodamine and dextran-fluorescein were made into the interposed and dorsal column nuclei, and labelled terminals in the thalamus were examined in the same sections. The labelled cerebellar and lemniscal terminals were located in separate areas throughout most of the ventral lateral and ventral posterior lateral nuclei, and there was only a limited region around the rostral border between these nuclei where the two groups of terminals came in close proximity to each other. In this common projection zone, however, cerebellar and lemniscal terminals seldom intermingled, and they mostly occupied separate, discreet areas. The results show that cerebellar and lemniscal fibres do indeed project to the border zone between the sensory and cerebellar thalamic nuclei, but they show practically no overlap in this region and are likely to influence separate thalamic neurones. 4, 1996 Wiley-Liss, Inc.

Indexing terms: cerebellothalamic, lemnothalamic, thalamic nuclei, anatomy, dextrans

It is generally agreed that the cerebellar and dorsal column nuclei project mainly to separate functional subdivi- sions of the thalamus. Fibres from the cerebellar nuclei are known to terminate predominantly in the “motor thala- mus,” which comprises the ventrolateral nucleus (VL) in the rat and cat and the posterior subdivision of the VL complex, or VLp, in the monkey. The main target of lemniscal projections from the gracile and cuneate nuclei is the lateral part of the ventral posterior complex or “sensory thalamus” comprising the ventral posterior lateral nucleus (VPL) in rat and cat and the caudal subdivision (VPLc) of the VPL complex in monkey (for a review, see Jones, 1985). However, the question as to whether or not cerebellar and lemniscal fibres converge onto thalamic neurones in the border zone between motor and sensory regions is unre- solved. Findings from some studies using anterograde

tracing have indicated that the projections are likely to be completely separate (Hendry et al., 1979; Tracey et al., 1980; Kalil, 1981; Asanuma et al., 1983), whereas others have suggested that there may be convergence (Cohen et al., 1958; Angaut, 1970; Angaut and Bowsher, 1970; Ha- roian et al., 1981; Berkley, 1983). More recently, we have shown that the interposed cerebellar nuclei provide a substantial projection to the cytoarchitecturally defined rostral pole of VPL in the rat, indicating the possibility of overlap with lemniscal projections (Aumann et al., 1994).

One way of resolving the problem of overlap is to label both projections in the thalamus simultaneously, and, in

Accepted December 18,1995. Address reprint requests to Dr. T.D. Aumann, Department of Anatomy,

Monash University, Wellington Road, Clayton, Victoria, Australia, 3168.

o 1996 WILEY-LISS, INC.

CEREBELLAR AND LEMNISCAL INPUTS TO THALAMUS 609

the present study, we have used a double anterograde tracing technique in the rat which has enabled us to compare these projections directly. The results show that the cerebellar and dorsal column nuclei do indeed project to common regions of thalamus, but the terminals from the two sources generally do not overlap.

MATERIALS AND METHODS This study was carried out on 14 adult (250-400 g) male

Sprague-Dawley rats that were deeply anaesthetised with sodium pentobarbitone (60 mgikg, i.p.1 for the surgical procedures. Eleven animals were used in double antero- grade labelling experiments in which bilateral injections of dextran-tetramethylrhodamine (DT; Schmued et al., 1990) and dextran-fluorescein (DF; Aumann et al., 1994) from Molecular Probes Incorporated were made into the cerebel- lar and dorsal column nuclei. The tracers were dissolved in 0.1 M neutral phosphate buffer (PB) at a concentration of 10% for the DT and 15% for the DF. Bilateral injections of 10% dextran-biotin (DB; Molecular Probes Inc.) were also made in three animals to compare the locations of cerebel- lar and lemniscal terminals with thalamic cytoarchitecture in counterstained sections. Tracers were pressure injected from calibrated glass micropipettes (tip diameter = 10-20 pm) with a picospritzer I1 (General Valve Corporation).

Details of the techniques for localising and injecting the tracers into the cerebellar nuclei have been published elsewhere (Aumann et al., 1994). In eight of the animals, a single injection of about 100 nl was made into the nucleus interpositus, and this was combined with injections aimed for one of the dorsal column nuclei to enable us to compare cerebellar projections with those from either the gracile or main cuneate nucleus. In three animals, we attempted to fill the entire extents of both the interposed and dorsal column nuclei with tracer. In these cases, two larger injections of one tracer (about 200 nl, each separated by about 0.5-1.0 mm in the mediolateral plane) were made into the interpositus, and these were combined with mul- tiple injections of the other tracer to encompass both the gracile and main cuneate nuclei. We concentrated on the interposed nuclei, as we have previously shown that termi- nals from the lateral cerebellar nucleus do not project to the VPL region in the rat (Aumann et al., 1994).

The dorsal column nuclei were approached dorsally via the foramen magnum, and they were identified by the tubercules formed on the surface of the medulla. Two or three small injections (about 40-50 nl each) were made

about 300 pm apart in the rostral-caudal direction to label a substantial proportion of either the gracile or main cuneate nucleus or both nuclei in some cases.

In pilot studies, we examined axonal transport rate with a range of survival times to assess the suitability of the tracers for double labelling. Following injections into the rat dorsal column nuclei, labelling by either tracer was present in lemniscal axons below the thalamus by 3 days, and clearly labelled terminals were evident in the thalamus by 4 days. We estimated that both tracers are transported at about 4 mm/day, and their fluorescence did not appear to diminish with survival periods as long as 15 days. The transport distances to the thalamus in the present animals were about 15 mm and 12 mm from the dorsal column and cerebellar nuclei, respectively, and survival periods of be- tween 5 and 10 days were used to ensure complete filling of the t halamic terminals.

After the survival period, the animals were anaesthetized with sodium pentobarbitone (60 mg/kg, i.p.) and perfused via the aorta with 300 ml of warm heparinised 0.1 M neutral phosphate buffered saline (PBS), followed by 500 ml of cold 4% paraformaldehyde in PB, and finally with 500 ml of cold 4% paraformaldehyde and 10% sucrose in PB. The brains were removed and left overnight at 4°C in 4% paraformaldehyde and 20% sucrose in PB.

The cerebellum, brainstem, and diencephalon were sec- tioned at 50 km in the coronal plane on a freezing micro- tome. Sections for fluorescence microscopy were washed in PBS, mounted serially on gelatinised slides, and examined mostly with a Leitz fluorescence microscope using excita- tion wavelengths of 540 nm and 490 nm for the DT and DF, respectively. In later experiments, we used a Nikon micro- scope fitted with a multiple bandpass filter (61002, Chroma Technology Corp.), which enabled the two tracers to be viewed and photographed simultaneously.

The sections from animals injected with DB were incu- bated with avidin-peroxidase and processed for horseradish peroxidase histochemistry using the diaminobenzidine (DAB) reaction (for details, see Aumann et al., 1994). These sections were counterstained with neutral red, dehydrated, cleared and coverslipped, and examined under brightfield illumination.

A microscope stage digitising system (MD-2 Minnesota Datametrics Corporation) was used to map the outlines of the sections and the locations of the injections and contra- lateral labelled terminals. The injection sites were defined as the region where neurones were intensely stained with tracer. Following data plotting and photography, the fluores-

AIP AM APT AV cf CL DAB DB DF DLG DT EC fr Gr L LD LP

anterior interposed cerebellar nucleus anterior medial nucleus anterior pretectal area anterior ventral nucleus cuneate fasciculus central lateral nucleus diaminobenzidine dextran-biotin dextran-fluorescein dorsal lateral geniculate nucleus dextran-tetramethylrhodamine external cuneate nucleus fasciculus retroflexus gracile nucleus lateral cerebellar nucleus lateral dorsal nucleus lateral posterior nucleus

Abbreviations

MC ml PBS PC PF PIP Po R SCP VL VLG VM VPL VPM VPPC 21

main cuneate nucleus medial lemniscus phosphate buffered saline paracentral nucleus parafascicular nucleus posterior interposed cerebellar nucleus posterior nuclear group reticular nucleus superior cerebellar peduncle ventral lateral nucleus ventral lateral geniculate nucleus ventral medial nucleus ventral posterior lateral nucleus ventral posterior medial nucleus ventral posterior nucleus (parvocellular) zona incerta

610 T.D. A U M A ” ET AL.

Fig. 1. Dextran-biotin labelled cerebellar and dorsal column nuclei projections. A: Injection into the anterior interposed nucleus (top) and resulting labelled terminals in the rostral pole of the ventral posterior lateral nucleus (bottom) at about the level of section 7 in Figure 5. B: Injection into the gracile nucleus (top) and resulting labelled terminals

in VPL (bottom) at about the level of section 5 in Figure 5. C: Injection into the main cuneate nucleus (top) and resulting labelled terminals in VPL (bottom) at about the level of section 5 in Figure 5. In A-C, arrowheads indicate labelled terminals and insets are higher magnifica- tions of terminals in squares. See Abbreviations list.

cent-labelled sections were counterstained with neutral red. They were then superimposed on the computer plots with a projection microscope, and the outlines of the nuclei were traced onto the mapped sections.

RESULTS Identification of nuclei and shape of terminals

The thalamic subdivisions outlined in the diagrams that follow were identified according to the cytoarchitectural descriptions of Jones (1985). Where we were unsure of the precise boundaries between nuclei, the division is indicated only tentatively by interrupted lines as was the case for the border between the caudal part of the ventral lateral nucleus (VL) and the posterior complex (Po), between parts of the ventral posterior medial (VPM) and ventral posterior lateral (VPL) nuclei, and between rostral VL and VPL.

It is generally accepted that VPL is characterised by the presence of numerous bundles of penetrating fibres and a rather patchy distribution of neurones. According to these criteria, VPL in the present animals extended along the

entire lateral aspect of the thalamus. However, evidence will be presented which suggests that the rostralmost part of VPL (see Fig. lA, section 6 in Fig. 4A, section 7 in Fig. 5, and section 7 in Fig. 7) is actually a part of the “cerebellar thalamus” and does not receive lemniscal afferents. But for the present diagrams, we have maintained our cytoarchitec- tural criteria and labelled it as VPL.

Figure 1 shows photomicrographs of injections of DB into the interposed, gracile and main cuneate nuclei and ex- amples of anterogradely labelled terminals in the contralat- era1 thalamus.

In the main projection areas of VL and VPL and in the rostral part of Po, cerebellar and lemniscal terminals tended to be located in clusters that were concentrated around the somata of thalamic neurones. Terminals from both sources were generally large (up to 10-15 p,m long), elongated varicosities that were formed mostly at the ends of fibre branches. Cerebellar and lemniscal terminals were also located in other regions of the diencephalon such as the zona incerta, parafascicular nucleus, and pretectum. In these regions, terminals from both sources were generally

CEREBELLAR AND LEMNISCAL INPUTS TO THALAMUS 611

much smaller than in VL, VPL, and rostral Po. They were mostly oval or round in shape with diameters of 1-2 pm, and many varicosities were formed en passant along fibre branches that coursed through the neuropil. Some photomi- crographs of these terminals are shown in Figure 2.

Double labelling experiments The combinations of successful double injections of fluo-

rescent tracers are illustrated schematically on standard diagrams in Figure 3, where the cross hatching indicates regions where neurones were intensely stained by the tracers. The injections in the interposed nuclei are shown in a single horizontal section, and those in the dorsal column nuclei are shown on a dorsal view of the medulla. Generally, these injections extended throughout the entire depth of each nucleus. Colour photomicrographs of fluorescent tracer injections are shown in Figure 6A-B.

We will describe first the results from the series of injections in eight cases (r61-r21 in Fig. 3) in which tracer was largely or exclusively confined to one of the dorsal column nuclei.

Figure 4A presents sample plots of the locations of terminals in the contralateral diencephalon for case r41 in Figure 3, where the injections encompassed the entire extent of the main cuneate nucleus and the medial portion of the anterior and posterior interposed nuclei. In these and

subsequent plots, the symbols (triangles for the cerebellum and circles for the dorsal column nuclei) indicate the location of clusters of terminals in VL or VPL (as shown in Fig. 11, and the main concentrations of terminals in other regions are shown with corresponding symbols. In Figure 4A, it can be seen that lemniscal fibres were distributed mostly along the medial aspect of VPL, but they did not extend into the rostral-most portion of the nucleus. The cerebellar terminals labelled by the injection into the interpositus were located mostly in the rostral part of VL and then extended rostrally and laterally around the lemnis- cal receiving area into the rostral-most portion of VPL. Throughout the VL/VPL complex terminals from the two sources were well separated except in the rostral-most part of the lemniscal receiving area where a few patches of the two sorts of terminals were in close proximity.

With injections in the gracile nucleus (Fig. 4B shows plots from case r61), lemniscal terminals were located mostly laterally in VPL and came close to cerebellar terminals only in rostral VPL. A similar picture emerged from the other six sets of injections in this series, in that terminals from the cerebellum and dorsal column nuclei were clearly separated except in the rostral border between the cerebellar and lemniscal projection areas. The findings for these cases are summarised in Figure 5, where labelling resulting from the different injections has been transferred to standard dia-

Fig. 2. Dextran-biotin labelled cerebellar and dorsal column nuclei terminals in other parts of the diencephalon. A Examples of terminals from the interposed nuclei in the pretectum. B: Gracile terminals in the pretectum. C: Terminals from the interposed nuclei in the zona incerta. D: Gracile terminals in the zona incerta.

r6r

r4r

r581

r58r

r41

r2r

RI

1231

R3r

a01

1-30 r

1 mm

EC

MC

t obex

Gr 1 mm

DT

DF lateral

Fig. 3. Diagrammatic representations of injections made into the interposed and dorsal column nuclei in double labelling experiments. Left: Horizontal section through the lateral, anterior interposed, and posterior interposed nuclei. Right: Dorsal surface of the medulla showing the gracile, main cuneate, and external cuneate nuclei. Shaded areas indicate the location and extent of the injections of dextran-tetramethylrhodamine and dex- tran-fluorescein.

Al.

(ca

udal

) ,

4.

\

2.

81.

6.

3.

5.

#I

2.

7.

(ros

tral

)

-

1 m

m

Fig.

4.

A:

Lw

tio

n o

f ter

min

als l

abel

led

from

the

inte

rpos

ed a

nd m

ain

cune

ate

nucl

ei (1

-7,

left

-han

d si

de). B: T

he in

terp

osed

and

gra

cile

nuc

lei

(1-6

, rig

ht-h

and

side

). T

he in

ject

ion

site

s ar

e sh

own

in c

ases

r41

and

r61

, res

pect

ivel

y, o

f Fig

ure

3. T

hese

dia

gram

s (a

nd th

ose

of F

ig. 7

)

3.

(cau

dal)

L? V

PM

,I

4. 5.

6.

(rost

ra11

indi

cate

lab

ellin

g in

sam

ple

coro

na!

sect

ions

, ea

ch 0

.5 m

m a

part

, thr

ough

the

con

trda

tera

l di

ence

phal

on. C

ircl

es in

dica

te le

mn

id te

rmin

als,

and

tria

ngle

s ind

icat

e cer

ebel

lar

term

inal

s.

614 T.D. AUMANN ET AL.

grams of the diencephalon.' I t can be seen that cerebellar and lemniscal fibres share only a rather limited common projection zone around the rostral VLIVPL border (section 6 in Fig. 5). However, in this common projection region, terminals from the two sources were nearly always concen- trated in separate patches as can be seen from the photomi- crographs of Figure 6C-H. Quite often, such patches of cerebellar and lemniscal terminals could be traced through two to five successive sections or about 100-250 pm in the rostral-caudal direction, and they appeared to be organised into separate interdigitating rods or short columns. It was also notable that cerebellar and lemniscal projections to other diencephalic nuclei such as the parafascicular nuclei, anterior pretectal area, and zona incerta were also largely concentrated in separate areas, and there was only a small amount of overlap, generally where projections from the two sources were sparse.

Although the series of injections just described covered the entire extent of the interposed and dorsal column nuclei, none of the nuclei were completely filled with tracer in any one case, and it is possible that some terminal overlap in the thalamus may have been overlooked. An attempt was therefore made to saturate each nucleus with tracer, and the location of these injection pairs are shown as the last four cases (r231-r30r) of Figure 3.

Each of these cases confirmed the impression that cerebel- lar and lemniscal fibres terminate in essentially exclusive zones within the thalamus, and plots and photomicro- graphs of labelling are shown in Figures 7 and 8 for cases r301 and r30r, where there was complete or almost com- plete filling of both nuclei with tracers. It can again be seen from Figure 7 that cerebellar and lemniscal terminals come close together only at the rostral border of the lemniscal receiving area. This is also apparent in the photomicro- graphs of Figure 8, which were taken through a multiple- band filter, and both projections can be compared directly in sections through different levels of the thalamus. Figure 8A-B shows labelling at about the mid-rostrocaudal levels of VL and W L , respectively, and it is evident that no lemniscal fibres or terminals (labelled with orange-red) enter the cerebellar receiving area. Likewise, no cerebellar fibres project into the lemniscal zone at this level of the thalamus. Figure 8D shows a more rostral section through the VLIVPL border, where fibres from the two sources start to come together. At the rostral border between the cerebellar and lemniscal receiving areas (Fig. 8C,E), the two groups of fibres terminate in common projection zones, but the terminals remain largely separate. In Figure 8C, a few green-labelled cerebellar terminals can be seen scattered amongst those from the dorsal column nuclei in the upper half of the photomicrograph, but this represents an ex- ample of the largest extent of intermingling seen in the present experiments. Figure 8F indicates the rostrolateral part of the lemniscal zone, where patches of cerebellar terminals are interspersed with some from the gracile nucleus.

'The caudal part of the external cuneate nucleus was involved in two injections (r2r and r21 in Fig. 3). However, control injections localised to this nucleus (the results of which will be reported separately) indicated that it provides very few terminals to VF'L in comparison with the main cuneate nucleus and is unlikely to be a major source of contamination.

DISCUSSION In the present study, we have used two anterograde

tracers to label simultaneously efferent fibres from the interposed and dorsal column nuclei to assess whether or not lemniscal and cerebellar projections overlap in the thalamus. In the main receiving areas of VL and W L in the rat, we found no evidence that terminals from the two sources converged, and there was only a limited region around the rostral border of the lemniscal termination zone that received a common projection. However, even in this region, very few of the cerebellar and lemniscal terminals intermingled, and they were mostly arranged in short, rostrocaudally oriented columns that interdigitated.

Although we initially identified VPL as extending along the entire lateral aspect of the thalamus, the anterograde labelling results showed that lemniscal fibres do not end in the rostral part of this region, which seems to be more appropriately classified a part of motor thalamus on the basis of its inputs from the interposed nuclei.

The present findings thus indicate that cerebellar and lemniscal fibres share a common projection region in VPL around its border with VL. This is in accord with observa- tions by Berkley (1983), which suggest that a similar situation exists in W L in the cat and monkey. In these experiments, Berkley (1983) used a combination of autora- diographic and degeneration tracing techniques that did not permit cerebellar and lemniscal terminals to be exam- ined in the same sections. However, she did deduce that there was unlikely to be much overlap between terminals from the two sources. This is confirmed by the present finding in the rat, which showed that cerebellar and lemniscal terminals are indeed almost entirely separate within the common projection zone in VPL.

We did, however, find that clusters of cerebellar and lemniscal terminals in this part of VPL were often located quite close to one another (sometimes within 50 p,m; see Figs. 6, 81, and this raises the question as to whether they might contact common thalamic neurones. In the rat, the dendritic arborisations of VL neurones that receive cerebel- lar input extend about 215 p,m into the frontal plane (Yamamoto et al., 1991). If we assume the dimensions are similar for neurones in rostral VPL, then it is quite possible for the dendritic trees of individual cells to span patches or rods of both cerebellar and lemniscal terminals. However, even allowing for wider dendritic trees for these neurones, a major degree of convergence seems very unlikely as the large terminals from the cerebellum make contact largely or exclusively with the proximal dendrites of VL/VPL neurones (Aumann et al., 1994) and lemniscal terminals in VPL also synapse proximally on thalamocortical neurones (Lieberman and SpaEek, 1971; Ralston, 1983; Jones, 1985). I t thus appears likely that thalamic neurones in the border zone will be influenced predominantly by one or the other input.

The smaller terminals supplied by cerebellar and lemnis- cal afferents to other regions of the diencephalon (chiefly the zona incerta, parafascicular nucleus, and the pretec- tum) were also mostly segregated, and, again, it seems probable that they supply separate populations of neu- rones. However, this conclusion can only be regarded as tentative at present because it is not known where these terminals contact their target neurones or how far the dendrites of these neurones extend across the nuclei.

CEREBELLAR AND LEMNISCAL INPUTS TO THALAMUS 615

- 1 mm

EC

MC

Gr

H 1 mm

Fig. 5. Summary diagram of the main areas of termination of cerebellar and lemniscal afferents in the diencephalon. The top left-hand diagrams indicate the total areas of the interposed and dorsal column nuclei encompassed by the series of injections. Sections 1-7 indicate labelling transferred to a series of standard diagrams (each 0.5 mm apart) through the contralateral diencephalon.

Fig. 6. Fluorescent tracer injections and labelled terminals in the rostra1 VL/WL border (coronal sections). A Example of an injection of dextran-tetramethylrhodamine into the posterior interposed nucleus. B: Dextran-fluorescein injection into the main cuneate nucleus. C-D,

E-F, and G-H are pairs of photomicrographs of the same areas of the thalamus. The red-labelled terminals are from the interpositus, and the green terminals are from the cuneate. The arrowheads indicate blood vessels.

CEREBELLAR AND LEMNISCAL INPUTS TO THALAMUS

Fig. 7. Locations of labelled terminals

617

7. (rostral)

from case r30r.

618 T.D. AUMANN ET AL.

Fig. 8. Dual wavelength fluorescence photomicrographs of thalamic labelling. Cerebellar projections were labelled with DF and fluoresce green and the DT-labelled lemniscal fibres and terminals appear orange-red. A,B: Labelling at the mid rostrocaudal level of VL and W L , respectively. C: Rostromedial level of the lemniscal receiving zone.

D. Slightly more caudal section at point where cerebellar and lemniscal fibres and terminals start to come together. E: Another example at the rostromedial lemniscal area. F: Rostrolateral border of the lemniscal receiving area. Examples are taken from case 1-301. Dorsal is up and lateral to the right in each section.

The regions of VPL which received projections from both the cerebellar and dorsal column nuclei in the present animals appears to correspond with the proprioceptive receiving area of the thalamus in the monkey and cat. This consists of a shell around rostral VPL (cf. the “deep shell” of Friedman and Jones, 1981) where neurones respond predominantly to input from muscles and joints (Anderson et al., 1966; Rosen, 1969; Landgren and Silfvenius, 1970; Maendly et al., 1981) and which projects to areas 3a and 2 of

somatosensory cortex (see Jones, 1985). In contrast, the deeper parts or “central core” of VPL receives cutaneous input and projects to cortical areas 3b and 1 (see Jones, 1985). It is not yet clear whether VPL in the rat is similarly organised into specialised proprioceptive and cutaneous receiving areas, and subdivisions of somatosensory cortex that are both anatomically and functionally distinct seem to be less clear-cut in this species than in the cat and monkey. Available evidence suggests there is extensive convergence

CEREBELLAR AND LEMNISCAL INPUTS TO THALAMUS 619

of muscle and cutaneous inputs onto cells in the cortical sensory representation of the forelimb, but there does appear to be a purely proprioceptive receiving zone that overlaps with part of the motor representation (Lamour and Jobert, 1982; Angel and Banks, 1983; Chapin and Lin, 1984; Gioanni, 1987). However, the thalamic connections of these subregions have not yet been determined. For the hindlimb area, the study by Hall and Lindholm (1974) indicated that there is almost complete overlap between the motor and sensory representations. Indeed, Donoghue et al. (1979) found labelled cells in both VL and VPL following injections of horseradish peroxidase into the centre of the hindlimb area, which is consistent with the notion that the hindlimb is represented as a sensorimotor amalgam in the cerebral cortex. However, a more recent study by Hum- melsheim and Wiesendanger (1985) suggests that overlap in the hindlimb area may be far less than previously supposed. These workers found that cells responsive to hindlimb muscle stretch were concentrated in granular cortex just caudal to the region where intracortical micro- stimulation evoked movements at lowest threshold, and there appeared to be only a very limited region of overlap between these motor and proprioceptive zones. In view of these findings, it seems possible that some of the labelling observed by Donoghue et al. (1979) in VL and VPL could have arisen from separate thalamic projections to adjacent sensory and motor representations, and it would be of interest to re-examine the connections of the rat sensorimo- tor cortex with injections localised to the potentially differ- ent functional subregions.

ACKNOWLEDGMENTS We thank Michelle Mulholland for her valuable assis-

tance in preparing the plates. This study was supported by a grant from the National Health and Medical Research Council of Australia.

LITERATURE CITED Andersson, S.A., S. Landgren, and D. Wolsk (1966) The thalamic relay and

cortical projection of group I muscle afferents from the forelimb of the cat. J. Physiol. 183.576-591.

Angaut, P. (1970) The ascending projections of the nucleus interpositus posterior of the cat cerebellum: An experimental anatomical study using silver impregnation methods. Brain Res. 24:377-394.

Angaut, P., and D. Bowsher (1970) Ascending projections of the medial cerebellar (fastigialj nucleus: An experimental study in the cat. Brain Res. 24:49-68.

Angel, A,, and D. Banks (1983) The functional organization of the forepaw sensorimotor cortex in the adult rat. J. Physiol. 343352-82.

Asanuma, C., W.T. Thach, and E.G. Jones (1983) Distribution of cerebellar terminations and their relation to other afferent terminations in the ventral lateral thalamic region of the monkey. Brain Res. Rev. 5237- 265.

Aumann, T.D., J.A. Rawson, D.I. Finkelstein, and M.K. Horne (1994) Projections from the lateral and interposed cerebellar nuclei to the

thalamus of the rat: A light and electron microscopic study using single and double anterograde labelling. J. Comp. Neurol. 349.165-181.

Berkley, K.J. (1983) Spatial relationships between the terminations of somatic sensory motor pathways in the rostra1 brainstem of cats and monkeys. 11. Cerebellar projections compared with those of the ascend- ing somatic sensory pathways in lateral diencephalon. J. Comp. Neurol. 220:229-251.

Chapin, J.K., and C.S. Lin (1984) Mapping the body representation in the SI cortex of anaesthetised and awake rats. J. Comp. Neurol. 229:199-213.

Cohen, D., W.W. Chambers, and J.M. Sprague (1958) Experimental study of efferent projections from the cerebellar nuclei to the brainstem of the cat. J. Comp. Neurol. 109:233-259.

Donoghue, J.P., K.L. Kerman, and F.F. Ebner (1979) Evidence for two organizational plans within the somatic sensory-motor cortex of the rat. J. Comp. Neurol. 183t647-664.

Friedman, D.P., and E.G. Jones (1981) Thalamic input to areas 3a and 2 in monkeys. J. Neurophysiol. 45:59-85.

Gioanni, Y. (1987) Cortical mapping and laminar analysis of the cutaneous and proprioceptive inputs from the rat foreleg: An extra- and intra- cellular study. Exp. Brain Res. 67.510-522.

Hall, R.D., and E.P. Lindholm (1974) Organization of motor and sensory neocortex in the albino rat. Brain Res. 66:23-28.

Haroian, A.J., C. Massopust, and P.A. Young (1981) Cerebellothalamic projections in the rat: An autoradiographic and degeneration study. J. Comp. Neurol. 197317-236.

Hendry, S.H.C., E.G. Jones, andd. Graham (1979) Thalamic relay nuclei for cerebellar and certain related fibre systems in the cat. J. Comp. Neurol. 185679-714.

Hummelsheim, H., and M. Wiesendanger (1985) Is the hindlimb representa- tion of the rat’s cortex a “sensorimotor amalgam”? Brain Res. 346: 75-81.

Jones, E.G. (1985) The Thalamus. NewYork: Plenum Press. Kalil, K. (1981) Projections of the cerebellar and dorsal column nuclei upon

the thalamus of the rhesus monkey. J. Comp. Neurol. 195t25-50. Lamour, Y., and A. Jobert (1982) Laminar distribution and convergence of

deep and superficial peripheral inputs in the forelimb representation of rat SI somatosensory cortex. J. Physiol. (Paris) 78.158-162.

Landgren, S., and H. Silfvenius (1970) The projection of group I muscle afferents from the hindlimb to the contralateral thalamus of the cat. ActaPhysiol. Scand. 80:lOA.

Lieberman, A.R., and J. SpaEek (1971) Synaptic glomeruli in the thalamus of the rat: Three-dimensional relationships between glomerular compo- nents. Experientia (Basel) 27:788-789.

Maendly, R., D.G. Ruegg, M. Wiesendanger, R. Wiesendanger, J. Lagowska, and B. Hess (1981) Thalamic relay for group I muscle afferents of forelimb nerves in the monkey. J. Neurophysiol. 461901-917.

Ralston, H.J. I11 (1983) The synaptic organisation of the ventrobasal thalamus in the rat, cat and monkey. In G. Macchi, A. Rustioni, and R. Spreafico (edsj: Somatosensory Integration in the Thalamus. Amster- dam: Elsevier, pp. 241-250.

Rosen, I . (1969) Excitation of group I activated thalamocortical relay neurones in the cat. J. Physiol. 205237-255.

Schmued, L., K. Kyriakidis, and L. Heimer (1990) In uzuo anterograde and retrograde axonal transport of the fluorescent rhodamine-dextran- amine, Fluoro-Ruby, within the CNS. Brain Res. 526r127-134.

Tracey, D.J., C. Asanuma, E.G. Jones, and R. Porter (1980) Thalamic relay to motor cortex: Afferent pathways from brainstem, cerebellum and spinal cord in monkeys. J. Neurophysiol. 44.532-554.

Yamamoto, T., Y. Kishimoto, H. Yoshikawa, and H. Oka (1991) Intracellular recordings from rat thalamic VL neurones: a study combined with intracellular labelling. Exp. Brain Res. 87245-253.