The ipsilateral cortico-cortical connexions between the cytoarchitectonic subdivisions of the primary somatic sensory cortex in the monkey

Brain Research Reviews, 9 (1985) 67-88

67

Elsevier BRR 90026

The Ipsilateral Cortico-Cortical Connexions Between the Cytoarchitectonic Subdivisions of the Primary Somatic Sensory Cortex in the Monkey M. F. SHANKS, R. C. A. PEARSON and T. P. S. POWELL department

of ~~rnan Anatomy,

university

of Oxford, South

Parks Road, Oxford OXI 3QX (il. K.)

(Accepted December 18th, 1984) Key words: cortico-cortical

connexions - cytoarchitectonic

subdivisions-somatic

sensory cortex -monkey

CONTENTS 1. Introduction

.............................................................................................................................................

2. Material and Methods

.................................................................................................................................

67 68

3. Results ..... ...............................................................................................................................................

70

4. Discussion ................................................................................................................................................

80

5. Summary .................................................................................................................................................

86

AcknowIedgement References

........... ..............................................................................................................................

...................................................................................................................................................

1. INTRODUCTION In an earlier

study,

the ~~oarchit~~tural

subdivi-

sions of the primary somatic sensory cortex (Sl) of the monkey were found to be interconnected by ipsi-

lateral cortico-cortical fibres, and these connexions were considered to link neurons with different functional properties within the same topographical representation*o. Several questions about the organization of these connexions have been raised by observations made in recent studies, and they are of sufficient interest to require a re-investigation with more refined methods. The arrangement of the callosal connexions of Sl has been shown to be arranged in a manner more intricate than was formerly thought16JsJ3,a; there are accentuations at the architectonic bounda~es at the representations of the midline and proximal limbs and all the separate repCorrespondence:

R. C. A. Pearson, Department

86 87

resentations of a part of the body are connected, through heterotopical projections, acrbss the midline@. There is therefore the possibility that the ipsilateral connexions may also have some special relationship to the boundaries and that all the multiple representations that have been described in Sl may be interconnected. The differences in functional properties of the cells in the various architectonic subdivisions have been known for some timea, and there is now evidence that the cells in progressively more posterior parts of Sl respond to increasingly complex stimuli, with large and complicated receptive fields, submodality convergence and direction selectivityiJJ3. The posterior parts of Sl, areas 1 and 2, receive few fibres from the thalamus and only areas 3a and 3b receive a heavy thalamic projectionlSJlJ9. The complex properties of the cells in areas 1 and 2 have therefore been attributed to intra-

of Human Anatomy, South Parks Road, Oxford OX1 3QX

016%0173/85/$03.3O@J1985 Elsevier Science Publishers B.V. (Biomedical Division)

68 2. MATERIAL AND METHODS

The brains of 20 macaque monkeys have been used in this study and many of them have also been used in concurrent

studies on the commissural

connexions

of

the somatic sensory cortex (Sl) and the projection of this sensory area upon area 5 (refs. 30,31,38,40). All the operations were done under Nembutal anaesthesia and with strict aseptic precautions. Most of the medio-lateral extent of the postcentral gyrus of one hemisphere

was exposed

and

small

lesions

were

made in Sl by one of two methods. On the lateral surface of the post-central gyrus, an area of cortex

Fig. 1. The sites of 3 microelectrode penetrations on the lateral surface of the postcentral gyrus (above), and the local intracortical fibre and terminal degeneration in a sagittal section of the cortex at the level of the penetration at A (below). The lesion is in area 1 and goes through the full depth of the cortex. In this and subsequent figures the lesion is shown in solid black, the fibre degeneration by short lines and the terminal degeneration by dots. The laminae of the cortex are indicated by Roman numerals on the right of the figure and the cytoarchitectonic subdivisions by Arabic numerals on the surface of the section with the boundaries marked by arrows. Note that the degeneration extends appreciably further posteriorly than anteriorly. CS, central sulcus.

cortical processing of sensory information initially reaching area 3b (ref. 13); successive stages of the processing occur in different architectonic subdivisions which are linked together by cortico-cortical connexions. It is possible that these cortico-cortical fibres form the major afferent pathway to areas 1 and 2, particularly in the distal limb representations which are devoid of callosal connexionsi6J*J*,40. Finally, a more detailed knowledge of these inter-area1 connexions has been necessary for the interpretation of the findings in a study of the projection of Sl upon area 5 (refs. 30, 31), and it would appear that any transformation of sensory information that underlies the differences in functional properties between the cells of this area of association cortex and those in the primary somatic sensory cortex begins within areas 1 and 2. A preliminary report of these findings has been published37 and accounts of other similar studies have appearedi7,46.

l-2 mm in diameter was devascularized by removal of the overlying pia-arachnoid or an insulated tungsten microelectrode was inserted and a current of 5-10 ,uA passed during the traverse of the electrode through the cortex. The electrode penetrations were made under a dissecting microscope and care was taken to make them between blood vessels and as perpendicular to the surface as possible. The electrodes were inserted to a depth of l-2 mm for lesions in areas 1 and 2, but in attempts to selectively involve area 3, the penetrations were deeper and up to 4 mm. With each type of lesion, two or three were made in most brains but they were widely separated along the medio-lateral extent of the gyrus. After survival periods of 3-5 days, the animals were again deeply anaesthetized and were perfused through the heart with 0.9% saline and 10% formalin, and the brains were removed and immersed in the fixative for several weeks. The brains were photographed to record the sites of the lesions and the sulcal pattern. The cortex of the pre- and postcentral gyri and the superior and inferior parietal lobules of the operated hemisphere was cut into two or three blocks and 25 pm sections cut on a freezing microtome in the sagittal plane. A 1 : 10 series was stained with the Fink-Heimers or Wiitanen47 technique and where necessary closer series were stained, particularly in the case of brains with microelectrode lesions. An alternate series of sections was stained with thionin. Outlines of the sections were drawn on a projection apparatus and the precise site and extent of the lesion and the distribution of the degeneration were reconstructed upon tracings of the photographs of the brains and upon orthogonal reconstructions of the architectonic subdivisons of Sl.

Fig. 2. Photomicrograph x50.

of the same microelectrode

lesion as shown in the section in the previous figure. Dark-field illumination.

by direct recording. and the degeneration

The relationship

of the lesions

to the architectonic

subdivi-

sions was determined by examination of the alternate thionin-stained sections. After each lesion, there was degeneration at several sites in the ipsilateral cortex, in the immediate vicinity of the lesion, in other parts of the first somatic sensory area (Sl), in areas 4, 5 and the second somatic sensory area @II). While the distribution of the degeneration away from the lesion differed depending upon the architectonic subdivision or area of representation of the degeneration

involved,

the pattern

in the cortex immediately

adjoin-

ing the damage was always essentially the same. This local degeneration will be described first, then the interconnections between the architectonic subdivisions of SI and finally the pattern of degeneration in the different topographic subdivisions. The connexions with areas 4 and 5 will be considered in separate papers.

Fig. 3. The local fibre and terminal degeneration in sagittal sections of the cortex at the levels of the microelectrode lesions shown at B and C in Fig. 1. The lesion in B is in area 3b of the leg representation and that in C passes down the posterior wall of the central sulcus through areas 1 and 3b.

3. RESULTS

The experimental material is made up of 16 small lesions of l-2 mm extent and 28 lesions made with a microelectrode; all of the architectonic areas and the different topographical subdivisions have been involved. As in the accompanying papers, the identification of the representation involved by a lesion was made by reference to the published maps from evoked potential and microelectrode studies and not

Local intracortical connexions A microelectrode lesion confined to the cortex of area 1 and perpendicular to the surface in the representation of the hand is representative of the cortex on the exposed surface of the postcentral gyrus (Fig. 1). Immediately around the whole depth of the track, which is approximately 100 pm wide, is a narrow zone of dense granular degeneration and beyond it much less dense fibre and terminal degeneration extend for a few millimetres; there are a few degenerating fibres in all layers, most of which are fine and orientated in all directions. In the deep part of layer III, the adjoining part of layer IV and the deeper part of layer V there are more fibre fragments and these are coarser. These fibres are clearly running horizontally and are grouped to form distinct bands associated with light terminal degeneration. The granular degeneration in layer I remains denser and extends further from the lesion than that in the other layers with the exception of the degeneration associated with the horizontally running bands of fibres. The extent of degeneration is different anterior and posterior to the lesion; posteriorly it extends for approximately 2 mm, whereas anteriorly it is found over only half of this distance. Both anterior and posterior to the lesion, the horizontal fibres reach further than the moderate degeneration in the other layers. On adjoining sections there are vertical degener-

Fig 4. Another microelectrode lesion in the cortex of area 1; the electrode penetration was slightly oblique as 1iayer IV. The bundle of degenerating efferent fibres can be clearly seen passing perpendicularly through the ‘white matter at the lower margin of the photomicrograph. Dark-field illumination. x90.

and only reached as deep layers V and VI to reach

72 ating fibres restricted

to the width of the dense de-

generation. After

perpendicular

penetrations

areas 3b and 3a, close to the medial hemisphere

confined margin

(Fig. 3), the fibre and terminal

ation within the cortex adjacent

to

of the

degener-

to the lesion is simi-

lar to that in areas 1 and 2. The horizontal bres in layers III, IV and V are particularly

bands of ficlear and

to the deeper layers. The degeneration

in layers III

and IV superficial to a lesion is restricted a few hundred pm and this appearance

to a width of may reflect

the

descriptions

of Golgi-impregnated

filled axons from the thalamus

and

branching

HRP

to form a

dense plexus over an extent of 3-4 pm in these layers*“; the clear and sharp restriction of the dense degeneration

in these layers indicates

the precise and

follow a curved course ascending on either side of the shallow central sulcus. There is again asymmetry in

localized termination of these afferent fibres. The origin of most of the horizontally disposed fibres in

the extent of the degeneration posteriorly.

layers III, IV and V is in those layers themselves as degeneration of these fibres appears only when the

which extends further

As none of the blocks containing the electrode tracks has been cut coronally, it is not possible to give precise measurements of the extent of the local degeneration in this dimension. However, examination of adjoining sections shows that there is very little, if any, degeneration which can be related solely to fibres passing within the cortex from the lesion beyond a distance of approximately 1 mm on either side of the track. Thus, like area 4 of the motor cortex9 the extent of the degeneration is greater antero-posteriorly than medio-laterally. A number of penetrations passed vertically through the posterior bank of the central sulcus to a varying depth in the posterior wall, crossing the radial cell columns in area 1 obliquely and passing almost tangentially through either the superficial or deeper layers of area 3b. In some of these, the sections have been cut in the plane of the penetrations and most of the length of the track throughout the depth of the posterior wall appears on a single section (Fig. 3), while in others successively deeper portions of the track appear on adjoining sections. The sections containing the tracks cut in these different planes show aspects of the degeneration brought out by penetrations crossing the predominantly radial arrangement of the cortex. The very restricted vertically arranged degeneration is shown clearly; in sections where only part of the track is present, the degenerating fibres both superficial and deep to the lesion run strictly parallel to the radial columns of cells and coincide with the width of the damage. When the damage is in layers I, II and the superficial part of III, few fibres pass into the deeper layers but when the deep part of layer III is involved, a large number pass into the deeper layers and white matter. The number entering the white matter increase with increasing damage

portion

of the track in the section

is at the corre-

sponding level. When the track penetrates to near the fundus of the sulcus, numerous fibres can be seen running in layers V and VI into the rest of area 3b and 3a. Essentially the same distribution is found within a

1

Fig. 5. The degeneration within SI at the level of a lesion in area 1 near the lateral part of the postcentral dimple in the trunk representation. There is degeneration throughout virtually the entire antero-posterior extent of SI, but it is accentuated at the boundaries between the architectonic subdivisions and it is much lighter in area 3b than in the other areas.

9JptiOr -

bp

9 _med@ margtn

Fig. 6. TO show the medio-lateral spread of the degeneration within SI resulting from a microelectrode lesion in area 3b within the foot representation. The site of the lesion and the levels of the sagittal sections are shown on a drawing of the surface of the hemisphere (above left), and the extent of the degeneration on a planar reconstruction (below left). Note that the degeneration is mainly in the middle of the antero-posterior extent of the cytoarchitectonic areas and with little at the boundaries. In this and the subsequent figures only the ‘feed-forward’ degeneration has been placed on the planar reconstruction. CS, central sulcus; gs, cingulate sulcus; pcd, postcentral dimple.

few millimetres tents involved

of larger

lesions,

by the different

are not so clear because

types

of the more

although

the

ex-

of degeneration irregular

outline

of the damage

and the varying

ent layers of the cortex.

involvement

of differ-

74 Interareal connexions Further from a lesion in both the medio-lateral antero-posterior

dimensions

and

the pattern of degenera-

sensory area differed with the architectonic subdivision and area of representation which was damaged. The marked differences tern of termination

tion is different. Projections between different areas and between parts of the same area are marked by a

subdivisions

sudden increase in the density of degeneration

of a simple reciprocal

which

in both the intensity

of fibres between

indicate that the interconnexions

large lesions,

nature. l-2

and pat-

the different are not

as is clearly shown by

coincides with the appearance of degenerating fibres entering the deeper layers of the cortex from the

relatively

mm in extent,

white matter. The connexions made by fibres passing through the white matter to other parts of the somatic

‘extensive intracortical degeneration posterior to the damage, extending through all layers of area 2, ante-

the cortex of areas 1 and 2 (Fig. 5). Although

involving there is

+ medial.

2

Fig. 7. The degeneration within SI following a lesion in area 1 at the level of the lateral third of the postcentral dimple. At the level of the lesion the degeneration is continuous throughout the antero-posterior extent of SI, but more medially and laterally it extends along the boundaries between the cytoarchitectural subdivisions. ips, intraparietal sulcus.

Fig. 8. The distribution of the degeneration after a microelectrode lesion in area 2 in the lateral part of the arm representation. to the lesion the degeneration now extends along the middle of the cytoarchitectonic areas with little at the boundaries.

rior to the lesion the degeneration spreads for only a millimetre or so, bringing it to the boundary with area 3b. Within area 3b there are relatively few degenerating fibres entering the deep aspect of the cortex but there is fine terminal degeneration in both the supra- and infragranular layers. It is present throughout the depth of layers I and II and in the superficial half of layer III, and is diffusely scattered in layers V and VI. Over most of the antero-posterior extent of area 3a there is quite heavy fibre and terminal degeneration, and while all layers of the cortex contain some degeneration the terminal degeneration is clearly densest in layer IV and the deep part of layer III. The differences in the density and pattern of the degeneration are quite clearcut at the boundary of areas 3a and 3b. The projections of individual cytoarchitectonic areas is best shown by small microelectrode lesions

Medial

limited to one area. It is difficult to do this for area 3b except near the medial margin of the hemisphere, and in the representation of the leg, where the central sulcus is very shallow, and this area extends up on to the posterior bank of the sulcus. The electrode track shown in Fig. 6 has entered almost vertically, and runs down through the deep layers of the cortex in the posterior wall of the central sulcus. Anterior to the track there is degeneration in the middle of area 3a but the anterior and posterior margins of this area have very little; the terminal degeneration is heaviest in layer IV and the deep part of layer III but there is some fragmentation in all layers. Posteriorly, similar patches of fragmentation are seen in areas 1 and 2, the densest degeneration again being in layer IV and the deep part of layer III and with much less at the boundaries of these areas; there is less in area 2 than in area 1. In each of these 3 areas the degenerating fi-

76 bres clearly enter the cortex from the white matter as

degenerating

distinct from the local degeneration

from the white matter and there is fine granularity

lesion. The degeneration

surrounding

in area 3a anterior

the

to the le-

fibre is seen entering

layers I, II and the superficial

the deep layers in

part of layer III and dif-

whereas that in areas 1 and 2 is maximal lateral to the

fusely in layers V and VI. Further forwards, degeneration is seen in layer IV and the deep part of layer III

track. After similar lesions more laterally

throughout

sion increases

on sections

medial

to the damage, in area 3b,

in the trunk and face representations, which pass through the posterior lip of the central sulcus and have only minimally

damaged

of area 1, the distribution areas is essentially

the superficial

of connexions

layers

to other

the same, except that the bound-

ary zones between these areas are not spared show an increased density of granularity.

but

A slightly larger lesion in area 1 of the trunk representation is shown in Fig. 7. In front of the lesion, intracortical degeneration spreads forwards as far as the boundary with 3b. and within 3b an occasional

the antero-posterior

but more densely

:,. I: ::

Posteriorly,

the

intra-cortical spread, then with definite degenerating fibres entering from the white matter to end mainly in layer IV and the deep part of layer III. On either side of the lesion,

medially

and laterally,

the degenera-

tion becomes concentrated at the 1-2 and 2-5 boundaries with only scattered fragments throughout the central portion of area 2. A small area of damage in area 2 (Fig. 8) in the lat-

.: E ..

..’ .. :: .: :: .:

0

;:.

.. . .; I.2‘107 :: t

at its boundaries.

of area 3a,

degeneration is continuous through area 2 and into the anterior part of area 5, at first with the features of

,:

.. .

extent

:’ .

swerior

: .

‘;

:

1 :.

:‘..

: :. ::

-

8

.’ i.

.”.....

:

3

..I.

..

. . .

-n-e&

Fig 9. The degeneration after a lesion at the boundary of areas 3b and 1 near the medial end of the postcentral dimple in the representation of the caudal trunk. The degeneration extends medially along the boundaries of the cytoarchitectonic subdivisions. and at the anterior and posterior margins it continues into the other representation of the caudal trunk in the cingulate sulcus.

77

2 ips Fig. 10. The degeneration folIowing a lesion at the bounda~ between areas 3b and 1 in the lateral part of the trunk representation. From the level of the Iesion the degeneration extends laterally along the architectonic boundaries and furthest at the anterior and posterior margins of SI.

era1 part of the arm representation results in intracortical spread of degeneration throughout most of the antero-posterior extent of area 2 and anteriorly just extends into area 1. Separate from this, however, and due to spread through the white matter, is degeneration in the more anterior part of area 1, in most of area 3b and in 3a. In areas 1 and 3b it is fine, -diffuse granularity in layers I, II and the superficial part of III and in layers V and VI with the occasional degenerating fibre as well in these deeper layers, while in area 3a it is in all layers but much denser in layer IV and the adjoining deep part of layer III. The inter-area1 degeneration which has been described has been essentially that in the antero-posterior direction. Degeneration spreading in the mediolateral plane differs with the topographic region which has been damaged, and there is a close correlation between the distribution of this degeneration

and the pattern of termination of the commissural fibres when the same region of the contralateral hemisphere is damagedss@. The extent of the medio-latera1 degeneration was the same whichever cytoarchitectonic area was involved by the lesion within a particular topographical subdivision. In two animals, microelectrode lesions in the medial part of the trunk representation at the level of the upper end of the postcentral dimple gave virtually identical patterns of degeneration. The track entered the cortex of the posterior bank of the central sulcus at the boundaries of areas 3b and 1, and passed vertically down through area 3b in the posterior wall (Fig. 9). There is degeneration throughout the antero-posterior extent of Sl at the level of the damage and laterally this extends for a millimetre or two but diminishes rapidly, so that at the level of the lower end of the dimple there is only an occasional fragmented fibre. Medially, de-

78 generation chitectonic

co~fi~ues along the boundaries of the arareas, and at the anterior and posterior

margins of Sl this passes uninterruptedly on to the medial surface and into the upper wall of the cingulate sulcus to join the separate representation of the cauda1 trunk. At the boundaries of areas 3b, 1 and 2, the degeneration forms narrow prolongations into the representation of the hind-limb, but most of the cortex containing

the representation

medial surface is bare. In another damage is in the same architectonic previous experiment end of the postcentral

quickly becomes restricted to the architectonic boundaries; at the anterior and posterior margins of Sl it continues

down to the level of the junction

of the

head and arm representations at the level of the lower end of the intraparietal sulcus, but at the bounda-

of the foot on the

ries between areas 3b, 1 and 2 it extends for a much shorter distance as narrow prolongations into the arm region. The face representation is also strongly connected

experiment, the regions as in the

by commissural fibres, and lesions here result in degeneration confined within this topographic region

but it is at the level of the lower dimple (Fig. 10). At the level

of the lesion, degeneration is again found throughout the antero-posterior extent of Sl, with definite accentuations at the boundaries between the cytoarchitectonic areas, particularly between areas 3a and 4 anteriorly and areas 2 and 5 posteriorly. Medially, the degeneration continues for a millimetre and ceases rapidly close to the upper end of the postcentral dimple whilst still within the topographic representation of the trunk. Laterally, the degeneration

lateral to the tip of the intraparietal

sulcus. Damage

of area 1 approximately a millimetre in diameter behind the lower part of the central sulcus results in degeneration throughout areas 1 and 2, up to the level of the lower end of the intraparietal sulcus and down to the upper bank of the lateral sulcus below the central sulcus (Fig. 12). Anteriorly, there is degeneration in layer IV at the boundaries of area 3a, and in area 3b, including that part of it in front of the lower end of the central sulcus, there is fine granularity in the superficial and deep layers. When other lesions

Fig. 11. The band of degeneration extending laterally at the boundary between areas 1 and 2 after the lesion shown in the previous figure. Dark-field illumination. x50.

79

1

2

3b Fig. 12. A lesion in area 1 in the central part of the face representation,

and the distribution of the resulting degeneration.

Is, lateral

sulcus

made in the face representation extended more medially to overlap the junction with the hand region close to the lower end of the intraparietal sulcus, the degeneration extended more medially along the cytoarchitectonic boundaries of Sl (Fig. 13). At the anterior and posterior margins of Sl this fragmentation extends as far as the level of the lateral end of the post-central dimple but between areas 3b, 1 and 2 it forms narrow prolongations into only the lower half of the hand representation. It is probable that this additional medially directed degeneration along the anterior and posterior boundaries of Sl and the cytoarchitectonic boundaries is due to involvement of the cortex containing the representation of the occiput at the junction between the hand and head representations, as it was present after lesions of either the head or hand only if they overlapped this region. From this group of experiments in which the damage has been in the parts of the cortex of Sl which are commissurally connected, it is clear that their intrinsic connexions are also confined within the commis-

surally connected regions. The separate representations of the occipital and caudal trunk regions in Sl are connected by fibres which pass along the anterior and posterior margins of Sl around the distal limb regions, and fibres also pass from the trunk and occipital regions as narrow prolongations along the cytoarchitectonic boundaries within Sl into the limb representations. This relation of the callosal and ipsilateral cortico-cortical connexions is shown clearly by an experiment in which two small lesions were placed in Sl, one in the trunk representation and the other in the upper part of the face region; the distribution of the degeneration is essentially the same as after removal of the entire somatic sensory cortex of the opposite side. Several experiments with lesions within the distal limb representations show that the pattern of connexions in these non-commissurally connected regions is complementary with that in the trunk and head representations. Five microelectrode lesions were made in the hand representation in different an-

80 imals and the resulting

pattern

of degeneration

was

essentially

the same in all cases. An electrode

track in

the central

part of the hand representation

passed

through the anterior margin of area 1 and vertically down through area 3b on the posterior wall of the central

sulcus, almost to the boundary

with area 3a

(Fig. 14). On sections at the level of this lesion there is degeneration within each of the architectonic areas of Sl, but except in the immediate vicinity of the elec-

ary between

areas 3b and 1 also becomes free of de-

generation. Similar microelectrode lesions in the cortex containing the representation of the foot close to the medial margin of the hemisphere

sence of degeneration is also seen after larger lesions of areas 1 and 2 within the distal limb regions. The distribution

of the degeneration

lesions of different

trode track, at the boundary between area 3b and area 1, the fragmentation is confined to the central

summarized

part of each cytoarchitectonic area with a virtual absence at the boundaries, including those at the ante-

4. DISCUSSION

rior and posterior margins of Sl. The degeneration extends medially and laterally for a few millimetres, and progressively further from the lesion it diminishes so that the bare boundary regions become wider and more definite; away from the damage the bound-

result in a simi-

lar distribution of degeneration avoiding the boundaries between the cytoarchitectonic areas, and this ab-

architectonic

diagrammatically

within Sl following subdivisions

can be

(Fig. 15).

These observations on the fibre and terminal degeneration within the subdivisions of the primary somatic sensory cortex indicate that the connexions are of two main types, those that extend ~~~~ucortically

Fig. 13. Microelectrode lesion in area 1 at the medial margin of the face representation within Sf. At the level of the track, degeneration is found throughout SI but medially it continues at the architectonic boundaries, with the central parts of the subdivisions quickly becoming free of fragmentation.

81

Fig. 14. Microelectrode lesion of are 3b in the central part of the hand representation within SI. Degeneration is concentrated in the central part of each of the architectonic subdivisions of Sl with little at the boundaries. This pattern extends medially and laterally for a few millimetres where the absence of boundary fragmentation becomes more marked.

for a few milhmetres in the immediate vicinity of their origin and those that enter the white matter to pass to another site in this sensory area. The features

fundamental feature of intracortical connexions, certainly in all the areas where they have been studied. It was first seen and emphasized6 after lesions in area

of the local intracortical

17 and has been clearly shown in the same area by, tracing the axons of cells after they have been injected with HRPia. A focus of increased density of autoradiographic grains in layer IV of area 3a at a distance of a few millimetres from an injection of labelled amino-acids in area 3b has been interpreted as being due to spread along intracortical axons17, and although foci of degeneration are seen in area 3a after lesions in 3b we interpret this as terminal degeneration of fibres which have passed through the white matter because these can clearly be seen entering the cortex of 3a immediately deep to the focus. The fibres that enter the white matter may be considered to form two quite distinct groups and each requires separate discussion: those which pass in a pre-

connexions

appear to be the

same not only within any subdivision and topographic representation of Sl but also with those of area 17 of the visual cortexb, of area 4 (ref. 9) and of area 5 (unpublished observation). As they have already been discussed in detail in the accounts of other areas and in studies of Sl (refs. 17,46) only a few points need to be considered. Their greater extent in the antero-posterior dimension than in the medio-latera1 in Sl is similar to that found in area 4, as also is an asymmetry in extent in their antero-posterior dimension. In Sl, the maximum spread is posterior whereas in area 4 it was the reverse, and although the significance of this difference between Sl and area 4 is not known, such asymmetry would now appear to be a

82

Efferent

3b

7

0

+

I

*

3a

3b

1

2

5

?

‘3

9 I

I ~__________L_________J

Afferent

Efferent

I

---_-----, ?

1

I

;

30

3b

A

0

1

+

v

2

5

0 ’

L

Afferen t

Efferent

2

l

I

*____

‘““‘T”’

!

1 I__________

--.w_-,

?

f

I

I

1

3a

Feed forward

(-)

Feed back(----)

Fig. 15. Schematic summary of the ipsilateral cortico-cortical connexions passing between the architectonic subdivisions of SI as found in this study. ‘Feed-forward’ connexions, whether as efferent fibres from an area or as afferent to an area are in continuous lines, and ‘feed-back’ connexions, both as effetent and afferent, are indicated in broken lines. The different laminar terminations of these two types of cortico-cortical connexions are shown diagrammatically below. The efferent connexions of area 3a have not been determined.

83 dominantly

antero-posterior

direction

different architectonic subdivisions extend mainly in the medio-lateral the architectonic been selectively

between

the

and those which direction. As all

areas with the exception of 3a have involved in each representation, it

layers V and VI. This pattern sidered

to represent

of degeneration

‘feed-back’

connexions

is conand is

similar to the distribution of grains in autoradiographic sections of Sl after injection of labelled ami-

has been possible to conclude that both these groups of connexions differ according to the particular area

no-acids in SIIs. If these interpretations of the observations are correct it means that area 3b is the only subdivision not receiving feed-forward cortico-corti-

and

cal connexions,

representation

involved,

with

the

important

quali~cation that the representation has been identified by a correlation of the site of the lesion with the

that after receiving

(and processing)

maps based upon evoked potential and microelectrode recordingss~*7~*9~50and without direct electro-

the major part of the somatic sensory info~ation from the thalamus there is a predominantly posteriorly directed sequence of connexions from area 3b anteriorly to areas 1 and 2 posteriorly and with each

physiological

of these subdivisions

recording.

In general,

the present

ob-

servations and conclusions are in agreement with other studiesi7.@j, and only important points of difference will be discussed; these are probably due largely to differences in technique or interpretation. Perhaps one of the more important conclusions that can be drawn from the experimental findings is that there is no simple reciprocal interrelationship between the cytoarchitectonic subdivisions, which is in contrast to our earlier conclusion*0 and is probably due largely to the use of smaller lesions involving most of the areas and representations. Instead, from a comparison with observations on the cortico-cortical connexions of the second somatic sensory area (SII)s and of those between the subdivisions of other sensory areas’,*5,34,41,45,49,the fibres passing in the antero-posterior dimension between the subdivisions of Sl may be considered as either ‘feed-forward’ or as ‘feed-back’. Except for area 3a, which has not been selectively damaged here and about whose efferents we can say nothing, a lesion in any subdivision results in some degenerating fibres passing both anteriorly and posteriorly into all of the other subdivisions (and also area 5), although the density and laminar pattern of the degeneration varies in the different areas. All the degeneration in areas posterior to a lesion and in area 3a anteriorly is comparatively heavy, affects all layers and the terminal degeneration is markedly heavier in layer IV and the deep part of layer III; this type of degeneration would be considered to be typical of ‘feed-forward’ connexions. In area 3b after a lesion in area 1 and in both areas 3b and 1 after damage of area 2, there is fine degeneration which has a laminar arrangement that is virtually complementary to the ‘feed-forward’ involving layers I, II and the superficial part of III and diffusely in

of Sl also projecting

to area 3a

anteriorly and to area 5 posteriorly. This arrangement of intrinsic connexions could be taken to support the suggestion that area 3b is the ‘core’ of Sl with the other areas a ‘fringe’**; but more importantly they make area 3a, as well as area 2, the termination of fibres from all other subdivisions and provide further experimental evidence for the validity of this area of transitional cortex between Sl and area 4 as a separate functional region with its own representation of the peripheryi9. The only evidence on the further efferent cortical connexions of area 3a is that of Jones et al.17 who have described fibres passing to area 1. There is evidence from detailed recording experiments for separate representations of the entire body in each of the structural subdivisions of Sl (refs. 27, 29) and for reversal of the representations at the boundaries of the subdivisions27. That the fibres passing in the antero-posterior dimension from one architectonic area to another are linking up the separate representations of the same part of the periphery is indicated by the findings in several. of the figures. These show that when the microelectrode lesion is close to the posterior boundary of one area, the resulting degeneration is maximum at the anterior boundary of the immediately adjoining area posterior to it and vice versa. The predominantly posteriorly directed sequential progression of connexions and the reversal between adjoining areas is found in all the topographic regions of Sl but there is one important difference in the arrangement of these intercortical connexions between the trunk and face representation on the one hand and the distal limbs on the other. The degeneration in an architectonic area after damage of another area in the trunk or face re-

84 gions was present throughout its antero-posterior extent and it was accentuated at the boundary regions; this was so both in areas posterior to a lesion and in area 3a anteriorly. In the distal limb regions, however, a small lesion within one architectonic

area and

which avoided the boundary resulted in foci of degeneration in the middle of the antero-posterior ex-

the finding of some neurons showing convergence from cutaneous and deep receptors in 3a and 2 (refs. 1,12,13). The pattern

of distribution

of the degenerating

bres which pass medially or laterally white matter also differs quite markedly the topographic

representation

fi-

through the according to

in the region of cor-

tents of the other areas with little or no fragmentation at the boundaries.

tex that has been damaged, and important tions can be made between the arrangement

The cells in the 4 structural subdivisions of Sl respond differentially to stimulation of receptors in cutaneous or deep tissues of the body32.33.42; and because there is an increasing proportion of neurons

ipsilateral connexions and the callosal fibres passing between Sl of the two hemispheres. After a lesion in any architectonic area in any representation, degen-

with more complex response properties in more posterior parts of Sl it has been suggested that ‘successive levels of information processing occur in the different cytoarchitectonic areas’*“. There is, for example, a progressive increase in size of receptive field of the neurons as one passes from area 3b to 1 and 2 and also a higher frequency of cells which respond to movement of a cutaneous stimulus and which show directional selectivity. It would be reasonable to suggest that the sequence of cortico-cortical connexions passing posteriorly from area 3b to areas 1 and 2 plays an important role in these changes. The major afferent pathway to any cortical area ends predominantly in layer IV and the deep part of layer III, and in the koniocortex of area 3b this comes from the thalamus with only a few from other cortical areasis.21.3“. In areas 1 and 2 there are few fibres from the thalamus and the posteriorly directed corticocortical fibres end in layers IV and III, so the latter form the major afferent pathway. An important consequence of this interpretation would be that any transformation in the relay of sensory information from the primary somatic sensory area to the association areas of the parietal lobe is beginning in areas 1 and 2 (ref. 46). This conclusion correlates well with the fact that area 3b sends fewer fibres to area 5 than area 1, and particularly area 2 (refs. 30,31), and with the finding that both areas 2 and 5 have a high proportion of cells responding to stimulation of deep tissuesi.4.13J6.33.35. In addition to these cortico-cortical connexions passing between them, individual areas share afferent pathways as thalamo-cortical axons may send branches to more than one area24. The observation that each area sends fibres both posteriorly to area 2 and anteriorly to area 3a may be related to

correlaof these

erating fibres extend medially or laterally for at least a few millimetres, and they end in medio-laterally disposed bands of cortex approximately a few hundred pm wide, antero-posteriorly. It is in the disposition and degree of spread of these bands that differences are found between the representations. Lesions in the representations of the trunk, occipital regions and proximal limbs result in degeneration that extends predominantly along the boundaries of the areas; between areas 3b, 1 and 2 the degeneration passes a few millimetres between the distal limb representation in these areas, but not completely separating them, but at the boundaries of area 3a anteriorly and between areas 2 and 5 posteriorly the spread is greater and continues to the other and quite distant representation of the same and immediately adjacent parts of the body surface. Thus, there is continuation of these bands along the anterior and posterior margins of Sl from the representation of the caudal trunk and proximal leg near the postcentral dimple to the depth of the cingulate sulcus on the medial surface, linking up the two representations of these parts of the body. There are similar continuities between the equally separate representations of the occiput, jaw and adjoining regions at the post-central dimple and at the lower end of the intraparietal sulcus. The same and immediately adjoining representations in different architectonic areas are therefore connected by the antero-posteriorly directed fibres passing between the areas at the same mediolateral level and by fibres passing medially or laterally at the anterior and posterior margins of Sl when these same representations are at widely separated parts of the cortex. This linkage of multiple representations, either at closely adjacent sites but different architectonic areas or at quite distant sites is an

85 important

feature

not only of the ipsilateral

connex-

terior

progression

of connexions

between

the arfrom a

chitectonic

been shown that a lesion confined to one area in a part of SI that is callosally connected sends fibres to the corresponding site in the opposite hemisphere

single cortical

and to the same representation

cussion upon the differences between the topographic representations and the correlation with callosal

distant sites; for example, taining

the representation

the postcentral the postcentral the opposite

in other areas and at

damage of the cortex conof the caudal

trunk

near

dimple results in degeneration near dimple and in the cingulate sulcus of hemisphere.

A further

similarity

be-

tween the callosal and ipsilateral fibres is the accentuation of both the ipsilateral and callosal fibres at the bounda~es of areas. The cortex with the representations of the trunk, face and proximal limbs is callosally connected and the distribution of the ipsilateral intrinsic connexions of these parts of the cortex is restricted to these regions and coincides with that of the callosal connexions. There is a complementary arrangement of the connexions of the distal limbs, which are not callosally connected, as lesions in these parts of the cortex cause degeneration which spreads medially and laterally in the middle of the same and other areas (including 3a) and clearly avoids the boundaries so that they are confined within cortex that is not callosally connected. Thus, in addition to the antero-posterior division into 4 architectonic and functional areas and the medio-lateral topographic organization, the primary somatic sensory area may be divided into two distinct regions on the basis of their callosal and ipsilateral fibre connexions. The correlation that has been found between the callosal and ipsilateral fibre connexions in the trunk and head region, and particularly the accentuation at the boundary regions, does not necessarily indicate that the terminal fields of these fibres precisely overlap. The terminal degeneration after interruption of either group of fibres is always broken up into bands of varying density and with irregular gaps, and it is of particular significance that the degeneration in layer III is in separate little ‘prongs’. It is quite possible, therefore, that these two groups of fibres (and those from area 4 which have a similar distribution) terminate in immediately adjoining parts of the cortex like that which has been shown in the frontal cortex by double-labelling methodsii. The major conclusions of a sequential antero-pos-

areas and of a divergent

projection

ions within Sl but also of the callosal fibres, as it has

focus to multiple

media-laterally

dis-

posed bands are in general agreement with other recent studiesi7,46. The emphasis in the present dis-

connexions

is probably

due largely to differences

in

technique and inte~retation. The projections to the separate representations of the caudal trunk and occiput may not have been identified reasons as the density of grains graphic experiments

for such technical in the autoradio-

not being sufficiently

above the

background level or because the majority of the injections seem to have been in the limb regions or because of different interpretations of the architectonic areas. The distribution of the cortico-cortical connexions of SI has been described in terms of ‘colurnns’l6,ls, but for the same reasons that have been given for not considering the callosal connexions in these terms we do not think it appropriate to interpret the ipsilateral connexions in this way either. The concept of columnar organization of the cortex is functional and is based upon the connexions from the thalamus, but all the connexions of the cortex must contribute to this organization, although the details are not known nor is the functional significance of the cortico-cortical connexions. This interpretation of the origin and termination of these connexions, because they are disposed in bands of a few hundred pm, has already led to considerable confusion about the definition of a ‘column’. There is a definite discrepancy between the conclusion of Vogt and Pandya46 that there is a marked diminution in the degeneration at the boundaries of architectonic areas and the observations in the present study of accentuations at these regions after damage in callosally connected parts of the cortex. It is probable that this is also a matter of different interpretation as the illustrations of Vogt and Pandya indicate serious differences between their positioning of boundaries from what is generally accepted; as examples, in their Fig. 1, area 3 extends too posteriorly and is apparently surrounded by area 1 while in Figs. 3 and 4, area 2 occupies the anterior wall of the middle third of the intraparietal sulcus, configurations not previously described. As was the fact with the callosal connex-

86 ions36.38.~.5*, there are many similarities in the arrangement of the ipsilateral connexions of the somatic and visual sensory areas. In both cases, there is a predominantly

outward

from the areas receiving

sequence the heaviest

of

connexions

thalamic

input

cortex but mainly in layer IV and the deep part of layer III, whereas the feed-back connexions are lighter and end in layers I, II, the superficial part of layer III and in layers V and VI. In the antero-posterior dimension, feed-forward fibres from area 3b go to

and there are far fewer fibres passing back*s. There

areas 3a, 1 and 2; area 1 sends feed-forward

are separate representations

ions to areas 3a and 2 and feed-back 2 sends a feed-forward projection

in each area, with rever-

sals of these at the boundaries, and a precise arrangement of the cortico-cortical fibres to link corresponding parts of each of thesesr. The callosally connected regions at the boundaries are connected with each other and the non-callosal parts of one area likewise with those of other areaGsJ2. The cortico-cortical fibres passing out from the area with a heavy thalamic input from each of these sensory systems end in layer IV and the deep part of layer III, are heavier than the thalamic input to these areas and form the major afferent pathways. Several of these features of the ipsilateral cortico-cortical and callosal connexions are also present in those passing between the primary auditory area, AI, which receives the major thalamic projection48, and the other subdivisions of the auditory cortex7.r4. 5. SUMMARY

The ipsilateral cortico-cortical connexions passing between the architectonic subdivisions of the primary somatic sensory cortex, Sl, of the monkey have been studied with axonal degeneration methods after the placement of small lesions. All architectonic subdivisions except area 3a, and all the topographic representations, have been involved by the lesions. The degeneration of local intrucortical fibres has the same features that have been described in other cortical areas: dense terminal degeneration for about 200 pm immediately around the lesion and moderate degeneration extending for a few millimetres with that in layers I, IV and the deep part of V being the most marked and reaching furthest; the degeneration extends further in the antero-posterior than in the medio-lateral dimension, and further posteriorly than anteriorly. The arrangement of the intercortical fibre connexions varies with the architectonic subdivision and with the topographic representation, and as in other sensory areas these fibres may be considered as either feed-forward or feed-back. The feed-forward projections are heavy, terminate in all layers of the

feed-back

connex-

to area 3b; area to area 3a and

to areas 3b and 1; all areas also send fibres

to area 5. A lesion in one of the architectonic subdivisions in the trunk and face representations results in degeneration throughout the antero-posterior extent of Sl, but after damage within an architectonic area in the distal limb regions, there are foci of degeneration in the middle of the antero-posterior extents of the other areas but with little or none at the boundaries. The cortico-cortical fibres also extend medially or laterally for a few millimetres, in bands a few hundred pm wide. After damage of the trunk, occiput or proximal limb representations, the degenerating fibres pass predominantly along the boundaries; the separate representations of the caudal trunk, at the postcentral dimple and cingulate sulcus, are connected by continuous bands along the boundaries of area 3a and at the 215 boundary, and those of the occiput region at the levels of the postcentral dimple and lower end of the intraparietal sulcus are similarly linked. The distribution of the ipsilateral cortico-cortical connexions in the trunk, face and proximal limb regions is restricted to those parts of the cortex that are connected across the midline by callosal fibres; there is a complementary arrangement of the connexions in the distal limb representations, which are not callosally connected, as lesions in these parts result in degeneration that remains within the cortex that is not callosally connected. There are many similarities in the arrangement of the ipsilateral corticocortical connexions passing between the subdivisions of the primary somatic sensory cortex and of those between the subdivisions of the visual and auditory areas. ACKNOWLEDGEMENT

This work was supported cal Research Council.

by a grant from the Medi-

87

REFERENCES 1 Burchfiel, J. L. and Duffy, F. H., Muscle afferent input to single cells in primate somatosensory cortex, Bruin Res., 45 (1972) 241-246. 2 Costanzo, R. M. and Gardner. E. P., A quantitative analysis of responses of direction-sensitive neurons in somatosensory cortex of awake monkeys, J. Neurophysiol., 43 (1980) 1319-1341. 3 Dreyer, D. A., Loe, P. R., Metz, C. B. and Whitsel, B. L., Representation of head and face in postcentral gyrus of the macaque, J. Neurophysiol., 38 (1975) 714-733. 4 Duffy, F. H. and Burchfiel, J. L., Somatosensory system: organizational hierarchy from single units in monkey area 5, Science, 172 (1971) 273-275. 5 Fink, R. J. and Heimer, L., Two methods for selective impregnation of degenerating axons and their synaptic endings in the central nervous system, Brain Res., 4 (1967) 369-374. 6 Fisken, R. A., Garey, L. J. and Powell, T. P. S., The intrinsic association and commissural connections of area 17 of the visual cortex, Phil. Trans. roy. Sot. B, 272 (1975) 487-536. 7 Fitzpatrick, K. A. and Imig, T. J., Auditory cortico-cortical connections in the owl monkey, J. camp. Neurol., 192 (1980) 589-610. 8 Friedman, D. P., Laminar patterns of termination of cortico-cortical afferents in the somatosensory system, Brain Res., 273 (1983) 147-151. 9 Gatter, K. C. and Powell, T. P. S., The intrinsic connections of the cortex of area 4 of the monkey, Brain, 101 (1978) 513-541. 10 Gilbert, C. D. and Wiesel, T. N., Clustered intrinsic connections in cat visual cortex, J. Neurosci., 3 (1983) 1116-1133. 11 Goldman-Rakic, P. S. and Schwartz, M. E., Interdigitation of contralateral and ipsilateral columnar projections to frontal association cortex in primates, Science, 216 (1982) 755-757. 12 Heath, C. J., Hore, J. and Phillips, C. G., Inputs from lowthreshold muscle and cutaneous afferents of hand and forearm to areas 3a and 3b of baboon’s cerebral cortex, J. Physiol. (Lord), 257 (1976) 199-227. 13 Hyvlrinen, J. and Poranen, A., Receptive field integration and submodality convergence in the hand area of the postcentral gyrus of the alert monkey, J. Physiol. (Lord), 283 (1978) 539-556. 14 Imig, T. J. and Reale, R. A., Patterns of cortico-cortical connections related to tonotopic maps in cat auditory cortex, J. camp. Neural., 192 (1980) 293-332. 15 Jones, E. G., Lamination and differential distribution of thalamic afferents within the sensory-motor cortex of the squirrel monkey, J. camp. Neural., 160 (1975) 167-204. 16 Jones, E. G., Burton, H. and Porter, R., Commissural and cortico-cortical ‘columns’ in the somatic sensory cortex of primates, Science, 190 (1975) 572-574. 17 Jones, E. G., Coulter, J. D. and Hendry, S. H. C., Intracortical connectivity of architectonic fields in the somatic sensory, motor and parietal cortex of monkeys, J. camp. Neural., 181 (1978) 291-348. 18 Jones, E. G., Coulter, J. D. and Wise, S. P., Commissural columns in the sensory-motor cortex of monkeys, J. camp. Neural., 188 (1979) 113-136. 19 Jones, E. G. and Porter, R., What is area 3a?, Brain Res.

tev.,2 (1980) l-43. 20 ‘ones, E. G. and Powell, T. P. S., Connexions of the somaic sensory cortex of the rhesus monkey. I. Ipsilateral corti:a1connexions, Brain, 92 (1969) 477-502. 21 lones, E. G. and Powell, T. P. S., Connexions of the somatic sensory cortex of the rhesus monkey. III. Thalamic connexions, Brain, 93 (1970) 37-56. 22 1Kaas, J. H., What, if anything, is SI. Organization of first ;omatosensory area of cortex, Physiol. Rev., 63 (1983) !06-231. 23 Killackey, H. P., Gould, H. J., Cusick, C. G., Pons, T. P. md Kaas, J. H., The relation of corpus callosum connec.ions to architectonic fields and body surface maps in sensoimotor cortex of new and old world monkeys, J. camp. ‘veurol., 219 (1983) 384-419. 24 Landry, P. and Deschenes, M., Intracortical arborizations and receptive fields of identified ventrobasal thalamocorti:a1 afferents to the primary somatic sensory cortex in the :at, J. camp. Neural., 199 (1981) 345-371. 25 Maunsell, J. H. R. and Van Essen, D. C., The connections If the middle temporal visual area (MT) and their relationship to a cortical hierarchy in the macaque monkey, J. Neuposci., 3 (1983) 2563-2586. 26 Mountcastle, V. B., Lynch, J. C., Georgopoulos, A., Sakata, H. and Acuna, C., Posterior parietal association cortex of the monkey: command functions for operations within extrapersonal space, J. Neurophysiol., 38 (1975) 871-908. 27 Nelson, R. J., Sur, M., Felleman, D. J. and Kaas, J. H., Representations of the body surface in postcentral parietal cortex of Macaca fascicularis, J. camp. Neural., 192 (1980) 611-643.

28 Olavarria, J. and Montero, V. M., Relation of callosal and striate-extrastriate cortical connections in the rat: morphological definition of extrastriate visual areas, Exp. Brain Res., 54 (1984) 240-252.

29 Paul, R. L., Merzenich, M. and Goodman, H., Representation of slowly and rapidly adapting cutaneous mechanoreceptors of the hand in Brodmann’s areas 3 and 1 of Macaca mulatta, Brain Res., 36 (1972) 229-249.

30 Pearson, R. C. A. and Powell, T. P. S., The cortico-cortical connections to area 5 of the parietal lobe from the primary somatic sensory cortex of the monkey, Proc. roy. Sot. B, 200 (1978) 103-108. 3i Pearson, R. C. A. and Powell, T. P. S., The projection of the primary somatic sensory cortex upon area 5 in the monkey, Bruin Res. Rev., 9 (1985) 89-107. 32 Phillips, C. G., Powell, T. P. S. and Wiesendanger, M., Projection from low-threshold muscle afferents of hand and fore-arm to area 3a of baboon’s cortex, J. Physiol. (Land.), 217 (1971) 419-446. 33 Powell, T. P. S. and Mountcastle, V. B., Some aspects of the functional organization of the cortex of the postcentral gyrus of the monkey: a correlation of findings obtained in a single unit analysis with cytoarchitecture, Bull. Johns Hopk. Hosp., 105 (1959) 133-162. 34 Rockland, K. S. and Pandya, D. N., Laminar origins and terminations of cortical connections of the occipital lobe in the rhesus monkey, Bruin Res., 179 (1979) 3-20. 35 Sakata, H. Y., Takorka, Y., Kawarasaki, A. and Shibutani, H., Somatosensory properties of neurons in superior parietal cortex (area 5) of the rhesus monkey, Brain Res., 64 (1973) 85-102.

36 Segraves, M. A. and Rosenquist, A. C., The afferent and efferent callosal connections of retinotopically defined

88 areas in cat cortex, J. Neurosci., 2 (1982) 1090-1107. 37 Shanks, M. F., Pearson, R. C. A. and Powell, T. P. S., The intrinsic connections of the primary somatic sensory cortex of the monkey, Proc. roy. Sot. B, 200 (1978) 95-101. 38 Shanks, M. F., Pearson, R. C. A. and Powell, T. P. S., The callosal connexions of the primary somatic sensory cortex in the monkey, Brain Res. Rev., 9 (1985) 43-65. 39 Shanks, M. F. and Powell, T. P. S., An electron-microscopic study of the termination of thalamocortical fibres in areas 3b, 1 and 2 of the somatic sensory cortex in the monkey, Brain Res., 218 (1981) 35-47. 40 Shanks, M. F., Rockel, A. J. and Powell, T. P. S., The commissural fibre connections of the primary somatic sensory cortex, Bruin RET., 98 (1975) 166-171. 41 Spatz, W. B., Topographically organized reciprocal connections between areas 17 and MT (visual area of superior temporal sulcus) in the marmoset Cullirhrix jucchus, Exp. Brain Res., 27 (1977) 559-572. 42 Tanji, J. and Wise, S. P., Submodality distribution in sensorimotor cortex of the unanaesthetized monkey, J. Neurophysiol., 45 (1981) 467-481. 43 Tigges, J., Spatz, W. B. and Tigges, M., Reciprocal pointto-point connections between parastriate and striate cortex in the squirrel monkey (Saimiri), J. camp. Neural., 148 (1973) 481-490. 44 Tigges, J., Spatz, W. B. and Tigges, M., Efferent corticocortical fiber connections of area 18 in the squirrel monkey (Saimiri), J. camp. Neural., 158 (1974) 216-236. 45 Tigges, J., Tigges, M., Anschel, S., Cross, W. D., Letbetter, W. D. and McBride, R. L., Area1 and laminar distribu-

tion of neurons interconnecting the central visual cortical areas 17, 18, 19 and MT in the squirrel monkey (Saimiri), J. camp. Neurol., 209 (1981) 539-560. 46 Vogt, B. A. and Pandya, D. P., Cortico-cortical connections of somatic sensory cortex (areas 3, 1 and 2) in the rhesus monkey, J. camp. Neurol., 177 (1978) 179-192. 47 Wiitanen. J. T., Selective silver impregnation of degenerating axons and axon terminals in the central nervous system of the monkey (Macaca mulatta), Brain Res, 14 (1969) 546-548. 48 Winer, J. A., Diamond, I. T. and Raczkowski, D., Subdivi-

49

50

51 52

sions of the auditory cortex of the cat: the retrograde transport of horseradish peroxidase to the medial geniculate body and posterior thalamic nuclei, J. camp. Neurol., 176 (1977) 387-418. Wong-Riley, M., Reciprocal connections between striate and prestriate cortex in squirrel monkey as demonstrated by combined peroxidase histochemistry and autoradiography, Brain Res., 147 (1978) 159-164. Woolsey, C. N., Organization of somatic sensory and motor areas of the cerebral cortex. In H. B. Harlow and C. W. Woolsey (Eds.), Biological and Biochemical Bases of Behaviour, University of Wisconsin Press, Madison, 1958. Zeki, S. M., Cortical projections from two prestriate areas in the monkey, Brain Res., 34 (1971) 19-35. Zeki, S. M., Simultaneous anatomical demonstration of the representation of the vertical and horizontal meridians in areas V2 and V3 of rhesus monkey visual cortex, Proc. roy. Sot. B, 195 (1977) 517-523.