thalamic projections to the somatosensory cortex of the echidna, tachyglossus aculeatus

THE JOURNAL OF COMPARATIVE NEUROLOGY 229:153-170 (1984)

Thalamic Projections to the Somatosensory Cortex of the Echidna, Tachyglossus

aculeatus

PHILIP S. ULINSKI Department of Anatomy and Committee on Neurobiology, The University of Chicago,

Chicago, Illinois 60637

ABSTRACT Evoked potential studies (Lende, '64) suggest that echidnas have a

single, topographically organized somatosensory area (SMI) that spans a mediolaterally oriented sulcus called sulcus alpha. A motor area (MI) is situated on the prealpha gyrus. This study examines the cytoarchitecture and thalamic afferents of SMI in the echidna, Tachyglossus aculeatus.

SMI contains two cytoarchitectonic fields. A caudal field extends across the postalpha gyrus and onto the floor of sulcus alpha. It has a well-developed layer 4 and a relatively small number of medium-sized pyramidal cells in layer 5 . The rostral field extends from the floor of sulcus alpha onto its rostral bank. It also has a well-developed layer 4 but has a large number of large pyramidal cells in layer 5 . Layer 4 thins as it is followed onto the crown of the prealpha gyrus. The remainder of this gyrus contains a single cytoarchitectonic field with a thin layer 4 and a layer 5 heavily populated with larger pyramidal cells. This field corresponds to the physiologically defined motor area MI.

Thalamic afferents to SMI were examined by placing pressure injections of horseradish peroxidase into the two cytoarchitectonic fields. An injection that involved both fields retrogradely labeled neurons throughout the ventral posterior nucleus of the thalamus. An injection restricted to the caudal field labeled a band of neurons that extends rostrocaudally throughout the ventral part of the ventral posterior nucleus. An injection restricted to the rostral field labeled a band of neurons situated dorsally in the ventral posterior nucleus. No other thalamic groups contained labeled neurons comparable to the labeling seen in the intralaminar or posterior nuclei following a horseradish peroxidase injection into SI of marsupial or placental mammals.

These results indicate that SMI in Tachyglossus contains two cytoarchitectonic fields that resemble areas 3a and 3b in some placental mammals, suggesting that the constellation of cytoarchitectonic fields corresponding to areas 4,3a, and 3b is a basic mammalian character which has been modified in marsupial and many placental mammals.

Key words: SI, ventrobasal nucleus, ventral posterior nucleus

Several lines of evidence suggest that mammals divided into two radiations shortly after their origin from therapsid reptiles in the late Triassic era (e.g., Griffiths, '68, '78; Marshall, '79; Clemens, '79). The first radiation includes the therian mammals now represented by the living marsupials and the dozen odd orders of placental mammals. The second radiation includes the prototherian mammals

which were once a widespread and successful lineage en- compassing groups such as the triconodonts, docodonts, and the multituberculates (e.g., Lillegraven et al., '791, but are now represented solely by the three genera of living mono-

Accepted May 29,1984.

0 1984 ALAN R. LISS, INC.

154

tremes-the duckbilled platypus (Ornithorhynchus anatinus) and the echidnas of the genera Tachyglossus and Zaglossus. Although the monotremes show a number of typically reptilian features (Griffiths, '68, '78), they are characteristically mammalian in other ways, including the possession of a well-developed neocortex.

The neocortex of both the platypus and echidnas is six- layered and histologically similar to therian neocortices. Several physiological studies of the neocortex in monotremes show that both the platypus and Tachyglossus have an "excitable" cortex that generates movements upon electrical stimulation (Martin, 1898; Abbie, '38; Goldby, '39; Lende, '64; Bohringer and Rowe, '77) and modality specific areas responsive to visual, auditory, and somatosensory stimuli (Lende, '64; Allison and Goff, '72; Bohringer and Rowe, '77). However, the nature of thalamic projections to the neocortex in monotremes has been the subject of only a single retrograde degeneration study (Welker and Lende, '80). The present study consists of a cytoarchitectonic description of the areas of the cortex identified as somatosensory (SMI) and motor (MI) by Lende ('64), followed by an investigation of the thalamic projections to area SMI using the retrograde transport of horseradish peroxidase. Area SMI is shown to contain two cytoarchitectonic fields. The horseradish peroxidase experiments indicate that both of these fields receive projections from a single thalamic nucleus. These observations raise the possibility that the somatosensory-motor cortex in Tachyglossus consists of three cytoarchitectonic fields that resemble the complex of areas 4, 3a, and 3b of many placental mammals, both in their cytoarchitecture and connections.

P.S. ULINSKI

were prepared by the tetramethyl benzidine procedure for HRP histochemistry (Mesulam, '78). These sections were counterstained with neutral red. Sections in two other bins were prepared by a cobalt-enhanced Hanker-Yates procedure. After rinsing in 0.1 M Tris buffer at pH 7.4, these sections were placed in a solution of 1% cobalt chloride in Tris buffer for 15 minutes at 35°C. They were then rinsed successively in Tris buffer, phosphate buffer, and preincu- bated for 15 minutes at 35°C in freshly made Hanker-Yates solution (Hanker et al., "77). Incubation was in two changes of 0.25 ml 3% hydrogen peroxide in 50 ml of Hanker-Yates solution for a total of 15 minutes. The sections were finally rinsed in phosphate buffkr for 30 minutes.

In addition to this material, two sets of serial sections of echidna forebrain were available from the collection of brains in the Department of Neurophysiology at the Uni- versity of Wisconsin. These brains were embedded in celloidin, sectioned at 25 pm and stained with thionin.

RESULTS Cortical cytoarchitecture

Figure 1A shows the map of cortical motor and sensory areas derived from Lende's 1964 study of Tachyglossus. A motor area (M) is largely separated from the sensory areas by sulcus alpha, which rims obliquely from dorsal to rostra1 across the caudal surface of the hemisphere. The only somatosensory area (SS) identified lies caudal to sulcus alpha and is separated from the more dorsally situated visual area (V) by a shallow sulcus. The auditory area (A) is positioned ventrally OM the hemisphere's caudal pole. Evoked potential maps (Fig. 1B) of the somatosensory area indicate that it extends across sulcus alpha and onto the caudal bank of the prealpha sulcus. It is topographically organized with the representations of the beak and the apices of the limbs extending onto the caudal bank of the prealpha sulcus.

Figure 2A is a photomicrograph of a section passing through sulcus alpha. 1 t shows that the physiologically defined SMI area contains two cytoarchitectonic fields. A

MATERIALS AND METHODS Injections of horseradish peroxidase into the posterior

cerebral hemispheres were attempted in three echidnas. The animals were premedicated with a single dosage of 38 mg of ketamine hydrochloride (Ketalar, Parke Davis). Each animal was fitted with a mask fashioned by cutting the tip of a surgical glove and slipping it over the animal's snout. It was anesthetized for the duration of the surgery with a mixture of halothane and nitrous oxide. Spines over the dorsum of the head were removed with bone forceps. The skin and muscles on the midline of the head were incised and the thick pad of fat that lies over the skull was removed. A burr was used to make a hole over the posterior aspect of the cerebrum. Two of the experiments yielded useful results. In the first animal (Echidna l), the cranium was opened unilaterally. In the second successful case (Echidna 3), a bilateral craniotomy was performed. The dura was cut and a series of injections of HRP (30% Bohrin- ger-Manheim grade I in pH 7.6 Tris buffer) were made into the cortex using a 10- or 5-pl syringe with a 25- or 32-gauge needle. The craniotomies permitted visualization of the major sulci and gyri and injections were placed on either side of sulcus alpha. The wounds were closed and the animals survived for 4 1 / 2 4 days.

Echidnas were then anesthetized with Nembutal and per- fused intracardially with a series of three solutions. The first was 0.2 M phosphate buffer at pH 7.4, the second was 1.25% glutaraldehyde and 1% paraformaldehyde in phosphate buffer, and the third was 5% sucrose in phosphate buffer. The brain was removed, photographed, and then stored overnight in sucrose buffer. A block containing the cortex and thalamus was sectioned into 18 bins. Sections in four bins were stained with cresyl violet. Those in two bins

C CP Hb IC IPN LGNa

LGNb

MI MN OP Pt r RF SM SMI SN V VM VP 1 2 3 4 5

Ahbreuiations Caudal field of SMI Cerebral peduncle Habenular nuclei Internal capsule Interpeduncular nucleus Lateral geniculate niucleus a of Campbell and Hayhow- Lateral peniculate nucleus b of Campbell and Hayhow- Motor area of Lende Mammillary nuclei of hypothalamus Nucleus OP of Welker and Lende Pretectal area Rostra1 field of SMI Fasciculus retroflexus Stria medullaris somatosensory area of Lende Substantia nigra Ventral nucleus of thalamus Ventromedial nucleus of thalamus Ventral posterior nucleus of thalamus Layer 1 of cortex Layer 2 of cortex Layer 3 of cortex Layer 4 of cortex Layer 5 of cortex

A

Fig.

1.

Cor

tical

org

aniz

atio

n in

Tuchyglossus. T

his f

igur

e su

mm

ariz

es fo

r th

e co

nven

ienc

e of

the

read

er t

he re

sult

s of

Len

de's

('64)

stu

dy w

ith

elec

tro-

ph

ysio

logi

cal t

echn

ique

s of

the

cor

tical

are

as in

ech

idna

s. A

. The

pos

ition

of

the

visu

al N

), au

dito

ry (

A),

som

atos

enso

ry (S

S), a

nd m

otor

(M

I are

as a

re

show

n on

a v

iew

of t

he le

ft c

ereb

ral h

emis

pher

e of

Tachyglossus. B

. Det

ails

of

the

bod

y re

pres

enta

tion

in

the

som

atos

enso

ry a

rea

(SM

I) ar

e sh

own.

The

in

set

show

s th

e di

stri

buti

on o

f re

cord

ing

site

s ov

er S

MI o

n a

view

of

the

righ

t ce

rebr

al h

emis

pher

e. T

he m

ajor

par

t of

the

fig

ure

show

s th

e lo

ci o

n th

e bo

dy s

urfa

ce t

hat

elic

ited

resp

onse

s up

on s

omat

ic s

tim

ulat

ion.

Sul

cus

alph

a ex

tend

s do

wn

the

cent

er o

f th

e re

cons

truc

tion.

Not

ice

that

the

som

a-

tose

nsor

y ar

ea e

xten

ds r

ostr

a1 to

sul

cus

alph

a on

to t

he p

real

pha

gyru

s.

From

Len

de ('

69).

h?

B

156 P.S. ULINSKI

Fig. 2. Cytoarchitivture of somatosensory-motor cortex in Tuchyglossus. A. A coronal section through sulci alpha and beta shows the cytoarchitecture of the cortex t h n t surrounds sulcus alpha. The physiologically defined somatosensory cortex (SMI) contains two cytoarchitectonic fields. The caudal field (ci coniprisi-s most of the postalpha gyrus. The rostral field (r)

extends from the floor of sulcus alpha onto the caudal bank of the prealpha gyrus. B-D show higher-magnification views of the three cortical fields. B. The cortex in the physiologically defined motor area (MI). C. The cortex in the rostral field (r) of SMI. D. The cortex in the caudal field (c) of SMI.

ECHIDNA THALAMIC PROJECTIONS 157

caudal field (C) is present throughout the postalpha gyrus and extends onto the floor of sulcus alpha. It is characterized by a well-developed fourth layer (Fig. 2D). Layer 5 contains the somata of relatively few, small pyramidal neurons. A rostral field (r) extends from the floor of sulcus alpha onto the caudal bank of the prealpha gyrus. Its fourth layer is only slightly less well developed than that of the caudal field (Fig. 2C). However, its fifth layer contains the somata of a large number of medium-sized pyramidal neurons. The junction between the SMI and MI areas defined physiologically by Lende is apparently marked by a transition to a third cortical area characterized by a very poorly developed layer 4 and a thick layer 5 containing the somata of many large pyramidal neurons (Fig. 2B). This area extends rostrally across the prealpha gyrus.

The border of the caudal and rostral fields was defined more precisely by preparing detailed drawings of the pre- and postalpha gyri on every 30th 25-pm serial section of a normal echidna brain. Examples of these drawings appear in the center of Figure 3. The bottom of Figure 3 shows how the pre- and postalpha gyri would appear if flattened into a sheet. In preparing this reconstruction, sections were aligned relative to sulcus alpha, so sulcus alpha appears as a straight line. The border between the caudal and rostral fields meanders relative to sulcus alpha. Ventrally, it is situated caudal to the floor of the sulcus in the head representation. More dorsally, it crosses the floor of the sulcus and extends onto the caudal face of the prealpha gyrus within the trunk and hindlimb representations. The caudal edge of the caudal field lies within the rhinal fissure. The rostral extent of the rostral field runs along the crown of the caudal extent of the prealpha gyrus.

Thalamic cytoarchitecture The principal result of this study is an examination of the

thalamic projections to the caudal and rostral fields of SMI. However, before these experiments can be considered it is first necessary to describe certain features of the caudal thalamus in echidnas. The nuclei of the caudal thalamus are illustrated in Figures 4 and 5 . Figure 4 shows a series of drawings of the caudal thalamus at regular intervals to illustrate the arrangement of the major nuclei. Figure 5 shows photomicrographs through two key levels.

The caudal pole of the thalamus in Tuchyglossus is formed by the nucleus designated OP by Welker and Lende ('80). This is a large, irregularly shaped structure that extends caudally as far as the rostral pole of the oculomotor complex (Figs. 4J-L). It forms the lateral surface of the diencephalon and is bounded medially by the pretectum. It increases in size as it is traced rostrally to a maximum just rostral to the posterior commissure, where it then begins to decrease in size. It extends as far rostral as the rostral pole of the interpeduncular nucleus (Fig. 45). OP is characterized throughout its extent by relatively large, oval neurons with a rim of cytoplasm containing distinct Nissl bodies. The neurons are arranged into clearly bounded islands and bands of cells. Welker and Lende ('80) chose the term "OP" because neurons in this nucleus show retrograde degeneration following lesions of the occipital pole of the cortex. Its neurons do not degenerate following lesions restricted to the primary visual area defined physiologically by Lende (Fig. 1A:V).

The caudal pole of OP overhangs the lateral geniculate nucleus a of Campbell and Hayhow ('71). This is a small,

rostrocaudally elongate nucleus that reaches from the level of the posterior commissure caudally, to the point where the internal capsule joins the telencephalon and diencephalon, rostrally (Fig. 4J-L). It contains small, spherical neurons with little Nissl substance. The Nauta studies of Campbell and Hayhow ('71) show that this nucleus receives a bilateral retinal projection. None of the cortical lesions performed by Welker and Lende ('80) produced detectable retrograde degeneration in lateral geniculate nucleus a.

OP is replaced by two nuclei as it is followed rostrally in serial sections. The first is the lateral geniculate nucleus b of Campbell and Hayhow ('71). This is a small, elongate and elliptical-shaped mass of cells situated dorsal to OP on the dorsolateral surface of the diencephalon (Fig. 41). It overlaps the caudal pole of the habenular complex and contains neurons that are relatively small with little Nissl substance. It receives a projection from the contralateral retina (Campbell and Hayhow, '71) and projects to the visual cortex on the ipsilateral cerebral hemisphere (Welker and Lende, '80).

The second nucleus that replaces OP at rostral levels is the ventral posterior nucleus (VP; Fig. 4A-L; Fig. 5a,b). Its caudal pole is situated at the level of the posterior commissure, positioned between OP and the pretectum (Fig. 4L). The ventral posterior nucleus increases in size as it is traced rostrally so that it makes up the largest fraction of the lateral thalamus throughout most of the rostrocaudal extent of the habenular complex (Fig. 4B-J). It is bounded medially by the ventromedial nucleus (VM) of the thalamus (Figs. 4A-6; Fig. 5a,b) and laterally by the internal capsule (IC, Fig. 4A-J, Fig. 5). It appears to extend to the ventral boundary of the dorsal thalamus at caudal levels, but begin- ning at the rostral pole of the substantia nigra (SN) it is bounded ventrally by the ventral nucleus (V) of the thalamus (Figs. 4A-D, 5). The neurons in the ventral posterior nucleus vary in shape from round to elliptical (Fig. 6). They appear slightly smaller than do the neurons in OP and significantly larger than those in the lateral geniculate nucleus b and the ventromedial nucleus. They stain less intensely in Nissl preparations than do the neurons in OP.

The ventral nlicleus (V) appears as a lens-shaped structure near the rostral pole of the habenular complex (Figs. 4A-D, 5) . It increases in size as it is followed rostrally and quickly replaces the ventral posterior nucleus at levels rostral to the habenular complex. It is distinguished from the overlying ventral posterior nucleus by the smaller size and paucity of Nissl substance of its neurons (Fig. 6c).

The nuclei present more rostrally in the thalamus of Tuchyglossus are illustrated by Welker and Lende ('80) but are not relevant to the experimental results reported here.

Thalamocortical projections Having described the nuclei of the caudal thalamus, we

can now consider the thalamic projections to the two cytoarchitectonic fields of SMI. These projections were studied using the retrograde transport of HRP in echidnas 1 and 3. Echidna 1 received a large injection of HRP in one hemisphere. Echidna 3 received injections in both the left and right hemispheres, which are treated as separate cases in the absence of any evidence for bilateral thalamocortical projections. All three cases showed a patch of retrogradely labeled neurons in the caudal thalamus (Fig. 7). The nature of labeling in the Hanker-Yates sections varied from the presence of a granular reaction product to the Golgi-like

158

1 5 I I

P.S. ULINSKI

2 0 mm

7 MI S M I \ 7 ,~~,,,,

7- :;;;US R C

S A U

SUICUS D fissure.. A -alpha

L

i

t t

%''l\\,

, ' . ! : I

5mm

Fig. 3. C o r t ~ c a l fii,ids in Tach,yg/ossus. This figure shows the overall position of the rostral and caudal fields of SMI on the cerebral hemispheres in 7'uchygkjssu.s. Thv drawing of the brain in the upper part of the figure shows thc positions of sulci alpha and beta. The lines designated 1 and 5 show the planes ol'svctrons 1 and 5 in the center of the figure. These, and the intcrveniriq 1hrt.t. swlions, show the configuration of the cortex at

spaced intervals along the pre- and postalpha gyri. The arrows mark the lateral and medial borders of the rostral field of SMI. Layer 5 of the cortex is indicated by stipple on these sections. The lower part of the figure shows a flattened reconstruction of the caudal and rostral fields based on sections like those shown in the center of the figure. Notice in particular that the lateral border of the rostral field extends across sulcus alpha.


- 4 mm

a

RF

"

OP 3

K

L

OP

L

CAUDAL Fig. 4. Caudal thalamus in Tachyglossus. The positions of the major

structures in the caudal thalamus and rostra1 midbrain of Tachyglossus are shown on a series of drawings made from serial sections at intervals of 750 pm.

160 P.S. ULINSKI

Fig. 5. Cytoarchitectiire of caudal thalamus. The positions of major nuclei in the caudal thnianiux of Tuchyglossus are illustrated by two coronal sections. a. The planc. of this section corresponds approximately to section B in Figure 4. Notice that, the ventral posterior nucleus (VP) makes up the major mass of the lateral thalamus hut that the ventral nucleus (V) is

present ventrally. Its dorsal border is marked by arrowheads. It is distinguished from the overlying ventral posterior nucleus by its smaller and more densely packed neurons. b. The plane of this section corresponds approximately to that of section H in Figure 4. Notice that the ventral posterior nucleus (VP) is the only nucleus present in the lateral thalamus.

filling that included the somata and proximal dendrites of labeled neurons. The number and distribution of retrogradely labeled neurons was similar in sections prepared by both the Hanker-Yates and tetramethyl benzidine procedures.

Echidna 1 received a large HRP injection in the left hemisphere (Fig. 8). Comparison of the position of the injection site to Lende's ('64) map of the somatosensory cortex (Fig. 1B) suggests that the injection site encroached on most of the head, trunk, and limb representations. However, it may have involved the underlying white matter so that uptake by fihers passing to the large beak representation cannot be exclud(xd The injection spans sulcus alpha and

involves both the rostral and caudal fields. Labeled axons can be traced from the injection site into the subcortical white matter, through the internal capsule, and into the lateral thalamus. The lateral thalamus contains a large number of neurons retrogradely filled with HRP reaction product within the cytoarchitecturally defined borders of the ventral posterior nucleus. They are present at the caudal pole of the thalamus (Fig. 8F) medial to the OP nucleus, fill the major part of the lateral thalamus at intermediate levels (Fig. 8E-C), and are displaced dorsally by the caudal pole of the ventral nucleus at rostral levels. Labeled neurons extend to the extreme lateral border of the ventral posterior nucleus. The medial border of the ventral poste-


Figure 5 (continued)

rior nucleus is more difficult to define, but comparison of the positions of labeled neurons in echidna 1 to the cytoarchitecture of the thalamus in celloidin-embedded material suggests that no labeled neurons are present in the medial third of the ventral posterior nucleus at intermediate and rostral levels. It is not possible to determine from the material at hand whether this is due to an error in defining the borders of the ventral posterior nucleus or to subtotal involvement of the somatosensory cortex. In any event, this case indicates that neurons in most, if not all, of the cytoarchitecturally defined ventral posterior nucleus project to the rostral and caudal fields of SMI.

The two injections in echidna 3 were designed to study the differential connections of the rostral and caudal sub- fields. The injection in the left hemisphere (Fig. 9) is situ-

ated entirely rostral to sulcus alpha and is confined principally to the part of the rostral field lying on the crown of the prealpha gyrus. Comparison to Lende’s map of the somatosensory area (Fig. 1B) suggests that the injection is in the representation of the hindlimbs. Retrogradely labeled neurons are present in the ventral posterior nucleus arranged in a slab-shaped column extending from the dor- somedial corner of the nucleus’s caudal pole (Fig. 8F) rostrally along its dorsal surface (Fig. 8D,E) and shifting ventrolaterally at more rostral levels (Fig. 8A-C). Although labeled neurons are present throughout the rostrocaudal extent of the ventral posterior nucleus, they are most prev- alent caudally.

The injection in the right hemisphere (Fig. 10) is situated caudal to sulcus alpha in the caudal field. Comparison to

162 P.S. ULINSKI

Fig. 6. Cytology of ncwrons in caudal thalamus. a. This low-power view of the caudal thalanius shows the character of neurons in the ventral posterior (VP) and ventral (v) nuclei. The boundary between the two nuclei is marked by arrowheads. Notice that neurons in the ventral posterior nucleus tend to be lerger and less densely packed than those in the ventral

nucleus. b. Neurons in the ventral posterior nucleus shown at higher magnification. Note the presence of some relatively large neurons. c. Neurons in the ventral nucleus. Note the absence of relatively large neurons. b and c are a t the same magnification.


Fig. 7. Neurons retrogradely labeled from a cortical injection. This figure shows a band of neurons lying in the ventral posterior nucleus retrogradely labeled from a cortical injection of HRP in the SMI cortex.

Lende’s somatosensory map suggests that it lies in the representation of the forelimbs or the head. Retrogradely labeled neurons are present along the lateral border of the ventral posterior nucleus situated more ventrally than those in echidna 3L. They do, however, form a rostrocaudally elongate slab that extends along the length of the nucleus. The majority of labeled neurons are present caudally.

The two injections in echidna 3 demonstrate that both the rostra1 and caudal fields receive projections from the ventral posterior nucleus. None of the three injections provide any hint of projections to the somatosensory cortex from other thalamic nuclei.

DISCUSSION The observations and experiments reported here show

that the somatosensory cortex SMI defined physiologically by Lende in Tachyglossus comprises two cytoarchitectonic areas and that both receive projections from the ventral posterior nucleus of the thalamus. Although the difficulty of obtaining echidnas limits the number of cases available, the results do provide information on the organization of the somatosensory cortex and thalamus in echidnas and

raise the possibility that monotremes have a complex of cortical area comparable to areas 4, 3a, and 3b of placental mammals.

Organization of somatosensory cortex in Tachyglossus

In contrast to the lissencephalic isocortex of Orrzithorhyn- chus (Ziehen, 1897; Elliot Smith, 1899; Hines, ’29), the isocortex in Tachyglossus is highly convoluted by fissures that were given a series of Latin names by Ziehen (1897) and Greek letters by Elliot Smith (’02). Ziehen (‘08) went on to describe the histology of the isocortex, establishing that its lamination pattern generally resembles that of therian mammals and shows areal variations in cytoarchitecture. Brodmann (‘09) and Schuster (‘10) also provided brief accounts of isocortical cytoarchitecture, but Abbie (‘40) carried out the only extensive description of the cytoarchitecture of monotremes. He recognized 14 cortical areas in both Ornithorhynchus and Tachyglossus, using a nomencla- ture motivated by the theory (e.g., Dart, ’34; Abbie, ’40) that the isocortex is derived from the parahippocampal (PH) and pyriform (PPy) cortices.

164

E C H I D N A 1

P.S. ULINSKI

A

1

c - 2 m m

s u l c u s t o .

Fig. 8. Thalanioc(irtical projections in echidna 1. This and the following two figures illustrate the results of experiments in which HRP injections were made into the SMI cortex in Tachyglossus. The format ofeach illustration is the same. The position of the injection site is shown on the cortical surface in a view of the lateral surface of the brain. These drawings were prepared from brain photographs taken prior to histological preparation. The planes of sections through the injection site are indicated by numbered lines. The sections show t,hc, location of the injection site (stipple) relative to

sulcus alpha. A series of spaced sections through the caudal thalamus are shown a t the side of each illustration. The positions of retrogradely labeled neurons are plotted onto these sections. Each dot represents a single labeled neuron. This particular injection extended across sulcus alpha and involved both the rostral and caudal fields of SMI. Labeled neurons are distributed throughout the rostrocaudal extent of the ventral posterior nucleus (VPI hut are absent from the ventral nucleus (V).

Descriptions of isocortex in Tachyglossus have all recognized that the nature of the cortical cytoarchitecture changes in the vicinity of sulcus alpha. In particular, Abbie ('40) placed the boundary between his PH4 and PPy3 cortices in the floor of sulcus alpha. The description of PPy3 cortex corresponds closely to what I have designated the caudal field, indicating that its rostral border lies in the floor of sulcus alpha within the representation of the hind-

limbs. By contrast, analysis of the entire region surrounding sulcus alpha indicates that the border of the rostral area shifts onto the caudal bank of the prealpha gyrus as the sulcus is followed ventrorostrally. Two cytoarchitectonic fields can be recognized within Abbie's PH4 cortex. The rostral field lies principally on the caudal bank of the prealpha gyrus while an aganular field extends across the gyrus to sulcus beta. There are, thus, three cytoarchitec-

ECHIDNA THALAMIC PROJECTIONS

E C H I D N A 3L

165

\\\ A

B

C

D

- 2 m m

Fig. 9. Thalamocortical projections in echidna 3L. The format of this illustration is the same as that of Figure 7. In this particular case, the injection site is situated rostral to sulcus alpha and lies in the rostral field of SMI.

tonic fields surrounding sulcus alpha: an agranular field occupies most of the prealpha gyrus, the rostral field occupies the caudal bank of most of the prealpha gyrus, and the caudal field occupies the postalpha gyrus, extending across the floor of sulcus alpha and onto the caudal bank of the prealpha gyrus in the temporal lobe of the hemisphere.

It is, of course, of considerable interest to understand the functional significance of these three cytoarchitectonic fields. The first investigations to address the issue were

electrical stimulation studies of the “excitable” cortex using relatively large electrodes and stimulation currents (Abbie, ’38; Goldby, ’39). They reported a single area of the cerebral hemisphere which responds to electrical stimulation. It is situated almost completely on the prealpha gyrus and contains a complete representation of the body with the hindlimbs represented dorsocaudally and the trunk, forelimbs, head, and beak represented in sequence along the length of the prealpha gyrus. Abbie (‘38, ’40) thus con-

166 P.S. [JLINSKI

E C H I D N A 3R

A

B

C

D

E

2 1

2mm

5

. . +.. .'..

..5:.;< . ..: >:-. ... :>.

Fig. 10. Thalamocortical projections in echidna 3R. The format of this illustration is the same as that of Figure 7. In this particular case, the injection site is situated caudal t o sulcus alpha and lies in the caudal field of SMI.

cluded that the excitable cortex in Tuchyglossus corresponds to his PH4 cortex and has its rostra1 border in the floor in sulcus alpha. He did report that a small part of the hindlimb representation extends caudal to sulcus alpha.

Lende ('64) repeated the electrical stimulation studies and extended them to include an account of the somatosensory cortex using evoked potential techniques. He confirmed the existence of a region of the hemisphere situated on the postalpha gyrus that causes movements of specific muscles with low thresholds. As in the earlier studies, the

hindlimbs were found to be represented dorsally and the head medially, but it was not possible to determine the orientation of the head and limbs along the rostrocaudal axis of the sulcus. The evoked potential studies showed the existence of a somatotopically organized representation of the body surface along the postalpha gyrus. The hindlimbs were represented dorsalljr and the head medially. The apices of the limbs and beak were directed rostrally and extended across sulcus alpha onto the caudal bank of the prealpha gyrus. It was a.lso possible to elicit movements


with electrical stimulation within the somatosensory representation, but the thresholds were higher. Lende was unable to find a secondary somatosensory representation or SII. This arrangement of cortical sensory areas has been confirmed with evoked potential techniques by Allison and Goff ('72).

Within the limits of resolution imposed by the evoked potential techniques (e.g., Kaas, '821, Lende's ('64) results suggest that there is a precise relationship between the physiologically defined somatosensory-motor areas and the three cytoarchitectonic fields identified here. Lende's area MI corresponds to the highly agranular field on the prealpha gyrus. His area SMI corresponds to both the rostral and caudal fields. In particular, the rostral border of SMI in Lende's map extends onto the caudal bank of the prealpha gyrus, as does the rostral border of the rostral field. A caveat here is that, because of the large electrodes used, slow waves originating in the postalpha gyrus could have been recorded across sulcus alpha. However, the good cor- respondence between the cytoarchitecture and evoked po- tentials suggests that this is not the case. Thus, it appears- subject to this qualification-that the single body representation reported in SMI by Lende overlies two cytoarchitectonic fields.

Organization of the somatosensory thalamus in Tachyglossus

Ziehen ('08) and Abbie ('34) provided descriptions of the thalamus in Tuchyglossus based principally upon material prepared with myelin stains that was not adequate to define the cytoarchitectonic borders between nuclei. Thus, both of these authors recognized a single, large nucleus within the caudal thalamus (nucleus lateralis and nucleus ventralis, respectively) and did not report the distinction between the ventral posterior and ventral nuclei as defined here. Welker and Lende ('80) recognized ventrobasal and ventrolateral nuclei in the caudal thalamus, with the ventrolateral nucleus positioned rostral to the ventrobasal nucleus. These two nuclei, taken together, probably correspond to my ventral posterior plus ventral nuclei. The caudal part of Welker and Lende's ventrobasal nucleus clearly corresponds to the caudal pole of my ventral posterior nucleus, and the rostral pole of their ventral lateral nucleus is equiv- alent to the rostral pole of my ventral nucleus. The differ- ence between the two nomenclatures is that I recognize the rostral pole of the ventral posterior nucleus as lying dorsal to the ventral nucleus, based on differences in neuronal size and density, so that the ventrobasal nucleus of Welker and Lende includes the caudal pole of my ventral nucleus.

The delineation of the lateral and dorsal borders of the ventral posterior nucleus provide little difficulty. It is distinguished from the OP nucleus of Welker and Lende by the larger neurons present in OP and from the lateral geniculate nucleus b of Campbell and Hayhow ('71) by the smaller neurons present in that nucleus. It is, however, difficult to be certain where to draw the medial border of the ventral posterior nucleus. The medium-size neurons that characterize this nucleus extend some distance medially, and the relatively large HRP injection in echidna 1 retrogradely labeled neurons only in the lateral regions of this field of medium-sized neurons. It is therefore possible that the medial border of the ventral posterior nucleus as drawn in Figure 4 is placed too far medially and should conform to the medial border of the field of labeled neurons in Figure 7.

Because of these relatively subtle variations in neuronal size and density, any cytoarchitectonic scheme for the caudal thalamus in Tuchyglossu must be regarded as provi- sional pending the results of appropriate experimental studies. The data available to date are the three HRP cases reported here and the cases 64-229, 64-230, and 64-232 of Welker and Lende ('80). All six cases are consistent with the hypothesis that the ventral posterior nucleus projects to Lende's ('64) area SMI while the ventral nucleus projects to his area MI. The injection in echidna 1 involves both the rostral and caudal fields of SMI and retrogradely labeled neurons throughout the rostrocaudal extent of the ventral posterior nucleus. In particular, labeled neurons are situated dorsally, overlying the cytoarchitecturally defined ventral nucleus, at the rostral pole of the habenular complex. The pattern of labeling obtained in echidna 3L and 3R is consistent with that obtained in echidna 1. Fewer neurons are labeled, as would be expected from the smaller injection sites. The cortical lesion in Welker and Lende's case 64-232 is restricted to the caudal field of the body representation within area SMI. The lesion resulted in a patch of retrograde degeneration that slants ventrorostrally through the caudal thalamus. Its position resembles closely the column of retrogradely labeled neurons in echidna 3R, which received an injection situated slightly ventrolateral relative to the lesion in case 64-232. The lesion in case 64-229 involves most of the caudal field of SMI and appears to extend ventrally below sulcus alpha into the head representation of MI. It results in a band of retrograde degeneration in the caudal thalamus that slants from dorsocaudal to ventrorostral within the confines of the ventral posterior nucleus, consistent with the involvement of SMI. The patch of degeneration continues rostrally past the level of the habenular complex within the ventral nucleus, consistent with the involvement of MI. The lesion in case 64-230 involves almost all of the prealpha gyrus. It thus must have damaged both the rostral field of SMI and MI so retrograde degeneration is present in the ventral posterior nucleus from its caudal pole rostrally to beyond the level of the habenular complex, consistent with the involvement of the rostral field of SMI. Degeneration is seen ventrally in the thalamus rostral to the habenula within in the ventral nucleus, consistent with the involvement of MI. A small patch of degeneration in the ventral, caudal pole of the thalamus is not accounted for by this hypothesis.

In addition to suggesting the general hypothesis that the ventral posterior nucleus projects to both the rostral and caudal fields of the somatosensory cortex, the six experimental cases also bear upon several more detailed aspects of the organization of thalamocortical projections in Tuchy glossus. The first of these is that all three HRP injections in SMI of Tachyglossus produce only a single patch of retrogradely labeled neurons in the thalamus. By contrast, injections in the primary somatosensory cortex of placental mammals (e.g., Nelson and Kaas, '81; Spreafko et al., '81; Donoghue et al., '79) except some primates (e.g., Jones et al., '79) and marsupial (Donoghue and Ebner, '81; Haight and Neylon, '78, '81; Neylon and Haight, '83) mammals produce at least three patches of retrogradely labeled neurons: one each in the ventrobasal, intralaminar, and posterior nuclear complexes. An additional patch of retrogradely labeled neurons is seen in the ventral lateral nucleus in those species that show an overlap of all or part of the somatosensory and motor representations (e.g., Donoghue

168 Y.S. U1,INSKI

et al., '79). Since neurons were well labeled in the ventral posterior nucleus, it seems unlikely that the three injections (particularly the large injection in echidna 1) would have failed to demonstrate existing projections from other thalamic nuclei. The difficulties surrounding definition of the medial border of the ventral posterior nucleus make it difficult to eliminate the possibility that the intralaminar and posterior nuclear complexes are present immediately adjacent to the ventral posterior nucleus and were labeled by my injections, but the small injections in echidna 3 should have produced disjunct patches of labeled neurons and this is not the case. There is presently, then, no cytoarchitectonic or connectional evidence for the existence of intralaminar and posterior nuclei in the thalamus of Tachy- glossus. The implications of this possibility have been dis- cussed elsewhere (Ulinski, '85).

A second aspect of the organization of the thalamocortical projections has to do with the representation of body parts within the ventral posterior nucleus in Tachyglossus. The small lesion in Welker and Lende's case 24-232 and the injections in echidna 3 are all restricted to the cortical representation of a particular region of the body and suggest that a column of neurons extending some distance rostrocaudally through the ventral posterior nucleus project to a discrete region of SMI. This resembles the situation in placental (e.g., Saporta and Kruger, '77; Nelson and Kaas, '81) and marsupial (Haight and Neylon, '78, '81) mammals in which focal injections of HRP into the primary somatosensory cortex produce elongate rods of retrogradely labeled neurons in the ventrobasal complex. It is likely, then, that information from a particular locus on the body surface in Tachyglossus is mapped onto an elongate column of neurons within the somatosensory thalamus and then to a patch in the cortex. The available data are not sufficient to determine how the body surface is represented along the dorsoventral and mediolateral axes of the ventral posterior nucleus.

A final organizational aspect concerns the relation of neurons in the ventral posterior nucleus to the two cortical fields defined within SMI. All of the available data indicate that both the rostral and caudal fields receive projections from the ventral posterior nucleus, but it is not possible without double labeling studies (e.g., Spreafico et al., '81) to be certain whether or not a single thalamic neuron projects to both cortical fields. The two small injections in echidna 3 suggest that neurons situated dorsally in the ventral posterior nucleus project to the rostral field while those situated more ventrally project to the caudal field, but interpretation of these two cases is complicated by the fact that the injections were placed into regions of the cortex representing different parts of the body, so that the varia- tion in labeling might reflect in part a somatotopy within the ventral posterior nucleus. However, taken at face value, the data resemble the pattern of thalamocortical projections from the ventrobasal complex to areas 3a and 3b reported in macaque monkeys and cats. (Spreafico et al., '81; Friedman and Jones, '81; Kaas, '83). The ventrobasal complex in these species consists of a shell of neurons along the dorsorostral face of the complex that projects to area 3a and a central core of neurons that projects to area 3b. Careful microelectrode mapping studies in macaques (Jones and Friedman, '82) show that units in the dorsorostral shell respond preferentially to stimulation of the deep tissues of the body while core neurons respond to cutaneous stimulation. The organization of the ventrobasal complex and the

pattern of thalamocortical projections is thus consistent with a tendency for submodality segregation within areas 3a and 3b. Microelectrode mapping experiments in Tachy- glossus designed to determine how deep and cutaneous information are represented in the rostral and caudal fields and in the ventral posterior nucleus would then be an important contribution.

Organization of somatosensory-motor cortex in mammals

These findings suggest some striking similarities between area SMI in Tachyglossus and Brodmann's area 3 in monkeys and cats (reviews: Jones and Porter, '80; Kaas, '82, '83). First, both areas contain two cytoarchitectonic fields, the more anterior of which (rostral field and 3a) is generally characterized by an attenuated layer 4 and large pyramidal neurons in layer 5. The posterior field (caudal field and 3b) is characterized by a well-developed layer 4 and smaller, less numerous pyramidal neurons in layer 5. Second, the posterior field contains a complete representation of the body surface with the apices of the limbs point- ing anteriorly. Third, both areas are situated immediately posterior to an area of agranular cortex which is electrically excitable with low threshold (Lende's area MI and Brod- mann's area 4). Fourth, both areas receive projections from a nucleus in the caudal thalamus. Fifth, based on my ten- tative data in Tachyglossus, the rostral field in both areas receives projections from thalamic neurons situated rostro- dorsally while the caudal field receives projections from more ventrally placed neurons. In contrast to the situation in cats (Hassler and Muhs-Clement, '641, raccoons (Johnson et al., '821, and primates (Jones, '73, a distinct cell-poor zone corresponding to the outer layer of Baillarger does not seem to be evident in Tachyglossus.

Less information is available on the somatosensory cortex in placental mammals other than carnivores and primates, but there is evidence for the existence of fields corresponding to areas 3a and 3b in prosimian primates (Krishna- murti et al., '76; Carlsori and Welt, '80; Sur et al., '80a), tree shrews (Sur et, al., '80b), and possibly squirrels (Sur et al., '78). Based on the distribution of such areas across several phylogenetic lineages of placental mammals, Kaas ('83) has recently suggested that this configuration is char- acteristic of at least placental mammals. My findings on Tachyglossus prompt an extension of this concept to the hypothesis that the division of the somatosensory cortex into two fields is a primitive character that is common to mammals generally.

However, evaluation of this hypothesis is complicated by the tendency for some degree of overlap between the somatosensory and motor cortices to occur in all three subclasses of mammals. Thus, Bohringer and Row ('77) report a sub- stantial overlap between the hindlimb and trunk components of the somatosensory and motor cortices in Ornithorhynchus. There is evidence for either an extensive (Didelphis: Lende, '63a-c; Magalhaes-Castro and Saraiva, '71) or partial (Trichosurus: Haight and Neylon, '77, '78, '79, '81) overlap between the somatosensory and motor cortices in marsupial mammals. Also, either extensive (armadillo: Dom et al., '71; Royce et al., '75) or partial (rat: Hall and Lindholm, '74; Welker, '71) overlap between somatosensory and motor cortices occurs in placental mammals. When the appropriate connectional studies have been carried out (Donoghue et al., '79; Killackey and Ebner, '73; Donoghue and Ebner, '81) the areas of overlap have been


found to receive thalamic projections from both the ventrobasal and ventrolateral thalamic nuclei.

Since both a 4-3a-3b pattern and patterns with a t least some degree of overlap between somatosensory and motor cortices occur in all three subclasses of mammals, it is impossible at this point to conclude that either pattern is primitive in the sense of being present in the common ancestors of therian and nontherian mammals. Lende (‘691, for example, had suggested that the distinct sensory and motor cortices of primates evolved by a process of parcellation (e.g., Ebbesson, ’80) or segregation from a “sensorimotor amalgam” present in primitive mammals. However, the presence of an area 4-3a-3b pattern in Tachyglossus makes it equally likely that the pattern of three distinct cortical areas is primitive and that the overlap pattern is derived independently in each subclass of mammals.

Evaluation of the two possibilities will require further information. It will first be necessary to confirm with microelectrode recording techniques that the rostra1 area in Tachyglossus indeed does receive information from deep mechanoreceptors and can be legitimately compared to area 3a. Second, it will be necessary to collect information on the organization of the somatosensory and motor cortices in a greater range of placental and marsupial mammals to understand more completely the range and distribution of different patterns.

ACKNOWLEDGMENTS This work was supported by NSF grant INT 81-02645 and

funds from the School of Anatomy at the University of New South Wales. Debra Randall and Dorothy Crowder typed the manuscript. Maryellen Kurek helped to prepare the illustrations. I am indebted to Dr. W.I. Welker for use of celloidin-embedded material and to Drs. C.R.R. Watson and M.-C. Holst, with whom the HRP experiments were performed.

LITERATURE CITED Ahbie, A.A. (1934) The brainstem and cerebellum of echidna. Phil. Trans.

R. SOC. Lond. LBio1.J 224:l-74. Abhie, A.A. (1938) The excitable cortex in the monotremata. Aust. J. Exp.

Biol. Med. Sci. 16t143-152. Abbie, A.A. (1940) Cortical lamination in the monotremata. J. Comp. Neu-

rol. 72:429-467. Allison, T., and W.R. Goff (1972) Electrophysiological studies of the echidna,

Tachyglossus aculeatus. 111.-Sensory and interhemispheric evoked re- sponses. Arch. Ital. Biol. 11Ot185-194.

Bohringer, R.C. and M.J. Rowe (1977) The organization of the sensory and motor areas of cerebral cortex in the platypus (Ornithorhynchus anatinus). J. Comp. Neurol. 174;l-14.

Brodmann, K. (1909) Vergleichende Lokalisationslehre der Grosshirnrinde. Leipzig: Barth.

Campbell, C.B.G., and W.R. Hayhow (1971) Primary optic pathways in the echidna Tachyglossus aculeatus: An experimental degeneration study. J. Comp. Neurol. 143t119-136.

Carlson, M., and C. Welt (1980) Somatic sensory cortex (SMI) of the prosimian primate Galago crassicaudatus: organization of mechanoreceptive input from hand in relation to cytoarchitecture. J. Comp. Neurol. 189249-271.

Clemens, W.A. (1979) Notes on the monotremata. In J.A. Lillegraven, Z. Kielan-Jaworowska, and W.A. Clemens (eds): Mesozoic Mammals. Berkeley: Univ. California Press, pp. 309-311.

Dart, R.A. (1934) The dual structure of the neopallium: Its history and significance. J. Anat. 69:3-19.

Donoghue, J.P., and F.F. Ebner (1981) The organization of thalamic projections to the parietal cortex of the Virginia oppossum. J. Comp. Neurol. 198t365-388.

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. 183:647-669.

Dom, R., G.F. Martin, B.L. Fisher, A.M. Fisher, and J.K. Harting(1971)The motor cortex and corticospinal tract of the armadillo (Dasypus nouem- cinctus). J. Neurol. Sci. 14225-236.

Ehbesson, S.O.E. (1980) The parcellation theory and its relation to interspe- cific variability in brain organization, evolutionary and ontogenetic development and neuronal plasticity. Cell Tissue Res. 213t179-212.

Friedman, D.P., and E.G. Jones (1981) Thalamic input to areas 3a and 2 in monkeys. J. Neurophysiol. 4559-85.

Goldby, F. (1939) An experimental investigation of the motor cortex and pyramidal tract of Echidna aculeatus. J. Anat. 73t509-524.

Grifiths, M. (1968) Echidnas. Oxford Pergamon. Griffiths, M. (1978) The Biology of the Monotremes. New York: Academic

Press. Haight, J.R., and L. Neylon (1977) A craniocaudal gradient of representa-

tional overlap in the motor-somatic sensory neocortex of the brush-tailed possum, Trichosurus uulpecula. J. Anat. 124:531.

Haight, J.R., and L. Neylon (1978) The organization of neocortical projections from the ventroposterior thalamic complex in the marsupial brush- tailed possum, Trichosurus uulpecula: A horseradish peroxidase study. J. Anat. 126:459-485.

Haight, J.R., and L. Neylon (1979) The organization of neocortical projections from the ventrolateral nucleus in the brush-tailed possum, Tricho surus uulpecula, and the problem of motor and somatic sensory convergence within the mammalian brain. J. Anat. 129t673-694.

Haight, J.R., and L. Neylon (1981) An analysis of some thalamic projections to parietofrontal neocortex in the marsupial native cat, Dasyurus uiuer- rinus (Dasyuridaei. Brain Behav. Evol. 19:193-204.

Hall, R.D., and E.P. Lindholm (1974) Organization of motor and somatosensory neocortex in the albino rat. Brain Res. 6623-38.

Hanker, J.S., P.E. Yates, C.B. Metz, K.A. Carson, A. Light, and A. Rustioni (1977) A new specific, sensitive and noncarcinogenic reagent for the demonstration of horseradish peroxidase (HRP). Neurosci. Abstr. 3t30.

Hassler, R., and K. Muhs-Clement (1964) Architektonischer Aufbau des sensomotorischen und parietalen Cortex des Katze. J. Hirnforsch. 6t377- 420.

Hines, M. (1929) The brain of Ornithorhynchus anatinus. Philos. Trans. R. SOC. Lond. [Biol.] 217t155-287.

Johnson, J.I., E.M. Ostapoff, and S. Warach (1982) The anterior border zones of primary somatic sensory (S-I) neocortex and their relation to cerebral convolutions, shown by micromapping of peripheral projections to the region of the fourth forepaw digit representation in raccoons. Neurosci- ence 7t915-936.

Jones, E.G. (1975) Lamination and differential distribution of thalamic afferents within the sensory-motor cortex of the squirrel monkey. J. Comp. Neurol. 160:167-204.

Jones, E.G., and D.P. Friedman (1982) Projection pattern of functional components of thalamic ventrohasal complex upon monkey somatic sensory cortex. J. Neurophysiol. 48:521-544.

Jones, E.G., and R. Porter (1980) What is area 3a? Brain Res. Rev. 2 - 4 3 , Jones, E.G., S.P. Wise, and J.D. Coulter (1979) Differential thalamic rela-

tionships of sensory-motor and parietal cortical fields in monkeys. J. Comp. Neurol. 1835333-882.

Kaas, J.H. (1982) The segregation of function in the nervous system: Why do sensory systems have so many subdivisions? Contrih. Sens. Physiol. 7201-240.

Kaas, J.H. (1983) What, if anything, is SI? Organization of first somatosensory area of cortex. Physiol. Rev. 63906-231.

Killackey, H.P., and F. Ebner (1973) Convergent projection of three separate thalamic nuclei to a single cortical area. Science 179282-285.

Krishnamurti, A., F. Sanides, and W.I. Welker (1976) Microelectrode mapping of modality-specific somatic sensory cerebral neocortex in slow loris. Brain Behav. Evol. 13:367-383.

Lende, R.A. (1963a) Sensory representation on the cerebral cortex of the opossum (Didelphis uirginiana). J. Comp. Neurol. 121t395-403.

Lende, R.A. (1963b) Motor representation in the cerebral cortex of the opossum (Didelphis uirginiana). J. Comp. Neurol. 121:405-415.

Lende, R.A. (1963~) Cerebral cortex: A sensorimotor amalgam in the mar- supialia. Science 141t730-732.

Lende, R.A. (1964) Representation in the cerebral cortex of a primitive mammal. Sensorimotor, visual and auditory fields in the echidna (Tach- yglossus aculeatus). J. Neurophysiol. 27t37-48.

170 P.S. ULINSKI

Lende, R.A. (1969) A comparative approach to neocortex: Localization in monotrcmes, marsupials and insectivores. Ann. N.Y. Acad. Sci. 167262% 275.

Lillegraven, J.A., Z. Kielan-Jaworska and W.A. Clemens (1979) Mesozoic Mammals: The First Two-Thirds of Mammalian History. Berkeley: Univ. of California Press.

Magalhaes-Castro, R., and P.E.S. Saraiva (1971) Sensory and niotor representation in the cerebral cortex of the marsupial Didelphis azarue azarae. Brain Res. 31.291 -299.

Marshall, L.G. (19791 Evolution of metatherian and eutherian (mammalian) characters: A review based on cladistic methodology. 2001. J. Linn. Soc., 66:364-410.

Martin, C.J. (18981 Cortical localisation in Ornithorhynchus. J. Physiol. (Land.) ISnppl. I 2.?:383-385.

Mesulam, M.-M. (1978) TetrameLhyl benzidine for horseradish peroxidase non-carcinogenic blue reaction-product with su-

sualizing neural afferents and efferents. J. His- tochem. Cytochcm. 26: 106-117.

Nelson, R.d., and J.H. Kaas (1981) Connections of the ventroposterior nucleus of the thalamus with the body surface representations in cortical areas 3b and 1 qf the cynomolgus macaque (Macaca fascicularis). J. Camp. Neurnl. 1!42:611L64:3.

Neylon, I,., and J.R FIaight (1983) Neocortical projections of the suprageni- culate and postrrior thalamic nuclei in the marsupial brush-tailed possum, Trwhosuriis culparula iPhalangeridae), with a comparative commentary on the organization of the posterior thalamus in marsupial and placental mammals. d. Camp. Neural. 227~357-375.

Royce, G.J., G.F. Martin, and R.N. Dam (1975) Functional localization and cortical architecture in the nine-banded armadillo (Dasypus nouemcinr- tus). J. (‘amp. Neural. 164:495--522.

Saporta, S., and I,. Kruger (1977) The organization of thalamocortical relay neurons in the rill ventrobasal complex studied by the retrograde transport of horseradish peroxidase. J. Camp. Neurol. 174r187-208.

Schuster, E. (1910) Preliminary note upon the cell lamination of the cerebral

cortex of echidna, with an enumeration of the fibres in the cranial nerves. Proc. Roy. SOC. Land. B. 82113-123.

Smith, G.E. (1899) Further observations on the anatomy of the brain in the monotremata. J. Anat. Physiol. 33:309-342.

Smith, G.E. (1902) Descriptive and illustrated catalogue of the physiological series of comparative anatomy contained in the museum of the Royal College of Surgeons of England, Val. 2, 2nd ed. London: Taylor and Francis, pp. 110-510.

Spreafico, R., N.L. Hayes, and A. Rustioni (1981) Thalamic projections to the primary and secondary somatosensory cortices in cat: Single and double retrograde tracer studies. J. Comp. Neurol. 203:67-90.

Sur, M., R.J. Nelson, and J.H. Kaas (1978) The representation of the body surface in somatosensory area 1 of the grey squirrel. J. Camp. Neural. 179:425-450.

Sur, M., R.J. Nelson, and J.H. Kaas (1980a) The representation of the body surface in somatic koniocortNex in the prosimian Galago. J. Camp. Neu- ral. 189:381-402.

Sur, M., R.E. Weller, and J.H. Kaas (1980b) Representation of the body surface in somatosensory arsea 1 of tree shrews, Tupa~a glis. J. Camp. Neural. 194:71-95.

Ulinski, P.S. (1985) Neurobiology of the therapsid-mammal transition. In: J.J. Roth, E.C. Roth, P.D. MacLean and N. Hotton 111 (eds): The Ecoloby and Biology of Mammal-Like Reptiles. Washington, DC: Smithsonian Institution, (in press).

Welker, C. (1971) Microelectrode delineation of fine grain somatotopic organization of Sml cerebral neocortex in albino rat. Brain Res. 26r259-275.

Welker, W., and R.A. Lende (1980) Thalamocortical relationships in echidna (Tachyglossus aculeatus). In S.O.E. Ebbesson (ed): Comparative Neurol- ogy of the Telencephalon. New York: Plenum, pp. 449-481.

Ziehen, T. (1897) Das Centralnervensystem der Monotremen und Marsupi- alier. Semon’s Zool. Forsch. Jena: Fischer, Val. 3, part 2, pp. 1-187.

Ziehen, T. (1908) Das Centralnervensystem der Monotremen und Marsupi- alier. Semon’s Zool. Forsch. Jean: Fischer, Val. 3 , part 2, pp. 789-921.

thalamic projections to the somatosensory cortex of the echidna, tachyglossus aculeatus

Documents