ABSTRACT

REVIEW OF
THE PHYLOGENETIC POSITION
or THE TREE SHREW
(TUPAIA GLIS DIARD),
wx—‘m NIB—W "o" 85' E' "RVATI ONS
ON THE
ARTERIA CAROWIS INTERNA

by Erla Anne Steuerwald

This study was made to evaluate all avail-
able evidence, including that provided by the
branching pattern and distribution of the internal
carotid arterial system, to determine the problemati-
cal phylogenetic position of the tree shrews (family
Tupaiidae).

Arteries of the cranial region of Tupaia
glis Diard were perfused with colored latex or sili-
cone-rubber injection mass prior to dissection.

The observations indicated that the internal

carotid arterial pattern of Tupaia glis is as primi-

 

tive as that of any mammal, including monotremes and
marsupials. Tupaia glis deviates very little from

the early post-branchial embryological pattern found
in all mammals studied. The following features are

regarded as primitive:

2 Erla Anne Steuerwald
l. The branching pattern of the aortic arch is nearly
symmetrical, typically displaying a left, as well as a
right, brachiocephalic trunk.
2. The lingual artery is still the major branch of
the external carotid stem. Facial and superficial
temporal arteries are not well developed.
3. The pterygoid portion of the maxillary artery
passes through an external alisphenoid canal.
4. The promontory (cerebral carotid) branch is as
well developed as the stapedial branch of the in-
ternal carotid stem; both are of considerable diam-
eter. (This observation conflicts with all pre-
viously published observations on Tupaiidae.)
5. The superior ramus of the stapedial artery is
well developed, retaining continuity with its
meningeal and orbital ramifications.
6. The proximal inferior ramus of the stapedial
artery is usually present but attenuated.
7. Ethmoidal rami of the infraorbital, supraorbital,
and intracranial anterior cerebral arteries all
remain well developed; thus the latter vessel still
reaches the nasal septum by way of the ethmoidal rete
lying within the dura mater on the lamina cribrosa.
8. A transverse anterior communicating artery gives
rise to a common anterior cerebral artery that bifur-

cates dorsal to the corpus callosum.

3 Erla Anne Steuerwald
9. The bifurcation of the basilar artery usually
produces a common stem, on each side, for the SUperior
cerebellar and mesencephalic arteries.
10. The posterior cerebral and mesencephalic branches
of the internal carotid-basilar anastomosis (posterior
communicating artery) arise independently and still

display a fairly metameric arrangement.

Peculiarities of the carotid branching pattern
of TUpaia glis that are difficult to evaluate from
a phylogenetic point of view include the following:
1. The orbital ramus of the infraorbital artery joins
the ethmoidal branch of the superior ramus of the
stapedial artery in the orbit.
2. The internal carotid and vertebro-basilar systems
appear to share more or less equally in supply of the
brain.
3. The true ophthalmic artery (branch of the cranial
ramus of the cerebral carotid artery) is exceedingly
reduced in calibre and joins the ciliary trunk arising

from the stapedial artery (superior ramus).

Comparisons of the above-described features
with Recent Primates and Insectivora indicate that
Tupaiidae-are generally more primitive than either,
with no clear-cut indications of primate trends.

Review of the literature on all other known

4 Erla Anne Steuerwald
features of tUpaiids indicates an essentially primitive
mammalian status, with a considerable degree of re-
semblance to marsupials. The numerous similarities
with Primates are probably the result of two factors:
1) parallel adaptation to an arboreal way of life and
2) derivation from a common ancestral stock with

similar genetic potentialities for change.

REVIEW 01:
THE PHYLOGENETIC POSITION
OF THE TREE SHREW
(TUPAIA GLIS DIARD),
WITH NEW OBSERVATIONS
ON THE
ARTERIA CAROTIS INTERNA

By
Erla Anne Steuerwald

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTCR OF PHILOSOPHY
Department of Anatomy

1969

ACKNOWLEDGMENTS

Appreciation is due Dr. Daris R. Swindler,
Dr. M. Lois Calhoun, Dr. Thomas W. Jenkins, Dr.

J. B. Thomas, and Dr. J. Alan Holman for their
advice and especially for helpful suggestions
after reading the first draft of this manuscript.

Correspondence with Dr. T. W. Williams pro-
vided information that greatly improved the effective-
ness of blood-vessel injections.

The general interest of fellow graduate
students in the anatomy of the tree shrew frequently
provided encouragement, useful information, and
assistance. Mr. Duane Haines, Mr. Kenneth Holmes,
and Mr. Charles Wood were especially helpful.

The funds needed for purchase of animals
and materials were provided by a grant (GM 12635) to
Dr. Daris R. Swindler from the National Library of
Medicine, National Institutes of Health, Bethesda,

Maryland.

ii

thfilb

IIICIGAIIIIIIIIIIIIIIII IIIIaIIIIII IIIIIIIIIIIJIII J W-.- W

1067
31293 6 If‘fifil.n?’ g“!

 

 

 

 

 

 

 

 

 

This is to certify that the

thesis entitled
IEVIEN OF THE FHYLOGENETIC POSITION
OF TIE TREE SHEEN (FUPe‘IIfi. GLIS DI 3.33 ),
NITd NEW 03333 JATIOI" I3 ON THE
ARIERIA CIBOTIS IN'IE3I~:A

presented by
Erla Anne Steuerwald

has been accepted towards fulfillment
of the requirements for

P

Major professor

Date April 11, 1969

0-169

(1|:

 

 

 

 

ABSTRACT

REVIEW OF
THE PHYLOGENETIC POSITION
OF THE TREE SHREW
(TUPAIA GLIS DIARD),
WI T—H N'E‘w'o' BS ERVATI ONs
ON THE
ARTERIA CAROTIS INTERNA

by Erla Anne Steuerwald

This study was made to evaluate all avail-
able evidence, including that provided by the
branching pattern and distribution of the internal
carotid arterial system, to determine the problemati-
cal phylogenetic position of the tree shrews (family
Tupaiidae).

Arteries of the cranial region of Tupaia
glis Diard were perfused with colored latex or sili-
cone-rubber injection mass prior to dissection.

The observations indicated that the internal

carotid arterial pattern of Tupaia glis is as primi-

 

tive as that of any mammal, including monotremes and
marsupials. Tupaia glis deviates very little from

the early post-branchial embryological pattern found
in all mammals studied. The following features are

regarded as primitive:

2 Erla Anne Steuerwald
1. The branching pattern of the aortic arch is nearly
symmetrical, typically displaying a left, as well as a
right, brachiocephalic trunk.
2. The lingual artery is still the major branch of
the external carotid stem. Facial and superficial
temporal arteries are not well deveIOped.
3. The pterygoid portion of the maxillary artery
passes through an external alisphenoid canal.
4. The promontory (cerebral carotid) branch is as
well developed as the stapedial branch of the in-
ternal carotid stem; both are of considerable diam-
eter. (This observation conflicts with all pre-
viously published observations on Tupaiidae.)
5. The superior ramus of the stapedial artery is
well developed, retaining continuity with its
meningeal and orbital ramifications.
6. The proximal inferior ramus of the stapedial
artery is usually present but attenuated.
7. Ethmoidal rami of the infraorbital, supraorbital,
and intracranial anterior cerebral arteries all
remain well developed; thus the latter vessel still
reaches the nasal septum by way of the ethmoidal rete
lying within the dura mater on the lamina cribrosa.
8. A transverse anterior communicating artery gives
rise to a common anterior cerebral artery that bifur-

cates dorsal to the corpus callosum.

3 Erla Anne Steuerwald
9. The bifurcation of the basilar artery usually
produces a common stem, on each side, for the SUperior
cerebellar and mesencephalic arteries.
10. The posterior cerebral and mesencephalic branches
of the internal carotid-basilar anastomosis (posterior
communicating artery) arise independently and still

display a fairly metameric arrangement.

Peculiarities of the carotid branching pattern
of Tupaia glis that are difficult to evaluate from
a phylogenetic point of view include the following:
1. The orbital ramus of the infraorbital artery joins
the ethmoidal branch of the SUperior ramus of the
stapedial artery in the orbit.
2. The internal carotid and vertebro-basilar systems
appear to share more or less equally in supply of the
brain.
3. The true ophthalmic artery (branch of the cranial
ramus of the cerebral carotid artery) is exceedingly
reduced in calibre and joins the ciliary trunk arising

from the stapedial artery (superior ramus).

Comparisons of the above-described features
with Recent Primates and Insectivora indicate that
Tupaiidae are generally more primitive than either,
with no clear-cut indications of primate trends.

Review of the literature on all other known

4 Erla Anne Steuerwald
features of tupaiids indicates an essentially primitive
mammalian status, with a considerable degree of re-
semblance to marsUpials. The numerous similarities
with Primates are probably the result of two factors:
1) parallel adaptation to an arboreal way of life and
2) derivation from a common ancestral stock with

similar genetic potentialities for change.

REVIEW OF
THE PHYLOGENETIC POSITION
OF THE TREE SHREW
(TUPAIA GLIS DIARD),
WITH NEW OBSERVATIONS
ON THE
ARTERIA CAROWIS INTERNA

By
Erla Anne Steuerwald

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Anatomy

1969

- I
\

t ‘2‘

f;\

ACKNOWLEDGMENTS

Appreciation is due Dr. Daris R. Swindler,
Dr. M. Lois Calhoun, Dr. Thomas W. Jenkins, Dr.

J. B. Thomas, and Dr. J. Alan Holman for their
advice and especially for helpful suggestions
after reading the first draft of this manuscript.

Correspondence with Dr. T. W. Williams pro-
vided information that greatly improved the effective-
ness of blood-vessel injections.

The general interest of fellow graduate
students in the anatomy of the tree shrew frequently
provided encouragement, useful information, and
assistance. Mr. Duane Haines, Mr. Kenneth Holmes,
and Mr. Charles Wood were especially helpful.

The funds needed for purchase of animals
and materials were provided by a grant (GM 12635) to
Dr. Daris R. Swindler from the National Library of
Medicine, National Institutes of Health, Bethesda,

Maryland.

ii

TABLE OF CONTENTS

Page
I NIRwUCTI w 0 O 0 O O 0 O O O O O O O O O O O O 1
REVIEW OF THE LITERATURE . . . . . . . . . . . . 5

Geographic Distribution (5)
Genetic Features (6)
Behavior (7)
Reproduction (11)

Female Organs (12)

Male Organs (l4)

Gestation period (18)

Embryogenesis (l8)

Extra-embryonic Membranes

and Placentation (18)
Integumentary System (25)
Endocrine System (27)
Urinary System (29)
Respiratory System (29)
Digestive System (30)

Teeth (30)

Palatine Ridges (30)

Tongue (31)

Gall Bladder (33)

Small Intestine 33;

Large Intestine 34

Mesenteric Relations (36)
Sensory Structures (37)

Cutaneous Receptors 37

Olfactory Receptors 38

Ears (38)

Eyes anthccessory Structures (39)
Nervous System (43)

Brain in General (43)

Telencephalon (43)

Diencephalon (51)

Mesencephalon (54)

Rhombencephalon (56)

Corticospinal Tracts (57)

Peripheral Nervous System (60)

iii

Skeletal S stem--Axial Division (61)
Skull 61)
Dentition (79)
vertebral Column (83)
Sternum (84)
Ribs (84)
Skeletal System--Appendicu1ar Division (84)
Anterior Appendages (85)
Posterior Appenda es (87)
Muscular System (95
Muscles of the Head and Neck (96)
Muscles of the Abdominal Wall (99)
Muscles of the Anterior Extremities (lOO)
Muscles of the Posterior Extremities (107)
Circulatory S stem (110)
Serology 110)
Circulatory Organs (116)

MATERIAL AND METHODS . . . . . . . . . . . . . . 119

DESCRIFIIVE ANATWY O O O O O O O O O O O O O C 121

The Arch of the.Aorta (121)

The Common Carotid Artery (122)

The External Carotid Artery (123)
The External Carotid Arter
(Maxillary Division) (127
The External Carotid Artery
(Infraorbital Division) (130)

The Internal Carotid Artery

(Stem and Major Branches) (131)
The_Stapedial Artery
(Tympanic Portion) (132)
The Stapedial Artery--
Inferior Ramus (133)
The Stapedial Arter --Superior Ramus
(Cranial Portion) 133)
The Stapedial Arter --Superior Ramus
(Orbital Portion) 133)

The Arteries at the Base of the Brain (138)
The Vertebral Artery (140)
The Basilar Artery (140)
The Basilar Artery (Bifurcation) (142)
The Superior Cerebellar Artery (144)
The Mesencephalic Artery (145)
The Posterior Communicating Artery (146)
The Posterior Communicating Artery--
Posterior Cerebral Branch (147)
The Posterior Communicating Artery--
Anterior Choroidal Branch (147)

iv

The Circle of Willis--

Cerebral Carotid Entrance (148)
The Cerebral Carotid Artery--
Anterior Ramus (149)

The Anterior Choroid Artery (150)

The Middle Cerebral Artery (150)

The Anterior Cerebral Artery (152)

DISQJSSION C O O O O O O O O O O O O O O C O O O 155

The Arch of the Aorta (156)
The Common Carotid Artery
(Bifurcation) (161)
The External and Internal Carotid Stems
(Relative Size) (161)
The External Carotid Artery
(Relations with the Stapedial Branch of
the Internal Carotid ArterY) (163)
The Maxillary Artery
(Pterygoid Portion) (171)
The Infraorbital Artery--
Ramus Orbitalis (172)
The Internal Carotid Artery
(Stem and Major Ramification) (174)
The Internal Carotid Artery--
Medial Entocarotid Branch (175)
The Internal Carotid Artery--
Promontory (Cerebral Carotid) Branch (176)
The Internal Carotid System
(Relation to Tympanic Walls) (178)
The Internal Carotid Artery--
Stapedial Branch (179)
The Internal Carotid Artery--
Inferior Ramus of the Stapedial Branch (181)
The Internal Carotid Artery--
Superior Ramus of the Stapedial Branch (182)
The Middle Meningeal Artery (182)
Orbital Blood Supply (Relationship of
the Internal and
External Carotid Arteries) (187)
The Circle of Willis
(Relationship of the Vertebro-Basilar
and Cerebral Carotid Systems)' (196)
The Circle of Willis
(Nature of the Posterior Region) (202)
The Circle of Willis
(Mode of Anterior Closure) (206)
The Arteries of the Olfactory Region (211)

CONCLUSIONS..................213

SUMMRY O O O O O O O O O O O O O O O O O O O O 2 2 3

LIST OF REFERENCES 0 O O C O O O O O O O O O C O 224

vi

LIST OF FIGURES

Figure Page
1. Lateral View of the Cranial Region . . . . 128
2. Ventral View of the Brain and Middle Ear . 139

143

3. Ventral View of the Brain (Tree Shrew 3)

4. Carotid Arterial Branching Patterns . . . 167
168

vii

INTRODUCTION

In the earliest published sketch and descrip-
tion of tree shrews by Ellis in 1782, they were re-
ferred to as "squirrels" (Sciurus dissimilis),
probably due to their arboreal habitat and feathery
tails (Lyon, 1913).

That tree shrews were early regarded as
members of the order Insectivora is evidenced by
Diard's naming of a species as Sorex glis, in 1820.
In 1822, the name Tupaia was pr0posed by Raffles for
the genus that included the above-named species and
two others. It was common practice among Europeans
at that time to call these animals insectivores
(Lyon, 1913). This may have been based on their
variably insectivorous habits, dentition, pointed
snouts, and high level of physical activity.

During the nineteenth century additional
genera and species were described within the family
TUpaiidae (Lyon, 1913). By 1884, Leche (as cited by
Gregory, 1910) had decided that the families Tupaiidae
and Macroscelididae (elephant shrews) were so widely
separated from other Insectivora as to constitute a

distinct order Menotyphla. Haeckel, in 1866 (cited

2
by Simpson, 1945), established a more lasting precedent
by considering Menotyphla as a suborder, with Lipo-
typhla (all other insectivores) as the second sub-
order, of the order Insectivora.

Gregory (1910), although prOposing that
Menotyphla constitute a distinct order, indicated a
trend toward consideration of tupaiids as morphologi-
cally annectent forms between the order Insectivora
and Lemuroidea (of the order Primates). Although
these two groups may have separated very early,
developing in parallel under the stimulus of similar
arboreal habits, they still show true generic rela-
tionship.

Carlsson (1922) and Le Gros Clark (1925, 1926a,
1959) went even further, on the basis of morphological
evidence, to place the superfamilies Tupaioidea and
Lemuroidea within the suborder Prosimii of the order
Primates.

By 1945, the accumulated evidence had persuaded
Simpson to establish this taxonomic position by placing
the superfamilies Tupaioidea and Lemuroidea within
the infraorder Lemuriformes of the suborder Prosimii
of the order Primates.

Fiedler (1956) and Buettner-Janusch (1966)
elevated tree shrews to the status of a distinct

infraorder Tupaiiformes of the suborder Prosimii.

2
by Simpson, 1945), established a more lasting precedent
by considering Menotyphla as a suborder, with Lipo-
typhla (all other insectivores) as the second sub-
order, of the order Insectivora.

Gregory (1910), although prOposing that
Menotyphla constitute a distinct order, indicated a
trend toward consideration of tUpaiids as morphologi-
cally annectent forms between the order Insectivora
and Lemuroidea (of the order Primates). Although
these two groups may have separated very early,
deveIOping in parallel under the stimulus of similar
arboreal habits, they still show true generic rela-
tionship.

Carlsson (1922) and Le Gros Clark (1925, 1926a,
1959) went even further, on the basis of morphological
evidence, to place the superfamilies Tupaioidea and
Lemuroidea within the suborder Prosimii of the order
Primates.

By 1945, the accumulated evidence had persuaded
Simpson to establish this taxonomic position by placing
the superfamilies TUpaioidea and Lemuroidea within
the infraorder Lemuriformes of the suborder Prosimii
of the order Primates.

Fiedler (1956) and Buettner-Janusch (1966)
elevated tree shrews to the status of a distinct

infraorder Tupaiiformes of the suborder Prosimii.

3

Straus (1949) and Martin (1966) would have
established Tupaioidea as a separate order.

Recent evidence based on skull develOpment,
fossil teeth, central nervous system structure, and
fetal membrane morphogenesis has tended to reverse
this trend. Romer (1966) considered the family
Tupaiidae as the only living representatives of the
superfamily Leptictoidea, suborder Proteutheria,
order Insectivora. Romer (1966) emphasized that it
is very difficult to trace phylogenetic lines among
the primitive eutherian mammals of late Cretaceous
and Paleocene times. Often there are only fragmentary
remnants of dentition. Groups that later became
quite divergent were then very similar and closely
related. All placental mammals of the late Cretaceous
period might well be placed in a single family; and
even in the Paleocene epoch, within a common order
(Insectivora).

The order Insectivora includes not only the
modern, variably adapted types but also the basic
eutherian stock from which all modern eutherian
types have evolved. The family Leptictidae is the
basal stock from which most later, and all modern,
placental mammals have been derived, according to

the currently prevalent view among paleontologists

(Romer, 1966).

4

On this view, the problem becomes one of
determining the proximity of the basic primate and
tupaiid stocks to their leptictoid forebears.
To what extent do similarities among anthropoids,
lemuroids, and tupaiids represent parallelism,
convergence, or lineal ancestry? Nor are these
evolutionary patterns mutually exclusive. By defini-
tion, parallelism is "the evolution of related organisms
along similar lines so that they develop similar
adaptations.“ It is to be distinguished from conver—
ggggg, which is "the like response of quite unrelated
organisms to like ecologic opportunities." (Defini-
tions are quoted from Colbert, 1949.) Pertinent to
the phenomenon of convergence is the-observation
that ”the directions of possible mutations are not per-
fectly ad libitum but are limited to changes in such
directions as can find expression in the gene
complex actually at hand" (JEpsen, 1949).

The purpose of the present study is two-fold.
First, since an organism adapts to its environment as
a functioning whole, the literature on the various
facets of tUpaiid adaptation is reviewed. Second,
since a gross anatomical study of the internal
carotid arterial system of a tree shrew has apparently
never been published, results of such a laboratory
project on Tupaia glis are reported here in phylo-

genetic perspective.

REVIEW OF THE LITERATURE

Since there are no unquestioned tUpaiid
fossils, evidence regarding the phylogenetic position
of this family may well be supplemented by less direct
lines of evidence (van Valen, 1965; McKenna, 1966;
Szalay, 1968).

The tree shrew resembles the primitive
placental mammal as closely as any animal alive today.
Since different functional systems evolve at differ-
ing rates, some features remain more generalized
than others.

The genetic features and geographic distribu-
tion of Tupaiidae provide a theoretical basis for con-
siderable Speciation in the face of a fundamental con-

servatism of structure.

Geographic Distribution
Tree shrews range, geographically, from eastern
India on the west, to Palawar (Philippine Islands) on
the east; from southern China on the north, southward
to Java and many small islands off the southwestern
coast of Sumatra. No single genus exists throughout
this range (Lyon, 1913; Buettner-Janusch, 1966). Since

possible tUpaiid fossils remain questionable, the

5

6
significance of their Paleocene to Eocene distribu-

tion in EurOpe (Messelina and Adapisoriculus) and

 

Mongolia (Anagale) cannot now be interpreted (van
Valen, 1965; Romer, 1966; Szalay, 1968).

Genetic Features
Observation of tupaiid chromosome complements
shows a high proportion of acrocentric chromosomes,
indicative of a generalized condition (Bender and
Chu, 1963; Klinger, 1963). There appears to be
chromosomal polymorphism or even speciation among
morphologically similar members of a Species, such

as Tupaiagglis (Bender and Chu, 1963; Egozcue,

 

Chiarelli, Sarti-Chiarelli, and Hagemenas, 1968).
Any viable variants would probably have a better chance
of perpetuating themselves in the small, isolated
pOpulations so prevalent among tupaiids. Buettner-
Janusch (1966) pointed out that such ”mutual isolation
will have emphasized tendencies toward divergence.“

In discussion of his isoimmunization theory,
Goodman (1963) stated, however, that the possible
acquisition of a hemochorial placenta by tupaiids
would have tended to reduce their genetic plasticity,
at least during the period of gestation (of approxi-

mately forty-five to sixty days).

7
Behavior
Patterns of behavior observed among
Tupaiidae correlate well with other data used to
speculate on phylogenetic position.

The “resting attitude" of Tupaia glis re-

 

sembles that of primates with bushy tails. The
tail is frequently brought forward alongside the
body (Andrew, 1964).

Tupaia displays special behavior patterns for
cleaning the face, especially after eating. Similar
patterns are observed in most small mammals with
generalized limbs, such as Macroscelididae (elephant
shrews) and Sciuridae (squirrels). This specialized

movement has never been observed in Galago, Loris, or

 

Lemur (Andrew, 1964).

Regarding the source of facial expression, it
has been suggested that the major primitive function
of facial muscles was movement of vibrissae. Andrew
(1964) noted that although vibrissal movement is
rather extensively used for environmental scanning
in rodents and in such insectivores as elephant
shrews, it is absent in Tupaia and prosimians except
as incidental to other facial movements. This may
possibly be due to descent from a Sgggx-like ancestor
in which the entire snout was continually moved from

side to side during exploration. In such a situation,

8
independent movement of vibrissae would only make
tactile data more difficult to interpret.

Andrew (1964) pointed out that although Urogale
(Philippine tree shrew) has stereosc0pic overlap of
fifty to sixty degrees, it may not actually use such
vision to any appreciable degree. Observation of
Tupaia shows that the animal looks frontally in
curiosity but generally uses monocular vision. All
primates, including Galago and Lemur, use stereosc0pic
vision almost all the time. Andrew (1964) also noted
that reduction of the muzzle is seemingly not
correlated with acquisition of stereosc0pic vision.
Simons (1962) traced the known phylogeny of stereo-

scopic vision in fossil Primates. Plesiadapis (the

 

oldest primate skull described to date), although
lacking a postorbital bar, closely resembles modern
tree shrews in the lateral direction of the orbits.
There is no high correlation between color
vision and advanced stereosc0pic vision among primates

(Andrew, 1964). Tree shrews (Tupaia_glis) display

 

effective color vision (von Tigges, 1963, 1964; Noback
and Shriver, 1967).

Insectivorous forms make less Specialized use
of the incisor teeth than do most mammals. The incisors
may seize or tear, but the longer canines are better

suited to such tasks. In the absence of cheeks, even

9
premolars or molars may be so used. Since the in-
cisors are not essential to feeding, it has been pos-
sible for them to become procumbent and specialized
for grooming in Tupaiidae. In Lemuroidea and Lorisidae
the procumbent incisors have become solely related to
grooming, with important socializing consequences (An-
drew, 1964). Simons (1962) cited evidence from both
fossil and living prosimians and tUpaiids that this
tooth-scraper has arisen by parallelism in at least
three stocks.

Tupaias may invent rather complex motor play,
although showing a strong prOpensity for stereotyped
activity (Vandenbergh, 1963; Kaufmann, 1965; Sorenson
and Conaway, 1966).

No living Primates observed (including
TUpaiidae) are as solitary as Insectivora. The degree
of sociability observed by Andrew (1964) among tupaias
(Tupaia glis) is greater than that among rats and
mice. It contrasts more markedly with the intolerance
displayed among elephant shrews (Macroscelididae).

The solitary habits of the latter resemble those of
most other insectivores.

Size and pattern of social grOUpS varies some-
what among Tupaia Species (Sorenson and Conaway, 1966).
Lorisid societies are at a level of organization

similar to those of Tupaiidae. Galago crassicaudatus

10
is in some ways less social than Tupaia. With reSpect
to mutual grooming, however, Andrew (1964) observed

that both Galago and Perodicticus (potto) show greater

 

sociability. Although this behavior is rare in most
Tupaia Species, males of all species attempt to groom
females whenever the latter near estrus (Sorenson and

Conaway, 1966). Tupaia longipes (Sorenson and Conaway,

 

1966) displays a social pattern that includes female-
female consorts. Consorting females groom each other
daily. Microcebus (mouse lemur), in rarely displaying
mutual grooming behavior, is not far from a tUpaia-like
condition. Lemuroids, especially the diurnal species,
tend to be more social than lorises (Andrew, 1964).
Sprankel (1961) described a form of scent
signal called ”urine washing." Andrew (1964), how-
ever, found the tUpaias' behavior a bit “too unspecific
in form" to homologize it. A somewhat similar behavior
pattern occurs during courtship in Microcebus (Andrew,
1964).
Scent marking by means of cutaneous gland
secretions is shared by Tupaia and Lemuroidea with
some insectivores (Soricidae and Solenodon). Since
the males of the insectivore and Lemur species have
flank glands, while Tupaia bears its odoriferous

glands on chin and throat, the movements by which

11
the scent is applied to the female during courtship
must also have evolved separately in Tupaiidae
(Sprankel, 1961; Andrew, 1964).

COpulation patterns in TUpaia glis resemble
those of Galago crassicaudatus in a number of ways
(Andrew, 1964).

Martin (1966) observed the very unusual
breeding behavior displayed by Tupaia (species not
indicated). The male is entirely reSponsible for
nest-building activity. This is unique among mammals.
The female suckles the young only once every two
days. Long intervals between brief suckling visits
by the mother have also been observed in rabbits
(Lagomorpha).

Young tupaias grow rapidly, reaching adult
size by the age of three months and attaining sexual
maturity at four months. This speed of maturation
contrasts markedly with all Primates (Sprankel,

1960; Martin, 1966).

Reproduction
According to Butler (1964), it is generally
agreed that reproductive organs are less likely to
show structural changes dependent on differing environ-
ments than are most other organ systems.
In general, the genital organs of TUpaiidae

are quite primitive for Eutheria. Strongest

l2
resemblances are with Metatheria on the one hand, and
with higher Primates on the other. Insectivora, as
well as Lemuroidea, appear to be somewhat deviant
and specialized (Wood Jones, 1917; Eckstein, 1958).
Within Tupaiidae the genital system is some-

what less Specialized in Dendrogale than in Ptilocercus,

 

with both correspondingly different from the seemingly

more advanced Tupaia (Davis, 1938).

Female Organs

 

An 05 clitoridis is present in the tip of
the clitoris of Tupaia javanica. This may be a prim-
itive mammalian feature since it is found in some
insectivores (Talpa, Erinaceus) and in some prosimians
(LSEEE9 Cheirogaleus, Galago) (Pehrson, 1914).

In Tupaia glis and Tupaia ferruginea the

 

penis-like clitoris is elongate and pendulous (Hill,

1958a). Although the urethra of Tupaia javanica does

 

not perforate the clitoris, Pehrson (1914) did observe
a ventral groove throughout its length, continuous
with the urogenital sinus. Lorisiform lemurs have

the entire length of the clitoris tunnelled by the
urethra (e.g., Galago). In lemuriform lemurs the
urethral orifice lies at the root of the clitoris or
may extend a short distance along its length (Butler,

1964). These variants indicate a rather primitive

13

mammalian condition of urethral relations (Harms,
1956). In all living Primates the urethra and vagina
Open separately into the vestibule. The urethra and
vagina of Tupaia join, as in many other primitive
Eutheria (Martin, 1966).

Small litters (one to two individuals) are
the rule among Primates (Petter-Rousseaux, 1964).
Reduction in litter size may well be an adaptation
to arboreal life. For tupaias, the long inter-
suckling intervals would also probably limit litter
size (Martin, 1966). The range is from one to four
individuals per litter (Lyon, 1913; Wood Jones, 1917;
Hendrickson, 1954; Sprankel, 1959). There has evi-
dently been no immediate effect on anatomy of
reproductive organs, except possibly to reduce the
number of mammary glands (Lyon, 1913; Wood Jones, 1917;
Carlsson, 1922). Insectivores tend to have larger-
sized litters, with a recorded range from two
(Macroscelididae) to twenty (Tenrecidae) (Asdell,
1964).

The reduction in number of mammary glands
(one to three pairs), if related to the long inter-
suckling periods in Tupaia, might well be a parallel-
ism with the reduction in Primates, in which there is
protracted gestation and postnatal care (Lyon, 1913;
Wood Jones, 1917; Carlsson, 1922; Harms, 1956;

l4
Asdell, 1964; Butler, 1964; Martin, 1966).

Elephantulus (Macroscelididae) demonstrates
a type of uterine bleeding that may represent a
functional parallelism with primate menstruation
(Harms, 1956; Asdell, 1964).

There is no correlation between the presence
of coiled arterioles in the uterine wall and menstrua-
tion. The distribution of the coiled arterioles in
tupaiids differs from that in Old World primates
'(including humans and prosimians) in that these
arterioles do not penetrate into the ”functional
zone" of the tupaiid endometrium (Luckett, 1967).

Luckett (1967) cites some evidence that
captive tree shrews Show slight loss of endometrial
pad tissue following pseudopregnancy. This may be
comparable to the slight menstrual bleeding observed
in New World monkeys. Pseudopregnant cycles are
probably rare among wild tree shrews. Little is
known in general, however, about possible micro-
scopic menses in Insectivora, Prosimii, or other

mammalian orders.

Male Organs

The main features of the male ducts and acces-
sory glands in tree shrews are of generalized primate
form (Garrod, 1879; Engle, 1926; Le Gros Clark, 1926a;
Eckstein, 1958; Butler, 1964). Although reproductive

l5
organs are reputed to be conservative, phylogenetic
patterns are very difficult to follow in the refer-
ences cited above. This may be due to preservation of
series of mutations resulting from lack of any marked
selection pressure. Any adaptive trends of long
duration would probably yield patterns easier to
interpret. Histochemical and histogenetic studies
are needed to resolve numerous phylogenetic problems
(Price, 1963; Butler, 1964).

In Tupaia the testes have permanently de-
scended into a hairy scrotum. In Ptilgcercus, how-
ever, descent is intermittent (seasonal), suggesting
a restricted breeding season (Le Gros Clark, 1926a;
Eckstein, 1958). Garrod (1879) observed that in a
male specimen of Tupaia belangeri the large cauda
epididymis alone descended into the rudimentary
scrotum; the testes did not.

The patent but narrowed processus vaginalis
of Tupaia also occurs in lower Primates, including
most monkeys (Eckstein, 1958).

Most Insectivora show permanent retention of
the testes intra-abdominally, with no true scrotum
present. Testes in 13123 are intra-abdominal gen-
erally, although descending into small peritoneal
pouches during the breeding season (Eckstein, 1958;

Asdell, 1964).

16

Gregory (1910) remarked that Tupaiidae stand
higher than “any other Insectivores” in permanent
descent of the testes into a hairy scrotum, as Opposed
to temporary descent into a ”cremaster sack”; yet
he found Tupaiidae more primitive than typical
eutherians in the following features: (1) the two
testes are not appressed, each being enclosed within a
separate cremaster sack in the scrotum; (2) the testes
fail to descend beyond the penis, remaining anterior

to it, as in MacrOpus (Marsupialia).

 

There is permanent descent of the testes in
practically all Primates; exceptions include marmosets
and some lemurs (Eckstein, 1958).

In Tupaiidae the position of the testes after
descent into the scrotum is rather unique among euther-
ian mammals, suggestive of an intermediate phase
between Metatheria and Eutheria (Wood Jones, 1917).
Carlsson (1922) noted this position even in Inpaia
embryos. Since the testes lie both cephalic and
lateral to the body of the penis, the position is
described by some as parapenial (Kaudern, 1910; Le
Gros Clark, 1926a). Although usually described as
prepenial, the position has also been likened to the
mons veneris area of the female. In Metatheria the
scrotal sacs lie cephalic to the site of the mons

veneris area. In Eutheria, as a rule, the testes

1?
lie caudal to the mons veneris area (Hill, 1958a),
except in Lagomorpha (Martin, 1966).

Among Insectivora, including Macroscelididae,
the testes lie in a post-penial position (Gregory,
1910).

Although most Primates show the post-penial
position, marmosets and certain lemurs diSplay a
parapenial position of the scrotal sac. Scrotal
sacs may be parapenial or even prepenial in some
Species of gibbon. It is probable that this seem-
ingly primitive condition must be a secondary
adaptation in the gibbon species (Eckstein, 1958).
This may also be true in tUpaiids and lagomorphs
(Martin, 1966).

The penis of Tupaiidae is semi-pendulous (Wood
Jones, 1917; Eckstein, 1958). Hill (1958a) observed
that the primate penis is invariably pendulous in
some degree. Primates thus differ from the majority
of mammals, with the exception of bats and bears.

The penis of Tupaia is somewhat bent in a
caudal direction. In Insectivora (disregarding
Chrysochloris), the penis shows sigmoid flexure.
Similarity to insectivores is no greater in the embryo,
according to Carlsson (1922).

An 05 penis (baculum) is found in some insec-

tivores, most primates, and many other scattered groups

18
of placental mammals (Eckstein, 1958; Hill, 1958a);
thus it may be a primitive eutherian feature. Its
absence in Tupaiidae and numerous other mammals would

then be due to secondary loss.

Gestation Period

Gestation periods are long in lemurs, com-
pared with other mammals of similar size. Tupaiidae
’appear to be intermediate between Insectivora and
Primates, showing a trend toward prolongation
(forty-five to sixty days) (Snedegar, 1949; Hendrick-
son, 1954; Sprankel, 1959; Asdell, 1964).

Embryogenesis

It is difficult to evaluate, at this time,
the phylogenetic significance of early embryogenesis
in Tupaiidae. There are several seemingly unusual

features. Only one species, Tupaia javanica, has been

 

investigated. Little information is available on most
Insectivora on the one hand, and many Prosimii on

the other (Starck, 1956).

Extra-embryonic Membranes and Placentation

Extra-embryonic membranes and placentae in
eutherian mammals must ultimately derive from membranes
of the reptilian (cleidoic) egg; thus one would expect
a primitive eutherian mammal to display: (1) a large

yolk sac, (2) a choriovitelline placenta or homologue,

l9
(3) a large allantois, (4) amnion formation by folding
(Butler, 1964).

Hill (1965) believed that the basic mammalian
stock from which Primates diverged must have had an
arrangement similar to that observed in modern
monotremes and marsupials:

The outer wall of the embryonal formation con-
sisted, over the lower hemisphere or thereabouts,
of omphalopleure, vascularized in part or
throughout by the vitelline or yolk-sac vessels,
and over the remainder, vascularized by the um-
bilical vessels of the allantois. The vestigial
Shell membrane and the zona having already dis-
appeared during the earlier stages of develop-
ment, the primitive placenta was simply constituted
by the close apposition of these two regions with
the vascular lining of the uterus, and so was
partly allanto-chorionic, partly omphalOpleural in
nature. Here, then, we have the simplest possible
type of appositional placenta, the trophoblast be-
ing as yet a single layer, functional in absorp-
tion and gas exchange, but exhibiting no prolifera-
tive or invasive properties (Hill, 1965).

The occurrence of a diffuse epitheliochorial,
non-deciduate placenta in groups as different as
Lemuroidea, Pholidota, Artiodactyla, Perissodactyla,
Proboscidea, Cetacea, and Sirenia could be interpreted
as due to persistence of an ancestral condition
(Grosser, 1933; Hill, 1965).

Among Insectivora an epitheliochorial
placenta occurs in Scalopus aquaticus (Talpidae)
(Mossman, 1939). Epitheliochorial paraplacental
arrangements are found in Centetidae, Potamogalidae,

and Macroscelididae (Starck, 1956).

20

Since an embryo gains access to a uterus with
an intact epithelium, there must be an initial
epitheliochorial connection. Yet, probably already
in Promammalia, several placental arrangements
appeared. Reptiles Show epitheliochorial, syndes-
mochorial, and endotheliochorial types of placentae,
including a major portion of one type and associated
accessory structures of another type (Starck, 1956).
Similar combinations are found in Insectivora, in
which the hemoendothelial type may also occur.
Echinosorex, one of the most generalized of the lower
Erinaceidae, shows a hemoendothelial, discoidal
chorioallantoic placenta, with a choriovitelline
placenta lying peripheral to it (Meister and Davis,
1953).

Since the hemochorial type of placenta occurs
in some insectivores, Hubrecht (1899) and Wislocki
(1929) regarded this invasive type as primitive for
eutherian mammals. On this view, the non-invasive
epitheliochorial type (as in lemurs) would exemplify
specialized reduction.

Tupaia species possess an endotheliochorial
placenta, according to van der Horst (1949), Hill
(1965), and Luckett (1967). Other workers (Hubrecht,
1899; de Lange, 1933; Mossman, 1937; Strauss, 1942;

Meister and Davis, 1956) have classified it as

21

hemochorial, at least near term. Since he supervised
Luckett's (1967) work, Mossman's view has changed.

Also in dispute is the presence and degree of
development of a choriovitelline placenta at an
early stage of gestation. Although Meister and Davis.
(1958) found no evidence of a yolk-sac placenta at
any stage in Tupaia tana, Starck (1956), Hill (1965),
and Luckett (1967) have found well-formed, functional
choriovitelline placentae in several Species of

Tupaia, Lyonogale, Urogale, and Dendrogale. The

 

yolk-sac placenta persists to the early limb—bud stage.
Features of the tupaiid placenta and fetal
membranes that are regarded as primitive are listed
below:
1. The depth of implantation is superficial; there
is no decidua capsularis (Mossman, 1937).

2. In Tupaia minor, Meister and Davis (1956) found

 

the uterine epithelium to remain intact beyond the
placental margin. Over the entire decidua parietalis,
chorionic epithelium is in contact with the uterine
epithelium.

3. Hill (1965) stated that ”the occurrence of
functionally active glands in the endometrium of the
pregnant uterus is a feature which links the placenta
of Tupaia with those of the epithelio-chorial type.”

Luckett (1967) found these glands active in secretion

22
throughout pregnancy.

4. In Tupaia no fetal or maternal giant cells appear
at any stage of development (Hill, 1965).

5. At parturition, in Tugaia, the placenta is not
expelled as an after-birth, according to Starck (1956).
Rather, as in 13123, it is locally resorbed ("contra-
deciduate" type). Luckett (1967), however, refutes
this view.

6. Amniogenesis is by folding (cavitation to form
a primitive cavity, followed by folding) (Mossman,
1937). This also occurs in [alga (Insectivora) and
in Galago senegalensis (Prosimii) (Starck, 1956;
Butler, 1964). I

7. The yolk sac is well develOped and completely
vascularized. There is evidently a true yolk-sac
placenta for a time.

Among mammals a large, free, vascular yolk
sac is characteristic of Carnivora, Tupaiidae, and

Strepsirhini (e.g., Galago senegalensis) (Starck,

 

1956; Butler, 1964; Luckett, 1967). Centetidae
(Insectivora) and most Primates have a reduced or ves-
tigial, free, vascular yolk sac. Most Insectivora
have inverted yolk sacs (Luckett, 1967).

Starck (1956), Meister and Davis (1958), and
Hill (1965), regarded the tupaiid yolk sac as waning

Since omphalopleural mesenchyme shows little or no

23

tendency to penetrate the "villi,” whereas allanto-
chorionic mesenchyme does.

8. Refuting Meister and Davis (1958), Luckett
(1967) found the allantois of tUpaiid Species to be

a large, flattened vesicle persisting in the late
stages of pregnancy. It fuses with the chorion
to cover most of the surface of the placental discs.

A large, lobed allantois has been observed in

Galago senegalensis (Strepsirhini) (Butler, 1964).

 

9. Trophoblastic villi are rather poorly develOped.
Hill (1965) found them to be variable in size and

Shape in Tupaia dorsalis.

 

10. In Dendrogale, Luckett (1967) observed small

 

depressions in the smooth chorio-allantoic membrane.

He believed these to be homologous to the paraplacental
absorptive areas found in all groups with an epi-
theliochorial placenta, as well as in Carnivora,
Talpidae, and Tubulidentata.

Luckett (1967) found the development of tUpaiid
fetal membranes to be most similar to that in Carniv-
ora, Talpidae, and Soricidae (in the order given).
Common to these groups are an endotheliochorial pla-
centa and retention of both a large allantoic ves-
icle and yolk sac. It is the morphogenetic rela-
tionships of the most conservative features of fetal

membranes that are of phylogenetic significance

24

and not simply their appearance in late stages of
pregnancy. Since these are primitive features, the
resemblance is believed to be due to conservatism.

Mossman (1937) believed that the earliest
placentation in mammals was diffuse epitheliochorial
in type. Correlation of evidence from the fossil
record with that of the morphogenesis of fetal
membranes has recently led Mossman to postulate a
dichotomy of fetal membrane characteristics in the
primitive eutherian stock (late Cretaceous to early
Paleocene) (Luckett, 1967). The ancestral insecti-
vore-like stock giving rise to all taxa with a
deciduate placenta presumably had a broad zonary,
labyrinthine endotheliochorial placenta, with large
allantoic vesicle and large, free, vascular yolk sac.
Among living mammals this situation is found only in
Carnivora and TUpaiidae. No problems of homology
are posed by the double discoid placentae of
tupaiids since carnivore placentae vary from complete
zonary, through incomplete zonary, to double and
single discoid types (Luckett, 1967). Repression of
portions of the broad zone would lead to retention of
localized disc(s).

It is further postulated by Mossman and
Luckett (1967) that the order Condylarthra (basic

ungulate stock) retained the more primitive, diffuse

25
epitheliochorial placenta, with large allantois and
yolk sac. Since there is no good fossil evidence
that primitive lemuroids ever gave rise to anything
but other lemuroids, having evolved separately since
late Cretaceous or early Paleocene times, they may
well have retained the primitive, diffuse epithelio-
chorial placenta of ancestral mammals. On this
view, lemuroids and primates may Show convergent
evolution in other features for adaptation to an

arboreal habitat.

IntegumentaryVSystem

Evidence from histology of the tupaiid skin
is indicative of probable primitive insectivore
status. Montagna, Yun, Silver, and Quevedo (1962)
demonstrated that, unlike that of primates, the tree
Shrew's skin is rather thin; the sebaceous gland
cells contain small, rounded lipid drOplets of
rather uniform size; hair follicles are arranged in
rows and Show no glycogen and phOSphorylase in the
quiescent phase. Some of these features are Shared
with other primitive orders, suOh as bats and rodents.
More information is needed concerning the gland types
found in the hairy skin of Insectivora, DermOptera,
and Chiroptera before one can evaluate the signifi-
cance of the presence of eccrine sweat glands in

Tupaia glis and in Anthropoidea, despite their absence

26

in Prosimii generally.

The phylogenetic significance of the volar
integument has been discussed by Biegert (1961,
1963). In adaptation to an arboreal life, prosimians
and tUpaiids (no other modern insectivores being tree
dwellers) have develOped three-dimensional vision and
touch. As eXpected, Tupaiidae approximate a primitive
primate model showing distinct advances over all
Insectivora. While volar pads appeared in very early
mammals and friction skin appeared secondarily by
convergence in several mammalian lines, all Recent
primates are distinguished by the rather stable pattern
of characteristic ridges. The configuration of the
embryonic pad determines the ridge pattern thereby
established for life. This pattern is shared with
all primates, only by Tupaiidae. If due to parallel-
ism, there seems to be some common genetic background.
Elephant shrews (Macroscelididae, the second meno-
typhlous family) Show a far more primitive condition
resembling that of Reptilia. (In many other
features Macroscelididae Show greater specialization
or divergence from the basic mammalian stock; thus
it appears that Tupaiidae and Macroscelididae must
have diverged from each other very early.)

A resemblance between Tupaia and Lemur lies

in the pattern of flexion creases on palmar and plantar

27
skin surfaces (Biegert, 1961).

In all primates, at least one pair of digits
bears nails. As in all primitive and generalized
mammals, tupaiids possess claws on all the digits.
Although McKenna (1963) recently questioned the
status of Anagale (lower Oligocene genus of eastern
Asia) as a fossil tUpaioid, it was long considered
so (Simpson, 1931). Although the menus in Anagale
has claws, ”the terminal phalanges of the pes are
flattened and spatulate distally rather closely
'approximating the distal phalanges of some primitive
lemuroids" (Le Gros Clark, 1936).

The arrangement of hairs (in tufts associated
with scales) on the tail of Ptilocercus, the pen-
tailed tree shrew, is a very primitive mammalian
characteristic (Gregory, 1910; Le Gros Clark, 1926a).

Among mammals, only most Marsupialia possess
the full complement of facial vibrissae in typical
arrangement (Pocock, 1914). The Tupaiidae retain a
nearly complete and well-develOped facial complement.
Within many orders, including Insectivora and Primates,
vibrissae show reduction, especially in the higher,

derivative species (Pocock, 1914; Carlsson, 1922).

Endocrine System
Histologically, the eminentia mediana (of

hypothalamus) in most mammals, including most

28
non-hominoid primates, is a simple, two-layered
structure. Tupaia glis shows incipient stages of
the primate trend toward a breaking up of the external
layer into ”separate perivascular regions around the
capillaries of the portal vessels in the interior
of the eminentia" (Hanstrgm, 1958).

Another feature shared by TOpaiidae and some
Primates, but not by any of six genera of Insectivora
examined, is a common caudal fold formed by pars dis-
talis and pars intermedia (Hanstrgm, 1958).

A feature of the tupaiid suprarenal glands
indicative of the primitive mammalian condition
(shared with MarSUpialia and Insectivora, e.g.) is
the variation in their position. Constancy of po-
sition is greater, even in lemurs, among primates
(Le Gros Clark, 1926a; Bourne, 1936, 1949; Bachmann,
1958).

Although unusually broad for a primitive

mammal, the adrenal cortex of Tupaia montana shows

 

an indefinitely fasciculated arrangement of cells.
(All mammals show some degree of fasciculation.) No
zona reticularis has been nOticed, although present
in most mammals. In lack of definition of cortical
zones, lemurs Show little advance over tree shrews.
Tarsiers and the higher primates Show three clearly

defined zones. In all Insectivora examined (three

29
genera), the three zones typical of higher Primates
and most Mammalia are present. It seems that the lack
of differentiation in TuEaia species is a primitive
feature; yet it may be secondary (Bourne, 1949;

Bachmann, 1958).

Urinary75ystem
In major features--smooth contour and uni-
papillate medulla-~Tupaiidae resemble primitive
Mammalia, such as Monotremata, Marsupialia, Insec-
tivora, and Rodentia, as well as most Primates.
The apex of the medulla is intra-ureteric, as in some
lower mammals including Lorisiformes and Tarsiiformes

among Primates (Straus and Arcadi, 1958).

Respiratory System

 

Although the elongated upper portion of the
larynx projects into the naSOpharynx in Insectivora,
Tupaia, and many lower Primates, the structure of
this vestibular region differs among them. In Insec-
tivora (e.g., Erinaceus, the EurOpean hedgehog) the
wallof this funnel-like projection is composed of
the middle portion of the epiglottis and the aryepi-
glottic folds. In Tupaia, Loris, Perodicticus (potto),

 

and Galago, however, the lateral epiglottic folds are
well develOped, with the smaller aryepiglottic folds
lying in the inner wall (Starck and Schneider, 1960).

3O
Digestive System
There is considerable indirect evidence that
primitive placental mammals were small and semi-
arboreal. The foods readily available to such animals
would be eggs, nestlings, insects, and fruits.
Such insectivorous-omnivorous habits would prevent
over-specialization of any one part of the digestive
tract. The tupaiid diet of insects and fruit may
be considered a primitive marSUpio-placental feature.
Lipotyphlous insectivores early became specialized
for a more exclusively insectivorous diet; primates,
including lemurs, became more omnivorous or herbivo-

_rous (Sonntag, 1925; Wharton, 1950, Grassé, 1955).

Teeth

Tupaiid cheek teeth are rather typically
insectivorous. In the arboreal Tupaiidae and Lemur-
oidea, the lower incisors are procumbent in some
degree, and these animals use their hands to hold
food while eating (Lyon, 1913; Buettner-Janusch and
Andrew, 1962).

Palatine Ridges

Schultz (1958) stated that palatine ridges are
found in practically all mammals, although number and
size vary. The probable primitive condition persists

in many marsupials and insectivores. The opossum,

31
e.g., has seven to eight pairs of well-developed
_rugae. In Tupaia species, seven pairs of strongly
developed, regularly arranged ridges are present
(Garrod, 1879; Schultz, 1958). In Ptilocercus, Le
Gros Clark (1926a) found nine transverse pairs, each
corresponding to a tooth, from the second incisor
backwards. Prosimians remain unspecialized, having
seven to eight pairs of well-develOped rugae with
regular symmetry (Le Gros Clark, 1926a; Schultz,
1949).

Tongue
Sonntag (1921, 1925) hypothesized that the

primitive mammalian tongue had two major components;
lingua and sublingua, each with considerable inde-
pendence of movement. The sublingua may be the mor-
phological equivalent of the infra-mammalian tongue.
With increased development of muscles and papillae

in and near the dorsum, a mammalian tongue would

have emerged. The OS entoglossum (cylindrical axial
rod near lower surface) and the remainder of the lower
part of the tongue would then become the lytta (lyssa)
and the sublingua, respectively. With subsequent adap-
tive radiation of primitive mammalian orders, certain
(inferior) parts tended to disappear by absorption
into the septum linguae and under surface of the

tongue. Only Menotyphla and Primates retained both

32
lytta and sublingua, in some members. In Tupaia
and Lemuroidea there is far less reduction--perhaps
even secondary specialization--of the sublingua, as
compared with Macroscelididae and Tarsius. In its
greater length and muscular develOpment, the tupaiid
lytta remains better develOped than in prosimians
(Sonntag, 1921; Le Gros Clark, 1926a; Machida,
Perkins, and Giacometti, 1967). Since lytta and
sublingua are derived from the reptilian tongue, they
can scarcely be interpreted as new acquisitions
appearing independently in Prosimii and Menotyphla;
thus Sonntag (1921) concluded that we must assume a
common origin for both groups from Mesozoic Meno-
typhla, a group designated by Gregory (1910)
as ”insectivorous pre-Primates.”

Marsupialia retained only the sublingua;
Lipotyphla, only the lytta. Groups having lost both
portions of the tongue include Chiroptera, DermOptera,
and Monotremata (Sonntag, 1925).

Like the primitive mammalian tongue, that of
tupaiids is long and slender, tapering somewhat toward
the rounded tip. Indicative of the primate trend is
a concentration of fungiform papillae on the tip. As
in Marsupialia, most Insectivora, and smaller Pri-
mates generally, there is a triad of vallate papillae
(Sonntag, 1921).

33

Gall Bladder

Carlsson (1922) found the prominent gall
bladder of Tupaia species lying in the cystic fissure,
on the diaphragmatic surface of the liver, at the
right margin of the central lobe. This resembles
prosimians (especially Galago and Lemur) and differs
from Insectivora. In Erinaceus, for example, the
gall bladder lies in the middle of the central lobe.
However, some variation exists within Tupaiidae.
Le Gros Clark (1926a) pointed out that Ptilocercus
has the gall bladder situated in the middle of the
right central lobe, as in lipotyphlous Insectivora.
This is regarded as the more primitive relationship.

In the position of the gall bladder, Dendrogale

 

(smooth-tailed tree shrew) resembles Tupaia (Davis,
1938).

Sma11_lntestine
According to Hill (1958b), the small intestine
is primitively of the same calibre throughout its
length. This condition persists in Tupaia and Tarsius.
In primitive mammals the entire length of the
small intestine is freely suSpended by a median
dorsal mesentery from the abdominal wall. Such a
simple series of mobile coils, arranged without dis-

cernible order, is found in tree shrews (Dendrogale

 

and Ptilocercus), some lemurs, Tarsius, and an occasional

34
platyrrhine monkey (e.g., Saimiri, the squirrel mon-
key) (Le Gros Clark, 1926a, 1959; Davis, 1938).

The phylogenetic stages in develOpment of the
duodeno-jejunal flexure and consequent gut rotation
are recounted by Hill (1958b). The sharp bend at the
duodeno-jejunal junction occurs in all Primates,
as well as Tupaiidae, and is caused by fixation of
the gut to the dorsal body wall at this point. This
duodeno-jejunal flexure becomes the pivot of clock-
wise rotation of the lower gut. Ptilocercus shows
minimal rotation: the colon remains on the left
(Le Gros Clark, 1926a); migration in Dendrogale is
similar (Davis, 1938). Tupaia and Tarsius (with some
individuals showing more advance than others) show
lengthening of the large intestine, with the ileo-
colic junction swinging dextrad around the periphery
of the coiled jejuno-ileum. Within the right flank
the caecum points dextro-craniad in Tupaia; in
Tarsius, where longer, caudad. Lemurs, lorises, and
most simians show more advanced ileo-colic migration.
(Note, however, that certain lower mammals Show the
greatest ileo-colic migration. An example is VOmbatus,

a marsupial.)

Large Intestine
The primitive mammalian large intestine is

believed to be a short, straight, simple tube of

35
relatively wide diameter and not clearly differen-
tiated into colon and rectum. It proceeds directly
caudad, to the anal canal. It is suspended in the
midline by a continuous dorsal mesentery. In re-
taining this pattern, Ptilocercus was regarded by
Le Gros Clark (19263) as "almost reptilian.“ Hill

(1958b) also found a young specimen of Tupaia tana

 

tana to be almost reptilian in hind-gut features. As
in iguana, e.g., he noted the especially capacious
cranial end, with ileum entering on the right
anterior aSpect, and with the caecum being a slight
bulge in the antero-Sinistral wall. Davis (1938)
noted that in Dendrogale the colon is even less dif-
ferentiated than in Ptilocercus, for its diameter
was found to be no greater than that of the small
intestine; Saimiri and Tarsius are somewhat similar
(Starck, 1958; Le Gros Clark, 1959).

Chapman (1904) found the large intestine of

Tupaia ferruginea and Tupaia pictum to be of uniform

 

diameter throughout. In Tupaia, as in Loris, Saimiri,
and Callicebus, taeniae and haustra are lacking;
the colon cannot readily be distinguished from the
rectum. One, two, or three taeniae first appear in
certain lemur Species (Hill, 1958b).

Le Gros Clark (1926a) detected no differen-

tiation within the colon of Ptilocercus, except by

36
observing the blood supply. In Tugaia, however, the
upper end of the colon bends dextrad to give the
appearance of an incipient demarcation between a
transverse and descending colon. Tarsius shows a
further step in having a quite well-defined trans-
verse colon (Le Gros Clark, 1959).

A slightly more advanced differentiation
may be seen in Tupaia (Anathana) ellioti and TuEaia
tana. Here, Hill (1958b) found the ileocolic junc-
tion in the right mid-flank area. Thus, a short
ascending colon runs craniad, toward the liver, and
then arches over the duodenum as a complete trans-

verse colon.

Mesenteric Relations ‘

Tree shrews Show the primitive mammalian
arrangement, with even the duodenum suspended freely
by the dorsal mesentery. This condition of intestinal
motility from stomach to rectum persists in some
lemurs, in Tarsius, and in a few platyrrhine monkey
Species, such as Saimiri. All Insectivora also Show
continuity of the dorsal mesentery (Hill, 1958b).

Starck (1958) observed that all Primates and
Tupaiidae possess a mesotyphlon (a peritoneal fold
between caecum and ileum). In primitive mammals
this fold is non-vascular and devoid of fat. VSssels,

when present, enter via adjacent vascular plicae.

37
Basically, the vascular folds are symmetrical and
short, lying to right and left of the true non-vascular
mesotyphlon. Although most Callithricidae (marmosets
and tamarins) retain this symmetry, Tugaia, like most
Primates, Shows enlargement of the left (dorsal)

vessels. The ventral caecal artery remains small.

Sensory Structures

 

Cutaneous Receptors

 

Winkelmann (1963) observed that in Urogale
everetti there were no papillary nerve endings of
the Meissner or mucocutaneous type, such as are
found in all Primates, including Lemuroidea. More
difficult to explain phylogenetically is the presence
of such specialized endings in only one other mammdfian

group, namely, Didelphis virginiana (Marsupialia)

 

(Winkelmann, 1964). Until studies have been made of
other marsupial and insectivore SpeCies, no inter-
pretation seems in order. Also of interest is
Winkelmann's (1963) failure to find any typical
mammalian end organs in Urogale; although he believed
further study would reveal them.

Tupaiidae retain the primitive mammalian type
of rhinarium: the external nares are surrounded by

an area of naked, moist glandular skin (Hill, 1953b).

38

Olfactory Receptors

Being primarily ground-dwellers, most
primitive mammals (including insectivores) have a
keen olfactory sense (Polyak, 1957; Negus, 1958).
Since tree shrews are in some measure arboreal, it is
notable that vision has increased concomitant with
suppression of olfaction. Least apparent in tree
shrews, this suppression increases as the primate
scale is ascended, through lemurs and monkeys, to
higher apes and man (Negus, 1958).‘

The cartilages associated with the vomero-
nasal organ (of Jacobson) are discussed with the

skull.

Ears

 

As in most lower Primates and mammals general-
ly, the pinna is more or less freely movable.
Lasinski (1960) maintained that the primitive form
of the primate ear was of a rounded or oval shape,
as in Tupaia. This general form is also found in
Primates that are diurnal. Even the incurling
of the peripheral margin corresponds to the helix
found in monkeys, apes, and man. Ckher landmarks,
such as tragus, antitragus, antihelix, and crura, also
resemble those of higher Primates.

Bony features of the ear region are discussed

with the skull.

39
Eyes and Accessory Structures

The tree shrew has larger eyes than other
shrews. This is usually considered to be a primate-
like feature (Polyak, 1957).

The well-developed Harderian gland and the
fair (covering one-third of corneal surface) develop-
ment of the nictitating membrane, in tupaias, are
generally found among primitive mammals (Le Gros
Clark, 1926a).

Tupaia has monocular vision, primarily.

As in all infra-mammalian vertebrates and many
mammals, there are few or no uncrossed nerve fibers
in the optic chiasma (Polyak, 1957; Campbell,

Jane, and Yashon, 1967).

In mammals generally, the cornea is thicker
than the sclera. The most extreme difference known
is found in Callithrix jacchus, with a corneal diameter
of 256-360 micra, and only 56 micra for the sclera.
This is five to six times as great as the difference
previously found to be unique among mammals. Tu aia,
however, shows similar values: the cornea is 200
micra in diameter; the sclera, only 35 micra. Similar
relations, although not as extreme, are found in

Perodicticus, Galago, and Nycticebus (Rohen, 1962).

 

Most Primates lack a Bowman's membrane; Tupaia

glis and Hominoidea, however, have such a cell-free,

4O
homogeneous-appearing layer beneath the basement
membrane of the corneal epithelium (Rohen, 1962).
Samorajski, Ordy, and Keefe (1966) have made

a careful study of the retina of Tupaia glis,

 

using fundus photography, histochemistry, and elec-
tron microscopy. Polyak (1957), Le Gros Clark (1959),
and Rohen (1962) had indicated the presence of
incipient "central area" formation in Tupaia and
Urogale, based on the diminished number of finer
retinal vessels observed here, in contrast with more
peripheral retinal regions. Samorajski, Ordy, and
Keefe (1966), however, have found the branches of

the prominent central retinal artery to radiate
uniformly over the retinal surface.‘ The single

layer of short, thick cones extends uniformly across
the central and peripheral zones of the retina. Cells
of the ganglion layer extend in a double row across
the retina, with no thinning so characteristic of the
fovea of higher Primates. (No comparable studies have
yet been done on Prosimii.)

The high visual acuity of higher Primates and
>man is matched by Tupaia, as indicated by behavioral
study (Ordy and Keefe, 1965). In the absence of a
fovea centralis, it has been suggested that in this
tree shrew the entire retina may function as a crude

sort of central area (Samorajski, Ordy, and Keefe,

'41
1966). This would indicate a kind of evolutionary
parallelism at the functional level.

Certain ultrastructural features (e.g.,
the nature of saccules in the cone outer segments)
are shared by Tupaia with monkeys, apes, and man,
in contrast with any other lower animals studied.

On the other hand, a unique feature (observed in

no other animals) is the highly ordered, whorled
lamellar configuration of concentrated cristae found
in the mitochondria of the cone inner segments.
Without comparable observations on other diurnal
insectivores and prosimians, it is difficult to
evaluate these findings of Samorajski, Ordy, and
Keefe (1966).

In Urogale everetti there is a long, thin
cone-shaped formation (almost entirely of neuro-
glia) with its base on the Optic papilla and pointing
into the vitreal cavity. A similar structure is
found in lizards (Polyak, 1957).

Insectivora generally Show some degree of
microphthalmia and poor sight. Vision shows close
correlation with mode of life among mammals generally.
This poses a problem regarding vision in relation to
mammalian origins, as discussed by Polyak (1957).
Insectivora are acknowledged as the most primitive

of living Eutheria. The earliest mammalian fossils

42
are insectivorous; thus, if early mammals derived
from large-eyed, keen-sighted reptiles, Insectivora
must have undergone secondary reduction due to their
way of life. On the other hand, if the earliest mammals
or mammal-like reptiles underwent reduction of vision,
an “evolutionary bottlenéck” would have resulted, with
marked reorganization of the sensory faculties.
Whereas certain insectivores and bats showed reduction
or loss of vision, perhaps certain tropical shrews,
becoming arboreal, would have acquired a dominant
visual sense.

The modern tree-shrew eye can scarcely be
considered generalized mammalian in type, being almost
exclusively specialized for photopic vision; thus
Polyak (1957) believed that tree shrews illustrate the
stage just preceding the coming of true Primates,
only in a general way.

Parallelism or not, as in unquestioned Pri-
mates, there must have been considerable eye-brain
reciprocity in the phylogenetic development of modern
tree shrews. -- Polyak (1957) called them ”born busy-
bodies,” for they are constantly examining everything
around them, even when not hungry. -- The nervous
system shows much evidence of such concomitant

develOpment.

43

Nervous System

 

Brain in General

In brain weight relative to total body weight,
tree shrews are comparable to elephant shrews on the
one hand, and to lemurs of comparable size (e.g.,
Microcebus) on the other (Le Gros Clark, 1928, 1959;
Evans, 1942; Campbell, 1966). More data, based on
a larger variety of mammalian groups, are needed
before this characteristic can be evaluated from a

phylogenetic point of view.

Telencephalon

 

In contrast with Macroscelides and Centetes,
the cerebral hemispheres of tree shrews extend back
over the anterior cerebellum, leaving no part of the
diencephalon or midbrain visible in a dorsal view.
This is Similar to the situation in lemurs (Le Gros
Clark, 1959).

In relation to body weight, Stephan and
Andy (1964) found that the telencephalon (of the five
secondary brain vesicles) shows the greatest phylo-
genetic development. In monkeys, it is six and
one-half to seven and one-half times the size found
in basal insectivores (Tenrec, Setifer, Erinaceus,
Crocidura, Sorex Species). Following in descending
order are prosimians, Tupaia, Macroscelididae and

Galemys, trailed by other insectivores.

44

In relation to body weight the neocortex of
monkeys undergoes tremendous growth, reduced by
half in prosimians. Tugaia has a neocortical size
between the lower primates and the highest insec-
tivores (Galemys and Macroscelididae). The other
insectivores Show little neocortical development in
relation to the basal forms; smallest of the latter
are Tenrec and Setifer (Staphan and Andy, 1964).

The corpus striatum shows the next greatest
development (among telencephalic subdivisions). It
is largest in monkeys, somewhat less developed in
prosimians, approximately equal for Tupaia and
Galemys, with all other insectivores following
(Stephan and Andy, 1964).

The hippOcampus is larger in specialized
insectivores (e.g., Macroscelididae) than in monkeys
and some prosimians; yet in Tupaia and Lari; its
size is close to that of basal insectivores (Stephan
and Andy, 1964).

Andy and Stephan (1966) pointed out that the
true septum appears to decrease in size when compared
with corpus callosum or with total brain volume, as
one progresses from primitive insectivores through
tupaiids and prosimians to higher primates and man.
However, in comparison relative to body size, using a

regression line based on primitive insectivores, the

45

septum is observed to undergo a definite increase in
size, phylogenetically.

Expansion of the neopallium in tUpaiids
(beyond most insectivores) appears to be primarily
in a posterior direction, due to an enlarged visual
cortex. This has tended to displace both the rhinal
sulcus and the temporal cortex downwards. While it
is true (Le Gros Clark, 1959) that the position of
the rhinal sulcus is more comparable to Microcebus
than to Centetes or Echinosorex, there is equivalent
or almost-equivalent downward displacement in a

variety of other mammals: Trichosurus, Dasyurus,

 

Petrogale, Onychogalea, Aepyprymnus (order Marsupi-
alia); Myocastor, Qélié: Dasyprocta, Tamias, Sciurus
(order Rodentia); cmyctolagus (order Lagomorpha)
(Campbell, 1966) .

 

While neocortical expansion in Tupaia does
involve chiefly the occipital cortex, the temporal
cortex shows some expansion, too. A distinct tem-
poral pole is formed. The frontal lobe, however,
remains small. Frontal and parietal association
areas are not elaborated. Such areas first become
recognizable in lemurs, and are especially character-
'istic of Anthropoidea (Le Gros Clark, 1959).

Overall, the neocortex of tree shrews shows

increased cellularity and better differentiation of

46
laminae than in insectivores, although poorer than in
most lemurs. The various cortical areas can be
distinguished by cytoarchitectural features (Le Gros
Clark, 1959).

In Insectivora increase in size of the
brain as a whole involves enlargement of the pyriform
lobe disproportionate to that of the neocortex (e.g.,
Echinosorex). In Tupaia, as in Primates generally,
neopallial expansion accompanies relative regression
of the pyriform lobe (Le Gros Clark, 1959).

Except for the calcarine (limiting) sulcus of
the visual cortex, there is little indication of neo-
cortical convolution in tupaiids. Le Gros Clark
(1924a) noted intercalary and presylvian sulci, as
well as some other incipient sulci (subzonal foldings)
in Tupaia. A lateral (Sylvian) fissure is not observed
below lemurs; a central sulcus, although occurring
sporadically among lemurs, is not a constant feature
until one reaches anthropoid levels above marmosets.

The corpus callosum of TuQaia is thin and

nearly straight. Although Microcebus adds anterior

 

(genu) and posterior (splenium) thickenings, the
insectivores and tarsioids show poorer development
than Tupaia. A notable exception is found in flagge-
scelides (Elliot Smith, 1902). Here the corpus

callosum shows quite extraordinary develOpment for

47
an insectivore. Although it is large, with definite
Splenium, it is peculiar in being high Up near the
dorsal margin of the hemiSphere, as well as in having
a long, hook-like genu ending in a sharp rostrum.
This would appear to be a case of rather specialized
parallélism with lemurs and other primates.

.Von Bonin and Bailey (1961), in discussing
patterns of the cerebral cortex, noted that although
the neocortex of Insectivora is much clearer than in
reptiles (where it first appeared), the cellular
layers and subdivisions into fields are difficult to
relate to those of higher mammals. Like Brodman
(whose work of 1909 they cite), they found the sub-
divisions to be vague. Only the insular cortex is
definitely distinguishable. (Archipallium still
constitutes over three-fourths of the cortex in

insectivores.) Elephantulus was found to resemble

 

Erinaceus in these reSpects. Von Bonin and Bailey

(1961) further stated: "The brain of Elephantulus is

 

clearly that of an Insectivore while that of Tupaia,
in marked contrast, has the unmistakable primate
pattern." I

Like Brodman, von Bonin and Bailey (1961)
found the motor area of Tupaia clearly recognizable;
giant pyramidal cells in area four are very clear.

There is, in Tupaia, a temporal area which

48

may have to do with hearing. There are other
recognizable areas: parietal, insular, and frontal
on the lateral aspect; retrosplenial (small) and
limbic areas (over the anterior two-thirds of corpus
callosum) on the medial aspect (von Bonin and Bailey,
1961).

Von Bonin and Bailey (1961) believed that

the cortex of Elephantulus shows aberrant features in

 

areas and their histology. As is generally true of
insectivores and rodents (but not of higher forms),
the retrOSplenial areas are large in Elephantulus.
The thickness of cortical layers I and IV are not
at all in the same range as for Primates, including
tree shrews. I

The prominence of the occipital pole in
Tupaia contrasts with Macroscelides and Lipotyphla.
In Tupaia the occipital pole overlaps the corpora
quadrigemina; however, this also occurs in numerous
other mammals (Campbell, 1966). Histologically, the
striate area is quite large and covers the dorso-oc-
cipital pole of the cerebral hemisphere, extending for-
ward to the calcarine sulcus on the medial aspect.
In Tupaia the calcarine sulcus is a limiting sulcus,
not an axial sulcus of the striate area; thus the
presence of a calcarine sulcus may be questioned as

indicative of affinity with higher Primates. Further,

49
a calcarine sulcus is absent in the presumably more
primitive Ptilocercus. The phalanger, Trichosurus,
has a much more prominent and extensive calcarine
sulcus; in a marsupial, this trait would appear to
be clearly due to convergence.

Tupaiidae show no retrocalcarine sulcus, found
in primitive lemurs (e.g., Microcebus) and all higher
Primates (Le Gros Clark, 1959).

The striate area of tupaiids shows greater
cellularity and better differentiation of laminae
than in any insectivores. Although interpretation
of the cell layers is controversial, there is some
evidence of an incipient ”white line“ of Gennari,
as found in Primates generally. This is evidently
effected by duplication of the inner granular layer,
which is very conspicuous and sharply defined.

Von Bonin and Bailey (1961) confirmed Le Gros Clark's
view (19243) that the striate area does begin to show
the primate pattern in which layer IV is duplicated.
They found the striate area of Insectivora to have
none of the features observed in Primates. On the
other hand, Campbell (1966) did not believe Le Gros
Clark (1924a, 1925) demonstrated any detailed simi-
larity between lamina IV of Tupaia and that of any
primate. Although the presence of well-defined

laminae in the tupaiid striate cortex is unquestioned,

50
it is not yet possible to say whether this similarity
to Primates is due to lineal ancestry or to parallel-
ism.

The olfactory bulb of Tugaia remains rather
large, compared with primates generally; compared
with most terrestrial insectivores, it is somewhat
reduced. Elephant shrews (with jumping habits),
show a comparable reduction in the olfactory areas
of the brain. Arboreal squirrels Show even greater
reduction than tree shrews and elephant shrews
(Le Gros Clark, 1959).

In tupaiids the temporal pole is still formed
by the pyriform lobe, as in some lower primates
(lemuroids and tarsioids) (Le Gros Clark, 1959).

The tuberculum olfactorium of Tugaia is only
slightly raised from the forebrain surface, in con-
trast with its prominent rounded shape in most small
macrosmatic mammals of similar size (Le Gros Clark,
1959).

In Macroscelides there is a psalterium of
crescentic shape, found elsewhere only in Metatheria.
The hippocampal formation extends forward onto the up-
per surface of this (anterior) commissure, as in the
Metatheria. In TuEaia the psalterium ends anteriorly
in a slight bulbous swelling. A similar configura-

tion is found in Microcebus. The hippocampal complex

51
is reduced in Tupaia, relative to Insectivora (Le

Gros Clark, 1959).

Diencephalon

In contrast with insectivores, there is an
increased degree of differentiation of the thalamic
nuclei in Tupaia (Le Gros Clark, 1929). Lamination
of the lateral geniculate nucleus, although present,
is not sharply defined (Woollard, 1926). Seemingly
progressive features shared-by tupaiids with primates
include: (1) well-developed dorsal and ventral
portions of the lateral geniculate nuclei, with rudi-
mentary lamination in the dorsal portion; (2) enlarge-
ment of the pars posterior Of lateral nucleus; (3)
prominence of the pretectal nucleus (Le Gros Clark,
1932). The first two features are also found, however,
in Trichosurus vulpecula, with dorsolateral geniculate
nuclei even better differentiated than in Tupaia. The
shape Of the laminae in the phalanger approaches that
of lemurs, in which optic tract fibers enter along
the lateral convexity and Optic radiation fibers
leave the medial concave aspect (inverted type of
lateral geniculate body--as Opposed to everted type
in Anthropoidea). In the phalanger, crossed (75 per
cent) retinal fibers and chiefly in laminae one and
three; uncrossed (25 per cent), chiefly in laminae two

and four. Laminae one and two do show some Overlap.

52
Higher Primates show a much more complete segrega-
tion of crossed from uncrossed fibers (Goldby,
1941).

Packer (1941) suggested that the degree of
lamination is probably related to the degree of
development of binocular vision in an animal.

Having studied sections Of the visual cortex,
Packer (1941) doubted if this area in the phalanger
differs significantly from other primitive mammals.
While lamina IV has fewer cells along its outer
border, there is no suggestion of a subdivision
within this layer (well marked in Tarsioidea and
Anthropoidea, less clearly defined in Tupaia and
Lemuroidea).

Bauchot (1959) made a cytoarchitectural study
of the diencephalon of Talpa europaea (mole),
generally quite similar to that of Marsupialia,
Rodentia, and ChirOptera. The lateral geniculate
body is rather reduced, as in most Insectivora,
except Macroscelides and Tupaia. It lacks the
lamination found in Tupaia.

Campbell, Jane, and Yashon (1967) made a com-
parative study of retinal projections in tree shrew

and hedgehog. The visually advanced Tupaia glis

 

showed a tendency toward restriction of mesencephalic

and diencephalic terminations.

53

Glickstein, Calvin, and Doty (1966) compared
the laminar structure of the dorsal geniculate body
of Saimiri and Tugaia by degeneration experiments.
Saimiri (squirrel monkey) shows a six-layered
structure similar to that of Old World Primates.
Layers one, four, and six Show transneuronal degenera-
tion contralateral to the enucleation; layers two,
three, and five show degeneration ipsilaterally.
Tupaia, with five layers in the nucleus, shows trans—
neuronal degeneration Of the (few) uncrossed fibers
in layers one and five. The decussating fibers
(Overwhelming majority) connect with internal layers
two and four Of the nucleus. Layer three did not
seem to be differentially degenerated on the crossed
and uncrossed sides.

According to Campbell (1966), every primate
examined shows a pattern resembling Saimiri (reverse
of Tupaia), with crossed fibers terminating in the
innermost and outermost lamellae. This suggests that
the lamination observed in the dorsal lateral genicu-
late nucleus of tree shrews is the result of conver-
gent Or parallel evolution.

Schroeder, Yashon, Becker, and Jane (1968) have
compared the medial lemniscus of Erinaceus europaeus,

Tupaia glis, Nycticebus coucang, and Saguinus oedipus

 

(marmoset). The major difference between Erinaceus

54
(hedgehog) and the other species is in the progressive
restriction of diencephalic terminations to the ventral
posterior nucleus in the tree shrew, loris, and
marmoset, as opposed to a wider distribution to other

nuclei in the hedgehog.

Mesencephalon

 

The anterior colliculi.of Tupaia remain
relatively large, deSpite advances in the striatal
cortex. These structures are relatively larger in
tupaiids than in any other mammal, with the possible

exception of the "flying lemur" (Cynocephalus, order

 

Dermoptera). This is a primitive feature, in view
of the great size of these structures (as Optic
lobes) in reptiles. As in many non-primate mammals,
visual function at this level remains important

(Le Gros Clark, 1959).

Von Tigges (1966) found that 80 per cent of
retinal fibers reach the anterior colliculus; the
other 20 per cent terminate in the dorsal nucleus
of the lateral geniculate body. This demonstrates
clearly the persisting dominance Of visual function
by the midbrain.

Feremutsch (1965) had made a preliminary study
of the anterior colliculus in Tenrec ecaudatus,

Tupaia glis, Blarina brevicauda, and certain primate

 

 

Species.

55

In Tenrec ecaudatus, the anterior colliculus
extends well into the thalamic region, as in Tupaia,
but to a lesser degree. Cells of the anterior
colliculus Show ill-defined stratification in Tenrec.
In Blarina brevicauda arrangement into strata externum,
intermedium, and internum is better defined. The
overall structure is more like that of Tupaia or

Microcebus than like that Of Tenrec. Tupaia glis

 

displays the laminated structure very nicely. As in
Cercopithecidae, Pongidae, and Hemp, the three strata
can be clearly distinguished (Feremutsch, 1965).

Among Prosimii, only Microcebus murinus re-
sembles Tupaia. Strata intermedium and internum are
not sharply divided, however. Lemur macaco and Larig
tardigradus show the three strata, if one looks care-
fully (Feremutsch, 1965).

Le Gros Clark's (1926b) comparative study Of
the mammalian oculomotor nucleus showed a very simple
pair of lateral nuclei in Erinaceus. There is no dif-
ferentiation of a recognizable Edinger-Westphal nucle-
us. Macroscelides is similar, except for the addition
of a distinct Edinger-Westphal nucleus like that of
Tupaia. In Tugaia the whole oculomotor nucleus is
more elaborate. While the small cells of the Edinger-
Westphal nucleus are not sharply circumscribed, the

cluster is quite recognizable. The large-celled

56
lateral nuclei are distinctly subdivided into dorsal,
ventral, and paramedian grOUpS. Although this

pattern represents a definite advance over Erinaceus

 

and Macroscelides, it is still a simple plan, detect-
ible in the oculomotor nuclei of other mammals. With
increasing elaboration Of the nucleus in higher
Primates and some other mammals, this arrangement of
three longitudinal cell grOUps is progressively ob-
scured.

The pattern in Cynocephalus is indicative
of a trend in the direction observed in Tupaia.
Sciurus is similar to Tupaia, except for certain
differences in disposition of the parts of the
oculomotor nucleus (Le Gros Clark, 1926b).

_The Edinger-Westphal nucleus of Nycticebus

 

and Galago remains relatively.undifferentiated

and poorly circumscribed. It is more conspicuous in
Tarsius, becoming very distinct and separable into two
cell grOUps in Hapale and Macacus. It reaches its
peak in anthropoids, especially in man. Such a small-
celled element, rather diffuse in outline, is found in
a number of other mammals studied, including Pinni-

pedia and Cetacea (Le Gros Clark, 1926b).

Rhombencephalon
The cerebellum is rather poorly develOped in

tupaiids, being of a simple, generalized mammalian

57
type. Its convolutional pattern, however, is more com-
plex than that found in Microcebus (a primitive lemur).
Tarsius, and insectivores (Macroscelides) of compara-
ble size (Le Gros Clark, 1959).

Cell counts of the superior Olivary complex in
tree shrew (Tupaia minor), marmoset, squirrel monkey,
and gibbon disclose regression Of the lateral nucleus
and nucleus of the trapezoid body, as compared with
the medial nucleus, among these primates, in the order

named (Sehmsdorff, 1966).

Corticospinal Tracts

Corticospinal tracts.are found mainly in
the dorsal funiculi Of monotremes, marSUpials,
edentates, and rodents (Jane, Campbell, and Yashon,
1965).

Bautista and Matzke's (1965) study of the
corticospinal tract Of opossum shows the upper regions
to be similar to other mammals down through the level
of the pyramidal decussation. It then descends, how-
ever, in the dorsal funiculus of the entire cervical
Spinal cord. Fibers terminate at the base of the
dorsal gray column. The cortiCOSpinal fibers show a
rather broad area of origin. In marsupials the corti-
cospinal tract has barely made its phylogenetic
appearance. The degree Of cortical influence Over

motor control appears to be slight.

58

Three insectivores (Erinaceus, Talpa, and
Scalopus) Show uncrossed ventral tracts. (There is
no information on elephant shrEws.) Crossed dorsal
pyramidal tracts are not characteristic of insectivores
(Campbell, 1966).

Jane, Campbell, and Yashon (1965) made a study
of corticospinal tracts in Tupaia glis. The decussae
tion in the medulla is apparently incomplete. Some
individuals show a small bundle passing into the
ipsilateral tract. Corticospinal fiber bundles
are found in the dorsal funiculi, extending downward
only to the mid-thoracic level of the spinal cord.

The distribution of the terminations of the cortico-
spinal fibers in the gray matter is Only to the dorsal
horn and intermediate zone, with_none to the ventral
horn.

Shriver and Noback (1966) studied the cortical
projections to the brain stem and spinal cord in
Tupaia glis by producing lesions in frontal and
parietal cortex. Uncrossed terminations are found in
the reticular formation Of the entire brain stem.

At spinal cord level, two pathways of degeneration
occur: (1) large crossed dorsal tracts (throughout
the cord), (2) small uncrossed dorsal tracts (in
cervical segments only). The majority of terminations

of these corticospinal fibers are in the dorsal horn

59
of the gray matter, with a lesser number in the zona
intermedia.
I Cortical influence on cranial and spinal
motor nuclei in tree shrews is therefore limited
to the phylogenetically Older "indirect” projections
to internuncial neurons located in the reticular
formation of the brain stem and in the dorsal horn of
the spinal cord. The more advanced ”direct” projec-
tions have not been added (Shriver and Noback,
1966).

In all Primates observed, most corticospinal
fibers decussate to form the lateral cortiCOSpinal
tracts, located in the lateral funiculi of the spinal
cord, and extending the length of the cord (Shriver
and Noback, 1966). -

Nycticebus, the only prosimian examined, shows

 

complete decussation. Fiber terminations are most
numerous at the base of the dorsal horn and the inter-
mediate gray column, with some fibers extending into
the ventral horn (Shriver and Noback, 1966).

The terminal distribution of corticospinal
fibers within the gray matter of the spinal cord of
higher Primates is to the dorsal horn and intermediate
zone, as well as to the ventral horn (Shriver and
Noback, 1966). '

Campbell (1966) stated: "As far as is now

60
known, the localization of the major crossed cortico-
Spinal tract does not appear to vary within the same
order, although it does vary from order to order.“
The corticospinal system of Tu2aia differs from all
known primates and insectivores. It cannot simply be

regarded as primitive mammalian in type.

Peripheral Nervous System

There is one recorded feature in which Tupaia
is transitional from generalized mammals to higher
Primates: this is the innervation of the infrahyoid
musculature. In general, among mammals, these
muscles are supplied by the descending ramus Of nervus
hypoglossus and the descending cervical nerve (from
upper cervical spinal nerves). In Tupaia, as in
monotremes, marsupials, and most insectivores, this
innervation includes cervical segments I and II (only
the first cervical segment in Sorex araneus and
Erinaceus albiventris). In Lemur macaco and all higher
Primates observed, cervical segments I, II, and III are
involved. In Prosimii and Tupaia javanica hypo-
glossal fibers supply only musculus thyrohyoideus,
all other infrahyoid muscles being supplied by cer-
vical nerve fibers. In Simiae all infrahyoid muscles
are SUpplied by the first three cervical nerves

(Fieandt, 1914).

61
Skeletal System—~Axial Division

 

The similarity between Tupaiidae and Macro-
scelididae in develOpment Of the chondrocranium is
striking (Henckel, 1928; Roux, 1947). From his in-
vestigation of the development of the macroscelidid
skull, which resembles that of Tupaiidae very closely,
Roux (1947) was convinCed that it would be impossible
for a primate skull to develop from this type, already
on an ”Insectivore line of Specialization." Seem-
ingly close affinities displayed by adult Skulls of
tupaiids and lemuroids are due, for the greater part,
to parallel modifications resulting from their
arboreal habits. Roux (1947) believed that the
Menotyphla became separated from the primitive in-
sectivore stem very early, to develOp along a diver-
gent line.

Broom's (1915) study of the vomeronasal organ
(of Jacobson) indicated that the morphology of the
associated cartilages are of value in determining
affinities among mammals. The structure is almost
the same in members of diverse orders. Since the
arrangement of these cartilages is complex, Broom
(1915) concluded that similarity indicates common
origin rather than independent convergent variations.

Among Mammalia, Broom (1915) found two

principal types: (1) the primitive (marsupial) type,

62
representing a simplification of the monotreme organ,
and retained in rodents with slight modification;
(2) the higher (eutherian) type, found in Lemur,
ungulates, carnivores, Procavia, some bats, and some
insectivores.

Broom (1915) found that the vomeronasal
(supporting) cartilages of Tupaia are very similar
to those of Macroscelides, with différences being
no greater than are found in different families of
diprotodont marSUpials or of artiodactyls. Both
closely resemble polyprotodont marsupials and Be;-
ameles. They differ in almost every respect from
Gymnura, an erinaceid insectivore. Gymnura, almost
identical with Erinaceus, also agreEs well with most
Eutheria. Thus, Broom (1915) proposed that Meno-
typhla constitute a distinct order, closely related
to marSUpials, and showing little affinity with
the ”more typical Insectivores.“

In general form, the tupaiid adult cranium
has rounded contours, although a sagittal crest is
present in Urogale (Lyon, 1913; Le Gros Clark,

1925). In showing broad convexity of the cranial

roof between the orbits, Gregory (1910) found Inpaiidae
differ widely from any Lipotyphla and approach Lemur-
oidea. Saban (1956-1957) regarded the tendency to in-

crease the size of the cranium relative to the face in

63

the same way.

Based on his study of development Of the nasal
skeleton, Spatz (1963) found the seeming reduction
of the incisura pretrabecularis, in Tupaia, indicative
of greater resemblance to Prosimii than to Insectivora.
This semicircular notch of the lower border of the
nasal septum marks the rostral end Of the trabecular
portion of this Septum. Well developed in Insectivora,

this incisure is scarcely perceptible in Tupaia glis,

 

and completely lacking in Prosimii. The loss may be
correlated with reduction in the Olfactory apparatus.
Tupaia and Lemur, as Opposed to Lipotyphla,
have only two ectoturbinal bones (Le Gros Clark,
1925; Gaughran, 1954;’Saban, 1956-1957; Sharma, 1958).
Reduction of the nasal fossae, anterior nasal orifices,
and choanae in Tugaia and Lemur are likewise correlated
with diminution Of olfaction (Saban, 1956-1957).
The presence Of palatal vacuities is regarded
as a primitive,metatherian feature of tupaiids

(Le Gros Clark, 1925); Dendrogale, however, lacks

 

such palatal fenestration (Davis, 1938).

The pattern of differentiation of foramina
(for nerves and blood vessels) on the skull base re-
sembles that of prosimians (Le Gros Clark, 1925). The
presence of malar foramina was described by Le Gros

Clark (1925) as a primate feature.

64

Observation of the basicranial axis shows
some similarities between tupaiids and lemuroids.
Le Gros Clark (1925) correlated a degree of flexion
of the face on the basicranial axis in the ethmoid
region with reduction in size of the olfactory bulbs.
A study by Hofer and Spatz (1963) on embryonic and neo-
natal heads Of representative primates, including
Tupaiagglis and Lemur, showed that the kyphosis at
the base always lies rostral to the sella turcica.

The plane of the foramen magnum in tupaiids
is somewhat depressed from the more primitive vertical
position. This has been correlated by Le Gros Clark
(1925) with the develOpment of occipital poles on
the cerebral hemiSpheres in tUpaiids and lemuroids.
It may also be functionally related to the common
assumption of the sitting posture in these groups.

The presence of a tympanic wing or process of
the aliSphenoid bone is regarded as a metatheroid
feature (Gregory, 1910; Le Gros Clark, 1925). Al-
though this process is much reduced in Tupaia, it re-
tains more of the posterior concavity, characteris-
tic Of primitive marSUpials, in Ptilocercus (Gregory,
1910).

The form of the ectOpterygoid fossae, on the
other hand, was found by Gregory (1910) to resemble

that of Lemuroidea.

65

The presence of an aliSphenoid canal is
another primitive metatherian trait (Le Gros Clark,
1925). This canal persists in scattered, primitive
placental grOUps, including most Insectivora and many
Carnivora (van der Klaauw, 1931; van Valen, 1965).
This foramen pierces the base of the pterygoid flange
(external pterygoid process) of the aliSphenoid bone.
In Tupaia glis the internal maxillary artery passes
through it along a short portion of its course.

Ancient mammal-like reptiles had an enclosed
orbit and a wide-angle Optic axis. This was evident-
ly retained in early marSUpials. (Later mammal-like
reptiles and all modern marsupials have had an open
orbit.) Early placental mammals also had an open
orbit, as did some early lemuroids; yet tree shrews
have an enclosed orbit, as in modern primates gen-
erally. Primates, with the exception Of tree shrews,
possess narrow-angle Optic axes. With a more complete
fossil record, this sequence will be Of greater phylo-
genetic interest (Prince, 1956).

Le Gros Clark (1925) and Saban (1956-1957)
pointed to the considerable forward rotation of
the orbits, eSpecially in Ptilocercus, as indicative
of a tUpaiid link with Lemur. An index, based on the
angle between optic axis and the sagittal plane, is

intermediate, for Tupaia, between insectivores and

66
primates. Saban found the same to be true of the
interorbital distance in relation to the breadth of
the face. A broad interorbital region is a primi-
tive characteristic for Eutheria. Van Valen (1965),
however, did not find the orbits of Tupaia to be
relatively more anterior and close to each other
than in other primitive mammals. He did concede
that the snout is somewhat shorter than usual.

Saban (1956-1957) believed that there are
a number of secondary effects to be correlated with
forward migration of the orbits. These, he said,
result in greater similarity between TUpaiidae and
Lemuridae than between TUpaiidae and Insectivora
generally. These secondary skull modifications are
as follows:

1. The malar bone is extensively involved in forming
the orbital margin.

Van Valen (1965) regarded the posterior ex-
tension of the malar bone as a primitive feature,
however.

2. The malar bone possesses a frontal process that,
together with the frontal bone, completes the posterior
bony wall of the orbit.

Although it is agreed (Gregory, 1910; Le Gros
Clark, 1925) that a well-developed postorbital bar

involving frontal and malar bones is typical of

67
Primates, van Valen (1965) points out that this bar
is variable in Ptilocercus, and is apparently not

primitive in Primates. It is absent in Plesiadapis.

 

Postorbital bars, although differing in constitu-
tion, haVe developed in some other placental orders.

3. A malar foramen is present.

This statement is inaccurate. The foramen is
minute in most tupaiids (Le Gros Clark, 1925).
Further, as van Valen (1965) noted, this foramen is
absent in Plesiadapis and in a number of Recent
prosimians.

4. The lacrimal sulcus and infraorbital canal (the
anterior Openings of the orbit) have been shifted
toward the external orbital margin.‘ The optic foramen
has been elevated to a position above the sphenoidal
fissure.

Van Valen (1965), on the contrary, maintained
that the lateral positions of the lacrimal foramen
and infraorbital canal are primitive features.

In Ptilocercus and many prosimians, the Optic
foramen does not lie above the sphenorbital fissure.
Erinaceids, on the other hand, do show the superior
position of the optic foramen (van Valen, 1965).

5. The medial orbital wall, in contrast to Insec-
tivora, displays broad palato-frontal contact, but

no sutural contact between the temporal and parietal

68
bones.

Van Valen (1965) argued that the extension of
the palatine to the lacrimal bone and the extensive
frontOpalatine suture are primitive features.

Groups in which they are lacking, however, include

Plesiadapis, Adapis, and Pronycticebus among fossil

 

forms, as well as Daubentonia of Recent times.

In commenting further on the nature of the
inner orbital wall, van Valen (1965) emphasized the
following observations:

1. Exclusion of the squamosal and parietal bones is
evidently primitive.

2. Although Evans (1942) stated that the malar-
1acrimal contact is shared by tupaiids and prosimians,
this feature is primitive and shows variation within
the two groups.

3. Subdivision of the lacrimal bone into facial and
orbital parts (Gregory, 1910) is not clearly defined
in Ptilocercus and is apparently primitive.

4. An 05 planum, although present in a Tupaia embryo
described by Carlsson (1922), is not found in the adult.
It is found in a number of mammals. It is limited,
among prosimians, to certain species only.

5. Although the lacrimal foramen pierces the lacrimal
bone immediately anterior to the orbital margin in

tupaiids and lemurs (Gregory, 1910), this feature shows

69
much variation among prosimians, as among Eutheria
generally.

6. While a supraorbital foramen is present in
Tupaia, it does not appear in Ptilocercus, Plesiadapis,
or in many Recent Prosimii.

7. It is true that tupaiids and prosimians have
larger orbits than most primitive mammals; yet such
enlargement has also occurred in a number of other
eutherian groups.

Another area of current dispute between van
Valen (1965) and long-accepted authorities on tree
shrews involves the auditory ossicles. Doran (1879)
wrote a treatise on the variation of these bones
among mammals. He stated that the malleus of Tupaia
“differs from that Of any other insectivorous
mammal" and resembles that of Primates; neck, anterior
process, and pars brevis are absent. In Madagascan
lemuroids these parts are slightly developed. The
malleus of Tarsius also lacks a neck. He concluded
that the tupaiid malleus most closely resembles that
of such lower primates as some lemurs and marmosets,
especially in the shortness of processus gracilis.

As in many monkeys and lemurs, the tupaiid
incus was described by Doran (1879) as having a high,
narrow body.

Although the presence of a bony canal in the

7O
stapes is shared with Lipotyphla and prosimians,
the stapedial crura are straight in tupaiids but
curved in Lipotyphla (Doran, 1879).

In general, Doran (1879) found the ossicles
to differ from any of the varied forms observed
among Insectivora.

After examination of the published figures
and descriptions (Doran, 1879; Le Gros Clark, 1926a;
Werner, 1960), van Valen (1965) failed to see any
special resemblance to Primates. In addition,
van Valen (1965) noted differences between the

ossicles of Tupaia and Ptilocercus. The processus

 

gracilis, which is well develOped in Tupaia, is
lacking in Ptilocercus. Van Valen (1965) concluded
that the similarities of tupaiids to Primates is no
greater than to other Eutheria.

The Leptictidae had an ear region closely
resembling the basic type from which tupaiid and
primate types must have develOped in Cretaceous to
Tertiary times (McKenna, 1966).

Werner (1960) emphasized the value of the
middle and inner ear features in determining phylo-
genetic relationships. Their recessed and protected
position removes them somewhat from external environ-
mental influences. Of course, neighboring structures

of more direct adaptive significance, such as jaws

71
and brain, must have modified their development.

Although the middle ear enclosure arises
separately from the cranium, its bony elements
either are part of the cranial wall or merge with
it secondarily.

Anterior to the tegmen tympani, the alis-
phenoid bone contributes to the wall Of the middle
ear cavity in TuEaia and in young Perodicticus
individuals. After the later fusion with the
petrosal bone, evidence of aliSphenoid participation
is no longer demonstrable (Werner, 1960).

The ventral (jugular) wall of the middle
ear cavity varies among primates. In Lemuroidea,
it is formed by a thin and arched bony plate, the
bulla ossea.

In Tupaiidae, the greater part of the bulla
wall develops from a separate endochondral bone,
the os bulla or entotympanicum, apposed to the
petrosal bone below, and to the cochlear region
medially. The tree-shrew bulla evidently is a
compound structure containing some additional minor
contributions. There is a small contribution from
the aliSphenoid, and perhaps from the basisphenoid
bone as well (van der Klaauw, 1929). Saban (1956-
1957) denies this, finding that the presence of

the latter is typical of insectivores generally, but

72
not of Tupaia and Lemur. Spatz (1966) found slight
supplementation of the bulla wall by a small proces-
sus tympanicus petrosi caudalis in Tupaia glis.

All fusions of the greatly eXpanded ento-
tympanicum with other skull elements are secondary.
Saban (1956-1957), however, argued for absence of
an entotympanicum in Tupaia and Lemur, believing
that the bulla floor is formed solely by an ex-
pansion of the petromastoid region of the temporal
bone.

The os entotympanicum was regarded by Werner
(1960) as a new acquisition among mammals. Spatz
(1966) agreed with this, and argued for affinity
between tupaiids and Malagasey lemurOids, based
on belief that the prosimian bulla was derived from
a tupaiid bulla by a primordial fusion of ento-
tympanicum with petrosum. He also noted similarity
in the overall form of the bulla; van Valen (1965),
however, did not find greater similarity between
these bullae than is commonly found between un-
related groups of placental mammals.

Van der Klaauw (1929) noted that although a
free entotympanic bone is present in Menotyphla,
it is not found in any modern Lipotyphla or in the
ontogeny of the Primates.

The early fusion of entotympanicum to

73
petrosum has been the basis of continuing controversy
through the present century. If, as van Kampen (1905)
suggested, the entotympanic bone represents merely
the expanded and secondarily independent tympanic
wing of the petrosal bone observable in many
maISUpials and insectivores, then the difference
in bulla formation Of Menotyphla and Lipotyphla
is a difference of degree, not of kind. There is
no doubt that the great difference now existing
implies a long period of separation between Meno-
typhla and Lipotyphla.

Van der Klaauw (1929) demonstrated that in
Macroscelididae the bulla is formed by the two
entotympanic bones (caudal and rostral), supple-
mented by aliSphenoid, basisphenoid, petrosal, and
ectotympanic bones.

McKenna (1966) pointed out that in the course
of tupaiid evolution, the medial border of os
entotympanicum apparently has fused (in adult) to
the os petrosum. Some traces of such fusion do
exist in adult tupaiids. In juvenile tupaiids (e.g.,
Tupaia glis modesta), the entotympanic portion Of
the bulla remains long separated from the petrosal
bone.’ The adult fusion of entotympanicum to
petrosum is thus believed to have been a relatively

recent phylogenetic event among tupaiids. It is on

74
the basis of the adult structure that the tupaiid
bulla is often erroneously considered to be an
outgrowth of the petrosal bone, even as in Malagasey
lemurs and other primates.

Van Valen (1965) has assembled much evidence
against tupaiid-primate affinity based on bulla
composition. He confirmed the virtually exclusive
contribution Of an OS entotympanicum. He contradicted
Saban's (1956-1957) statement that no suture is found
lateral to the carotid foramen (in os petrosum).

In almost all young skulls (and even a few presum-
ably mature ones) Observed by van Valen (1965),

such a suture is present. Since ossification of os
entotympanicum evidently begins anteromedially (near
the pharyngotympanic tube), van Valen (1965)

believed this bone is homologous to the anterior
entotympanic bone described by van der Klaauw (1929).

In lemurids and lorisids, on the contrary,
the major portion of the bulla grows out (ontogeneti-
cally) from the medial aspect of the OS petrosum
(van Kampen, 1905; van Valen, 1965). Ossification
here is not initially limited to the anteromedial
region, but occurs medially throughout the length
of the bulla. The medial portion of the bulla is
evidently derived from both mastoid and petrosal re-
gions, at least in lorisids. Van Valen's (1965)

75
evidence indicated this probably is also true for
Isms. _

Van der Klaauw (1931) pointed out that
bullae may develop from any bone or cartilage
adjacent to the middle ear region, in one or
another mammalian group; bullae derived from a given
element have developed independently, by convergent
evolution, a number Of times. Thus, even if the
bullae Of tupaias and lemurs are homologous, this
alone would not be conclusive evidence of proxi-
mate common ancestry.

According to Gregory (1910), the true
tympanic bone remains as a delicate ring inside the
entrance of the bulla in both Tupaiidae and Lemuri-
formes. Hill (1953b) noted that in all Malagasey
lemurs, including Daubentonia, the annulus tympanicus,
although not involved in bulIa formation, becomes
enclosed to lie free within the bulla during develOp-
ment. This condition is described as unique among
living mammals, except for Tupaiidae. He noted that
it is also found in the fossil Adapidae and Plesiadap-
idae. McKenna (1966) added fossil Leptictidae to
the list.

A free tympanic ring is certainly primitive;
yet, as van Valen (1965) emphasized, no direct

process can be envisaged by which a Malagasey

76
lemuroid bulla could develop into one similar to
that of Perodicticus, the bulla of the latter being

comparable to the bullae of Plesiadapis, Tarsius, and

 

anthropoids. The problem is to get the tympanic
element incorporated into the already completed
bulla surrounding it. One ontogenetic possibility,
suggested by van Valen (1965), involves failure of
the petromastoid bulla to grow out over the tympanic
ring. This, however, would require an abrupt transi-
tion between the two adult conditions. Some function-
al disturbance would be likely. In searching for
possible intermediate stages, van valen (1965)
observed that, at times, old LEEEE individuals dis-
play calcification, and possibly ossification, of
the annulus membrane. Such occurrence, ontogeneti-
cally, could lead to lateral expansion of the
tympanic bone, following incorporation into the bulla.
Tupaia has been said to differ widely from
Lipotyphla and to resemble Malagasey lemurs in the
shape and position of the ectotympanic bone (van
Kampen, 1905). Gregory (1910) noted that the
ectotympanic bone extends laterally to form a
very short external auditory meatus. Although most
conspicuous in Catarrhinae and Hominoidea, this

development also occurs in Loris and Nycticebus.

 

Some degree Of such develOpment variably occurs

77
among Insectivora (Saban, 1956-1957).

Among the features of the bulla listed as
similarities between Tupaia and LEEEE are the
following (Saban, 1956-1957):

1. The orifices (carotid foramen, stylomastoid
foramen, orifice of pharyngotympanic tube) are
peripherally located.

2. The septa delimiting the diverticula of the
bulla are arranged in the same manner.

Van Valen (1965) stated that similar septa
commonly develop even in non-homologous bullae.
Lyon (1913) had indicated that these septa are
lacking in Ptilocercus.

3. The counterpart Of the basisphenoidal divertic-
ulum is absent;-yet it is always present in Insec-
tivora.

4. The hypotympanic sinus is large.

Van valen (1965) maintained (citing van der
Klaauw, 1931) that such an inflated bulla is a
common evolutionary develOpment among mammals.

5. .A bony canal for the pharyngotympanic tube is
present here, but absent in Insectivora.

Yet, such a bony covering is a common evo-
lutionary development (van der Klaauw, 1931).

6. There is early synostosis of the fissure of

Glaser. In Insectivora, the fissure remains widely

78
Open.

7. The epitympanic recess has straightened to be-
come integrated with the interior of the bulla. This
is correlated with a realignment of the tympanic
framework.

Van Valen (1965) considered this relatively
internal position of the epitympanic recess as a
primitive feature. It also appears in some other
mammalian groups.

8. The recessus meatus is enlarged.

It is likewise enlarged along other mammalian
lines, according to van Valen (1965).

9. Portions of the vestibular complex diSplay
similar interrelations.

10. The inferior petrosal (venous) sinus Opens into
the internal jugular vein. Among Insectivora, this
Opening always leads into the lateral sinus.

Extracranial passage of the inferior
petrosal sinus is also to be found in marsupials,
according to van Valen (1965).

11. The petro-squamous sinus is of very large diame-
ter.

Van Valen (1965) regarded this feature as
being probably primitive, although rather variable.

Saban (1956-1957) measured angles of inclina-

tion of the petrous temporal bone. These indicated

79
that tupaiids are about equidistant from most
Insectivora on the one hand, and from Primates on
the other. He related these angles to the tilting
of the occiput, presumably the result of increased
brain size. Van Valen (1965) pointed out, however,
that since the brain is relatively small in
Plesiadapis, the greater size found in Recent
tupaiids and Recent lemuroids has probably arisen
independently.

A mastoid region of the petrous temporal
bOne is not yet distinctly set apart in tree
shrews. There are no air Spaces here. Lemuriformes
show less pneumatization of the expanded petrous
and squamous temporal regions than Lorisiformes

(Gregory, 1910; Werner, 1960).

Dentition

McKenna (1966) noted the scarcity of early
Paleocene insectivore remains, with which early
primates must be compared. He also pointed out
that Eocene deposits of southern Asia and Borneo,
which may be expected to contain fossil tupaiids,
have not been searched carefully, using appropriate
collecting techniques. Even so, for thin-skulled
inhabitants Of tropical forests, teeth will probably
remain the best index to the phylogenetic position

of tree shrews.

80

Although available evidence indicates
that living adult tupaiids show more similarity to
living adult Malagasey lemuroids than tO living
adult insectivores, Primates Obviously arose from
fossil, not living, insectivores of late Cretaceous
and early Tertiary times. A later insectivore
radiation gave rise to modern Lipotyphla. Even
in the Eocene epoch, leptictids, as well as moles,
shrews, and hedgehogs, still resembled tupaiids in
dentition and certain features of the ear region.
Leptictid fossils date from Cretaceous to Oligocene
times (McKenna, 1966).

Van Valen.(l965) observed that the evolu-
tionary path from Procerberus (the most primitive
known eutherian mammal, dating from the late

Cretaceous period) to Adapisoriculus (accepted as

 

leptictid) does not closely resemble that leading to
Paleocene Primates. There is greater similarity to
trends leading to primitive erinaceoids. Recent
tupaiids agree well with the trends exhibited by
Adapisoriculus of the Paleocene epoch.

The tritubercular teeth of Adapisoriculus

 

and Of Recent tupaiids more closely resemble those
of primitive forms of Didelphidae and Dasyuridae
than those Of any Recent or extinct placental

mammal (van Valen, 1965).

81

The primitive condition of the upper molar
tooth, with a raised trigone and a talon basin,
persists only in Tarsiidae, among Prosimii (Napier
and Napier, 1967).

In dentition, there appears to be greater
similarity between Adapisoriculus and either Tupaia
or Ptilocercus than there is between the latter two
genera; thus van Valen (1965) would place all three
genera in the family Tupaiidae.

The lower molars of Tupaiidae are also
primitive in being tuberculosectorial (Gregory,
1910), with sharply pointed cuSps and enlarged
styles. All prosimian families, except Tarsiidae,
have modified this trigonid to produce a quadricuspid
tooth (Napier and Napier, 1967).

Adapisoriculus, Plesiadapis (most primitive
known primate), and Anagale (questionable Oligocene
tupaioid) all retained the full placental comple-
ment of teeth:

--—--—-————3 : i : 2 : 3 =44 (Le Gros Clark, 1959; van Valen,
1965).
In modern tree shrews, the dental formula is

as follows: 2 - l - 3 - 3 ==38. This represents
3 - l - 3 - 3

less reduction than in any living primate. Reduction
is common in many mammalian orders, including Insec-

tivora. No primate has more than two incisors in

82
the lower jaw. In tupaiids, the third lower incisor
is reduced, having almost disappeared in Urogale.

Absence of the first premolar teeth is also
frequently observed among insectivores. Reduction
of canine teeth, typical of insectivores, also
occurs in many primate groups.

The lower incisors of tupaiids are somewhat
procumbent to form a comb or scraper. Generally,
the lower canines are not involved in this formation.
In one genus, Anathana, the canine is also quite
procumbent and may be interpreted (Buettner-Janusch,
1966) as related to an earlier stage in evolutionary
development of the tooth-comb found in modern
Lorisidae, Indriidae, and Lemuridae. In these prosim-
ian families the lower incisors, together with the
canines, form a closely packed series of horizontally
projecting teeth. (Certain Cebidae also show moderate
procumbency of the lower incisors.)

This specialization of Tupaiidae and many
Prosimii probably represents evolutionary parallelism.
This feature was not present in any early Cenozoic
prosimians. Ptilocercus has the tooth-Scraper less
well developed than in other tree shrews (Simons,
1962). Some lipotyphlous insectivores, such as
Blarina brevicauda (short-tailed shrew), display

”pronate” incisors (Gaughran, 1954).

83

Szalay (1968) has emphasized the primitive
nature of transversely oriented preparacrista and
postmetacrista for Metatheria and Eutheria. These
are correlated functionally with a transverse
shearing mechanism. This feature is not found in the
earliest primate fossils. Leptictids Show reduction
of this feature; early primates show even less of it.

Mixodectids, with probable close leptictid
ties and some similarities with Recent tupaiids and
Adapisoriculus, differ considerably from the Eocene
microsyOpids, which are probably closely related to
Paleocene prosimians (Szalay, 1968).

The dentition of Ptilocercus (most primitive
living tupaiid) “does not greatly resemble the den-
titions of primitive and early prosimians,“ according
to Szalay (1968). Tupaiids completely lack conules.
Strong, bulbous conules are present in microsyopids
and early primate families (paromomyids, plesiadapids,
and carpolestids); indeed Szalay (1968) would include
microsyopids within the order Primates. He would not
give leptictids the central position in relation
to eutherian evolution that McKenna (1966) and van

Valen (1965) have ascribed to them.

VSrtebral Column
The numbers of vertebrae in the different

regions of the tupaiid column is typical for primitive

84
mammals (modal numbers: 7 cervical, 13 thoracic,
6-7 lumbar, 3 sacral, 25 caudal vertebrae). In this,
Tupaiidae resemble prosimians more closely than
insectivores. Correlated with climbing and leaping
habits are the anteriorly directed lumbar para-
pophyses, shared with Macroscelididae and prosimians.
The overall arched condition of the back, in
Tupaiidae, is a primitive marsupio-placental condi-
tion (Gregory, 1910; Lyon, 1913; Le Gros Clark,
1926a; Davis, 1938; Schultz, 1961).

Sternum

In Tupaia, as in all prosimians and lower
simians, the sternum is very slender. The length
is comparable with prosimians and many insectivores
and is probably a primitive mammalian feature

(Gregory, 1910; Schultz, 1961).

Riga

The number of eternal ribs lies within the
prosimian primate range but probably approximates
the primitive mammalian condition (Gregory, 1910;

Davis, 1938; Schultz, 1961).

Skeletal System--Appendicu1ar Division
It has been rather generally accepted that

the most primitive marsupial and placental mammals

85
were somewhat arboreal in habitat; thus the modern

Opossum (Didelphis) and the tree shrew would be close

 

to these stem forms. Ckher authorities have pic-
tured the most primitive mammals as terrestrial
(Haines, 1958). In any event, tree-shrew limbs
conform closely to primitive arboreal mammals
generally. The length Of the hind limbs exceeds
that of the fore limbs. Limb-joint mobility closely
resembles that of plesiadapid Primates of the Paleo-
cene epoch (Le Gros Clark, 1959).

.Anterior Appendages

 

The tupaiid clavicle permits a full range of
prehensile activity (Lyon, 1913; Le Gros Clark,
1926a).

In Tupaiidae the humerus is approximately
equal in length to the forearm; in Lipotyphla the
humerus is relatively shorter (Gregory, 1910;

Le Gros Clark, 1926a).

Primitive mammalian features of the tUpaiid
humerus include the low, elongated deltoid crest,
the short and low supinator crest, as well as the
entepicondylar foramen and pronounced entepicondyle
characteristic of Ptilocercus. In Didelphis, Lemur
macaco, Erinaceus, and Sciurus, the deltoid and supi-
nator crests are relatively more sharply defined.

Anagale, the questioned tupaioid, resembles primitive

86
lemuroids more closely than the typical tupaioids
in having a stronger supinator crest (Gregory,
1910; Carlsson, 1922; Simpson, 1931).

Primate features of the distal tUpaiid
humerus include the distinct separation of the
capitellum from the trochlea by a low external
lip on the latter (Le Gros Clark, 1926a). As in
Lemuridae, the trochlea and capitulum of Tupaia
lie side by side in a transverse plane, indicative

Of a primitive position of radius and ulna; Erinaceus

 

differs in this respect (Carlsson, 1922).

Radius and ulna tend to remain distinct,
although there may be a varying amount of ossifica-
tion extending into the interosseous membrane; this
may be correlated with leaping habits (Lyon, 1913;
Carlsson, 1922; Le Gros Clark, 1926a).

Overall, the carpus of Tupaiidae is morpho-
logically intermediate between that of lipotyphlous
Insectivora and Lemuroidea.

In Tupaiidae and Lemuridae, as in marsupials
and most lower placental mammals, the dorsal surface
of the lunate bone makes broad contact with the ham-
ate bone. The area of contact is reduced in Erinacei-
dae and Sciurus (Carlsson, 1922).

A free as centrale, also regarded as a

primitive mammalian feature, is found in Tupaiidae

87
and Lemuridae, as well as in Sciurus, Talpa, and
many other Lipotyphla; it is absent in Erinaceus and
all marsupials (Carlsson, 1922).

As in Sciurus, Erinaceidae, and Setifer (a
primitive tenrecid insectivore), the navicular and
lunate bones are united in Tana tana and Tupaia to
form a broad, shallow area. Scaphoid (navicular)
and lunate bones are separate, however, in Ptilocer-
£3; and Dendrogale. This latter condition is both
more specialized and more lemurine (found in Lemur)
(Le Gros Clark, 1926a; Davis, 1938).

In Marsupialia the capitate and hamate
bones together are larger than navicular and lunate
bones; the size relations are reVersed for most
primitive placental mammals (Gregory, 1910).

Yet, Tupaia and Lemu£_resemble Marsupialia in this
respect, while Sciurus and Erinaceus resemble primi-

tive Eutheria (Carlsson, 1922).

Posterior Appendages

In tree shrews the os innominatum generally
shows the form characteristic of primitive arboreal
mammals (Gregory, 1910).

The ilium of Tupaia and PtilOcercus is more
spatulate than in elephant shrews, thus being further
differentiated from the Simple trihedral shape Ob-

served in Didelphis. In Tupaia the sacral surface

 

 

 

88
is much wider than the iliac surface. An enlarged
sacral surface (for ligament and muscle attachments)
is found in all Lemuroidea and monkeys. The rela-
tive width of the ilium increases but little in the'
lower Lemuroidea (Galago, Microcebus, Cheirogaleus)
(Straus, 1929).

 

The overall height-width index of the ilium
in Tupaiidae resembles Lemgr. AS in Lemar, the iliac
surface in tupaiids is absolutely widest in the
lower posterior part, definitely narrowing toward
the crest, where pubic and acetabular borders are
confluent (Straus, 1929).

The ilium of Ptilocercus is more primitive
than in Tupaia, showing less concavity of the lateral
Surface, and a simple rod-like form. Lipotyphlous
ilia are even more rod-like and non-spatulate. The
gluteal surface is nearly flat in Rhynchocyon and
Macroscelides. Iliac surfaces, in these elephant
shrews, are directed ventrolaterally, as in opossum
(Straus, 1929).

The acetabular border of the ilium is
straight in opossum, almost straight in elephant
shrews, but somewhat concave in tupaias. Generally,
this border remains straight in Lemuroidea. It is
quite concave, however, in Lemar, Cheirogaleus,

Lichanotus (woolly lemur), Propithecus, and Indri

89
(Straus, 1929).

There is a prominent antacetabular Spine
(anterior inferior iliac spine) in Tupaia. This
spine is not found in most marSUpials and lipo-
typhlous insectivores. While large in all Meno-
typhla, the antacetabular spine shows much varia-
tion among Prosimii, from a trace to considerable
enlargement (Straus, 1929).

The sacroiliac articular surface in Meno-
typhla, Lemuroidea, and monkeys resembles that of
lower mammals generally. The surface takes a
shallow U Shape, with ventral arm somewhat the longer
(Straus, 1929).

Among primitive mammals it is the rule to
have only one or two (sacral) vertebral segments
articulating with the ilium. Marsupials Show one
or two. Lipotyphla Show one to three. Among
Menotyphla, Tupaia and Rhynchocyon have just one,

 

Ptilocercus and Macroscelides have two. The variation
among Lemuroidea ranges from one to three (Straus,
1929).

Straus (1929) concluded that minor changes
could transform a tupaioid ilium into a lemuriform
ilium. The simplest change would lead to Lepidolemur
and Lemur, and thence to Indris and Propithecus. If
this did occur, however, one must explain the appar-

ently more primitive ilia found in most other

90
Lemuroidea. If, as an alternative hypothesis,
one assumes evolution of the tupaioid ilium in
parallel with Lemar and allies, one must search
for the ancestral lemuroid type among other Meno-
typhla or related forms. The fossil evidence re-
mains inconclusive.

The symphysis in Menotyphla (Tupaiidae and
Macroscelididae) involves union of both pubis and
ischium. This is a primitive mammalian condition.
Lipotyphla show some degree of reduction, or even
complete absence, of such a symphysis. Prosimii have
a strictly pubic symphysis (Carlsson, 1922; Le Gros
Clark, 1926a; Davis, 1938).

The gluteal ridge ("third trochanter“) of
Tupaia is prominent, extending farther distally than
the lesser trochanter. This is a primitive condition

shared with Sciurus and Didelphis. In Erinaceus, it

 

 

is represented by a crest (Gregory, 1910; Carlsson,
1922).

Another primitive feature of Tupaia is the
shallow patellar trochlea (Gregory, 1910).

Having tibia and fibula completely separated
throughout their length is a primitive feature shared
with Didelphis and Prosimii (except Tarsius). Fusion
is constant in Macroscelididae (jumping forms) and is

common among Lipotyphla (with many fossorial forms)

91
(Gregory, 1910; Lyon, 1913; Carlsson, 1922).

The lemuroid features of the tupaiid tarsus
may be correlated with adaptation to an arboreal
life. The strong malleoli are associated with sub-
vertical facets on the astragalus. This arrangement
is also found in Sciurus (Carlsson, 1922).

The broad confluence of the sustentacular
facet with the navicular facet of the talus, in
Tupaiidae, is also observable in Lemgr and Sciurus;
the facets are clearly set apart in Erinaceus. In
Tupaia, as in Lemar, the ectal facet is the larger;
the two facets are about equal in Sciurus. Since,
in the primitive Marsupialia, the sustentacular facet
is in continuity with that for the navicular bone,
Carlsson (1922) felt justified in viewing this
common feature of £2222 and Tupaia as a Sign of gene-
tic relationship.

The facet Of metatarsal I for articulation
with the entocuneiform bone is oblique and somewhat

saddle-shaped in tupaiids. The obliquity is slight,

 

however, in comparison with Lemur, in which the facet
is turned medially for great opposability of the hal-
lux (Carlsson, 1922; Buettner-Janusch, 1966).

The terminal phalanges of the pes in Anagale
were flattened and spatulate. Since this resembles

the condition in various advanced prosimians more

92
closely than that of modern tree shrews, it has been
thought that this fossil form would tend to lessen
the morphological gap between tree shrews and lemurs
(Simpson, 1931; Le Gros Clark, 1959).

Haines (1958) believed that the primitive
placental hand was probably of the convergent type,
in which the tips of the digits converge to lie close
together in flexion Upon the palm. ‘This is still
commonly found in reptiles, most insectivores (except
burrowers and tUpaiids), as well as carnivores. A
claSping hand, as found in Didelphis among other
marsupials, is used in climbing. The entire hand
grasps the branch. A prehensile hand with opposable
pollex is found in some marsupials, primates (including
all lemurs), and (incipiently) in tupaias. Haines
(1958) pointed out, however, that the opposability in
tree shrews may have resulted from parallel evolution.
Among the unique features may be noted the articula-
tion of the trapezium with the scaphoid, centrale,
and trapezoid bones. The trapezium also shows close
ligamentous attachment to metacarpal II, and receives
some fibers from the abductor pollicis longus muscle.
The articular surface for the first metacarpal is
saddle-shaped, although the articular cartilage
and synovial cavity are continuous with those of the

scaphoid. This peculiar type of opposability may have

93
been superimposed on the clasping type of hand,
Haines (1958) believed.

Whereas Haines (1958) thus regarded the
Tupaiidae as too Specialized (at least in the adult)
to approach the basic mammalian stock, Steiner (1951)
regarded Tupaiidae as the most primitive of living
mammals, based on embryological study of the hand
and foot skeleton of Tupaia. Steiner (1951) also
believed it is the arboreal mode of life that
"stamps” the adult structural organization of pri-
mates and tupaiids as well as the basic mammalian
type. Among the features--paralle1ing many other
mammalian features--first appearing in the basal
cheiridial elements of theromorph reptiles are the
elongated form of the first carpal and tarsal bones
and a degree of opposability of the pollex and
hallux. Steiner (1951) found that ontogeny of
the tUpaiid hand and foot skeleton demonstrates deri-
vation of mammalian features directly from reptiles.
He found all four centralia of primitive tetrapods
present in the mammalian carpus. Overall, the em-
bryonic skeleton of hand and foot is surprisingly
similar to that of the lower tetrapods. The later
deviations, generally regarded as typical mammalian
features, manifestly are adaptations to an arboreal

way of life, according to Steiner (1951).

94

Based on his ontogenetic study of the hand,
Lorenz (1927) concluded that Tupaia is more primitive
than, and probably not ancestral to, the Primates.
Although the adult hand is narrow and adapted for
"claw-climbing," the embryonic hand is broad and
of a primitive "clasping-hand” construction. Primates
(barring certain Specialized "claw-climbing,“
“callosity-climbing,” and brachiating variants) are
still more or less clearly ”clasping-hand climbers,”
with a relatively much broader adult hand than in
Tupaia. The embryonic tupaiid hand approximates most
closely (of any known living form) the primitive
clasping hand of the hypothetical ”climbing-creeping
primitive mammal.”

Napier (1961) classified primate hands as either
"convergent" or "prehensile with pseudo-opposable or
opposable thumbs.“ If an animal habitually picks
up and holds food in one hand, Napier (1961) would
regard it as having a "prehensile" hand. On this

basis, the hand of Tupaia glis is considered to be

 

“convergent.” Despite some divergence of the thumb,
the tree shrew does not usually pick up food in one
hand; thus Tupaiidae would be the only Primates
with convergent hands.

Napier (1961) pointed out that the consider-

able degree of divergence of the pollex indicated in

95

Eocene mammals need not imply Opposability, for
such divergence was a feature of primitive reptiles
as well. The feature may be considered as pre-
adaptive for later opposability.

Prehension, in Lemuriformes and Lorisiformes,
shows some increase in complexity over Tupaiidae.
In reaching for an Object, most prosimians Open the
hand and then close the fingers around the object.
Fingers are not simply flexed in one plane, as in
Tupaia. The advancement here involves hand structure,
rather than any improvement in nervous control; there
is still no variability in prehensive pattern

(Bishop, 1962).

Muscular System,

 

In their studies of tupaiid musculature,
Carlsson (1922) and Le Gros Clark (1924b) used the
squirrel (Sciurus) as a kind of “control" to deter-
mine whether features peculiar to Tupaia among insec-
tivores might be arboreal adaptations. There is also
a difference in degree of arboreal activity among
Tupaiidae. The small Species, such as Inpaia minor,
with relatively longer tails, are so arboreal as to
frequent the higher branches of large trees; the
larger Species, such as Tupaia ferruginea, with
short tails, tend to be ”bush animals.” Despite

these differences among tUpaias and squirrels, there

96
are no corresponding differences in musculature; thus
the primate-like features found in tupaias cannot be
readily dismissed as secondary (convergent) adapta-
tions to an arboreal life. The features in which
Tupaia agrees with Lipotyphla and differs from
Lemuroidea are generally regarded as primitive

mammalian characters.

Muscles of the Head and Neck

In the head and neck region, features of
Tupaia that have been regarded as more or less
primitive and not shared with prosimians include the
following:

1. Musculus temporalis is composed of three parts
(Carlsson, 1922).

Although Arnbgck-Christie-Linde (1907) found
three parts in Crocidura and Sgrex, no separation
into three parts could be demonstrated in Scalopus
aquaticus by Gaughran (1954). The muscle shows three
regions but is not split into three parts in Suncus
murinus (Sharma, 1958).

2. Musculus digastricus has its bellies separated
only by a thin tendinous intersection in Ptilocercus,
although Tupaia shows a well-develOped central ten-
don. The tupaiid digastric muscle joins that of

the opposite side for a short distance in front of

the hyoid bone (Le Gros Clark, 1926a).

97

Sharma (1958) pointed out that "this is one
of the most variable muscles in the Insectivora.”
In Suncus (Sharma, 1958) and Blarina (Gaughran,
1954) a very delicate tendinous intersection is
present. No tendinous inscription was observed in
ScalOpus or in'IEEEE (Gaughran, 1954). Thus, among
insectivores, three types of musculus digastricus
exist: (1) with tendinous intersection, (2) without
tendinous intersection, and (3) with intermediate
tendon. Tupaiidae alone, among insectivores, may have
the type with intermediate tendon (Sharma, 1958).

The interrelations of the paired anterior
bellies also is variable. In Blarina, the antimeres
are attached by a fascial sheet. In Talpa, fibers
of the left tendon run deep to those of the antimere.
In ScaloEus, anterior belly fibers interlace along
the mid-sagittal plane with those of the antimere.
This condition in Scalopus is regarded (Gaughran,
1954) as more primitive than that in Blarina.
3. Differentiation of a scalenus anticus muscle
has occurred in Tupaia, although it is not yet
distinct in Ptilocercus (Le Gros Clark, 1926a).

Musculus scalenus Shows three distinct
parts (ventral, medial, and dorsal), but all lie
dorsal to the brachial plexus in Blarina and Suncus

(Gaughran, 1954; Sharma, 1958). No distinct

98
subdivisions are demonstrable in ScaloEus (Gaughran,
1954).
4. The medio-ventral border of musculus platysma,
although very poorly defined in Ptilocercus, does
become definable in Tupaia (Lightoller, 1934).
5. The notoplatysma muscle (with origin posteromedial
to the acromioclavicular joint) is well developed,
but trachelOplatysma (with origin anteromedial to
the acromioclavicular joint) is still lacking in

Tupaia javanica (Lightoller, 1934).

Features of head and neck musculature
indicative of primate trends are the following:
1. Musculus digastricus shows a well-develOped cen-
tral tendon in Tupaia (Le Gros Clark, 1924b).
2. Musculus scalenus anticus is differentiated in
Tupaia, as in Lemuroidea (Le Gros Clark, 1924b).
3. The medio-ventral border of the platysma muscle
is defined in TUEaia and lemurs (Le Gros Clark,
1924b).
4. Lightoller (1934) stated that all four grOUps he
examined (Tupaia, Hapale, Lemur, and Tarsius species)
show ”the primate modification of the mammalian plan
of facial musculature.” This includes the extension
of the naso-labial muscle into the medial frontal
region, the presence of a SUperficial orbito-

auriculo-labial musculature, simplification of the

99
retro-auricular musculature, and subdivision of the
orbicularis oris muscle into peripheral and marginal
portions (Le Gros Clark, 1924b, 1926a; Lightoller,
1934).

Muscles of the Abdominal:wall

A tupaiid feature of the abdominal wall mus-
culature that is considered primitive, and not shared
with Prosimii, is the presence of a pyramidalis
muscle (Carlsson, 1922).

TUPaiidS share the following features of
abdominal wall musculature with Lemuroidea:
1. External oblique muscle fibers continue over the
rectus abdominis muscle almost to the linea alba

in Tupaia. In Lemuridae (especially Cheirogaleus)

 

and in Nycticebus muscle fibers Spread over the rec-

 

tus abdominis muscle to a varying extent. In Lipo-
typhla, however, the aponeurosis is broad (Carlsson,
1922).

2. The rectus sheath of Tupaia and Ptilocercus shares
a structure with Madagascan lemuroids that is found
in no other mammals except Charis (sea lion). The
aponeurosis of the tUpaiid internal oblique muscle
lies dorsal to the rectus abdominis muscle in its
cranial portion; in its caudal portion, however, the
internal oblique aponeurosis lies ventral to rectus

abdominis (Kaudern, 1910; Carlsson, 1922; Le Gros

100
Clark, 1924b, 1926a). This peculiarity is most
strongly marked in Tupaia, is somewhat less evident

in Lemuridae (Lemur, Cheirogaleus, PrOpithecus), and

 

 

is absent in Lorisiformes and higher Primates
(Kaudern, 1910; Carlsson, 1922; Le Gros Clark,
1924b). Kaudern (1910) believed this to be a very
ancient feature. Carlsson (1922) observed that the
internal oblique aponeurosis lies ventral to the
rectus abdominis muscle in all Insectivora except

Talpinae, in which it lies dorsal to that muscle.

Muscles of the Anterior Extremities

Probable primitive features of the muscula-
ture of the anterior extremities of TUpaiidae include
the following:
1. The tendon of insertion of the extensor carpi
radialis muscle Splits into two parts as it crosses
the abductor pollicis longus tendon in Tupaia species,
in Erinaceus (Carlsson, 1922), and in Suncus (Sharma,
1958). The larger tendon attaches to the third, and
the smaller to the second, metacarpal bone. In
Rhynchocyon (elephant shrew), the single tendon
attaches only to the third matacarpal bone. The

muscle is completely divided into two in Didelphis

 

(opossum), Sciurus (squirrel), Lemuroidea, and most
higher Primates (Carlsson, 1922).

2. Musculus abductor pollicis brevis has but one head

101

in Tupaia species, Erinaceus, Tenrec, Macroscelides

 

 

 

(elephant shrew) (Carlsson, 1922), and Suncus
(Sharma, 1958). It arises by two heads in Lemuroidea
(Carlsson, 1922).

3. Musculus opponens pollicis has not differentiated

in Tupaia, Sciurus, Didelphis, or Erinaceus. While

 

 

remaining inconstant in Prosimii, the muscle is
present in Callithrix (marmoset) (Carlsson, 1922).
4. Musculus opponens digiti minimi is undifferenti-
ated in Tupaia, in Lipotyphla (e.g., Tenrec), and in
Sciurus. This muscle is usually developed in
Didelphis, Lemuroidea, and Callithrix (Carlsson,

1922).

 

5. Of the deep digital extensor muscles, extensor
pollicis longus and extensor indicis proprius together
form one muscle, as in Erinaceidae (also in Suncus,
according to Sharma, 1958), many other mammalian
orders, and the South American monkeys. The Old World
Primates, at least Lemur among Lemuroidea, differ. In

Didelphis, also, both muscles appear. In Sciurus

 

and Macroscelididae, the extensor pollicis longus

muscle is absent (Carlsson, 1922).

Other primitive features of tupaiid forelimb
musculature shared with lemuroids include the follow-
ing:

1. A brachioradialis muscle is completely absent in

102
Ptilocercus. In this, it resembles most Lipotyphla
and differs from Tupaia, in which this muscle is pre-

sent, at least in Tupaia minor and Tupaia ferruginea.

 

 

Normally, it is also found in lemurs and other Prosimii.
It is possible that the brachioradialis muscle has
disappeared in Ptilocercus; or, if the muscle has not
yet differentiated from the extensor carpi radialis
mass (prominently developed in this genus), the
absence may be primitive (Le Gros Clark, 1924b,
1926a).

2. In Tupaia species, the extensor digitorum communis
and extensor digitorum lateralis muscles arise by a
common head that quickly divides into a stronger
radial extensor digitorum communis and a weaker

ulnar extensor digitorum lateralis. In the proximal
half of the forearm, the tendon of extensor digitorum
communis splits into four parts, continuing on to the
second through fifth digits. The relatively longer
tendon of extensor digitorum lateralis Splits into
two parts, leading to the two ulnar fingers. Both
Rhynchocyon and Callithrix lack a tendon to the fifth

 

digit. The two tendons are present in Didelphis,
Erinaceus, Petrodromus, Macroscelides, Sciurus, and
Prosimii (Carlsson, 1922), as well as in the soricid,
Suncus (Sharma, 1958).

3. The three contrahentes muscles are attached to

103
the first, second, and fifth fingers in Tupaia, as

also in Didelphis, Tenrecidae (Madagascan insec-

 

tivores), Macroscelididae, and Lemuroidea. None

are usually retained in Erinaceus (Carlsson, 1922)

or in Suncus (Sharma, 1958). Primitively, in uro-
deles and Emyg (Tertiary marsh-dwelling turtles),

the contrahentes muscles attach to all five fingers.
Their reduction in Tupaia is thus less than in
Sciurus, with only contrahens V remaining (Carlsson,
1922).

4. The omohyoid muscle of Tugaia shows a distinct
intersection, as in Primates and Chiroptera. This
muscle is apparently absent in many lipotyphlous
Insectivora (Krnbgck-Christie-Linde, 1907; Le Gros
Clark, 1924b; Gaughran, 1954; Sharma, 1958). In
Blarina, however, Gaughran (1954) found a thin, strap-
like omohyoid muscle, without tendinous intersection.
5. Tupaia species have four lumbricales muscles, as
do Didelphis, Gymnura, Macroscelides, and Lemuroidea,
as well as Sciurus. In Erinaceus, the number is var-
iable or the muscles may be completely absent
(Carlsson, 1922).

6. Since dorsal interosseous muscles are found in
urodeles, Trichosurus (common phalanger or Australian

"opossum"), and Didelphis (common Virginia Opossum),

 

as well as several other marSUpials, they are regarded

104
as primitive muscles. While clearly apparent in
Tupaia, Lemuroidea, and Callithrix, they are not ob-
servable in Sciurus (Carlsson, 1922).
7. Lower mammals have musculi flexores breves
profundi for each finger. The full primitive com-

plement persists in Tupaia, as well as in Didelphis,

 

Erinaceus, Prosimii, and Callithrix (Carlsson,
1922), and also in Suncus (Sharma, 1958). Sciurus
shows some reduction in having these muscles only
for the four ulnar fingers (Carlsson, 1922).

8. Musculus flexor pollicis brevis sublimis is

present in Tupaiidae, Didelphis, Tenrec, and Lemur-

 

oidea (Carlsson, 1922; Le Gros Clark, 1924b, 1926a).
It is absent from Erinaceus, Sciurus (Carlsson,
1922), and Suncus (Sharma, 1958).

Prosimii have a far more advanced differen-

tiation of the hand muscles generally than is attained

in Tupaia.

Tupaiids and lemuroids share a few features

that probably represent advances. These include the
following:
1. The coracobrachialis muscle is normally fully
developed in tupaiids and lemurs, but frequently is
lacking (wholly or in part) among Insectivora
(Davis, 1938).

Sharma (1958) did not find this muscle in

105
Suncus, and noted that it has not been reported for
any soricoid genus. Musculus coracobrachialis,

however, is present in the talpid, Uropsilus, which

 

is regarded as primitive.
2. In Tupaia, the dorso-epitrochlearis muscle is
completely duplicated. Two heads of origin are

found in Dendrogale (smooth-tailed tree shrew) and

 

in lemurs. This similarity is probably due to loco-
motor patterns for climbing. Although double

here, many animals more arboreal than tree shrews

and lemurs have a simple dorso-epitrochlearis

muscle, as do many cursorial animals. The single
head unites with the triceps brachii muscle in insec-
tivores such as Erinaceidae, Macroscelididae, and
Suncus (Soricidae) (Carlsson, 1922; Le Gros Clark,
1924b; Davis, 1938; Sharma, 1958).

3. In Tupaia and Dendrogale, as in Lemur and other

 

Prosimii, the lateral head of the triceps brachii
muscle is double, having both superficial and deep
parts. Only the upper part (arising from the neck of
the humerus) apparently is represented in lipotyphlous
insectivores and sciuromorph rodents (Le Gros Clark,l
1924b; Davis, 1938; Sharma, 1958).

4. It has been observed that both the first and fifth
digits of Tupaia can diverge widely. Musculi contra-

hentes II and V here arise from a common median raphe.

106
Musculus contrahens I is independent of this raphe,
to form a small muscle controlling pollex alone.
(A similar pattern is found in the foot.) This
feature has been interpreted by Le Gros Clark
(1926a, 1927) and Napier (1961) as indicative
of close affinity between Tupaiidae and Lemuri-
formes. To Haines (1958), however, this similarity
between Tupaia and Lemur arose by convergence for
arboreal locomotion. In Tupaia the main raphe
lies between contrahentes II and V, with a smaller
one occurring between I and II. In Lemur, however,
the major raphe lies between contrahentes I and IV.
The major raphe, providing some functional Oppos-
ability, thus shows differing anatomical develOpment
in these two groups. Tupaia also shows a far more
specialized arrangement of its hypothenar muscles
(for opposition of radial digits to hypothenar pad)
than does Lemur (Haines, 1958).
5. Some muscle fibers called "Carpalballenmuskeln"
(muscles of the carpal pads) are found in both carpal
pads of Tupaia species. Such have not been described
for lipotyphlous insectivores. An ulnar Carpalball-
enmuskel is differentiated in Sciurus, Lemuroids have
both the ulnar and radial muscles develOped. Such

muscle fibers are absent in Didelphis on the one hand,

 

and in Callithrix on the other (Carlsson, 1922).

107
Muscles of the Posterior Extremities
Muscles of the lower extremities show
similar trends, thus demonstrating greater similarity
of Tupaiidae to Lemuroidea than to most Insectivora.
Among the similarities to Lemuroidea are the follow-
ing:

1. The elements of the iliopsoas muscle are fully
differentiated at their insertion in Tupaia and
Ptilocercus (Le Gros Clark, 1924b, 1926a). The form
and disposition of the iliacus muscle in Tupaia and
Galago are identical; it is similarly disposed in
Tarsius (Carlsson, 1922; Le Gros Clark, 1924b).
Davis (1938) found the iliacus muscle of Dendrogale

 

to be quite similar to that of Tupaia, with the
usual lateral and medial heads.
2. A distinct pyriformis muscle is present in

Ptilocercus, Tupaia, and Dendrogale, as well as in

 

Macroscelididae and Prosimii, but absent in all
Lipotyphla (Le Gros Clark, 1924b, 1926a; Davis,
1938; Sharma, 1958).

3. Davis (1938) described the gluteus medius and
gluteus minimus muscles as quite readily separable

in Dendrogale, with gluteus medius being bilaminar.

 

Carlsson (1922) found gluteus medius superior to be
a relatively more independent muscle in Tupaia,

Macroscelididae, and Lemuridae than in Lipotyphla.

108
4. Musculus sartorius is normally present in
tupaiids and lemurs but lacking in lipotyphlous

insectivores except Erinaceus (Le Gros Clark, 1924b,

 

1926a; Sharma, 1958).
5. The extent of the tibial attachment of the

popliteus muscle in Ptilocercus agrees with Tupaia.

 

This is considered to be a lemuroid feature since
the muscle does not extend downward beyond the
upper fifth of the tibial shaft. Other groups

display greater length: Didelphis and Suncus,

 

one-third; Erinaceus, one-half; Callithrix, one-
fourth (Carlsson, 1922; Le Gros Clark, 1924b, 1926a;
Sharma, 1958).

6. Musculus plantaris in Tupaia is united with the
lateral head of the gastrocnemius muscle, as in
Lemuridae, but not as in Erinaceus (Carlsson, 1922).

7. Musculus flexor brevis digitorum in Tupaia re-
sembles Lemuroidea but is unlike Lipotyphla and
Sciurus in having two heads of origin from the deep
flexor tendons and the plantar fascia. In having the
muscle attached directly to 05 calcis, Tugaia parallels
only Primates and some Chiroptera (Carlsson, 1922).

8. Musculus crureus frequently is absent among
Lipotyphla, although found in Tupaia and Lemuroidea
(Le Gros Clark, 1924b).

9. Occasionally, Tupaia shows a distinct musculus

109
subcrureus. Its presence is normal in Prosimii.
It is completely absent in LipOtyphla (Le Gros
Clark, 1924b).
10. Tupaia, with three musculi contrahentes, shows
less reduction of this muscle group than any insec-
tivore. Tarsius and Lemur retain musculi contra-
hentes for all five toes (Carlsson, 1922).
ll. Musculus abductor hallucis is well develOped in
Tugaia and Prosimii. It is absent in Insectivora,
according to Le Gros Clark (1924b). Sharma (1958)
found a long, slim musculus abductor hallucis in
Suncus murinus.
l2. Musculus peroneus digiti quarti is differentiated
in Tupaia and lemurs but not in insectivores,
according to Le Gros Clark (1924b). This muscle
is present, however, in Suncus murinus (Sharma, 1958).
13. Musculus peroneus quinti and musculus quinti
digiti, in being differentiated into parts, shows
similarity between Tupaia and Lemuroidea (Le Gros
Clark, 1924b; Sharma, 1958).

Primitive features of the musculature of
the posterior extremities include the following:
1. A tensor fasciae femoris muscle is undifferenti-
ated (from gluteus maximus) in both Ptilocercus and
Tupaia, although Le Gros Clark cites Leche as having
found it in Tupaia ferruginea (Le Gros Clark,

110
1924b, 1926a). It is of interest that this muscle
is differentiated in Suncus (Sharma, 1958).

2. The tenuissimus muscle is present in Ptilocercus

 

and Tupaia; it is generally present in Lipotyphla
but absent in Lemuroidea. Since this muscle repre-
sents the femoral head of musculus biceps femoris

of higher Primates, it is clearly a primitive
feature (Le Gros Clark, 1924b, 1926a; Sharma, 1958).
3. Musculus extensor digitorum communis longus of
Ptilocercus and Tupaia is evidently primitive in its
origin from the femur (lateral condyle). Lipotyphla
and Sciurus also retain a femoral origin. In
Lemuroidea, the origin has shifted (presumably

secondarily) to the tibia and fibula. Didelphis shows

 

a somewhat parallel secondary shift to the head of
the fibula (Carlsson, 1922; Le Gros Clark, 1924b,
1926a; Sharma, 1958).

Circulatory System

 

Serology

J. Buettner-Janusch and V. Buettner-Janusch
(1964) have reported an electrOphoretic study of cell
hemolysates from different primates and insectivores.
One can thereby differentiate certain taxa above

generic levels. Hemoglobin of Tupaia glis has one

 

component that moves more slowly than hemoglobin of

any primate studied. It does not resemble the

111

hemoglobin of any insectivores studied. Rhynchocyon

 

chrysopygus, an elephant shrew, has two very fast-
moving components, resembling the hemoglobin of Perodic-
ticus potto. A feature common to adult prosimians and
elephant shrews is the relatively high level of
alkali-resistant hemoglobin. In Anthropoidea, only
fetal blood contains alkali-resistant hemoglobin.

In 1963, Maisel and Goodman reported an
immunoelectrophoretic study of proteins in primate
lenses. Precipitin bands decrease with phylogenetic
distance from the homologous species; thus bands
decrease as one proceeds from Old World monkeys,
to hominoids, to New World monkeys, to prosimians
(including tree shrew), to non-primates. No reac-
tion occurs with a great variety of non-primate
mammals. This would tend to confirm primate affinity
for the tree shrew. There was no indication that
insectivores were included in the study.

In the immunochemical study made by Hafleigh
and Williams (1966), relative cross-reactivity was
determined by the quantitative precipitation reaction.
Serum albumins were tested against human serum albumin.
The immunochemical correspondence value for Tupaia
glis (31 per cent) is well within the range of

Prosimii, being a few curve area percentage points

112
lower than for Lemur species (34-37 per cent), but
similar or a few points higher (closer to human)
for Perodicticus_potto (31 per cent), Galago

crassicaudatus (28 per cent), and Propithecus verrauxi

 

(22 per cent). The only insectivore tested was
Erinaceus (17 per cent), with a correspondence
several points below that of any prosimian, although
in a range expected for a near relative of primates.
(Pig shows 8 per cent correspondence.)

If the isoimmunization theory of evolutionary
arrest (Goodman, 1963) is correct, the high cross-
reaction of the tree shrew albumin might not be an
accurate indication of degree of relationship. The
correspondence to other prosimians might be a coin-
cidence. Further qualitative and quantitative analy-
sis of prosimian and insectivore albumins is neces-
sary to clarify the nature of the relationship. Good-
man's (1964) report indicates that lorisids are
closer to lemurids than to either tupaiids or anthro-
poids. With respect to serum proteins at least,
lemurid genes have evidently shown more rapid diver-
gence from the ancestral or basal primate type than
have the homologous genes in tupaiids and lorisids.

In his 1962 report of similar results, Goodman
indicated that some serum protein types evolve more

rapidly than others. Albumins apparently change

113
slowly, showing marked divergence in Primates,
only as the lemurid-lorisid-tupaiid level is
reached; gamma globulins, however, have evolved much
more rapidly. Albumin is synthesized early in fetal
life; gamma globulins, late.

In cross-reactions of antisera to human
serum proteins (Goodman, 1963), tupaiid serum
globulins show greater resemblance to those of
hedgehog than to those of primates; yet tUpaiid
serum albumins resemble those of primates more
closely. Human albumin correSponds more closely
with tree shrew albumin than with lemur albumin.
Goodman (1963) postulated that albumin evolution
would have been more rapid in lemur (with epithelio-
chorial placenta) than in tupaiids (with endothelio-
chorial placenta) and human females (with hemo-
chorial placenta). In the latter two situations,
maternal isoimmunizations would have caused reten-
tion of very primitive genetic patterns. Since serum
globulins develop at a later ontogenetic stage, they
would not have been so much affected, even though
hedgehog also possesses a hemochorial placenta.

Based on production of chicken antisera to
the serum of prosimians and insectivores, Goodman
(1963) postulated a kind of phylogeny. Although

there are antigenic differences between lorisids

114
(Galago, Loris, Perodicticus) and lemurids (Lemur),

 

each family resembles the other more closely than
either resembles tupaiids or anthropoids. It may

be concluded that Lorisiformes and Lemuriformes did
not separate until after their common ancestor had
separated from antecedents of Anthropoidea. There ap-
pears to be no Special relationship between Tupaioidea
and Lemuroidea. Using anti-hedgehog and anti-tree-
shrew sera, Goodman (1963) found closer correspondence
between hedgehog and tree shrew than either shows

to other mammals tested (kangaroo, galago, lemur,

man, rat, rabbit, dog, elephant, horse, cow); Yet

the tree shrew-hedgehog affinity is not at all

close since the heterologous cross-reaction, in

each case, was far less than the homologous one.

Thus, Tupaioidea may well have separated from other
mammals during the initial major eutherian radiation
of the late Cretaceous period.

Goodman (1962) used two-dimensional starch-
gel electrOphoresis to separate macromolecular com-
plexes within serum. These separations show con-
stellations of components forming patterns that
are distinctive for taxonomic groups. Tree shrew
(Tupaia glis) and hedgehog (Erinaceus) have very
different patterns.

Johnson and Wicks (1962), using

115

electropherograms to characterize serum proteins, found
each Species has a qualitatively and quantitatively
distinct pattern. In some orders the albumin migrates
farther than in the human control; in others, less
far. In Didelphidae (Marsupialia), albumin migrates
at a rate which is 90 per cent of that of the human
control. The rate in Soricidae (shrews) is about
the same as in man, while the albumin moves slightly
faster in Talpidae (moles). Cerc0pithecidae (Cfld
World monkeys) are more similar to Hominidae than
are Tupaiidae; the latter differ no more from
Hominidae than one family does from another within
other orders.

.An electrophoretic study by Lange and
Schmitt (1963) showed that Tupaia glis has no

 

recognizable prealbumins. Prealbumins are also
lacking in Cgbgg and Ateles and are weak in Cacajao--
all New-World genera. Lgmg£_has but one prealbumin;
Pongidae and most Cerc0pithecidae have two.

Tupaia shows an eSpecially short gamma-globulin

zone, one shorter than any primate tested. .A uniform,
haptoglobin 1-1 phenotype is found in Tupaia, as in
all Primates observed. In contrast, the transferrin

system diSplays great polymorphism among Primates.

 

Tupaia shares the A5 transferrin with Presbytis and

Lemur, and Shares the Bo transferrin with Europide

116

Homo, Symphalangus, Macaca, Papio, and Presbytis.

 

Layton (1965) pointed out that human atopic
reagins (serum antibodies responsible for hyper-
sensitivity of mucous membranes, as in hay fever)
can be used to sensitize (passively) other Primates.
He was able to induce passive allergy in a variety

of monkeys and prosimians but not in Tupaia glis

 

or in any of a variety of non-primates, including

Opossum and bat but no insectivores.

Circulatory Organs

Starck (1960) noted that the spleen displays
different gross forms showing evolutionary trends.
The elongate, ribbon-like form observable in Tupaiidae
may be regarded as the ancestral primate (also
mammalian?) model. This form is found in many
Prosimii and in Insectivora; however, among
prosimians, a number of specializations appear.

Frick's (1960) description of the heart of
Tupaia glis indicated that it Shares in the variabil-
ity of prosimians in several features. Without com-
parable descriptions of insectivore hearts, little
can be said regarding phylogeny. Le Gros Clark
(1926a) stated that the heart Shows the ”generalized
mammalian condition." In SUpport of this is the
observation (based on Frick's illustrations) that

base-apex length exceeds base diameter considerably,

117
and that the heart axis deviates but little from
the long axis of the thorax. Hill (1953b) noted
that the following are usually absent in lower
mammals: coronary (Thebesian) valves, caval
(Eustachian) valves, fossa ovalis, and annulus
ovalis. In Tupaia, Prick (1960) found that fossa and
annulus ovalis are both present; the caval valves
are present, but share in the variability usual
among strepsirhines; typical coronary valves are
lacking.

In connection with the system of blood
vessels, few phylogenetic trends can be indicated
due to lack of adequate data (Davies, 1948).

Le Gros Clark (1926a) noted that the absence
of a common iliac artery in Ptilocercus may be
regarded as primitive. Such a common vessel is
usually present in strepsirhines (Hill, 1953b).

Ptilocercus Shares with Lemur catta the
arrangement of arteries in the arm (brachium),
where the brachial artery iS relatively small and the
principal vessel is the SUperficial brachial
artery (Le Gros Clark, 1926a).

The presence of paired precavae, as in Tupaia
and Ptilocercus, is characteristic of primitive
mammals (Le Gros Clark, 1926a).

Ptilocercus has a single (left) azygos

 

118
vein. This is found in many marsupials, including
Phalangeridae. (A left vena azygos is also found
in certain rodents and artiodactyls.) (Le Gros
Clark, 1926a)

Except for description of the mode of
branching of the arch of the aorta and the relations
of the internal carotid system in the middle ear
cavity and walls, little information has been
published on cranial arteries of Ptilocercus (Le Gros
Clark, 1926a), and virtually no information is

to be found for Tupaia or Dendrogale species.

 

The observations of phylogenetic interest will
be noted in the discussion of results of the present

study.

MATERIAL AND METHODS

For the study of the carotid vessels, four
mature individuals (three males and one female) of

the Malayan tree shrew, Tupaiagglis Diard, were

 

used.

The first three tree shrews (numbered: '1,
2, and 3) were embalmed after death. The preserving
fluid used was a standard mixture of potassium
nitrate, borax, boric acid, thymol, menthol,
salicylic acid, oil of Wintergreen, ethylene glycol,
formalin, and water. Then the arterial system was
perfused with acetone to flush out the blood. The
arterial systems of these three tree shrews were
subsequently injected with red latex. Additional
tree shrews injected in this manner did not display
adequate penetration of latex into the arteries
of the cranial region. This was probably due to
contraction of vessels by embalming fluid or acetone.
Only the three animals described here were adequately
injected for observation.

The fourth tree shrew (number 4) was there-

fore injected with a more finely dispersed silicone

119

120
rubber compound called Microfil.l Another preliminary
procedure substituted in the case of tree shrew 4
included perfusion with Ringer's warm-blooded saline
solution containing heparin. Penetration of finer
vessels was greatly improved by use of these pro-
cedures.

Tree shrew 1 had been purchased as a dead
Specimen; thus cause and time of death were unknown.
Animals numbered 2, 3, and 4 were obtained in the
living state. Perfusion, for flushing out of
blood and for embalming, was carried out while the
animals were anaesthetized by ether. Injection of
the colored latex was carried out after intervals
of twelve to twenty-four hours. Microfil was in-
jected immediately after perfusion, within fifteen
minutes after death, in tree shrew 4.

Observations have been limited to the gross
anatomy of the carotid vessels. Due to the small
size of the animal, most dissection required the
use of micro-dissecting instruments and a Zeiss

binocular dissecting microsc0pe.

 

lMicrofil is the trademark of the Canton Bio-
Medical Products Company, Swarthmore, Pennsylvania.

DESCRIPTIVE ANATOMY

The branching pattern and distribution of
the carotid system will be described in the following
order: arch of aorta, common carotid artery,
external carotid artery, internal carotid artery.
Description of the stapedial branch of the internal
carotid artery precedes that of the arteries at
the base of the brain, which include the vertebro-

basilar and cerebral carotid groups.

The Arch of the Aorta

The mode of origin of the major branches
of the arch of the aorta shows minor variation in
the four tree shrews observed. In three out of
four individuals, there are two main trunks: (l)
a long truncus brachiocephalicus dexter and (2)
a very short (about one millimeter in length)
truncus brachiocephalicus sinister. In one case,
when the common areolar connective tissue sheath
around the left common carotid and left subclavian
arteries was cleared away, the left brachiocephalic
trunk was found to be only a fraction of a milli-
meter in length. In the fourth individual, three
main trunks arise from the arcus aortae:

121

122
(1) truncus brachiocephalicus, (2) arteria carotis
communis sinistra, and (3) arteria subclavia sinistra.
These arise in order from right to left, with the
second and third lying close together.

The (right) brachiocephalic trunk divides
into the arteria subclavia dextra and the arteria
carotis communis dextra. The division occurs at
the level of the sternoclavicular joint, at the
right lateral margin of the trachea, the brachio-

cephalic trunk having crossed the trachea obliquely.

The Common Carotid Artery

There is only one collateral branch of the
left common carotid artery: the arteria thyroidea
inferior, which is given off about five to six
millimeters above the origin of the left common
carotid artery, also receives a contribution from
the (right) brachiocephalic artery. The brachio-
cephalic contribution was not observed in two cases.

Each common carotid artery ends at a level
deep to, or just caudal to, the posterior belly
of the digastric muscle. The hypoglossal nerve
crosses just lateral to the carotid bifurcation. The
latter division results in formation of the arteria
carotis externa and arteria carotis interna. The
arteria carotis externa is detectably larger in

calibre than the arteria carotis interna in all

123
but one case: in tree shrew 2, they are approximately
equal. The difference is always slight. These two
diverging vessels continue their course, with the
vagus nerve, deep to the posterior belly of the

digastric muscle.

The External Carotid Artery

 

Although the external carotid artery is not
of primary interest here, certain general features
must be described to demonstrate its relations with
the internal carotid system.

Within one millimeter of its origin (except
on the right side of tree shrew 3), the arteria caro-
tis externa sends off the arteria occipitalis.
Arising within the V-shaped carotid bifurcation,
this branch parallels the course of the arteria
carotis interna to reach the muscles supplied. On
the right side in tree shrew 3, the occipital artery
originates, at a considerably higher level, from a
common trunk with the posterior auricular artery
and the major branch of supply to the parotid gland.
This common trunk lies inferolateral to the auricle,
deep to the parotid gland, and caudal to the masseter
muscle.

In all cases, the arteria thyroidea superior
is given off by the arteria carotis externa within one

to three millimeters beyond the carotid bifurcation.

124
The hypoglossal nerve crosses at, or just beyond,
this level. The superior thyroid artery curves
mediad and then runs rather straight caudally to
reach the thyroid gland and the larynx.

The next major division of the arteria
carotis externa (two to five millimeters farther
on), in three animals, is a bifurcation into a
more superficial branch that courses ventrolaterad as
the arteria lingualis and a deeper branch that
courses dorsolaterad as the continuation of the
arteria carotis externa. The latter is usually some-
what larger than the arteria lingualis; they may be
almost equal in caliber.

The arteria lingualis stem very quickly
sends off a large and rather superficial arteria
submandibularis to supply the submandibular and
sublingual glands. The main continuation of the
arteria lingualis remains at a deep level, coursing
toward the tongue.

Tree shrew 4, however, displays a variation
of this branching pattern. Here, one and one-half to
two millimeters beyond the origin of the SUperior
thyroid artery, the external carotid artery sends
a large, independent submandibular artery to the
submandibular and sublingual salivary glands. About

two millimeters further on, after a sharp dorsal turn,

125
the external carotid artery gives off a common
trunk (about equal in calibre to the external
carotid continuation) for the arteria lingualis
and arteria facialis. The common trunk courses
rostrally for about Six millimeters, sending off
muscular twigs to the masseter and digastric muscles
and some supplemental twigs to the submandibular
and sublingual salivary glands. The facial artery
appears to be the continuation of,the common trunk.

In tree shrews l, 2, and 3, however, the
arteria carotis externa continues for about three
to six millimeters beyond the exit of arteria
lingualis before sending off the rather small arteria
facialis deep to the posteroventralmargin of the
masseter muscle.

Whatever its mode of origin, the slender
facial artery continues along the typical rostral
course deep to the lower border of the masseter
muscle. En route, it sends off muscular branches
and then turns dorsally to cross the mandible near
the cephalic border of the masseter muscle. In tree
shrew 4, one could observe the very rich vascular
supply to the vibrissal area of the snout; however,
these branches appeared to arise mainly from the
infraorbital artery.

In all cases, the external carotid artery

takes a sharp dorsal turn after exit of the lingual

126
artery and here lies deep to the parotid salivary
gland. Caudal to the temporomandibular joint, it ends
by breaking Up into a series of branches. Whereas
the major one does continue the stem as the arteria
maxillaris, it is not at all evident that the arteria
temporalis SUperficialis is the other major terminal
branch. The superficial temporal artery is not a con-
spicuous vessel in this Species.

This area of terminal ramification of the
arteria carotis externa appears to be quite variable.
A common pattern shows the major branch of SUpply to
the parotid gland, the ramus parotideus, arising from
the arteria carotis externa very close to the arteria
auricularis posterior (in two animals bilaterally,
and in one animal unilaterally). There may be a
common stem for the ramus parotideus and the arteria
auricularis posterior (in one animal bilaterally,
in one animal unilaterally). As mentioned previously,
the right Side of tree Shrew 3 shows a common stem
for the ramus parotideus, arteria occipitalis, and
arteria auricularis posterior. The arteria temporalis
superficialis arises, typically, a Short distance
further along the main stem, which may now be

designated as the arteria maxillaris.

127
The External Carotid Artery (Maxillary Division)

The maxillary artery continues rostrad, ven-
tral to the pinna, and then markedly mediad as it
crosses the anterolateral wall of the tympanic bulla.
As it courses deep to the dorsal part of the masseter
muscle, it sends twigs to that muscle, to the pterygoid
muscles, and to the parotid gland. As the arteria
maxillaris reaches the line of junction between the
cranial wall and the dorsal bulla wall, deep to the
lateral pterygoid muscle, it gives off a slender
ramus anastomoticus cum arteria stapedia, which may
be regarded as the proximal portion of the ramus
inferior arteriae stapediae. (See Figure 1.) This
anastomotic branch, relative to thearteria maxillaris,
courses posterodorsally and passes through an opening
lying just anterodorsal to the inner end of the
external acoustic meatus to join the stem of the
arteria stapedia. The latter lies within the dura
mater, having just emerged from its canal in the
medial wall of the tympanic cavity. This is also
approximately the level at which the stapedial
artery (superior ramus) gives off a large meningeal
branch coursing dorsally along the line of the fissure
between cerebrum and cerebellum. On one side in
one individual there appear to be two diverging

branches of the common anastomotic trunk penetrating

.hhmpnm Hmpfioeguo .pm “humans HmpnOHm .om “hyopam Hmaflnoma .mm “mowhopnm
Hmomcwcme mapcfla .bm “museum Hmwvommpm mo mason noanmmdm .mm “unease
Aampmv uneven Hmwpommpm .vm “hnmunm Ahnoucosonmv vapopmo Hmhnmnoo

.mm “Aaopmv humane canopmo Hmcpopzw .mm “humans Hmaoobam nowanSm .Pm "humane Hmpwnnownmcw mo mossy
Hmpfinno .om “humans Hmpflnhomnmcfi .mw “amuse vwocmnmmwam Hmnpopxo on» gm:092v mcwmmmg humans humaafixma

.mp “humane hmaoobam Hoflnomcfi .bp "knoppm hmmaawxms mamcnmpawv .mp "humans Hmsmqwa .mp "nympho Hmufi
ngaooo .vF "Asmpmv humane caponmo Hmnhopxo .m— “hnouhm dwpohmo coaaoo .N— uo>pmc Hammoawomhn .F9 “was
lacuna .OF "Hamnmho .m no>nmn owpmo .m "maomss mzpooh HownoQSm .h uvaho mo Hams .m “psocm mo :HMm .m
.m “maasn oficmmahp mo Ham: Hmwvoa .N “azaaononoo .P "poms weaned: o» hem

Hmmwsflcma knommooom .mm “nymphs
Hmfipommpm mo mason noflnomcfl .mm “

"mnoa hpopommfio .v “asunmnoo
.coawom Hmacmho on» go 36H> Hmhopmq .H ohswam
3
3.3.3
96 .
mm a... .....mH
.. a. 2.3
. 00000.”
a mm
cocoomm
DQOOOON
...l~mw

 

  

O
9‘ t

129
the skull wall separately. Only one of these branches
joins the stapedial artery as the latter emerges from
the middle ear region. On one side in another
individual the anastomotic branch joins the
stapedial artery farther posteriorly by piercing the
anterior bulla wall. Near the maxillary end, this
anastomotic branch has collateral twigs SUpplying the
bulla wall itself. The connection with the maxillary
artery was not observable in this case nor on one
side of another individual. The connection may or
may not have been present in these two instances.
Connection with the arteria stapedia was demonstrable
in all cases. Collateral branches to the bulla wall
were commonly seen. .

In the region medial to the mandible and
rostral to the temporomandibular joint, the arteria
maxillaris gives off the large arteria alveolaris
inferior that enters the mandibular foramen almost
immediately, lying directly ventral to the inferior
alveolar nerve.

As the arteria maxillaris itself continues
its rostromedial course, it sends off a series of
branches to pterygoid and temporal muscles. In
one case there is a major muscle branch with several
collateral twigs.

As the arteria maxillaris approaches the

130
skull base, it enters the external aliSphenoid canal
in the small, laterally projecting external ali-
sphenoid process. (See Figure 1.) After its brief
passage through this canal (no branch being
detectible within the canal), the arteria maxillaris
again appears briefly in the infratemporal region
before entering the orbit to cross its floor,
accompanying, and lying immediately lateral to, the

large maxillary division of the trigeminal nerve.

The External Carotid Artery_(Infraorbital Division)

From the afore-mentioned juncture, the
arteria maxillaris may be called the arteria
infraorbitalis. This portion sends branches to
the palate. The artery also sends off a major
branch, the arteria alveolaris superior, which
accompanies a comparable nerve branch.

A large medial branch of the arteria infra-
orbitalis extends across the orbital floor to the
ethmoidal foramen in the ventromedial orbital wall.
Before entering the cranial cavity (in the region of
the olfactory lobe), this anastomotic ramus orbitalis
of arteria infraorbitalis joins the ethmoidal
branch of the stapedial artery (superior ramus), to
be described later. This anastomosis was observed
only in tree shrews 3 and 4, bilaterally. (Unfor—

tunately, it could not be demonstrated with certainty

131

in tree shrews l and 2, due to the earlier method
of dissection used. Here it was largely or completely
destroyed, presumably.) Ramus orbitalis, in tree
shrew 4, could be seen to send off a few small branches
to inferior ocular muscles.

The infraorbital artery leaves the orbit,
by way of the infraorbital canal and foramen, to
enter subcutaneous areas of the snout region.

The Internal Carotid Artery
(Stem and Major Branches)

The arteria carotis interna ascends obliquely
cephalodorsad from its origin deep to, or below,
the caudal border of the posterior belly of the
digastric muscle. After a rather straight, un-
branched course to the posteromedial bulla wall,
the artery passes through the carotid foramen to
enter a thin-walled (translucent) bony canal in the
medial wall of the tympanic cavity. For the remainder
of its course through the tympanic cavity, the
arteria carotis interna and its branch, the arteria
stapedia, remain within such bony canals. The
bifurcation occurs about four to five millimeters
after entrance to the bulla. The two arteries
Show no appreciable Size difference; if one is
slightly larger, it is the arteria stapedia. The

continuation of the arteria carotis interna (arteria

132

promontorii) passes rostrally and medially to
cross the convexity of the promontorium and enter
the cranial cavity by passing through the foramen
lacerum medium, located where aliSphenoid and
basiSphenoid join the-petrosal portion of the
temporal bone. The arteria promontorii is not at
all "reduced" or attenuated.

The further course of the arteria carotis
interna (arteria promontorii), as the arteria
carotis cerebri, will be described with the circulus
arteriosus cerebri and associated arteries at the

base of the brain.

The Stapedial Artery (Tympanic Portion)

As the arteria stapedia courses rostrodor-
sally, it passes between the crura of the stapes
and thence continues its course within its bony
canal in the incomplete septum separating the
epitympanic sinus from the loculus. It enters the
cranial cavity at the anterior margin of the os
petrosum. (See Figure l.) The cranial portion of
the arteria stapedia is entirely intradural in
position and in distribution. Immediately upon enter-
ing the cranial cavity, the arteria stapedia sends
off the arteria meningea media that extends dorsally
along the groove between the flocculus and the

temporal lobe of the brain, continuing on toward the

133

longitudinal fissure.

The Stapedial Artery-~Inferior Ramus

At about the same level, the ramus inferior
(anastomotic branch to the maxillary artery) arises
from the stem of the stapedial artery. This branch
extends through the cranial wall and courses rostro-
ventrally along the anterior bulla wall. ,Although
this vessel is constant in all individuals observed,
it is very slender, being one-third or less of the
size of the stapedial stem. It sends collateral

twigs to the bulla wall.

The Stapedial.Arteryr-Superior Ramus

(Cranial Portign)

The continuation of the main stem or ramus
superior of the arteria stapedia courses rostrally
and somewhat ventrally, across the temporal lobe,
within the dura mater. In tree shrew 1, this portion
sends Off one dorsally directed branch, possibly
comparable with a ramus meningeus accessorius.
Tree Shrew 2 Shows three additional, smaller, ascend-
ing dural branches of this intracranial portion of
arteria stapedia (ramus superior).
The Stapedial.Artery--SUperior Ramus
{Orbital Portion)

On arrival in the orbit, the ramus superior

arteriae stapediae gives Off a slender arteria

134
lacrimalis that courses along the upper border Of
the posterior rectus muscle to reach the lacrimal
gland. (See Figure 1.) About three millimeters
deep to the conjunctival fornix, the lacrimal
artery bifurcates, sending a branch to each of the
separate glandular lobes.

At the same point or within one to two
millimeters, the main stem Of the ramus SUperior
arteriae stapediae sends Off a more substantial
branch, the arteria frontalis, that remains close
to the orbital roof, above the SUperior rectus
muscle. (See Figure 1.) On reaching the dorsolateral
region Of the orbit, the arteria frontalis joins
a branch Of the maxillary division Of the trigeminal
nerve. The vessel and nerve continue through the
supraorbital foramen to the subcutaneous tissue of
the dorsolateral snout region. NO collateral branches
Of the frontal artery could be observed within the
orbit.

The main trunk (larger in calibre than the
frontal artery) of the stapedial artery continues
its course anteriorly between the superior rectus
muscle and the retractor bulbi muscle that thinly
covers the Optic nerve. The vessel crosses this
areolar Space at a right angle with the superior

rectus fibers. (See Figure l.) Branches to extrinsic

135
eye muscles arise from this stapedial trunk and
certain Of its larger branches.

As the stapedial stem nears the rostral
end Of the zone between the superior rectus muscle
and the retractor bulbi muscle, the overall rostro-
medial course is continued by the slender arteria
ethmoidalis that reaches the dura mater covering
the Olfactory lobes, by way of the ethmoidal foramen
in the anteromedial orbital wall. (No anastomoses
with branches Of the internal carotid artery
supplying the substance of the lobes themselves
could be detected in this area.)

The ethmoidal artery could be seen (through
the translucent orbito-cranial wall) to pass dorsally
along the cephalic border of the Olfactory lobe.
Dissection revealed that, on reaching the dorsal
'surface of this lobe (within the covering dura
mater), it crosses the lobe anteromedially. About
one millimeter lateral to the medial border Of the
lobe, this dural vessel makes a sharp turn forward
through the cribriform plate to enter the region
of the nasal mucosa, which, in an excellent injection
(tree shrew 4), is observed to contain a highly
complex Olfactory or ethmoidal rete of fine arteries.

It will be recalled that the infraorbital

artery sends an anastomotic branch, the ramus

136
orbitalis, to join this ethmoidal artery as it
enters the ethmoidal foramen.

In caliber, the real continuation Of the
stapedial artery appears to be the division that
bends sharply downward and laterally toward the
bulbus oculi, close against the anterior aSpect of
the Optic nerve and deep to the anterior rectus
muscle. Having given Off muscular branches and
additional twigs to the Optic nerve sheath and
sclera, this trunk terminates in branches Of supply
to the inferior and posterior rectus muscles. In
addition to these small twigs, the arteria stapedia
also sends off some larger collateral branches to
be described now.

While still lying deep to the SUperior
rectus muscle, the stapedial artery gives off a
large branch coursing distally, more or less
parallel with the parent trunk. This branch
Spends itself in sending Off a long series of
collateral branches to Harder's gland, as it con-
tinues deep to that gland along its considerable
length. Before this branch reaches Harder's gland,
however, it sends Off a large collateral branch
at a right angle to the trunk. This collateral
branch courses ventral to the Optic nerve, just

deep to the inferior rectus muscle, supplying extrinsic

137

muscles Of this inferior region.

At the anterior border of the SUperior
rectus muscle, the stapedial artery sends Off a
very Short common trunk from its lateral surface;
the common trunk, in turn, gives rise to four branches.
These supply extrinsic muscles, including SUperior
rectus, superior Oblique, anterior rectus, and
retractor bulbi muscle slips.

Another series of collateral branches of
the stapedial trunk constitute the arteriae ciliares.
These Spring from a collateral branch that serves
as their common trunk Of origin. These branches
tend to be prolonged downward along the optic
nerve and onto the bulb in a brush-like formation.
0n reaching the bulb, further branching of this more
or less parallel type occurs; thus there are (in
tree shrew 4) three major branches and three to
four lesser branches of this ciliary artery group.
Many of them course within the sheath of the bulbus
oculi to reach the region just deep, and proximal,
to the insertion of the retractor bulbi muscle slips.
Although some finer branches penetrate the bulb en
rggtg, the major branches penetrate the sclera in
the terminal region indicated above. A large proximal
branch Of the ciliary common trunk crosses the ventral

surface of the optic nerve obliquely, to reach the

138

ventrocaudal and dorsocaudal aspects of the nerve
and bulb.

The arteria ophthalmica (branching from
the intracranial arteria carotis interna) anastomoses
with.the branch of the arteria stapedia that serves
as the common trunk for the ciliary branches; thus
the exceedingly small Ophthalmic artery is functional-
ly in a position to SUpply blood only to the ciliary
arteries. (See Figure 2.) It supplies no other
structures within the orbit. The arteria
Ophthalmica was very difficult to locate and trace
in all individuals except tree shrew 4. In tree
shrew 3, it could not be found on the right side.
In tree shrew 1, it could not be traced beyond the
Optic nerve. Failure to be injected with latex
could make it all but impossible to find, even with
a binocular microsc0pe. 'Normally, its path may be
traced across the ventrolateral surface of the Optic
chiasma and thence obliquely across the ventral
surface of Optic nerve to the ciliary trunk lying

at the anterior margin of that nerve.

The Arteries at the Base of the Brain
In describing the arteries at the base of
the brain, the posterior vertebro-basilar group
will be taken up first, to be followed by the internal

carotid group at more anterior levels. The two groups

  

(Dura mater is intact
except to reveal the
anterior rami of the
cerebral carotid
arteries.)

Figure 2. Ventral View of the Brain and Middle Ear.

Key to numbers used: 1, medulla Oblongata; 2, cerebellum;
3, cerebrum; 4, Olfactory lobe; 5, dorsomedial wall of tym-
panic bulla (remnant); 6, trigeminal nerve; 7, optic nerve:
8, optic chiasma; 9, internal carotid artery (stem); 10,
superior ramus of stapedial artery: 11, cerebral carotid
(promontory) artery; 12. anterior ramus of cerebral carotid
artery; 13, ophthalmic artery; 14, ethmoidal artery; 15,
ciliary artery (stem).

140
are joined at the transition from rhombencephalon

to mesencephalon levels.

The Vertebral Artery

 

The arteriae vertebrales curve around the
lateral surfaces of the caudal medulla Oblongata,
extending obliquely cephalad as they pass from the
dorsolateral to the mid-ventral surface.

Each arteria vertebralis, as it extends
obliquely cephalad toward the arteria basilaris, gives
off three branches from its lateral aspect: ‘these
supply the lower, lateral medulla oblongata.
Anastomotic interconnections occur among their
smaller branches as they continue on to SUpply the
dorsal aSpect of the medulla. }

All medially directed branches Of the verte-
bral arteries anastomose freely with one another

on the ventrolateral surface of the medulla oblongata.

The Basilar Artery

The median arteria basilaris forms by
confluence of the two arteriae vertebrales at the
posterior border of the pons. It sends Off five
pairs of rami ad pontem as it courses cephalad
across the pons. The most anterior pair arises
immediately caudal to the bifurcation of the basilar

artery.

141

The second pair of branches emanating from
the arteria basilaris constitutes the larger arteriae
cerebelli inferiores anteriores. The collateral
branches supply not only the ventrolateral pons
but much of the flocculus as well. As each inferior
anterior cerebellar artery crosses the pons, it
sends small branches to the base of the trigeminal
nerve trunk. The abducens nerve arises just cephalic
to the level of the inferior anterior cerebellar
artery on each side.

In a well-injected individual (tree shrew 4),
one branch Of each inferior anterior cerebellar
artery can be Observed to supply the inner ear
region as the arteria labyrinthi. VAlong the ventral
margin Of the flocculus, a slender branch anastomoses
with a similar branch coursing posteriorly from the
SUperior cerebellar artery. This long anastomotic
channel sends off small twigs to both pons and
flocculus. Some small branches continue laterally
from the posterior choroid plexus area to supply
the caudolateral cerebellar hemisphere and deep caudal
aSpect of the vermis.

All branches of the basilar artery display
anastomotic interconnections on the ventrolateral

surface of the pons.

142
The Basilar Artery (Bifurcation)

The bifurcation Of the arteria basilaris
occurs at the rostral border Of the pons. (See
Figure 3.) In three of the four tree shrews
Observed, the bifurcation is quite symmetrical in
arrangement of branches. In these cases there is a
short truncus communis (one-half to two-thirds of
a millimeter in length) on each Side. From this
trunk the superior cerebellar and mesencephalic
arteries diverge laterally, while the main trunk
continues to curve rostrally as the posterior
communicating artery on each side. In tree shrew 3,
such a common trunk is observable on the right side,
but not on the left. On the left Side, the superior
cerebellar artery seems to be the terminal branch
of the basilar artery, arising at a right angle to
the parent vessel. Then, the somewhat reduced
bifurcation product appears to be the proximal
portion of the mesencephalic artery, as the latter
curves anterolaterally, just rostral to the oculo-
motor nerve origin. The posterior communicating artery
continues the cephalic curve, so that the posterior
cerebral artery emerges at a right angle.

The bifurcated arteria basilaris (proximal
arteria mesencephali on left side Of tree shrew 3)

sends Off several, slender, parallel rami centrales

 

c...
o
.........
o...

   

(Dura mater
has been removed.)

Figure 3. Ventral View of the Brain (Tree Shrew 3).

Key to numbers used: 1, spinal cord; 2, medulla Oblongata;
3, pons; 1+, cerebrum; 5, hypophysis cerebri; 6, vertebral
artery; 7. ventral (anterior) spinal artery; 8, basilar ar-
tery; 9, anterior inferior cerebellar artery; 10, pontine
arteries: 11, common trunk for right superior cerebellar and
mesencephalic arteries; 12, superior cerebellar artery: 13,
mesencephalic artery: 14, posterior communicating artery;
15, posterior cerebral artery; 16, cerebral carotid (promon~
tony) artery (cut Off): 17, anterior ramus of cerebral car-
otid artery; 18, a branch of the middle cerebral artery.

144
entering the posterior perforated substance.
Where a common trunk results from basilar
bifurcation, it sends off a large arteria
cerebelli superior that is not quite as large as

the main continuation of the trunk.

The Superior Cerebellar Artery

The SUperior cerebellar artery courses
laterally along the upper border of the pons to the
cephalic aspect Of the flocculus, sending collateral
branches to the pons, the posteriOr colliculus, and
deep parts of the flocculus. There are three major
terminal branches of the superior cerebellar artery.
The first of these diverges caudally along the dorsal
surface of the flocculus near the junction between
the latter and the cerebellar hemisphere, SUpplying
twigs to both and communicating with small branches
of the two inferior cerebellar arteries. This branch
continues over to the posterior deep aSpect of the
cerebellar hemisphere, crossing the latter to
terminate in branches supplying the posterior deep
region of the vermis. This long trunk also sends
off collateral branches to the choroid plexus of
the fourth ventricle.

The other major terminal branches of the
superior cerebellar artery SUpplY the rostral

deep surfaces of the cerebellar hemisphere and the

145
vermis. One branch follows a course deep to the
transverse venous sinus. The other major branch
passes posteriorly between the cerebellar hemisphere

and vermis, on the deep surface.

The Mesencephalic Artery

 

The main continuation of the bifurcated
basilar artery appears to be the proximal arteria
mesencephali (or the distal arteria communicans
posterior) connecting with either the superior
cerebellar artery or their shared common trunk.

This vessel continues the circulus arteriosus

cerebri for another one and one-half to one and
three-fourths millimeters, on each side, before the
distal mesencephalic artery extends laterad across
the ventrolateral midbrain cephalic to the attachment
of nervus oculomotorius. This proximal mesencephalic
artery (distal posterior communicating artery)

sends small twigs to the midbrain and hypothalamus.

The mesencephalic arterial trunk extends
laterad one to two millimeters before sending'off
two to four major branches deep to the overhanging
temporal lobe. These branches tend to course in
parallel ("penicillus") formation as they ramify
further. These continue across the lateral, and
then the dorsal, aSpect of the midbrain at the level

of the anterior colliculus, which is the major area

146
supplied. The most rostral branch of the mesen-
cephalic artery generally follows the groove along
the cephalic margin of the anterior colliculus
to terminate in the dorsal thalamus region. As
this rostral division approaches the dorsal
midline, it diverges dorsally to SUpply the Splenium
of the corpus callosum. It anastomoses with the
more caudal divisions of the mesencephalic artery.
It also apparently sends some terminal twigs
(arteria choroidea posterior medialis) to the
choroid plexus of the third ventricle.

The lower or more caudal divisions Of the
arteria mesencephali course more directly laterad,
obliquely crossing, but not following, the groove
between the anterior and posterior colliculi. The
main continuation extends obliquely across the
anterior colliculus, giving numerous branches to it.
A few branches extend as far caudad as the posterior
colliculus. This major caudal division forms several
anastomoses with branches of the posterior cerebral

artery.

The Posterior Communicating Artery

 

The apparent continuation of the bifurcated
basilar artery cephalic to the level of the mesen-
cephalic branch may apprOpriately be called the

arteria communicans posterior. It continues

147

rostrally, lateral to the hypothalamus and deep to
the overhanging medial border Of the temporal lobe.
The Posterior Communicating Artery--
Posterior Cerebral Branch

The arteria cerebri posterior arises one-
half to three-fourths of a millimeter cephalic to
the arteria mesencephali. Proximally, the posterior
cerebral artery sends twigs to the cerebral peduncle.
As the posterior cerebral artery courses laterad,
it lies parallel with the mesencephalic artery for
about two millimeters. Having crossed the cerebral
peduncle, the artery diverges cephalolaterally to
cross the anterior colliculus and optic tract. Before
reaching the Optic tract level, the posterior cerebral
artery sends Off three (two Often arising from a
common stem) branches that course along the deep sur-
face of the posterior cerebral hemiSphere, curve
around the caudal margin, and then ramify over the
posterior occipital and temporal regions. En route,
these branches anastomose with mesencephalic

branches (on the deep aspect of the hemisphere).

The Posterior Communicating Artery--
Anterior Choroidal Branch

 

In tree shrew 4, what appears to be the
arteria choroidea anterior arises from the arteria

communicans posterior, about one-fourth of a millimeter

148
rostral to the origin of the arteria cerebri
posterior. It courses laterally, deep to the
overhanging temporal lobe, and gives off numerous
long branches to the posteromedial hemiSphere
region and posterodorsal thalamus. Many of these
branches apparently supply the choroid plexus of
the lateral ventricle. (In tree shrews 1, 2,
and 3, a comparable vessel arises more rostrally,
above the level of the arteria ophthalmica.)
The Circle of Willis--
Cerebral Carotid Entrance

The arteria communicans posterior continues
its rostral course for two to three millimeters
before receiving the arteria carotis interna, of
which it is the proximal posterior ramus. En route,
it sends more branches to the hypothalamic region and
to the cerebral peduncle and optic tract. The
arteria carotis interna enters the circulus
arteriosus cerebri just caudal to the Optic tract,
as the latter is prolonged caudolaterad from the
Optic chiasma.

It should be noted, too, that the portion of
the arteria communicans posterior between the level
of entry of the arteria carotis interna and the exit
of the posterior cerebral and mesencephalic arteries

is somewhat attenuated in comparison with the

149

bifurcated basilar portion of the circulus arteriosus
cerebri more caudally and in comparison with the ramus
anterior of the arteria carotis interna more
cephalically. These more caudal and cephalic por-
tions are about equal in calibre. If a difference
can be detected, the caudal portion appears to be
slightly larger.
The Cerebral Carotid Artery--
Anterior Ramus

The anterior ramus of the internal carotid
artery continues cephalad along the groove between
the optic chiasma and the temporal lobe. It continues

to send off fine branches to these adjacent structures.

The Ophthalmic Artery

The first longer branch arising from the
medial aspect of the internal carotid artery (about
one and one-half millimeters rostral to the internal
carotid entrance) is the extremely reduced arteria
Ophthalmica. The latter arises ventral to the
Optic chiasma, continuing obliquely across the ventral
surface of the chiasma and Optic nerve to the point
of anastomosis with the ciliary trunk of the stapedial
artery (SUperior ramus) on the rostroventral aspect

of the nerve near the bulbus oculi. (See Figure 2.)

150
The Anterior Choroid Artery

In tree shrews l, 2, and 3, the most prom-
inent branch of the internal carotid artery (anterior
ramus), extending laterally across the optic
tract, may be regarded as an anterior choroid
artery; a Sizeable collateral branch continues over
to the choroid plexus of the lateral ventricle.

The main continuation of this artery ends in the
region Of the geniculate bodies and the pulvinar
of the thalamus.

After coursing rostrally another one-half
millimeter, the anterior ramus of the internal or
cerebral carotid artery sends off (laterally)

a fairly small, but constant, branch supplying the
temporal pole area (pyriform lobe) of the cerebral

hemisphere.

The Middle Cerebral Artery

Immediately cephalic to this (within one
millimeter), the anterior ramus of the internal
carotid artery sends off the arteria cerebri media.
The latter is almost as large as the main continua-
tion, which is designated as the arteria cerebri
anterior.

Coursing craniolaterally, the middle cere-
bral artery enters the Sylvian fossa of the cerebral

hemisphere. Collateral branches supply the entire

151
fossa area. The main stem continues around the
lateral margin onto the superolateral surface of
the hemisphere, where the main stem continues
mediocaudally. Its collateral branches supply
the upper temporal and much of the parietal areas,
to within three millimeters of the longitudinal
fissure. The remaining medial strip is supplied
by branches of the anterior cerebral artery. The
branches supplying the superolateral portions of
the frontal lobe originate from the part of the
middle cerebral artery lying within the Sylvian
fossa; thus these branches must curve around the
lateral border of the hemisphere individually. They
supply all but the most medial two millimeters Of
the dorsal area of the frontal lobe. The remaining
medial strip is, again, supplied by branches of
the anterior cerebral artery.

Some rostrally directed branches of the
middle cerebral artery (within the Sylvian fossa
region) supply the proximal portions of the olfactory'
tract.

The middle cerebral artery sends off a pair
of branches (more or less at right angles) to course
rostrally and caudally in the rhinal sulcus. This
sulcus continues well back into the temporal lobe

region.

 

 

.
\
o
1 I
I
m, r
r t

152
The Anterior Cerebral Artery

The anterior cerebral arteries and the
anterior communicating artery send out small twigs
of supply to the Optic chiasma.

Immediately cephalic to the optic chiasma,
the arteria communicans anterior transversely
interconnects the two arteriae cerebri anteriores.
The right half may be somewhat larger in caliber
than the left, or the reverse. The arteria communi-
cans anterior gives off, at right angles to
itself, the single, median arteria cerebri anterior
communis (arteria corporis callosi) that courses
dorsally over the anterior surface of the rostrum
of the corpus callosum. On reaching the dorsal
surface of the corpus callosum, this common anterior
cerebral artery bends posteriorly to course along
the upper surface of the corpus callosum. The
collateral branches given off en route supply the
medial surfaces of the frontal lobe areas. (No
branches extend rostrally to SUpply the olfactory
bulbs.)

About one-third of the distance back along
the upper surface of the corpus callosum, the common
anterior cerebral artery begins a dorsocaudal ascent,
away from the corpus callosum and toward the dorsal

surface of the cerebrum, within the longitudinal

 

153
fissure. En route, it gives Off additional branches
to the medial surfaces of the hemispheres. The
major branches curve around the dorsomedial margin
of the cerebral hemisphere to supply the medial two
to three millimeters of the dorsolateral surfaces.

As the common anterior cerebral artery nears
the dorsal surface (while still within the longitu-
dinal fisSure), it bifurcates into two equal-sized
branches (Often referred to as arteriae corporis
callosi) about half-way back along the length of the
corpus callosum. Each branch ascends over the dorso-
medial border of the hemiSphere to supply the medial
three millimeters of the dorsum of each hemisphere.
(This distribution is apparent in latex-injected
Specimens. Anastomoses among the branches of the
anterior, middle, and posterior cerebral arteries
occur on the dorsolateral surfaces of the cerebral
hemispheres. In tree shrew 4, with the more pervasive
Microfil injection, the three areas of distribution
are not readily distinguishable because of the high
degree of overlap in areas of branching.)

The apparent continuation of the anterior
cerebral arteries, rostral to the level of the
anterior communicating artery, sends branches of
supply only to the olfactory tract and bulb; thus

they may be called arteriae bulbi Olfactorii mediales.

154

Shortly after exit of the arteria cerebri
media, each arteria cerebri anterior gives off a
ramus ethmoidalis or arteria bulbi olfactorius
lateralis that courses rostrad to reach the lamina
cribrosa, where it connects with an intradural
rete ethmoidale. The ramus orbitalis from the
arteria infraorbitalis and the ramus ethmoidalis
from the arteria stapedia, joining within the orbit,
form a common ethmoidal branch that also connects
with this rete ethmoidale. From the rete, in turn,
small branches extend into the nasal cavity with
the fila olfactoria; thus blood from the arteria
carotis cerebri may reach the nasal mucosa, with
its exceedingly rich plexus of vesSels.

The arteria bulbi olfactorius lateralis
also sends branches to the lateral aspects of the
olfactory tract and lobe. Its terminal branches
anastomose with those of the arteria bulbi
olfactorius medialis and with cephalic branches of

the arteria cerebralis media.

DISCUSSION

There is no characteristic arrangement of
cranial arteries for a given order of mammals:
certain types are repeated in several orders; animals
in closely related groups frequently present very
different conditions. Thus, one cannot trace
alterations through a serial progression of phylo-
genetic stages (Tandler, 1899; Hofmann, 1900;
de Vriese, 1905; Stephan, 1954; Guthrie, 1963).~

There is, nevertheless, in the class Mammalia,
a basic cranial arterial pattern, in terms of which
variant types may be interpreted. Retention Of
one portion or obliteration of another portion of
the vascular system in different animal grOUps
cannot yet be explained. Studies of what may be
called the "ecology“ of head structures are required.
When all the facets of the dynamic processes of
development of this head complex are clarified,
laws of head structure will emerge. Thus, effort
expended in gathering and correlating comparative data
will be repaid.

Discussion of the tupaiid arteries, from a

phylogenetic point of view, here follows in the same

155

156
sequence as in the preceding section on descriptive

anatomy.

The Arch of the Aorta

The pattern of branching of the aortic arch
in Tupaia glis--with two brachiocephalic trunks
or with left common carotid and left subclavian
trunks arising separately from the aorta--is a
rather generalized, and possibly close-to-primitive,
condition among mammals.

Observations of other anatomists on Tupaiidae
are noted here. In the two Specimens of Ptilocercus
lgwii examined by Le Gros Clark (1926a), there was
a (right) brachiocephalic artery (about six milli-
meters in length) and a very short (one-half milli—
meter to two and one—half millimeters in length) left
brachiocephalic trunk for the left common carotid
and left subcalvian arteries. Le Gros Clark (1926a)
stated that the occurrence of a short brachiocephalic
trunk on the left Side is usual in Tugaia.
Recognizing variation, he observed (1926a) one

Specimen of Tupaia minor in which the left common

 

carotid and left subclavian arteries had separate
origins from the aortic arch.
Platzer (1960) cited Hill and Davies as

having observed, in Tupaia glis (number of Specimens

 

not given), a pattern that represents a further step

157

in the concentration of branches in the (right)
brachiocephalic trunk. Here, on one side, there
is a common trunk for the (right) brachiocephalic
and the left common carotid arteries. On the
other side is found the left subclavian artery.
This would resemble the most common human variant.

Among Marsupialia, examples of the greatest
separation of trunks are found correlated with
symmetry of spacing. Phascolarctos (marSUpial

bear) may have four separate trunks; Thylacinus

 

(Tasmanian wolf) may Show two subclavian trunks

and an intervening bicarotid trunk (Parsons, 1902).
Considerable symmetry is also displayed

among Insectivora and Chiroptera, the most primi-

tive eutherian orders (Parsons, 1902). A left

brachiocephalic artery (of varying length) has been

commonly observed among bats (VeSperugo, Plecotus,

 

Pteropus, Cynonycteris) (Grosser, 1901; Parsons,

 

 

1902). In other bats (Rhinolophus, VeSpertilio),

 

 

the left common carotid stem arises separately, but
closer to the left subclavian than to the (right)
brachiocephalic artery (Grosser, 1901; Arnbgck-
Christie-Linde, 1907). A symmetrical arrangement
with two brachiocephalic trunks is also the more

common arrangement among Insectivora, being usual in

 

Talpa (mole) and very frequent in Chrysochloris

158
(golden mole) and Erinaceus (hedgehog). The common
variant in the latter two genera is that in which
the left common carotid lies closer to the left
subclavian than to the (right) brachiocephalic
trunk (Dobson, 1882-1890; Parsons, 1902). This
latter situation is regularly found in other insec-

tivores, such as Centetes, Myogale, Potamogale,

 

Rhynchocyon, and Soricidae (Dobson, 1882-1890; .
Parsons, 1902; Arnback-Christie-Linde, 1907). Other
primitive mammals in which the latter arrangement

is common or the rule would include armadillo,
three-toed sloth, and aard-vark (Parsons, 1902).

As a variant in some of these forms, the left common
carotid artery may lie closer to the brachiocephalic
than to the left subclavian artery. Monotremata
commonly are included in the latter group (Parsons,
1902; Arnbgck-Christie-Linde, 1907).

The possible sequence of patterns is suggested
by Lewis's (1923) study of Sheep embryos. He found
that aortic branches pass from an early symmetrical
condition, through asymmetrical phases, to reach the
adult pattern of maximum concentration of branches
typical for ungulates.

Based on generality of presence in most
mammalian orders (as rule or as variant) one might,

however, tend to regard the asymmetrical, three-trunk

159
pattern as more primitive than two-trunk or four-
trunk patterns (de Garis, 1941).

Among higher Primates, the situation closest
to the above-described "primitive" patterns would be
that with three trunks, the left common carotid
artery lying near the brachiocephalic trunk. This

is the predominant arrangement in Homo, Gorilla, Pan,

 

 

Cebu; (capuchin), Callithrix (marmoset), Ateles
(Spider monkey), and Saimiri (squirrel monkey)
(Owen, 1866; Sonntag, 1923, 1924; Beattie, 1927;
Glidden and de Garis, 1936; Hill and Davies, 1959;
Platzer, 1960).

Most prosimians appear to be more Specialized.
The only prosimians regularly diSplaying the three-
trunk arrangement are Tarsius (Woollard, 1925;
Hill, 1953a; Platzer, 1960) and Lgris (Chapman,
1900; Rau and Rao, 1930; Davies, 1947; Hill and
Davies, 1959). This is a common variant in large'
mammalian orders, such as Carnivora and Rodentia
(Parsons, 1902).

The general arrangement in Lemuriformes is
the two-trunk pattern, in which the (right)
brachiocephalic and left subclavian trunks arise
from the aortic arch (Chapman, 1900; Davies, 1947;
Hill, 1953b; Hill and Davies, 1959; Platzer, 1960).

This two-trunk arrangement is also the usual one for

160

a number of higher Primates, such as Hylobates

 

(gibbon), Macaca, Alouatta (howler monkey),
Mandrillus, Pongo (orang-utan) (Chapman, 1900;
Sonntag, 1922; Lineback, 1933; Platzer, 1960).

Primates illustrate well, within the range
of their variation, the seemingly general mammalian
trend toward concentration of trunks on the right
by fusion with the (right) brachiocephalic trunk.
This reaches its peak in the pattern (uncommon

anomaly in Lemur, usual situation in Propithecus)

 

showing a Single (right) brachiocephalic trunk
arising from the aorta (Hill and Davies, 1959).
Although concentration of branches is
highly correlated with narrow thoracic outlet and
increasing distance of the aortic arch from that
outlet (Parsons, 1902; Sawin and Edmonds, 1949), it
does not seem indifferent that the Shift has been to
the right--as for veins--in the heart region. There
may.be correlation between heart orientation within
the thorax and the asymmetrical concentration of
trunks on the right. Man is closer to the primitive
primate condition (neotenic?) than many lemurs
known to be Specialized (as primates) in other ways.
With respect to branching of the adult aortic

arch, Tupaia glis is like many Insectivora, but most

 

closely resembling those showing the trend (primate

161
and general mammalian) toward asymmetry with concen-

tration of branches on the right.

The Common Carotid Artery (Bifurcation)

The level of bifurcation of the common
carotid artery is of no Special phylogenetic signifi-
cance. In most primitive mammals, the division
into external and internal carotid arteries occurs
deep to the musculus digastricus or just caudal to
this. This applies to TUpaiidae, Prosimii, Insectivora
(except Talpa, with division occurring deep down in
the neck), ChirOptera, Rodentia, and Marsupialia. In
Monotremata and Edentata, the bifurcation tends to be
somewhat lower, lying opposite the hyoid bone or the
larynx (Tandler, 1899, 1901; Grosser, 1901; Sicher,
1912; Hafferl, 1919; Woollard, 1925; Le Gros Clark,
1926a; Davies, 1947; Hill, 1953a, 1953b; Platzer,
1960).

The Externalyand Internal Carotid Stems
(Relative Size)

The relative sizes of the external and in-
ternal carotid arterial stems may be of some phylo-
genetic significance when more facts are known.

In Tupaiidae the external and internal carotid
arteries are approximately equal at their origin.
According to Hill (1953b), the external carotid is

by far the larger of the two arteries in Strepsirhini.

162

Davies (1947) confirmed this for Nycticebus.

 

Tandler (1899) found the two vessels to be about

equal in size for Loris (Stenops) gracilis and

 

Galago (Otolicnus) senegalensis. In Daubentonia

 

(Chiromys) madagascariensis and Lemur varius there

 

is a rudimentary internal carotid artery and a nor-
mal external carotid artery (Tandler, 1899; Hill,
1953b). In Tarsius, however, the external carotid,
as such, is much smaller than the internal carotid
artery (Platzer, 1960). This is misleading since
a number of typically external carotid branches
arise from a separate common trunk at the bifurca-
tion point (Hafferl, 1919; Woollard, 1925; Hill,
1953a).

There is also variability in this feature

among Insectivora. Whereas in Erinaceus europaeus

 

the two arteries are almost equal in size, in Talpa
europaea the internal carotid is markedly larger
than the external carotid artery (Tandler, 1899).

In Pteropus (ChirOptera), the two vessels
are approximately equal in size (Grosser, 1901), as
also in Ornithorhynchus (Monotremata) and Edentata
(Tandler, 1899). The external is larger than the
internal carotid artery in Echidna (Monotremata)
and in Marsupialia (Tandler, 1899; 1901).

The variation in size of external and internal

163
carotid stems depends upon two principal factors:
(1) the relative share of internal carotid and
vertebral arteries in supplying the circulus
arteriosus cerebri and (2) the degree Of (adult)
develOpment of the stapedial branch of the
internal carotid artery. In lizards, possessing
the primitively limited distribution Of the external
carotid artery and the rudimentary intracranial
development of vertebral arteries, the internal
carotid artery is at least twice the size Of the

external carotid artery (Bhatia and Dayal, 1933).

The External Carotid Artery (Relations
with the Stapedial Branch of the
Internal Carotid Artery)

The external carotid artery of lower ver-
tebrates is weak: its distribution area is limited,
in amphibians and lizards, to the hyoid and tongue
musculature, some muscles of mastication and the
pharynx wall, and subcutaneous connective tissue of
the lower lip and region medial to the mandible
(Bhatia and Dayal, 1933; Stephan, 1954). Thus, in the
primitive tetrapod condition, the arteria carotis
externa is essentially an arteria lingualis. It has
been noted that the lingual artery differentiates
very early in the mammalian embryo (Grosser, 1901;
Tandler, 1902; Sicher, 1912; Hafferl, 1919). During

embryogenesis Of higher mammals, anastomoses form

164
in such a manner as to permit take-over by the
arteria carotis externa of branches (primarily,
arteria stapedia) of the internal carotid system

(Tandler, 1899; Stephan, 1954).

Preliminary to further discussion of the
external carotid artery, the phylogenetic role of
the stapedial branch of the internal carotid artery
must be indicated.

The arteria stapedia is a primary artery
of the mammalian embryo (Grosser, 1901; Tandler,
1902; Evans, 1912; Sicher, 1912; Hafferl, 1919;
Congdon, 1922). Its precursor is found in lower

vertebrates: in the lizard, Hemidactylus flaviviridis

 

(Bhatia and Dayal, 1933), the stapedial artery arises
from the internal carotid artery at the posterior end
of the ear. Although the stapedial artery sends off

a small auricular branch, its two major rami show

a distribution Similar to that observed in primitive
mammals. The more dorsal ramus (temporal artery)
supplies occipital and temporal regions (still lacking
branches from the external carotid artery) and then
presents a pattern very Similar to the supraorbital
division (superior ramus) in mammals; that is, it
enters the orbit to send off lacrimal, ophthalmic,

and frontal branches. The second major ramus (mandib-

ular artery) of the lizard's stapedial artery is

165
comparable in distribution with the inferior ramus of
primitive mammals and mammalian embryos; thus it has a
branch comparable with the inferior alveolar artery and
one branch (superficial Ophthalmic artery) resembling
the mammalian infraorbital artery in its distribu-
tion. Of great interest is the presence of an anastomo-
sis interconnecting the mandibular ramus and the exter-
nal carotid artery. A similar anastomosis occurs very
early in mammalian embryos (Grosser,'l901; Tandler,

1902; Evans, 1912; Sicher, 1912; Hafferl, 1919).

The Simplest adult mammalian external
carotid pattern is found in Monotremata (Tandler,
1899, 1901). The arteria carotis externa is not
yet completely differentiated as to its branches,

especially in Ornithorhynchus. The latter genus

 

lacks an arteria maxillaris (interna). The arteria
stapedia, arising below the tympanic cavity, still
sends a branch to occipital-artery territory. Medial
to the nervus trigeminus, the inferior ramus of
arteria stapedia receives the slender ramus anasto-
moticus from the arteria carotis externa. On emerging
from the pterygoid canal, this inferior ramus bifur-
cates into the ramus infraorbitalis (accompanying
nervus infraorbitalis into the face) and the ramus
orbitalis that apparently supplies all structures of

the orbit except the bulbus oculi. In the absence of

166
a superior ramus, there are no supraorbital branches.
(There is, however, good presumptive evidence of
the presence of the superior ramus in Echidna;
thus embryonic presence of both rami of the arteria
stapedia is assumed for Monotremata.) (Tandler,
1899, 1901)

Marsupialia, including Didelphis, Show the

 

direct continuation of the arteria carotis externa
in the form of a typical (Similar to human) arteria
maxillaris (interna). Embryonically, it represents
the distal (to mandibular division of nervus trigem-
inus) portion of the inferior ramus of arteria
stapedia. As flow through the arteria carotis
externa increased, the proximal inferior ramus of
arteria stapedia evidently atrOphied. This pattern
is typical for most Eutheria, including all Primates
(Tandler, 1899, 1901).

In this reSpect, some Insectivora appear
to be more primitive than Marsupialia; secondary

reversion is another possibility. In Erinaceus

 

europaeus (hedgehog) and Talpa europaea (mole),

 

 

the arteria carotis externa does not continue as

an arteria maxillaris. (See Figure 4.) Since the
arteria stapedia is initially strong and the arteria
carotis externa weak among mammals, atrOphy of the

proximal inferior ramus (arteria stapedia) would not

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Lemur

Figure 4. Carotid Arterial Branching Patterns.

Ilt‘l‘l

 

168

Figure 4. Key to letters used:

A

"UOZZHNHHIEQ’IIMUOW

FBI/J 311.0

stapes

mandibular ramus of the trigeminal nerve
common carotid artery (stem)

external carotid artery (stem)

superficial temporal artery

inferior alveolar artery

infraorbital artery

orbital ramus of the infraorbital artery
internal carotid artery (stem)

cerebral carotid (promontory) artery

anterior ramus of the cerebral carotid artery
ophthalmic artery

ciliary artery (ramification)

posterior ramus of the cerebral carotid artery
stapedial artery (stem)

(proximal) inferior ramus of the stapedial
artery

superior ramus of the stapedial artery
middle meningeal artery
ethmoidal artery

frontal artery

169

be expected. In the hedgehog and mole, even the
connection between them evidently breaks down (by ob-
literation of a part of the arteria carotis externa
proximal to the anastomosis with the ramus inferior
of arteria stapedia) (Sicher, 1912). Thus, the arteria
carotis externa ends as the arteria temporalis super-
ficialis in Erinaceus europaeus. The stapedial in-
ferior ramus retains the anterior mandibular and
infraorbital territories (including probable
embryonic take-over, by anastomosis, of the inferior
alveolar artery) (Sicher, 1912). The involution of
the arteria carotis externa goes a step further in
Ifgpa, in which even the superficial temporal artery
has become part of the stapedial System (Tandler,
1899; Sicher, 1912).

In Suncus murinus (Soricidae) (Sharma, 1958)

the pattern conforms with that of Ptilocercus (Le Gros

 

Clark, 1926a) and at least some Tupaia glis individuals

 

(present study). That is, the proximal stem of the
inferior ramus (arteria stapedia) persists in direct
continuity with the arteria maxillaris. (See Figure
4.)

In general, Insectivora most nearly retain
the basic embryonic pattern for mammals. Some tUpaiids
and soricids retain it completely. In hedgehog and

mole, the inferior ramus (arteria stapedia) is

170

retained completely; a portion of the arteria
carotis externa has been lost, apparently by secondary
regression. The attenuation and possible discon-
tinuity observed in tupaiids may represent an in-
cipient primate trend; it is also a rather general
eutherian trend, leading to such forms as edentates,
ungulates, some carnivores, and some bats.

It is of interest that a phase of develop-
ment similar to that described by Sicher (1912)
for Talpa also occurs in Tarsius (Prosimii) (Hafferl,
1919); thus there appears to be, at least in some
primate embryos, an early phase of rapid extension
of the arteria carotis externa, followed by a
growth lapse in which connection with the arteria
stapedia is lost and finally by a marked length-
ening of the arteria carotis externa to re-establish
connection. The last one of these three phases is

evidently lacking in Talpa and Erinaceus. The

 

three-step succession has been Observed also in the
pig by von Hofmann (1914) and in bats by Grosser
(1901). In certain Rodentia (Mpg and Pedetes) the
arteria carotis externa has also lost connection,

in adult, with the arteria maxillaris (Tandler, 1899).

The status of Tupaia glis appears (without

 

benefit of embryological study) to be as primitive

as that of any mammal. Like Insectivora, it retains

171
both SUperiOI and inferior rami of arteria stapedia
as well as the anastomosis of the latter with the
arteria carotis externa. This anastomosis, present
in Ptilocercus, may be lacking or interrupted in
some Tupaia individuals, at least in the (serially

sectioned) Specimen of Tupaia minor observed by Le

 

Gros Clark (1926a). In contrast with Erinaceus and

 

Talpa, among Insectivora, the arteria carotis externa
of Tupaiidae does continue as an arteria maxillaris.
The principal difference between Lemur

varius and Tupaia glis is the loss, in Lemur, of the.

 

proximal stem of the inferior ramus of arteria
stapedia. (See Figure 4.) Otherwise, both show a
well-developed stapedial system and continuation of
the arteria carotis externa as an arteria maxillaris
(Tandler, 1899). In this pattern the arteria meningea
media remains definitively attached to the SUperior
ramus of arteria stapedia. This is also the case in

Nycticebus (Davies, 1947) and Daubentonia (Chiromys)

 

 

(Platzer, 1960).

The Maxillary Artery(Pterygoid Portion)

A feature of unknown phylogenetic significance
is the path taken by the pterygoid portion of the
arteria maxillaris through an external aliSphenoid

canal in Tupaia glis. The external aliSphenoid

 

process and its canal are very small. The arteria

172
maxillaris seems to deviate mediad to enter it
and no visible branches are given off by the artery
within the canal. An external aliSphenoid canal

is present in some insectivores (Gymnura, Centetes,

 

Myogale, and Solenodontidae) and absent in others
(Erinaceus, Potamogalidae) (Dobson, 1882-1890).

This canal may or may not be present in Macroscelididae
(Muller, 1935; Evans, 1942). It is also present in
elephants, some Carnivora, and some Perissodactyla
(Turner, 1848). Thus, an external aliSphenoid

canal is probably a primitive mammalian feature.

Since it has no apparent function in Tupaiidae, it
seems most reasonable to regard it as a vestige.

When present in living mammals, it evidently transmits
a (masticatory) muscle branch of nervus trigeminus
(Turner, 1848). It is of interest that in some

bats and rodents the (internal) maxillary artery

runs through the cranial cavity for a part of its
course (Grosser, 1901; Guthrie, 1963). The ramus
inferior of arteria stapedia may formerly have had

a more medial position.

The Infraorbital Artery--Ramus Orbitalis
On reaching the orbit, the arteria maxillaris

of Tupaia glis becomes the infraorbital artery. This

 

is the characteristic pattern for many mammals,

including Marsupialia, Insectivora, and Primates.

173
The orbital ramus, generally present, is a collateral
branch of the infraorbital artery. In Marsupialia
and in higher Primates the ramus orbitalis takes
the form of a few small muscle branches (Tandler,
1899).

In Tupaia glis the ramus orbitalis of
arteria infraorbitalis joins the ethmoidal branch
of the superior ramus of arteria stapedia just
outside of the foramen conducting the arteria
ethmoidalis into the anterior cranial cavity
(region of the olfactory lobe). (See Figure 4.)

The only comparable description found in
the literature is that for Tarsius. Hill (1953a)
described a large branch of the infraorbital artery
that courses across the floor to the medial wall
of the orbit, where it ramifies Over the os planum
and sends one of these branches through the ethmoidal
foramen.

Tandler (1899) simply described the large
ramus orbitalis of Lemur varigs as arising in the
infratemporal fossa and joining the superior ramus

of arteria stapedia. (See Figure 4.) In Daubentonia

 

(Chiromys) madagascariensis a ramus orbitalis is
not demonstrable. In Loris (Stenops) gracilis and
Galago (Otolicnus) crassicaudatus the ramus orbitalis

is reduced to a small muscle branch.

174

In rodents and bats the orbital ramus is
more variable in its point of origin from the
arteria carotis externa or its terminal continuation
as the arteria maxillaris and the arteria infra-
orbitalis. In many bats it is the major vessel
of the orbit, often referred to as the arteria
ciliaris communis, equalling arteria infraorbitalis
in size (Grosser, 1901; Guthrie, 1963).

The Internal Carotid Artery (Stem
and Major Ramification)

 

 

The arteria carotis interna is a primary
artery in all vertebrates. As expected, it is well-
develOped in Monotremata and Marsupialia, most
Insectivora (known exceptions: Suncus murinus
and probably some TuEaia Species), most Primates
(known exceptions: LEEEE and Daubentonia species),
as well as some members of many other orders
(Edentata, Perissodactyla, Pinnipedia, Carnivora,
Rodentia, Chiroptera) (Tandler, 1899, 1901; Le
Gros Clark, 1926a; Stephan, 1954; Sharma, 1958).

Even when the internal carotid SUpply to
the brain is reduced or virtually absent, the initial
portion giving rise to the stapedial artery remains
well developed; examples include lemur, aye-aye,
horseshoe bat, and guinea pig (Tandler, 1899;

Grosser, 1901). Reduction usually occurs at a

175
relatively late stage of embryonic life (Wolff, 1938).
The primitive mammalian internal carotid
artery had four principal branches: (1) the medial
entocarotid artery, (2) the promontory artery, (3)
the SUperior ramus of the stapedial artery, (4) the

inferior ramus of the stapedial artery (van Valen,

1965).

The Internal Carotid Artery;-
Medial Entocarotid Branch

 

The medial entocarotid artery (typically
entering the cranium at the junction between
petrosal and basioccipital-basiSphenoid bones) is
most commonly lost. It is not present in Recent
Tupaiidae (van Valen, 1965; McKenna, 1966).
Although usually said to be absent in all Recent
Primates, Saban (1963) has Observed a carotide

antérieure in Cheirogaleus (dwarf lemur). This

 

vessel has the appropriate location and pathway for
a medial entocarotid artery. Further, the rameau
tubaire of Lamp; is possibly a persistent adult branch
of a medial entocarotid artery. There is also some
speculation that the medial branch of the internal
carotid artery is not the promontory artery (as in
other Primates) but, rather, is the medial entocarotid
artery (van Valen, 1965).

Although it has been postulated that rodents

176

have descended from an early primate stock (McKenna,
1961), Plesiadapis (earliest known primate skull of
the late Paleocene epoch) did not possess a medial
entocarotid artery (McKenna, 1966). The artery
occurs in some Recent rodents (Guthrie, 1963).

Oligocene leptictids (at a time when primates
had already diverged from this stock) possessed a
reduced medial entocarotid artery as well as an
unreduced lateral branch giving rise to promontory
and stapedial arteries (van Valen, 1965; McKenna,
1966).

Based on the observations cited and Tandler's
(1899) finding that loss of a medial entocarotid
artery is common among Eutheria, it is quite possible
that the loss of this artery has occurred independent-
ly in the history of Primates and Tupaiidae.

The Internal Carotid Artery-~Promontory
(Cerebral Carotid) Branch

Of some phylogenetic interest is the size
of the promontory artery (the branch of the internal
carotid artery that generally joins the circulus
arteriosus cerebri in modern tUpaiids and primates).

In Leptictidae (of the Oligocene epoch) the
lateral branch of the internal carotid artery
divided at the promontorium into two large vessels:

stapedial and promontory arteries. This pattern

177

persisted in the Microsyopidae (resembling Plesiadapis)

 

and in primitive Primates, such as Plesiadapis,

 

Necrolemur, and Tarsius (McKenna, 1966). The large
calibre of the promontory artery has persisted in

all Recent Primates except Lemur and Daubentonia,

 

as well as other Madagascan lemurs, including

Hapalemur, Lepidolemur, Microrhynchus, Propithecus,

 

 

and Lichanotus (Hill, 1953b; Winge, 1941).

Sharma (1958) has observed that Suncus
murinus (Soricidae), the house shrew of India, is
at least one modern insectivore possessing a
promontory artery of small diameter correlated with
a stapedial artery of greater diameter.

There is some apparent variation among Tupai-
idae. In Ptilocercus the promontory and stapedial
arteries are about equal in Size. This also is the
case for TUpaia glis specimens observed in the present
study. However, the promontory artery has been gen-
erally described as reduced in calibre (not more than
one-third of the diameter of the stapedial artery) and
thus comparable with that of lemurs (Le Gros Clark,
1926a; Saban, 1956-1957). Saban (1963) has demon-
strated that the stapedial artery of lemurs may be
quite reduced or even absent. In any case, lack of
Special resemblance of tupaiids to lemurs does not

preclude tupaiid-primate affinity.

178

The Internal Carotid System (Relation
to Tympanic Walls)

 

Another phylogenetic issue is the freedom
with which the internal carotid system conducts it-
self through the ear region in relation to the
bony walls.

The primitive condition, present in the
ultimate primate ancestors, was that in which the
internal carotid artery and its branches coursed

through grooves; hence "free." In Plesiadapis

 

(late Paleocene) the artery ran through an Open
groove (McKenna, 1966).

In Oligocene leptictids there was still a
groove for the internal carotid artery. The trench,
to be sure, became quite deep as the promontory
artery approached the foramen into the cranial
cavity but has been observed to close over (only
briefly), in one known fossil Specimen, near the
bifurcation into promontory and stapedial branches
(McKenna, 1966).

Early Tertiary insectivores generally diSplayed
Similar grooves (McKenna, 1966).

Thus, although both modern Primates and
Tupaiidae have the internal carotid system completely
enclosed within bony tubes in the medial wall of
the tympanic cavity, the condition probably arose

independently in these two groups. Actually, such

179

bony enclosure (partial or complete) must have been
a common evolutionary trend occurring in other
grOUps: some Pantolestidae (Short-lived side branch of
Proteutheria) and some members of modern Erinaceidae,
Chrysochloridae, Talpidae, and Macroscelididae (van
Kampen, 1905; McKenna, 1966).

The location of the carotid foramen, opening
into the bulla, although common to tupaiids and lemurs,

is a common one among mammals generally (van Valen,

1965).

The Internal Carotid.Artery--
Stapedial Branch

 

Basically, the stapedial branch of the inter-
nal carotid artery has long been the artery accom-
panying the trigeminal nerve ramification (Tandler,
1899).

In the lizard, Hemidactylus flaviviridis,

 

where the internal carotid artery lies on the dorsal

Side of the basiSphenoid bone it sends one branch

into the cranial cavity while the major portion
diverges somewhat to continue anteriorly, sending
branches to the masseter muscle, palate, nose, and
maxilla (Bhatia and Dayal, 1933).

In another lizard, Crotaphytus collaris,

 

Burda (1966) described the internal carotid artery

as passing through a canal in the floor of the

180
basiSphenoid bone. Just before entering the cranial
cavity, it sends off a large naso-palatine branch
running laterad through a foramen in the antero-
lateral wall of the canal and then sends branches
to the floor of the orbit, the nasal region, and
the palate.

Tandler's (1902) embryological study of the
rat indicated that although the stapedial artery orig-
inates from the second aortic arch, it takes over
the branches of the primitive mandibular (first
arch) artery, giving rise to the supraorbital,
infraorbital, and mandibular arteries. (The external
carotid artery annexed the mandibular artery later.)

In the primitive state, therefore, the
branches of the first branchial-arch artery accompany
the divisions of the trigeminal nerve. These periph-
eral branches would not alter if only the main stem
of the primitive mandibular artery was to be replaced
by the stapedial artery. In that case the supra-
orbital division of arteria stapedia would have an
extensive intracranial course from petrous bone to
orbit. _ThiS is observable in such primitive forms as
Lgmgg, Erinaceus, Tal a, and Arctomys (Tandler,
1899), as well as in Tupaiidae. This may also have
been true in some degree for the infraorbital ramus

(lizards and some rodents), and to a Slight degree

181
for the mandibular ramus (branches to muscles of
mastication emerging through the cranial wall in a
number of primitive mammals) (Tandler, 1899; Guthrie,
1963). The presumably vestigial aliSphenoid canal
of Tupaia may be most plausibly SXplained on this
assumption since it still probably conducts a small

nerve to a masticatory muscle (Turner, 1848).

The Internal Carotid Artery-~1nferior
Ramus of the Stapedial Branch

 

 

Loss of the proximal inferior ramus of the
stapedial artery may have occurred independently
along primate and tupaiid lines. In the ontogeny
of all Primates the proximal stem of this vessel
atrOphies and the distal ramification is taken
over by the external carotid artery (Tandler,

1902; Hafferl, 1919). This may also be the case in
some TuEaia Species or individuals, as indicated by
the present study and by Le Gros Clark's Observa-
tions (1926a). However, in Ptilocercus and most

Tupaia glis Specimens observed here the proximal

 

stem of the stapedial inferior ramus persists in the
adult, if only as a very slender anastomotic
connection with the maxillary artery. It also per-
sists in Suncus murinus (Soricidae), according to
Sharma (1958). No evidence is available for

Dendrogale (smooth-tailed tree shrew), which Davis

 

182
(1938) regarded as more primitive than Tupaia and
slightly less primitive than Ptilocercus in general

structural features.

The Internal Carotid Artery-Superior
Ramus of the Stapedial Branch

 

 

In mammals the primitive distribution areas
of arteria stapedia included the upper jaw, the
orbit, and the dura mater, as well as peripheral
regions of the lower jaw. The ramus superior of
arteria stapedia is essential to development of
the arteria meningea media and the orbital arteries.
Although the distal ramification Of these vessels
may become secondarily attached to some other
definitive parent vessel, the ramus SUperior of
arteria stapedia served as their original stem
(Tandler, 1899). Only on this assumption can many
peculiar secondary conditions and anomalies be
explained. The limited evidence from embryology
of representative forms reaffirms this view
(Tandler, 1902, 1906; Grosser, 1901; Hafferl, 1919;
Wolff, 1938; de la Torre and Netsky, 1960).

The Middle Meningeal Artery

 

Upon entering the cranial cavity (posterior
corner of middle cranial fossa) the ramus superior
of arteria stapedia sends off the arteria meningea

media in Tupaia glis, as in a number of other

 

183

primitive mammals: Talpa, Erinaceus, Suncus (Insec-

 

 

tivora); Daubentonia, Lemur (Primates); VeSpertilio,

 

 

Noctilio, Plecotus, Phyllostoma, Mormoops, Taphozous

 

 

(Chiroptera); Rattus, Sciurus, Marmota (Rodentia).

 

(See Figure 4.) The main stem of the intracranial
portion of the SUperior ramus of arteria stapedia
then continues its subdural course anteriorly to
reach its terminal orbital ramification (Tandler,
1899, 1901; Grosser, 1901; Sharma, 1958). This
pattern conforms most closely with the primary
embryological condition of the ramus superior of
arteria stapedia (Grosser, 1901; Tandler, 1902).
Monotremata, Marsupialia, and Edentata are
not included in this primitive group. These animals
Show an apparent loss of the ramus superior (arteria
stapedia) as adults. In lower mammals generally
there is a tendency for the occipital artery to
remain as a branch of the arteria carotis interna,
arising just below the posterior skull base. The
arteria (stylo)mastoidea, a branch of the arteria
occipitalis, takes over the distal ramification
of the superior ramus of arteria stapedia in its
middle meningeal and orbital portions. This distal
portion of arteria occipitalis is Often called the
arteria diplogtica magna in these three primitive,

but aberrant, groups (Tandler, 1899, 1901).

184

A more usual modification among mammals, and
likewise probably acquired in parallel more than
once, is an apparent secondary attachment of the
arteria meningea media to the arteria maxillaris when
the major intracranial portion of the ramus superior
of arteria stapedia atrOphies ontogenetically.

Such "secondary attachment" occurs in Pterogus
(Chiroptera), Lagomorpha, Carnivora, ungulates, and
many Rodentia.

Along the primate line only Lemuridae and
Daubentoniidae retain the primitively complete
SUperior ramus of arteria stapedia for life
(Tandler, 1899; Hill, 1953b; Platzer, 1960). The
definitive connection of the arteria meningea
media, as found in man, was evidently a rather late
phylogenetic develOpment. As noted above, the
orbital portion of the superior ramus of arteria
stapedia generally persists in mammals. When the
major part of the intracranial portion breaks
down, secondary connections must be made. Along
the primate line the Ophthalmic artery tends to
become the dominant stem for vessels of the orbit
(Tandler, 1899).

In Nycticebus (slow loris) the ophthalmic
artery sends off a recurrent branch re-entering the

middle cranial fossa more laterally. This recurrent

185

branch then courses posterodorsally along the usual
path of the middle meningeal artery (Davies, 1947).
Obviously, this is the middle meningeal artery,
which has retained rostral connection by way of its
early anastomosis with the Ophthalmic artery but
has lost connection with its original stem; thus
there is no attachment to the arteria maxillaris.

A similar arrangement is found in Callithrix

 

(Hapale) penicillata (marmoset), in which Tandler
(1899) observed a recurrent branch from the lacrimal
artery. (This only requires a shift within the
orbit of the early anastomosis between the arteria
stapedia and the arteria Ophthalmica. The arteria
lacrimalis may be a branch of either, depending
on the Species.) Why Davies (1947) thought the con-
nection of the middle meningeal artery with the
Ophthalmic artery must represent persistence of
the inferior ramus of arteria stapedia is difficult
to understand; this is not the simplest explanation.
Hill (1953a) found no such recurrent artery
in Tarsius. He did observe relatively large anterior
meningeal branches of the Ophthalmic artery. The
arteria meningea media of Tarsius retains connection
with its original stem, and thereby, with the arteria
maxillaris, of which it is a definitive branch. It

is enclosed by the roots of the auriculo-temporal

186
nerve, as in man, but enters the cranial cavity
through the foramen ovale. This raises a question
as to its homology with the arteria meningea media
of other Primates, eSpecially in view of the Obser-
vation that Tarsius has a conspicuously develOped
arteria stylomastoidea with an intracranial area of
distribution correSponding with that of the usual
arteria meningea media. Nothing comparable with
this has been observed among mammals except in
marsupials and edentates (Tandler, 1899; Hill,
1953a).

Although the proximal stem of arteria
stapedia is larger in Lemuroidea than in Lorisoidea,
it is generally (except in Lemuridae and Daubentoni-
idae) used up in supplying structures in the temporal
bone, such as mucous membrane of the tympanic walls
(Hill, 1953b).

In this reSpect the meningeal-artery pattern
of monkeys, apes, and man seems closer to the pattern

in Nycticebus. As a rule for monkeys and apes,

 

there is still considerable inflow of blood from
arteria Ophthalmica into arteria meningea media, the
proximal connection with the arteria maxillaris
remaining quite weakly developed. In man the latter
connection gains the ascendancy; yet the anastomosis

with the arteria ophthalmica does typically persist,

187
even if rather attenuated (Tandler, 1899; Wolff,
1954; de la Torre and Netsky, 1960).

Tandler (1899) postulated, on the basis
of comparative anatomic observations, that the human
arteria meningea media has been put together from
a portion of the ramus inferior, a portion of the
ramus superior, plus the meningeal ramification that
usually Springs from the ramus SUperior of the
arteria stapedia. Based on studies of human
embryos, Evans (1912) concluded that the proximal
stem of the middle meningeal artery represents the
remains of the common maxillo-mandibular stem of
the stapedial artery. This would explain apparent
secondary attachment to the maxillary artery.

As with most features of the cranial arterial
system, these observations on the middle meningeal
artery demonstrate that Tugaia is among the most
primitive of all mammals: from its pattern could
be derived that of Primates-éor of Monotremata,
Marsupialia, and all higher Eutheria.

Orbital Blood Supply (Relationship of

the Internal and External
Carotid Arteries)

 

 

The true arteria Ophthalmica, as a branch
of the intracranial portion of the arteria carotis
interna, is a primary artery in vertebrates generally.

In all vertebrate classes, this artery varies in

188
degree of development (Tandler, 1899, 1901;
Hofmann, 1900; Beddard, 1904b, 1905a, 1905b;
Allen, 1905; Stephan, 1954).

In lizards the orbital contents are
supplied by the true Ophthalmic artery as well as
by lacrimal and Ophthalmic branches of the supra-
orbital and infraorbital (mandibular) divisions
of the stapedial or facial carotid system derived
from the extracranial stem of the internal carotid
artery (Bhatia and Dayal, 1933; Burda, 1966).

In general among mammals there are three
sources of arterial blood supplying orbital contents:
(1) the arteria Ophthalmica from the (intracranial)
arteria carotis interna; (2) the orbital portion of
the ramus superior of arteria stapedia from the
(extracranial) arteria carotis interna; (3) the
ramus orbitalis of arteria maxillaris (almost
constantly from the arteria carotis externa in
adults but derived from the extracranial inferior
stapedial branch of the arteria carotis interna,
embryologically). Any one of these three paths of
inflow may be eSpecially develOped or degenerated.
A complete continuum of possibilities can be
demonstrated among mammals (Tandler, 1899).

In Erinaceus europaeus all three paths are

Supplied equally; similar relations are Observed in

189
lgmgg and Daubentonia (Chiromys) (Tandler, 1899).
Within a number of orders, certain forms
Show secondary reduction of the true arteria Ophthal-

mica. In some Rodentia (e.g., Lagostomus), the arteria

 

Ophthalmica produces only an arteria centralis
retinae (Beddard, 1904b). The arteria Ophthalmica
is rather weak in edentates and their allies
(armadillo, sloth, ant-eater, pangolin), being
nutritive to the bulbus oculi only. (Judging from
its source, course, and embryological development,
optic nerve and bulb appear to be this artery's
primary distribution areas.) The arteria Ophthal-
mica is not demonstrable in (virtually blind) Talpa
europaea (Tandler, 1899). An arteria ophthalmica is
retained, among Chiroptera, only in PterOpidae. In

Rhinolophus (horseshoe bat) even the optic nerve

 

artery may arise from the orbital ramus of arteria
maxillaris (Grosser, 1901).

Thus, fairly well-developed ophthalmic arteries
are found in prosimians, insectivores with normal
vision, and in monotremes and marsupials generally.

One can only Speculate regarding the extreme reduc-

tion of these arteries in Tupaia glis. Assuming

 

that other tupaiids are similar, one might suppose
that the stapedial pathways, seemingly more directly

in line with the main stem of the arteria carotis

190
interna, would be more efficient than the more in-
direct intracranial branch. This view is also
based on the assumption that, in the face of con-
siderable expansion of the brain and eye, the .
vertebral arteries had not yet develOped adequately,
and the intracranial or cerebral carotid arteries
had not readjusted apprOpriately. It would seem
that Daubentogig and Lgmg; were even less adequate
in this adjustment; here, however, the vertebral
arteries have gained ascendancy to compensate for
diminished cerebral carotid supply to the Ophthalmic
arteries. In all other Primates the adjustment was
made at the expense of the stapedial system. Perhaps
the develOpment of a carotid canal was an important
factor in this primate adjustment. Perhaps the detour
of the stapedial stem through the stapes made one or
the other less efficient. It is also of interest to
note that the middle ear cavity of Tupaia is both rela-
tively and absolutely larger than in man. In any
event, the explanation for the reduced arteria ophthal-
mica is not directly apparent from the vessel rela-
tions. It may be noted that in Daubentonia and Lgmg;
(with reduced arteria carotis cerebralis), the arteria
Ophthalmica arises at the point where the arteria
carotis cerebralis bends posteriorly to give rise to

the arteria communis posterior. This arrangement

191
leaves the impression that the arteria communicans
posterior gives off the arteria ophthalmica, and
that the slender part of arteria carotis cerebralis
also opens into the arteria Ophthalmica (Tandler,
1899). It has been demonstrated, in Daubentgnia
and nggg, that the widened portion of the
arteria carotis cerebralis, between the arteria
Ophthalmica and the circulus arteriosus, actually
carries blood in a reversed direction: out of the
circulus arteriosus into the arteria ophthalmica
(Platzer, 1960). In Tupaia glis the Ophthalmic
artery arises somewhat farther rostrally, from the
anterior ramus of the arteria carotis cerebralis.
In all prosimians, other than Daubentoniidae and
Lemuridae, the arteria ophthalmica is the first
branch of the arteria carotis cerebralis within
the cranial cavity (Platzer, 1960).

In Tupaia glis the orbital contents are
primarily supplied by supraorbital (superior ramus)
branches of the arteria stapedia. The arteria
Ophthalmica probably supplies a small volume of blood
to the ciliary branches. The arteria centralis
retinae may arise from it or from a ciliary
trunk. Such arrangements do vary among mammals
(Rochon-Duvigneaud and Rode, 1943). Gross dissection

could not reveal the source here. ,Although some

192

bats completely lack an arteria centralis retinae
(Grosser, 1901), the fundus photograph study of
Samorajski, Ordy, and Keefe (1966) demonstrated
that this artery is present in Igpaia glis. The
ramus orbitalis of arteria maxillaris (infraorbital
portion) appears to be practically excluded from
orbital supply by its anastomosis with the ethmoidal
branch of the ramus superior of arteria stapedia.
The ramus orbitalis does send Off a few small
muscle branches, en rgute across the orbital floor.

The relative importance of infraorbital and
supraorbital stapedial sources of supply to the
orbit is quite variable among mammals. Monotremes
have lost the major portion of the superior ramus
of arteria stapedia, leaving the orbital ramification
to be annexed by the arteria ophthalmica and the
ramus orbitalis of arteria maxillaris (infraorbitalis).
It is the latter source that evidently dominates in
Chnithorhynchus, supplying all of the orbital contents
except the bulbus oculi (Tandler, 1899). In
Echidna rami Ophthalmici of the arteria diplogtica
magna (modified superior ramus of arteria stapedia)
share in supplying soft parts (Tandler, 1899, 1901).
In Marsupialia the pattern is described by Tandler
(1899) as very Similar to that in man; thus only

a few weak branches extend to the inferior ocular

193
muscles from the arteria maxillaris. No anastomosis
with upper orbital vessels is demonstrable. The
arteria ophthalmica is strong and has taken over
the orbital ramification of the ramus superior
arteriae stapediae. Edentates are rather variable
in orbital supply from the three possible sources,
with the well-developed ramus orbitalis of arteria
maxillaris tending to take the lead. Microchiroptera
(bats), while losing the arteria ophthalmica, vary
in degree of dominance of the inferior or superior
ramus of arteria stapedia. Although the superior
ramus is primary, its main stem frequently drops
out, and its orbital ramification persists to be
annexed by the ramus orbitalis of arteria maxillaris
(Tandler, 1899, 1901).

Thus, it is difficult to say whether the
three-way balance observed in Erinaceus is more
primitive, or whether the balance favoring the ramus
superior, as in Igggig, is more primitive.

One may also ask whether the "ramus orbitalis”
is primarily orbital or primarily ethmoidal in its
distribution area. This, too, is difficult to de-
termine from available evidence. Both supraorbital
and infraorbital branches of the arteria stapedia
typically send off ethmoidal branches, among lizards
as well as mammals (Tandler, 1899, 1901; Hofmann,

194
1900; Bhatia and Dayal, 1933; Burda, 1966). An
ethmoidal branch of the ramus superior may be ob-
served extending into the nasal cavity of a mole
embryo (nine millimeters in length). In a some-
what older embryo (thirteen millimeters in length),
the extension of the arteria ophthalmica takes over
the ramus orbitalis of arteria maxillaris. Later
still, the arteria Ophthalmica disappears completely
(in an embryo seventeen and one-half millimeters
in length) (Sicher, 1912). A Similar sequential
pattern occurs in bats (Grosser, 1901).

Among Prosimii, there is rapid embryologi-
cal ascendance of the Ophthalmic artery. In
Daubentonia, this artery gives rise to two equal-
Sized branches in the orbit. The more SUperficial
one anastomoses with the superior ramus of arteria
stapedia. The united trunk sends off arteriae
lacrimalis, frontalis, and ethmoidalis. The deeper
branch of the Ophthalmic trunk produces all of the
arteriae ciliares (Tandler, 1899; Platzer, 1960).

Tandler (1899) observed no ramus orbitalis
from the arteria maxillaris in Daubentonia. In
Lgmgg, however, he did find a large ramus orbitalis
joining the superior ramus, directly upon entry into
the orbit. The main trunk of the ramus superior

arteriae stapediae also communicates with the arteria

195
Ophthalmica before breaking Up into the arteriae
lacrimalis, frontalis, and ethmoidalis. The
arteriae ciliares arise from the arteria Ophthalmica
itself.

.Arctomys marmota (woodchuck), as a fairly

 

primitive rodent, should be closer to the primate
line than most other modern mammals. Here, the
arteria stapedia has two well-developed rami. The
superiOr ramus is the main source of blood to the
orbit and eye. Within the orbit, it is joined by
a well-develOped anastomotic branch from the arteria
carotis interna. This, according to Wolff (1938),
closely resembles an embryonic stage in the origin
of the ophthalmic artery in man. It also resembles
Daubentonia and Lgmgr, as well as Tupaia and
Erinaceus. However, in the adult 122313, the
Ophthalmic artery is reduced; in Erinaceus, the
superior ramus is not dominant in supply of the orbit.

In Lgrig, galagg, and Tarsius, there is no
ramus superior of arteria stapedia, as such. The
arteria Ophthalmica is well developed. There is,
apparently, no longer a ramus orbitalis arising from
arteria maxillaris. .A few small muscle branches
may represent it, in its reduced form (Tandler,
1899; Platzer, 1960).

This does not fully explain the loss of an

196
ethmoidal branch of arteria maxillaris, still
present in Tupaia. Barring the possibility of
species and individual variability in the Site of
attachment of the ramus orbitalis with the ramus
superior, it seems that loss of an ethmoidal
branch of arteria maxillaris is a definite primate
trend, one not observable in Tupaia glis or in most
higher mammalian groups. It may also be noted
that the “ethmoidal” extension of the intracranial
anterior cerebral artery to the nasal septum,
still prominent in Tupaia, regresses early in human
embryos (Tandler, 1899, 1902). Reduction of the
olfactory sense may be correlated with this. Such
reduction is slight in Tupaiidae.

In monkeys, apes, and man, all orbital
branches arise from the Ophthalmic artery, with the
exception of a few muscle branches from the arteria
infraorbitalis to the inferior orbital region
(Tandler, 1899; Lineback, 1933; Weinstein and
Hedges, 1962).

The Circle of Willis (Relationship of
the vertebro-Basilar and

Cerebral Carotid Systems)

In all inframammalian vertebrate classes
the arteria carotis interna (as the arteria carotis
cerebralis) alone SUpplies the brain. The arteriae

vertebrales are rudimentary with reSpect to supply

197

of the central nervous system. In all vertebrates
the arteria carotis cerebralis terminates on the
ventrolateral aSpect of the brain, bifurcating
into a ramus anterior (ramus cranialis) and a ramus
posterior (ramus caudalis). The two rami posteriores
unite as an arteria basilaris, continuing as an
arteria Spinalis ventralis (anterior) that diminishes
in calibre caudally. Thus, although there is
anastomotic connection between the vertebral and
carotid arterial systems in all vertebrates,
the former does not directly supply the brain in
forms below mammals. Only in some reptiles (snakes)
do lateral Spinal arteries even penetrate into the
cranial cavity in forming the basilar artery
(Hofmann, 1900; de vriese, 1905).

Comparative study of cerebral arteries
alone will not establish an evolutionary scale.
.A similar disposition of arteries is to be found
in grOUpS phylogenetically far removed from one
another. Within a single order (e.g., Rodentia)
one finds patterns of the circulus arteriosus
cerebri ranging from one extreme to another (Tandler,
1899; Hofmann, 1900; de vriese, 1905).

Tandler (1899) distinguished three major
types of arterial circle of Willis among mammals:

l. The internal carotid arteries supply the brain

198
more or less exclusively.

This arrangement is found in Monotremata
and in kangaroos among MarSUpialia. The pattern
may be primary or secondary in Artiodactyla,
Perissodactyla, Pinnipedia, Cetacea, and many
Carnivora.

2. The vertebral and internal carotid arteries
share, in varying proportions, in formation of
the arterial circle.

Early vertebral artery development,
along the mammalian line. led to appearance of
this arrangement in some Marsupialia (opossum, e.g.)
and in Insectivora generally. It is also found
in some Chiroptera, some Rodentia (jerboa, rat),
Lagomorpha, some Edentata (armadillo), as well as
a number of Primates (some Lemuroidea and all
.AnthrOpOidea).

3. The internal carotid arteries are almost or
completely excluded from involvement in the arterial
circle, which is thereby formed by vertebral arteries
alone.

Although Obviously a secondary condition,
this third arrangement appeared early in a number of
primitive mammalian groups, including some ChirOptera,
numerous Rodentia, some Edentata, and some Lemuroidea.

This pattern has also been acquired by more highly

199

evolved groups, such as some Carnivora.

The arrangement in Erinaceus and Talpa_
falls into the second category, above. The pattern
resembles that of higher Primates in the slight
develOpment of the proximal stem (posterior communi-
cating artery) of the posterior ramus of the cerebral
carotid artery; thus the anterior cerebral and middle
cerebral arteries appear to be mainly supplied by
the anterior ramus, and the posterior cerebral
arteries largely supplied by the bifurcation of the
basilar artery (Tandler, 1899; Hofmann, 1900). The
pattern observed in Tupaia glis (present study)
closely resembles this insectivore pattern. Although
the arteria communicans posterior is not small, its
calibre is somewhat less than that of the anterior
ramus or of the basilar branches of bifurcation
and the stem of the posterior cerebral arteries
arising from the latter. This observation conflicts
with the published statements regarding the very
reduced size of the arteria carotis interna after
release of arteria stapedia in the tympanic region
(Le Gros Clark, 1926a; Saban, 1956-1957). Although
there are no published accounts of the arteria
carotis cerebralis, it would presumably be small,
as in Lemuroidea. In Lgmg£_and Daubentonia, the

arteria carotis cerebralis is attenuated and

200
presumably supplies the arteria ophthalmica
primarily. (See discussion of arteria ophthalmica,
above.) The vertebral arteries and the basilar
artery are well developed. The posterior com-
municating arteries are large, appearing as direct
continuations of the branches of bifurcation of
the basilar artery; thus the anterior, middle,
and posterior cerebral arteries appear to be branches
of these paired trunks (Tandler, 1899; Hill, 1953b;
Platzer, 1960).

It seems that the reduction of the arteria
promontorii (and of arteria carotis cerebralis)
is a lemuroid Specialization (Hill, 1953b; Lamberton,
1947). This condition probably occurs in some
Tupaiidae (Le Gros Clark, 1926a; Saban, 1956-1957).
The condition could have arisen, in parallel, in
the two groups.

In other Prosimii (Lorisiformes and Tarsii-
formes) the arteria promontorii is large. In ngig
(Stenops) gracilis, the posterior communicating artery
is described by Tandler (1899) as still quite large,
much larger than in man. This most closely approxi-
mates the condition observed for Tgpaia glis in the
present study. Tarsius Shows greater reduction
of the posterior communicating arteries (Hill,

1953a).

201

It appears that both Tupaiidae and prosimians
reached a critical phase in achieving balance be-
tween the ascending vertebro-basilar system and
the well-established internal carotid system that
was just keeping up with it in the supply of the
rapidly eXpanding brain. Two divergent paths may
have followed, perhapsin parallel. Adong one
path, leading to higher Primates, the internal
carotid system maintained an important role. .Along
another path (not necessarily a single ancestral
line) leading to Lemuridae, Daubentoniidae, and
some Tupaia Species or individuals, the vertebro-
basilar system gained the ascendancy. Perhaps the
progressive accentuation of the bending of the
telencephalon (related to its great develOpment)
played a role in the recovery of the importance Of
the internal carotid system. Decérisy (1952) Observed
(by study of corrosion preparations) that as one
progresses from lemuroids to anthropoids the circle
of Willis gradually abandons its plane aSpect to
achieve a progressive folding of two half-rings that
tend to close in the manner of pages in a book.

In Chiroptera and Rodentia the circle of
‘ Willis varies in relation to the degree of atrophy
of the internal carotid artery; in the majority the

vertebral system outweighs the internal carotid

202
system (Grosser, 1901; de Vriese, 1905).

The Circle of Willis (Nature of
the Posterior Region)

The posterior region of the circulus

arteriosus cerebri of Tupaia glis appears to be

 

more primitive than in Erinaceus and Talpa. A
few features resemble those of certain prosimians.
The greatest Similarity is to such forms as Macropus

(kangaroo), MyrmeCOphaga jubata (giant ant-eater)

 

and Echidna (spiny ant-eater). Adequate comparisons
cannot be made.‘ Tandler's descriptions (1899, 1901)
and those of most primatologists are limited to the
prOportions of the circle of Willis itself, without
description of all branches and their distribution.
Hofmann's descriptions (1900), although detailed,
do not include monotremes, marsupials, or primates.
.Abbie's detailed descriptions (1933, 1934) of
certain arteries of the brain do not include any
prosimians or insectivores. NO definite conclusions
can be drawn from the few Observations to be cited
here.

The bifurcation of the basilar artery into
a common trunk, on each Side, for the SUperior cere-
bellar and posterior cerebral arteries is a feature
Shared by Tupaia glis with Myrmecophagg jubata (il-

lustration closely resembling asymmetrical tree

203
shrew 3), Macropus (Shellshear, 1930), Dasypus
novemcinctus (Hofmann, 1900), and certain prosimians.
Such common trunks (two millimeters in length) were
described by Beddard (1904a) for Perodicticus
EELL- Tandler (1899) stated that the two large
posterior communicating arteries of qubentonia

(Chiromys) madagascariensis arise from the basilar

 

artery. Each then sends off the arteria cerebelli
superior and the arteria cerebri posterior as side
branches. The description for Lemur varius was
much the same.

In Erinaceus europaeus and Talpa europaea,
the pair of arteriae cerebelli superiores arise at
the bifurcation of the basilar artery into the stems
of the posterior cerebral arteries (Hofmann, 1900).
This arrangement was observed unilaterally in one
Specimen of Tupaia glis (present study).

The presence of a truncus communis, as
observed in the limited sampling of marsupials,
edentates, prosimians, and tUpaias, would seem to
be the result of incomplete fusion of the arteria
basilaris elements, embryologically. Various forms
of incomplete fusion, including island formation,
are common in reptiles and primitive mammals (Hofmann,
1900; de Vriese, 1905, Shindo, 1914).

The presence of a few branches representing

204

”posterior cerebral” elements, in Tupaia glis, is

 

clearly a primitive mammalian feature.

In Reptilia, the posterior cerebral artery
still arises from the anterior ramus of the arteria
carotis cerebri (Hofmann, 1900; Shindo, 1914; Abbie,
1934). In general among Mammalia the posterior
cerebral artery or arteries tend to arise more
and more posteriorly, along primate and other
evolutionary lines. These successively more
posterior positions have been designated as alpha,
beta, gamma, and delta, by Hofmann (1900). The
dorso-posterior extension of each cerebral hemi-
sphere shifts the major field of supply posteriorly;
thus the stem shifts to the nearest source of
supply. This source is the stem of the longitudinal
anastomotic channel extending from the arteria
basilaris to the arteria carotis cerebralis on
each side. In man the posterior cerebral artery
has reached its ultimate posterior stem of origin
(Hofmann's “arteria cerebralis posterior delta").
Since the area of supply lies almost directly above
the midbrain, it utilizes mesencephalic vessels,
exclusively, as a source of blood (Abbie, 1934).
Shellshear (1930) postulated that ”in the Mammalia the
invasion of the cerebral hemispheres by visual, tac-

tile, and auditory tracts from the midbrain and

205

thalamus would seem to have had the effect of grouping
together the metameric branches to the corpora
quadrigemina, the geniculate bodies and the thalamus.”

In Echidna, Shellshear (1930) saw branches
arranged in parallel series for distribution to
corpora quadrigemina and the lateral aspect of the
midbrain. He described (1930) a similar array of
metameric branches from a longitudinal channel in
Myrmecophgga jubata.

Tupaia glis is somewhat less metameric;
several, fairly parallel branches share a common
stem from the posterior communicating artery
(longitudinal channel from the basilar artery to
the internal carotid artery).

The mesencephalic artery of Tupaia glis may

 

correspond with the “arteriae cerebri posteriores
gamma and delta” of Hofmann (1900). It may be con-
sidered as, in some degree, “cerebral" in distribu-
tion: it supplies the splenium region of the

corpus callosum and forms anastomoses with branches
of cortical distribution. The artery here described

(for Tupaia glis) as the arteria cerebri posterior

 

may represent the "arteria cerebri posterior beta”
of Hofmann (1900). The arteria choroidea anterior
of Tupaia glis would correspond, in source and in
part of its distribution, to Hofmann's (1900)

206
”arteria cerebri posterior alpha}; however, this
artery appears to have no cerebral distribution
in Tupaia glis.

In view of the greater cerebral development
of Tugaia, it seems strange that this animal would
show less fusion of branches than is observed in
Erinaceus (Hofmann, 1900); the anterior colliculi
of Tupaia do remain almost reptilian in relative
size, however. More thorough tracing of branches,
in serial sections of the brains of representative
primitive groups, will be required to make adequate

comparisons.

The Circle f Willis Mode of
Anterior Closure)

The mode of anterior closure of the circulus
arteriosus cerebri is of some phylogenetic interest.

In Tupaia glis the two arteriae cerebri
anteriores (terminations of the anterior rami of
arteriae carotis cerebralis) are interconnected,
anterior to the optic chiasma, by a transverse
arteria communicans anterior. From this, in turn,
a median, initially unpaired arteria cerebri
anterior communis extends posteriorly to bifurcate
again about half of the distance backward over the
corpus callosum.

Although Tandler (1899) used the term

207
“arteria communicans anterior“ in his description
of Erinaceus europaeus and Talpa europaea, the
account of the vessel is too brief to be sure of
its nature, other than that it closes the circle
anteriorly. Hofmann's (1900) detailed description
of these two insectivore species indicates differences
from Tupaia glis. He stated that the anterior
ramus of each arteria carotis cerebralis converges
toward its counterpart as it courses anteriorly
within the longitudinal fissure anterior to the
optic chiasma. The two arteries there unite as an
unpaired arteria cerebri anterior communis; thus
there would be no anastomotic arteria communicans
anterior. The arteria cerebri anterior communis
then gives off a large arteria bulbi olfactorii
medialis, distributed primarily to the medial
surfaces of the olfactory tracts and bulbs. The
arteria cerebri anterior communis bifurcates into
the arteriae corporis callosi in this olfactory
region. The arteriae corporis callosi then bend
backward and ramify on the upper surface of the

corpus callosum, as in Tupaiaglis.

 

Tandler's (1899) descriptions of Lemur varius,
Daubentonia (Chiromys) madagascariensis, Loris

(Stenops) gracilis, and Galago (Otolicnus)

crassicaudatus indicate presence of a median artery

208
that arises by merging of the anterior cerebral
arteries. For £392; and Daubentonia he indicated
that each anterior cerebral artery sends a
branch forward to supply the bulbus olfactorius.
Regarding Simiae, Tandler (1899) mentioned that
the unpaired median continuation of the anterior
cerebral fusion winds around the genu of the corpus
callosum and bifurcates distally. For man, he
stated that the anterior cerebral arteries are
united by an anterior communicating artery.

Beddard's (1904a) description of Perodicti-
cus potto indicated formation of an unpaired
arteria corporis callosi before turning upward
over the surface of the corpus callosum. He stated
that no arteria communicans anterior is present;
union was described as oblique. The arteria
corporis callosi divides again, farther posteriorly,
above the corpus callosum.

Lemur macaco and Lemur coronatus were both
described by Beddard (1904a) as possessing a stout
anterior communicating artery, from which arises
the briefly single anterior cerebral artery. After
a very short course over the corpus callosum, paired
anterior cerebral arteries (arteriae corporis
callosi) are formed again. Of all descriptions

found in the literature, this most closely corresponds

209
with that in Tupaia glis.

Similar variability in union of the anterior
cerebral arteries occurs in other orders, such
as Rodentia and.Artiodacty1a, as well as in the
class Reptilia (Tandler, 1899; Hofmann, 1900;
Beddard, 1905b; de vriese, 1905; Shindo, 1914; Abbie,
1934; Burda, 1966).

De vriese (1907) concluded that, within the
vertebrate series, the anterior cerebral arteries
show three major patterns:

1. The two anterior cerebral arteries run parallel,
with no connection between them. This most primitive
arrangement is found in fishes, some amphibians,
some reptiles, and all birds.

2. The two anterior cerebral arteries are inter-
connected by one or more anterior communicating
arteries in front of the optic chiasma; beyond this,
the two run parallel again. This pattern occurs in
some amphibians and some mammals.

3. The two anterior cerebral arteries converge,
anterior to the optic chiasma, to form an unpaired
arteria cerebralis anterior communis (arteria
corporis callosi, in some cases), which dissociates
again, after a varying distance, into two branches.
This arrangement is the common one, being found in

most reptiles and most mammals.

210

De vriese (1907) observed the embryological
sequence, as found in an animal (rabbit) with the
third arrangement described above. Initially, the
arteriae cerebrales anteriores lie parallel to
each other; then these two arteries become inter-
connected by a transverse vascular network; subse-
quently the two anterior cerebral arteries become
connected by an arteria communicans anterior.
Extending anteriorly from this anastomosis are one
deep, and two superficial, median anterior cerebral
arteries. In the final developmental phase, the
two anterior cerebral arteries merge as an unpaired
stem, due to atrophy of one of the primitive super-
ficial anterior cerebral arteries and the deep
median anterior cerebral artery anterior to the
level of the anterior communicating artery, which
is also lost.

Tupaia glis appears to have reached a stage
between the third and fourth, as described above.
Man's development appears to have stepped at the
third stage, with some loss of branches. Barring
secondary regression, Tupaia seems more primitive
in this respect than modern insectivores and most

prosimians.

211
The.Arteries of the Olfactory Region

 

The literature reveals the primitive nature
of the collaterals that the terminal anterior ramus
of arteria carotis cerebralis sends to the olfac-
tory lobe. Names and descriptions are so varied as
to make comparison difficult. There is little doubt
that the anterior ramus of the arteria carotis
cerebralis primitively passes through the lamina
cribrosa, from the medial surface of the olfactory
lobe, to terminate in the septum narium. The arteriae
bulbi olfactorii of Tupaia glis retain this condition,
also described for vespertilio and Vesperugo species

 

among bats (Grosser, 1901). The intracranial arteria
carotis interna is therefore the artery of the frontal
process as well as the brain.

Hofmann (1900) described the large anterior
continuation of the anterior ramus of arteria carotis
cerebralis, in Erinaceus and 13123, as the arteria
bulbi olfactorii medialis. Although most of the
branches of this vessel, in Tupaia, are concen-
trated medially, some also extend to the lateral
surface.

Tupaia, like Erinaceus and Talpa, has a

 

ramus ethmoidalis (arteria bulbi olfactorii lateralis)
given off by the arteria cerebri anterior, shortly

after exit of arteria cerebri media. This ramus

212
ethmoidalis courses craniad to end in a rete
ethmoidale lying within the dura on the lamina
cribrosa of the ethmoid bone. In Tupaia the
ethmoidal branch of the superior ramus of arteria
stapedia joins the ramus ethmoidalis in the dura
mater covering the olfactory lobe. Hofmann (1900)
described a connection of the rete ethmoidale with
vessels of the orbit, through a foramen ethmoidale,

in Erinaceus and Talpa. The rete ethmoidale sends

 

numerous small twigs into the nasal cavity with
the file olfactoria in all three genera (Tupaia,
Erinaceus, and IElEE)‘ This pattern of ethmoidal
branches is probably primitive. .A description

of these vessels in Crocodilia is quite similar
(Beddard, 1905b).

The anterior ethmoidal artery of man is
evidently the homologue of the orbital ethmoidal
ramus (Wolff, 1954; de la Torre and Netsky, 1960).
The literature does not reveal the level of primate
phylogeny at which the extension of the arteria
cerebri anterior to the nasal septum underwent

regression.

CONCLUSIONS

The primitive mammalian features observable in
the branching pattern and distribution of the cranial
arteries of Tupaia glis include the following:

1. The branching pattern of the arch of the aorta is
nearly symmetrical.

2. The lingual artery is still the major branch of
the external carotid stem. The facial and superfi-
cial temporal arteries are rather slender vessels.

3. The superior ramus of arteria stapedia is well
developed, retaining continuity with its meningeal
and orbital ramifications.

4. The pterygoid portion of the arteria maxillaris
passes through an external aliSphenoid canal.

5. The ethmoidal rami of the arteria infraorbitalis,
arteria supraorbitalis, and arteria carotis cerebralis
all remain well developed, with that of the arteria
cerebralis anterior still terminating on the nasal
septum.

6. The promontory and stapedial arteries have re-
mained about equal, both being of considerable
diameter. (The literature contains conflicting

observations of this feature, however.)

213

214

7. An anterior communicating artery is formed
between the anterior cerebral arteries. The forma-
tion of a common anterior cerebral artery, beyond
this point, is also quite primitive.

8. Well-developed arteriae bulbi olfactorii appear
as continuations of the anterior cerebral arteries
themselves.

9. The superior cerebellar arteries arise from the
branches of bifurcation of the basilar artery.

10. There is a fairly segmental arrangement of origins
of the posterior cerebral artery components, with
mesencephalic arteries arising directly, and inde-

pendently, from the posterior communicating artery.

Features of the cranial arteries of Tupaia
glis that show some similarity to the Primates are
the following:

1. There is a trend toward concentration of aortic
arch branches and shifting of their origins toward
the (right) brachiocephalic trunk.

2. There is reduction and possible loss of the
proximal inferior ramus of the stapedial artery.

3. The orbital ramus of the arteria infraorbitalis
may no longer be significant in supply of orbital
contents.

4. The internal carotid artery and its branches

course through bony canals in the medial wall of

215

the tympanic cavity.
5. The vertebro-basilar arterial system appears
to contribute to the circulus arteriosus cerebri
about equally with the internal carotid arterial
system. (On this point, there is some implied
disagreement in the literature.)

All of the features listed above, with
the possible exception of the third, have also
appeared along other mammalian lines of evolution;
thus they may also have arisen independently here.
There is good fossil evidence for this, in the
case of the fourth feature. There is no con-
vincing evidence favoring direct derivation of pri-
mates from a tupaioid stock; it does remain a possi-

bility since no features preclude such ancestry.

Since lipotyphlan ancestors and leptictids
are more tupaiid-like in dentition in the Eocene
than in the Oligocene epoch, and since early
leptictids are often difficult to distinguish from
primitive marsupials (van Valen, 1965), tupaioid
forms probably arose in Cretaceous or Paleocene times.
In many features, including cranial arteries, a
tree shrew is as primitive as any mammal. The
numerous features in which tree shrews resemble mar-
SUpials also are indicative of great antiquity.

Tree-shrew-like animals may have appeared close to

216
the time of eutherian-metatherian divergence in the
late Cretaceous period.
Although primitive mammalian features are
numerous, a few are indicated below:

1. There is a high proportion of acrocentric
chromosomes.

2. Tufts of hair are associated with scales on the
tail (Ptilocercus).

3. Seven or eight well-developed palatine ridges are
present.

4. Lytta and sublingua are both present in the
tongue.

5. The entire length of the intestine is freely sus-
pended by a median dorsal mesentery.

6. The large intestine is a short, wide, straight
tube, undifferentiated into colon and rectum
(Ptilocercus, Dendrogale, and some Tupaia species).

7. Paired precavae are present in the venous system.

8. The branching pattern of the internal carotid
arterial system closely resembles that of all mam-
malian embryos at an early post-branchial stage.

9. The suprarenal cortex lacks a zona reticularis.
10. Harder's glands are well develOped.

11. The nictitating membrane is fairly well developed.
12. Most optic nerve fibers cross at the optic

chiasma.

217
13. The anterior colliculi remain relatively very
large.
14. Amniogenesis involves folding.
15. The allantois is a very large, folded vesicle.
16. The yolk sac is well developed.
17. A pubo-ischial symphysis is present.
18. The embryonic skeleton of the hand and foot is
similar to that of lower tetrapods. Four centralia

are present in carpus and tarsus.

Seemingly close affinity of tupaiids to
marsupials is indicated by the following features:

1. The branching of the aortic arch is fairly
symmetrical.

2. There is a single (left) azygos vein, commonly.

3. The sublingua is retained.

4. The facial complement of vibrissae is almost
complete.

5. The corticospinal tract courses in the dorsal
funiculus of the upper spinal cord only, with fibers
terminating in the dorsal or intermediate gray
column.

6. The testes remain anterior to the penis.

7. Each testis is enclosed in a separate cremaster
sac. Descent into the scrotum is generally permanent.

8. There is extracranial passage of the inferior

petrosal (venous) sinus.

218
9. The vomeronasal supporting cartilages are quite
similar.
10. Palatal vacuities are generally present (absent

in Dendrogale).

 

11. A tympanic wing of the alisphenoid bone is
present and shows a posterior concavity.

12. An external aliSphenoid canal is commonly pre-
sent.

13. The combination of an enclosed orbit and a wide-
angle optic axis is present in some members of both
groups.

14. The tritubercular molar teeth of Adapisoriculus

 

and Recent tupaiids show similarity to those of
primitive marsupials. ’

15. Capitate and hamate bones, together, are larger
than navicular and lunate bones combined.

16. There is broad confluence of the sustentacular
facet with the navicular facet of the talus.

17. .A choriovitelline placenta is present during the

early gestation period.

Features of Tupaiidae that are primitive for
Eutheria include the following:
1. The annulus tympanicus lies free within the
tympanic cavity.
2. The corpus uteri is very short.

3. The depth of implantation of the embryo is

219
superficial; no decidua capsularis is present.
4. The endometrial glands remain functionally active
in the pregnant uterus.
5. A choriovitelline placenta persists during an

early stage of gestation.

The very substantial number of similarities,
whether parallelisms or not, between tupaiids and
primates seems to indicate a rather proximate common
ancestor, possibly in leptictid stock. Many of these
similarities are adaptations to an arboreal habitat
or may be the indirect fruits of that adaptation.
Some of these traits can be regarded as probable
parallelisms on the basis of available paleontolog-
ical evidence; yet there are different evolutionary
routes to an arboreal way of life, as indicated by
squirrels and phalangers. There must have been a
large number of genes and chromosomes with similar
diapositions for change in the early tupaiid and
primate stocks. Features of Tupaiidae that do not
seem highly correlated with arboreal habits and that
appear to have a common genetic basis with Primates
are the following:

1. There is a characteristic pattern of friction
ridges of the volar integument.
2. A pendulous penis is found in both groups.

3. There is retention of the combination of lytta

220

and sublingua.

4. Incipient demarcation exists between a trans-
verse and a descending colon (TuEaia).

5. .A vascular mesotyphlon is present.

6. The ratio of corneal to scleral diameter falls
within the same range, unique among mammals.

7. A free annulus tympanicus is retained.

8. There is development of a hypoconulid on the
last molar tooth (Ptilocercus).

9. A mesostyle is present on the molar tooth
(Tupaia).

10. Musculus digastricus has a well-developed central
tendon (Tupaia).

11. The naso-labial facial musculature extends into
the median frontal region.

12. Superficial orbito-auriculo-labial musculature
is present.
13. The internal oblique aponeurosis lies dorsal to
musculus rectus abdominis in its anterior portion and
lies ventral to musculus rectus abdominis in its
caudal portion.

14. .A pyriformis muscle is present.
15. There is early synostosis of the fissure of
Glaser.

16. There is reduction and possible loss (partial

or complete) of the proximal inferior ramus of

221

arteria stapedia.
l7. Cytoarchitectural patterns of the cerebral
cortex are similar.
18. Immunochemical correspondence of serum albumins
is within the prosimian range.
19. Proteins in tupaiid lenses are more similar to
those of primates than to any non-primates.

It seems evident that Special resemblances
to lemurs, as opposed to other prosimians and primates
generally, are eSpecially suspect as possible parallel-

isms.

It is the conclusion of the writer, based on
the survey of the literature and the observation
of the cranial arteries, that the current evaluation
of the phylogenetic position of TUpaiidae by many
paleontologists (Romer, 1966) is the most plausible;
hence Tupaiidae are representative of a very primi-
tive eutherian insectivore stock. This stock is
presently designated as the suborder Proteutheria
of the order Insectivora. It must be emphasized
that the order Insectivora is an exceedingly broad
category including all eutherian stem forms.
Szalay (1968) has indicated that many paleontologists
currently tend to utilize the Insectivora as a catch-
all category pending adequate fossil evidence. Until

such evidence is available it does seem unwarranted,

222
however, to set Up a separate order TUpaioidea,

as suggested by Straus (1949) and by Martin (1966).

SUMMARY

Based on evidence from a study of the internal
carotid arterial system and a survey of the literature
on tupaiid affinities, it is concluded that the family
Tupaiidae remains fairly representative of an ances-
tral stock of insectivores diverging from metatherian
stock in late Cretaceous times. Superimposed upon
this fundamentally primitive condition are numerous
arboreal adaptations and sequelae that closely
resemble those found in some Prosimii. Many of
these similarities have an apparent genetic basis;
thus they have probably arisen by parallelism from
a common stock possessing numerous genes with

similar potentialities for change.

223

LIST OF REFERENCES

LIST OF REFERENCES

Abbie, A. A. 1933. The blood supply of the lateral
geniculate body, with a note on the morphology
of the Choroidal arteries. J. Anat. 67:491-
521.

Abbie, A. A. 1934. The morphology of the forebrain
arteries with especial reference to the
evolution of the basal ganglia. J. Anat.
68:433-470.

Allen, w. F. 1905. The blood-vascular system of
the Loricati, the mail-cheeked fishes.
Proc. Wash. Acad. Sci. 7:27-157.

Andrew, R. J. 1964. The displays of the primates.
Evolutionary and Genetic Biology of Primates,
vol. II. Edited by J. Buettner-Janusch.
Academic Press, New York.

Andy, 0. and H. Stephan. 1966. Primate septum in
phylogeny. Anat. Rec. 154:310.

Arnbgck-Christie-Linde, A. 1907. Der Bau der
Spriciden und ihre Beziehungen zu anderen
Saugethieren. Morph. Jahrb. 36:463-514.

Asdell, S. A. 1964. Patterns of Mammalian Repro-
duction. Cornell Univ. Press, Ithaca, New
York.

Bachmann, R. 1958. Nebenniere. Primatologia,
Schultz, and D. Starck. S. Karger, Basel.

Bauchot, R. 1959. Etude des structures cytoarchi-
tectoniques du diencephale de Talpa europaea
(Insectivora, Talpidae). Acta Anat.

(Basel) 39:90-140.

Bautista, N. S. and H. A. Matzke. 1965. A degen-
eration study of the course and extent of
the pyramidal tract of the Opossum. J. Comp.
Neurol. 124:367-376.

224

225
Beattie, J. 1927. The anatomy of the common

marmoset (Ha ale jacchus). Proc. 2001.
Soc. London (I927):593-718.

Beddard, F. E. 1904a. Note on the brains of the
potto (Perodicticus potto) and the slow loris
(Nycticebus tardigradus), with some observa-
tions Upon the arteries of the brain in
certain primates. Proc. 2001. Soc. London
(1904):157-163.

Beddard, F. E. 1904b. On the arteries at the base
of the brain in certain mammals. Proc.
2001. Soc. London (1904):183-197.

Beddard, F. E. 1905a. A.contribution to the knowl-
edge of the arteries of the brain in the
class Aves. Proc. 2001. Soc. London (1905):
102-117.

 

Beddard, F. E. 1905b. A contribution to the knowl-
edge of the encephalic arterial system
in the Sauro sida. Proc. 2001. Soc.
London (1905 :59-70.

Bender, M. A. and E. H. Y. Chu. 1963. 'The chromo-
somes of primates. Evolutionary and Genetic
Biology of Primates, vol. 1. Edited by J.
Buettner-Janusch. Academic Press, New York.

Bhatia, M. L. and J. Dayal. 1933. On the arterial
s stem of the lizard Hemidact lus flaviviridis
R ppel. Anat..Anz. 76:417-437.

Biegert, J. 1961. Volarhaut der Hande und Ffisse.
Primatologia, Bd. II, T. 1. Edited by H.
Hofer, A. H. Schultz, and D. Starck.
S. Karger, Basel.

Biegert, J. 1963. The evaluation of characteris-
tics of the skull, hands, and feet for
primate taxonomy. Classification and Human
Evolution. Edited by S. L. Washburn.
Aldine, Chicago.

Bishop, A. 1962. Control of the hand in lower pri-
mates. Ann. N. Y. Acad. Sci. 102:316-337.

226

Bonin, G. von and P. Bailey. 1961. Pattern of the
cerebral isocortex. Primatologia, Bd. 11,
T. 2. Edited by H. Hofer, A. H. Schultz,
and D. Starck. S. Karger, Basel.

Bourne, G. H. 1936. The phylogeny of the adrenal
gland. Amer. Nat. 70:159-178.

Bourne, G. H. 1949. The Mammalian Adrenal Gland.
Oxford Univ. Press, London.

Broom, R. 1915. On the organ of Jacobson and its!
relations in the Insectivora. Part 1.
Tu aia and G nura. Proc. 2001. Soc.
Lon on (1915):ISV-163.

Buettner-Janusch, J. 1966. Origins of Man. John
Wiley and Sons, New York.

Buettner-Janusch, J. and R. J} Andrew. 1962. The
use of the incisors by primates in grooming.
Amer. J. Phys. Anthrop. 20:127-129.

Buettner-Janusch, J. and V. Buettner-Janusch. 1964.
Hemoglobins of primates. Evolutionary and
Genetic Biology of Primates, vol. II.

Edited by J. Buettner-Janusch. Academic
Press, New York.

Burda, D. J. 1966. Embryonic modifications of
lacertilian intracranial arteries. Amer.
J. Anat. 118:743-754.

Butler, H. 1964. The reproductive biolo y of a

strepsirhine (Gala o sene alensis . Int.
Rev. Gen. Exp. Zbo§., vol. 1. Edited by

w. J. L. Felts and R. J. Harrison. Academic
Press, New York.

Campbell, C. B. G. 1966. The relationships of the
tree shrews: the evidence of the nervous
system. Evolution 20:276-281.

Campbell, C. B. 6., J. A. Jane, and D. Yashon. 1966.
The retinal projections of tree shrew and
hedgehog. .Anat. Rec. 154:326.

Campbell, C. B. G., J. A. Jane, and D. Yashon. 1967.
The retinal projections of the tree shrew and
hedgehog. Brain Res. 5:406-418.

227

Carlsson, A. 1922. fiber die Tupaiidae und ihre
Beziehungen zu den Insectivora und den
Prosimiae. ,Acta‘Zool. 3:227-270.

Chapman, H. C. 1900. Observations upon the anatomy
of Hylobates leuciscus and Chiromys mada as-
cariensis. ‘Proc. Acad. Nat. SCi. PEiIageIphia
52:414-423.

Chapman, H. C. 1904. Observations on Tu aia, with
reflections on the origin of primates. Proc.
Acad. Nat. Sci. Philadelphia 56:148-156.

Colbert, E. H. 1949. Some paleontological principles
significant in human evolution. Studies
in Physical Anthropology, No. 1: Early Man
in the Far East. Edited by'w. w. Howells.
American Association of Physical Anthropolo-
gists, Detroit.

Congdon, E. D. 1922. Transformation of the aortic-
arch system during the development of the
human embryo. Carnegie Inst. Wash., Contrib.
Embryol. 14:49-110.

Davies, D. V. 1947. The cardiovascular system of

the slow loris (N cticebus tardi radus
malaianus) . Proc. 2001. Soc. Eonaon 117:
377 -410 0

Davies, D. V. 1948. The cardiovascular system of
the primates. Brit. Med. Ass., Proc. Ann.
Mtg. (1948): 371-374.

Davis, D. D. 1938. Notes on the anatomy of the
treeshrew Dendrogale. Fieldiana: 2001.
20:383-404.

Decérisy, J.-L. 1952. Remarques a propos du cercle
de des primates. Mammalia 14:213-215.

de la Torre, E. and M. G. Netsky. 1960. Study of
persistent primitive maxillary artery in
human fetus: some homologies of cranial
arteries in man and dog. Amer. J. Anat.
106:185-195.

Dobson, G. E. 1882-1890. A Monograph of the Insec-
tivora, Systematic and Anatomical. J. Van
Vborst, London.

228

Doran, A. H. G. 1879. Morphology of the mammalian
ossicula auditus. Trans. Linn. Soc. 2001.

Eckstein, P. 1958. Internal reproductive organs.
Primatologia, Bd. III, T. 1. Edited by H.
Hofer, A. H. Schultz, and D. Starck. S.
Karger, Basel.

Egozcue, J., B. Chiarelli, M. Sarti-Chiarelli, and
F. Hagemenas. 1968. Chromosome poly-
morphism in the tree shrew (Tu aia glis).
Folia Primat. (Basel) 8:150:I§8.

 

Elliot Smith, G. 1902. Notes on the brain of
Macroscelides and other Insectivora. J.
Linn. Soc. 2001. 28:443-448.

Engle, E. T. 1926. The copulation plug and the
accessory genital glands of mammals. J.
Mammal. 7:119-126.

Evans, F. G. 1942. The osteology and relationships
of the elephant shrews (Macroscelididae).
Bull. Amer. Mus. Nat. Hist. 80:85-125.

Evans, H. M. 1912. The develOpment of the vascular
system. Manual of Human Embryology, vol.
11. Edited by F. Keibel and F. P. Mall.
J. B. Lippincott, Philadelphia.

Feremutsch, K. 1965. Mesencephalon. Primatologia,
Bd. II, T. 2. Edited by H. Hofer, A. H.
Schultz, and D. Starck. S. Karger, Basel.

Fieandt, E. 1914. fiber das‘flurzelgebiet des Nervus
hypoglossus und den Plexus hypoglosso-
cervicalis bei den Saugetieren. Morph.
.Jahrb. 48:513-624.

Fiedler, w. 1956. fibersicht fiber das System der
Primates. Primatologia, Ed. I. Edited by
H. Hofer, A. H. Schultz, and D. Starck.
S. Karger, Basel.

Prick, H. 1960. Das Herz der Primaten. Primatologia,
Bd. III, T. 2. Edited by H. Hofer, A. H.
Schultz, and D. Starck. S. Karger, Basel.

Garis, C. F. de. 1941. The aortic arch in primates.
Amer. J. Phys. Anthrop. 28:41-74.

229
Garrod, A. H. 1879. Notes on the visceral anatomy

of the Tu aia of Burmah (Tu aia belan eri).
Proc. 2001. Soc. London (I579):30I-585.

Gaughran, G. R. L. 1954. A comparative study of
the osteology and myology of the cranial and
cervical regions of the shrew, Blarina
brevicauda, and the mole Scalopus aquaticus.
Mich. Univ. Mus. 2001., Miscellaneous

Publications No. 80. Univ. of Mich. Press,
Ann Arbor.

 

 

Glickstein, M., W. Calvin, and R. w. Doty. 1966.
Laminar structure of the dorsal lateral
geniculate body of Saimiri and Tupaia.
Anat. Rec. 154:348.

Glidden, E. M. and C. F. de Garis. 1936. Arteries
of the chimpanzee (Egg,spec. 7). I. The
aortic arch; II. Arteries of the upper
extremity; III. The descending aorta;

IV. Arteries of the lower extremity. Amer.
J. Anat. 58:501'5270

Goldby, F. 1941. The normal histology of the
thalamus in the phalanger, Trichosurus
vulpecula. J. Anat. 75:197-224.

 

Goodman, M. 1962. Immunochemistry of the Primates
and primate evolution. Ann. N. Y. Acad.
Sci. 102:219-234.

Goodman, M. 1963. Man's place in the phylogeny of
the primates as reflected in serum proteins.
Classification and Human Evolution. Edited
by S. L. Washburn. Aldine, Chicago.

Goodman, M. 1964. Problems of primate systematics
attacked by the serological study of proteins.
Taxonomic Biochemistry and Serology. Edited
by C. A. Leone. Ronald Press, New York.

Grassé, P.-P. 1955. Ordre des Insectivores. Traité
de Zoologie. Edited by P.-P. Grassé. Masson
et Cie, Paris.

Gregory, W. K. 1910. The orders of mammals. Bull.
Amer. Mus. Nat. Hist. 27:1-524.

Grosser, O. 1901. Zur Anatomie und Entwicklungs-
geschichte des Gefgsssystems der Chiropteren.
Anat. Hefte 17:207-424.

230

Grosser, O. 1933. Human and comparative placenta-
tion. Lancet (1933):999-1001.

Guthrie, D. A. 1963. The carotid circulation in
the Rodentia. Bu11., Mus. Comp. 2001.
Harvard Univ. 128:457-481.

Hafferl, A. 1919. Zur Entwicklungsgeschichte der
.Aorteangen und der Kopfgefgsse von Tarsius
spectrum. Morph. Jahrb. 50:19-48.

Hafleigh, A. S. and C. A. Williams, Jr. 1966.
Antigenic correspondence of serum albumins
among the primates. Science 151:1530-1535.

Haines, R. W. 1958. Arboreal or terrestrial ancestry
of placental mammals. Quart. Rev. Biol.
3331-230

Hanstr3m, B. 1958. Hypophysis. Primatologia, Bd.
III, T. 1. Edited by H. Hofer, A. H.
Schultz, and D. Starck. S. Karger, Basel.

Harms, J. W. 1956. Fortpflanzungsbiologie.
Primatologia, Ed. I. Edited by H. Hofer,
A. H. Schultz, and D. Starck. S. Karger,
Basel.

Henckel, K. o. 1928. Studien fiber das Primordial-
cranium und die Stammesgeschichte der
Primaten. Morph. Jahrb. 59:105-178.

Hendrickson, J. R. 1954. Breeding of the tree shrew.
Nature (Lond.) 174:794-795.

Hill J. P. 1965. On the placentation of Tu aia.
’ J;‘Zool., Lond. 146:278-304. "Jl‘T'

Hill, W. C. O. 1953a. The blood vascular system
of Tarsius. Proc. 2001. Soc. London 123:
655-394 0

Hill, W. C. O. 1953b. Primates. I. Strepsirhini.
Wiley, New York.

Hill, W. C. O. 1958a. External genitalia. Prima-
tologia, Bd. III, T. 1. Edited by H. Hofer,
A. H. Schultz, and D. Starck. S. Karger,
Basel.

231

Hill, W. C. O. 1958b. Pharynx, esophagus, stomach,
small and large intestine. Form and position.
Primatologia, Bd. 111, T. 1. Edited by
H. Hofer, A. H. Schultz, and D. Starck.

S. Karger, Basel.

Hill, W. C. O. and D. V. Davies. 1959. The heart
and great vessels in the Strepsirhini. Trans.
Roy. Soc. Edinburgh 63:115-127.

Hofer, H. and W. Spatz. 1963. Studien zum Problem
des Gestaltwandels des SchSdels der
gaugetiere, insbesondere der Primaten. II.

ber die K hosen fetaler und neonater
Primatensch del. Z. Morphol. u. Anthropol.
53:29-52.

Hofmann, L. von. 1914. Die Entwicklung der
Kopfarterien bei Sus scrofa domesticus.

Hofmann, M. 1900. Zur vergleichenden Anatomie der
Gehirn- und Rackenmarksarterien der vertebraten.
Z. Morphol. u. Anthropol. 2:247-322.

Horst, C. J. van der. 1949. The placentation of

Tupaia minor. Proc. Acad. Sci. Amsterdam

Hubrecht, A. A. W. 1899. fiber die Entwicklung der
. Placenta von Tarsius und Tu aja nebst
Bemerkungen fiber deren Beaeutung als
haematOpoietische Organe. Proc. 4th Int.
Congr. 2001. Cambridge.

Jane, J. A., C. B. G. Campbell, and D. Yashon. 1965.
Pyramidal tract: a comparison of two prosimian
primates. Science 147:153-155.

Jepsen, G. L. 1949. Selection, 'orthogenesis,“ fossil
record. Yearbook of Physical Anthropology,
vol. V. Edited by G. W. Lasker and C. I.
Shade. The Viking Fund, New York.

Johnson, M. L. and M. Wicks. 1964. Serum-protein
electrophoresis in mammals: significance
in the higher taxonomic categories. Taxonomic
Biochemistry and Serology. Edited by C. A.
Leone. Ronald Press, New York.

232

Kampen, P. N. van. 1905. Die Tympanalgegend des
Sgugetierschadels. Morph. Jahrb. 34:
321-722.

Kaudern, W. 1910. fiber einige Ahnlichkeiten
zwischen Tupaia und den Halbaffen. Anat.
A112. 37: " 2.

Kaufmann, J. H. 1965. Studies on the behavior of
captive tree shrews (Inpaia glis). Folia
Primat. 3:50-74.

Klaauw, C. J. van der. 1929. On the development
of the tympanic region of the skull in the
Macroscelididae. Proc. 2001. Soc. London
(1929):491-560.

Klaauw, C. J. van der. 1931. On the auditory
bulla in some fossil mammals, with a
general introduction to this region of
the skull. Bull. Amer. Mus. Nat. Hist.
62:1-352.

Klinger, H. P. 1963. The somatic chromosomes
of some primates (Tupaia glis, Nycticebus
coucang, Tarsius bancanus, Cercocebus

§*errimus, S Ihilan us 5 ndaét lus.
Cytogenetics (EaseI) 2:I%U-ISI.
Lamberton, M. C. 1947. Quelques traits de
1'évolution de la circulation endo- et
exocranienne chez les Primates. Bull.

Acad. Malgache, Tananarive, N. S. 26:
141-1420 -

 

Lange, D. de, Jr. 1933. Plazentarbildungen.
Handbuch der Vergleichenden Anatomie der
Wirbeltiere, Bd. VI. Edited by L. Bolk,
E. Gsppert, E. Kallius, and W. Lubosch.
Urban und Schwarzenberg, Berlin.

Lange, V. and J. Schmitt. 1963. Das Serumeiweissbild
der Primaten unter besonderer Berucksichtigung
der Haptoglobine und Transferrine. Folia
‘Primat. 1:208-250.

Lasinski, w. 1960. Ausseres 0hr. Primatologia,
Bd. II, T. 1. Edited by H. Hofer, A. H.
Schultz, and D. Starck. S. Karger, Basel.

233

Layton, L. 1965. Passive transfer of human atopic
allergies into lemurs, lorises, potto, and
galagos. J..Allergy 36:523-531.

Le Gros Clark, W. E. 1924a. On the brain of the
tree-shrew (Tu aia minor). Proc. 2001.
Soc. London (I923):1053-1074.

Le Gros Clark, w. E. 1924b. The myology of the
tree-shrew (Tu aia minor). Proc. 2001.
Soc. London (I924):ZEI-497.

Le Gros Clark, W. E. 1925. On the skull of Tu aia.
Proc. 2001. Soc. London (l925):559-5 .

Le Gros Clark, W. E. 1926a. On the anatomy of
the pen-tailed tree-shrew (Ptilocercus lowii).

Proc. 2001. Soc. London (1926):1179-1309.

Le Gros Clark, W. E. 1926b. The mammalian oculo-
motor nucleus. J. Anat. 60:426-448.

Le Gros Clark, W. E. 1928. On the brain of the
Macroscelididae (Macroscelides and Elephantu-
lus). J. Anat. 62:245-275.

Le Gros Clark W. E. 1929. The thalamus of Tupaia
minor: J. Anat. 63:177-206.

Le Gros Clark, W. E. 1932. A.morphological study
of the lateral geniculate body. Brit. J.

Le Gros Clark, W. E. 1936. The problem of the claw
in primates. Proc. 2001. Soc. London
(1936):1-24.

Le Gros Clark, W. E. 1959. The Antecedents of Man.
Harper and Row, New York.

Lewis, F. T. 1923. .A note on symmetry as a factor
in the evolution of plants and animals.
Amer. Nat. 57:5-41.

Lightoller, G. G. 1934. The facial musculature
of some lesser primates and a tupaia. A
comparative study. Proc. Zool.‘Soc.
London (1934):259-309.

Lineback, P. 1933. The vascular system. The Anatomy
of the Rhesus Monkey. Edited by C. G. Hartman
and W. L. Straus, Jr. Williams and Wilkins,
Baltimore.

234

Lorenz, C. F. 1927. Ueber Ontogenese und Phylogenese
der Tupajahand. Morph. Jahrb. 58:431-439.

Luckett, W. P. 1967. Morphogenesis of the Fetal
Membranes and Placenta of the Tree Shrews.
Unpublished Ph. D. thesis, University of
Wisconsin.

Lyon, M. W., Jr. 1913. Tree-shrews: an account
of the mammalian family Tupaiidae. Proc.
Us 50 Natl. MUS. 45:1'1880

Machida, H., E. Perkins, and L. Giacometti. 1967.
The anatomical and histochemical preperties
of the tongue of primates. Folia Primat.
(Basel) 5: 264- 279.

McKenna, M. C. 1961. A note on the origin of
rodents. Amer. Mus. Novitates 2037:1-5.

McKenna, M. C. 1963. New evidence against
tupaioid affinities of the mammalian family
Anagalidae. Amer. Mus. Novitates 2158:
1-160

McKenna, M. C. 1966. Paleontology and the ori in
of the primates. Folia Primat. (Basel
431-250

Maisel, H. and M. Goodman. 1963. Serological
specificity of primate lenses. Fed. Proc.
22:497.

Martin, R. D. 1966. Tree shrews: unique reproductive
mechanism of systematic importance. Science
152:1402-1404.

Meister, W. and D. D. Davis. 1953. Placentation
of primitive insectivore, Echinosorex
gymnura. Fieldiana: Zool. 35:9-26.

Meister, W. and D. D. Davis. 1956. Placentation
of the pygmy tree-shrew, Tupaia minor.
Fieldiana: Zool. 35:73-84.

Meister, W. and D. D. Davis. 1958. Placentation
of the terrestrial tree-shrew (Tupaia tana).
Anat. Rec. 132: 541- 554.

Montagna, W., J. S. Yun, A. F. Silver, and‘W. C.
Quevedo, Jr. 1962. The skin ofp imates.
XIII The skin of the tree hshrewZKTu aia
9115). Amer. J. Phys. Anth rop.

235

Mossman, H. W. 1937. Comparative morphogenesis
of the foetal membranes and accessory
uterine structures. Carnegie Inst. Wash.,
Contrib. Embryol. 26:129-246.

Mossman, H. W. 1939. The epitheliochorial placenta
of an American mole, Scalo us a uaticus.
Proc. Zool. Soc. London (I939):373-375.

Muller, J. 1935. The orbito-temporal region of

the skull of the Mammalia. Arch. Néerland.
Zool. 1:118-259.

Napier, J. R. 1961. Prehensility and opposability
in the hands of primates. vertebrate Loco-
motion, Symp. No. 5 Zool. Soc. London.
Edited by J. E. Harris. Zool. Soc. London,
London.

Napier, J. R. and P. H. Napier. 1967. .A Handbook
of Living Primates. Academic Press, New York.

Negus, V. 1958. The Comparative Anatomy and
Physiology of the Nose and Paranasal Sinuses.
E. and S. Livingstone, Edinburgh.

Noback, C. R. and J. E. Shriver. 1967. Color vision
in Tu aia and the phylogeny of color vision
in e primates. Amer. J. Phys. Anthrop.
27:241.

Ordy, J. M. and J. R. Keefe. 1965. Visual acuity
and retinal specialization in primates:
a comparison of the transitional primate
tree shrew with the diurnal rhesus monkey.
Anat. Rec. 151:394.

Owen, R. 1866. On the Anatomy of vertebrates. Vol.
11: Birds and Mammals. Longmans, Green,
and Co., London.

Packer, A. D. 1941. An experimental investigation
of the visual system in the phalanger,
Trichosurus vulpecula. J. Anat. 75: 309-329.

Parsons, F. G. 1902. On the arrangement of the
branches of the mammalian aortic arch. J.
Anat. Physiol. 36:389-399.

Pehrson, T. 1914. Beitrgge zur Kenntnis der gusseren
weiblichen Genitalien bei Affen, Halbaffen
und Insektivoren. Anat. Anz. 46:161-179.

236

Petter-Rousseaux, A. 1964. Reproductive physiology
and behavior of the Lemuroidea. Evolutionary
and Genetic Biology of Primates, vol. 11.
Edited by J. Buettner-Janusch. Academic
Press, New York.

Platzer, w. 1960. Das Arterien- und Venen-system.
Primatologia, Bd. 111, T. 2. Edited by H.
Hofer, A. H. Schultz, and D. Starck. S.
Karger, Basel.

Pocock, R. I. 1914. On the facial vibrissae of
Mammalia. Proc. Zool. Soc. London
(1914):889-9l2.

Polyak, S. 1957. The vertebrate Visual System.
Univ. of Chicago Press, Chicago.

Price, D. 1963. Comparative aspects of develop-
ment and structure in the prostate. Biology
of the Prostate and Related Tissues, Nat.
Cancer Inst. Monogr. No. 12.

Prince, J. H. 1956. Comparative Anatomy of the
Eye. C. C. Thomas, Springfield, Illinois.

Rau, A. S. and P. K. Rao. 1930. Contributions to
our knowledge of the anatomy of the Lemuroidea.
I. Arterial system of Loris lydekkerianus.
J. Mysore Univ. 4:90-121.

Rochon-Duvigneaud, A. and P. Rode. 1943. valeur
systématique de la vascularisation rétinienne
des mammiferes. Mammalia 7:56-73.

Rohen, J. W. 1962. Sehorgan. Primatologia, Bd. II,
T. 1. Edited by H. Hofer, A. H. Schultz,
and D. Starck. S. Karger, Basel.

Romer, A. S. .1966. vertebrate Paleontology. Univ.
of Chicago Press, Chicago.

Roux, G. 1947. The cranial development of certain
Ethiopian "Insectivores” and its bearing
on the mutual affinities of the group. Acta
Zool. 28:165-397.

Saban, R. 1956-1957. Les affinites du genre Tupaia
Raffles 1821, d'apres les caractéres
morphologiques de la téte osseuse. Ann.
Paléont. 42:173-224; 43:1-99.

237

Saban, R. 1963. Contribution a 1'étude de 1'os
temporal des primates. Description chez
l'homme et les prosimiens. Anatomie
comparée et phylogénie. Mém. Mus. Hist.
Nat., Paris, N. S. Zool. 29:1-377.

Samorajski, T., J. M. Ordy, and J. R. Keefe. 1966.
Structural organization of the retina in
the tree shrew (Tupaia glis). J. Cell
Biol. 28:489-504.

Sawin, P. B. and H. w. Edmonds. 1949. Morphogenetic
studies of the rabbit. VII. Aortic arch
variations in relation to regionally '
specific growth differences Anat. Rec.
105:377-397. '

Schroeder, D., D. Yashon. D. P. Becker, and J. A.
Jane. 1968. The evolution of the primate
medial lemniscus. Anat. Rec. 160:424.

Schultz, A. H. 1949. The palatine ridges of
primates. Carnegie Inst. Wash., Contrib.
Embryol. 33:43-66.

Schultz, A. H. 1958. Palatine ridges. Primatologia,
Bd. III, T. 1. Edited by H. Hofer, A. H.
Schultz, and D. Starck. S. Karger, Basel.

Schultz, A. H. 1961. vertebral column and thorax.
Primatologia, Bd. IV. Edited by H. Hofer,
A. H. Schultz, and D. Starck. S. Karger,
Basel.

Sehmsdorff, J. 1966. The primate superior olivary
complex. Anat. Rec. 154:421.

Sharma, D. R. 1958. Studies on the anatomy of
the Indian insectivore Suncus murinus.
J. Morph. 102:427-553.

Shellshear, J. L. 1930. A study of the arteries
of the brain of the Spiny anteater (Echidna
aculeata), to illustrate the principles 0
arterial distribution. Phil. Trans.
Roy. Soc. London, Ser. B. Biol. Sci. 218:
1-36.

Shindo, T. 1914. Zur verg eichenden Anatomie der
arteriellen Kopfgef sse der Reptilien. Anat.
Hefte 51:267-355.

238

Shriver, Joyce E. and C. R. Noback. 1966. Cortical
projections to the brain stem and s inal
cord in the tree shrew (Tupaia glis .
Anat. Rec. 154:423.

Sicher, H. 1912. Die Entwicklung der Kopfarterien

von Talpa europaea. Morph. Jahrb. 44:
465-4 0

'Simons, E. L. 1962. Fossil evidence relating to
the early evolution of primate behavior.
Ann. N. Y. Acad. Sci. 102:282-294.

Simpson, G. G. 1931. A new insectivore from the
Oligocene, Ulan Gochu horizon, of Mongolia.
Amer. Mus. Novitates 505:1-22.

Simpson, G. G. 1945. The principles of classifica-
tion and a classification of mammals. Bull.
Amer. Mus. Nat. Hist. 85:1-272.

Snedegar, R. 1949. Breeding of the Philippine

tree shrew, Urggale everetti Thomas. J.
Mammal. 30:194-195.

Sonntag, C. F. 1921. The comparative anatomy of
the tongue of the Mammalia. V. Lemuroidea
and Tarsioidea; VI. Summary and classifica-
tion of the tongues of the primates. Proc.
Zool. Soc. London (1921):741-755, 757-767.

Sonntag, C. F. 1922. On the anatomy of the drill
(Mandrillus leuco haeus). Proc. Zool.
Soc. London (I928):329-453.

-Sonntag, C. F. 1923. On the anatomy, physiology,
and pathology of the chimpanzee. Proc.
Zool. Soc. London (1923):323-429.

Sonntag, C. F. 1924. On the anatomy, physiology,
and pathology of the orang-utan. Proc.
Zool. Soc. London (1924):349-450.

Sonntag, C. F. 1925. The comparative anatomy of
the tongues of the Mammalia. XII. Summary,
classification and phylogeny. Proc. Zool.
Soc. London (1925):701-762.

Sorenson, M. W. and C. H. Conaway. 1966. Observa-
tions on the social behavior of tree shrews
in captivity. Folia Primat. (Basel) 4:
124-145.

239

Spatz, W. B. 1963. Zur Entwicklung des Nasenskeletts
bei Sgugetieren. Naturwiss. 50:454.

Spatz, W. B. 1966. Zur Ontogenese der Bulla
tympanica von Inpaia glis Diard 1820
(Prosimiae, Tu aiifOrmesT. Folia
Primat. (Basel 4:26-50.

Sprankel, H. 1959. Fortpflanzung von Tu aia lis
Diard 1920 (Tupaiidae, Prosimiae) In Gefangen-

schaft. Naturwiss. 46:338.

Sprankel, H. 1960. Zucht von Tupaia glis Diard
1820 (Tupaiidae, Prosimiae} in Gefangenschaft.
Naturwiss. 47:213.

 

Sprankel, H. 1961. fiber Verhaltensweisen und Zucht
von Tu aia lis Diard 1820 in Gefangenschaft.
Z. WIssen. 3001. 165:186-220.

Starck, D. 1956. Primitiventwicklung und Plazenta-
tion der Primaten. Primatologia, Bd. 1.

Edited by H. Hofer, A. H. Schultz, and D.
Starck. S. Karger, Basel.

Starck, D. 1958. Bauchraum und TOpographie der
Bauchorgane. Primatologia, Bd. III, T. 1.
Edited by H. Hofer, A. H. Schultz, and D.
Starck. S. Karger, Basel.

Starck, D. 1960. Die gussere Form der Primatenmilz.
Primatologia, Bd. III, T. 2. Edited by H.
Hofer, A. H. Schultz, and D. Starck. S.
Karger, Basel.

Starck, D. and R. Schneider. 1960. Larynx.
Primatologia, Bd. III, T. 2. Edited by H.
Hofer, A. H. Schultz, and D. Starck. S.
Karger, Basel.

Steiner, H. 1951. Die embryonale Hand- und Fuss-
Skelettentwicklung von Tu aia. verh. Schweiz.
Naturforsch. Ges. 131: - 4.

Stephan, F. 1954. Morphologie générale du systéme
circulatoire. Traité de Zoologie--Anatomie,
Systématique, Biologie. Tome XII: vertébres:
Embryologie, Grands Problémes d'Anatomie
Comparée, Caractéristiques Biochimiques.
Edited by P.-P. Grassé. Masson et Cie, Paris.

240

Stephan, H. and O. J. Andy. 1964. Quantitative
comparisons of brain structures from insec-
tivores to primates. Amer. Zool. 4:59-74.

Straus, W. L., Jr. 1929. Studies on primate ilia.
mere J. Anat. 43:403'4600

Straus, W. L., Jr. 1949. The riddle of man's
ancestry. Quart. Rev. Biol. 24:200-223.

Straus, W. L., Jr. and J. A. Arcadi. 1958. Urinary
system. Primatologia, Bd. III, T. 1. Edited
by H. Hofer, A. H. Schultz, and D. Starck.

S. Karger, Basel.

Strauss, F. 1942. vergleichende Beurteilung der
Placentation bei den Insektivoren. Rev.
Suisse Zool. 49:269-282.

Szalay, F. S. 1968. The beginnings of primates.
Evolution 22:225-230.

Tandler, J. 1899. Zur vergleichenden Anatomie der
Kopfarterien bei den Mammalia. Denkschr.
Kaiserlichen Akad. Wiss. 67:677-784.

Tandler, J. 1901. Zur vergleichende Anatomic der
Kopfarterien bei den Mammalia. Anat.
Hefte 18:327-368.

Tandler, J. 1902. Zur Entwicklungsgeschichte der
Kopfarterien bei den Mammalia. Morph. Jahrb.
30:275-373.

Tandler, J. 1906. Zur Entwicklungsgeschichte der
arteriellen Wundernetz. Anat. Hefte 31:
236'- 267 o -

Tigges, J. von. 1963. Untersuchun en fiber den
Farbensinn von Tu aia lis Diard, 1820).

Z. Morphol. u. Afitfiropol. 53:109-123.

Tigges, J. von. 1964. On visual learning capacity,
retention and memory in Tu aia lis (Diard
1820). Folia Primat. (BaseI) 3:232-245.

Tigges, J. von. 1966. Ein experimenteller Beitrag
zum subkortikalen optischen System von

Tugaia glis. Folia Primat. (Basel) 4:103-

 

241

Turner, H. N. 1848. Observations relating to some
of the foramina at the base of the Skull in
Mammalia, and on the classification of the
order Carnivora. Proc. Zool. Soc. London
(1848):63-88.

Valen, L. van. 1965. Treeshrews, primates, and
fossils. Evolution 19:137-151.

Vandenbergh, J. G. 1963. Feeding, activity and
social behavior of the tree shrew, Tu aia
lis, in a large outdoor enclosure. EoIia
mat. (Basel) 1: 199- 207.

vriese, B. de. 1905. Sur la signification morpho-
logique des artéres cérébrales. Arch. Biol.
(Liége) 21. 357-457.

vriese, B. de. 1907. Zur Entwicklungsgeschichte
der Arteriae cerebrales anteriores. verh.
Anat. Ges. 21:125-129.

Weinstein, J. D. and T. R. Hedges, Jr. 1962. Studies
of intracranial and orbital vasculature of
the rhesus monkey (Macaca mulatta). Anat.
Rees 144: 37-410

Werner, C. F. 1960. Mittel- und Innenohr.
Primatologia, Bd. 11, T. 1. Edited by H.
Hofer,.A. H. Schultz, and D. Starck. S.
Karger, Basel.

Wharton, C. H. 1950. Notes on the Philippine tree
shrew, Urogale everetti Thomas. J. Mammal.
31: 352-354.

Winge, H. 1941. The Interrelationships of the
Mammalian Genera, vol. II. Translated by
E. Deichmann and G. M..A11en. Reitzels
Forlag, Kopenhagen.

Winkelmann, R. K. 1963. Nerve endings in the Skin
of primates. Evolutionary and Genetic
Biology of Primates, vol. 1. Edited by J.
Buettner-Janusch. Academic Press, New
York.

Winkelmann, R. K. 1964. Nerve endings of the North
American Opossum (Didelphis virginiana):
a comparison with nerve endings o primates.
Amer. J. Phys. Anthrop. 22:253-258.

 

 

242

'Wislocki, G. B. 1929. On the placentation of the
primates, with a consideration of the phylogeny
of the placenta. Carnegie Inst. Wash.,
Contrib. Embryol. 20:51-80.

Wolff, E. 1954. The.Anatomy of the Eye and Orbit.
McGraw-Hill, New York.

Wolff, H. G. 1938. The cerebral blood vessels--
anatomical principles. Res. Publ. Ass._
Nerv. Ment. Dis. 18:29-68.

Wood Jones F. 1917. The genitalia of Tupaia. J.
Anat. 51:118-126.

Woollard, H. H. 1925. The anatomy of Tarsius

sgectrum. Proc. Zool. Soc. London (1925):
71-1184.

Woollard, H. H. 1926. Notes on the retina and
lateral geniculate body in Tu aia, Tarsius,
Nycticebus, and Hapale. Brain 9:77-IUZ.

MICHIGAN smn: UNIV. LIBRARIES
llHI"H‘IWIIWIMMWWW[IllWIINI‘IHWIWI
31293106731130