close

Вход

Забыли?

вход по аккаунту

?

Cranial anatomy of Ignacius graybullianus and the affinities of the Plesiadapiformes.

код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 89:477498 (1992)
Cranial Anatomy of lgnacius graybullianus and the Affinities of
the Plesiadapiforrnes
RICHARD F. KAY, J.G.M. THEWISSEN, AND ANNE D. YODER
Department of Biological Anthropology and Anatomy, Duke University
Medical Center, Durham, North Carolina 27710
KEY WORDS
Primate evolution, Paleoprimatology, Archonta,
Dermoptera
A nearly complete cranium of Zgnacius graybullianus proABSTRACT
vides increased understanding of the cranial anatomy of Plesiadapiformes. In
nearly all details of cranial anatomy, Ignacius differs markedly from primates. USNM 421608 exhibits a long tapering snout, small widely spaced
orbits, and a complete lack of postorbital process or bar. Large olfactory bulbs
are inferred from the wide interorbital space. The marked flare of the zygomatic arches suggests that Ignacius possessed large and powerful temporal
muscles. The basicranial region is particularly well preserved and reveals a
distinct suture between the petrosal bone and a n entotympanic bulla. This
suture is visible on both the left and right sides of the skull and dispels the
hypothesis that Ignacius and, by inference, other Plesiadapiformes share the
primate synapomorphy of a petrosal bulla. To test the phylogenetic position of
Zgnacius, cranial characters were identified and scored for Ignacius, Plesiadapis, Cynocephalus, and a number of primates, bats, and scandentians. Two
erinaceomorph insectivores were also included to allow the assessment of
archontan monophyly. These characters were incorporated into a maximumparsimony analysis to determine the phylogenetic position of Plesiadapiformes. There are several important phylogenetic conclusions that can be
inferred from this analysis: 1)Ignacius and Plesiadapis make up a monophyletic clade; 2) Plesiadapiformes may be the sister group of Dermoptera; 3)
Scandentia, not Plesiadapiformes, is the sister group of Primates; and 4 )
Primates. plesiadapiforms. bats, colugos. and scandentians may not form a
monophyletic clade Archonta. Consequently, the taxon Archonta is in need of
review. 0 1992 Wiley-Liss, Inc
Central to debates about primate origins
has been the Plesiadapiformes, a Holarctic
Paleogene group of mammals. Initial claims
for the primate status of Plesiadapiformes
were based on shared dental similarities,
particularly of Plesiadapis with Eocene primates. With recovery of cranial and postcranial bones, further synapomorphies have
been proposed based especially on the structure of the foot skeleton and the anatomy of
the middle ear and its vessels (see, e.g., Szalay, 1972,1975).
Arguments about the status of Plesiadapiformes (herein sensu Szalay and Delson
[19’79] construed to include [but not re0 1992 WILEY-LISS, INC.
stricted to] Plesiadapidae and Paromomyidae, but not Microsyopidae) have intensified recently with new recoveries of cranial
and postcranial material of Ignacius. Most
authors agree that the affinities of Zgnacius
are with one of the orders of Archonta (Primates, Chiroptera, Dermoptera, Scandentia), but consensus ends there. Traditionally
Zgnacius was considered a paromomyid plesiadapiform closely related to primates (see,
e.g., Szalay and Delson, 1979). Recently
Beard (1990) proposed on the basis of post~~
Received September 20, 1991; accepted June 6, 1992
478
R.F. KAY ET AL.
cranial evidence that Ignacius and other
paromomyids were the sister group of the
living flying lemur (Cynocephalus, Order
Dermoptera) and that the two major plesiadapiform families, Plesiadapidae and
Paromomyidae, do not form a monophyletic
clade. Some of the evidence for this conclusion has been challenged by Krause (1991).
Kay et al. (1990) published a preliminary
discussion of the lgnacius skull described in
this paper and also concluded that paromomyids were the sister group of Dermoptera,
but without proposing that Plesiadapidae
were to be excluded from such a relationship.
We present here the anatomical details of
a particularly well-preserved skull of Ignacius graybullianus and clarify and correct
prior cranial reconstructions of this animal
based on more fragmentary material. We examine the relations among Primates, Plesiadapiformes, Chiroptera, Scandentia, and
Dermoptera. To approach this question, we
made detailed cranial comparisons among
early eutherians, Ignacius, other Plesiadapiformes, Eocene Primates, and various living mammals, including Chiroptera,
Lipotyphla (e.g., hedgehogs and tenrecs),
Dermoptera, Primates, and Scandentia.
MATERIALS AND METHODS
TABLE 1. Taxa and specimens used in phylogenetic
analysis (Those specimens with no museum numbers are
in priuate collections)
Order Proteutheria
Asioryctes nemegetensrs (Kielan-Jaworowska, 1981)
Order Plesiadapiformes
Plesiadapis tricuspidens (per Russell, 1964)
Ignaczus graybul2ianu.s (see list in text)
Order Primates
Leptadapis magnus (per Stehlin, 1916)
Adapis parisiensis (per Stehlin, 1916)
Notharctus tenebrosus (per Gregory, 1920)
Lemur fuluus
Loris tardigradus
Tarsius hancanus
Order Scandentia
Tupaia tuna
Ptilocercus lowii [USNM (MA) 481103 and 4880681
Order Dermoptera
Cynocephalus uariegatus [USNM (MA) 317119 and
4810111
Order Chiroptera
Pteropus uanipyrus [USNM (MA) 83277 and 4775871
Nyctzmene albioenter WSNM (MA) 5432371
N major LUSNM (MA) 1246351
Macroglossus lagochilus [USNM (MA) 5431991
Rhcnopoma mzcrophyllum iUSNM (MA) 2824581
Taphozous nudiuentris LUSNM (MA) 3002101
Noctilio albiuenter (USNM (MA) 4071361
Lauza f r o m LUSNM (MA) 164921 and 4528361
Rhanolophus chaseni [USNM (MA) 352581
R euryale [USNM (MA) 4762611
Eptesicus fuscus
Myotis lucifugus
Order Lipotyphla
Echinosorex gymnurus lUSNM (MA) 4488601
Erznaceus europaeus
Tenrec ecaudatus [USNM (MA) 328589 and 3416181
Institutional abbreviations
The following abbreviations are used:
ANS, Academy of Natural Sciences, Philadelphia; MNHN, Museum
National
d’Histoire Naturelle, Paris; UM, University
of Michigan, Museum of Paleontology;
USNM (MA), National Museum of Natural
History, Smithsonian Institution, Division
of Mammals; and USNM (VP), National Museum of Natural History, Smithsonian Institution, Division of Vertebrate Paleontology.
Materials
Most of the observations reported here are
based on USNM (VP) 421608, a nearly complete skull (Kay et al., 1990).This skull was
found at Houde site 24 (near University of
Michigan fossil vertebrate site SC 1251, SW
1/4 of NW 1/4 of section 24, T56N, R102W,
Clark Quadrangle. The site is in the Cardiolophus radinskyi range zone described by
Gingerich (19911, Wasatchian, Sand Coulee
beds, Clarks Fork Basin, Park County, Wyoming. Another specimen consulted in making the reconstructions that appear here
[USNM (VP) 2147771 includes a lower jaw
and parts of the face of a n immature animal
from Houde site 14 (within UM site SC 26),
N 112 of section 4,T55N, RlOlW, from the
same range zone. Additional information
comes from UM 65569, the rostra1 part of a
skull with most of the dentition from the
Sand Coulee beds, Willwood Formation,
early Eocene, described by Rose and Gingerich (1976). Table 1lists the comparative material used in this study.
Cranial anatomy of lgnacius
The following description of the cranial
anatomy of Ignacius graybullianus is based
on USNM (VP) 421608 unless otherwise
stated. Cranial dimpnsions of this specimcn
are given in Table 2. Terminology follows
IGNACIUS AND PRIiMATE ORIGINS
TABLE 2. Skull dimensions
of Ignacius graybullianus’
Dimension
Interorbital breadth
Prosthion-nasion
Internasal suture length
Dental arcade length
Preorbital rostrum length
Dental arcade width (maximum)
Prosthion-inion
USNM(W)
421608
UM
65569
19.5
21.@
16.4
22.12
16.3
14.4
48.2’
-
21.1
28.1
-
-
‘All measurements are in millimeters. Measurements and measuring
techniques follow those of Kay and Cartmill (1977).
a Based on the skull reconstruction.
479
They may be scratch marks of a small predator or scavenger.
Dentition
The upper canine through M3 are preserved in this specimen. Dental morphology
of Ignacius graybullianus is described by
Rose and Gingerich (1976) and Bown and
Rose (1976).
Face and orbit
The facial skeleton of the new specimen IS
very similar to that described in UM 65569
(Rose and Gingerich, 1976). The snout
tapers markedly rostrally. The nasals are
long and convex mediolaterally. They widen
State of preservation
caudally to form a broad contact with the
The skull (see stereophotographs in Figs. frontals at a point caudad to the anterior
1 4 and reconstruction in Fig. 5) was re- orbital margin and at a similar level a s the
moved from a limestone matrix by acid etch- maxillofrontal suture. Although the preing. It is slightly crushed in the interorbital maxillae were lost before fossilization, their
region, with the face displaced dorsally rela- outlines and sutural contacts are clear. The
tive to the braincase. The right and left pre- premaxillae overlapped the nasals medially
maxillae are not preserved. The medial and butted against the maxillae caudally.
sides of both orbits are badly fragmented They tapered caudally and are widely sepaand their sutures obscured. The zygomatic rated from the frontals by a maxillonasal
arches, dorsal cranial vault, and occipital re- contact. This is unlike the condition region are complete. The midline basal ele- ported in UM 65569 where the premaxilla
ments of the skull and lateral pterygoid contacts the frontal (Rose and Gingerich,
plates are well preserved, as are the bony 1976).Reexamination of this specimen leads
walls of the middle ear and the articulations us to conclude that the location of the suwith surrounding bones. The premaxilla tures in this area is confounded by cracks in
was reconstructed in Figure 5 based on UM the critical areas.
The maxilla makes a broad contribution
65569 (Rose and Gingerich, 1976). The
lower jaw is reconstructed based on USNM to the orbital margin and wall. Its sutural
(VP) 214777, and the palate is well pre- contacts in the orbit are obscured by breakserved on the skull. The canines and cheek age. The maxilla excludes the lacrimal from
teeth are preserved on both sides. The post- contact with the zygomatic. The suture becanines were acid etched during prepara- tween the maxilla and zygomatic is broad
tion, slightly obscuring some anatomical de- and extends obliquely from rostrodorsal to
tail. The teeth appear worn due to slight caudoventral. There is a large infraorbital
acid etching, leaving the mistaken impres- foramen dorsal to P3.The orbital foramen of
sion that the skull was of a n old individual, the infraorbital canal is roofed by a shelfwhich clearly is not the case, as the distinct- like process of the maxilla. The lacrimal is
ness of the cranial sutures demonstrates. confined to the antorbital rim and lacks a
Part of the right auditory bulla was removed distinct facial process. The lacrimal has a
in the course of preparation to reveal details rugose orbital rim with a lacrimal tubercle.
The lacrimal foramen opens posterolaterally
of the middle ear (Figs. 6,7).
A number of more or less perpendicular within the orbital rim. A foramen medially
scratches occur on the dorsal aspect of the adjacent to M3 could be the caudal palatine
right parietal. These match a similar set on or sphenopalatine foramen, which would
the orbital aspect of the same bone (Fig. 8). mark the presence of the palatine in the or-
the Nomina Anatomica Veterinaria, a s cited
in Miller2 Anatomy of the Dog (Evans and
Christensen, 1979).
480
R.F. KAY ET AL.
Fig. 1. Ignacius graybullianus [USNM (VP) 4216081. Stereopair of skull in ventral view
bit. A similar foramen occurs in UM 65569. ble between the foramen ovale and the orHowever, the sutural contacts of the pala- bital fissure, indicating that neither a n alar
canal nor a foramen rotundum was present.
tine cannot be observed in either specimen.
The orbits are widely separated by a broad The latter is probably confluent with the orinterorbital space. The orbit was small but bital fissure. A foramen rotundum has been
its precise limits are unclear. Laterally and described for Plesiadapis (Russell, 1964) but
ventrally, the raised orbital margin merges we believe that this foramen in Plesiadapis
smoothly with the sharply edged dorsal (MNHN CR 965) is more likely the suboptic
margin of the zygomatic arch. Medial and fnramen. The alisphenoid IS not strongly in
dorsal t o the lacrimal tubercle, the orbital Slated laterally in this region, implying that
margins merge smoothly with temporal the middle cranial fossa was relatively
crests. There is no postorbital process o r bar. small. No optic foramen is visible in USNM
There is no trace of a n orbital process of the (VP) 421608 but a small foramen, possibly
zygomatic bone.
the optic foramen, is present in the ventroThe precise limits of the bones of the or- caudal part of the orbit on UM 65569. The
bital wall are not discernible in USNM (VP) small size of this foramen in Ignacius would
421608. However, the absence of sutures in imply that the eye must have been small like
well-preserved orbital fragments when that of Erinaceus and much smaller than
other cranial sutures are so clearly defined seen in living or fossil primates. A foramen
suggests that the frontal must have ex- that probably transmitted the frontal diploic
tended far down the lateral wall of the vein occurs in the right orbit of UM 65569;
braincase in the infratemporal fossa and breakage obscures this region in USNM
probably contacted the maxilla broadly, as (VP) 421608.
in Plesiadapis (Russell, 1964). The foramen
ovale is lateral t o the root of the lateral Neurocranium
pterygoid plate and faces ventrally. Further
anteriorly is a large, oval, forwardly directed
The postorbital constriction is extreme in
orbital fissure. There are no foramina visi- the region of the frontal just rostra1 t o the
IGNACIUS AND PRIMATE ORIGINS
481
Fig. 2. Ignacius graybullianus [USNM (VP) 4216081. Stereopair of palate and basicranium from
ventrolateral view
back of the skull, where it is confluent with
sharply raised nuchal crests. The point a t
which the temporal lines converge appears
to be further rostra1 than in UM 65569.
Above and below the suture between parietal and squamosal on the side wall of the
braincase are several vascular foramina
that open into shallow, caudally directed
surface grooves. These probably connected
with intracranial dural venous sinuses. A
large suprameatal foramen (Novacek, 1986)
is present on the lateral side of the zygomatic arch above the mandibular fossa, posMasticatory anatomy
sibly for a vein that was confluent with the
The zygomatic arches are board dorsoven- intracranial venous sinuses.
trally and bowed laterally. The zygomatic
The mandibular fossa of the squamosal is
articulates with the squamosal along a n ob- large, transversely concave, and somewhat
lique suture. Ventrally the suture reaches elongate mediolaterally. Medially, the
caudally to the caudolateral border of the squamosal laps onto the craniolateral side of
mandibular fossa. Lateral flaring of the zy- the auditory bulla. The mandibular fossa is
gomatic arches indicates that the temporal bordered caudally by a weak retroarticular
muscles were large. Temporal lines con- (=postglenoid) process and the cranial parts
verge strongly to form a single sharp mid- of the auditory bulla and external auditory
sagittal ridge beginning on the frontal bone meatus. A well-developed retroarticular
a t the level of the maximum postorbital con- (=postglenoid) foramen occurs medial to the
striction. This crest extends distally to the retroarticular process. The meatal surface of
suture between the frontal and parietal.
This marks the level of the most posterior
extension of the olfactory bulbs and the most
anterior extension of the cerebral hemispheres. The olfactory bulbs were probably
large, as indicated by the great breadth between the orbits.
The cranial cavity is partly filled by matrix. However, a n ossified tentorium cerebelli is visible through the foramen magnum, as is a large subarcuate fossa on the
left side.
482
R.F. KAY ET AL.
Fig. 3. lgnacius graybullianus [USNM (VP) 4216081. Right: Skull in dorsal view. Left: Facial skeleton
in coronal view.
Fig. 4. Ignucius gruybullianus [USNM (VP) 4216081. Stereopair of braincase in posterolateral view
the squamosal is extremely short mediolaterally. The squamosal laps caudally onto the
mastoid process.
The palate is broad between the molars
and lacks palatal fenestrae. The palate is
delineated by a sharply raised edge rostral
to the cheek teeth that extends rostral and
medial to a small single-rooted canine. Several transversely oriented palatal ridges are
visible in front of the cheek teeth. The palatine bone reaches forward to P4. Palatal processes extend distally from the palate behind M3, bordering the greater palatine
notch. On the left side of the skull, this notch
is bridged by the broken and displaced lat-
eral pterygoid plate. Anterior and middle
palatine foramina are indistinct. The posterior margin of the palate is nearly straight
and buttressed by a strong rounded postpalatine torus.
The vomer is extended caudally into a
high midline keel in the fossa between the
medial pterygoid plates. The sphenoidal
processes of the palatine are strong. Medially they are continuous with short and
poorly defined medial pterygoid processes.
Strong and deep lateral pterygoid plates occur laterally. On each side they reach caudally t o the medial side of the foramen ovale
and the lateral side of the auditory bulla.
IGNACIUS AND PRIMATE ORIGINS
483
Fig. 5. Reconstruction of t h e cranium and mandible of Ignaczus graybullzanus based on USNM (VP)
421608 and USNM (VP) 214777. A Lateral view. B: Dorsal view. C: Ventral view.
The lateral pterygoid plates diverge widely,
producing large fossae for the medial pterygoid muscles. A foramen is present in the
lateral pterygoid plates opposite the foramen ovale, which may have transmitted the
nerves to the medial pterygoid muscle.
Basicranium and braincase
The auditory bulla and circumbullar
bones and foramina are well preserved. The
bulla is large and completely ossified. It is
continuous laterally, with a fully ossified ec-
R.F. KAY ET AL.
484
,
Foramen ? for newes to
medial pterygoid muscle
Gutter for "auditory" tube
Promoiitoriuni
Epityiupanic recess
PeLrosal-entotympanic suture
Tubular ectotympanic
Promontory shield
Entry for Carotid foramen
Hypoglossal foramen
Fig. 6. Ignacius graybullianus [USNM (VP) 4216081, ventral view of right auditory bulla and basicranium
totympanic that is prolonged laterally into
an ossified external auditory meatus. The
ectotympanic is fused to the bulla, but parts
of its sutures are visible. In Plesiadapis, the
middle ear cavity is prolonged laterally ventral to the ectotympanic to form a subtympanic recess (Fig. 15B in MacPhee and
Cartmill, 1986), whereas, in Ignacius, the
tympanic ring is fused with the bulla, and no
subtympanic recess is apparent. The bulla
extends medially close to the midsagittal
plane. Caudally, it balloons out over the basioccipital and obscures the lateral parts of
the occipital condyles. The bulla is composed
of an entotympanic bone and lacks basisphenoid, basioccipital, ectotympanic, or petrosal
contributions (Fig. 6). Three observations
indicate this. 1) Sutures separate medial,
anterior and posterior bullar walls from the
basisphenoid and basioccipital. 2 ) Rostrally
and caudally, partial sutures are visible between the ectotympanic (which forms the
external auditory meatus) and the bony
bulla. The more cranial of these shows a
small depression for the tympanohyal. 3) In
both bullae, on the roof of the tympanic cavity, the promontorial part of the petrosal is
separated from the bulla medially by a suture (Fig. 7 ) . The petrosal partly overlaps
the bone of the bulla in this area. Both petrosal and entotympanic thin out at their junction rather than butting up against one another as might appear in postmortem
breakage. The position of the suture is identical on right and left sides. These observations suggest that there is a true suture be-
ZGNACZUS AND PRIMATE ORIGINS
Fig. 7. Scanning electron micrograph of a cast, of the
inside of the right auditory bulla of USNM (VP) 421608
showing part of the right promontorium (petrosal) and
entotympanic bones. The orientation is similar to that in
Figure 6. The ventral overlap of the petrosal onto the
Fig. 8. Scratch marks on the right side of USNM
(VP) 421608. See text for discussion, Midsagittal line is
towards top of page.
tween the petrosal and bulla and that the
auditory bulla is composed of the entotympanic.
485
entotympanic is matched by a depressed area on the
central surface of the entotympanic. This and the thinning of the pretrosal edge indicate that this is a suture
rather than a crack. The overall morphology of this suture is the same in the left auditory bulla.
The principal blood supply to the brain
was probably from the vertebral arteries via
the foramen magnum. A significant supply
from the internal carotid artery (ICA) is
ruled out by the small size of the carotid
foramen. This foramen is located ventromedial to the “dual” stylomastoid foramen (see
below) on the caudolateral side of the bulla
just medial to the suture between the entotympanic and ectotympanic (Fig. 9). The diameter of the right carotid foramen of
USNM 421608 is -0.17 mm, comparable to
that or lorises (Table 3, Fig. 10) in which the.
internal carotid artery degenerates during
development and only the plexus of nerves
that travels with the internal carotid artery
is enclosed by a bony canal. In contrast, in a
‘pecirnen Of Erinaceus Of
length, the diameter of the internal carotid
groove that transmits a patent ICA is 0.43
mm. The small size of the foramen in Ignacius suggests that it transmitted only the
nerve plexus (Fig. 10).
R.F. KAY ET AL.
486
Mastoid foramina
al stylomastoid foramina
Suprameatal foramen
Entotympanic
'
I
Entotympanic-
ectotympanic suture
Fig. 9 Ignacius graybullianus (USNM VP 421608). Right side of neurocranium, auditory bulla a n d
zygomatic a r c h viewed posteriorly.
T A B L E 3 . Internal diameter of carotid foramen in selected primates and other mammals
Taxon
Specimen
Carotid foramen
Skull length
~~~
Ignacius sp.
Erinaceus sp.
Tupaia glis (n = 3)
Tupaia tuna
Nycticebus coucang
Perodicticus potto
Galago senegalensis
Galago demidouii
Lemur fuluus
Lemur sp.
USNM (VP) 421608
Cartmill coll.
Cartmill coll.
Cartmill coll.
USNM (MA) 488076
USNM (MA) 450056
Cartmill coll.
Cartmill coll.
Cartmill coll.
Kay coll.
Tarsius sp.
CaZlithrix argentata
Callicebus sp. (Bolivia)
Saguinus mystax
Saimiri sciureus
Aotus tnuirgatus
Cebus apella
Pithecia pithecia
Ateles geoffroyi
Alouatta pigra
Kay coll.
USNM (MA) 239457
A N S 14306
USNM (MA) 397877
USNM !MA! 398676
USNM [MA) 396724
USNM (MA) 261319
USNM (MA) 545891
USNM (MA) 211240
USNM (MA) 544508
0.17
0.43'
0.62
0.81
0.453
0.18
0.20
0.14
0.68
0.75
0.56
0.74
1.05
0.99
1.24
1.30
2.00
1.37
2.42
2.50
~
48.2
45.0
49.1
59.4
59.2
61.9
46.8
37.3
87.3
83.7
39.3
45.7
59.0
50.7
59.7
62.7
95.5
82.6
102.9
104.4
'All foramen measurements were made using Wild M5 microscope fitted with an eyepiece reticle a t either 25 x or 50 x magnification.
'This is a measure ofthe breadth ofthe internal carotid sulcus immediately proximal to Its bifurcation into stapedial and promontory arteries
'The internal carotid foramen 1s slit-like; the maximum diameter is 0.45 mm.
A significant blood supply to the brain
from the ascending pharyngeal artery, proposed by MacPhee et al. (19831, is also unlikely. There is no groove on the basioccipital or basisphenoid along the cranial base
that would indicate a n extrabullar course
(sensu Wible, 1984) for ICA, nor is there a
petrooccipital fissure (=middle lacerate foramen) to transmit the ascending pharyngeal artery from the neck to the inside of the
braincase as in lorises (Cartmill, 1975). Foramina and grooves identified by MacPhee
et al. (1983) for these structures may be for a
ventral petrosal venous sinus extending intracranially in the fissure between the medial side of the petrosal and the basisphenoid and basioccipital (Szalay et al., 1987).
The carotid foramen is situated lateral to
the middle ear cavity so that the sympathetic nerve plexus associated with the invo-
IGNACIUS AND PRIMATE ORIGINS
I
.
,
I6
, .
.
,
17
.
,
1.8
,
19
.
.
.
.
2
Log Skull length
Fig. 10. Lug-lug bivariate plot of size of the carotid
canal versus skull length based on the taxa listed in
Table 3. The line above the points represents a slope of
isometry (1.0). Filled circles represent small to mediumsized anthropoids, Tarsius, and lemurs; open squares
represent Erinaceus and two species of Tupaia; open
circles represent four galagines and lorisines. Ignacius
is the solid circle within the lorisid cluster. A polygon is
drawn around species whose carotid canal transmits an
internal carotid artery (above); another polygon is
drawn around the lorises and galagos whose carotid foramen transmits only a sympathetic nerve plexus in the
adult.
luted internal carotid artery entered the
bulla posterolateral to the fenestra cochleae.
The course of the carotid canal close to the
carotid foramen is obscured by matrix and a
lamina of bone that projects ventrolaterally
from the promontorium obscuring also the
fenestra cochleae in ventral view. It is possible that this lamina formed ontogenetically
when a medially expanding middle ear cavity enveloped the short bullar portion of the
carotid canal. This distinctive feature was
previously reported for Ignacius, Plesiadapis, and Phenacolemur (Szalay et al., 1987).
The free edge of the lamina may contain a
small carotid canal in Plesiadapis (Szalay
et al., 1987; but see MacPhee and Cartmill,
1986). Matrix obscures the lateral aspect of
the promontorium, and it is unclear if there
is a stapedial groove. Medially, there is no
groove on the promontorium for a promontorial artery.
A septum of bone (longitudinal septum;
Szalay, 1972) stretches rostrally from the
promontorium to a point just medial to the
entrance of the auditory tube. Szalay (1972)
suggested that the promontorial artery may
487
have traveled through a similar septum in
Phenacolemur. However, this septum shows
no lumen rostrally where it is broken and is
thin caudally. Phenacolemur also lacks a lumen (MacPhee and Cartmill, 1986). Moreover, in Ignacius, rostrally this longitudinal
septum projects laterally away from the position of the cerebral arterial circle where
the promontory artery commonly ends.
In Ignacius and other Plesiadapiformes,
the auditory bulla is extremely large. Right
and left bullae are separated medially by
only 1 4 mm. and the basisphmoid and basioccipital bones are narrow. Caudally, the
bulla extends ventrally over the jugular foramen and hypoglossal foramen. The promontorium is laterally positioned within the
auditory bulla owing to the enormous medial expansion of the hypotympanic cavity
a s in Phenacolemur, Plesiadapis, and
adapids (Szalay et al., 1987). Plesiadapiformes differ from adapids in having a narrow basisphenoid across which the right and
left bullae nearly touch, whereas in adapids
the basisphenoid is much broader and the
bullae are widely separated (compare Fig. 1
ofIgnacius with Adapis, Fig. 8, in Gingerich
and Martin L198ll).
A single hypoglossal foramen is adjacent
to the occipital condyles. A large jugular foramen is posterior to the bulla and lateral to
the hypoglossal foramen. The former transmitted the internal jugular vein and cranial
nerves IX, X, and XI. Further lateral is a
recess that contains two foramina. One of
these is the stylomastoid foramen, and the
second is associated with a vascular trace. It
could be the opening for the stylomastoid
artery connecting the occipital or posterior
auricular artery with the ramus posterior of
the stapedial artery (Wible, 1987). A similar
“dual” stylomastoid foramen is reported in
Plesiadapis (Russell, 1964). Mastoid foramina are present dorsolaterally in the nuchal
plane between the mastoid and occipital.
This foramen may have transmitted the
large diploic artery (Wible, 1987) or a vein
(Novacek, 1986). Either or both the stylomastoid artery and the large diploic artery
could have served as a n additional source of
blood for the brain or other areas supplied
by the distal branches of the stapedial artery.
488
R.F. KAY ET AL.
The braincase bears several crests and
surfaces for the attachment of the nuchal
muscles. A midline sagittal crest merges
with strong nuchal crests on the parietal
bones. Laterally, the occipitals extend onto
the lateral wall of the braincase from the
nuchal crests. Still further laterally, the parietals again reach caudad to the nuchal
crest. Here, they contact broad nuchal exposures of the mastoids and do not extend on
the caudal aspect of the skull.
The foramen magnum is broadly oval and
faces directly posteriorly. The diameter of
the foramen magnum is 4.8 mm midsagittally and 6.8 mm mediolaterally. Strong
paroccipital processes occur immediately
adjacent to the dorsal occipital facets.
formal phylogenetic analyses because the focus of our study is the skull of Ignacius. The
outcome of the analysis is thus based only on
a portion of the available evidence for most
taxa. Our results for the affinities of all
other taxa included in our analysis remain
tentative and are open to testing on the basis of noncranial material.
Version 3.0 of the PAUP program (Swofford, 1989) was used to analyze the data matrix of Table 4 based on the characters of
Appendix A. Because the full character/
taxon matrix exLeedo ihe limits of the
exhaustive search and branch-and-bound
search algorithms of PAUP, we approached
the analysis in several steps.
1. As a starting point for the analysis, we
accept the monophyly of the following
To investigate the phylogenetic position of groups: Primates, Scandentia, MicrochiIgnacius, we surveyed the cranial anatomy roptera, Megachiroptera, and Erinaceomorof extant and some fossil archontans and pha in accordance with the consensus view
lipotyphlan insectivores (Table 1).We based current in the recent literature of the monoobservations of extant taxa on original ma- phyly of these clades. Next, we condensed
terial but relied primarily (but not exclu- the taxa composing each of these clades into
sively) on the descriptions of Gregory (1920) a single representative. The scores for these
for scoring Notharctus, Stehlin (1912) and hypothetical ancestral representatives were
Forsyth-Major (1901) for scoring Adapis, those character state distributions at each of
and Russell (1964) for scoring Plesiadapis. the basal nodes in a n initial heuristic
To determine if Ignacius is a member of Ar- search. The character matrix for these hypochonta, we also studied two erinaceomorph thetical taxa (HTUs) is given in Table 5. In
insectivores (Erinaceus and Echinosorex) addition to the five clades mentioned, Chiroptera (Megachiroptera and Microchiand a tenrecomorph (Tenrec).
We chose Asioryctes a s the single out- roptera) and Lipotyphla ( Erinaceomorpha
group for our analysis because it is cranially and Tenreci were also found t o be monophjilwell-known and was probably close to the etic in the heuristic analysis, but we did not
ancestry of all later placental mammals use condensed HTUs for these possible
(Kielan-Jaworowska, 1984). Several skulls clades in later analyses because their monohave been described ofAsioryctes and closely phyly has been challenged in the literature
related Kennalestes (Kielan-Jaworowska, (see Fig. 12).
2. An exhaustive search was executed on
1981). These taxa are the only proteutherians for which adequate cranial material is Ignacius, Cynocephalus, and Tenrec, toknown. Palaeoryctids, once considered to be gether with HTUs representing Erinaceoancestral placentals, are now generally morpha, Primates, Scandentia, Microchithought to be lipotyphlan insectivores (Mc- roptera, and Megachiroptera. This search
Kenna et al., 1988; Butler, 1988; Thewissen resulted in a single most-parsimonious claand Gingerich, 1989). The great antiquity of dogram of 61 steps and a consistency index
Asioryctes (Santonian-Coniacian, late Cre- of 0.69 (Fig. 11).We found 21 cladcgrams of
taceous) is consistent with its position as 62 steps.
3. Plesiadapis was included in a second
close to the ancestry of placentals.
Each of these taxa was scored for 33 cra- exhaustive search resulting in ten most-parnial characters (Table 4). Dental and post- simonious cladograms of 67 steps and a concranial characters do not form part of the sistency index of 0.59. Plesiadapis and IgPHYLOGENETIC ANALYSIS
0 0 ~ 0 0 0 0 ~ 0 3 - ~ 3 0 0 0 0 ~ - 0 0 0 0
-- -
0 0 0 ~ 3 3 3 3 ~ 1 0 0 0 0 0 0 0 0 0 0 0 0 0
0
3
3
0 0
3
m
3 C . 3
0 0 0 3 3 ri
i 0 4 c. m c?
---
ri
0 3
3
r(
3
o i3
3 3 3 3 3
NC. N 3
3
ri
~ 3 0 0 0 1 r i 3 0 0 3 ~ 0 0 ~ 0 3 3 3 0 ~ 3 ~
c.e.3
-
3 ri ri
c-. 0 0 0
0 0
000 0 0 3 0 0
0 0 0 - 0 0 0 0 0 - 0 0 0 0 0 - 0 0 0 0 0 0 0
13
-
3 4 3
0 ri
0 0 0 0 0 3 0
ri
a
w I
0 0 0 r- 0 ?.
c-,
3
000
3
0 0 00
c-.
C.
c. 0
0 0
c
C-.
c-0 c . 0 e -o c -0 0 0
0
C. 0 0 0
- -
0 0
c
0 N N N N N N N N
c.
3 ri ri
N N N 3 C - N 3 3 0 ri 0
e -0
3 3 3 i
3
R.F. KAY ET AL.
490
,Ignacius
P
i c 0 0 0 -
ori100c
0 0 O O 0 t
Cynocephalus
Tenrec
%
Erinaceomorpha
Primates
Scandentia
- 0 c 0 0 c
13000c
,
,Microchiroptera
Y Megachiroptera
mm.lmi-*
1 0 0 0 o c
0 3 0 - 1 c
Fig. 11. Maximum-parsimony cladogram of the
characters in Appendix A. This cladogram is discussed
in the text as analysis 2.
0 0 0 0 3 c
0--00c
1
1
d
d
N
C
0 0 0 - o c
0 - c 0 0 c
1 0 0 0 o c
C N N r i N C
0 3 0 0 c c
O N N O - I C
0 0 0 0 0 -
0 0 0 0 o c
c c,
--
-
i
oc000r
3 0 0 - 1 c
d d - - N N
3 0 0 0 0 0 1 i i 0 0
0 0 0 0 - 0
0 0 0 0 0 0
0 0 0 0 1 0
0 3 1 1 1 1
nacius formed a monophyletic clade in nine
of these ten most parsimonious cladograms.
In the remaining cladogram they formed a
paraphyletic base taxon to all other included
groups except Dermoptera. There seems to
be overwhelming evidence for monophyly of
Paromomyidae and Plesiadapidae to the exclusion of Cynocephalus. This is borne out
not only by the cranial evidence subsumed
within our analysis but also by dental synapomorphies described in the literature but
not used in our analysis (Gingerich, 1976;
Kay and Cartmill, 1978; Szalay and Delson,
1979). Therefore, a n additional HTU was
created from the character distribuliuris "1'
the basal node that unites Plesiadapiformes.
Using this HTU with the others mentioned
in this paragraph, a further PAUP exhaustive search yields a single most parsimonious cladogram of 59 steps and consistency
index of 0.71. This cladogram is identical to
the cladogram in Figure 11 (with Plesiadapiformes substituted for Ignacius) and
represents in our opinion the best supported
hypothesis of the relations of Ignacius and
Plesiadap is.
4. Several hypotheses have been proposed for the relations among Archonta.
These are tested with our data by calculating the minimum number of transformations based on our character matrix when
these tree topologies are forced. We evaluated the cladograms proposed by Gregory
IGNACIUS AND PRIMATE ORIGINS
C ynocephalus
Microchiroptera
Megachiroptera
Primates
Scandentia
Tenrec
Erin aceomorph a
Gregory, 1910
length: 56
shortest length: 54
Microchiroptera
Megachiroptera
Scandentia
Erinaceomorpha
Primates
Miyamoto and Goodman, 1986
length: 46
shortest lenght: 39
49 1
Cynocephalus
Microchiroptera
Megachiroptera
Primates
Scandentia
Tenrec
Erinaceomorpha
Novacek and Wyss, 1986
length: 56
shortest length: 54
Cynocephalus
Megachiroptera
Primates
Scandentia
Microchiroptera
Pettigrew et al., 1989
length: 48
shortest length: 41
Fig. 12. A test of four published cladograms representing the phylogenetic relationships proposed by
Gregory (1910; Fig. 31), Novacek and Wyss (19861, Miyamoto and Goodman (19861, and Pettigrew et al.
(1989).Each cladogram is presented as a user tree in PAUP, and the characters in Table 4 are plotted onto
it. Below each is given the number of steps (length)of each tree compared with the shortest tree using our
characters and the taxa presented.
(1910), Novacek and Wyss (19861, Miyamoto
et al. (1986), and Pettigrew et al. (1989; Fig.
8). These authors did not include all of the
same taxa in their analyses a s we did, so we
also executed a n exhaustive search for the
most parsimonious cladograms for their
subsets of taxa. Cladograms, numbers of
steps, and consistency indices are summarized in Figure 12.
DISCUSSION OF CLADISTIC ANALYSIS
We will restrict our discussion to the single most parsimonious cladogram in the exhaustive search without Plesiadapis (search
2), the 21 cladograms of the same analysis
that were one step longer, the ten most parsimonious cladograms of the analysis that
included Plesiadapis (search 3), and the single most-parsimonious cladogram based on
the HTU for Plesiadapiformes.
Our most significant results are that 1)
Ignacius and Plesiadapis are sister taxa, 2)
Plesiadapiformes may collectively group
with Dermoptera, and 3) Ignacius and Plesiadapis are never linked with Primates. We
will also discuss 4)the sister-group relation
of Primates and Scandentia and 5) the concept of Archonta.
R.F. KAY ET AL
492
lgnacius and Plesiadapis form a
clade, Plesiadapiformes
ply the brain and that the ventral petrosal
sinus is endocranial. Ignacius does not differ
in these features from Plesiadapis.
One potential difference between Ignacius
In nine of ten cladograms of search 3, Plesiadapis was the sister group to Ignacius, and Plesiadapis is the composition of the
whereas, in the remaining cladogram, these bulla. The auditory bulla (character 24) of
two taxa formed a paraphyletic root group to Ignacius is completely ossified by the entoall taxa in the analysis (except Cynoceph- tympanic bone (interpreted a s a synapomoralus). This corroborates the monophyly of phy of Ignacius and Plesiadapis in 11 of 22
the two core families of Plesiadapiformes: cladograms). Plesiadapis reportedly has a
Plesiadapidae and Paromomyidae. Depend- petrosal bulla, but the detailed similarities
ing on which of the ten trees is selected (each between the two taxa in the middle ear arvarying in the remaining topology of the cla- teries hugged that Plesiadapis actually had
dogram), the following are possible shared a n entotympanic bulla. If the bulla of Plederived features of Plesiadapiformes: maxil- siadapis is entotympanic, then the suture
lary-frontal contact in orbit (character 6; between petrosal and entotympanic must
two of ten cladograms); suboptic foramen have been obliterated-a common phenomepresent (character 16; nine cladograms); os- non in ontogeny (MacPhee and Cartmill,
sified external auditory meatus (character 1986).
23; eight cladograms); promontory artery
absent (character 25; eight cladograms); staPlesiadapiformes are the sister
pedial artery absent (character 26; eight clagroup of Dermoptera?
dograms); mastoid tubercle strong (character 29; one cladogram).
Our analysis weakly supports the notion
Several other characters not used in the of a monophyletic Plesiadapiformes as a sisanalysis appear to further support the hy- ter group to Dermoptera but does not suppothesis of plesiadapiform monophyly. Ad- port the hypothesis of Beard (1990) that
ditional cranial similarities of Ignacius to Paromomyidae (but not Plesiadapidae) and
Plesiadapis, the cranially best known paro- Cynocephalus (Order Dermoptera) are sismomyid and plesiadapid respectively, in- ter groups and its corollary that Plesiadapiclude the carotid foramen being situated in a formes is a paraphyletic group. Three basic
posterolateral position of entry into the patterns emerge from our analyses of the
bulla in both taxa and the fenestra cochleae relations of Plesiadapiformes. In the analybeing shielded ventrally by a bony shelf (see, sis without Plesiadapis (search 21, Dere g , Gingerich, 1976; Kay and Cartmill, moptera is the sister group ofIgnacsus in ? I
1977; Szalay and Delson, 1979; and Szalay of the 22 most-parsimonious cladograms (toet al., 1987).Ignacius and Plesiadapis, as do pology A) or of Lipotyphla in nine claall other known paromomyids and ple- dograms (topology B). In the remaining two
siadapids, also share apomorphies of the cladograms, Ignacius is the primitive sister
dentition including the presence of a Nanno- to a clade consisting of all other taxa except
pithex fold on the upper molars, reduction in Cynocephalus (topology C). Topologies A
the number of anterior teeth and a n en- and C occur only once in the analysis that
larged lanceolate lower incisor (Gingerich, included Plesiadapis, while topology B was
1976; Kay and Cartmill, 1977; Szalay and common (seven cladograms). Thus these results are ambiguous with respect to the conDelson, 1979).
Moreover, the skull oflgnacius, unlike the clusion of Kay et al. (1990) that Plesiadapiprevious reconstruction suggestions of formes and Dermoptera are closely related.
MacPhee et al. (1983), demonstrates that
Possible characters supporting monothe arteries of the basicranium are very sim- phyly of Plesiadapis, Ignacius, and Cynoilar to those of Plesiadapis. The new mate- cephalus are ossified auditory meatus (charrial demonstrates that Ignacius has no acter 23; 11 of 22 cladograms), promontory
petrooccipital fissure and implies that the artery absent (character 25; 11 of 22 claascending pharyngeal artery does not sup- dograms), and stapedial artery absent
IGNACIUS AND PRIMATE ORIGINS
(character 26; 11of 22 cladograms). Another
possible cranial synapomorphy not included
in this parsimony analysis also suggests a
dermopteran-plesiadapoid clade. While
many living or extinct mammals have or had
a n ectotympanic bulla, Cynocephalus and
Ignacius share one apparently unique feature for this structure. In both taxa, the entotympanic bone apparently contacts the basioccipital medially (see cross-sectional
diagram in Hunt and Korth, 1980, for Cynocephalus). In other mammals with entotympanic bullae, the entotympanic contacts only
the petrosal, which in turn contacts the basioccipital.
This hypothesis of the affinities of Ignacius is also being debated with reference
to postcranial evidence not used in our analysis (Beard, 1990; but see Krause, 1991). In
summary, the link between Dermoptera and
Plesiadapiformes is plausible but weak:
Dermoptera was the sister group in one-half
of the most-parsimonious trees when Plesiadapis was excluded, and Lipotyphla was
the sister group to Plesiadapiformes in most
analyses that included Plesiadapis. This
link needs to be further evaluated on the
basis of new fossil material for Plesiadapiformes and Dermoptera.
A Plesiadapiformes-Primate link?
In our analyses, Plesiadapiformes was
never the sister group to Primates, a widely
held view in the past (see, e.g., Szalay and
Delson, 1979). For more than 100 years, paleontologists have recognized Eocene primates on the basis of synapomorphies with
living primates, including especially a postorbital bar, a n orbital convergence, a welldeveloped internal carotid artery in a promontorial position enclosed in a bony canal,
a n auditory bulla composed of the petrosal, a
Nannopithex fold on the upper molars, nails
rather than claws on some digits, and a n
opposable hallux. There is continued debate
about which nonprimate group is sister to
Eocene-Recent primates, and Plesiadapiformes are often nominated to this position.
One thorough recent analysis (Szalay et al.,
1987) lists nine shared-derived features
uniting Plesiadapiformes with adapids and
omomyids. Six characters are cranial or dental: 1)auditory bulla inflated and formed by
493
the petrosal bone, 2) ectotympanic expanded
laterally and fused medially to the wall of
the bulla, 3) promontorium centrally positioned in the bulla, and large hypotympanic
sinus widely separating promontorium from
the basisphenoid, 4)internal carotid entering the bulla posteriolaterally and enclosed
in a bony tube, 5 ) Nannopithex fold on the
upper molars, and 6) loss of one pair of incisors.
This new information on the cranial anatomy of Ignacius puts the similarities between Plesiadapiformes and Primates in a
new light. The bulla of Ignacius is entotympanic and not petrosal, suggesting that the
bulla of Plesiadapis is also entotympanic.
The ectotympanic is expanded laterally (tubular) and fused to the bulla only in omomyids. Adapids show the more primitive
mammalian condition with the ringlike ectotympanic. This suggests that the omomyid-plesiadapiform similarity is a homoplasy. The internal carotid artery is
absent in Plesiadapiformes, with the principal blood supply to the brain probably coming from the vertebral arteries. In primates,
the ICA is well developed. The reconstructed
last common ancestor of Plesiadapiformes
and Primates had a n arterial arrangement
similar to that of the inferred primitive condition for eutherians (ICA and its promontory and stapedial branches well developed)
(Wible, 1987). Reduction of dental formula
seems to be a common parallel occurrence
among many mammalian groups and is
therefore a poor guide in phylogenetic inference.
Of the other features of similarity between Plesiadapiformes and Primates, the
central position of the promontorium in a
large hypotympanic sinus is a n interesting
similarity but probably occurred by different
developmental pathways. In plesiadapiforms, the brain is relatively small and the
promontoria appear to be situated closer to
the midsagittal plane than in primates. The
central position of the promontorium within
the plesiadapiform bulla seems to result
from a n ontogenetic medialward hypertrophy of the hypotympanic sinus. In contrast,
in Eocene primates the brain is expanded
and the promontoria are carried laterally
during development. In this case the central
494
R.F. KAY ET AL
position of the promontorium is more likely
the result of this lateral displacement inside
an unhypertrophied bulla.
The shared presence of a Nannopithex fold
is the most likely possible primate-plesiadapiform synapomorphy. However, here
it should be noted that this feature is poorly
developed in the more primitive members of
each clade (i.e., Teilhardina and Cantius for
Primates and Purgatorius for Plesiadapiformes), suggesting homoplasy in this feature.
Primates and Scandentia are a clade
Our analyses support the hypothesis that
Primates and Scandentia are a monophyletic group. This corroborates the once widely
held view that primates and tree shrews are
closely related (see, e.g., Le Gros Clark,
1959). Monophyly is supported by four characters in the single most-parsimonious
cladogram: presence of postorbital bar
(character 13), maxillary artery pierces ectopterygoid plate (character 20), arteries of
middle ear enclosed in bony canals (character 28), and jugular foramen dual (character
32).
Unity of the Archonta
In only one of the 31 cladograms under
consideration was Archonta a monophyletic
group. The uniting characters of Archonta
have always been few. Novacek and Wyss
11986) cited a pendulous penis and confluent sustentacular and navicular facets of the
talus a s uniting characters. The cladograms
of Gregory (1910) and Novacek and Wyss
(1986) support monophyly of Archonta (Fig.
9), but these are less parsimonious than our
cladograms (56 vs. 54 steps for the taxa they
included). The monophyly of Archonta is in
need of review.
Several other suggestions as to the phylogeny of Primates and archontans can be
evaluated with our data. Pettigrew et al.
(1989) proposed that Megachiroptera was
the sister group of Primates, a view not endorsed by our analysis of cranial data. Pettigrew’s cladogram is considerably less parsimonious than our results (48 vs. 41 steps).
Miyamoto and Goodman (1986) proposed
that ScandentEa is the sister group to Lipotyphla; their cladogram took 46 steps based
on our character matrix, whereas our most
parsimonious cladogram took only 39.
ACKNOWLEDGMENTS
We thank Dr. Peter Houde who collected
and prepared the Ignacius specimens and
kindly allowed us to study them. Figure 5
was drawn by Ms. Jo-Ellen Trecartin; Figures 6 and 9 were prepared by Dr. Bradley
Smith. Photographs in Figures 1-4 were
made by Mr. Victor Krantz of the Smithsonian Institution. The authors profited
greatly from conversations with Drs. Matt
Cartmill, Ross MacPhee, John Wible and Richard W. Thorington, J r . , about cranial
anatomy and primate evolution generally.
Much of the research on the fossils was undertaken by R.F.K. under a Smithsonian Institution Fellowship in 1987. Original specimens were made available for our study by
the kindness of Dr. Matt Cartmill, Duke
University; Drs. R. Emry and R. Thorington,
Jr., Smithsonian Institution; and Dr. D.E.
Russell, MNHN, Paris.
LITERATURE CITED
Beard KC (1990) Gliding behaviour and palaeoecology of
the alleged primate family Paromomyidae (Mammalia, Dermoptera). Nature 345:340-341.
Bown TM, and Rose KD (1976) New early Tertiary primates and a reappraisal of some Plesiadapiformes.
Folia Primatol. 26:109-138.
Bugge J (1974) The cephalic arterial system in insectivores, primates, rodents and lagomorphs, with special
reference to the systematic classification. Acta Anat.
871 Suppl. 62 ):1-160.
Butler PM (1948) On the evolution of the skull and teeth
in the Erinaceidae, with special reference to fossil material in the British Museum. Proc. Zool. Soc. London
I 18:446-50 1.
Butler PM (1956) The skull of Ictops and the classification of the Insectivora. Proc. 2001. Soc. London
126t453-481.
Butler PM (1988) Phylogeny of the insectivores. In M J
Benton (ed.):The Phylogeny and Classification of the
Tetrapods, Vol. 2: Mammals. Oxford: Clarendon
Press, pp. 117-141.
Cartmill M (1975) Strepsirhine basicranial structures
and the affinities of the Cheirogaleidae. In WP Luckett and FS Szalay (eds.): Phylogeny of the Primates.
New York: Plenum Publ. Corp., pp. 313-354.
Cartmill M (1978) The orbital mosaic in prosimians and
the use of variable traits in systematics. Folia Primatol. 30:89-114.
Cartmill M, and MacPhee RDE (1980) Tupaiid affinities: The evidence of the carotid arteries and cranial
skeleton. In PW Luckett (ed.): Comparative Biology
and Evolutionary Relationships of Tree Shrews. New
York: Plenum Pub. Corp., pp. 95-132.
IGNACIUS AND PRIMATE ORIGINS
Conroy GC, and Wible JR (1978)Middle ear morphology
of Lemur uariegatus: Some implications for primate
paleontology. Folia Primatol. 29:81-85.
Evans HE, and Christensen GC (1979)Miller’s Anatomy
of the Dog. Philadelphia: WB Saunders.
Forsyth Major CI (1901)On some characters of the skull
in the lemurs and monkeys. Proc. Zool. SOC.London
Zt129-156.
Gingerich PD (1976) Cranial anatomy and evolution of
early Tertiary Plesiadapidae (Mammalia, Primates).
Papers Paleontol. Univ. Michigan 15:l-140.
Gingerich PD (1989)New earliest Wasatchian mammalian fauna from the Eocene of northwestern Wyoming:
Composition and diversity in a rarely sampled highfloodplain assemblage. Papers Paleontol Univ. Michigan 28.1-97.
Gingerich PD (1991) Systematics and evolution of early
Eocene Perissodactyla (Mammalia)in the Clarks Fork
Basin, Wyoming. Contrib. Paleontol. Univ. Michigan
28t181-213.
Gingerich PD, and Martin RD (1981) Cranial morphology and adaptations in Eocene Adapidae. 11. The
Cambridge skull of Adapis parisiensis. Am. J. Phys.
Anthropol. 561235-257.
Gregory WK (1910) The order of mammals. Bull. Am.
Mus. Nat. Hist. 27:l-362.
Gregory WK (1920) On the structure and relations of
Nothai-ctus, an American Eocene primate. Mem. Am.
Mus. Nat. Hist. 3t51-243.
Hershkovitz P (1977) Living New World Monkeys (Platyrrhini): Vol. 1.Chicago: University of Chicago Press.
Hunt RM J r , and Korth WW (1980) The auditory region
of Dermoptera: Morphology and function relative to
other living mammals. J . Morphol. 164r167-211.
Kay RF, and Cartmill M (1977) Cranial morphology and
adaptations of Palaechthon nacimienti and other
Paromomyidae (Plesiadapoidea, ?Primates), with a
description of a new genus and species. J . Hum. Evol.
6:19-53.
Kay RF, Thorington RW Jr. and Houde P (1990) Eocene
plesiadapifol-m shows affinities with flying lemurs
not primates. Sature 345t342-344.
Kielan-Jaworowska Z (1981) Evolution of the therian
mammals in the Late Cretaceous of Asia. Part IV.
Skull structure of Kennalestes and Asioryctes. Palaeontol. Polon. 42:25-78.
Kielan-Jaworowska Z (1984) Evolution of the therian
mammals in the late Cretaceous of Asia. Part VII.
Synopsis. Palaeontol. Polon. 46:173-183.
Krause DW (1991) Were paromomyids gliders? Maybe.
Maybe not. J . Hum. Evol. 21:177-188.
Le Gros Clark WE (1959) The Antecedents of Man. Edinburgh: Edinburgh University Press.
MacPhee RDE (1981) Auditory regions of primates and
eutherian insectivores: Morphology, ontogeny, and
character analysis. Contrib. Primatol. 19:l-282.
MacPhee RDE, and Cartmill M (1986)Basicranial structures and primate systematics. In DP Swindler and J
Erwin (eds.): Comparative Primate Biology, Vol. 1:
Systematics, Evolution, and Anatomy. Plenum Press
New York pp. 219-275.
MacPhee RDE, Cartmill M, and Gingerich PD (1983)
New Palaeogene primate basicrania and the definition of the order Primates. Nature 301t509-511.
495
McKenna MC, Xue X and Zhou M (1984) Prosarcodon
lonunensis, a new Paleocene micropternodontid
palaeoryctoid insectivore from Asia. Am. Mus. Nov.
1278O:l-17.
Miyamoto MM, and Goodman M (1986) Biomolecular
systematics of Eutherian mammals: Phylogenetic
patterns and classification. Syst. 2001.35:230-240.
Muller J (1934)The orbitotemporal region of the skull of
the Mammalia. Arch. Neerl. Zool. 1.118-259.
Novacek MJ (1986) The skull of leptictid insectivorans
and the higher-level classification of eutherian mammals. Bull. Am. Mus. Nat. Hist. 183:3-111.
Novacek MJ, and Wyss AR (1986) Higher-level relationships of the recent eutherian orders: morphological
evidence. Cladistic? 2257-287.
Novacek MJ, Wyss AR, and McKenna MC (1988) The
major groups of eutherian mammals. In MJ Benton
(ed.): The Phylogeny and Classification of the Tetrapods, Vol. 2. Oxford: Clarendon Press, pp. 31-71.
Pettigrew JD, Jamieson BGM, Robson SK, Hall LS,
McAnnally KI, and Cooper HM (1989) Phylogenetic
relations between microbats, megabats and primates
(Mammalia: Chiroptera and Primates). Phil. Trans.
R. SOC.London [Series Bl325t489-559.
Rose KD, and Gingerich PD (1976) Partial skull of the
plesiadapiform primate Ignacius from the early
Eocene of Wyoming. Contrib. Mus. Paleontol. Univ.
Michigan 24:181-189.
Russell DE (1964) Les Mammiferes Paleocenes
#Europe. Mem. Mus. Nat. Hist. Nat. Series C 13:l324.
Stehlin HG (1912, Die Saugetiere des schweizerischen
Eocanes. Kritischer Catalog der Materialien. Siebenter Teil, erste Halfte: Adapis. Abhandl. schweizerischen Palaontol. Ges. 38:1165-1209.
Straney DO (1984) The nasal bones of Chzroderma
(Phyllostomidae).J . Mammal. 65:163-165.
Swofford DL (1989)PAUP: Phylogenetic Analysis Using
Parsimony, Version 3.0. Champaign, IL: Privately
published.
Szalay FS (1972) Cranial morphologl of the early Tertiary Phenacolemur and its bearing on primate phylogeny. Am. J. Phys. Anthropol. 36:59-76.
Szalay FS (1975) Phylogeny of primate higher taxa. The
basicranial evidence. In WP Luckett and FS Szalay
(eds.): Phylogeny of the Primates. A Multidisciplinary
Approach. New York: Plenum Publ. Corp., pp. 91-125.
Szalay FS, and Delson E (1979) Evolutionary History of
the Primates. New York: Academic Press.
Szalay FS, Rosenberger AL, and Dagosto M (1987) Diagnosis and differentiation of the order primates. Yrbk.
Phys. Anthropol. 30:75-105.
Thewissen JGM (1985) Cephalic evidence for the affinities of Tubulidentata. Mammalia 49:257-284.
Thewissen JGM and Gingerich PD (1989) Skull and endocranial cast ofEoryctes melanus, a new palaeoryctid
(Mammalia, Insectivora) from the early Eocene of
western North America. J. Vert. Paleontol. 9,459470.
Wible JR (1984) The Ontogeny and Phylogeny of the
Mammalian Cranial Arterial Pattern. PhD Thesis,
Duke University.
Wible JR (1986) Transformations in the extracranial
course of the internal carotid artery in mammalian
phylogeny. J. Vert. Paleontol. 6,313-325
496
R.F. KAY ET AL
Wible J R (1987) The eutherian stapedial artery: Character analysis and implications for superordinal relationships. 2001. J. Linnean Soc. 91:107-135.
APPENDIX A: CHARACTERS USED IN
THE CLADISTIC ANALYSES DESCRIBED
IN THE TEXT.
Many of the characters described below
are based on Novacek (1986), Wible (1984,
19871, Novacek and Wyss (19861, and Butler
(1948,1956). These authors are often not the
original source for the characters they use.
We have included only informative characters (i.e., those that may represent synapomorphies of some of the investigated taxa.)
Characters 5, 7,10,12,13, and 15 are scored
as ordered multistate. Other multistate
characters are considered to be unordered.
with a n r2 of 0.624. For each taxon the “expected” IOF size was calculated from this
regression equation. The observed (measured) size of IOF for each species was compared with the expected and expressed a s a
residual:
100 . (observed
-
expected)/(expected).
This residual ranges from -22.1 (Loris)to
+ 17.6 (Erinaceus).Residuals were grouped
into three categories: large (0)for taxa with
residuals greater than 7.0; moderate (1)for
taxa with residuals smaller than 7.0 but
greater than -8.0; small (2) for taxa with
residuals smaller than -8.0. Ignacius has a
value of 12.1, essentially identical to the
value for Echinosorex (12.2). From observation of the figures and descriptions of &e1. Primitively the caudal part of the na- lan-Jaworowska (19811, Asioryctes has a
sal flares laterally (0); in the derived condi- large IOF. Plesiadapis was scored a s modertion the nasals are of uniform width and ate based on Kay and Cartmill (1977).
6. Contact between lacrimal and palahave a narrow suture with the frontal (1).
The nasals are wider than long in certain tine in the orbit (0; Butler, 1956; Novacek
Microchiroptera (e.g., Rhinolophus) due to and Wyss, 1986). Derived conditions include
extreme shortening of the rostrum, but this a contact between maxilla and frontal (1)or
does not bear on the scoring of this charac- large ethmoid exposure (2), which reduce
ter. This and other sutural characters of the the size ofthe orbital process ofthe palatine.
microbats are difficult to score because the The condition of this character in Ignacius is
cranial bones fuse very early in ontogeny unknown. The condition in Tarsius was determined by Cartmill (1978). Tupaia and
(Straney, 1984).
2. Left and right incisive foramina sepa- Ptilocercus were scored according to Le Gros
rated by a bony process in the median plane Clark (1959).
7. Caudal midsagittal margin of palate
(0) or fused in the midsagittal plane (1 NOCrostra1 to M3 (O), at M’ (11, caudal to M3 ( 2 ) .
tilio lacks a n incisive foramen iscored as 2 j .
8. Posterior palatine torus present (0) or
3. Rostra1 edge of palate complete (0) or
interrupted a t the suture between premax- absent (1).Unlike Novacek (19861, we conilla and maxilla (1).The incisive foramina sider the presence of the postpalatine torus
are open rostrally a t the suture between pre- primitive because it is present in Asioryctes.
9. Lacrimal foramen surrounded by the
maxilla and maxilla in certain Microchilacrimal bone (0) or lies on the lacrimal/
roptera (2).
4. Infraorbital canal long (0) or short, de- maxillary suture (1).Noctilio lacks a lacrimal foramen and consequently could not be
veloped as just a foramen in the maxilla (I).
5. Infraorbital canal large. The relative scored.
10. Lacrimal foramen opens into orbit (0),
size ofthe infraorbital foramen was based on
measurement of its area, the product of onto orbital rim (l),or onto rostrum (2).
11. Absence (0) or presence (1) of a lagreatest diameter times diameter a t right
angles to that measurement. Log IOF area crimal tubercle. The tubercle is absent
was regressed against log prosthion-inion in Cretaceous proteutherians (Kielan-Jalength. The resulting least-squares regres- worowska, 1981).
12. The zygomatic is primitively large
sion is
and forms a significant portion of the zygolog IOF = 1.583(log skull length) - 2.673, matic arch (0). We recognize three states:
IGNACIUS AND PRIMATE ORIGINS
497
zygomatic separates squamosal and maxilla cuspidens (MNHN CR 965) but was labelled
(01, squamosal and maxilla share a suture a s “t.d.a.” by Russell (1964: Fig. 19).
17. Foramen rotundum absent (0) or
on the zygomatic arch (implying that the zygomatic is small, scored a s (11, and zygo- present (1).Muller (1934) established that
matic absent (2). In Echinosorex, maxilla the fusion of foramen rotundum with the
and squamosal are in contact on the medial sphenorbital fissure is primitive for placentals.
side of the zygomatic arch.
18. Lateral pterygoid plate present (0) or
13. Postorbital process of frontal absent
(O), present but not touching zygomatic arch absent (1).
19, 20. Alar canal and maxillary artery.
(I), or touching zygomatic arch to form a
postorbital bar ( 2 ) . Intermediates between The course of the maxillary artery from the
these states occur. The size ofthe postorbital basicranium to the orbit may lead through a
process is variable in Ptero-pus.where i t may bony foramen or canal in the ahsphenoid.
or may not reach the zygomatic process This canal is often developed a s a bridge on
the lateral side of braincase and lateral
(scored a s ‘‘?,,).
14. Ethmoid foramen present (0) or ab- pterygoid plate (e.g., Tenrec). In other mammals (e.g., Tarsius) the maxillary artery
sent (1).
15. Optic foramen (OF) small (01, moder- may pierce the lateral pterygoid plate. Both
ate (l),or large (2). The relative size of the canals or foramina have been described a s
optic foramen was based on measurement of alar or alisphenoid canal, and they are
its area, the product of greatest diameter equivalent in that they carry the maxillary
times diameter a t right angles to that mea- artery in a bony structure from which the
surement. A line was assigned to a bivariate lateral pterygoid muscle originates. Howcluster of log skull length vs. log OF area. ever, both structures can be present in the
The line was assigned slope 2.0 (slope of same taxon (e.g., Rattus), so we distinguish
isometry) and passed through mean log two characters.
We retain the name alar canal for the caskull length and mean log OF area. The
nal on the lateral wall of the braincase
equation expressing this line is
(character 19). Primitively this canal is
Log (optic foramen diameter) = 2Uog skull present (01, but it can also be absent (1).The
alar canal is present but lacks a medial wall
length) - 3.167.
in Megachiroptera and thus lies within the
For each taxon the “expected” OF size was cranial cavity (Wible, 1984). This condition
calculated from this equation. The observed is scored as 2. The alar canal is commonly
(measured) size of OF for each species was prePent in Echinosorex but is sunietimes abcompared with the expected and expressed sent (Thewissen, 1985).
The foramen in the lateral pterygoid plate
as a residual: 100 . (observed - expected)/
(expected). This residual ranges from -93.5 differs from the alar canal in that it trans(Tenrec)to +283.1 (Tarsius).Residuals were mits the maxillary artery and/or other strucgrouped into three categories: small (0) for tures from medial to lateral side of the lattaxa with residuals less than -75.0, moder- eral pterygoid plate (character 20), whereas
ate (1)for taxa with residuals greater than the alar canal does not pierce the plate. This
-75.1 but less than -25.0, large (2) for taxa foramen is apparently absent in Asioryctes.
with residuals greater than -10.1. Ignacius Absence is considered primitive (0).
The foramen of the lateral pterygoid plate
has a value of 76.6, essentially identical to
is usually a simple window, but in Ptilocerthe value for Erinaceus (-76.1).
16. Suboptic foramen absent (0) or cus it is elongated to a canal. In addition to
present (1).The suboptic foramen carries a the artery, this foramen may also carry the
vein that interconnects the orbits (Butler, branch of the trigeminal nerve to the medial
1948). This character was scored a s 1 if the pterygoid muscle, or this nerve may travel
venous canal is completely separated from through a second foramen in the lateral
the optic foramen by a bony process. A sub- pterygoid plate, a s in Tupaia. Dissections
optic foramen is present in Plesiadapis tri- confirm that the foramen does not carry the
~
498
R.F. KAY ET AL.
maxillary artery in Lemur, and instead carries only a branch of the mandibular branch
of the trigeminal nerve. Because we cannot
distinguish the soft structures transmitted
on the basis of osteological evidence, we
score the presence of one or more foramina
in the lateral pterygoid plate a s (1).Other
perforations of the lateral pterygoid plate
include the foramen crotaphiticum and foramen Civini described by Hershkovitz (1977).
21. Postglenoid process present (0) o r absent (1).The process is well-developed in
Asioryctes .
22. Suprameatal foramen present (0) or
absent (1).This foramen dorsal to the mandibular fossa in the lateral aspect of the
mandibular process of the squamosal transmits veins. It is present in Asioryctes.
23. External auditory meatus membranous (0) or bony (1).Primitively the external
auditory meatus is unossified. In several
taxa there is a distinct bony tube lateral to
the tympanic membrane. In primates, this
tube is often formed by the ectotympanic.
24. Auditory bulla unossified (O), parfully
tially ossified by the ectotympanic (l),
ossified by the petrosal (2), fully ossified by
the entotympanic (3), or fully ossified by the
alisphenoid (4).Cynocephalus was scored
according to Hunt and Korth (1980), Tupaia
and Ptilocercus according to MacPhee
(1981).
25. Promontory artery present in adult
(0) or absent in adult (1).For those taxa in
which the promontory artery has involuted
in the adult, Ignacius and Loris (Cartmill,
1975), Cynocephalus (Wible, 1986), and Lemur (Bugge, 1974; Conroy and Wible, 1978),
we could nonetheless determine its course in
fetal life by the location of the sympathetic
plexus of the internal carotid artery. The
location of this plexus is a reliable indicator
of the course of the promontory artery. All of
the studied taxa exhibit a transpromontorial course (sensu Wible, 1984) of the internal carotid artery or its sympathetic plexus.
26. Stapedial artery present in adult (0)
or absent in adult (1). A groove on the caudal
promontory for the stapedial artery is primitive for placentals (Wible, 1987). Grooves
are not always visible for taxa with stape-
dial arteries. We were able to score three
taxa because their morphology relating to
this character has been previously published: Myotis received a score of 0 (Wible,
19841, Pteropus a score of 1 (Wible, 19871,
and Ptilocercus a score of 0 (Cartmill and
MacPhee, 1980). The Lipotyphla were
scored according to MacPhee (1981).
27. Vascular foramina in parietal or
squamosal present (0) or absent (1).Primitively these foramina, which may be venous
or arterial, are present.
28. Arteries of middle ear cavity exposed
(0) or enclosed in bony canals (1). The arteries of the middle ear are primitively in ventrally open grooves.
29. Mastoid tubercle weak or absent (0) or
strong (1).A weak mastoid process if present
in Asioryctes. Loris and Cynocephalus both
received a score of 0 even though the scoring
is somewhat ambiguous because of the extreme mastoid inflation in both taxa. Pteropus and Macroglossus were scored a s 1.
However, the tubercle is formed by the
squamosal rather than the mastoid.
30. Mastoid foramen present in mastoid
bone (O), on suture of mastoid and occipital
(11, in occipital bone (21, or absent (3). Mastoid foramen is used to describe any venous
foramen that drains a dural sinus and
pierces the occipital part of the mastoid or
its suture with the occipital bone. Such a
foramen is present in the mastoid bone of
Asioryctes.
31. Paroccipital process weak or absent
(0) or strong (1).Asioryctes lacks a ventral
process of the occipital caudal to the external auditory meatus.
32. Jugular foramen single (0) or dual (1).
The jugular foramen transmits venous
drainage from the dural sinuses a s well a s
cranial nerves IX, X, and XI. Although
scored a s having the derived state, Loris and
Tupaia are potential intermediates. In both
taxa, two foramina are visible within a shallow fossa. It is not clear, however, whether
the two foramina are separating venous
drainage from nerves o r if they are both
channels for venous drainage.
33. Supraorbital foramen absent (0) or
present (1).
Документ
Категория
Без категории
Просмотров
2
Размер файла
1 918 Кб
Теги
anatomy, cranial, ignacius, graybullianus, plesiadapiformes, affinities
1/--страниц
Пожаловаться на содержимое документа