\Reabody Museum
| of Natural History
Yale University
New Haven, CT 06520

Posti | la Number 177

30 November 1979

Revision of the Genus Palatobaena
(Testudines, Baenidae), with the Descrip-
tion of a New Species

J. David Archibald
J. Howard Hutchison

(Received 12 May 1978)
Abstract

Recovery of three new partial skulls of Pa/a-
tobaena bairdi from the latest Cretaceous
(Lancian) and middle Paleocene (Torrejonian)
of Montana and from the early Paleocene of
Colorado adds greatly to the knowledge of
this peculiar baenid turtle. Although there is
some variation in the temporal emargination,
Palatobaena primitively appears to have had
a deep emargination as seen in eubaenines.
The wide triturating surfaces align Palato-
baena with the Eubaeninae. A skull of anew
Species found in the Wasatch Formation
(Wasatchian), Wyoming, belongs to this
genus and extends the record of Palatobaena
into the early Eocene. The new species has a
relatively shorter skull and its jaw mechanics
represents a further specialization from the
Morphology seen in P. bairdi.

© Copyright 1979 by the Peabody Museum of
Natural History, Yale University. All rights
reserved. No part of this publication, except
brief quotations for scholarly purposes, may
be reproduced without the written permission
Of the Director, Peabody Museum of Natural
History.

Abbreviations

The following institutions are referred to in the
text:

AMNH The American Museum of Natural
History
CCM Carter County Museum, Montana

Field Museum of Natural History
PA, Princeton University

UCM University of Colorado Museum

UCMP University of California Museum
of Paleontology

YPM Yale Peabody Museum of Natural

History

Introduction

Recently, Gaffney (1972a) revised the turtles
of the family Baenidae. Unlike earlier workers
he relied heavily on cranial morphology and
convincingly showed (1972a and 1975) that
baenids are true cryptodires. A major portion
of his revision (1972a) was devoted to the
description of new genera and species. The
most unusual new taxon described was Pala-
tobaena bairdi. Due to this species’ unusual
features and also the fragmentary nature of
the material, Gaffney (1972a, p. 304-308) was
unable to determine the relationships of this
taxon.

Palatobaena bairdi was based largely on the
right half of a skull (type specimen) from
Cedar Point Quarry, Wyoming (Tiffanian) and
the anterior portion of a skull from the Creta-
ceous of Montana. Gaffney (1972a, figs. 22:
23) restored the skull of Palatobaena with a
rather circular outline in dorsal view and witha
relatively shallow temporal emargination. In
1975, a joint University of California Museum
of Paleontology (UCMP) - Los Angeles County
Museum field party recovered portions of a
skull (UCMP 1145339) of this genus from the
Hell Creek Formation (Lancian), Montana.
Though fragmentary, this specimen pre-
serves the posterior portion of the skull roof

Palatobaena, Postilla 177
Revision and New Species

3 Palatobaena,
Revision and New Species

Postilla 177

@Fig. 1

Palatobaena bairdi Gaffney. Restored skull
based on UCMP 114539 with additions from
FMNH PR 829, PU 16839 (type), and UCM
37738: a. dorsal view, b. ventral view, c. pos-
terior view; d. lateral view.

Abbreviations for this and following figures:
ant. crs., anterior crista of processus inferior
Parietalis; aper. nar. int., apertura narium
interna: bo, basioccipital; bs, basisphenoid:
Cav. acust. jug., Cavum acusticojugulare; ex,
€xoccipital; for. ant. can. car. int., foramen
anterior canalis carotici interni; for. car. post.,
foramen posterior canalis carotici interni; for.
ner. hypo., foramen nervi hypoglossi; for. pal.
Post., foramen palatinum posterius; for.
Prepal., foramen praepalatinum; for. stap.,
foramen stapediotemporale; for. supmax.,
foramen supramaxillare; fos. orb., fossa
Orbitalis: fr, frontal; ju, jugal; /at. crs., lateral
Crista of processus inferior parietalis; mx,
Maxilla; na, nasal; op, opisthotic: pa, parietal:
Pal, palatine; pf, prefrontal; pm, premaxilla;
PO, postorbital; pr, prootic; proc. inf. par.,
Processus inferior parietalis; pt, pterygoid; gj,
Quadratojugal; qu, quadrate: so, supraoccipi-
tal; sq, squamosal; sul. cav., sulcus caverno-
SUS; vo, vomer.

allowing for a somewhat different restoration
(Fig. 1). Another skull fragment (UCM 37738)
from the Dawson Formation (Puercan), Colo-
rado, agrees in most characters with UCMP
114539.

In 1977 during the preparation of this study,
Dr. Eugene S. Gaffney called our attention toa
skull of this genus recovered by Mr. Marshall
Lambert from the Tongue River Formation
(Torrejonian), southeastern Montana. This
skull (CCM 77-11) is approximately three-
quarters complete and is undistorted (Figs. 2;
3; 7a, b). In 1972 and 1975, pieces of a

skull of this genus were recovered by UCMP
parties in the Wasatch Formation, Wyoming.
Though fragmentary when found, this spec-
imen represents an almost complete skull of a
new species of Palatobaena. It also extends
the range of this genus into the early Eocene.

Family Baenidae Cope, 1882
Subfamily Eubaeninae (Williams, 1950)
Palatobaena Gaffney, 1972a

Type species Palatobaena bairdi Gaffney,
1972a

Known distribution Latest Cretaceous
(Maestrichtian correlative, Lancian) of Mon-
tana, early (Puercan) Paleocene of Colorado,
middle (Torrejonian) Paleocene of Montana,
late (Tiffanian) Paleocene and early (Wasat-
chian) Eocene of Wyoming.

Revised Diagnosis Eubaenines with wide,
subrectangular skulls; anterior portion of skull
broadly curving in dorsal view; preorbital skull
length shortest of any baenoid; labial ridges of
maxilla and lower jaw forming larger angle
with sagittal plane than other baenoids; tem-
poral emargination deep to moderate; ridge
separating orbital floor from cheek; nasals
small and not contacting in midline; distinct,
smooth sulcus surrounding narial opening
ventrally and laterally; upper triturating sur-
faces very wide forming triangular crushing

Palatobaena, Postilla 177
Revision and New Species

5 Palatobaena,
Revision and New Species

Postilla 177

areas, surfaces flat to concave, narrowing in
anteromedial portion and forming small con-
tinuous lingual ridge on premaxillae; pro-
cessus pterygoideus externus very small:
anterior process of jugal extending below
Orbit; posterolateral process of frontal distinct;
processus trochlearis oticum more prominent
than in other baenids, except possibly
Eubaena; antrum postoticum slightly narrow-
ing dorsoventrally; lower jaws with wide tri-
angular triturating surfaces sloping labially,
Processus coronoideus tall and massive; two
deep fossae on labial side of jaw, first large
one posteroventral to processus coronoideus
and second small one anteroventral to area
articularis manibularis. Lower jaws known only
for P. bairdi.

Palatobaena bairdi Gaffney, 1972a
(Figs. 1; 2; 3; 4a, b; 6d, e; 7a, b)

Type Specimen PU 16839, the right half ofa
skull, partially distorted (Gaffney, 1972a, p.
269).

Type Locality Cedar Point Quarry, Bighorn
Basin, Wyoming.

Horizon Fort Union Formation, early Tiffan-
lan (late Paleocene).

Collector Princeton Party, 1949.
Referred Specimens Only specimens not

listed by Gaffney, 1972a, p. 269-272 are
noted.

UCMP 114539, fragmentary skull and jaws.
Locality: V75180, Baenid Skull Locality, Gar-
field County, Montana. Horizon: Hell Creek
Formation, Maestrichtian correlative (Lan-
Clan). Collector: David Lawler, 1975.

@Fig. 2
Palatobaena bairdi Gaffney. Restored skull
based onCCM 77-11: a, dorsal view; b, lateral
ae C, anterior view. Abbreviations as in
ig. 1.

UCMP 114644, left maxilla, UCMP 114656,
right quadrate. Locality: V65127, Bug Creek
Anthills-General, McCone County, Montana.
Horizon: Hell Creek Formation, Maestrichtian
correlative (Lancian). Collector: UCMP Party,
1970.

UCMP 114680, right and left jaw halves;
UCMP 114686, left maxilla. Locality: V70201,
Bug Creek Anthills C, McCone County, Mon-
tana. Horizon: Hell Creek Formation, Maes-
trichtian correlative (Lancian). Collector:
UCMP Party, 1970.

AMNH 8277, complete lower jaws. Locality:
Bug Creek Anthills, McCone County, Mon-
tana. Horizon: Hell Creek Formation, Maes-
trichtian correlative (Lancian).

AMNH 2603, nearly complete lower jaws.
Locality: Lismas, Montana (label). Horizon:
Hell Creek Formation, Maestrichtian cor-
relative (Lancian).

UCM 37738, skull and jaw fragments. Locality:
Corral Bluffs, El Paso County, Colorado.
Horizon: Dawson Formation, Paleocene
(Puercan). Collector: UCM Party, 1974.

CCM 77-11, three-quarters of undistorted
skull. Locality: Medicine Rocks, Site 1, Carter
County, Montana. Horizon: Tongue River
Formation, Paleocene (Torrejonian). Collec-
tor: Marshall Lambert, 1977.

Revised Diagnosis Temporal emargination
deep to moderate; tomial ridge deflecting
ventrally about 10° and vomer about 25° from
the plane of the basicranium:; skull roof over
crista supraoccipitalis varying from sharp
point with little dorsal exposure to laterally
wide triangular surface; crista supraoccipi-
talis terminating posteriorly to level of occipital
condyle and about in line with, or slightly
anteriorly to posterior termini of squamosals:
processus inferior parietalis forming anteriorly
two distinct cristae (anterior and lateral):
frontals wider than long and terminating pos-
teriorly to anterior margin of premaxillae; narial

6 Palatobaena,
Revision and New Species

Postilla 177

Fig. 3

Palatobaena bairdi Gaffney. Restored ventral
view of skull, facing page, based on CCM
77-11. Abbreviations as in Fig. 1.

sulcus extending across anterior margin of
frontals; anterior termination of maxillae well
anterior to orbits; dorsal margin of cheek
emargination with distinct tubercle on medial
side and no smooth shelf on lateral side;
tubercula basioccipitalia divergent and
extending posteriorly as far as occipital con-
dyle articulation; processus paroccipitalis
with distinct transverse ridge extending onto
squamosal; processus paroccipitalis exten-
sively contacting squamosal posteriorly;
posterior surface of squamosal and quadrate
above the incisura columellae auris forming a
broadly U-shaped surface. Features of the
lower jaws same as those given in the generic
diagnosis.

Discussion The record of Palatobaena has
been greatly augmented since the original
description of Gaffney (1972a). Material
referable to this genus is now known from
each of the land mammal ages from Lancian
(latest Cretaceous) through Wasatchian (early
Eocene), except possibly for the Clarkforkian
(late Paleocene or early Eocene; see Ginger-
ich and Rose, 1977). The genus is still rep-
resented only by skull material, but Archibald
(1977) has speculated that its shell, probably
unrecognized in existing collections, resem-
bles that seen in eubaenines and baenines.

Although the stratigraphic record of this

genus has greatly improved, the sample in
each interval is usually represented by only
one specimen. Thus, variability can only be

7 Palatobaena,
Revision and New Species

Postilla 177

estimated for each sample. Conservatively we
Nave recognized only two species, but the
recovery of additional material may warrant
further subdivision. All but the new Eocene
Skull described below are referred to Palato-
baena bairdi.

The depth and shape of the temporal emar-
gination are the most variable characters in
the skulls referred to Palatobaena bairdi. In
Gaffney's restoration of the type specimen
(1972a, figs. 22, 23) he tentatively recon-
Structed the skull with a shallow temporal
€Margination. The skull roof of this specimen
'S Somewhat distorted and the posterolateral
Margin of the parietal is quite thick indicating
More bone may have been present poster-
lorly. However, in the Eocene skull (new

species), which preserves the temporal emar-
gination, the parietal remains thick until almost
the very border of the emargination.

In the new Cretaceous specimen of Pa/ato-
baena bairdi, UCMP 1145339, the posterior
part of the skull roof is relatively intact and
undistorted. Here the temporal emargination
is deep, the anterior border reaching anter-
iorly beyond the level of the cavum tympani
(Fig. 1d). The bone near the emargination is
thin, and the dorsal exposure of the crista
supraoccipitalis is extensive, but much nar-,
rower than in the Eocene species. In the por-
tion of the skull roof preserved in the early
Paleocene specimen, UCM 37738, the pari-
etal is slightly wider and thicker along the
temporal emargination than in UCMP 114539.

8 Palatobaena,
Revision and New Species

Postilla 177

In the well-preserved middle Paleocene skull,
CCM 77-11, the left temporal emargination is
complete (Fig. 2a). The bone surrounding the
emargination is as thick as in UCM 37738.
However, the anteroposterior depth in CCM
77-11 is noticeably less than in UCMP 114539
(and probably UCM 37738) with the anterior
border only reaching the level of the anterior
margin of the cavum tympani. This depth is
comparable to that seen in the new Eocene
species described below.

The shape of the temporal emargination is
very distinctive in CCM 77-11. In the other
skulls of Palatobaena preserving the emar-
gination, including the new Eocene species,
the lateral border of the emargination is
approximately in a parasagittal plane while
the medial border diverges anteriorly. In CCM
77-11 both the lateral and medial borders are
in an approximately parasagittal plane. The
distinctive parallel-sided temporal emargina-
tion in CCM 77-11 is the result of the lateral
expansion of the supraoccipital and the pos-
terior portion of the parietal. Although this
distinctive shape of the temporal emargina-
tion may prove to be of taxonomic value upon
recovery of additional specimens, we prefer to
regard it as variability within one species.

The general shallowing of the temporal emar-
gination and the thickening of the bordering
bones going from the latest Cretaceous to
early Eocene specimens of Palatobaena
suggest that the temporal emargination in the
type specimen of P. bairdi, PU 16839, may
have been shallower than in UCMP 114539,
but not as shallow as restored by Gaffney
(1972a, fig. 22A).

Additional variability can be observed in the
position and in the presence or absence of the
anterior and lateral cristae of the processus
inferior parietalis. Three character states are
present within the sample of Palatobaena. |n
the first condition, present in the latest Cre-
taceous skull fragments, UCMP 114539 and
FMNH PR 829, and the early Paleocene spec-
imen, UCM 37738, the anterior and lateral

cristae form an angle of 80—90° to one another
(Fig. 6d). The second state is represented in
the middle and late Paleocene skulls, CCM
77-11 and PU 16839, respectively. In these
specimens the cristae are both more anterior
to anterolateral in orientation and parallel
each other in their more dorsal portions (Fig.
6e). The third character state is present in the
new Eocene species described below. Here
lateral cristae are totally absent and the ante-
rior cristae, although present, are less promi-
nent (Fig. 6f). Only in this last skull are the
differences such as to probably be of taxo-
nomic value in view of the small sample size of
this genus.

A marked sulcus almost completely encircles
the apeturanarium externain all specimens of
Palatobaena bairdi preserving this region of
the skull. Only CCM 77-11 preserves the pre-
maxillae showing that the sulcus, although
less distinctive in this region, continues on to
the anteroventral surface of the premaxillae.
There is some variation in the development of
the rim that surrounds the sulcus and the
apertura. As can be seen in dorsal view of
CCM 77-11 (Fig. 2a), the lateral rim of this
sulcus is slightly flared, interrupting the other-
wise rounded outline of the anterior aspect of
the skull. Most of the other specimens of P.
bairdi including the type, PU 16839, do not
show this flaring (see Fig. 1a). This could be
due to the poorer preservation of the other
specimens, but it seems doubtful. The only
other specimen of P. bairdi that shows a simi-
lar flaring is FMNH PR 829, an anterior skull
fragment (Gaffney, 1972a, fig. 25A). The
overall size of FMNH PR 829 cannot be esti-
mated, but itwas probably a smaller individual
than CCM 77-11 which has a condyle—pre-
maxilla length of 48 cm. This compares to a
condyle—premaxilla length of 63 cm for
UCMP 114539 and 54 cm for PU 16839. The
apparent absence of flaring in these larger
skulls is probably due to one or more factors
related to allometry (e.g., differences in onto-
genetic age or sexual dimorphism). Similarly
the size of the apertura narium externa
appears to be relatively larger in CCM 77-11

2) Palatobaena,
Revision and New Species

Postilla 177

Fig. 4

Lower jaws. Palatobaena bairdi Gaffney,
AMNH 2603: a, dorsal view; b, lateral view.
Plesiobaena putorius Gaftney, PU 17108: c,
dorsal view; d, lateral view.

than as reconstructed here for UCMP 114539
(Fig. 1a) and by Gaffney for the type, PU
16839 (Gaffney 1972a, fig. 22).

Gaffney (1 972a) suggested that the frontals
and nasals may be fused in Palatobaena
bairdi. CCM77-11 is well enough preserved to
Show that the nasals are present (recognized
by Gaffney, personal communication)
although they are highly modified from the
arrangement present in other baenoids. In
CCM 77-11, and presumably P. bairdi in
general, the nasals are reduced to small,
''regular slivers of bone exposed in the dorso
lateral aspect of the apertura narium externa
(Fig. 2). They do not contact each other in the

midline. Some of the sutures between the
nasals and surrounding bones are difficult to
ascertain in CCM 77-11, but the nasals exten-
sively contact the frontals and prefrontals, and
probably the maxillae. The other specimens of
P. bairdi are not well enough preserved to
identify the nasals. In the new Eocene species
of Palatobaena a small pit is present at the
anterolateral corner of the left frontal that may
have received a small nasal bone (Fig. 6f).
Most of its contact was with the frontal bone,
but very reduced contact with the maxilla and
prefrontal cannot be eliminated as a
possibility.

Palatobaena,
Revision and New Species

Postilla 177

Lower jaw fragments were found with UCMP
114539 and UCM 37738, which are the first
associations of jaws and skulls of this genus.
This has allowed for a more certain identifica-
tion of complete isolated lower jaws of the
genus. No comparative study of baenid lower
jaws has been published; therefore, detailed
comparisons with other taxa are not
attempted here. However, lower jaws are now
known for all the baenid species (based on
skulls) described by Gaffney (1972a) except for
Trinitichelys hiatti and Hayemys latifrons, and
considerations in our paper are based ona
comparison with these other taxa. The bone
relationships in a complete lower jaw of Pala-
tobaena bairdi, AMNH 8277, will be dis-
cussed elsewhere by E. Gaffney. A second,
almost complete lower jaw of P. bairdi (AMNH
2603) from the Hell Creek Formation, Mon-
tana, and acomplete lower jaw of Plesiobaena
putorius (PU 17108) from the Polecat Bench
Formation (Torrejonian), Wyoming, are illus-
trated (Fig. 4) for comparison. The latter spec-
imen (mentioned by Gaffney 1972a, p. 264) is
included because of the close similarities of
this lower jaw to those assigned to Palato-
baena bairdi and both are discussed in the
section dealing with speculations on jaw
mechanics.

The scalation pattern on the skull of the type-
species was figured by Gaffney (1972a, figs.
22: 24). Of the new material only CCM 77-11
and the Eocene skull, UCMP 114529, pre-
serve much of this pattern. The scalation pat-
tern becomes more pronounced in going from
the Cretaceous to Paleocene to Eocene spec-
imens. This could be a phylogenetic trend, but
could also be a preservational bias or a
vagary resulting from small sample sizes.

Palatobaena gaffneyi Hutchison, new
species (Figs. 5; 6a, b, c, f; and 7c, d)

Type Specimen UCMP 114529, nearly
complete skull lacking right orbital region,
sutures not co-ossified, slightly distorted. Only
known specimen of species.

Type Locality 71238, Upper Menisco-
therium locality, Sweetwater County,
Wyoming.

Horizon Main Body of Wasatch Formation,
Eocene (Wasatchian).

Collectors J.A. Lillegraven, 1972, and J.H.
Hutchison, 1975.

Etymology For Dr. Eugene S. Gaffney, The
American Museum of Natural History, in rec-
ognition of his contributions to our Knowledge
of the Baenidae.

Diagnosis Temporal emargination moder-
ate; tomial ridge deflecting ventrally about 20°
and vomer about 60° from the plane of the
basicranium; skull roof over crista supra-
occipitalis terminating posteriorly at about
level of occipital condyle and anteriorly to
level of posterior termini of squamosals; pro-
cessus inferior parietalis forming anteriorly
only one distinct crista (anterior); frontals
about equally wide as long and terminating
anteriorly to anterior margin of premaxillae;
narial sulcus only on maxillae and premaxil-
lae; maxillae terminating immediately anterior
to orbits; dorsal margin of cheek emargination

11 Palatobaena,
Revision and New Species

Postilla 177

Fig. 5

Palatobaena gaffneyi Hutchison, n. sp.
Restored skull based on type, UCMP 114529:
&, dorsal view: b, ventral view; c, lateral view;
d, anterior view. Abbreviations as in Fig. 1.

12 Palatobaena,
Revision and New Species

Postilla 177

for. prepal. ,
A ? aper. nar. int.
E

COP

or. ant.can.

sul. cav.
mae

for. prepal.
fos. orb. aper. nar. int.

for. ner. hypo.

fossa for

13 Palatobaena,
Revision and New Species

Postilla 177

with smooth step along lateral side and no
medial tubercle; tubercula basioccipitalia
Parallel and falling short of occipital condyle;
processus paroccipitalis is without distinct
transverse ridge, contacting squamosal in
Small posterolateral area; posterior surface of
Squamosal and quadrate basally forming a
Narrow, sharply angled canal. Lower jaws are
not known.

Description All the noted differences
between Palatobaena gaffneyi and P. bairdi
seem to be related to the feeding mechanism
which involves a further shortening of the
rostrum and concomitant modifications of the
Masticatory musculature. In comparison with
P. bairdi, these resolve into different character
States in diverse parts of the skull.

The preorbital portion of the ventral part of the
Maxilla is markedly shortened in Palatobaena
Qaffneyi so that the labial ridge is within 15—
20° of perpendicular to the sagittal plane
Compared to an angle of 30-40? in P. bairdi.
The sulcus surrounding the nasal aperture in
P. bairdi is reduced in size and restricted to
the maxilla and premaxilla in P. gaffneyi. The
Orbit of P. gaffneyi appears to be more ver-
tically aligned. The fossa orbitalis forms a
Smaller and rather triangular basin in the dor-
Sal surface of the maxilla. Processus inferior
Parietalis forms only a simple ridge with the
anterior part of the parietal roof rather than
dividing into anterior and lateral cristae.

The frontals are relatively longer than in Pala-
fobaena bairdi, terminating in a simple, blunt
Point that extends anteriorly beyond the level
Of the premaxillae. As noted previously, evi-

dence from CCM 77-11 suggests P. gaffneyi

Fig. ¢

Palatobaena gaffneyi Hutchison, n. sp.
Restored skull based on type UCMP 114529:
&, dorsal view with skull roof removed; b andc,
Posterior view. Ventral view of skull roof: d, P.
bairdi, UCMP 114539: e, P. bairdi, PU 16839
(type); f, P. gaftneyi, UCMP 114529 (type).
Abbreviations as in Fier ab

also possessed reduced nasals. In the latter,
the anteriorly elongated frontals dorsally
override and more completely separate the
nasals.

The dorsal margin of the cheek emargination
has a smooth reflected rim probably for the
origin of an adductor muscle (see discussion
of jaw mechanics below). No such rim is
present in Palatobaena bairdi, but just medial
to the cheek rim lies an elongate and ventrally
projecting tubercle which probably serves for
attachment of the same muscle. No other
baenids possess either of these structures.

Palatobaena gaffneyi is less robust inthe area
of the opisthotic-quadrate suture. The dorsally
exposed area between the processus troch-
learis oticum and the tip of the processus
paroccipitalis is generally depressed in P.
gaffney/, whereas in P. bairdi there is a strong
transverse ridge across this region in the area
of the opisthotic-quadrate-squamosal suture.
In P. gaffneyi the squamosal has almost lost
contact with the opisthotic, touching only at
the extreme tip of the processus paroccipitalis
in contrast to the extensive contact between
these two bones in P. bairdi. The posterior
face of the squamosal-quadrate above the
incisura columellae auris is rather constricted
and Is marked by a narrow, sharply angled
channel in contrast to the shallow broadly U-
shaped surface in P. bairdi. In P. gaffneyi, the
tubercula basioccipitalia are less divergent,
more medial, and terminate more anteriorly
than in P. bairdi. The crista supraoccipitalis is
shorter, more broadly pointed (except in CCM
77-11) and has greater dorsal exposure than
in P. bairdi.

The expression of the scale pattern on the
skull roof varies in Palatobaena bairdi, but the
type of P. gaffneyi, despite its probably more
immature age, shows a well-defined scale
pattern as in CCM 77-11 and a better defined
pattern than the type of P. bairdi.

All character states distinguishing Palato-
baena gaffneyi from P. bairdi can be derived
from those seen in P. bairdi, most being exten-

14 Palatobaena,
Revision and New Species

Postilla 177

sions of those characters which distinguish P.
bairdi from other baenids. Morphologically
and temporally, P. bairdi is a suitable ances-
tral species for P. gaffneyi; however, there are
no specimens of P. bairdi that show any ten-
dency for the anterior extension of the frontals.

Speculations on Jaw Mechanics

Most of the characters in the skull of Pa/ato-
baena compared to those in the presumably
more primitive Plesiobaena antiqua (see
Gaffney, 1972a) point to changes in diet. The
majority of the speculations discussed below
indicate Palatobaena was molluscivorous,
Probably the most obvious single feature is
the development of large triangular triturating
surfaces on the maxilla and lower jaw. How-
ever, other baenids—Stygiochelys estesi,
Eubaena cephalica, and Plesiobaena putor-
jus—have variously developed wide triturat-
ing surfaces. Only in Palatobaena bairdi and
finally in P. gaffneyi are the width of the tritur-
ating surfaces plus other features in the skull
and Jaws carried to an extreme.

The reduction in the crista supraoccipitalis
from Palatobaena bairdi to P. gaffneyi is
probably related to the similar shortening and
downward rotation of the face in the latter
species. The shortening of the face in P.
gaffneyi is clearly seen in the more abbre-
viated preorbital region of the premaxillae and
maxillae, and in the greater angle the labial
ridges of the maxillae make with the sagittal
plane. The downward rotation of the face is
evident from the greater angles that the vomer
and tomial ridges make with the plane of the
basicranium in P. gaffneyi relative to P. bairdi.
Schumacher (1973) noted that the M. ptery-
goideus, which originates partially behind the
orbit, helps to counteract the force of the M.
adductor mandibulae externus, which par-
tially originates along bones in the temporal
emargination. Thus the shortening of the face
and of the region surrounding the temporal
emargination in P. gaffneyi probably occurred
simultaneously.

At least one other feature suggests the M.
pterygoideus has been shortened (or some-
what reduced) in P. gaffneyi. As noted earlier
only the anterior crista of the processus infe-
rior parietalis is present in P. gaffneyi. Other
baenids (i.e., Stygiochelys, Eubaena, Plesio-
baena antiqua, and Baena) have both anterior
and lateral cristae as in Palabobaena bairdi.
Two of the better-preserved UCMP baenid
specimens of Plesiobaena antiqua (UCMP
49759) and Baena arenosa (UCMP 117348)
possess a low curved ridge connecting the
two cristae anteriorly, thus forming a small
pocket. Probably this pocket and parts of the
lateral crista served as points of origin for
parts of the M. pterygoideus. The absence of
this pocket or a lateral crista in P. gaffneyi
suggests that the origin of M. pterygoideus
has moved ventrad along the processus infe-
rior parietalis or has become somewhat
reduced. It is doubtful that the muscles for
retraction or protraction (Schumacher, 1973)
have been relatively or absolutely reduced. It
is more likely that with the shortening and rela-
tive deepening of the skull the muscles have
moved closer to the site of insertion increasing
the power of the bite. It is also interesting to
speculate that the anteroposterior arching of
the palate in P. gaffneyi (partially caused by
downward flexure of the face) was produced
in part to offset powerful muscles of protrac-
tion and retraction.

The extremely large processus trochlearis
oticum in Palatobaena bairdi and P. gaffney!
over which the tendon for the M. adductor
mandibulae externus rides (Schumacher,
1973; Gaffney, 1975) is almost directly dorsal
to the large processus coronoideus, suggest-
ing the capability of a very powerful bite.

Other modifications in the skull of Palatobaena
probably are related to its mode of feeding
and are unique for baenids, and possibly
among cryptodires.

It cannot be determined if the Cretaceous and
early Paleocene specimens or the late Paleo-
cene type specimen of Palatobaena bairdi

15 Palatobaena,
Revision and New Species

Postilla 177

have a processus pterygoideus externus.
Both Gaffney’s (1972a, Fig. 23) and our res-
torations suggest it was present. Gaffney
(1972b, p. 31) noted that the “lateral edge [of
the process] is generally produced into a ver-
tical plate that acts asa guide for the lower jaw
during adduction of the lower jaw.” The well-
Preserved specimen of P. bairdi CCM 77-11,
possesses a remnant of the processus ptery-
goideus externus, but the vertical plate is
absent (Fig. 3). InP. gaffneyieven the remnant
Of the process is almost completely gone (Fig.
Sb). The reduction and final loss of the proc-
€Ss and vertical plate suggest other features
Probably arose in the skull, the lower jaws, or
IN both to stabilize the lower jaws during
adduction. At least part of the answer may be
found in the curious development of a slightly
Medially positioned tubercle in the dorsal
Margin of the cheek emargination in P. bairdi.
In P. gaffneyi this tubercle is absent, but a
Shallow sulcus positioned on the lateral mar-
YIN of the cheek emargination is present. Both
Of these structures presumably were for the
Origin of a superficial portion of the M. adduc-
tor mandibulae externus (Pars superficialis) or
Nn analog of the M. zygomaticomandibularis
Of trionychids (Schumacher, 1973). Depena-
INg on the point of insertion on the lower jaw,
this “de novo” muscle may have 1) served to
Stabilize the lower jaw during adduction, 2)
NCreased muscle mass for adduction, 3)
Served as a muscle for protraction to counter-
act the retraction of the M. adductor mandib-
Ulae ©xternus, or in a combination of these
functions, There are two probable places on
the lower jaw where this muscle may have
been inserted. The first is along the lower
Border of the large fossa on the labial side of
the Jaw ventral to the coronoid (see Fig. 4b). If
tinserted here, it probably served as an
&dductor and stabilizer, but not for protrac-
tion. The second possible point of insertion is
the Small fossa on the lateral side of the lower
law below the anterior edge of the area articu-
laris mandibularis (Fig. 4b). This fossa is
Posterior and ventral to the area of origin on
the skull, and if the “de novo” muscle inserted
ere it could serve for adduction, lower jaw
Stabilization, and protraction to counteract the

retraction of the M. adductor mandibulae
externus. It is also possible but unlikely that
the M. pteryoideus inserted here and served a
similar purpose. However, among baenids
only Palatobaena bairdi possessed this small
fossa on the lateral side of the lower jaw (lower
jaws are not known for P. gaffneyi, but it pre-
sumably possessed an homologous fossa)
plus a tubercle or sulcus along the cheek
emargination. This suggests, but by no means
proves, that these new structures were
respectively for the insertion and origin of the
“de novo” muscle.

A lower jaw from the middle Paleocene (Torre-
jonian) of Wyoming, PU 17108, was included
in the hypodigm of Plesiobaena putorius by
Gaffney (1972a, p. 264) and is shown here in
Figure 4c, d. This lower jaw could also belong
to Palatobaena bairdi, but because of
several characters it is probably best included
in Plesiobaena putorius as done by Gaffney
(1972a). It shows the high degree of parallel-
ism between Palatobaena and Plesiobaena
putorius.

In Gaffney’s comments and figures (1972a) it
can be seen that Plesiobaena putorius has not
developed the extreme specializations of
Palatobaena. Although it has wide triturating
surfaces on the maxilla and lower jaws as in
Palatobaena it still retains the presumably
more primitive triangular-shaped face of other
baenids. It is difficult to determine, based on
Gaffney’s illustration (fig. 16A); however, it
appears that the one specimen of Plesio-
baena putorius which preserves the cheek
emargination (PU 16837) does not show the
presence of a tubercle (as in Palatobaena
bairdi) or a sulcus (as in Palatobaena
gaffneyi). In addition, the lower jaw (PU
17108), probably assignable to this species,
lacks the small fossa below the area articularis
mandibularis as seen in Palatobaena bairdi
(compare Fig. 4b and d). PU 17108 is some-
what laterally compressed as evidenced by a
crack in the midline of the lower jaw; however,
even when this is accounted for the triturating
surfaces clearly do not rotate outwards from
the midline to the degree seen in the lower

16 Palatobaena,
Revision and New Species

Postilla 177

jaws of Palatobaena (compare Fig. 4a and c).
This is also true of the triturating surfaces on
the maxilla of Plesiobaena putorius (Gaffney,
1972a, fig. 15) compared to both species of
Palatobaena. These differences support
Gaffney’s (1972a) allocation of PU 17108 to
Plesiobaena putorius.

Hypotheses concerning jaw mechanics and
feeding habits based solely on skull and lower
jaw morphology are always speculative.
Nevertheless, the combination of structural
features in the skull of both species of Palato-
baena strongly suggest that this genus had
developed masticatory muscles especially
designed for a powerful bite (possibly at the
expense of rapid jaw movements). Combined
with greatly widened triturating surfaces this
indicates a probable molluscivorous diet.

A final point of speculation concerns the pres-
ence of asulcus around the apertura narium
externa variously developed in the two
species of Palatobaena. The simplest expla-
nation is that the sulcus was for the origin of
muscles inserting on amoveable proboscis. If
this is true, Palatobaena was undoubtedly a
very odd looking creature with a rounded face
and a moveable snout.

Intrafamilial Relationships of Palatobaena

Gaffney erected a new subfamily, Palato-
baeninae (1972a, p. 269) for his new genus
Palatobaena. The modifications in the anterior
region of the skull are strikingly different from
eubaenines (or other baenoids). The blunting
of the snout, the expansion of the triturating
surface, and the expansion of the muscle
attachment along the cheek emargination as
noted above suggest a very specialized
feeding mechanism, probably for crushing
molluscs. However, the recovery of material
preserving the posterior portion of the skull
indicates Palatobaena is very similar to

eubaenines. As in eubaenines the temporal
emargination is deep at least in stratigraph-
ically lower specimens and the parietal and
squamosal are separated by a flange of the
postorbital. As in Eubaena cephalica the skull
roof overlying the crista supraoccipitalis
narrows and extends beyond the foramen
magnum in the latest Cretaceous specimen of
P. bairdi, UCMP 114539. However, in this
specimen of P. bairdi the narrowing and pos-
terior extension are more exaggerated. The
quadratojugal is reminiscent of other eubae-
nines. There is no broad anterior process and
the quadratojugal is C-shaped with a dorsal
process.

Gaffney tentatively associated Palatobaena
with the baenines (1972a, p. 306 and Figs. 45
and 46). His criteria were the fusion of the
nasals to the frontals and a presumed shallow
temporal emargination. The additional mate-
rial, as described above, has shown that the
nasals are present but greatly reduced and
that the temporal emargination is like other
eubaenines.

The most important feature aligning Palato-
baena with the eubaenines is the expanded
triturating surface found in all members of the
subfamily, except Plesiobaena antiqua. The
polarity of this character state is more easily
adduced than others in the skull. The expanded
triturating surface is regarded as a derived
character. It is not clear whether this character
state is to be regarded as a shared derived
character of eubaenines or as a parallel
development. The former conclusion is favored
here and accordingly we place Palatobaena
with the eubaenines. If it can be shown that the
similarities in the posterior portion of the skull
of Palatobaena and other eubaenines are also
shared derived characters for the subfamily,

Fig.7 >

Stereophotographs of Palatobaena bairdi
Gaffney, CCM 77-11: a, ventral view; b, poste-
rior view and Palatobaena gaffneyi Hutchison,
n. sp., UCMP 114529 (type): c, ventral view; d,
posterior view.

Palatobaena,
Revision and New Species

Postilla 177

Palatobaena,
Revision and New Species

Postilla 177

then this could be used as additional evi-
dence for aligning these taxa. Our study indi-
cates that the character state seen in the
posterior region of the skull of eubaenines is
primitive for both eubaenines and baenines,
and possibly for the family Baenidae. Plesio-
baena antiqua lacks the expanding triturat-
ing surface and for that reason is tentatively
excluded from the Eubaeninae (Archibald,
1977). The slight expansion of the triturating
surface seen in the late Paleocene Plesio-
baena putorius is regarded as a development
parallel to that seen in the Cretaceous and
later eubaenines. The Eubaeninae is here
considered to comprise the following genera
recognized by Gaffney (1972a, p. 248)
Eubaena, Stygiochelys, and Palatobaena.

Acknowledgements

The authors wish to thank E.S. Gaffney for the
loan of material pertinent to this study and for
allowing us to use Figure 3 and outline draw-
ings used in Figures 2a and b; D.M. Bramble
for his observations on musculature; and the
curators of the following institutions for the use
of specimens: The American Museum of
Natural History; Carter County Museum, Mon-
tana; Field Museum of Natural History; Prince-
ton University; University of Colorado Museum;
and University of California Museum of Pale-
ontology. We would also like to thank D.M.
Bramble, W.A. Clemens, and E.S. Gaffney for
reading and providing helpful criticisms
throughout the preparation of this manuscript.
Financial support from NSF Grant BMS
75-21017 to W.A. Clemens, NSF Grant GB
40361 to D.E. Savage, a Grant-in Aid for
Research from Sigma Xi (to J.D.A.), and the
Annie Alexander Endowment to UCMP are
gratefully acknowledged. This study was
completed while the senior author was a J.
Willard Gibbs Instructor, Department of Geol-
ogy and Geophysics, Yale University.

Line drawings were prepared by the following
people: Figures 2a and b (outline drawings)
and Figure 3, Lorraine Meeker, AMNH, ini-
tialed figures, the authors; all other figures, Pat
Lufkin, UCMP. Photographs (Fig. 7) were
done by William K. Sacco, YPM.

19 Palatobaena,
Revision and New Species

Postilla 177

Literature Cited

Archibald, J.D. 1977. Fossil Mammalia and
Testudines of the Hell Creek Formation, and
the geology of the Tullock and Hell Creek
Formations, Garfield County, Montana. Ph.D.
dissertation, Univ. Calif., Berkeley.

Cope, E.D. 1882. Contributions to the history
of the Vertebrata of the Lower Eocene of
Wyoming and New Mexico, made during
1881. Proc. Amer. Philos. Soc. 20:139-197.
Gaffney, E.S. 1972a. The systematics of the
North American family Baenidae (Reptilia,
Cryptodira). Bull. Amer. Mus. Nat. Hist.
147:243 319,

as 1972b. An illustrated glossary of turtle
skull nomenclature. Amer. Mus. Novitates, no.
2486:1-33.

1975. Aphylogeny and classification of
the higher categories of turtles. Bull. Amer.
Mus. Nat. Hist. 155:389-436.

Gingerich, P.D. and K.E. Rose. 1977. Pre-
liminary report on the American Clark Fork
Mammal fauna, and its correlation with similar
faunas in Europe and Asia. Géobios. Mém.
Spécial 1:39-45.

Schumacher, G.H. 1973. The head muscles
and hyolaryngeal skeleton of turtles and croc-
Odilians, p. 101-199. In C. Gans and T.S.
Parsons [eds.] Biology of the Reptilia, Vol. 4.
Academic Press, New York.

Williams, E.E. 1950. Variation and selection
IN the cervical central articulations of living
turtles. Bull. Amer. Mus. Nat. Hist. 94:505-562.

The Authors

J. David Archibald. Department of Biology,

Yale University, and Peabody Museum of

Natural History, Yale University, New Haven,
Onnecticut 06520.

J. Howard Hutchison. Museum of Paleon-
lology, University of California, Berkeley,
California 94720.