Caninemys, a new side-necked turtle (Pelomedusoides, Podocnemididae) from the Miocene of Brazil

2009·1.6 MB·English

by  Meylan, Peter A. (Peter Andre)

A tamerican museum
Novitates
PUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORY
CENTRAL PARK WEST AT 79TH STREET, NEW YORK, NY 10024
Number 3639, 26 pp., 7 figures, 1 table March 31, 2009
Caninemys, a New Side-Necked Turtle
(Pelomedusoides: Podocnemididae) from the Miocene
of Brazil
PETER A. MEYLAN,1 EUGENE S. GAFFNEY,2 AND
DIOGENES DE ALMEIDA CAMPOS3
ABSTRACT
A new genus and species of podocnemidid pleurodire, Caninemys tridentata, is described on the
basis of a skull collected by L.I. Price in 1962 from the late Miocene of Acre, Brazil. It is unique
among podocnemidids (and all other turtles) in having greatly inflated maxillae, each with a
ventral, toothlike process. Along with a midline process of the premaxillae, these processes form a
tridentate condition in the upper triturating surface, also unique among podocnemidids but
comparable to the condition in the kinosternid Claudius. This skull has previously been identified
as the shell-based genus Stupendemys, but there are no associations supporting this assignment.
The type specimens are separated by over 2000 km, and other large podocnemidid taxa are known
from this time and region. Although relatively large with a condylobasal skull length of about
17 cm, Caninemys was probably about half the size of Stupendemys. Phylogenetic analysis of 63
skeletal characters provides evidence that this new taxon nests within the family Podocnemididae
as follows: (Bauruemys (“Roxochelys” (Podocnemis (Caninemys (Dacquemys ((Erymnochelys,
Peltocephalus) (Neochelys (Shweboemys, Stereogenys)))))))).
INTRODUCTION Gaffney et al., 2006 for summary). This
activity has been precipitated largely by new
During the last 20 years there has been efforts to examine both fossil and recent
important progress in our understanding of pleurodires, and by the discovery of high-
the diversity of fossil side-necked turtles (see quality fossil skull material. Many of these
1 Collegium of Natural Science, Eckerd College, St. Petersburg, FL 33711 ([email protected]).
2 Division of Paleontology, American Museum of Natural History, New York, NY 10024 ([email protected]).
3 Departmento Nacional da Produgao Mineral, Rio de Janeiro, Brasil.
Copyright © American Museum of Natural History 2009 ISSN 0003-0082
2 AMERICAN MUSEUM NOVITATES NO. 3639
fossils are not directly related to living side¬ Williams, 1954; Neochelys Bergounioux,
necked turtles but represent completely ex¬ 1954; Shweboemys Swinton, 1939; Stereogenys
tinct lineages. They are essential to our Andrews, 1901; Papoulemys Tong, 1998; and
realization that pleurodires have a more Bair demy s Gaffney and Wood, 2002.
complex history than previously thought. Portezueloemys de la Fuente, 2003, Hama-
Others represent extinct taxa that are closely dachelys Tong and Buffetaut, 1996, and
related to the few living lineages of side- Brasilemys Lapparent de Broin, 2000, constitute
necks. Together, they reveal a previously sister taxa to the family Podocnemididae and
overlooked, remarkable diversity in this were placed in the epifamily Podocnemidinura
clade. along with the family Podocnemididae by
Following current practice (Antunes and Gaffney et al. (2006). Higher taxa below the
Broin, 1988; Broin, 1988; Meylan, 1996; family level within the Podocnemididae are not
Gaffney et al., 1998; Tong, 1998; Gaffney et used here (see Lapparent de Broin, 2000);
al., 2006) the family Pelomedusidae is re¬ instead, we have made comparisons and discus¬
stricted to the living genera Pelusios and sions using genera.
Pelomedusa, and the family Podocnemididae This paper is part of a larger study of
is outside of the Pelomedusidae. Recognition pelomedusoid pleurodires that has so far
of the Podocnemididae as a family-level resulted in a revision of three of five fami¬
taxon is useful because there is clear evidence lies of the Pelomedusoides: Euraxemy-
that the long accepted family Bothremydidae didae, Araripemydidae, and Bothremydidae
(Baur, 1891) is the sister group to the (Gaffney et al., 2006). Additional data
Podocnemididae and not to the family have been assembled for members of the
Pelomedusidae. For discussion of the suffix Podocnemididae, and this description is un¬
for Podocnemididae see Gaffney et al. (2006: dertaken in light of our understanding of skull
45). Although there are only three living morphology in members of that family
genera belonging to the Podocnemididae, (Gaffney et al., 2006: 45). Podocnemis is
Podocnemis and Peltocephalus from included as the six living species, each of
South America, and Erymnochelys from which has been scored from multiple skulls, as
Madagascar, the fossil record of this is the case for the two other living genera,
family shows that it was quite large, mor¬ Peltocephalus and Erymnochelys. Dacquemys
phologically diverse, and geographically from the early Oligocene of Egypt is included
widespread. on the basis of our restudy of this taxon
In the past, the name Podocnemis has been (Gaffney et al., 2002). Shweboemys from the
applied to shell taxa that show overall Pliocene(?) of Burma and Stereogenys from
similarity of the shell to that of the living the Eocene of Egypt are included on the basis
members of this genus. Because of the highly of Gaffney’s studies of this clade that includes
conservative nature of the pelomedusoid shell the genus Bair demy s (Gaffney and Wood,
(Gaffney et al., 2006), turtles in what are now 2002; Gaffney et al., 2008). Neochelys from the
recognized as separate families of the Eocene of southern Europe is included on the
Pelomedusoides were placed in the genus basis of study of the complete skull of the type
Podocnemis (Schmidt, 1940; Zangerl, 1948). of N. arenarum in the MNHN and figures and
Thus, in this study we follow Gaffney (1988) descriptions in Broin (1977). Our understand¬
and use Podocnemis in a restricted sense ing of the somewhat problematic genus
to include only the living species plus P. “Roxochelys” is based on AMNH 14444 and
bassleri and an undescribed Podocnemis from THUg 2160, nearly complete skulls from the
Lewellyn Price’s Acre 34 locality, both of Paleocene of Bolivia. Hamadachelys (Tong
which are very similar to living P. expansa and Buffetaut, 1996) from the Cenomanian of
(there are others that are valid). Morocco has also been included on the basis
Described fossil genera that belong to of material listed in Gaffney et al. (2006).
the Podocnemididae and are known from Bauruemys from the Late Cretaceous of Brazil
skull material include Bauruemys Kischlat, is included on the basis of skulls and
1994; Roxochelys Price, 1953; Dacquemys postcranial material in the AMNH, MCZ,
2009 MEYLAN ET AL.: CANINEMYS, MIOCENE SIDE-NECKED TURTLE 3
and DNPM. The relevant outgroups are of greater Amazonia. Lapparent de Broin
discussed in Gaffney et al. (2006). (2000: 72) discussed the phylogenetic posi¬
Among the many fossils of the Podocne- tion of Stupendemys but apparently did this
mididae is a large skull, DNPM-MCT 1496- on the basis of two isolated cervical vertebrae
R, from the late Tertiary of western Brazil described by Bocquentin and Negri (1993)
that was mentioned by Lapparent de Broin that are considered to represent this taxon.
et al. (1993) and Gaffney et al. (1998). There is no mention of DNPM-MCT 1496-
Because there is no corroboration for an R in Lapparent de Broin’s (2000) assign¬
assignment to any shell-based taxon at this ment of Stupendemys to the subfamily
time, we describe this skull as new. Podocnemidinae.
Lapparent de Broin et al. (1993) included Bocquentin and Melo (2006) included the
a description of three isolated elements (a three isolated elements described by
costal bone, a peripheral bone, and a Lapparent de Broin et al. (1993), the two
humerus) representing one or more large isolated vertebrae described by Bocquentin
podocnemidids from the late Miocene-early and Negri (1993), and six other isolated
Pliocene of southwestern Amazonia. elements in their hypodigm of a new species,
DNPM-MCT 1496-R is mentioned under Stupendemys souzai. They did not mention
the description of a first left peripheral DNPM-MCT 1496-R. There was no attempt
(UFAC 1294). Lapparent de Broin et al. by these authors to justify their assumption
(1993) interpreted the morphology of this that these 11 solitary elements represent a
peripheral as indicating the presence of a single taxon, and some of them may not even
large, deep midline notch in the nuchal belong to Stupendemys. For maps and local¬
bone, as is known in the type of ities of isolated skeletal elements referred
Stupendemys geographicus Wood, 1976, to cf. Stupendemys, Stupendemys sp., and
and they used this morphology to assign Stupendemys souzai from the southwestern
the isolated peripheral to Stupendemys sp. Amazonian region (i.e., Alto Rio Acre, Rio
In their discussion of this peripheral ele¬ Alto Purus, Rio Alto Jurua), see Lapparent de
ment, they observed that the nearly fully Broin et al. (1993), Bocquentin and Negri
roofed condition in DNPM-MCT 1496-R (1993), Gaffney et al. (1998), and Bocquentin
would be consistent with the deeply notched and Melo (2006).
shell of Stupendemys, but they proceeded to We are in agreement with Lapparent de
say that other well-roofed genera in the Broin et al. (1993) that there is probably more
Podocnemididae, such as Podocnemis itself, than one very large podocnemidid in the late
lack an anterior notch in the shell. They Miocene of Amazonia. As most of this
concluded that (Lapparent de Broin et al., material is shell-based, fragmentary, and
1993:664): not even overlapping morphologically, we
describe DNPM-MCT 1496-R as a new skull-
it is impossible to link the two characters based taxon in order to provide a name that
[large nuchal notch and complete skull can be used in a phylogenetic framework of
roof] in every case and to refer the DNPM other Pelomedusoides. Associating the
skull to Stupendemys. It might represent a Stupendemys shell with DNPM-MCT 1496-
new form, more similar to Podocnemis by R could be adding a chimera to a phyloge¬
its shell. netic analysis. On the other hand, we do think
that there is a higher probability that the
Indeed, the isolated skulls of Podocnemis lower jaw, LACM 141498, does belong to
bassleri Williams (1956) from Amazonian Caninemys (see “Discussion”). We realize
Peru and a large and undescribed Podo¬ that naming this new skull-based species
cnemis skull from Lewellyn Price’s Acre 34 will make it impossible to identify postcranial
locality (in the Acre region), all of which are fragments of large podocnemidids from the
comparable in size to living P. expansa, Amazonian Tertiary to genus, but this is
suggest that other large podocnemidids were a more realistic reflection of the present
present in the late Miocene/early Pliocene situation.
4 AMERICAN MUSEUM NOVITATES NO. 3639
ABBREVIATIONS Anatomical Abbreviations
Institutional Abbreviations bo basioccipital
bs basisphenoid
AMNH American Museum of Natural
ex exoccipital
History
fr frontal
DNPM- Departmento Nacional de Produgao
ju jugal
MCT Mineral, Divisao de Geologia e
mx maxilla
Mineralogia, Museu de Paleon-
op opisthotic
tologia, Rio de Janeiro, Brazil
pa parietal
LACM Natural History Museum of Los
pal palatine
Angeles County, Los Angeles
MCZ Museum of Comparative Zoology- pf prefrontal
Harvard University, Cambridge, pm premaxilla
Mass. po postorbital
MNHN Museum National d’Histoire Na- pr prootic
turelle, Paris pt pterygoid
THUg Teikyo Heisei University, Chiba, qj quadratojugal
Japan qu quadrate
UFAC Vertebrate Paleontology Collection, so supraoccipital
Universidade Federal do Acre, Rio sq squamosal
Branco, Brazil vo omer
SYSTEMATICS ventral, toothlike process, which, together
with a single process formed on the midline
of the premaxillae, form a tridentate condition
ORDER TESTUDINES LINNAEUS, 1758 in the upper triturating surfaces, unique
among pleurodires. The entire animal was
MEGAORDER PLEURODIRA COPE,
probably smaller than Stupendemys geogra-
1864
phicus Wood, 1976.
HYPERFAMILY PELOMEDUSOIDES
COPE, 1868 Caninemys tridentata, new species
EPIFAMILY PODOCNEMIDINURA
Type Specimen: DNPM-MCT 1496-R, a
COPE, 1868
nearly complete skull (figs. 1-4) collected by
FAMILY PODOCNEMIDIDAE COPE, L.I. Price in 1962.
1868 Type Locality: Locality 28 of L.I. Price,
Volta de Pedra Pintada, upper Rio Jurua,
Caninemys, new genus Acre, Brazil (fig. 5).
Type Species: Caninemys tridentata, new Horizon: Vertebrate fossils from the up¬
species. per Rio Jurua are typically found in two beds
Distribution: Late Tertiary, Miocene, of associated with the Ucayali Unconformity
Acre, Brazil. (Gaffney et al., 1998, Campbell et al., 2000).
Etymology: Named for the bulldog ap¬ They come from late Miocene Red Beds of the
pearance of the skull and the large maxillary Contamana Group that he below the uncon¬
processes in the position of mammalian canines. formity (“Huayquerian Beds” of Lapparent
Diagnosis: A podocnemidid pleurodire de Broin et al., 1993), or they come from the
with a well-developed processus trochlearis Acre Conglomerate unit of the Madre de Dios
pterygoidei, quadrate-basioccipital contact, Formation, which overlies the unconformity
and a large cavum pterygoidei; unique among (Campbell et al., 1985, 2001). Paleochannels
podocnemidids (and all other turtles) in filled with younger sediments that are known
having greatly inflated maxillae, each with a to produce vertebrate fossils elsewhere in Acre
2009 MEYLAN ET AL.: CANINEMYS, MIOCENE SIDE-NECKED TURTLE 5
Fig. 1. Caninemys tridentata, new genus and species, DNPM-MCT 1496-R holotype. Partially restored
views of skull. A, Dorsal; B, ventral; C, lateral. (B. Degner, del.)
are apparently not accessible along the upper the Neogene of Acre. We have not been able to
Rio Jurua (Campbell et al., 2000). It is confirm an origin from the “Huayquerian
unlikely that this skull is from a younger Beds” below the Ucayali Unconformity.
horizon. The Acre Conglomerate is also Discussion of a lower jaw (see below) that
considered to be of late Miocene age by might be assigned to this taxon and other
Frailey (1986) and Campbell et al. (2001). geologic references are in Gaffney et al. (1998).
Thus, we assign DNPM-MCT 1496-R to a Diagnosis: Same as for the genus.
late Miocene age, even though we do not Etymology: The species epithet is based
know with certainty from which side of the on the tridentate appearance of the skull that
Ucayali Unconformity it has come. is most clearly seen in anterior view.
Lapparent de Broin et al. (1993) included this Discussion: Although this taxon cannot
skull among turtle material they reported from be differentiated from the shell-based
“Huayquerian Beds” and referenced the work Stupendemys geographicus Wood, 1976, be¬
of Campos and de Broin (1981). The latter cause there is no overlap in presently known
reference lists this material only as coming from morphology, it is likely that Caninemys is
6 AMERICAN MUSEUM NOVITATES NO. 3639
Fig. 2. Caninemys tridentata, new genus and species, DNPM-MCT 1496-R holotype. Partially restored
ventral view. (F. Ippolito, del.)
significantly smaller than the Venezuelan Stupendemys were present in the Acre region
Stupendemys. Using skull-shell ratios of re¬ and that Caninemys is the skull of one of these,
cent specimens of Podocnemis expansa and but this is only speculation.
other recent podocnemidid species, it is
hypothesized that the shell of Caninemys DESCRIPTION
would be less than 4-5 feet in length rather
than the 7-foot plus length of Stupendemys. Only the skull of the type, DNPM-MCT
For comparison, the largest skulls of recent 1496-R, is known (figs. 1-4). It is uncrushed
Podocnemis expansa have a condylobasal and nearly complete. The posterior margin of
length of about 12 cm (Williams, 1956) and the skull roof is missing on the right side. On
the Mio-Pliocene Podocnemis bassleri the left side, both the otic capsule and the
(Williams, 1956; very similar to P. expansa in posterolateral part of the skull roof have been
morphology) is 15.7 cm in length, compared to lost. However, since these areas are preserved
about 16.5-17.0 cm for Caninemys. It is of on the right side, it is possible to give a full
course possible that smaller species of description of the skull of this new taxon. The
2009 MEYLAN ET AL.: CANINEMYS, MIOCENE SIDE-NECKED TURTLE 7
Fig. 3. Caninemys tridentata, new genus and species, DNPM-MCT 1496-R holotype. A, Dorsal; B,
ventral; C, right lateral; D, posterior; E, left lateral; F, anterior. (B. Degner, del.)
front half of this skull is massive. The maxillae extensive, but is comparable to Podocnemis,
are remarkably large and thick, and the cheek and that the orbits were somewhat dorsally
emargination is reduced. In dorsal view it is oriented also as in Podocnemis. Measurements
clear that the temporal emargination was not of this skull are given in table 1.
AMERICAN MUSEUM NOVITATES NO. 3639
Fig. 4. Caninemys tridentata, new genus and species, DNPM-MCT 1496-R holotype. A, Dorsal; B,
ventral; C, right lateral; D, posterior; E, left lateral; F, anterior. (B. Degner, del.)
Dermal Roofing Elements were present. The absence of nasals is typical
of all members of the Pelomedusoides (char.
Nasals: The snout of Caninemys is very 1). Nasals are present in nearly all chelids and
well preserved, and it is clear that no nasals
in primitive cryptodires.
2009 MEYLAN ET AL.: CANINEMYS, MIOCENE SIDE-NECKED TURTLE 9
Fig. 5. Map of southwestern Amazonia showing the localities for Caninemys and other fossil turtles
discussed in the text. 1. Type locality of Stupendemys geographicus (Wood, 1976). 2. L.I. Price locality 28,
type locality of Caninemys tridentata (DNPM-MCT 1496-R). 3. LACM locality 5994, source of LACM
141498, a lower jaw that might represent Caninemys (see “Discussion”).
10 AMERICAN MUSEUM NOVITATES NO. 3639
TABLE 1
Measurements of Skull of Caninemys tridentata (DNPM-MCT 1496-R) (in cm)
(For positions of measurements see Gaffney et al., 2006: fig. 315)
A. Midline length as preserved (estimated original length 16.5-17.0) 16.la
B. Maximum width (right side half width doubled) 19.2
C. Width between orbits 3.9
Dl. Width of left orbit 2.6
D2. Width of right orbit 2.6
E. Width of external nares 4.0
F. Width of internal nares 4.7
G. Maximum height at quadrate 9.0
H. Width of skull at middle of orbits 11.8
I. Length from anterior margin of prefrontals to posterior margin of supraoccipital 16.5a
Jl. Height of left orbit 3.7
J2. Height of right orbit 3.6
K. Skull height at occipital condyle 7.2
L. Anterior width of triturating surface (including labial ridge) 3.4
Anterior width of triturating surface (medial to labial ridge) 2.3
M. Posterior width of triturating surface (including labial ridge) 4.2
Posterior width of triturating surface (medial to labial ridge) 3.6
N. Width of palate across foramina palatinum posterius 8.3
O. Length from front of skull to posterior edge of condylus articularis 15.6
aDamaged.
Prefrontal: The prefrontal is a relatively olfactorius and the medial part of a well-
large element making up the anterior part of developed septum orbitotemporalis. The sep¬
the snout. It contacts the other prefrontal tum orbitotemporalis of Caninemys is the
medially, the frontal posteriorly, and the same relative size as in Podocnemis. A defining
maxilla ventrolaterally. It forms the anterodor- feature of the members of the genus
sal quarter of the orbit and the dorsal part of Podocnemis is the presence of a groove on
the external nares and the fossa nasalis. A the dorsal surface of the frontal (char. 3). No
strong ventral ridge on the posterior part of the such groove is present in Caninemys or other
prefrontal continues onto the ventral surface of members of the Pelomedusoides.
the frontal and defines the lateral limits of the Parietal: The parietals are incomplete
fossa nasalis. This element in Caninemys differs posteriorly. This element makes up the largest
from Podocnemis but is like other podocnemi- part of the skull roof and contributes to the
dids in having no groove on the dorsal surface lateral wall of the braincase. It contacts the
at the midline (see “Frontal”). other parietal on the midline, the frontal
Frontal: The frontal is relatively small but anteriorly, the postorbital anterolaterally,
makes up much of the skull roof between the and the quadratojugal posterolaterally. The
relatively small orbits. The orbits of Caninemys contact of the parietal to the quadratojugal is
are oriented somewhat upward rather than broad in Caninemys (char. 4). This is a derived
directly outward (char. 2). This is the condition condition relative to that seen in pelomedu-
found in Podocnemis, Bauruemys, and other sids, Araripemys, and other pleurodires, in
podocnemidids except for the clade that which these elements are separated by deep
includes Shweboemys, Stereogenys, Neochelys, temporal emargination. Narrow contact be¬
Erymnochelys, Peltocephalus, and Dacquemys. tween the quadratojugal and parietal is
The frontal contacts the prefrontal anteri¬ observed in the primitive pelomedusoid,
orly, the other frontal on the midline, the Euraxemys, most Taphrosphyini (Gaffney et
postorbital laterally, and the parietal posteri¬ al., 2006), Hamadachelys, and Bauruemys. The
orly. In addition to forming the middle part of broad contact seen in Caninemys and all other
the skull roof and the dorsal margin of the podocnemidids is scored as a separate, further
orbit, it forms the posterior part of the sulcus derived condition.