5 Joyce Bell 2004

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/40663845
A review of the comparative morphology of extant testudinoid turtles (Reptilia:

Testudines)
Article · January 2004

Source: OAI
CITATIONS READS
71 735
2 authors:
Walter Joyce Christopher Bell

Université de Fribourg University of Texas at Austin
137 PUBLICATIONS 3,534 CITATIONS 82 PUBLICATIONS 1,960 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Quaternary Arvicoline Rodent Paleontology and Biochronology View project
Patterns of evolution of side-necked turtles (Pleurodira) View project
All content following this page was uploaded by Walter Joyce on 20 August 2015.
The user has requested enhancement of the downloaded file.

2004 Asiatic Herpetological Research Vol. 10, pp. 53-109
A Review of the Comparative Morphology of Extant Testudinoid Turtles

(Reptilia: Testudines)
WALTER G. JOYCE1 AND CHRISTOPHER J. BELL2
1Department of Geology and Geophysics, Yale University, New Haven, CT 06511

E-mail: [email protected]
2Department of Geological Sciences, The University of Texas at Austin, Austin, TX 78712
E-mail: [email protected]
Abstract. - With an expansive geographic distribution, an excellent fossil record, and over 140 recognized extant
species, testudinoid turtles constitute one of the most diverse and widespread clades of turtles. The current under-
standing of the distribution of morphological characters among testudinoid turtles is poor. Improved knowledge will
help to facilitate accurate identification of fossil remains, and to provide a reliable morphological data set for phylo-
genetic analyses. We provide a critical review of skeletal and scute characters commonly utilized in previous system-
atic analyses of Testudinoidea. Description and illustration of character states, discussion of their distribution within
Testudinoidea, and polarity determinations for 93 characters are provided. Our preliminary results indicate that onto-
genetic changes in skeletal structure are an important source of variation within Testudinoidea. Sexual variation, onto-
genetic variation, and intra- and inter-population variation are inadequately documented for most testudinoid taxa.
Furthermore, data matrices of morphologic characters in the existing literature must be carefully reconsidered.
Previously published morphologic data provide reasonably strong support for the monophyly of 'Testudinidae.' Strong
morphologic support for a monophyletic 'Emydidae' is lacking, and 'batagurid' monophyly has not been rigorously
tested in the literature. Because a new research cycle centered on testudinoid phylogeny is now under way, it is essen-
tial to critically re-examine the underlying assumptions and working hypotheses that have governed this field of study
over the last 20 years.
Key words. - Testudines, Testudinoidea, Testudinidae, Emydidae, Bataguridae, Geoemydidae, morphology, systemat-
ics
Introduction subclades of testudinoid turtles is desirable. A more

thorough understanding of morphological data sets will
Pond turtles and land tortoises (collectively, provide not only a means by which molecular trees may
Testudinoidea) form one of the largest and most wide- be independently assessed, but also will form an essen-
spread clades of living turtles, with more than 140 extant tial foundation for diagnosing and interpreting fossil
species and an almost worldwide distribution. The dis- specimens. This in turn will facilitate the integration of
covery and description of many new fossil testudinoids fossil taxa into future systematic analyses, and will
in the last half century, combined with the emergence enhance our understanding of the paleobiogeography
and ascendancy of molecular techniques in systematics, and divergence times of extant lineages.
provide new opportunities to explore the evolutionary The recent flurry of published works appears to rep-
history of the group in unprecedented detail. resent the beginning of a new research cycle (sensu
Concomitant with the appearance of these new data sets Kluge, 1991) in testudinoid systematics. We suggest that
and analytical techniques comes an increasing apprecia- an important part of this cycle will be a critical re-exam-
tion for conservation efforts to preserve these turtle lin- ination of the working hypotheses that have governed
eages and help to secure their future in the face of testudinoid systematics since the publication of
increasing human predation and habitat encroachment. McDowell’s (1964) seminal work on the group. A key
This is true especially for the Asian representatives of component of this will be the assessment of fundamen-
this clade (e.g., van Dijk et al., 2000) but also is relevant tal, often unstated, assumptions that underlie current
at a more generalized and inclusive level (e.g., Rhodin, hypotheses of relationship. Our contribution to this
2000). research cycle is the first critical reappraisal of morpho-
Our recent attempts to diagnose fossil testudinoids logical characters applied to testudinoid systematics
reliably and to place them within a phylogenetic context since the work of Hirayama (1985). The emerging
led to the recognition that a critical re-evaluation of mor- improvement in our understanding of testudinoid rela-
phological data and purported synapomorphies for the tionships based on molecular sequence data will certain-
© 2004 by Asiatic Herpetological Research

Vol. 10, p. 54 Asiatic Herpetological Research 2004
ly result in numerous new questions (e.g., regarding valid species by Ernst and Barbour (1989), with the
paleobiogeography, the timing of sequence and evolu- exception of texana, which they placed within ‘concin-
tionary divergences, and the evolution of morphological na.’ Generic allocations for testudinoid species varied
adaptations) that will demand a clearer understanding of widely over the last 50 years and are subject to differing
testudinoid morphology. opinions today, particularly because the monophyly of
The purpose of this paper is to present a preliminary many testudinoid genera remains untested. We conse-
revision and discussion of the morphological characters quently suppress the use of generic names wherever pos-
previously utilized in investigations of testudinoid sys- sible and use species epithets only. This procedure also
tematics. Our goal here is not to produce a phylogenetic has the advantage of precisely associating observations
hypothesis (indeed, we deliberately eschew such a pro- with species only instead of higher taxonomic cate-
duction), but rather to evaluate the morphological data gories. Most extant turtles have distinct species names,
that have been, and will be, used to generate such but among those turtles discussed in this review the
hypotheses. To enhance our discussion and facilitate species epithets insculpta, nelsoni, oculifera, ornata,
improved communication about testudinoid morpholo- and platynota each appear twice (insculpta under
gy, we provide illustrations of all characters states we Glyptemys and Carettochelys; nelsoni under Pseudemys
discuss. and Terrapene; oculifera under Graptemys and
Psammobates; ornata under Pseudemys and Terrapene;
Abbreviations. - Institution and collection abbrevia- platynota under Geochelone and Notochelys). For clari-
tions: CAS, California Academy of Sciences, San ty in these instances, we indicate our usage with a sin-
Francisco, California; CM, Carnegie Museum of Natural gle-letter generic abbreviation. A complete list of all cur-
History, Pittsburgh, Pennsylvania; FMNH, Field rently recognized testudinoid species and all outgroup
Museum of Natural History, Chicago, Illinois; IVPP, species used herein is provided in Appendix 2 together
Institute of Vertebrate Paleontology and with a list of their various generic assignments used in
Paleoanthropology, Beijing, China; KU, The University the last 50 years.
of Kansas Natural History Museum, Lawrence, Kansas; Our use of the classic higher categories is always
LMNH, Louisiana Museum of Natural History, Baton restricted to their phylogentic crown. ‘Emydinae’ (sensu
Rouge, Louisiana; MCZ, Museum of Comparative McDowell, 1964) are also referred to as ‘Emydidae,’
Zoology, Harvard University, Cambridge, ‘emydids,’ or North American pond turtles;
Massachusetts; TNHC, Texas Natural History ‘Batagurinae’ (sensu McDowell, 1964) as ‘Bataguridae,’
Collections, Texas Memorial Museum, Austin, Texas; ‘batagurids’ or Asian pond turtles; and ‘Testudinidae’
TUMNH, Tulane University Museum of Natural (sensu McDowell, 1964) as ‘testudinids’ or (land) tor-
History, New Orleans, Louisiana; YPM, Yale Peabody toises. We make no a priori assumptions of monophyly
Museum, New Haven, Connecticut. for any of these categories, and retain single quotations
Abbreviations used in figures: AB, abdominal around these names throughout the text to emphasize
scute; af, articular facet; an, angular; bo, basioccipital; our uncertainty.
CE, cervical scute; co, costal bone; ent, entoplastron; We attempted to examine most significant morpho-
epi, epipterygoid; fdm, foramen dentofaciale majus; fr, logical characters commonly utilized in systematic stud-
frontal; fpp, foramen palatinum posterius; HU, humeral ies of testudinoids, but the majority of our observations
scute; hyo, hyoplastron; hypo, hypoplastron; ju, jugal; concern the skeletal system and scute characters. Almost
MA, marginal scute; mx, maxilla; ne, neural bone; pa, all characters were derived from the literature. Major
parietal; pal, palatine; PEC, pectoral scute; pf, pre- sources for each category were: ‘Batagurinae’
frontal; PL, pleural scute; pm, premaxilla; po, postor- (Hirayama, 1985; McCord et al., 1995; Yasukawa et al.,
bital; pt, pterygoid; qj, quadratojugal; qu, quadrate; VE, 2001); ‘Testudinidae’ (Crumly, 1982, 1985, 1994); and
vertebral scute; vf, vomerine foramen; vo, vomer. ‘Emydinae’ (Gaffney and Meylan, 1988; Burke et al.,
1996). Additional characters were also found in
Material and Methods Mlynarski (1976), Shaffer et al. (1997), and other
sources cited in the character discussions. With few
exceptions, morphological features were examined on
We examined 309 testudinoid specimens representing
specimens themselves; evaluations based on previously
93 species, but focused our efforts on 46 representative
published literature are indicated where applicable.
species. The list of specimens examined is provided in
Sexual dimorphism, ontogenetically influenced poly-
Appendix 1. Turtle shell nomenclature follows Zangerl
morphisms, and geographic variation in morphology are
(1969) and cranial nomenclature follows Gaffney
not well explored in testudinoid turtles. These areas are
(1972). Of the 46 focal species, most were recognized as
in need of much more research. A full exploration of
2004 Asiatic Herpetological Research Vol. 10, p. 55
such variation is beyond the scope of this work, but we many authors. A brief summary is given here.
are able to make some preliminary observations regard- During the second half of the 19th century, a num-
ing morphological change through ontogeny in some ber of attempts were made to work out higher-level tes-
anatomical systems. tudinoid relationships and to apply taxonomic conven-
Polarizing characters with the help of outgroups tions that were designed (to a greater or lesser extent) to
proved to be a difficult task, mostly because all relevant communicate conceptualizations of these relationships.
extant sister taxa are highly specialized after more than In his synopsis on the turtles of North America, Agassiz
65 million years of independent evolution. Furthermore, (1857) united all pond turtles into the Emydoidae and
hypotheses of the systematic relationships of the major subdivided this group into a three monotypic subfami-
groups of cryptodires reveal a highly unstable picture lies (Deirochelyoidae for reticularia, Evemydoidae for
(e.g., Bickham, 1981; Gaffney, 1975, 1985; Gaffney et blandingii, Cistudinina for T. ornata and carolina) fol-
al., 1991; Shaffer et al., 1997) making it impossible to lowed by the subfamilies Clemmydoidae (for G. insculp-
make any a priori decisions regarding the succession of ta, guttata, marmorata, and muhlenbergii) and
outgroups. We consequently assessed polarity for most Nectemydoidae (for those species currently placed in the
characters by examining select outgroup taxa and the genera Pseudemys, Trachemys, Graptemys, Malaclemys
ingroup taxa. Where polarity is not clear from outgroup and Chrysemys). Most land tortoises were placed in
comparison, we sometimes relied upon ingroup com- Testudinidae by Theobald (1868); he also included all
monality. To allow full transparency, we discuss every ‘leaf turtles and tortoises’ (e.g., amboinensis, emys, den-
polarity decision at the end of each character descrip- tata, grandis, tricarinata) in Geoemydidae, and an
tion. eclectic group of aquatic turtles, including mega-
Outgroup taxa include the cryptodires caretta, cephalum, serpentina, kinosternids, and all remaining
odoratus, serpentina, and spinifera, and the pleurodires testudinoids, in Emydidae. Subsequently, all land tor-
gibba, siebenrocki, subglobosa, and subrufa. For a num- toises (including emys and impressa) were united in the
ber of characters, especially of the shell, neither ingroup Testudinidae by Gray (1870). Those species currently
nor outgroup analysis of extant taxa proved useful. In placed in Pseudemys and Trachemys were assigned by
these instances, polarity was based on literature descrip- Gray (1870) to the Pseudemydae; the Asian taxa baska,
tions of the “lindholmemydid” taxa Gravemys, borneoensis, thurjii, kachuga, and ocellata were
Lindholmemys, and Mongolemys. These Cretaceous, assigned to the Bataguridae, and all hinged pond turtles,
Asian, fossil taxa are not well described in the literature, ‘true terrapins,’ and ‘snail-eating pond turtles’ to the
but sufficient material and description exists to use these Holarctic families Cistudinidae, Emydidae, and
taxa to help polarize character states (e.g., Khosatzky Malaclemmydae, respectively.
and Mlynarski, 1971; Sukhanov, 2000; Danilov and Despite these early attempts, most subsequent
Sukhanov, 2001). The group may not be monophyletic, authors (e.g., Boulenger, 1889; Siebenrock, 1909;
but putative members currently are hypothesized to sit Lindholm, 1929; Smith, 1931; Bourret, 1941) ignored
along the phylogentic stem of Testudinoidea (Danilov (or were unaware of) these works and simply divided all
and Sukhanov, 2001). We purposefully did not use the testudinoid turtles into two speciose subgroups: tortois-
fossil taxon ‘Echmatemys’ as an outgroup taxon es (Testudinidae or Testudininae) and pond turtles
(Hirayama, 1985), because its phylogenetic position out- (Emydidae or Emydinae). This situation remained static
side of Testudinoidea or even ‘Batagurinae’ is not suffi- for nearly 100 years until the comprehensive and influ-
ciently demonstrated. ential work of McDowell (1964). He not only divided all
All figures were produced using digital photogra- known pond turtles into several species complexes
phy and processed using Adobe Photoshop. Images were (‘Hardella,’ ‘Batagur,’ ‘Orlitia,’ ‘Geoemyda,’
digitally enhanced using the burn and burnish tools and ‘Chrysemys,’ ‘Deirochelys,’ and ‘Emys’ complexes), but
the unsharp mask filter option. also concluded that pond turtles can be divided clearly
into two subgroups, the predominantly North American
Taxonomic and Systematic Background. - Despite the ‘Emydinae’ and the Asian and central American
increased attention directed towards testudinoids by sci- ‘Batagurinae.’ Furthermore, McDowell (1964) reasoned
entists, hobbyists, and nonprofessional enthusiasts in the that tortoises are not the sister group of pond turtles, but
last thirty-five years, our collective conceptualization of rather were likely derived from a ‘batagurine’ ancestor.
the higher-level (beyond the specific and generic) sys- These conclusions were later corroborated by the first,
tematics within this clade remained virtually unchanged and to date only, comprehensive morphological cladistic
since the work of McDowell (1964). The various tax- analysis of ‘batagurine’ systematics (Hirayama, 1985).
onomies in current use owe their existence in large part The influence of McDowell’s (1964) work is best
to historical contexts that are not well appreciated by understood when considering its continuous impact on
subsequently proposed phylogenies. Despite differences produce testable results. It is toward this end that we
of opinion regarding generic- and species-level system- offer our critical reappraisal of morphological characters
atic arrangements, virtually all major synthetic works in in testudinoids.
the last thirty years followed McDowell’s (1964) subdi-
vision of pond turtles into the ‘Batagurinae’ and Results and Discussion
‘Emydinae’ (e.g., Mlynarski, 1976; Pritchard, 1979;
Ernst and Barbour, 1989), even though ‘Batagurinae’ Cranium
may best be regarded as a paraphyletic taxon (1) Shape of the fissura ethmoidalis; 0 = narrow or
(McDowell, 1964; Hirayama, 1985). The fundamental closed, keyhole-shaped, Fig. 1; 1 = very wide, Fig. 2
division proposed by McDowell is also reflected in more (modified from Crumly, 1982, 13; Hirayama, 1985, 1;
recent studies centered on using various molecular tech- McCord et al., 1995, 5).
niques to elucidate phylogeny, the majority of which The general configuration of the fissura in ‘emy-
dealt with treatments of in-group relationships within dids’ and ‘batagurids’ is keyhole-shaped (McDowell,
one or the other of McDowell’s groups (Sites et al., 1964). Different proportions and widths are apparent
1984; Bickham et al., 1996; Carr and Bickham, 1986; (especially in the ventral part of the fissura), and were
Wu et al., 1999; McCord et al., 2000; Feldman and scored by Hirayama (1985) and Crumly (1982) as dis-
Parham, 2002; Honda et al., 2002; Iverson et al., 2002; crete character states. Our survey of many taxa reveals
Stephens and Wiens, 2003). morphological intermediates, and the expression of var-
Admittedly, the list of autapomorphic characters ious states appears to have an ontogenetic component in
compiled by Crumly (1985) for the ‘Testudininae’ com- which younger individuals exhibit a relatively larger fis-
pellingly corroborates the hypothesis of tortoise mono- sura, which corresponds to a less-ossified nasal cavity.
phyly. However, most characters that currently unite However, a rather significant morphological gap can be
‘Emydinae’ or ‘Batagurinae + Testudininae’ seem to observed between tortoises and pond turtles. For the
support these groupings weakly, because the derived purpose of this review, we lumped Hirayama’s (1985)
states typically are not found within all species of the states into our state 0, and Crumly’s (1982) into our state
ingroup and commonly also are observed in species of 1. The scoring presented by Hirayama (1985) and
the alleged sister group (e.g., Hirayama, 1985; Gaffney McCord et al. (1995) permitted phylogenetic resolution
and Meylan, 1988). In addition, several of the characters within ‘batagurids’, and that of Crumly (1982) within
that purportedly distinguish ‘Batagurinae + tortoises. Our revised scoring permits support only for
Testudininae’ from the ‘Emydinae’ probably should be the hypothesis of a monophyletic ‘Testudinidae.’
considered primitive for the entire group (Gaffney and Polarity: Pleurodires lack a defined fissura eth-
Meylan, 1988). Even if some characters do successfully moidalis. A keyhole-shaped fissura ethmoidalis is pres-
unite a group, monophyly is not established until the ent in spinifera, odoratus, caretta, serpentina, and
involved characters are demonstrated to be derived with- Mongolemys, and this condition is considered primitive
in Testudinoidea. Furthermore, the simple demonstra- for testudinoids.
tion of monophyly for a given group does not automati-
cally imply that it must be the sister to the remaining (2) Medial inflection of the inferior descending process-
taxa. For instance, ‘Emydinae’ may be monophyletic, es of the frontal; 0 = absent, or very small, Fig. 3; 1 =
but monophyly does not necessarily demand that present, well-developed, medial contact present or
‘Emydinae’ be regarded as the sister to ‘Batagurinae + almost present, Fig. 4 (modified from Hirayama, 1985,
Testudininae.’ It is at least plausible that ‘Emydinae’ is 2).
situated within ‘Batagurinae,’ a possibility that is not In most turtles, a gutter (the sulcus olfactorius) is
adequately explored and tested in the literature. formed along the ventral surface of the frontals. This
Similarly, most of the groupings considered by gutter transmits the olfactory nerve. The lateral rims of
Gray (1855, 1870) and Agassiz (1857) were not dis- the sulcus sometimes form processes that descend ven-
cussed in recent literature, even though they might be tromedially to surround the nerve from below
valid. For instance, given the considerable list of mor- (McDowell, 1964). According to Hirayama (1985) these
phological similarities that are shared by hinged turtles processes are well-developed, or are in contact medially,
of the New and Old World (e.g., development of a plas- in ocellata and hamiltonii. We confirm the presence of
tral hinge, reduction of posterior neural elements, fusion well-developed processes in these taxa and add petersi,
of the femoral trochanter, great reduction of the tempo- N. platynota, and all sampled tortoises to the list. We
ral arch) perhaps Gray (1870) was truly visionary in recommend that this character not be subdivided into
uniting these turtles as the ‘Cistudinidae.’ Only a global additional character states, because the descending
cladistic analysis with no a priori assumption regarding processes of the frontals grow larger through ontogeny.
internal relationships can evaluate these alternatives and
Polarity: A medial inflection is absent in all out- of a medial contact with the pterygoid were treated as
groups and the vast majority of the ingroup. We consid- two characters by Hirayama (1985) and Yasukawa et al.
er its presence to be derived. (2001). The scoring for the two characters appears to be
redundant and we followed the recommendation of
(3) Frontal contribution to the orbital rim; 0 = present, Gaffney and Meylan (1988) by combining them.
no prefrontal/postorbital contact on dorsal surface, Fig. We confirm the loss of a medial contact between the
5; 1 = absent, frontal excluded from orbital rim by pre- jugal and the pterygoid in galbinifrons, flavomarginata,
frontal/postorbital contact, Fig. 6 (modified from and mouhotii (Hirayama, 1985; Yasukawa et al., 2001),
Crumly, 1982, 17; Hirayama, 1985, 3; Shaffer et al., but we found this condition to be polymorphic in speng-
1997, 97; Yasukawa et al., 2001, 1). leri (also reported by McCord et al., 1995), and in triju-
Three states for this character were scored by ga. Our observations are concordant with those of Burke
Hirayama (1985) and Yasukawa et al. (2001): frontal et al. (1996).
contribution always or usually present, frontal some- Polarity: A contact between the jugal and pterygoid
times excluded from orbital rim, and frontal always in present in spinifera, odoratus, and serpentina, but is
excluded from orbital rim. Our sample size for many absent (although the two bones closely approach one
taxa does not permit a reliable assessment of intraspecif- another) in caretta. We conclude that the contact
ic variation in this character, and thus our initial scores between the two bones is the primitive condition for tes-
differed for some taxa from those of Hirayama (1985). tudinoids and that their separation is derived, a conclu-
We add petersi and N. platynota to the list of taxa in sion also reached by Hirayama (1985). Our polarity
which the frontal appears always to be excluded. For determination is opposite that used by McCord et al.
those taxa that sometimes exclude the frontals, we con- (1995), who mistakenly claimed to have derived their
firm this polymorphic condition in crassicollis, and add polarity assessment from Hirayama (1985).
agassizii and annulata. Our sample was too small to
confirm the reported polymorphic condition in (5) Contact between jugal and palatine; 0 = absent, Fig.
amboinensis, and pulcherrima by Hirayama (1985) and 9; 1 = present, Fig. 10 (Gaffney and Meylan, 1988,
Yasukawa et al. (2001), but we scored these taxa as F5.4).
polymorphic based on their observations. We also fol- The presence of a contact between the medial
lowed Crumly (1982) by coding pardalis as polymor- process of the jugal and the palatine was used previous-
phic, even though we were not able to observe this in our ly in support of a monophyletic Deirochelyinae
sample. Our coding differs from that of Shaffer et al. (Gaffney and Meylan, 1988). We confirm the formerly
(1997) for Heosemys and reevesii. In their analysis, they observed distribution of this character within ‘emydids’
used spinosa (in which the frontal contributes to the with the exception of reticularia, which does not exhib-
orbital rim), but we used grandis (in which it does not). it a contact. A contact is present in numerous
In both specimens of reevesii available to us, the frontal ‘batagurids,’ such as borneoensis, reevesii, and hamil-
clearly does not participate in the orbital margin. Given tonii, but was absent in all examined members of
the contrary statement by Shaffer et al. (1997), reevesii ‘Testudinidae.’
may be polymorphic for this character. Polarity: A contact between the jugal and the pala-
Polarity: The frontal participates in the orbital rim tine is present in caretta, odoratus, and pleurodires, but
in pleurodires and spinifera, it is excluded in odoratus is absent in spinifera and serpentina. A contact is absent
and caretta, and it is polymorphic in serpentina. No pat- in Mongolemys. We consequently consider its presence
tern is apparent within the ingroup. Given that the to be derived.
frontal clearly contributes to the orbital rim in
Mongolemys, we consider its absence to be derived. (6) Contact of the epipterygoid with the jugal; 0 = clear-
ly absent, Fig. 11; 1 = present, or almost present,
(4) Contact between jugal and pterygoid; 0 = present, epipterygoid forms a long lateral process that approach-
medial process of jugal well-developed and touching the es the jugal, Fig. 12 (Gaffney and Meylan, 1988, F8.1;
pterygoid, Fig. 7; 1 = absent, medial process reduced, Shaffer et al., 1997, 106).
Fig. 8 (modified from Hirayama, 1985, 11, 12; McCord According to Gaffney and Meylan (1988) the
et al., 1995, 3; Burke et al., 1996, 23; Yasukawa et al., epipterygoid and the medial process of the jugal
2001, 4, 5). approach one another or are in contact in reticularia and
The jugal of most testudinoid turtles is expanded at the various species they included in Pseudemys and
its ventral end to form a medial process that contacts the Trachemys. They also noted a contact between these two
pterygoid medially (McDowell, 1964). Presence or elements in species they classified in Graptemys, but the
absence of the medial process, and presence or absence condition in those taxa was interpreted to be a result of
the medial expansion of the jugal and not a lateral process of the parietal and the maxilla in all outgroups.
expansion of the epipterygoid, and consequently was Its presence is considered to be derived.
regarded as non-homologous (Gaffney and Meylan,
1988). We confirm the contact or near contact of these (9) Extent of quadratojugal; 0 = quadratojugal well
two elements in reticularia, decorata, scripta, alaba- developed, firmly attached to jugal, Fig. 15; 1 = quadra-
mensis, P. nelsoni, rubriventris, texana, flavimaculata, tojugal present, contact lost with jugal, Fig. 16; 2 =
geographica, kohnii, nigrinoda, G. oculifera, ouachiten- quadratojugal so heavily reduced that it appears to be
sis, and versa, and also report it in picta and terrapin. absent in many skeletal specimens, Figs. 17, 18 (modi-
Contact was clearly absent in the specimens of barbouri, fied from Hirayama, 1985, 16; Shaffer et al., 1997, 47;
ernsti, and gibbonsi we examined. We made no assess- Burke et al., 1996, 21; McCord et al., 1995, 6; Yasukawa
ments of homology, but accept any contact between et al., 2001, 7, 8).
these two elements as the derived state (as was done by Variation in the structure of the temporal region of
Shaffer et al., 1997). Among ‘batagurids,’ we also found turtles was discussed in detail by Zdansky (1924) and
a close approach in reevesii. comments specific to testudinoids were provided by
Polarity: A contact, or near contact, between the Zangerl (1948) and McDowell (1964). We originally
epipterygoid and the jugal is absent in all outgroups and scored the reduction of the quadratojugal as three differ-
the vast majority of the ingroup. We consider its present characters: loss of contact with the jugal, loss of con-
ence to be derived. tact with the squamosal, and the apparent loss of the
quadratojugal. All five logically possible combinations
(7) Contact of the inferior process of the parietal with were observed, but in most testudinoid turtles the tem-
the medial process of the jugal; 0 = absent, Fig. 13; 1 = poral arch is so slender that the contact between the
present, Fig. 14 (Hirayama, 1985, 13). quadratojugal and squamosal is commonly reduced to a
Our coding differs significantly from that of sliver that would have to be scored as ‘just barely pres-
Hirayama (1985). In reevesii, N. platynota, and bealei ent’ or ‘just barely absent.’ We therefore abandoned our
we found no contact between the parietal and jugal, efforts to evaluate the contact between the quadratojugal
although these were the only three taxa in which and squamosal. Our observations generally agree with
Hirayama (1985: table 2) scored it to be present. those of McDowell (1964), Hirayama (1985), Burke et
However, we found a pronounced contact between these al., (1996), McCord et al. (1995), and Yasukawa et al.
two elements in subtrijuga, a species scored by (2001).
Hirayama (1985: table 2) as lacking such a contact, but We purposefully avoid addressing the apparent lack
shown on his tree (Hirayama, 1985: fig. 2) as a unique of a quadratojugal in many species as an absence,
‘batagurid’ feature convergent with some ‘emydids.’ because previous work by Zdansky (1924) showed that
Among ‘emydids,’ a well-developed contact occurs in the quadratojugal of some ‘batagurids’ is so poorly ossi-
barbouri and other broad-headed species currently clas- fied and connected to the surrounding elements that it
sified in Graptemys. tends to be lost in skeletal specimens (Figs. 17, 18). An
Polarity: There is no contact between the inferior example of this problem can be found among the many
process of the parietal and the medial process of the conflicting statements made regarding the presence of
jugal in all outgroups and the vast majority of the this element in N. platynota (e.g., Smith, 1931; Bourret,
ingroup. A contact is considered to be the derived condi- 1941; McDowell, 1964; Ernst and Barbour, 1989).
tion. Polarity: The quadratojugal is present and firmly
attached to the jugal in all outgroups, with the exception
(8) Contact of the inferior process of the parietal with of chelids. Its reduction is considered to be derived.
the maxilla; 0 = absent, Fig. 13; 1 = present, Fig. 14
(Hirayama, 1985, 14). (10) Contribution of jugal to the rim of upper temporal
Our coding differs from that of Hirayama (1985). emargination (Hirayama, 1985, 15); 0 = absent, Figs.
According to his character matrix (table 2) a contact 19, 20; 1 = present, Fig. 21.
should be present between the inferior process of the Participation of the jugal in the rim of the upper
parietal and the maxilla in reevesii and mouhotii, but his temporal emargination was reported previously in
cladogram (fig. 2) indicated that the presence of a con- hamiltonii and ocellata (Hirayama, 1985). We confirm
tact should be regarded as a uniquely derived autapo- its presence in both species, but in one of the hamiltonii
morphy of subtrijuga. We found no contact in reevesii or specimens we examined (MCZ 120333) the jugal forms
mouhotii. Of the testudinoid species we examined, sub- a significant part of the rim only on one side of the skull;
trijuga is the only one that shows this feature. on the other side, which appears abnormal and likely
Polarity: There is no contact between the inferior represents a teratology, it does not. In all other species
available to us, the jugal does not participate in the rim. and as an independently evolved synapomorphy for
In subtrijuga, the jugal is excluded from the upper tem- Kinosternidae, whereas Shaffer et al. (1997) noted a
poral emargination by narrow extensions of the postor- contact to be present in Kinosternidae, C. insculpta, and
bital and quadratojugal (Fig. 20). megacephalum. Our observations confirm the presence
Polarity: The jugal participates in the upper tempo- of a contact in all of these extant groups. Among testudi-
ral rim of spinifera, but it is excluded in odoratus, caret- noids, subtrijuga is unique in having an extensive con-
ta, serpentina, Mongolemys, and most pleurodires. We tact in lateral view (Fig. 23). In several ‘emydids,’ a con-
consider the participation of the jugal in the rim of the tact is present on the inside of the temporal arch (bar-
upper temporal emargination to be the derived condition bouri, and nigrinoda; polymorphic in geographica, G.
within testudinoids. oculifera, and texana; Fig. 24). For now, we scored all
taxa as present, regardless of whether the contact is vis-
(11) Contact between the quadratojugal and the articu- ible in lateral view, medial view, or both. In several other
lar facet of the quadrate; 0 = absent, Fig. 22; 1 = pres- ‘emydid’ taxa, the bones closely approach one another,
ent, quadratojugal sends a process ventrally along the but do not actually meet, on the inside of the temporal
rim of the cavum tympani and touches the lateral edge arch (alabamensis, ernsti, flavimaculata, gibbonsi,
of the articular facet, Fig. 23 (modified from Hirayama, kohnii, P. nelsoni).
1985, 17). Polarity: A contact is present between the quadrato-
The original character definition (Hirayama, 1985, jugal and maxilla in odoratus, but absent in spinifera,
character 17) is inappropriate, because the jugal does not caretta, serpentina, and Mongolemys. We consider the
contact the articular surface of the quadrate in any turtle presence of a contact to be derived for Testudinoids.
except for madagascariensis and dumerilianus (Gaffney
and Meylan, 1988). However, because Hirayama indi- (13) Medial contact of the maxillae along the anterior
cated in his tree that the only ‘batagurid’ taxon to exhib- margin of the jaw; 0 = absent, Figs. 25, 26; 1 = present,
it this character is subtrijuga, we assume that he was Fig. 27 (modified from Hirayama, 1985, 20; McCord et
referring to a contact between the quadratojugal and the al., 1995, 2; Yasukawa et al., 2001, 10).
articular facet of the quadrate, a characteristic of subtri- In most testudinoids, the anteromedial ends of the
juga only among testudinoids. A contact between the maxillae are separated medially by the premaxillae
two elements was reported previously for reevesii in the along the anterior margin of the jaw (Fig. 25). Hirayama
character matrix published by Hirayama (1985), but we (1985) noted that the maxillae have a medial contact in
conclude that this must be a publishing error, because it some ‘batagurids,’ which was confirmed by McCord et
stands in conflict with his tree. In the specimens of al. (1995) and Yasukawa et al. (2001) for spengleri and
reevesii available to us, there is no contact. It is possible several other species that they included in the genus
that the scoring for subtrijuga and reevesii were flipped, Geoemyda. We found a broad medial contact of the max-
at least in part, in the Hirayama (1985) matrix (in which illae in spengleri and annulata (Fig. 27). In some
the taxa were listed next to one another). species, the maxillae approach one another along the
Polarity: A contact between the quadratojugal and ventral rim of the nasal opening (e.g., amboinensis,
the articular surface of the quadrate is absent in mouhotii, pulcherrima, crassicollis), but a well-devel-
spinifera, but present in odoratus, caretta, and serpenti- oped contact is never present (Fig. 26).
na, and consequently could be considered primitive. Polarity: The maxillae do not meet medially along
However, based on ingroup commonality and the the anterior margin of the jaw in odoratus, caretta, ser-
absence of a contact in Mongolemys, we consider a con- pentina, and Mongolemys. A medial contact is present in
tact to be derived for Testudinoidea. spinifera but only along the ventral border of the exter-
nal nares. We consider a medial contact along the ante-
(12) Contact between quadratojugal and maxilla; 0 = rior margin of the jaw to be the derived condition with-
absent, Fig. 22; 1 = present, Figs. 23, 24 (Hirayama, in Testudinoidea.
1985, 18).
According to Hirayama (1985), among ‘batagurids’ (14) Size of the foramen orbito-nasale; 0 = small, less
a contact between the quadratojugal and maxilla is only than 1/6 of orbit length, Figs. 28, 29; 1 = large, more
present in subtrijuga and reevesii. We did not find a than 1/6 of orbit length, Fig. 30 (modified from
contact in our specimens of reevesii, but confirm its Hirayama, 1985, 33; Gaffney and Meylan, 1988, F9.3,
presence in subtrijuga. F10.2, G10.3, H11.1, H16.3; Crumly, 1982, 25; Crumly,
Gaffney and Meylan (1988) listed a contact 1994, 12).
between the quadratojugal and maxilla as a synapomor- We were cautious when first approaching this char-
phy for Platysternina {megacephalum + ‘Chelydropsis’} acter due to the inconsistent usage and definition of
‘small’ and ‘large’ by various authors. However, after by Hirayama (1985) as one character that combined two
assessing the size of this foramen based on its size rela- morphological features: the size of the foramen palat-
tive to the length of the orbit, we were surprised to see inum posterius (f.p.p.) and participation of the pterygoid
that we were able to reproduce Hirayama’s (1985) scor- in the margin of the f.p.p. Although four possible com-
ing for the ‘batagurids’ without too many difficulties. In binations of these features are logically possible, only
contrast, our initial observations of tortoises were in two were originally included (participation present,
stark contrast to those of Crumly (1982, 1985, 1994) and f.p.p. large; participation absent, f.p.p. small). Gaffney
Gaffney and Meylan (1988). This may be due to the thin and Meylan (1988) also used this character within ‘emy-
nature of the palatine of many tortoises, and the relative dids,’ but their character applied only to the exclusion of
ease with which that part of the palate can be damaged the pterygoid from the f.p.p. We decided to subdivide
during skeletal preparation and handling. Furthermore, Hirayama’s (1985) character into one character that
the foramen becomes progressively more closed with describes the size of the f.p.p. and a second that address-
increased ontogenetic age (Crumly, 1982). We encoun- es the position of the pterygoid relative to the f.p.p.
tered similar problems in attempting to score reticularia We found no difficulty in identifying the f.p.p. as
and blandingii. Because we deem this character to be ‘large’ or ‘small,’ (Figs. 33 and 34, respectively), and no
potentially useful for helping to resolve phylogeny with- ambiguous condition was encountered. Because our
in ‘batagurids’ and ‘emydids,’ we decided to score all character definition only includes two character states,
testudinoids with delicate palatines (i.e., all tortoises, our scorings do not reflect those of other workers with a
reticularia, and blandingii) as ‘unknown.’ We acknowl- more limited target group (e.g., McCord et al., 1995). In
edge that our redefinition of the character is still subjec- juveniles the f.p.p. tends to be larger, but during later
tive and somewhat problematic, but using this definition ontogenetic stages the f.p.p. is slowly reduced in size.
we were able to unambiguously score all the ingroup Polarity: The f.p.p. is small in odoratus, spinifera,
taxa we examined. and most pleurodires; it is absent in caretta, but is large
Polarity: The foramen orbito-nasale is large in ser- in serpentina and Mongolemys. We consider a small
pentina, odoratus, and spinifera, but small in caretta. f.p.p. to be the derived state.
We consider presence of a large foramen orbito-nasale to
be the derived condition within testudinoids, because the (17) Position of the pterygoid relative to foramen palat-
foramen is small in Mongolemys. inum posterius (f.p.p.); 0 = pterygoid situated posterior
to the f.p.p., Fig. 33; 1 = pterygoid situated posterior to
(15) Contact between maxilla and vomer; 0 = present, the f.p.p., but sends a process anterior and lateral to the
Fig. 31; 1 = absent, vomer separated from the maxilla f.p.p., Fig. 34.
by the premaxilla, Fig. 32 (Hirayama, 1985, 31; Crumly, Our survey of testudinoids indicated that reliable
1982, 21; Yasukawa et al., 2001, 14). assessment of participation of the pterygoid in the f.p.p.
We generally agree with previous scorings for this may be difficult because many species show an ontoge-
character (Hirayama, 1985, Crumly, 1982, Yasukawa et netic change in configuration of this part of the palate. In
al., 2001). We confirm the absence of a contact in juveniles, the f.p.p. typically includes the pterygoid in its
amboinensis and pulcherrima, but our scoring differs posterior margin. During later ontogenetic stages the
slightly for those taxa that Hirayama (1985) coded as pterygoid is excluded. In spite of this, it appears that the
‘intermediate apomorphic,’ a character state that we relative position of the pterygoid tends to stay constant
interpret as polymorphism. Of those taxa that Hirayama during ontogeny.
(1985) and Yasukawa et al. (2001) scored as intermedi- Polarity: The anterior end of the pterygoid is situat-
ate (flavomarginata, caspica, annulata), our sample size ed posterior to the f.p.p. in odoratus, serpentina, most
is too small to confirm whether both character states are pleurodires, and Mongolemys, but is situated lateral to
present. We consequently follow these authors by scor- the f.p.p. in spinifera. We consider a posterior position to
ing those taxa as polymorphic. be primitive for Testudinoidea.
Polarity: A contact between the maxilla and vomer
is present in all outgroups. The loss of this contact is (18) Epipterygoid participation in the trigeminal fora-
derived for testudinoids. men; 0 = absent, Fig. 35; 1 = present, epipterygoid
clearly separates the parietal and pterygoid in lateral
(16) Size of the foramen palatinum posterius; 0 = large, view, Fig. 36.
Fig. 33; 1 = small, Fig. 34 (modified from Hirayama, The anteroventral rim of the trigeminal foramen of
1985, 22; Gaffney and Meylan, 1988, F2.2, F6.1; most testudinoid turtles is formed by the parietal and
McCord et al., 1995, 4; Yasukawa et al., 2001, 12). pterygoid. The epipterygoid commonly comes close to
Our characters 16 and 17 were published originally the foramen, but does not form part of it. In spengleri
and mouhotii, the epipterygoid consistently participates (21) Contact between pterygoid and basioccipital; 0 =
in the margin of the trigeminal foramen thus separating present, Fig. 41; 1 = absent, Fig. 42 (modified from
the parietal and pterygoid, at least in lateral view. Gaffney and Meylan, 1988, F1.1, F10.3, H18.3; Crumly,
Polarity: The epipterygoid forms part of the 1994; Shaffer et al., 1997, 103).
anteroventral rim of the trigeminal foramen in the Two of the most often-cited characters that purport-
majority of outgroups, with the exception of pleurodires edly help to distinguish the ‘Emydidae’ from the
that lack a definitive ossified epipterygoid (Gaffney, ‘Bataguridae’ are the batagurine process and the contact
1979). The condition is unclear for Mongolemys. between the pterygoid and the basioccipital (McDowell,
However, within the ingroup we found this character 1964). Both traits are commonly combined as one char-
only in the seemingly rather specialized turtles spengleri acter (e.g., Gaffney and Meylan, 1988) and even seem to
and mouhotii. We consequently consider its presence to have been confused with one another (e.g., Mlynarski,
be secondarily derived. 1976; Shaffer et al., 1997). The batagurine process is a
poorly-defined feature that, in McDowell’s original
(19) Vomerine foramen; 0 = absent, Fig. 37; 1 = pres- usage (1964) appears to consist of a lateral process of
ent, Fig. 38 (Gaffney and Meylan, 1988, H4.1; Crumly, the basioccipital that floors the recessus scalae tympani.
1994, 15). Many testudinoid species (including non-‘batagurines’)
The vomerine foramen (= anteromedial vomerine have a laterally-projecting process of the basioccipital; it
aperture of Crumly, 1982 and 1994 [in part]) is a small may or may not floor the recessus scalae tympani, but it
opening that pierces the vomer along the midline just is often obscured from view in articulated specimens.
posterior to the foramen praepalatinum (Bramble, 1971). When disarticulated material is examined, a broader dis-
The presence of a vomerine foramen was noted in agas- tribution of this feature (assuming it is interpreted as we
sizii and berlandieri by Bramble (1971), and was used have done above) across testudinoids is revealed.
by Gaffney and Meylan (1988) to unite various species Within Testudinoidea, the pterygoid commonly
currently placed in Gopherus as a clade. Its irregular sends a process posteriorly and contacts the basioccipi-
presence in elegans, elongata, chilensis, and radiata tal just lateral to the basisphenoid. This character
was reported by Crumly (1982). Specimens in our sam- appears to be absent in most ‘emydids,’ but is present in
ple enable us to confirm its presence in agassizii, terrapin and those species that are currently attributed to
berlandieri, and chilensis. Graptemys (Gaffney and Meylan, 1988). We noticed a
Polarity: The vomerine foramen occurs in only a strong ontogenetic component to this character within
few ‘testudinids’ and it is absent in all outgroups. Its ‘emydids.’ The pterygoid is commonly rather short dur-
presence is considered derived for Testudinoidea. ing younger ontogenetic stages, but finally reaches the
basioccipital in later stages. For instance, among our
(20) Development of the foramen praepalatinum as a specimens of terrapin, orbicularis, and texana, the
canal (canalis praepalatinum) that is concealed by a pterygoid does not contact the basioccipital in younger
bony secondary palate in ventral view; 0 = absent, Fig. individuals, but a clear contact is present in adults. We
39; 1 = present, Fig. 40 (modified from Hirayama, 1985, score such species as polymorphic, but note that ontoge-
24). netically influenced polymorphisms are not well
In most testudinoids, the foramen praepalatinum is explored in turtles.
a small opening that connects the nasal cavity with the Among ‘batagurids,’ we noted a similar pattern.
roof of the oral cavity (Fig. 39). However, in a number Contrary to general belief we were not able to observe a
of taxa with extensively developed secondary palates, contact between the pterygoid and the basioccipital in all
the anterior nasal artery passes through an elongated taxa traditionally classified in this group (e.g., trijuga,
canal that is concealed in ventral view by the bony sec- pulcherrima, sinensis). As with the ‘emydids,’ there
ondary palate (e.g., baska, tentoria, petersi; Fig. 40). We seems to be an ontogenetic effect in which younger
refer to this structure as the canalis praepalatinum. Our specimens tend not to have a contact. Where recognized,
scorings are fully consistent with those of Hirayama we scored these species as polymorphic.
(1985). We found much conflicting data regarding the dis-
Polarity: The foramen praepalatinum is absent in tribution of this character among tortoises (Crumly,
spinifera (Gaffney, 1979) and caretta (Nick, 1912), and 1982, 1985, 1994; Gaffney and Meylan, 1988). Among
is developed as a true foramen that is exposed in ventral the specimens we examined, we note the complete
view in odoratus, serpentina, and Mongolemys. The absence of a contact only in graeca; polymorphisms
development of a canalis praepalatinum is considered were observed in polyphemus and horsfieldi. Again, an
the derived condition within Testudinoidea. ontogenetic component is apparent.
Polarity: A contact between the pterygoid and
basioccipital is present in all outgroups. Its absence is Mandible

considered to be derived. (24) Angular contribution to the sulcus cartilaginis
Meckelii; 0 = present, the angular contributes to the sul-
(22) Contact of the pterygoid with the articular facet of cus and is as long or longer than the prearticular, Fig.
the quadrate; 0 = absent, Fig. 43; 1 = present, Fig. 44 47; 1 = absent, the angular is shorter than the preartic-
(Hirayama, 1985, 38). ular, Fig. 48 (modified from Gaffney and Meylan, 1988,
According to the data matrix published by F1.4).
Hirayama (1985) he only observed this contact in A broad contact of the angular with Meckel’s carti-
reevesii, however, in his tree (fig. 2) the contact is lage was used by McDowell (1964) to characterize the
mapped as an autapomorphy of subtrijuga. We confirm ‘Emydinae’ and later used by Gaffney and Meylan
the presence of a contact between the posterior process (1988) as a synapomorphy to unite the same grouping.
of the pterygoid and the articular surface of the quadrate As it was originally worded, this character is difficult to
in subtrijuga. It is the only taxon we examined that dis- observe in its literal sense for most museum specimens,
plays the derived condition. because the Meckel’s cartilage usually is not present in
Polarity: A contact between the pterygoid and the modern and fossil skeletal specimens. We suggest that
articular surface of the quadrate is absent in all out- the spirit of McDowell’s (1964) character can be evalu-
groups and the vast majority of the ingroup. We consid- ated by examining the participation of the angular in the
er its presence to be derived. sulcus cartilaginis Meckelii. We confirm that a small to
broad angular contribution is present in all ‘emydids’
(23) Closure and depth of the incisura columella auris; with the exception of rubriventris. In most ‘batagurids,’
0 = absent, incisura is open, Fig. 44; 1 = present, the angular is a short bone that does not participate in the
incisura closed, Fig. 45 (Crumly, 1985; Gaffney and sulcus and barely spans half the distance the prearticular
Meylan, 1988, H1.3). does. However, a small but clear contribution to the sul-
The incisura columella auris is a notch that is cus is present in an eclectic group comprised of baska,
formed by the quadrate and that holds the stapes and dentata, thurjii, punctularia, and some grandis. We were
eustachian tube (Gaffney, 1972). In a number of turtles, not able to carefully evaluate potential polymorphisms
the incisura closes to fully surround the stapedial shaft in these taxa.
(Gaffney and Meylan, 1988). Within testudinoids, the Polarity: An angular contribution to the sulcus car-
incisura evidently is closed in most tortoises (Crumly, tilaginis Meckelii is present in all cryptodiran outgroup
1985). We are able to confirm the presence of such a taxa. We consider is absence to be derived.
closed incisura in all ‘testudinids’ we examined with the
exception of one specimen of kleinmanni (CAS (25) Contact between surangular and dentary; 0 =sim-
228431), the smallest of the species now classified in ple contact, Fig. 49; 1 = strongly interdigitated suture,
Testudo. In a number of ‘batagurids’ and ‘emydids’ the Fig. 50 (Crumly, 1982, 12; Crumly, 1985; Gaffney and
incisura commonly is very narrow and even appears to Meylan, 1988, H6.1).
be closed, however, a closer look under the microscope In most testudinoids, the surangular and dentary
combined with a probing needle reveals that this appar- meet along the lateral side of the mandible in a simple,
ent closure is produced by dry tissues that remain in this overlapping contact. According to Crumly (1982, 1985),
area in many articulated skulls. The only ‘batagurid’ for this contact is stabilized through a finger-like process of
which we sometimes found a closed incisura is N. the surangular that interdigitates with the dentary in all
platynota; in that species the quadrate does not fuse tortoises except emys, impressa, and those species he
together forming a solid ring behind the incisura, but this classified in Gopherus. We confirm the absence of this
is also the case for many tortoises (e.g., some belliana, character in agassizii, berlandieri, emys, impressa, and
emys, some homeana, some kleinmanni). In some cases, polyphemus, but also did not observe it in areolatus.
the polymorphism we noted (e.g., in belliana and home- Polarity: An interdigitated contact between the
ana) appears to be a result of ontogenetic age, with older surangular and dentary is absent in all outgroups and the
individuals displaying a greater degree of fusion at the majority of the ingroup. We consider its presence to be
posterior part of the incisura. derived.
Polarity: The polarity of this character is somewhat
ambiguous, because the incisura columella auris is (26) Height of the processus coronoideus; 0 = as high as
closed in serpentina and spinifera, but open in caretta dentary, Fig. 51; 1 = rising significantly above the den-
and odoratus. We conclude that its presence is derived tary, Fig. 52 (modified from Hirayama, 1985, 45).
within testudinoids because it is absent in Mongolemys. The coronoid of most turtles is a small bone that
produces a minor knobby projection that rises only little
above the adjacent dentary, if at all. According to a clade within the ‘Emydidae.’ In our observations, how-
McDowell (1964) and Hirayama (1985) the coronoid is ever, this participation also is absent among other ‘emy-
larger and rises moderately above the dentary in dids’ such as flavimaculata, kohnii, nigrinoda, picta, and
borneensis and crassicollis, and in reevesii and subtriju- reticularia (we were not able to evaluate adequately the
ga the coronoid is very large and produces a robust potential for polymorphism in these taxa). In addition, it
process that sits high above the dentary. We confirm appears that the absence of a palatine participation rep-
these observations, however, we were also able to resents the plesiomorphic state for testudinoids, thus
observe moderately developed coronoid processes in eliminating its value for diagnosing monophyletic
kachuga and tentoria. Among ‘emydids,’ we also groups.
observed moderately developed coronoids in barbouri, Polarity: The palatine does not participate in the trit-
ernsti, flavimaculata, geographica, gibbonsi, kohnii, urating surface of spinifera, but does so in odoratus,
and terrapin. Unlike Hirayama (1985), we only utilize caretta, and serpentina. The palatine does not participate
one derived character state, because it is difficult to in the triturating surface of Mongolemys and only occurs
objectively measure and discretize the relative height of in Testudinoids with highly derived secondary palates.
the coronoid among turtles. We consequently consider the participation of the pala-
Polarity: The coronoids of caretta, serpentina, and tine in the triturating surface to be derived within tes-
odoratus are small and do not rise above the dentary, but tudinoids.
the coronoid of spinifera is well developed and forms a
moderate process. The lower jaw is not described for (29) Participation of the vomer in the triturating surface
Mongolemys. We consider well-developed coronoids to of the upper jaw; 0 = absent, Figs. 56, 57, 58; 1 = pres-
be derived within testudinoids. ent, Fig. 59 (Hirayama, 1985, 25).
The triturating surface is the grinding surface of the
(27) Foramen dentofaciale majus; 0 = small, Fig. 53; 1 jaw. In most turtles, it is formed on the upper jaw pre-
= large and situated within a large lateral fossa, Fig. 54 dominantly by the maxilla and premaxilla. However, in
(Hirayama, 1985, 47). turtles with extensive secondary palates the vomer may
The foramen dentofaciale majus of most testudi- also participate. In texana, the vomer may have a ventral
noids is a small opening that is situated on the lateral projection that barely separates the maxillae in the mid-
side of the mandible, just ventral and slightly anterior to line, but it does not participate in the triturating surface
the coronoid. The foramen dentofaciale majus is greatly proper because it sits in a dorsal concavity within the
enlarged in thurjii and ocellata and is situated at the palate (Fig. 56). Our scorings differ from those of
anterior end of an expanded lateral fossa (Hirayama. Hirayama (1985) for mouhotii (which Hirayama scored
1985). We confirm the enlargement in those taxa, and as “intermediate apomorphic” and we score as absent
further note its presence in petersi. because it does not have a secondary palate) and subtri-
Polarity: The foramen dentofaciale majus is small in juga (in our specimens the vomer does not descend to
all outgroups and the vast majority of the ingroup. Its the palatal surface, but this species may be polymor-
presence is considered to be derived. phic). This character is polymorphic in barbouri.
Polarity: Because the vomer does not participate in
Triturating Surfaces the triturating surface of odoratus, spinifera, caretta,
(28) Participation of palatine in the triturating surface serpentina, and Mongolemys, its participation is consid-
of the upper jaw; 0 = absent, Fig. 55; 1 = present, Figs. ered to be the derived condition for Testudinoidea.
56, 57 (Hirayama, 1985, 26; Gaffney and Meylan, 1988,
F2.1). (30) Presence and number of lingual ridges of the tritu-
In some testudinoids, the palatine has a ventrolater- rating surfaces of the upper and lower jaws; 0 = no lin-
al maxillary process that participates in the triturating gual ridges present, Fig. 60; 1 = one lingual ridge pres-
surface of the upper jaw. The degree of participation ent, Figs. 61-62; 2 = two lingual ridges present (modi-
varies among taxa, and within some species. A clear and fied from Hirayama, 1985, 29, 44; Gaffney and Meylan,
extensive participation is present in barbouri, ernsti, 1988, F7.2, F9.1)
geographica, gibbonsi, petersi, rubriventris, scripta, Most turtles lack lingual ridges on their triturating
terrapin, texana, and versa. It is weakly developed in surfaces (Fig. 60), but one or two such ridges are devel-
hamiltonii, ocellata, G. oculifera, ouachitensis, reevesii, oped in a number of testudinoids (Figs. 61-62). These
subtrijuga, and some individuals of baska. ridges run parallel to the labial surface of the maxilla
Participation was used by Gaffney and Meylan and dentary, and typically do not meet their counterpart
(1988) to unite Terrapene spp., blandingii, guttata, G. on the midline. They are not necessarily a continuous
insculpta, marmorata, muhlenbergii, and orbicularis as structure (Fig. 61), and may be divided into several com-
ponents. In some cases, an extensive ridge-like structure morphologies; this character needs to be critically
can create the appearance of an additional ridge at the reevaluated with adequate sample sizes for the relevant
extreme lingual margin of the maxilla bordering the taxa.
internal nares; however, we consider these to be the Polarity: All of our outgroup taxa and the majority
thickened rim of the internal nares rather than an addi- of the ingroup taxa lack strong serrations. We interpret
tional ridge. Among ‘batagurids’ and ‘emydids,’ one lin- their presence to be derived.
gual ridge is present in alabamensis, borneoensis, deco-
rata, kachuga, P. nelsoni, ocellata, petersi, rubriventris, (32) Median ridge or sulcus of the triturating surface of
scripta, sinensis, tentoria, texana, and thurjii. Two lin- the upper jaw; 0 = both structures absent, Fig. 60; 1 =
gual ridges are developed in baska. We found lingual median ridge present, Fig. 62; 2 = median sulcus pres-
ridges in all tortoises we examined except erosa, bel- ent, Fig. 63 (modified from Hirayama, 1985, 30;
liana, and homeana. Crumly, 1985, 1994, 4; Gaffney and Meylan, 1988,
Hirayama (1985) originally scored this character as H3.1).
two separate characters, one for the mandible and one In a number of testudinoid turtles with partially
for the maxilla. In our observations, the triturating sur- developed secondary palates and lingual ridges, addi-
face of the lower jaw closely mimics that of the upper tional structures are formed along the midline of the
jaw, creating an occlusal surface that closely reproduces upper jaw that typically correspond to reciprocal struc-
the function of cusps in mammalian cheek teeth. Both tures of the lower jaw. The upper jaw of petersi is char-
characters were scored in parallel in Hirayama’s matrix, acterized by a narrow sulcus (Fig. 63) and the mandible
and we see no reason to consider them independent. exhibits a low median ridge. On the other hand, in
Polarity: All outgroups lack lingual ridges on the baska, borneoensis, thurjii, kachuga, agassizii,
triturating surfaces. We consequently consider their berlandieri, and polyphemus, a ridge runs along the mid-
presence to be derived. line (Fig. 62), which typically corresponds to a sulcus in
the lower jaw. An incipient ridge also was reported in
(31) Well-developed serrations on labial or lingual emys (Crumly, 1994), but we were not able to confirm
ridges of the triturating surfaces of the upper and lower this on the specimen available to us.
jaws; 0 = absent, Fig. 60; 1 = present, Fig. 61 (modified Polarity: A median ridge is absent in all outgroups
from Hirayama, 1985, 21, 27, 41, 43, 46; Gaffney and and the vast majority of the ingroup. Its presence is con-
Meylan, 1988, F9.2; Yasukawa et al., 2001, 11). sidered to be derived.
Well-developed serrations on the lingual and labial
ridges of the upper and lower jaws are developed in sev- (33) Posterior extension of the lower triturating surface
eral ‘batagurids’ and ‘emydids’. A number of tortoises behind the symphysis of the dentary; 0 = absent, Fig.
and other ‘batagurids’ (e.g., carbonaria, pardalis, sulca- 64; 1 = present, Fig. 65 (Hirayama, 1985, 42; Gaffney
ta, annulata, and areolata) exhibit serrations on their and Meylan, 1988, G5.2).
ramphothecae, but unlike the bony, tooth-like serrations In several ‘batagurids,’ the triturating surface of the
of borneoensis, thurjii, petersi, or texana, these crenula- dentary forms a shelf along the midline that extends so
tions are weakly developed, leaving very little or no far posteriorly that the symphysis cannot be seen when
trace of serrations on the underlying bone. In compari- the mandible is observed in dorsal view (McDowell,
son to those taxa with well-developed serrations, it very 1964). Our scorings fully agree with those of Hirayama
difficult to establish a consistent scoring system for taxa (1985) for the ‘Bataguridae,’ but we disagree with
with fine crenulations, because many specimens will not Gaffney and Meylan (1988) who asserted that this char-
exhibit any serrations, probably due to wear of the ram- acter also occurs in some ‘Emydidae.’ Admittedly, sev-
phothecae. Unlike Hirayama (1985) and Yasukawa et al. eral species currently placed in the genera Graptemys,
(2001) we scored all taxa with such weak serrations as Pseudemys, and Trachemys have greatly expanded tritu-
absent. rating surfaces of the dentary, but in all of the specimens
Unfortunately, even in those taxa with well-devel- available to us, the symphysis is always visible in dorsal
oped serrations, the serrations are not always evenly view.
developed on all ridges. We consequently combined all Polarity: An extended triturating surface of the den-
of Hirayama’s (1985) characters relating to serrations tary does not occur in any outgroup taxon. We consider
into one character. Because serrations commonly occur its presence to be derived.
on all available ridges, this treatment will also help to
avoid unconsciously weighting the presence of serra- Carapace
tions with up to five characters. In those taxa that have (34) Carapace strongly tricarinate in adult; 0 = absent,
them, the ridges themselves often have very different Figs. 66, 67; 1 = present, Fig. 68 (modified from
Hirayama, 1985, F; McCord et al., 1995, 10; Yasukawa (36) Carapace of adult tectiform in cross-section with a
et al., 2001, 24). strong posterior projection on the third vertebral scute;
Three distinct carapacial ridges are present in the 0 = absent, Fig. 69; 1 = present, Fig. 70 (Hirayama,
adults of reevesii, hamiltonii, spengleri, subtrijuga, tri- 1985, N).
juga, and mouhotii. We cannot replicate Hirayama’s According to Hirayama (1985), this character only
(1985) placement of this character as an autapomorphy occurs in tecta and tentoria. For our sample, we were
in hamiltonii. In our observations, the carinae in hamil- able to confirm its presence in tecta and tentoria and
tonii are not better developed than in some other taxa. In also observed it in barbouri (barbouri and other species
fact, they are more weakly developed than those in now classified in Graptemys may be sexually dimorphic
mouhotii (Fig. 68). Because keels are present in the for this character). The descriptive term ‘tectiform’ is
young and subadults of such taxa as crassicollis (Fig. somewhat problematic, because any turtle shell can be
67), mutica, and sinensis, but disappear with age, and considered ‘roofed’ (Fig. 69). We regard a carapace as
because we were not able to observe the juveniles of tectiform if its sides are more-or-less flat and meet along
most species, we restricted this character to those the midline at a rather sharp angle (Fig. 70). Many tes-
species that exhibit well-developed tricarinae as adults. tudinoids, and notably those ‘emydids’ currently classi-
Three keels were reported to be present in the adults of fied within Graptemys, have a somewhat tectiform cara-
dentata (McCord et al., 1995), but we cannot confirm pace as juveniles, but that morphology typically is lost in
this (tricarinae are not present on our younger speci- the adults.
mens). Polarity: Because all outgroups and the majority of
Polarity: Tricarinae are absent in our outgroup the ingroup do not have a tectiform carapace, we consid-
species (caretta, odoratus, serpentina, spinifera, gibba, er its presence to be derived.
siebenrocki, subglobosa, and subrufa), but do appear
occasionally in some of their close relatives, such as (37) Shape and orientation of the second neural; 0 =
scorpioides, temminckii, and fimbriatus. We consider the second neural hexagonal, short sides positioned anteri-
presence of tricarinae to be derived within orly, Fig. 71; 1 = second neural hexagonal, short sides
Testudinoidea. positioned posteriorly, Fig. 72; 2 = second neural
octagonal, Fig. 73 (modified from Hirayama, 1985, G;
(35) Significant serration of the posterior peripherals; 0 Yasukawa et al., 2001, 25).
= absent, Fig. 66; 1 = present, Fig. 68 (modified from
Hirayama, 1985, D; McCord et al., 1995, 11; Yasukawa (38) Shape and orientation of the third neural; 0 = third
et al., 2001, 23). neural hexagonal, short sides positioned anteriorly, Fig.
We generally agree with previous observations 71; 1 = third neural hexagonal, short sides positioned
reported for this character (Hirayama, 1985). However, posteriorly, Fig. 72; 2 = third neural square, Fig. 74; 3
because the carapace rim is at least slightly serrated in = third neural octagonal, Fig. 75 (modified from
almost all turtles, we rephrase the character definition to Hirayama, 1985, G; Yasukawa et al., 2001, 25).
include only significantly serrated posterior peripherals Originally, Hirayama (1985) only discussed the ori-
as found, for example, in crassicollis, dentata, grandis, entation of the neurals in general, which is fully suffi-
mouhotii, N. platynota, and spengleri. Among the ‘emy- cient for his ‘batagurid’ ingroup, because almost all indi-
dids’ and ‘testudinids,’ the peripherals of barbouri, viduals exhibit his two suggested character states.
erosa, flavimaculata, geographica, homeana, kohnii, However, in most tortoises the second and/or third neu-
nigrinoda, oculifer, G. oculifera, pseudogeographica, rals are not hexagonal, but rather are square or octago-
and versa also are serrated. It is important to note that nal, making it impossible to assign them to one of
our scores are based on the peripheral bones; the amount Hirayama’s (1985) character states. We consequently
of carapacial serration greatly depends on the presence split Hirayama’s original character into two characters,
or absence of the marginal scutes in the specimens used, restricted their application to the second and third neu-
because the scutes greatly accentuate the amount of ser- ral, and added additional character states.
ration, if present. We find no conflict with the codings of Our observations generally agree with those of
McCord et al. (1995) and Yasukawa et al. (2001). Hirayama (1985) and Yasukawa et al. (2001) for
Polarity: Serrated posterior peripherals are present ‘batagurids,’ with the exception of annandalei in which
in caretta, but absent in odoratus, serpentina, and most we found the short side of the second and third neurals
pleurodires. However, due to ingroup commonality and to be positioned anteriorly, and not posteriorly as was
its absence in taxa placed within “Lindholmemydidae” indicated by Hirayama (1985).
we conclude that its presence is derived. Polarity: The short side of the second and third neu-
ral bones faces posteriorly in odoratus and spinifera but
anteriorly in caretta. The shape of the second and third Meylan (1988) that this scute is absent in carbonaria,
neurals is extremely variable in serpentina. However, in chilensis, elegans, nigra, pardalis, and sulcata and addi-
Lindholmemys and Mongolemys the short side of the tionally code homeana and erosa as polymorphic.
second and third neurals is positioned anteriorly. We Polarity: The cervical scute is present in all cryp-
consider that condition to be primitive for todiran outgroups that have scutes on their carapace. We
Testudinoidea. consider its absence to be derived.
(39) Medial contact of the seventh and/or eighth costal (41) Number of vertebral scutes; 0 = five, Fig. 80; 1 =
bones; 0 = absent, Fig. 76; 1 = present, Fig. 77 six or more, Fig. 81 (Hirayama, 1985, P).
(Hirayama, 1985, V; Yasukawa et al., 2001, 26). We confirm Hirayama’s (1985) observation that
In some testudinoid turtles, the posterior costal there are at least six vertebral scutes in N. platynota.
bones meet along the midline due to the reduction of the Additional scutes occasionally occur in other species,
posterior neural elements. The original character defini- but are best considered abnormalities; they typically
tion provided by Hirayama (1985) and Yasukawa et al. lack the symmetrical associations with adjacent pleural
(2001) was worded to indicate a contact between the scutes seen in N. platynota.
seventh and eighth costal bones among some Polarity: All testudinoids, except for N. platynota,
‘batagurids.’ We are unable to reproduce their results if have five vertebral scutes. We consider the presence of
the character definition is taken literally. For example, in six scutes to be derived.
all our specimens of amboinensis and galbinifrons, the
eighth costals meet on the midline, but the seventh (42) Position of the anterior sulcus of the fourth verte-
costals do not. However, if the character definition is bral scute; 0 = sulcus lies on the fifth neural, Fig. 82; 1
modified to include any contact of the seventh or eighth = sulcus lies on fourth neural, or on the suture between
costals, our results are concordant with those of the fourth and fifth neural, Fig. 83; 2 = sulcus lies on the
Hirayama (1985) and Yasukawa et al. (2001). In addition sixth neural, or on the suture between the fifth and sixth
to the Asian box turtles, we report a medial contact of neural, Fig. 84 (modified from Hirayama, 1985, L+M).
the posterior costals in baska, carolina, coahuila, T. nel-
soni, T. ornata and rubida. (43) Position of the posterior sulcus of the fourth verte-
Polarity: A medial contact of the seventh and/or bral scute; 0 = sulcus lies on the eighth neural, or on the
eighth costals is absent in serpentina, but present in homologue of the eighth neural, if the seventh is reduced
spinifera, caretta, many kinosternids, and many pleu- (e.g., in most tortoises), Fig. 85; 1 = sulcus lies on the
rodires. Although a composite reconstruction of a puta- seventh neural, or on the suture between the seventh and
tive “lindholmemydid” from the Early Cretaceous of eight neural, Fig. 86; 2= eighth neural absent, sulcus
Japan was illustrated with the seventh costals in contact overlies costals that meet at the midline, Fig. 87 (modi-
at the midline (Hirayama et al., 2000, fig. 11), such a fied from Hirayama, 1985, L+M).
contact is absent in other specimens of Lindholmemys The size of the fourth vertebral scute was addressed
and Mongolemys. Its presence within testudinoids is pre- with two characters by Hirayama (1985) but the total
dominantly in the highly derived box turtles, and we range of morphological variability in testudinoids is not
conclude that its presence is derived for Testudinoidea. encompassed by his character definitions. In most tes-
tudinoids, the fourth vertebral scute covers the posterior
(40) Cervical scute; 0 = present, Fig. 78; 1 = absent, half of the fifth neural bone, the sixth and seventh neu-
Fig. 79 (modified from Crumly, 1985, 1994, 34; Gaffney rals, and the anterior half of the eighth neural. A number
and Meylan, 1988, H5.2, H10.1; Shaffer et al., 1997, of variations are known, and simply counting the num-
41). ber of neural bones covered by this scute results in prob-
The presence and shape of the cervical scute is used lems by creating a false perception of homology. For
commonly to determine phylogenetic relationships instance, a fourth scute that partially overlies the fourth
within tortoises. According to Crumly (1985) the cervi- and seventh neurals and fully covers the fifth and sixth,
cal scute is very narrow or absent in all tortoises except can strictly be said also to cover four neurals, but the ele-
agassizii, berlandieri, emys, flavomarginatus, impressa, ments involved are only partially homologous with the
and polyphemus. We generally agree with these observa- common condition. We attempt to resolve these issues
tions, but when this character is applied to all testudi- by establishing two new characters that preserve what
noids intraspecific variation is so great that the character we think was Hirayama’s (1985) original intent, but that
becomes essentially useless. We consequently limit our permit a more accurate representation of the association
scoring to the mere presence or absence of the cervical between the fourth vertebral scute and the underlying
scute. We confirm the observations of Gaffney and neural bones.
Some problems that are associated with scoring this tebral scute commonly is restricted to the nuchal bone in
character include the prevalence of scute abnormalities several other taxa, but the scute shows no anterior nar-
among testudinoids (e.g., Coker, 1905, 1910; Newman, rowing, we limit the original character definition to the
1906; Zangerl and Johnson, 1957). Specimens exhibit- shape of the first vertebral scute only. This peculiar mor-
ing such abnormalities were scored as unknown. The phology seems to be the result of growth that is limited
notable exception to this is N. platynota, in which a to the anterior edge and the posterior half of the lateral
sixth, or even a seventh, vertebral scute is always pres- edge of the first vertebral scute. As a consequence, this
ent. character is not apparent in juveniles, but becomes
Polarity: Determining the polarity through outgroup increasingly accentuated in adults.
relationship is somewhat difficult, because almost every Polarity: Anterior narrowing of the first vertebral
outgroup exhibits a different condition, especially for scute is absent in all outgroups and within the large
the posterior sulcus. Based on ingroup commonality, and majority of the ingroup. We consider its presence to be
the presence of our zero state in both Lindholmemys and derived.
Mongolemys, we consider the sulci of the fourth verte-
bral scute to be primitively situated on the fifth and (46) Significant contact of the tenth marginal scute with
eighth neural bones. the fifth vertebral scute; 0 = absent, Fig. 92; 1 = pres-
ent, Fig. 93 (modified from Hirayama, 1985, K).
(44) Posterior margin of first vertebral scute significant- Contact of the tenth marginal scute with the fifth
ly narrower than its anterior margin; 0 = absent, Fig. vertebral scute was reported previously only in baska,
88; 1 = present, Fig. 89 (modified from Hirayama, 1985, smithii, tecta, and tentoria (Hirayama, 1985). We are
C). able to confirm the presence of a very well developed
When originally proposed, this character was contact in all but smithii (not seen), and we add spinosa
applied to a posterior constriction of all vertebral scutes to the list of species in which this contact may occur (it
(Hirayama, 1985). If strictly applied, this character is is polymorphic for spinosa; contact is present in CAS
absent in all taxa, because the fifth vertebral scute never 228368, but absent in the smaller CAS 228459, so onto-
is constricted along its posterior edge relative to the genetic differences may explain the polymorphism). We
anterior edge. If each scute is viewed by itself, it also note slight contacts in some specimens of other
becomes apparent that especially the fourth vertebral species (e.g., agassizii, borneoensis, carbonaria, home-
scute tends to be narrowed posteriorly, as can be ana, orbicularis, pardalis, and polyphemus), but by
observed in all species now classified in the genera rewording Hirayama’s (1985) original character to
Graptemys, Heosemys, Trachemys, and Testudo among include only significant contact, we are able to retain
others. According to Hirayama (1985), posterior narrow- what we believe was his original intent.
ing is limited to crassicollis and borneensis and unites Polarity: Due to the absence of contact in all out-
them as a synapomorphy. We were able to replicate this groups in which it is applicable and the predominance
distribution only if the character definition was restrict- within the ingroup, we consider its presence to be
ed to the first vertebral scute, in which the posterior mar- derived.
gin is significantly narrower than its anterior margin in
those two species only. In making this change, however, (47) Contact of the second marginal scute with the first
this character becomes at least partly redundant with vertebral scute; 0 = absent, Fig. 94; 1 = present, Fig. 95
characters 45 and 47. (Hirayama, 1985, O; see also Tinkle, 1962, table 1,
Polarity: Outgroup analysis reveals that posterior ‘Seam A’).
narrowing of the first vertebral scute is present only in According to Hirayama (1985), the first vertebral
odoratus; it is absent in Lindholmemys and scute usually (>90%) contacts the second marginal scute
Mongolemys. We regard the presence of a posterior nar- in japonica, leprosa, and caspica. He also noted that the
rowing of the first vertebral scute to be derived. scutes are sometimes in contact in N. platynota and
bealei. For our sample, we are able to confirm this con-
(45) Anterior half of the first vertebral scute much nar- tact as a polymorphism for bealei, caspica, japonica,
rower than posterior half, especially in adults; 0 = and leprosa, but the contact is clearly absent in all our
absent, Fig. 90; 1 = present, Fig. 91 (modified from specimens of N. platynota. We also note that these scutes
Hirayama, 1985, R). are sometimes in contact in picta, amboinensis, orbicu-
We confirm the clear presence of an anteriorly nar- laris, and terrapin. Together with all of the above, these
rowed first vertebral scute in dentata and spinosa as taxa were scored as polymorphic. The only taxa to
reported by Hirayama (1985), and note that grandis is exhibit a consistently well-developed contact are reticu-
polymorphic. Because the anterior sulcus of the first ver- laria and blandingii.
Polarity: A clear contact between the second mar- According to McDowell (1964, p. 240 and table 1
ginal scute and the first vertebral scute does not exist in number 4) members of the ‘Emydinae’ can be distin-
caretta or serpentina, but both morphologies occur in guished from the ‘Batagurinae’ based on an incomplete
kinosternids and pleurodires. The scutes are not in con- subdivision of the pygal bone by the median sulcus of
tact in Lindholmemys and Gravemys, but they are in con- the posterior-most marginals. We found exceptions with
tact in Mongolemys. The polarity for this character is picta, N. platynota, pulcherrima, reevesii, reticularia,
ambiguous. and spengleri, which do not always clearly exhibit the
pattern that would be predicted by McDowell’s (1964)
(48) Contact of the sixth marginal scute with the third statement, but we note that the expression of this charac-
pleural scute; 0 = absent, Fig. 96; 1 = present, Fig. 97 ter will depend significantly on the shape of the pygal
(modified from Hirayama, 1985, B; see also Tinkle, bone. In all tortoises except emys and impressa, the
1962, table 3, ‘Seam C’). twelfth marginal scutes are fused to from a single supra-
The contact between the sixth marginal and third caudal scute that covers the posterior part of the cara-
pleural scutes is easily enough rendered as a simple pace (Crumly, 1985). For this condition, we created a
‘presence or absence’ character, but this hides the range third character state.
of possible morphological variation. The degree of con- Polarity: The twelfth marginal scutes are fully sep-
tact can range from extensive to a condition where the arated in all outgroups that have them, and their com-
two scutes just barely contact at their corners. Several mon sulcus fully subdivides the pygal bone. The pygal
taxa exhibit a condition where these scutes either barely bone in “lindholmemydids” is polymorphic, with the 0
touch or do not touch one another at their corners, but state found in Lindholmemys and Mongolemys, and the
whenever several specimens were available to us, they 1 state in Gravemys. Either state 0 or state 1 is primitive
typically turned out to be polymorphic. For this reason for Testudinoidea; the midline fusion of the twelfth mar-
Gaffney and Meylan (1988) called this character ‘dubi- ginals is a derived feature for ‘Testudinidae’.
ous.’ We scored such borderline cases as polymorphic,
even if not enough specimens were available to corrob- Bridge
orate this. (50) Sutured contact between plastron and carapace; 0
We confirm Hirayama’s (1985) observations = present, plastron and carapace are tightly connected
regarding the absence of a contact between these scutes by an osseous bridge, Fig. 100; 1 = absent, plastron is
in baska, ocellata, tecta, tentoria, and thurjii, and we attached to carapace by connective tissue, Fig. 101
add petersi and spinosa to that list. Taxa that we score as (modified from Hirayama, 1985, Q; Shaffer et al., 1997,
polymorphic include annandalei, borneensis, borneoen- 58; Yasukawa et al. 2001, 21a).
sis, caspica, crassicollis, grandis, hamiltonii, japonica,
punctularia, reevesii, sinensis, subtrijuga, and trijuga. (51) Presence and development of anterior buttresses; 0
Whereas all ‘emydids’ lack a contact, tortoises exhibit = anterior buttresses absent, Fig. 102; 1 = anterior but-
both character states. tresses present but small, and not in contact with the first
The presence of a contact between the sixth margin- costal bones, Fig. 103; 2 = anterior buttresses long and
al scute and the third pleural scute was considered by thin and just barely in contact with the costal bones, if at
Hirayama (1985) to unite crown group ‘batagurids’ and all, Fig. 104; 3 = anterior buttresses well developed and
‘testudinids’ as a synapomorphy. Given the patchy dis- in clear contact with the first costal bones, Fig. 105; 4 =
tribution of this character, and widespread polymor- anterior buttresses very large and in direct contact with
phism, it seems to be of little use. the first dorsal rib, Fig. 106 (modified from Hirayama,
Polarity: A contact between the sixth marginal scute 1985, Q; Gaffney and Meylan, 1988, A14.2; Yasukawa
and the third pleural scute is absent in most outgroups. et al., 2001, 28).
The exception is caretta; this is not surprising because
caretta has five instead of four pleural scutes. We con- (52) Presence and development of posterior buttresses;
sider the presence of a contact to be derived. 0 = posterior buttresses absent, Fig. 107; 1 = posterior
buttresses present but small, and not in contact with the
(49) Twelfth marginal scute; 0 = two present, their com- costal bones, Fig. 108; 2 = posterior buttresses long and
mon sulcus only partially subdivides the pygal bone, thin and just barely in contact with the costal bones, if at
Fig. 98; 1 = two present, but their common sulcus fully all, Fig. 109; 3 =posterior buttresses well developed
subdivides the pygal bone, Fig. 98; 2 = both twelfth and in clear contact with costal bones V and VI, Fig.
marginal scutes fused along the midline, Fig. 99 (modi- 110; 4 = posterior buttresses well developed but only in
fied from Mlynarski, 1976; Crumly, 1985, 1994, 35; clear contact with costal bone V, Fig. 111 (modified from
Gaffney and Meylan, 1988, H2.1). Hirayama, 1985, Q; Gaffney and Meylan, 1988, A14.2,
Shaffer et al., 1997, 55; Yasukawa et al., 2001, 29).
(53) Medially-directed pivoting process for plastral (55) Contact between inguinal and femoral scutes; 0 =
hinge developed on fifth peripheral bone; 0 = absent, absent, Fig. 116; 1 = present, Fig. 117 (Crumly, 1985,
Fig. 112; 1 = present, Fig. 113. 1994, 42; Gaffney and Meylan, 1988, H3.3, H15.2).
Within tortoises the complete or frequent absence of
(54) Complete or almost complete overlap of hyoplas- a contact between the inguinal scute and the femoral
tron/hypoplastron suture by the pectoral/abdominal sul- scute was used previously to hypothesize the monophy-
cus; 0 = absent, Fig. 114; 1 = present, Fig. 115 (modi- ly of several smaller clades, for example {graeca + her-
fied Gaffney and Meylan, 1988, F3.2; Burke et al., 1996, manni + horsfieldi + kleinmanni + marginata + tornieri}
16; McCord et al., 1995, 13; Yasukawa et al., 2001, (Crumly, 1985) and {agassizii + berlandieri + flavomar-
21b). ginatus + polyphemus} (Gaffney and Meylan 1988). We
In most testudinoid turtles, the plastron is attached confirm the absence of a contact in representatives of the
to the carapace via a fully ossified bridge and variably first group, but not in the second. Among the second
developed plastral buttresses. In species with a kinetic group (traditionally classified together in Gopherus) the
plastron, the bridge is typically absent and the plastron is contact is strongly reduced, but still is present. Among
attached to the carapace via connective tissues (e.g., ‘batagurids’ and ‘emydids,’ a contact is absent in all taxa
amboinensis, blandingii, carolina, dentata, galbinifrons, with the noteworthy exception of hamiltonii.
mouhotii, orbicularis, T. ornata). The original configu- Polarity: Determining the polarity for this character
ration of this character tied the presence of plastral kine- is somewhat difficult because all living outgroups have
sis to the reduction of the buttresses (Hirayama, 1985). an arrangement of plastral scutes that is rather different
However, within testudinoid turtles the morphology of from testudinoids. However, based on ingroup common-
the buttresses varies significantly and independently ality and the absence of a contact in the
from plastral kinesis. We consequently split this charac- “Lindholmemydidae,” we conclude that the presence of
ter into three discrete characters concerned with the a contact between the inguinal and femoral scutes
morphology of the bridge and the buttresses. should be considered derived for Testudinoidea.
We also developed two new characters that pertain
to the morphology of the bridge region: the presence of (56) Presence of musk glands; 0 = inguinal and axillary
medially-directed processes on the fifth peripherals that gland present; 1 = axillary gland present only; 2 = musk
act as pivots for the plastral bones during shell closure glands absent (modified from Crumly, 1985).
(Bramble, 1974) and a revised plastral kinesis character
(Gaffney and Meylan, 1988; McCord et al., 1995) that (57) Presence of anterior musk duct foramina; 0 = musk
considers plastral-kinesis to be well developed only in glands and their foramina present, Fig. 118; 1 = musk
those taxa in which the pectoral/abdominal sulcus fully glands present, but foramina not developed; 2 = musk
overlaps the hyoplastron/hypoplastron suture, allowing glands and foramina absent (modified from Hirayama,
optimal movement between the two plastral lobes. Most 1985, A; Gaffney and Meylan, 1988, F1.2, F5.3; Burke
taxa with plastral kinesis also have well-developed piv- et al., 1996, 20).
oting processes on the fifth peripherals, but notable
exceptions are orbicularis and N. platynota. In (58) Presence of posterior musk duct foramina; 0 =
blandingii, the process is modified into an anteroposte- musk glands and their foramina present, Fig. 119; 1 =
riorly-elongated, flattened process that extends along musk glands present, but foramina not developed; 2 =
most or all of the length of the fifth peripheral (Fig. 113). musk glands and foramina absent (modified from
In some specimens of blandingii, a similar structure is Hirayama, 1985, A; Gaffney and Meylan, 1988, F1.2,
developed on the sixth peripheral as well. F5.3; Burke et al., 1996, 20).
Polarity: Reconstructing the basal condition for According to Hirayama (1985), the presence of
these characters within testudinoids is difficult, because musk duct foramina characterizes the paraphyletic
all living cryptodiran outgroups do not have plastral but- assemblage ‘Batagurinae’ (sensu Hirayama, 1985, not
tresses and commonly lack osseous bridges. However, Gaffney and Meylan, 1988). We believe the difference
the bridge of Gravemys, Mongolemys, and of opinion between Hirayama (1985) and Gaffney and
Lindholmemys is osseous, shows no signs of kinesis, and Meylan (1988) regarding this character is based on fail-
(at least in Lindholmemys) the anterior and posterior but- ure to make clear the distinction between the presence of
tresses are well developed and touch the costal bones. musk glands and the presence of musk duct foramina.
We consider that morphology to be primitive for Musk glands are developed in almost all extant turtles
Testudinoids. (Waagen, 1972), and we consequently agree with
Gaffney and Meylan (1988) that their presence should
be considered primitive for all cryptodiran turtles.
However, even though most turtles have musk glands, flat with a slight increase in thickness along the anterior
true musk duct foramina are developed only in some margin. In contrast, many tortoises have strongly thick-
pleurodires (e.g., Chelodina, Emydura), some ‘emy- ened epiplastral margins that sometimes form an over-
dids,’ and all ‘batagurids,’ making a monophyletic hanging lip along the interior rim of the plastron.
Testudinoidea (sensu Hirayama, 1985) possible. Distinct According to Gaffney and Meylan (1988) the presence
musk duct grooves are present on the anterior peripher- of such an overhang of the epiplastra unites all tortoises
als of Kinosternidae, and tiny foramina are sometimes except those classified in the genera Manouria (Fig.
associated with these (Hutchison, 1991). 120) and Gopherus, with a reversal occurring in giant
Because the presence of musk glands does not nec- insular forms. We confirm this general pattern, but we
essarily result in the development of musk duct forami- note the absence of an overhang in tornieri. A interior
na, we decided to score these two characters separately. overhang is absent also in all extant ‘emydids’ and
We relied on an unpublished thesis on the musk glands ‘batagurids,’ but a small overhang is present in extinct
of turtles (Waagen, 1972) to determine the presence of Ptychogaster and Echmatemys (Mlynarski, 1976, figs.
musk glands for most taxa. In scoring taxa not investi- 78, 81), taxa generally considered to belong to either the
gated by Waagen (1972) we only recorded presence of ‘Emydidae’ or ‘Batagurinae.’
musk glands if musk duct foramina provided positive Polarity: An overhanging lip on the epiplastra is
evidence for their presence (e.g., baska, bealei, borneen- absent in all outgroups and the majority of the ingroup.
sis, galbinifrons, kachuga, mouhotii, petersi, pulcherri- We consider its presence to be derived.
ma, spengleri, and tentoria). Many tortoises, conse-
quently, had to be scored as unknown, because they (60) Intersection of the entoplastron by the humeropec-
were not analyzed by Waagen (1972) and do not exhibit toral sulcus; 0 = absent, Fig. 122; 1 = present, Fig. 123
musk duct foramina (e.g., elongata, homeana, horsfiel- (Hirayama, 1985, X; Crumly, 1985; Gaffney and
di, and pardalis). Meylan, 1988, F5.1; McCord et al., 1995, 15).
Our scoring of the musk duct foramina is derived This character was used to help resolve relation-
from a combination of osteological observation and data ships within ‘batagurids’ by Hirayama (1985) but we
provided by Waagen (1972). Taxa not reported to pos- were unable to replicate his results in our analysis. We
sess musk glands (Waagen, 1972) were checked for agree with Crumly (1985) that the sulcus crosses the
musk duct foramina, but none were found. For those entoplastron in at least one species classified in the
species that Waagen reported as having musk glands, we genus Indotestudo (i.e., elongata), but the sulcus is at the
sought musk duct foramina on osteological specimens. entoplastron/hyoplastron suture in our specimen of
Several taxa with musk glands, but only lightly ossified forsteni. The condition in species now commonly classi-
bridges, do not exhibit musk duct foramina (e.g., fied in Testudo varies widely (e.g., the suture crosses the
blandingii, dentata, flavomarginata, orbicularis, and entoplastron in graeca and horsfieldi, but does not in
pulcherrima) or show an asymmetry with foramina only hermanni or kleinmanni). We also agree with Gaffney
visible anteriorly (e.g., N. platynota). In taxa that pos- and Meylan (1988) on their distribution of this character
sess them, the musk duct foramina are sometimes con- among the ‘Emydidae,’ however, our scoring for picta,
tained entirely within the peripherals (e.g., N. platyno- orbicularis, and blandingii is polymorphic, because the
ta), and sometimes between the peripheral and the plas- sulcus generally runs along the suture between the ento-
tral buttress (e.g., reevesii). plastron and the hypoplastra, but may barely fall on
Polarity: Given the presence of musk glands in all either side.
extant outgroups (Waagen, 1972), their absence should Polarity: The polarity is ambiguous if only extant
be considered derived. Musk duct foramina are not taxa are considered. The plastron of most outgroups is
described for “lindholmemydids” but J. H. Hutchison too different from that of testudinoids to be of any use
specifically searched for them in Mongolemys speci- for polarizing this character. For instance, the plastron of
mens housed at IVPP and found no trace of them. spinifera lacks scutes, and that of serpentina, odoratus,
Because musk duct foramina are developed in the vast and caretta is too heavily modified to enable a meaning-
majority of the ingroup, we consider their absence to be ful comparison. Both character states are commonly
derived for testudinoids. found in ‘batagurids,’ ‘emydids,’ and testudinids, mak-
ing an ingroup analysis futile. The humeropectoral sul-
Plastron cus is distinctly posterior to the epiplastron in Gravemys
(59) Extensive overhanging lip of the epiplastra; 0 = and Mongolemys, so we consider an intersection of this
absent, Fig. 120; 1 = present, Fig. 121 (Gaffney and suture with the entoplastron to be derived.
Meylan 1988, H5.1, 9.2).
In most testudinoid turtles, the epiplastra are rather
(61) Anal notch of the plastron; 0 = present, Fig. 124; 1 dis and spinosa.
= greatly reduced, Fig. 125; 2 = absent, Fig. 126 (mod- Polarity: Vibrant, radiated color patterns are miss-
ified from Hirayama, 1985, W; Yasukawa et al., 2001, ing in all outgroups and the majority of the ingroup.
22). Their presence is derived.
The plastron of most testudinoid turtles has a signif-
icant anal notch. The absence of such an anal notch for Postcranium and Soft Tissue
amboinensis, galbinifrons, and flavomarginata was (64) Development of a suprascapula; 0 = absent; 1 =
reported by Hirayama (1985) and we confirm those present, Fig. 130 (Gaffney and Meylan, 1988, F3.1;
observations. An anal notch also is absent in belliana, Burke et al., 1996, 11).
carolina, coahuila, erosa, homeana, T. nelsoni, and T.
ornata. To accommodate the presence of a reduced anal (65) Development of an episcapula; 0 = absent; 1 =
notch we modify Hirayama’s (1985) character by creat- present, Fig. 130 (Gaffney and Meylan, 1988, F4.1;
ing a third character state. A reduced anal notch is found Burke et al., 1996, 11).
in blandingii, orbicularis, N. platynota, and reticularia. The presence of both a suprascapula and an epis-
In at least one species (mouhotii), a distinct anal notch is capula apparently is limited to blandingii and the species
present in larger individuals, but small specimens have a currently classified in Terrapene. A suprascapula is also
reduced notch (scored as polymorphic in our matrix); present in orbicularis. Both structures are involved in
this suggests that development of an anal notch may be the locking mechanism of the anterior plastral lobe dur-
subject to ontogenetic variation in at least some testudi- ing shell closure (Bramble, 1974). These structures are
noids. difficult to verify in most osteological preparations,
Polarity: In caretta and serpentina an anal notch is because they may dissociate from the scapula and be dif-
not present, however, their plastra are narrow and ficult to recognize, and because they may ossify only in
tapered posteriorly. The fleshy plastron of spinifera is older individuals. The specimen we dissected to illus-
smooth along its posterior margin, but this cannot be trate these features (TNHC 62532, a T. ornata with cara-
observed in osteological preparations. A notch is weak- pace length of 103 mm) has a completely cartilaginous
ly developed in at least some Lindholmemys, but is episcapula, and a predominantly cartilaginous supras-
absent in Mongolemys. A notch is present in kinostern- capula (Fig. 130). It is therefore much easier to confirm
ids, pleurodires, most of the ingroup, and in Gravemys. their presence than verify their absence. We followed
We conclude that its absence should be considered Bramble’s (1974) account of these structures and scored
derived for our ingroup. our matrix accordingly, as probably did Gaffney and
Meylan (1988) and Burke et al. (1996).
(62) Anal scutes fused; 0 = absent, Fig. 127; 1 = pres- Polarity: Suprascapulae and episcapulae are absent
ent, Fig. 127 (Hirayama, 1985, Z). in all outgroups and the majority of the ingroup. Their
The anal scutes of adult galbinifrons, and flavomar- presence is considered to be derived.
ginata are at least slightly fused, especially along their
posterior medial border. We fully agree with Hirayama’s (66) Shape of coracoid blade; 0 = long and narrow, Fig.
(1985) treatment for this character. Anal scute fusion can 131; 1 = short and very wide, Fig. 131 (Crumly, 1985,
be identified easily in macerated specimens (Fig. 127), 1994; Gaffney and Meylan, 1988, H1.7).
because the anal scutes will not separate from one anoth- The coracoid blade of all ‘emydids’ and
er, as will all other scutes. ‘batagurids’ is an elongate bone with a narrow, short
Polarity: Anal scute fusion is absent in the vast shaft and a long, wedge-shaped coracoid blade that is
majority of turtles, and is considered to be the primitive about two times wider than the base. In tortoises, this
condition. bone is still wedge-shaped, but relatively much shorter
and with a blade that is considerably wider, typically
(63) Plastral scutes with vibrant, radiating color pat- four times the width of the base (Crumly, 1985, 1994;
tern; 0 = absent, Fig. 128; 1 = present, Fig. 129 Gaffney and Meylan, 1988). We agree with previously
(Hirayama, 1985, S; McCord et al., 1995, 16 Yasukawa published observations.
et al., 2001, 32). Polarity: The coracoid blade of caretta, serpentina,
Vibrant, radiating color patterns of the plastral and odoratus is long and narrow and that of spinifera is
scutes of dentata, grandis, and spinosa were noted by long, but not wedge-shaped. We consider a long and nar-
Hirayama (1985) and McCord et al. (1995). We add row coracoid blade to be primitive for Testudinoidea.
tcheponensis to this list, as well as the ‘testudinids’ geo-
metricus and P. oculifera. In our specimens, the pattern (67) Number of manual claws; 0 = five, Fig. 132; 1 =
of dentata and tcheponensis is not as vibrant as in gran- four, Fig. 133 (modified from Hirayama, 1985, J).
Most testudinoid turtles have five manual claws webbing). Among ‘emydids,’ we note that the webbing
with the exception of baska and horsfieldi, both of is heavily reduced in carolina, T. nelsoni, and T. ornata.
which have only four. We did not verify this character All tortoises lack webbing.
independently for all species, due to an overall lack of Polarity: All outgroups and the majority of ingroup
articulated skeletons and our limited access to pickled taxa have webbed hands and feet. We consider the
specimens. However, because the number of claws of absence of webbing to be derived.
the forelimbs is an easily determinable, discrete number
that is regularly noted and described in the literature, we (70) Sexual size dimorphism; 0 = absent; 1 = present,
scored all remaining taxa from the comprehensive infor- female much larger than male (Gaffney and Meylan,
mation provided by Ernst and Barbour (1989). 1988, F5.2; Burke et al., 1996, 37).
Polarity: Five manual claws are present in serpenti- In almost every species of turtle, there is some
na, odoratus, and almost all pleurodires (excluding expression of sexual size dimorphism (Berry and Shine,
species currently classified in Chelodina and 1980; Gibbons and Lovich, 1990). The difference in size
Hydromedusa); three are present in spinifera, and two in between the sexes can be expressed as a ratio and typi-
caretta. The condition in “lindholmemydids” is cally shows considerable variation depending on the
unknown. Based on ingroup commonality, we consider population (Gibbons and Lovich, 1990). We initially
five claws to be primitive for the ingroup. tried to score this character with three character states, as
done by Burke et al. (1996), differentiating between
(68) Number of phalanges of manus and pes; 0 = digi- species with larger males, larger females, and equally
tal formula of 2-3-3-3-3 or 2-3-3-3-2, Fig. 132; 1 = dig- sized sexes, but we abandoned that, because exact data
ital formula with less than 2-3-3-3-2, Fig. 133 (Crumly, are missing for most ‘batagurid’ taxa. We consequently
1985; Gaffney and Meylan, 1988, H1.1). only score taxa as being sexually dimorphic if females
The digital formula of most testudinoid turtles is 2- are at least 1.4 times larger than the males. Our scores
3-3-3-3 or 2-3-3-3-2. Among tortoises, the manus and are derived from Gibbons and Lovich (1990) and Ernst
pes are greatly shortened and the digital formula is typi- and Barbour (1989).
cally reduced to 2-2-2-2-2 or less (Auffenberg, Polarity: Sexual size dimorphism is prevalent in
1974:135-136; Crumly, 1985). Due to the dissociated most outgroups. In spinifera the female is much larger,
nature of most of the material we viewed, we were not in odoratus and serpentina the male tends to be slightly
able to verify the digital formulae of most of the turtle smaller, in caretta the sexes are of similar size. The out-
taxa we included. However, when articulated hands and group polarity is thus ambiguous, but in the majority of
feet were present, we never found anything to contradict the ingroup pronounced sexual dimorphism is absent.
the statements made above. We scored all tortoises
based on information provided by Auffenberg (1974) Problematic Characters
and Crumly (1985). We encountered difficulties in evaluating a number of
Polarity: All outgroups and the majority of the previously used characters, and we provide some sum-
ingroup do not have a reduced digital formula. We con- mary statements for those in this section. Most of these
sequently consider the reduced formula to be derived. characters were not pursued thoroughly in our study
because we were not able to understand the original
(69) Webbing between digits; 0 = present, well devel- descriptions, were unable recover discrete character
oped, Fig. 134; 1 = absent, or at least strongly reduced, states, or because at an early point in our investigation of
Fig. 135 (Hirayama, 1985, b). the character we detected significant variation in expres-
Due to their semi-aquatic nature, most testudinoids sion of character states within taxa. In the latter case
have well-developed webbing between the digits of their greater sample sizes or new methodological techniques
hands and feet. In more terrestrial species, however, the (e.g., Wiens, 1995; Smith and Gutberlet, 2001) will be
webbing often is reduced. Unfortunately, there seems to required to tease out a phylogenetic signal.
be a gradient in the development of webbing, from
extremely well developed (e.g., baska, reticularia) to (A) Frontal contribution to the supratemporal rim
moderately developed (e.g., dentata, guttata) to virtual- (Hirayama, 1985, 4).
ly non-existent (e.g., spengleri). We nevertheless were The anterior extent of the upper temporal emargina-
able to reproduce Hirayama’s (1985) distribution for the tion is difficult to define in many taxa, and is impossible
‘batagurids’ with the exception of grandis and spinosa, to determine in those with a fully emarginated temporal
which have reduced webbing (grandis is the only ‘bor- region (e.g., T. ornata). The result is a high degree of
derline’ taxon we found, but its webbing is reduced rel- ambiguity and a general lack of discrete character states.
ative to those taxa we scored as having well-developed
(B) Contact between postorbital and quadrate cant amount of variation can be observed in two speci-
(Hirayama, 1985, 10). mens of borneensis available to us, that exhibit left/right
In the vast majority of ingroup taxa, there is no con- asymmetry and the full spectrum from a fully intact (Fig.
tact between the postorbital and the quadrate. Such a 137), to partially subdivided (Fig. 138), to fully subdi-
contact was observed only in japonica and punctularia vided (Fig. 139) trigeminal foramen. Given that most
by Hirayama (1985). In specimens of japonica available taxa are represented by three or fewer skulls in our
to us, we were not able to confirm this contact. CAS study, it is apparent that we are not able to fully docu-
228348 is a skeleton from a diseased specimen of punc- ment the amount of variation exhibited by testudinoids.
tularia. On the right side of the skull there is a possible
(but only slight) contact. It is possible that the contact is (F) Contact between postorbital and squamosal
actually between the postorbital and the quadratojugal (Hirayama, 1985, 9).
(Fig. 136). We also found a minimal contact in one spec- Gaffney et al. (1991) noted that absence of this con-
imen of annulata. Given these diverging observations tact is associated with the upper temporal emargination
and the minute contact that is present in our material, we and considered it informative at the level of their analy-
regard (for now) any contact within the ingroup as an sis. Within our ingroup, all turtles have substantial upper
abnormality. temporal emarginations, resulting in the contact being
just barely present, or just barely absent, or polymorphic
(C) Absence of the “posterior process of the postor- (e.g., picta, petersi, texana, crassicollis). See comments
bital” (Hirayama, 1985, 8). above under character 9.
We cannot determine unambiguously what
Hirayama (1985) meant by this character. In our assess- (G) Median premaxillary notch (Hirayama, 1985, 19;
ment of testudinoids, both a posterolateral and a postero- Yasukawa et al., 2001, 9)
medial process of the postorbital can occur. In
Hirayama’s (1985) analysis, only grandis and spinosa (H) Large cusps near the suture of the premaxillae and
lack a “posterior process of the postorbital.” These maxillae (Hirayama, 1985, 28).
species are also the only ‘batagurid’ taxa to fully lack a Initially, we were faced with the problem of defin-
temporal arch. We suspect that this character may some- ing these two characters independently from one anoth-
how be referring to a lack of a bony temporal arch. er, because the presence of two tightly spaced, opposing
cusp-like structures along any margin will automatically
(D) Processus inferior parietalis “medially approximat- result in the development of a median notch. An addi-
ing each other, cranial cavity anteroventrally narrow- tional problem relating to these characters is the ques-
ing” (Hirayama, 1985, 5; McCord et al., 1995, 7; tion of whether these features should be observed on the
Yasukawa et al., 2001, 2). ramphotheca or the maxilla.
We acknowledge the validity of this character as Large, tooth-like cusps are clearly present in a num-
was originally worded by McDowell (1964). However, ber of taxa (e.g., thurjii) but so is the full spectrum of
we find it difficult to determine how strongly the con- smaller cusps, making it impossible to clearly define
striction of the brain case must be before it can be con- discrete character states. Furthermore, if all species were
sidered present. We were unable to develop unambigu- evaluated for medial notches that existed even if the
ous discrete character states for this feature. cusps were removed, all taxa in our sample would show
a medial notch. We were unable to develop a consistent
(E) Subdivision of the foramen nervi trigemini (Crumly, method for scoring this character for all testudinoid
1982; Hirayama, 1985, 6). species.
This character was used originally by Crumly
(1982) to infer phylogenetic relationships within (I) “Antero-medial portion of the upper triturating sur-
‘Testudinidae’. For his ingroup, Crumly (1982) face formed by premaxillae and maxillae” (Hirayama,
observed a great amount of polymorphism, with no sin- 1985, 23; Yasukawa et al., 2001, 13).
gle species either completely lacking or always exhibit- We are neither able to replicate the full meaning of
ing a subdivision of the foramen. He also noted asym- this character nor formulate truly discrete character
metry for this character between the left and right side of states. A connection with the development of the second-
some individuals. We confirm the common presence of ary palate is evident, but the morphology of this region
a subdivided trigeminal foramen in representatives of seems to be sufficiently covered by a number of other
‘Testudinidae,’ and the occasional presence in individu- characters.
als of ‘Emydidae’ and ‘Bataguridae’ (e.g., areolata, den-
tata, flavimaculata, N. platynota, and rubida). A signifi-
(J) Participation of the vomer in the foramen praepalat- (N) Bony sutures and sulci lost in old adults (Hirayama,
inum (Crumly, 1982, 10; Hirayama, 1985, 32; Yasukawa 1985, I).
et al., 2001, 15). According to Hirayama (1985), loss of sutures and
Within testudinoid turtles, the foramen praepalat- sulci occurs in baska, borneoensis, and borneensis only.
inum perforates the nasal cavity at the border between We are able to confirm this, but we do not have individ-
the premaxilla and the vomer. When the foramen is posi- uals of all other species that are sufficiently old enough
tioned slightly more anteriorly, it is fully surrounded by to positively confirm that they also do not exhibit this
the premaxilla, when it is minutely farther posterior it is feature at old age. In subsequent treatments, loss of
surrounded by the vomer. Considering the impact of sutures and loss of sulci should be treated as separate
such minute changes, it is not surprising that our scoring characters.
for this character generally seems to be in conflict with
that of Hirayama (1985) and Crumly (1982). This char- (O) Ossification of cornu branchiale II (Hirayama,
acter appears to be subject to great intraspecific varia- 1985, 48; Yasukawa et al., 2001, 20).
tion. This character was used previously to unite tortois-
es with a number of ‘batagurid’ taxa (Hirayama, 1985).
(K) Foramen palatinum posterius enclosed within the The hyoid apparatus of turtles is often disarticulated in
brain cavity (Hirayama, 1985, 34). skeletal preparations, making is difficult to positively
According to Hirayama (1985), in reevesii (only) confirm if an ossified cornu branchiale is present or
the foramen palatinum posterius is enclosed within the absent. However, for those taxa for which we were able
region of the brain cavity due to a flared descending to observe the hyoid apparatus, we were not able to con-
process of the parietal. We cannot confirm this observa- firm Hirayama’s (1985) observation of a reduced cornu
tion for any testudinoid turtles (including three speci- branchiale II in some ‘batagurids’ (e.g., galbinifrons,
mens of reevesii). spengleri). Instead, these taxa exhibit a cornu branchiale
II that is not significantly different from most other
(L) Participating bones in the processus trochlearis ‘batagurids.’
oticum (Hirayama 1985, 37; Gaffney and Meylan, 1988,
Gaffney et al. 1991, 6; McCord et al., 1995, 8; Shaffer et (P) Double articulation between the fifth and sixth cer-
al., 1997, 74, 258 Yasukawa et al., 2001, 18). vical vertebrae (Hirayama, 1985; Gaffney and Meylan,
The relative participation of the prootic, parietal, 1988, F1.5).
and quadrate to the processus trochlearis oticum was Most articular surfaces of the cervical column are
used previously by a number of authors to infer phylo- rather homogenous within all testudinoid turtles
genetic relationships within turtles. Our observations (Williams, 1950). A double articulation between the fifth
confirm the great variety of morphologies that can be and sixth cervical previously was considered to be a
observed in this region. However, the amount of unique character that unites the ‘Emydinae’ (McDowell,
intraspecific variation is considerable and the full spec- 1964). This character also was used by Hirayama (1985)
trum of possible morphologies seems to be filled, mak- and with reservations by Gaffney and Meylan (1988).
ing it difficult to discern discrete character states. Future Our observations generally confirm the presence of a
research in the area may result in more clearly defined more or less clear double articulation in most ‘emydids,’
discrete character states. however, this features is also present in a number of
‘batagurids’ confirming that this character is highly vari-
(M) Length of the crista supraoccipitalis (Hirayama, able within the ingroup (Williams, 1950; Gaffney and
1985, 40). Meylan, 1988). Unfortunately, discrete character states
A long crista supraoccipitalis was observed by are lacking; we were able to observe the full morpholog-
Hirayama (1985) for borneensis. The character states he ical spectrum from a clear singular articulation to a clear
used are defined by relative length of the crista supraoc- double articulation.
cipitalis to the “condylo-basal length.” Unfortunately,
we could not replicate this because it is not clear exact- (Q) Scapular prong with lateral concavity (Hirayama,
ly how the length of the crista was measured. 1985, E).
Furthermore, a true morphological gap seems to be Hirayama (1985) reported this character as an
missing between the admittedly very long crista of autapomorphy for subtrijuga only. However, we cannot
borneensis and other ‘batagurids’ with an elongated identify this morphology in any of our specimens of sub-
crista. This character is problematic, because it is poor- trijuga.
ly defined and lacks discrete character states.
(R) Large facet of the ilium for the testoscapularis and 1972) and Mauremys (Killebrew, 1977; Stock, 1972).
testoiliacus (Hirayama, 1985, T; Yasukawa et al., 2001, This conflict in primary data is probably best understood
34) when considering the nature of testudinoid chromo-
somes: whereas 14 pairs of chromosomes have a consid-
(S) Extensive development of both testoscapularis and erable size, all of the remaining ones are extremely
testoiliacus (Hirayama, 1985, U). small. Given these circumstances, it seems reasonable to
An extensive development of the testoscapularis hypothesize that one pair of chromosomes may be
and testoiliacus muscles together with an associated unrecognized during analysis.
large scar on the ilium was reported by Bramble (1974)
for Asian and North American box turtles. Whereas we (V) Plica media spade-shaped (Gaffney and Meylan,
have no reason to doubt his assessment of the develop- 1988, F7.1).
ment of these muscles for box turtles, we were not able The penile soft anatomy of turtles was comprehen-
to score this character for most of the remaining taxa. sively reviewed by Zug (1966) and one of his characters,
The shape of the ilium was explored and illustrated for the shape of the plica media, was used by Gaffney and
some ‘emydids’ and ‘batagurids’ by Yasukawa et al. Meylan (1988) to unify species placed in Chrysemys,
(2001:122-123). Deirochelys, Trachemys, and Pseudemys as a mono-
phyletic group. In his detailed description of the plica
(T) Ossification of the epipubis (Gaffney and Meylan, media, Zug (1966) referred to the shape of this structure
1988, F5.5). as being “spade-shaped” in those taxa, but made similar
The identification (and confirmation of presence or claims for other taxa too. Furthermore, based on the
absence) of an ossified epipubis (Fig. 141) is somewhat illustrations that were provided by Zug (1966) for other
difficult for most species, because it seems to ossify taxa, the plica media of species placed in Graptemys,
rather late in ontogeny, and can fall off during prepara- Malaclemys, Rhinoclemmys, and Platysternon appear
tion. Our tentative observations confirm the presence of “spade-shaped” also, even though Zug (1966) did not
an ossified epipubis in numerous adult ‘emydids’ and explicitly use those descriptive words. This anatomical
‘batagurids,’ typically terrestrial forms (e.g., T. ornata, system should be carefully reevaluated for all testudi-
G. insculpta, N. platynota, yuwonoi). An improved sam- noids, with special attention given to definition of dis-
ple of adult specimens of all taxa, however, is necessary crete characters.
to reveal the true distribution of this character.
(W) Ossifications within the fenestra postotica.
(U) Diploid Number of Chromosomes (Hirayama, 1985, In some taxa, portions of the fenestra postotica are
Ø; Shaffer et al., 1997, 43). closed or obscured by ossifications (noted, but without
The diploid number of chromosomes was used by exemplars, by Gaffney, 1972). In our largest specimen
Carr and Bickham (1986) to hypothesize a sister group of grandis (CAS 228443) a short, spike-like ossified
relationship between the ‘Emydinae’ and subtrijuga, fol- process extends posterodorsally from the dorsal edge of
lowed by borneensis and crassicollis and finally the rest the quadrate process of the pterygoid, and crosses the
of the ‘Batagurinae.’ Whereas most ‘batagurids’ alleged- fenestra postotica. It is situated ventral to the stapes (col-
ly have 52 chromosomes, subtrijuga, borneensis, crassi- umella auris), medial to the incisura columella auris, and
collis, and ‘emydids’ are supposed to have 50 chromo- lateral to the fenestra ovalis (Fig. 141). In some speci-
somes. We view these results with caution, because a mens, the dorsal tip of a similar structure approaches or
brief review of the relevant literature reveals great dif- meets a posterodorsally-inclined process that extends
ferences in chromosomal counts for a variety of taxa. from the dorsal surface of the pterygoid, near the suture
For instance, according to the work of Killebrew (1977) with the prootic. In our largest specimen of N. platynota
and Bickham (1981), amboinensis has 52 chromosomes, (CAS 228342) the two processes meet to enclose the
however, Gorman (1973) reported only 50. Similar con- stapedial shaft in a ring of bone situated at approximate-
flicts can also be found for dentata (Bickham, 1981; ly the midpoint between the fenestra ovalis and the
DeSmet, 1978; Gorman, 1973; Stock, 1972), subtrijuga medial opening of the incisura columella auris. Our
(Bickham, 1981; Killebrew, 1977), trijuga (DeSmet, other, smaller, specimen of N. platynota shows no devel-
1978; Carr and Bickham, 1986), sinensis (Bickham, opment of these processes. It seems likely that there is
1981; Killebrew, 1977; Stock, 1972), crassicollis an ontogenetic component in the expression of this fea-
(Killebrew, 1977; Stock, 1972; Bickham and Baker, ture. It does not appear to have any systematic signifi-
1976), and some of the species currently placed in cance, but in any case it is not widespread within
Graptemys (Killebrew, 1977; McKown 1972), Testudinoidea.
Trachemys (DeSmet, 1978; Killebrew, 1977; Stock,
Conclusions produce a phylogenetic hypothesis in this paper was

based primarily on two considerations. The first is the
Our observations can be used to draw some tenta- relatively small sample size we used for many taxa in
tive conclusions regarding the current level of under- this study (although it is comparable to sample sizes
standing about morphological variation within testudi- from other, previously published, studies), and the fact
noid turtles. In addition, we provide some cautionary that several taxa are not represented in our work. The
statements about the quality of morphological data now second consideration is our sense that the current under-
in use for assessing systematic relationships among standing of morphological variation in testudinoid tur-
these turtles. It is clear from a perusal of the relevant lit- tles is insufficiently mature to permit reliable phyloge-
erature, and from our data, that there is reasonably netic hypotheses based solely on morphological data.
strong morphological support for a monophyletic The most expedient way to address the need for greater
‘Testudinidae.’ Support for monophyly of ‘emydines’ documentation of variation within testudinoid species is
and ‘batagurines’ is not as impressive. The paraphyly of to utilize existing museum collections to the greatest
‘Batagurinae’ (with respect to ‘Testudinidae’) was extent possible, and secondarily to develop responsible
explicitly proposed by Hirayama (1985) and has been collecting programs that are designed with this need in
generally accepted since that time, although some mind.
strides have been made towards resolving relationships Our results also indicate that morphological data
among some ‘batagurine’ taxa. The monophyly of matrices currently in the literature should not be taken at
‘Emydinae’ seems to have been implicitly assumed by face value. We had particular difficulties replicating
many workers, but remains to be established in the con- some of the scoring in the Hirayama (1985) matrix. That
text of a rigorous phylogenetic analysis of all relevant seminal analysis (completed prior to the widespread use
taxa. The interpretation of several morphological fea- of computer-assisted analytical methods in systematics)
tures shared between some ‘emydines’ and some laid the foundation for nearly all subsequent work on
‘batagurines’ as either convergence or synapomorphy ‘batagurine’ morphology (including our own), and its
remains an important and interesting challenge. For importance in shaping our current conceptualization of
example, there are intriguing morphological similarities ‘batagurine’ phylogeny cannot be overstated. The work
between subtrijuga and some species classified in the of pathfinders in all fields of inquiry is often subjected
genus Graptemys (e.g., contact of the jugal and descend- to the greatest scrutiny by the next generation of
ing process of the parietal; contact of the quadratojugal researchers. Our statements and contradictory observa-
with the articular facet of the quadrate; contact between tions in this paper in no way denigrate Hirayama’s work;
the quadratojugal and the maxilla; ventral process of the instead, we view our efforts as minor attempts to correct
pterygoid approaching the articular surface of the the few inconsistencies in his analysis, and to contribute
quadrate). These similarities may be due to functional our observations to the body of knowledge that he began
convergence as a result of a molluscivorous diet, but to synthesize 20 years ago.
they raise questions about the propriety of utilizing sub- The accurate interpretation and understanding of
trijuga as an outgroup for systematic studies of ‘emy- the morphological descriptions of previous authors were
dines’ (e.g., Burke et al., 1996). Additional similarities among the great challenges we faced when we began our
are reported for chromosome numbers in subtrijuga and studies of testudinoid skeletal morphology. Much of our
‘emydines’ (see ‘Problematic Character’ U, above). confusion could have been averted if adequate illustra-
A seriously deficient understanding of morphologi- tions accompanied all published character descriptions,
cal variation is one of the greatest inadequacies of cur- but such documentation often is an expensive undertak-
rent perspectives on morphological data in turtles gener- ing. Our photographs of character states discussed in
ally, and testudinoids specifically. Few published studies this paper are intended to facilitate communication
have been conducted to evaluate the range and causes of among turtle enthusiasts, and to provide a baseline for
sexual, ontogenetic, intra- and inter-population variation future comparisons and discussions about testudinoid
in morphological characters within testudinoids. Our morphology. We hope that adequate illustration of all
preliminary considerations of ontogenetic variation, newly proposed characters will become standard prac-
combined with reports of sexual variation (e.g., Berry tice among turtle systematists. It seems likely that our
and Shine 1980; Gibbons and Lovich, 1990; Stephens, interpretations of characters will differ in some respects
1998) and new studies exploring the complex interac- from those of our colleagues, and we anticipate that our
tions of morphological evolution with behavioral char- decisions regarding ‘problematic characters’ (discussed
acteristics and environmental conditions in turtles (e.g., above), and our choices with respect to ‘lumping’ or
Lindeman, 2000; Herrel et al., 2002; Joyce and Gauthier, ‘splitting’ previously published character states, will
2003; Claude et al., 2003) emphasize the importance of generate much spirited discussion in the years ahead.
pursuing these questions further. Our decision not to
As practicing vertebrate paleontologists, we also Graduate Fellowships provided funding to WGJ.

hope that our efforts here will stimulate additional inves- Additional financial support was provided by the
tigation and publication of the extensive fossil record of Geology Foundation of The University of Texas at
testudinoid turtles. Morphology is, of course, of para- Austin.
mount importance for the interpretation of fossils. Our
visits to many museums in the last several years Literature Cited
revealed an abundance of unpublished testudinoid fossil
material. Although a phylogenetic analysis was not a Agassiz, L. 1857. Contributions to the Natural History
goal of this research, we adopt the convenient and now of the United States of America. First Monograph.
familiar means of summarizing morphological data by Part II. North American Testudinata. Pp. 235-452,
providing a character matrix that summarizes some of Plates 1-7. Little, Brown and Company, Boston.
our observations. Although character data can be used to
assist paleontologists in diagnosing fossil testudinoid Auffenberg, W. 1974. Checklist of fossil land tortoises
specimens, we also encourage paleontologists to utilize (Testudinidae). Bulletin of the Florida State
our study as a starting point for basic morphologic Museum, Biological Sciences 18:121-251.
descriptions. Description of differential diagnostic char-
acters generally is an adequate minimum for the erection Berry, J. F., and R. Shine. 1980. Sexual size dimorphism
of a new taxon. However, such a diagnosis is, in itself, and sexual selection in turtles (Order Testudines).
not particularly helpful to systematists trying to score a Oecologia 44:185-191.
matrix and place a fossil into a broader phylogenetic
context. Descriptions of new fossil specimens (and taxa) Bickham, J. W. 1981. Two-hundred-million-year-old
will be most useful if they include discussions of char- chromosomes: Deceleration of the rate of karyotyp-
acter state data for all preserved anatomical regions. ic evolution in turtles. Science 212:1291-1293.
This current summary of morphological characters that
traditionally are used in systematic treatments of testudi- Bickham, J. W., and R. J. Baker. 1976. Chromosome
noids can be used as a preliminary guide to the anatom- homology and evolution of emydid turtles.
ical regions and features that would be most useful when Chromosoma 54:201-219.
included as part of a description of new fossil material.
Bickham, J. W., T. Lamb, P. Minx, and J. C. Patton.
Acknowledgements 1996. Molecular systematics of the genus Clemmys
and the intergeneric relationships of emydid turtles.
We are greatly indebted to a number of institutions Herpetologica 52:89-97.
and their staff for providing generous access to skeletal
and wet specimens: Jens Vindum (CAS), Stephen Boulenger, G. A. 1889. Catalogue of the Chelonians,
Rogers and John Wiens (CM), Alan Resetar and Rhynchocephalians, and Crocodiles in the British
Maureen Kearney (FMNH), John Simmons and Linda Museum (Natural History). New Edition. Taylor
Trueb (KU), Jim McGuire (LMNH), José Rosado and Francis, London. 311 pp., Plates 1-6.
(MCZ), Travis LaDuc and David Cannatella (TMM),
Harold Dundee (TUMNH), and Gregory Caldwell- Bourret, R. 1941. Les tortues de l’Indochine avec une
Watkins and Jacques Gauthier (YPM). We extend our note sur la pêche et l’élevage des tortues de mer par
thanks to Rick Van Dyke for the gift of specimens to F. Le Poulain. L’Institut Océanographique de
support this research. Gabe Bever, Julien Claude, Jason l’Indochine. Imprimerie G. Taupin & Cie, Hanoi.
Downs, Jacques Gauthier, Jason Head, Pat Holroyd, 235 pp.
Howard Hutchison, Lyndon K. Murray, Jim Parham, and
Krister Smith (in alphabetical order) provided useful Bramble, D. M. 1971. Functional morphology, evolu-
comments and criticisms regarding this project. All tion, and paleoecology of gopher tortoises. Ph.D.
images containing YPM specimens were reproduced Thesis. University of California, Berkeley. 341 pp.
with courtesy of the Peabody Museum of Natural
History, Yale University, New Haven, CT. Special Bramble, D. M. 1974. Emydid shell kinesis:
thanks go to Ted Papenfuss and Jim Parham for provid- Biomechanics and evolution. Copeia 1974:707-727.
ing us with the unique opportunity to publish the results
of this study, and to Julien Claude and Howard Burke, R. L., T. E. Leuteritz, and A. J. Wolf. 1996.
Hutchison for their detailed and insightful reviews. The Phylogenetic relationships of emydine turtles.
Brett Sterns Award for Chelonian Research (CAS), the Herpetologica 52:572-584.
Visiting Scholar Fund (FMNH), and Yale University
Carr, J. L., and J. W. Bickham. 1986. Phylogenetic Gaffney, E. S. 1972. An illustrated glossary of turtle
implications of karyotypic variation in the skull nomenclature. American Museum Novitates
Batagurinae (Testudines: Emydidae). Genetica 2486:1-33.
70:89-106.
Gaffney, E. S. 1975. A phylogeny and classification of
Claude, J., E. Paradis, H. Tong, and J.-C. Auffray. 2003. the higher categories of turtles. Bulletin of the
A geometric morphometric assessment of the American Museum of Natural History 155:387-436.
effects of environment and cladogenesis on the evo-
lution of the turtle shell. Biological Journal of the Gaffney, E. S. 1985 (“1984”). Progress towards a natu-
Linnean Society 79:485-501. ral hierarchy of turtles. Studia Geológica
Salmaticensia Volumen Especial 1: Studia
Coker, R. E. 1905. Gadow’s hypothesis of “orthogenet- Palaeocheloniologica 1:125-131.
ic variation” in Chelonia. The Johns Hopkins
University Circular 178:489-504. Gaffney, E. S. 1979. Comparative cranial morphology of
recent and fossil turtles. Bulletin of the American
Coker, R. E. 1910. Diversity in the scutes of Chelonia. Museum of Natural History 164:65-376.
Journal of Morphology 21:1-75, Plates 1-14.
Gaffney, E. S., and P. A. Meylan. 1988. A phylogeny of
Crumly, C. R. 1982. A cladistic analysis of Geochelone turtles. Pp. 157-219. In M. J. Benton (ed.), The
using cranial osteology. Journal of Herpetology Phylogeny and Classification of the Tetrapods,
16:215-234. Volume 1: Amphibians, Reptiles, Birds. The
Systematics Association Special Volume 35A,
Crumly, C. R. 1985 (“1984”). A hypothesis for the rela- Clarendon Press, Oxford.
tionships of land tortoise genera (Family
Testudinidae). Studia Geológica Salmaticensia Gaffney, E. S., P. A. Meylan, and A. R. Wyss. 1991. A
Volumen Especial 1: Studia Palaeocheloniologica computer assisted analysis of the relationships of
1:115-124. the higher categories of turtles. Cladistics 7:313-
335.
Crumly, C. R. 1994. Phylogenetic systematics of North
American tortoises (Genus Gopherus): Evidence Gibbons, J. W., and J. E. Lovich. 1990. Sexual dimor-
for their classification. Pp. 7-32. In R. B. Bury and phism in turtles with emphasis on the slider turtle
D. J. Germano (eds.), Biology of North American (Trachemys scripta). Herpetological Monogrpahs
Tortoises. United States Department of the Interior, 4:1-29.
National Biological Survey, Fish and Wildlife
Research 13. Gorman, G. C. 1973. The chromosomes of the Reptilia,
a cytotaxonomic interpretation. Pp. 349-424. In A.
Danilov, I. G., and V. B. Sukhanov. 2001. New data on B. Chiarelli and E. Capanna (eds.), Cytotaxonomy
lindholmemydid turtle Lindholmemys from the Late and Vertebrate Evolution. Academic Press, New
Cretaceous of Mongolia. Acta Palaeontologica York.
Polonica 46:125-131.
Gray, J. E. 1855. Catalogue of Shield Reptiles in the
DeSmet, W. H. O. 1978. The chromosomes of 11 species Collection of the British Museum, Part 1:
of Chelonia (Reptilia). Acta Zoologica et Testudinata (Tortoises). Taylor and Francis,
Pathologica Antverpiensia 70:15-34. London. 79 pp.
Ernst, C. H., and R. W. Barbour. 1989. Turtles of the Gray, J. E. 1870. Supplement to the Catalogue of Shield
World. Smithsonian Institution Press. Washington, Reptiles in the Collection of the British Museum,
D. C. 313 pp. Part 1: Testudinata (Tortoises). Taylor and Francis,
London. 120 pp.
Feldman, C. R., and J. F. Parham. 2002. Molecular phy-
logenetics of emydine turtles: Taxonomic revision Herrel, A., J. C. O’Reilly, and A. M. Richmond. 2002.
and the evolution of shell kinesis. Molecular Evolution of bite performance in turtles. Journal of
Phylogenetics and Evolution 22:388-398. Evolutionary Biology 15:1083-1094.
Hirayama, R. 1985 (“1984”). Cladistic analysis of Lindholm, W. A. 1929. Revidiertes Verzeichnis der
batagurine turtles (Batagurinae: Emydidae: Gattungen der rezenten Schildkröten nebst Notizen
Testudinoidea); a preliminary result. Studia zur Nomenklatur einiger Arten. Zoologischer
Geológica Salmaticensia Volumen Especial 1: Anzeiger 81:275-272.
Studia Palaeocheloniologica 1:141-157.
McCord, W. P., J. B. Iverson, and Boeadi. 1995. A new
Hirayama, R., D. B. Brinkman, and I. G. Danilov. 2000. batagurid turtle from northern Sulawesi, Indonesia.
Distribution and biogeography of non-marine Chelonian Conservation and Biology 1:311-316.
Cretaceous turtles. Russian Journal of Herpetology
7:181-198. McCord, W. P., J. B. Iverson, P. Q. Spinks, and H. B.
Shaffer. 2000. A new genus of geoemydid turtle
Honda, M., Y. Yasukawa, and H. Ota. 2002. Phylogeny from Asia. Hamadryad 25:86-90.
of the Eurasian freshwater turtles of the genus
Mauremys Gray 1869 (Testudines), with special ref- McDowell, S. B. 1964. Partition of the genus Clemmys
erence to a close affinity of Mauremys japonica and related problems in the taxonomy of the aquat-
with Chinemys reevesii. Journal of Zoological ic Testudinidae. Proceedings of the Zoological
Systematics and Evolutionary Research 40:195- Society of London 143:239-279.
200.
McKown, R. R. 1972. Phylogenetic relationships within
Hutchison, J. H. 1991. Early Kinosterninae (Reptilia: the turtle genera Graptemys and Malaclemys. Ph.D.
Testudines) and their phylogenetic significance. Thesis. The University of Texas at Austin. 112 pp.
Journal of Vertebrate Paleontology 11:145-167.
Mlynarski, M. 1976. Handbuch der Paläoherpetologie.
Iverson, J. B., P. Q. Spinks, H. B. Shaffer, W. P. McCord, Teil 7. Testudines. Gustav Fischer Verlag, Stuttgart,
and I. Das. 2002 (“2001”). Phylogenetic relation- New York. 130 pp.
ships among the Asian tortoises of the genus
Indotestudo (Reptilia: Testudines: Testudinidae). Newman, H. H. 1906. The significance of scute and
Hamadryad 26:272-275. plate “abnormalities” in Chelonia. Biological
Bulletin 10:68-114.
Joyce, W. G., and J. A. Gauthier. 2003. Palaeoecology of
Triassic stem turtles sheds new light on turtle ori- Nick, L. 1912. Das Kopfskelet von Dermochelys cori-
gins. Proceedings of the Royal Society of London, acea L. Zoologische Jahrbücher, Abteilung für
Series B, Biological Sciences 271:1-5. Anatomie und Ontogenie der Tiere 33:1-238.
Khosatzky, L. I., and M. Mlynarski. 1971. Chelonians Pritchard, P. C. H. 1979. Encyclopedia of Turtles. T.F.H.
from the upper Cretaceous of the Gobi Desert, Publications, Inc., Ltd. Neptune, New Jersey. 895
Mongolia. Pp. 131-144, Plates 22-24. In Z. Kielan- pp.
Jaworowska (ed.), Results of the Polish-Mongolian
Palaeontological Expeditions – Part III. Rhodin, A. G. J. 2000. Publisher’s editorial: Turtle sur-
Palaeontologia Polonica 25. vival crisis. Turtle and Tortoise Newsletter 1:2-3.
Killebrew, F. C. 1977. Mitotic chromosomes of turtles. Shaffer, H. B., P. Meylan, and M. L. McKnight. 1997.
IV. The Emydidae. The Texas Journal of Science Tests of turtle phylogeny: Molecular, morphologi-
29:245-253. cal, and paleontological approaches. Systematic
Biology 46:235-268.
Kluge, A. G. 1991. Boine snake phylogeny and research
cycles. University of Michigan Museum of Zoology Siebenrock, F. 1909. Synopsis der rezenten
Miscellaneous Publications 178:1-58. Schildkröten, mit Berücksichtigung der in his-
torischer Zeit ausgestorbenen Arten. Zoologische
Lindeman, P. V. 2000. Evolution of the relative width of Jahrbücher, Supplement 10:427-618.
the head and alveolar surfaces in map turtles
(Testudines: Emydidae: Graptemys). Biological Sites. J. W. Jr., J. W. Bickham, B. A. Pytel, I. F.
Journal of the Linnean Society 69:549-576. Greenbaum, and B. A. Bates. 1984. Biochemical
characters and the reconstruction of turtle phyloge- Wiens, J. J. 1995. Polymorphic characters in phyloge-
nies: Relationships among batagurine genera. netic systematics. Systematic Biology 44:482-500.
Systematic Zoology 33:137-158.
Williams, E. E. 1950. Variation and selection in the cer-
Smith, E. N., and R. L. Gutberlet Jr. 2001. Generalized vical central articulations of living turtles. Bulletin
frequency coding: A method of preparing polymor- of the American Museum of Natural History
phic multistate characters for phylogenetic analysis. 94:505-562.
Systematic Biology 50:156-169.
Wu, P., K.-Y. Zhou, and Q. Yang. 1999. Phylogeny of
Smith, M. 1931. The Fauna of British India, Including Asian freshwater and terrestrial turtles based on
Ceylon and Burma. Reptilia and Amphibia. Vol. I. – sequence of 12S rRNA gene fragments. Acta
Loricata, Testudines. Taylor & Francis Ltd., Zoologica Sinica 45:260-267. (In Chinese with
London. 185 pp. English summary).
Stephens, P. R. 1998. Variation in the cranial osteologi- Yasukawa, Y., R. Hirayama, and T. Hikida. 2001.
cal morphology of turtles in the genus Graptemys Phylogenetic relationships of geoemydine turtles
(Reptilia; Anapsida; Testudines; Cryptodira; (Reptilia: Bataguridae). Current Herpetology
Emydidae; Deirochelyinae). M.S. Thesis. 20:105-133.
University of South Alabama. 197 pp.
Zangerl, R. 1948. The methods of comparative anatomy
Stephens, P. R., and J. J. Wiens. 2003. Ecological diver- and its contribution to the study of evolution.
sification and phylogeny of emydid turtles. Evolution 2:351-374.
Biological Journal of the Linnean Society 79:577-
610. Zangerl, R. 1969. The turtle shell. Pp. 311-339. In C.
Gans, A. d’A. Bellairs, and T. S. Parsons (eds.),
Stock, A. D. 1972. Karyological relationships in turtles Biology of the Reptilia Volume 1, Morphology A.
(Reptilia: Chelonia). Canadian Journal of Genetics Academic Press, New York.
and Cytology 14:859-868.
Zangerl, R., and R. G. Johnson. 1957. The nature of
Sukhanov, V. B. 2000. Mesozoic turtles of middle and shield abnormalities in the turtle shell. Fieldiana:
central Asia. Pp. 309-367. In M. J. Benton, M. A. Geology 10:341-362.
Shishkin, D. M. Unwin, and E. N. Kurochkin (eds.),
The Age of Dinosaurs in Russia and Mongolia. Zdansky, O. 1924. Über die Temporalregion des
Cambridge University Press, New York. Schildkrötenschädels. Bulletin of the Geological
Institution of the University of Upsala 19:89-114.
Theobald, W. 1868. Catalogue of reptiles in the museum
of the Asiatic Society of Bengal. Journal of the Zug, G. R. 1966. The penial morphology and the rela-
Asiatic Society, Extra Number, 1868:8-88. tionships of cryptodiran turtles. Occasional Papers
of the Museum of Zoology, University of Michigan
Tinkle, D. W. 1962. Variation in shell morphology of 647:1-24.
North American turtles I. The carapacial seam
arrangements. Tulane Studies in Zoology 9:331-
349.
van Dijk, P. P., B. L. Stuart, and A. G. J. Rhodin, eds.

2000. Asian Turtle Trade. Proceedings of a
Workshop on Conservation and Trade of Freshwater
Turtles and Tortoises in Asia, Phnom Penh,
Cambodia, 1-4 December 1999. Chelonian
Research Monographs 2.
Waagen, G. N. 1972. Musk glands in recent turtles. M.S.

Thesis. University of Utah. 64 pp.
Appendix 1 15482, YPM 15486; kachuga, FMNH 224128, FMNH

224152; kleinmanni, CAS 228422, CAS 228426, CAS
List of specimens used. 228431; kohnii, TUMNH 10237, TUMNH 12121,
agassizii, CAS 33867, FMNH 216746, FMNH 250843; TUMNH 14544, TUMNH 15678; leprosa, CM 137031;
alabamensis, CM 95968, CM 95991; amboinensis, CAS marmorata, CAS no number, CAS 188533, CAS
153872, CAS 228345, CAS 228369, CAS 228412, CAS 220052, FMNH 22076; mouhotii, CAS 228365, CAS
228419, FMNH 224009; annandalei, FMNH 258876, 228374, CAS 228420, CAS 228444; muhlenbergii,
FMNH 258879, FMNH Moll3036; annulata, CAS LMNH 55352, MCZ 52248; mutica, LMNH 54883; P.
SUR7425, CM 87903, YPM 15410; arachnoides, MCZ nelsoni, CM 67311; nigra, CAS 8125, CAS 8289;
54050; areolatus, MCZ 42214; areolata, CM 47957, nigrinoda, TUMNH 15147, TUMNH 15317, TUMNH
CM 87904; barbouri, CAS SUR12063, TUMNH 15400, 15408, TUMNH 15750; ocellata, CAS 228336; G.
TUMNH 15428, TUMNH 15429, TUMNH 16899; oculifera, TUMNH 26, TUMNH 3359, TUMNH 7548,
baska, FMNH 224095, FMNH 224097, FMNH 224213, TUMNH 12402, TUMNH 16928; P. oculifera, CAS
FMNH 224124, FMNH 224226; bealei, CM 118554, 165598, CAS 220645, CAS 220646; odoratus, YPM
FMNH 255270, FMNH 226542; belliana, CAS 228394, 13622, CAS 228351, CAS 228352, CAS 228353; orbic-
CAS 228404; berlandieri, TNHC 2813; blandingii, ularis, CAS 173223, CAS 228347, CAS 228415, CAS
CAS 12837, CAS 228346, CAS 228373, CAS 228434, 228447, LMNH 10347, YPM 15479; T. ornata, CAS
CAS 228448, YPM 14677, YPM 10382; borneensis, 228381, CAS 228382, CAS 228383, CAS 228384,
FMNH 224001, FMNH 224003, FMNH 224004, TNHC 62532; ouachitensis, CM 61656, CM 84696;
FMNH 224005, FMHH 224140; borneoensis, FMNH pardalis, CAS 148630, CAS 228349, CAS 228410,
224107, FMNH 224122, FMNH 224129, FMNH CAS 228418, CAS 228432, CAS 228449; petersi, CAS
251499, MCZ 42198; carbonaria, CAS 228397, CAS 8608, CM 124246, CM 124247; picta, CAS 13889, CAS
228411, CAS 228427, CAS 228430; caretta, CAS 8383, 228379, CAS 228380, CAS 228385; N. platynota, CAS
FMNH 31021; carolina, CAS 228375, CAS 228378; 228342, CAS 228450, CM 118586, FMNH 224216,
caspica, CAS 141118, CM 118517, FMNH 19714, YPM 14678; polyphemus, CAS 14090, LMNH 43344,
FMNH 74505, FMNH 98764; chilensis, CM 112252; LMNH 43354, LMNH 59320; pulcherrima, CAS
coahuila, KU 46929, KU 92623, MCZ 120335; crassi- 11754, CAS 228355, CAS 228366, CAS 228377; pul-
collis, CAS 228335, CAS 228451, FMNH 11091, MCZ chra, LMNH 48092; punctularia, YPM 465, CAS
7821, MCZ 134451; decorata, CM 118590; dentata, 228348; reevesii, CAS 31437, CAS 228361, CAS
CAS 134332, KU 47170, MCZ 29567, CAS 228333, 228364; reticularia, CAS 228338, CAS 228388, LMNH
CAS 228362, CAS 228414; elegans, CAS 228396; elon- 14565, LMNH 14569, LMNH 49145, LMNH 54856,
gata, CAS 228408, FMNH 183740, FMNH 257382; LMNH 78580; rubida, CM 87907, CM 87908;
emys, CAS 228344, FMNH 63749, FMNH 224034; ern- rubriventris, CM 34409, CM 45188; scripta, CAS
sti, TUMNH 13460, TUMNH 13462, TUMNH 16899; SUR8642, CAS 228436, CAS 228442, LMNH 15684;
erosa, YPM 12653; flavimaculata, TUMNH 15375, serpentina, CAS 228452, YPM 6369, CAS 228457;
TUMNH 15404, TUMNH 15747, TUMNH 15787; siebenrocki, CAS 228393; signatus, CAS 228405, MCZ
flavomarginata, CAS 18040, CAS 228356, CAS 42217, MCZ 42218; sinensis, CAS 18031, CAS
228357, CAS 228359, FMNH 216515; flavomarginatus, 228339; spengleri, CAS 21008, CAS 228331, CAS
MCZ 164926; forsteni, CAS 228433; galbinifrons, CAS 228332, CAS 228343, CAS 228371, YPM 11653, YPM
228358, CAS 228360, CAS 228367, CAS 228376, YPM 14445; spinifera, CAS 65705, CAS 228350, CAS
14074, YPM 14080; geographica, CAS 12809, 228354, YPM 656; spinosa, CAS 228368, CAS 228459;
TUMNH 15391, TUMNH 15392, TUMNH 15410, subglobosa, CAS 228334; subtrijuga, CAS 16996, CAS
TUMNH 15430; geometricus, MCZ 32184; gibba, CAS 228445, CAS 228446, CAS 228453, CAS 228454; sub-
228392; graeca, CAS 217732, CAS 228435, LMNH rufa, CAS 228389, CAS 228390, CAS 228391; tchepo-
54855, LMNH 54861; grandis, CAS 228341, CAS nensis, CAS 228363, CAS 228370; tecta, CM 89923;
228443, FMNH 224038, YPM 15431; guttata, CAS tentoria, FMNH Moll3026, FMNH Moll3028, FMNH
8696, CAS 228372, CAS 228386, FMNH 211573; Moll3032, FMNH 259430; tentorius, MCZ 41944,
hamiltonii, MCZ 120333, YPM 10399; hermanni, CAS MCZ 46604; terrapin, CAS 43640, CAS 228340, CAS
228400, CAS 228401, CAS 228402; homeana, CAS 228387, YPM 2983; texana, CAS 30965, CAS 228437,
228409, CAS 228423, CAS 228428, CAS 228429, CAS CAS 228438, CAS 228439, CAS 228440, CAS 228441,
228455, FMNH 19794; horsfieldi, CAS 120707, CAS CAS 228456, CAS 228458; thurjii, FMNH 224135,
228398, CAS 228399, CAS 228421, CAS 228425, YPM FMNH 224153, MCZ 62523, MCZ 62524, YPM 14072,
16450; impressa, CAS 228416; G. insculpta, CAS YPM 14073, YPM 14074; tornieri, CAS 139704, CAS
228406, CAS 228407, CAS 228413; japonica, YPM 228395, CAS 228403, CAS 228417, CAS 228424; tri-
fasciata, CAS 228337, MCZ 5218; trijuga, CAS 12463, dussumieri Aldabrachelys, Dipsochelys,
CM 89921, CM 124227, YPM 15453; versa, TUMNH Geochelone, Megalochelys
4484, TUMNH 10510, TUMNH 16192; yuwonoi, YPM elegans Geochelone
12626. elongata Geochelone, Indotestudo
emolli Chrysemys, Pseudemys, Trachemys
emys Geochelone, Manouria
Appendix 2 ernsti Graptemys
Generic synonymies for all currently recognized testudi- erosa Kinixys
noid species and all outgroup species used herein. femoralis Homopus
Synonymies are based only on common usage during flavimaculata Graptemys, Malaclemys
the last 50 years. flavomarginata Cistoclemmys, Cuora, Geoemyda
flavomarginatus Gopherus
Ingroup floridana Chrysemys, Pseudemys
adiutrix Chrysemys, Pseudemys, Trachemys forsteni Geochelone, Indotestudo, Manouria
agassizii Gopherus, Xerobates funerea Callopsis, Geoemyda, Rhinoclemmys
alabamensis Chrysemys, Pseudemys gaigeae Chrysemys, Pseudemys, Trachemys
amboinensis Cuora galbinifrons Cistoclemmys, Cuora
angulata Chersina, Goniochersus, Neotestudo, geographica Graptemys, Malaclemys
Testudo geometricus Psammobates
annamensis Annamemys, Mauremys gibbonsi Graptemys
annandalii Hieremys graeca Testudo
annulata Callopsis, Geoemyda, Nicoria, grandis Geoemyda, Heosemys
Rhinoclemmys guttata Clemmys
arachnoides Pyxis hamiltonii Geoclemys
areolata Callopsis, Geoemyda, Nicoria, hermanni Testudo
Rhinoclemmys homeana Kinixys
areolatus Homopus horsfieldi Agrionemys, Testudo
aurocapitata Cuora impressa Geochelone, Manouria
barbouri Graptemys, Malaclemys G. insculpta Calemys, Clemmys, Glyptemys
baska Batagur japonica Clemmys, Mauremys
bealei Clemmys, Sacalia kachuga Kachuga
belliana Kinixys kleinmanni Pseudotestudo, Testudo
berlandieri Gopherus kohnii Graptemys
blandingii Emydoidea, Emys, Neoemys leprosa Clemmys, Mauremys
borneensis Orlitia leytensis Geoemyda, Heosemys
borneoensis Callagur lobatsiana Kinixys
boulengeri Homopus marginata Testudo
caglei Graptemys marmorata Clemmys, Emys
callirostris Chrysemys, Pseudemys, Trachemys melanosterna Geoemyda, Rhinoclemmys
carbonaria Chelonoidis, Geochelone mouhotii Cuora, Cyclemys, Geoemyda, Pyxidea
carolina Terrapene muhlenbergii Calemys, Clemmys, Glyptemys
caspica Clemmys, Mauremys mutica Clemmys, Mauremys
chilensis Chelonoidis, Geochelone nasuta Callopsis, Geoemyda, Rhinoclemmys
coahuila Terrapene natalensis Kinixys
concinna Chrysemys, Pseudemys nebulosa Chrysemys, Pseudemys, Trachemys
crassicollis Siebenrockiella P. nelsoni Chrysemys, Pseudemys
decorata Chrysemys, Pseudemys, Trachemys T. nelsoni Terrapene
decussata Chrysemys, Pseudemys, Trachemys nigra Chelonoidis, Geochelone
dentata Cyclemys, Geoemyda nigricans Chinemys
denticulata Chelonoidis, Geochelone nigrinoda Graptemys, Malaclemys
depressa Geoemyda, Heosemys ocellata Morenia
dhongoka Kachuga G. oculifera Graptemys, Malaclemys
diademata Geoemyda, Rhinoclemmys P. oculifera Psammobates
dorbigni Chrysemys, Pseudemys, Trachemys oldhamii Cyclemys
orbicularis Emys yaquia Chrysemys, Pseudemys, Trachemys

T. ornata Terrapene yniphora Asterochelys, Geochelone
P. ornata Chrysemys, Pseudemys, Trachemys yunnanensis Cuora
ouachitensis Graptemys yuwonoi Geoemyda, Leucocephalon,
pani Cuora Notochelys
pardalis Geochelone zhoui Cuora
petersi Morenia
picta Chrysemys Outgroup
planicauda Acinixys, Pyxis caretta Caretta
G. platynota Geochelone dumerilianus Peltocephalus, Podocnemis
N. platynota Notochelys C. insculpta Carettochelys
polyphemus Gopherus, Xerobates fimbriatus Chelus
pseudogeographica Graptemys, Malaclemys gibba Phrynops, Mesoclemmys
pulcherrima Callopsis, Geoemyda, Rhinoclemmys madagascariensis Erymnochelys, Podocnemis
pulchra Graptemys, Malaclemys megacephalum Platysternon
pulchristriata Cyclemys odoratus Kinosternon, Sternotherus,
punctularia Geoemyda, Rhinoclemmys Sternothaerus
quadriocellata Clemmys, Sacalia scorpioides Kinosternon
radiata Asterochelys, Geochelone serpentina Chelydra
reevesii Chinemys siebenrocki Chelodina, Macrochelodina
reticularia Deirochelys spiniferaApalone, Trionyx
rubida Callopsis, Geoemyda, Nicoria, subglobosa Emydura, Tropicochelymys
Rhinoclemmys subrufa Pelomedusa
rubriventris Chrysemys, Pseudemys temminckii Macroclemys, Macrochelys
scripta Chrysemys, Pseudemys, Trachemys
signatus Homopus
silvatica Geoemyda, Heosemys Appendix 3
sinensis Ocadia Distribution of character states for 70 characters among
smithii Kachuga, Pangshura 46 species of testudinoid turtles; a = 0/1; b = 0/2; c = 0/3;
spekii Kinixys d = 0/1/2; e = 1/2; f = 1/3; g = 2/3.
spengleri Geoemyda
spinosa Geoemyda, Heosemys agassizii
stejnegeri Chrysemys, Pseudemys, Trachemys 11a10 00000 000?0 11010 00110 00001
subtrijuga Malayemys 01000 03200 00000 00120 22001 22200
sulcata Geochelone 00000 10110
sylhetensis Kachuga, Pangshura amboinensis
taylori Chrysemys, Pseudemys, Trachemys 00a00 00010 00001 00000 00010 00000
tcheponensis Cyclemys, Geoemyda 00000 01110 00200 0a011 11110 00001
tecta Kachuga, Pangshura 20000 00000
tentoria Kachuga, Pangshura annandalei
tentorius Psammobates 0010? 00020 00011 10001 00010 00000
terrapen Chrysemys, Pseudemys, Trachemys 00100 01100 00000 00a10 34000 00001
terrapin Malaclemys 00000 0000?
texana Chrysemys, Pseudemys annulata
thurjii Hardella 00a00 00010 0010a 00000 00010 00000
tornieri Malacochersus 00000 03100 00000 000?0 11000 0??01
travancorica Geochelone, Indotestudo 00000 00010
tricarinata Geoemyda, Melanochelys barbouri
trifasciata Cistoclemmys, Cuora 00001 11000 01010 10000 10000 101a0
trijuga Geoemyda, Melanochelys 00001 10000 00000 00000 34000 22200
trivittata Kachuga 00000 00001
venusta Chrysemys, Pseudemys, Trachemys baska
versa Graptemys, Malaclemys 00001 00000 00010 10001 00000 00a02
werneri Testudo 11100 00010 00000 10110 43000 00000
00000 01000 00100 00010

bealei guttata
00000 00010 00000 10000 00010 00000 00010 00000 00000 00000 10000 00000
00000 00000 0??00 0a010 34000 0000? 00000 00000 00000 00000 34000 10201
00000 00000 00000 00000
blandingii hamiltonii
00000 00000 000?0 00000 10000 00000 01101 00001 00000 10000 00010 00110
00000 0c000 00000 01001 11110 1120a 00111 00000 00000 00a10 33001 0000a
10011 00000 00000 00000
borneensis homeana
00100 00000 00010 00000 00010 10000 11010 00000 000?1 11000 00111 00000
00000 00000 00010 00a10 33000 00000 00001 0110a 00000 00020 22001 ???1a
00000 00000 20000 10110
borneoensis horsfieldi
00001 00000 00010 10001 00010 00001 11010 00000 000?0 11000 a0111 00001
11100 00000 00100 00a10 43000 00000 00000 03200 00000 00020 22000 ???11
00000 00000 00000 11110
carbonaria kachuga
11010 00000 000?a 11000 00111 00001 00101 00000 00010 10001 00010 10001
00000 01201 00000 00120 22001 22210 11100 00000 01000 10110 43000 00000
00000 10110 00000 0000?
caspica mouhotii
00000 00010 0000a 00000 00010 00000 00010 00010 00000 11100 00010 00000
00000 00000 00000 0aa10 34000 00001 00011 01100 00000 00011 11110 00001
00000 00000 a0000 00010
crassicollis P. oculifera
00a00 00000 00000 10000 00010 10000 11010 00000 000?? 11000 00111 00001
00001 00000 00010 00a10 33000 00001 00001 01200 00000 00120 22001 ???10
00000 00000 00000 10110
elongata orbicularis
11010 00000 000?a 11000 00111 00001 00000 00000 00000 00000 a0000 00000
00000 03200 00000 00020 22001 ???11 00000 00000 00000 0a001 11010 1120a
00000 10110 10010 00000
dentata T. ornata
00000 00010 00000 00000 00000 00000 00010 00010 00000 00000 10000 00000
00001 01100 00001 00011 11000 01101 00000 00010 00200 00001 00110 ed201
00100 00000 20011 00010
emys pardalis
11010 00000 000?a 11000 00110 00001 11010 00000 000?a 11000 00111 00001
00000 01000 02000 00aa0 22001 ???00 00000 0f201 00000 00120 22001 ???10
00000 10110 00000 10110
flavomarginata petersi
00010 00020 0000a 00000 00010 00000 01100 00000 00010 10001 00010 01111
00000 01110 00100 00011 11110 ?1101 12100 02000 00100 00110 34000 00000
21000 00010 00000 00001
graeca picta
11010 00000 000?1 11000 10111 00001 00001 10000 00010 00000 10000 00000
00000 03200 00000 00a20 22000 22211 00000 00000 00000 0a0a0 34000 2220a
00000 10110 00000 0000a
grandis N. platynota
0010a 00020 00000 00000 000a0 00000 01101 00010 00000 00000 00a10 00000
00001 01100 0000a 00a10 34000 00001 00001 00000 10100 00001 03010 00101
10000 00000 00000 10110

polyphemus
11010 00000 000?0 11000 a0110 00001
01000 0a200 00000 00120 22001 22200
00000 10110
pulcherrima
00a00 00010 00001 00000 a0010 00000
00000 01100 00000 000a0 14000 01?01
00000 00000
reevesii
00101 10000 00000 10000 00010 10110
00110 00000 00000 00aa0 33000 00001
00000 00000
reticularia
00000 10000 000?0 00000 10000 00000
00000 00000 00000 010a0 34000 22200
10000 00001
scripta
00001 10000 00010 10000 10000 00101
10000 00000 00000 00000 34000 22200
00000 0000a
sinensis
00001 00000 00010 10001 10010 00001
00000 00000 00000 00aa0 33000 00001
00000 00000
spengleri
000a0 00000 00100 10100 00010 00000
00011 01100 00000 00000 34000 00001
00000 00010
subtrijuga
00100 01100 11000 10000 01010 10100
00110 00g00 00000 00a10 33000 00000
00000 00000
tentoria
00001 00000 00010 10001 00010 10001
10100 10000 01000 10110 43000 00000
00000 0000?
terrapin
00001 10000 00010 10000 a0000 10100
00000 00000 00000 0a000 34000 ed200
00000 00001
texana
00001 10000 0a010 10000 a0000 00101
10000 00000 00000 00000 34000 22200
00000 0000a
thurjii
00001 00000 00010 10001 00000 01011
11100 00000 00100 00110 43000 00000
00000 00001
tornieri
11010 00000 000?1 11000 00111 00001
00000 0??00 00000 00020 22000 2220a
View publication stats

5 Joyce Bell 2004

Uploaded by

Copyright:

Available Formats

5 Joyce Bell 2004

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

5 Joyce Bell 2004

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

A review of the comparative morphology of extant testudinoid turtles (Reptilia:

Article · January 2004

Walter Joyce Christopher Bell

SEE PROFILE SEE PROFILE

Quaternary Arvicoline Rodent Paleontology and Biochronology View project

Patterns of evolution of side-necked turtles (Pleurodira) View project

The user has requested enhancement of the downloaded file.

A Review of the Comparative Morphology of Extant Testudinoid Turtles

WALTER G. JOYCE1 AND CHRISTOPHER J. BELL2

1Department of Geology and Geophysics, Yale University, New Haven, CT 06511

Introduction subclades of testudinoid turtles is desirable. A more

© 2004 by Asiatic Herpetological Research

basioccipital is present in all outgroups. Its absence is Mandible

Conclusions produce a phylogenetic hypothesis in this paper was

As practicing vertebrate paleontologists, we also Graduate Fellowships provided funding to WGJ.

van Dijk, P. P., B. L. Stuart, and A. G. J. Rhodin, eds.

Waagen, G. N. 1972. Musk glands in recent turtles. M.S.

Appendix 1 15482, YPM 15486; kachuga, FMNH 224128, FMNH

orbicularis Emys yaquia Chrysemys, Pseudemys, Trachemys

00000 01000 00100 00010

10000 00000 00000 10110

View publication stats

You might also like