Content uploaded by Graciela Piñeiro
Author content
All content in this area was uploaded by Graciela Piñeiro
Content may be subject to copyright.
A NEW PROCOLOPHONOID (REPTILIA, PARAREPTILIA) FROM THE UPPER PERMIAN
OF URUGUAY
GRACIELA PIN
˜EIRO, ALEJANDRA ROJAS, and MARTI
´N UBILLA
Departamento de Paleontologı´a, Facultad de Ciencias, Igua´ 4225, CP 11400, Montevideo, Uruguay, fossil@fcien.edu.uy
ABSTRACT—Pintosaurus magnidentis gen. et sp. nov., a latest Permian procolophonoid reptile from the Buena Vista
Formation of Uruguay, is described and its relationships are considered. The new taxon is characterized by the absence
of a lateral maxillary depression, a premaxilla with a dorsally expanded maxillary process, a maxilla-premaxillary junction
almost straight and marked by a shallow, but well-defined step that displaces the premaxillary alveolar margin slightly
medially with respect to the maxillary tooth row, and the presence of a large conical, fang-like palatine tooth, which is
positioned next to the suture with the vomer. The transitional nature of Pintosaurus is reflected in the retention of several
plesiomorphic features that contrast with the presence of synapomorphies that are shared with procolophonids. A
phylogenetic analysis suggests a basal position for Pintosaurus within Procolophonoidea, either as the sister taxon of a
clade that includes Coletta and procolophonids or as the sister taxon of Coletta itself. The results of our phylogenetic
analysis, together with the Permian age of Pintosaurus, are consonant with the hypothesis of a Gondwanan origin of
Procolophonoidea.
INTRODUCTION
Procolophonoids represent a group of small to medium-size
parareptiles that range in time from the Late Permian to the Late
Triassic. Most recent studies (Laurin and Reisz, 1995; Lee, 1995;
deBraga and Reisz, 1996; Modesto et al., 2001; Reisz and Scott,
2002; deBraga, 2003) recognize two clades within Procolopho-
noidea. One of them is Owenettidae, erected by Broom (1939) in
order to distinguish his new genus Owenetta from the more de-
rived Procolophon. Until recently, owenettid procolophonoids
were represented only by Owenetta rubidgei from the Upper
Permian Cistecephalus and Dicynodon Assemblage Zones of
South Africa (Broom, 1939; Gow, 1977a) and Barasaurus be-
sairei from Upper Permian strata of Madagascar. However, the
recent description of a new owenettid, Owenetta kitchingorum
from the Lystrosaurus Assemblage Zone (Reisz and Scott, 2002),
extends the stratigraphic record of owenettids into the Triassic.
The other monophyletic procolophonoid clade, represented
by Procolophonidae and its close relatives, appears to have ex-
perienced a low level of extinction in the Permo-Triassic mass
extinction, and achieved cosmopolitan distributions during the
Triassic Period (Modesto et al., 2001). Within Procolophonidae,
two distinct groups have been recognized, the Procolophoninae
and the Leptopleuroninae (Ivakhnenko, 1979; Modesto et al.,
2002). Procolophonines include the basal Gondwanan procolo-
phonid Procolophon, from the Lower Triassic of South Africa,
South America, and Antarctica and until recently the genus The-
legnathus from the Middle Triassic of South Africa (Modesto et
al., 2002). A recent phylogenetic analysis by deBraga (2003) sug-
gests that Procolophoninae include Procolophon and the Lower
Triassic Russian taxon Tichvinskia, whereas Thelegnathus is a
leptopleuronine. Leptopleuronines include the most derived
procolophonid taxa such as Leptopleuron, from the Late Triassic
of Scotland, and Hypsognathus and Scoloparia, both from the
Late Triassic of North America (Sues and Baird, 1998; Spencer,
2000; Sues et al., 2000; Modesto et al., 2002; deBraga, 2003).
Recently, the Lystrosaurus Assemblage Zone yielded Coletta
seca (Gow, 2000) and Sauropareion anoplus (Modesto et al.,
2001), two procolophonoids that may be considered transitional
between the owenettids and procolophonids in terms of skull and
dental characters, but remain plesiomorphic with respect to Pro-
colophon and its closest relatives. Coletta was initially described
by Gow (2000) as a member of the Procolophonidae, due to the
presence of several synapomorphies that characterize this group
(Laurin and Reisz, 1995). However, later phylogenetic studies con-
sidered it as the sister taxon of the Procolophonidae, a position that
supports the biogeographic hypothesis that Procolophonoidea orig-
inated in Gondwana during the Permian and achieved a cosmo-
politan distribution only in the Triassic (Modesto et al., 2002).
Another transitional taxon is Contritosaurus from the Lower Tri-
assic of Russia, which has a basal position in previous phylogenetic
analyses. Its primitive and transitional nature is suggested by its
weakly differentiated cheek dentition (Ivakhnenko, 1974).
Although procolophonids were a cosmopolitan clade during
the Triassic, only three taxa have been reported from South
America, and these are known exclusively from the Triassic of
Brazil. Procolophon pricei Lavina, 1983 and Procolophon
brasiliensis Cisneros and Schultz, 2002 come from the Lower
Triassic Sanga do Cabral Formation and Candelaria barbouri
Price, 1946 comes from the Middle Triassic Santa Maria Forma-
tion. Procolophon pricei and P.brasiliensis are considered close
relatives of Procolophon trigoniceps from the Lower Triassic
Lystrosaurus Assemblage Zone of the Karoo Basin, South Af-
rica (Barberena et al., 1985; Cisneros and Shultz, 2002). The
Sanga do Cabral Formation is considered a stratigraphic equiva-
lent of the Lystrosaurus Assemblage Zone due to the presence of
the genus Procolophon (Barberena et al., 1985). However, the
remaining elements of the two faunas are quite different.
The Permian record of continental tetrapods in Uruguay is
extremely poor. The Yaguarı´ and the Buena Vista formations
represent the Upper Permian sequences in this country and it is
possible that the latter constitutes a transitional Permo-Triassic
unit, including upper levels of earliest Triassic age (Goso et al.,
2001). Until recently, the entire Buena Vista Formation has been
correlated lithostratigraphically to the Lower Triassic Sanga do
Cabral Formation of Brazil (Bossi and Navarro, 1991). However,
only a partial skull of a temnospondyl amphibian related to the
Dvinosaurus-Tupilakosauridae clade (Marsicano et al., 2000)
had been described from the Buena Vista Formation. Since the
limited nature of this specimen precludes a more precise assess-
ment of its phylogenetic relationships, conclusive inferences
about the age of the Buena Vista Formation cannot be made
(Marsicano, pers. comm., 2002). Recent collecting efforts on the
intraformational conglomerates of the Buena Vista Formation
yielded a continental tetrapod fauna, which lacks therapsid syn-
apsids and includes varanopid “pelycosaurs” (Pin˜eiro et al.,
2003), procolophonoids, other reptiles not yet described, and
basal stereospondyls. The presence of varanopids and the ab-
Journal of Vertebrate Paleontology 24(4):814–821, December 2004
© 2004 by the Society of Vertebrate Paleontology
814
sence of Triassic index fossils are together suggestive of a Late
Permian age for the intraformational conglomerates of the
Buena Vista Formation.
Among the reptile material is a partial skull of a small para-
reptile, which represents a basal procolophonoid that is de-
scribed in this paper. The specimen consists of an incomplete
skull and articulated lower jaws, which display several differ-
ences with known procolophonoid reptiles and support the rec-
ognition of a new genus and species. We conduct a phylogenetic
analysis to determine the relationships of the Uruguayan taxon
to other procolophonoids, especially those of the Upper Permian
and Lower Triassic of South Africa.
Institutional Abbreviations—FC-DPV, Departamento de
Paleontologı´a de Vertebrados, Facultad de Ciencias, Monte-
video, Uruguay.
SYSTEMATIC PALEONTOLOGY
PARAREPTILIA Olson, 1947
PROCOLOPHONIA Seeley, 1888
PROCOLOPHONOIDEA Romer, 1956
PINTOSAURUS MAGNIDENTIS gen. et sp. nov.
(Figs. 2, 3)
Holotype—FC-DPV 1181; partial skull and lower jaws pre-
served in tight occlusion.
Diagnosis—Small procolophonoid reptile distinguished by
maxilla-premaxillary contact being almost straight and forming a
shallow but well-defined step, such that the lateral surface of the
premaxilla is slightly recessed relative to the lateral surface of the
maxilla, a premaxilla with a posterior (maxillary) dorsally ex-
panded process, a maxilla dorsoventrally shallow above the
tooth row and lacking both a premaxillary process and a lateral
maxillary depression, and the presence of one very large palatine
tooth positioned close to the vomerine-palatine suture. Pintosau-
rus is distinguished from owenettids by the presence of two rows
of parachoanal vomerine teeth and it is distinguished from pro-
colophonids by the following plesiomorphies: upper and lower
marginal tooth bases and crowns that are conical with pointed,
non-recurved tips and the absence of labiolingually expanded
teeth.
Locality and Horizon—A locality near Colonia Orozco,
Cerro Largo County, Uruguay (Fig. 1). The holotype comes
from intraformational conglomerate levels intercalated within a
red sandstone of the Buena Vista Formation. The conglomerates
are dated as Late Permian and lie conformably below strata of
the same formation that are probably lowermost Triassic in age
(Goso et al., 2001).
Etymology—The prefix of the generic name, Pinto-, honors
Dr. Iraja Damiani Pinto, for his substantial contributions to
South American paleontology, and for his support and friend-
ship over the years. The generic suffix, saurus, is Greek for ‘liz-
ard’. The trivial name refers to the large palatine tooth set next
to the palatine-vomerine suture.
DESCRIPTION
Almost all the materials that come from the conglomerate
levels of the Buena Vista Formation are fragmentary and highly
disassociated elements, with articulated bones being relatively
rare. The specimen described here consists of a fragmentary
skull and associated lower jaws (Figs. 2, 3) in which most of the
roofing elements, the skull table and the braincase are missing.
The skull of Pintosaurus magnidentis is small, approximately
30 mm in length, appears lightly constructed, and in dorsal and
ventral view shows a subtriangular outline more reminiscent of
that of Owenetta rubidgei (Reisz and Scott, 2002), than that of
Procolophon (Carrol and Lindsay, 1985) because it does not
appear so laterally expanded posteriorly.
Skull Roof
Unfortunately, the specimen is poorly preserved and most of
the dermal bones that form the skull roof are missing. The pre-
maxilla and the maxilla can be described from the left side of the
skull, where these elements are almost complete. The right max-
illa is weathered immediately posterior to the maxillary foramen,
but its complete alveolar margin and teeth are preserved in tight
articulation with the right mandibular ramus.
FIGURE 2. Pintosaurus magnidentis gen et sp. nov., holotype FC-DPV
1181, skull in left lateral view. Abbreviations:an, angular; d, dentary; ec,
ectopterygoid; j, jugal; m, maxilla; mf, antero-lateral maxillary foramen;
pal, palatine; pdp, premaxillary postero-dorsal process; prm, premaxilla;
sa, surangular; sp, splenial, v, vomer. Scale bar equals 5 mm.
FIGURE 1. Location of Colonia Orozco in Cerro Largo County,
northeastern Uruguay, where Pintosaurus magnidentis was collected.
PIN
˜EIRO ET AL.—A NEW PROCOLOPHONOID FROM URUGUAY 815
The left premaxilla is a slender and thin bone that forms the
ventral and at least part of the anterior border of the narial
opening. As in Coletta, the owenettids, and some procolophonids
(e.g., Hypsognathus), but unlike Procolophon trigoniceps, the
nasal process of the premaxilla extends directly dorsally and not
anterodorsally to form the internarial bar. The posterodorsal
extension of this process is missing because of the weathering of
the specimen. So, we are not able to say how much it contributes
to form the anterior border of the external nares. Posteriorly,
where the premaxilla contacts the anterior end of the maxilla, it
develops a dorsally expanded maxillary process (Fig. 2) that is
reminiscent of the condition seen in P.trigoniceps (Carroll and
Lindsay, 1985) and possibly in Hypsognathus (Sues et al.,
2000:fig. 2E, F).
There is a shallow but well-defined step where the premaxilla
contacts the maxilla (Figs. 2, 3), indicating that the alveolar mar-
gin of the premaxillary teeth is not continuous to the maxillary
tooth row, but is positioned slightly medially displaced with re-
spect to that of the maxillary teeth. Three conical teeth of sub-
equal size are present in each premaxilla. This is also the condi-
tion present in the Russian taxon Contritosaurus.Tichvinskia has
three premaxillary teeth, but the first tooth is the largest of the
series.
The maxilla of Pintosaurus magnidentis is dorsoventrally shal-
low and similar in outline to that of owenettids (Reisz and Scott,
2002) and Coletta (Modesto et al., 2002), but differs in having a
short, less dorsally extended anterodorsal process and the lack of
a distinct maxillary premaxillary (subnarial) process. Thus, the
contact between the premaxilla and maxilla is almost straight,
and this feature is reminiscent of that seen in Contritosaurus and
to some extent in Coletta, although a small subnarial process
remains in the latter. Owenetta kitchingorum and Owenetta ru-
bidgei both feature a relatively large subnarial process, but only
in O. kitchingorum does it overlie the premaxilla, below the exter-
nal naris as illustrated by Reisz and Scott (2002:fig. 2D, 4A, B).
Given the weathering that has removed most of the skull roof,
it is very hard to determine if the maxilla contributes to the
lateral border of the external naris; if it does, the contribution
was probably minimal. Posteriorly, the maxilla of Pintosaurus
seems less elongate than that of either Owenetta rubidgei (Gow,
1977a) or Owenetta kitchingorum (Reisz and Scott, 2002), and at
its posterior end there is a short edentulous segment. There is no
“maxillary lateral depression”(sensu Carroll and Lindsay, 1985)
antero-dorsally on the maxilla. Dorsal to the third tooth, there is
a large antero-lateral maxillary foramen (Fig. 2), which is present
in most parareptiles (deBraga and Reisz, 1996), and more pos-
teriorly, just dorsal to the eighth tooth there is a small supralabial
foramen.
Ten teeth are preserved in the right maxilla and nine in the
left. However, there are two empty tooth places on the right side
and three on the left (Fig. 2), indicating a total of 12 teeth. The
maxillary teeth are conical and sharply tipped as seen in Owen-
etta kitchingorum,Coletta, and Sauropareion, but in Pintosaurus
they are not recurved. Most teeth are subequal in size, although
the anteriormost and posteriormost teeth appear slightly shorter
than the others. Unlike the condition seen in Procolophon and
other procolophonids, the teeth of Pintosaurus are not inset from
the lateral margin of the skull. The teeth of Pintosaurus are
unique among procolophonoids in that they are ornamented by
delicate fluting (Fig. 2). The maxillary teeth (and also the pre-
maxillary teeth) are ankylosed to the jaw by spongy bone of
attachment, which is consistent with the identification of proto-
thecodont implantation in procolophonids (Gow, 1977b; Small,
1997).
The right jugal is slender anteriorly and broadens posteriorly,
but it is disarticulated and turned over the right splenial (Fig. 3),
and so is seen only in medial view. It appears to resemble very
closely that of other basal procolophonoids such as Coletta and
Sauropareion.
Palate
Premaxillae, vomers, palatines, and the left ectopterygoid are
preserved and exposed in ventral view (Fig. 3); the pterygoids
and the elements of the palatoquadrate complex are not pre-
served. The premaxillae are less developed posteriorly than in
Procolophon, such that the vomers meet the premaxillae more
anteriorly. The internal naris is partially covered by the lower
jaws, precluding a detailed description of this vacuity.
The vomers are relatively broad and contact each other along
their entire medial margin. There are several large, pointed
teeth, which are arranged in two parachoanal rows. The first
tooth of each series is larger than the succeeding teeth, as in
Procolophon trigoniceps (Carroll and Lindsay, 1985), but in Pin-
tosaurus there is also a large tooth in the end of each lateral row
(Fig. 3). Two parachoanal tooth rows are also present in Coletta
(Modesto et al., 2002) and Contritosaurus (Ivakhnenko, 1974),
but the teeth in those taxa appear to be subequal in size. Unlike
some Russian procolophonids (e.g., Contritosaurus and Tichvin-
skia) and owenettids, Pintosaurus magnidentis lacks a tooth row
set close to the intervomerine suture.
Only the left palatine is complete. Anteriorly it contacts the
vomers with a slightly overlapping, oblique suture. The palatine
gradually narrows posteriorly and the postero-medial contact
with the pterygoid is concave. The palatine dentition consists of
a single large, conical, fang-like tooth placed immediately pos-
terior to the vomer-palatine suture (Fig. 3). The pterygo-palatine
row of teeth that is present in most procolophonids is absent in
Pintosaurus. Postero-laterally, the palatine contacts a relatively
small ectopterygoid and on this suture, close to the contact with
the maxilla, opens the small suborbital foramen.
FIGURE 3. Pintosaurus magnidentis gen et sp. nov., holotype FC-DPV
1181, skull in ventral view. Arrow indicates the step at the maxilla-
premaxillary junction. See the legend of Figure 2 for abbreviations. Scale
bar equals 5 mm.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 24, NO. 4, 2004816
Lower Jaw
The lower jaws are preserved in tight articulation with the
skull, and the fragility of the material precludes its disarticulation
for better examination. The morphology of the posterior lower
teeth of the series can be viewed through a broken portion of the
right dentary. The teeth are conical like those of the upper den-
tition, and they are not inset from the lateral margin of the
dentary. The lower dentition bites entirely medial to the upper
tooth row, as a consequence there is no interdigitating arrange-
ment of the upper and lower teeth as is frequently observed in
procolophonids.
The symphysis is formed completely by the dentary. It is rela-
tively short anteroposteriorly and does not reach the level of the
internal naris. The labial surface of the dentary is not sculptured,
and there is no labial groove as described for Tichvinskia
(Ivakhnenko, 1973) and Contritosaurus (Ivakhnenko, 1974). The
alveolar margin extends parallel posteriorly to the ventral mar-
gin almost through the entire length of the dentary, and it di-
verges only slightly at the level of the suture with the angular and
surangular.
The number of dentary teeth cannot be established due to the
tight occlusion of the mandible and skull. Medial to the dentary
is the splenial (Fig. 3), a slender bone that extends posteriorly
and contributes to the anterior border of the small meckelian
foramen.
Only the anterior part of the angular and surangular are pre-
served, but a more detailed description cannot be made because
of the poor preservation of the mandible posterior to the den-
tary. Nothing remains of the coronoid and the articular.
PHYLOGENETIC RELATIONSHIPS
Coletta seca and Sauropareion anoplus, both from the Lower
Triassic of South Africa, may be considered, in morphological
terms, taxa that are transitional between the owenettids and the
procolophonids. Pintosaurus magnidentis from the Upper Permian
of Uruguay shows a similar transitional morphology that is sugges-
tive of a close relationship to these African procolophonoid species.
We conducted a parsimony analysis in order to determine the
phylogenetic position of Pintosaurus magnidentis with respect to
the new basal South African procolophonoids and to the Pro-
colophonidae. The following genera form the ingroup because
they are represented by specimens that preserve cranial charac-
ters facilitating comparison with our material: Coletta (Gow,
2000; Modesto et al., 2002); Contritosaurus (Ivakhnenko, 1974);
Tichvinskia (Ivakhnenko, 1973); Procolophon trigoniceps (Car-
roll and Lindsay, 1985; Colbert and Kitching, 1975); Thelegna-
thus (Gow, 1977c); Scoloparia (Sues and Baird, 1998), Hypsog-
nathus (Colbert, 1946; Sues et al., 2000) and Leptopleuron
(Huene, 1912, 1920; Sues et al., 2000). The taxon “Thelegnathus”
was retained in the ingroup, despite the recent declaration of the
type species T. browni as a nomen dubium, and reassignment of
the four species described by Gow (1977c) to four new genera
(Modesto and Damiani, 2003). Since in recent phylogenies (e.g.,
Modesto et al. 2002; de Braga, 2003) Thelegnathus has a contro-
versial position and, taking into account that the new monospe-
cific genera were neither exhaustively described nor their rela-
tionships evaluated yet, we decided to retain “Thelegnathus”in
our analysis and codified the state of the characters following
Gow (1977c). The parareptile Macroleter (Ivakhnenko, 1997)
and the owenettids Owenetta kitchingorum (Reisz and Scott,
2002) and Owenetta rubidgei (Broom, 1939; Gow, 1977b; Reisz
and Scott, 2002) represent the outgroups. The choice of Macro-
leter was based on previous phylogenetic studies that show the
basal position of this taxon with respect to the Procolophonoidea
(Reisz and Scott, 2002). Similarly, the choice of the owenettids
O.kitchingorum and O.rubidgei was based on their basal posi-
tion with respect to the Procolophonidae (Modesto et al., 2001).
The morphological differences between both species, as revealed
by their respective codings in the data matrix, argue for the
inclusion of both taxa in the analysis.
Sauropareion anoplus was excluded because several of the
characters used in the present analysis are indeterminate in the
holotype and only known specimen.
A taxon-character data matrix was constructed based on 23
characters (22 cranial and 1 postcranial) and 12 terminal taxa
(see Appendices 1 and 2). The analysis was run using PAUP
3.1.1 for Macintosh (Swofford, 1993). Character states were po-
larized by outgroup comparison. A branch-and-bound search,
followed by bootstrap and Bremer analyses, were performed and
the results of this analysis are shown in Figure 4. All the char-
acters remained unweighted, were treated as unordered, and op-
timized using the Deltran algorithm.
The analysis produced two most parsimonious trees, each with
a length of 41 steps. The consistency index (CI) is 0.756, the
retention index (RI) is 0.846 and the rescaled consistency index
(RC) is 0.640. A strict consensus of the two trees shows Pinto-
saurus in a trichotomy with Coletta and the Procolophonidae
(Fig. 4C). Pintosaurus is either the sister taxon of a clade com-
prising Coletta and the Procolophonidae, or Pintosaurus and Co-
letta form a clade that is the sister group of Procolophonidae
(Fig. 4A, B). A single ambiguous synapomorphy links Pintosau-
rus with Coletta: the absence of vomerine teeth along the medial
vomerine suture (character 12, state 1). The trichotomy shown in
the consensus tree is presumably due to the limited information
afforded by the single incomplete skull of Pintosaurus.
Another interesting point of our results is the observation that
in both trees, the owenettids Owenetta rubidgei and Owenetta
kitchingorum do not form a clade, with the latter as the most
basal member of Procolophonoidea. This is unexpected, because
a previous analysis (Modesto et al., 2001) that included all owen-
ettid taxa recovered the family as monophyletic but showed an
unresolved trichotomy between the two species of Owenetta and
Barasaurus.
The basal position of Contritosaurus and Tichvinskia within
Procolophonidae and the topology of the clades at nodes F (Pro-
colophon trigoniceps +“Thelegnathus”)(⳱Procolophoninae), G
(Scoloparia (Hypsognathus +Leptopleuron)) (⳱Leptopleuroni-
nae) and H (Hypsognathus +Leptopleuron) (Leptopleuronine)
are retained in all trees (Fig. 4), confirming previous phyloge-
netic work (Sues et al., 2000; Modesto et al., 2002). However, in
our phylogeny the characters that support the clades (Fig. 4C)
are slightly different.
The bootstrap and Bremer support values (Fig. 4) show that
the nodes corresponding to the procolophonines and leptopleu-
ronines (nodes F, G, and H) are weakly supported and each
require only one extra step to collapse. However, the clade at
node E is relatively stronger (decay index⳱3). The sister-group
relationship between Tichvinskia and the clade including pro-
colophonines and leptopleuronines (node D) appears slightly
more strongly supported (two extra steps) than in the phylogeny
of Modesto et al. (2002) in which Tichvinskia required one or
two extra steps to form a sister-group relationship with the lep-
topleruronines and procolophonines.
Regarding the topology shown in the consensus tree (Fig. 4C)
the different clades are supported by the following characters:
Node a—This clade is diagnosed on the basis of two synapo-
morphies but only one is unambiguous: all taxa lack postparietals
(character 18, state 1). The additional synapomorphy, the pres-
ence of a suborbital foramen enclosed by the palatine and
ectopterygoid (character 19, state 1) does not diagnose more
derived members of the clade (character 19, state 2). Paired
postparietals at the posterior end of the skull table and the sub-
orbital foramen formed by palatine, ectopterygoid, and ptery-
goid (Reisz and Scott, 2002) separate Owenetta kitchingorum
from O. rubidgei and the remaining taxa considered.
PIN
˜EIRO ET AL.—A NEW PROCOLOPHONOID FROM URUGUAY 817
FIGURE 4. Cladograms showing the relationships of Pintosaurus magnidentis within Procolophonoidea. Aand Bshow the two most parsimonious
trees discovered in a PAUP analysis, whereas Cshows the consensus of A and B. Tree length, 41; consistency index, 0.756; retention index, 0.846;
rescaled consistency index, 0.640. Lists of synapomorphies that diagnose the clades are detailed in the text. Bold numbers indicate reversals and an
asterisk indicates ambiguous optimization of the character. Bootstrap and Bremer values for each node: a, 84/2; b, 88/2; c, 79/2; d, 76/2; e, 98/3; f, 64/1;
g, 64/1; h, 69/1.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 24, NO. 4, 2004818
Node b—This clade is supported by two possible synapomor-
phies: three premaxillary teeth (character 2, state 2), reversed in
Coletta and the procolophonines (character 2, state1), whereas the
leptopleuronines have the other derived condition (character 2,
state 3). The other synapomorphy is the presence of one or two
rows of parachoanal teeth on the vomers (character 11, state 1).
Node c—There are two synapomorphies that diagnose Procolo-
phonidae: 10 or fewer maxillary teeth (character 6, state 1) and
cheek teeth bases labio-lingually expanded (character 7, state 1).
Node d—Five synapomorphies support this clade: first or sec-
ond premaxillary teeth larger than the others (character 3, state 1,
unknown in “Thelegnathus”); maxillary tooth crowns labio-
lingually expanded (character 8, state 1); pterygo-palatine teeth
present and aligned antero-medially (character 13, state 1); orbit
posteriorly extended well beyond the level of the pineal foramen
(character 15, state 2); and postfrontal absent and area ocuppied by
the parietal (character 16, state 2). The last two characters are re-
versed in Procolophon and “Thelegnathus”(character 16, state 1).
Node e—This clade is supported by five synapomorphies.
Three of them are unambiguous synapomorphies: premaxillary
nasal process extended antero-dorsally (character 1, state 1), re-
versed in Hypsognathus; maxillary cheek teeth inset from the
lateral margin of the skull and mandibles (character 9, state 1);
and anterior dentary incisiform teeth procumbent (character 20,
state 1). The last one could be reversed in some species of “The-
legnathus”. Of the remaining features, one is problematic be-
cause it is lost in most taxa and appears homoplastic: absence of
vomerine teeth along medial vomerine suture (character 12,
state 1). The other one, suborbital foramen delimited by pala-
tine, ectopterygoid, and maxilla (character 19, state 2), is present
only in P. trigoniceps and “Thelegnathus”, but remains unknown
in leptopleuronines.
Node f—Five synapomorphies group the Procolophoninae:
quadratojugal lateral surface with a single large and posteriorly
directed process (character 17, state 1); four premaxillary teeth
(character 2, state 1); dentary teeth not biting medial to the
maxillary teeth, but there is a contact between the tips of the
upper and lower dentition (character 10, state 1); orbit extended
posteriorly to the level of the posteriormost margin of the pineal
foramen (character 15, state 1); and postfrontal not contacting
postorbital (character 16, state 1).
Node g—The Leptopleuroninae are diagnosed by three un-
ambiguous synapomorphies: two premaxillary teeth (character 2,
state 3); quadratojugal lateral surface with two or more radiating
spines circumscribed at their basis by distinct sulci (character 17,
state 2); and jaw articulation positioned below level of marginal
dentition (character 22, state 1).
Node h—Two shared derived synapomorphies support this
clade: prefrontal confined entirely to the orbital rim (character
14, state 1) and the presence of only one incisiform tooth on the
dentary (character 21, state 1).
DISCUSSION
Despite the incomplete preservation of the specimen of Pin-
tosaurus magnidentis, enough information is preserved to allow
us to identify it as a new procolophonoid and to investigate its
relationships to other members of Procolophonoidea. Within
this group, the basal position of Pintosaurus appears to be re-
lated to the presence of several primitive characters shared with
owenettids, such as the absence of a maxillary depression, pre-
maxillary and maxillary teeth that are conical with subcircular
cross-sections at their base and crown, the absence of labio-
lingually expanded cheek teeth, and 12 maxillary tooth positions.
Except for the first character, the last two are shared with Coletta
and these similarities are expressed in the cladogram through the
possible sister-group relationship seen in one of the most parsi-
monious trees (Fig. 4 B).
Nevertheless, Pintosaurus shares many derived conditions
with basal procolophonids such as Contritosaurus,Tichvinskia,
and Procolophon. The presence of two vomerine parachoanal
tooth rows, also present in Coletta, is shared with both Contri-
tosaurus and Procolophon, whereas the dorsally-expanded max-
illary process of the premaxilla was only observed in Procolo-
phon (Carroll and Lindsay, 1985) and seems to be also present in
Hypsognathus (Sues et al., 2000:fig. 2E and F).
Additionally, the evaluation of the distribution of several char-
acter states used in the phylogenetic analysis allows a consider-
ation of the evolution of the following characters within Procol-
ophonoidea:
(1) A slender maxillary process of the premaxilla. This could be
a derived condition that was acquired early in procolopho-
noid evolution, because Pintosaurus has this character. How-
ever, it is also present in Procolophon trigoniceps and Hyp-
sognathus.
(2) “Maxillary depression”(sensu Carroll and Lindsay, 1985).
The absence of an antero-dorsal maxillary fossa is a plesio-
morphic state shared by Pintosaurus and owenettids. A max-
illary fossa is present in the procolophonoid Coletta and the
procolophonids Contritosaurus,Tichvinskia, and Procolo-
phon and is considered a procolophonid autapomorphy
(Gow, 2000 following Laurin and Reisz, 1995). However, it is
not clear if the “maxillary depression”(sensu Carroll and
Lindsay, 1985) is equivalent to the “narial shelf”listed by
Laurin and Reisz (1995) as a procolophonid autapomorphy,
for in Hypsognathus they are exclusive characters (only the
second is present) (Sues et al., 2000). On the other hand, the
maxillary depression also extends over the nasal in Coletta,
Procolophon,Contritosaurus, and Tichvinskia. This charac-
ter appears as homoplastic in our phylogenetic study and was
not considered as a synapomorphy, but only as an apomor-
phy of those taxa having a maxillary depression.
(3) Contribution of the maxilla to the external naris. This char-
acter refers to the variable contribution of the maxilla to the
ventral and postero-lateral border of the external naris. In
the primitive condition, as seen in owenettids (Reisz and
Scott, 2002), the maxilla has a long anterior premaxillary
ramus that extends below the external naris for half their
length. Coletta and Pintosaurus have successively more de-
rived conditions, in that the premaxillary ramus does not
extend anteriorly farther than the overlying septomaxilla, or
farther than the posterior border of the external naris,
respectively. Contritosaurus has a condition similar to that
present in Pintosaurus. Regarding the condition in Procolo-
phon and other procolophonids (e.g., Hypsognathus) there is
no contribution of the maxilla to the external naris. Thus, the
character state present in Coletta and Pintosaurus could rep-
resent intermediate stages between the owenettids and pro-
colophonids.
(4) Premaxillary tooth number and size. According to the results
of our phylogenetic analysis, the reduction of premaxillary
teeth from five to three was acquired early in procolopho-
noid evolution (Fig. 4), and the four premaxillary teeth pres-
ent in Coletta and the procolophonines represent reversed
conditions. However, as the node including Coletta,Pinto-
saurus, and Procolophonidae is unresolved, care must be
taken in the interpretation of this character. Procolophonids
exclusive of Contritosaurus have anterior premaxillary teeth
that are larger than the others and the phylogenetic analysis
considered this character to be a synapomorphy of the clade
that includes Tichvinskia, procolophonines, and leptopleu-
ronines. Regarding the current disagreements concerning
the phylogenetic position of Tichvinskia (Modesto et al.,
2002; deBraga, 2003; this study), this character could be a
synapomorphy of the clade comprising procolophonines and
leptopleuronines.
PIN
˜EIRO ET AL.—A NEW PROCOLOPHONOID FROM URUGUAY 819
(5) Absence of labio-lingually expanded teeth on both the max-
illary and dentary. This is a plesiomorphic condition of Pin-
tosaurus shared with owenettids, Coletta, and Sauropareion.
Procolophonids have maxillary and dentary teeth whose
bases and crowns are transversally expanded. However, in
Contritosaurus the maxillary and dentary teeth are labio-
lingually expanded only at their bases, whereas their crowns
retain pointed tips. Because in Tichvinskia the crown tips are
slightly transversely expanded (“quasicheek”teeth) (Ivakh-
nenko, 1973), these last two taxa could be regarded as exhib-
iting successive advanced states of the plesiomorphic condition,
as is suggested by their position in all the trees resulting from
the phylogenetic analysis performed in this study.
Pintosaurus has two autapomorphies that distinguish it from all
other procolophonoids: (1) the presence of a single, very large
palatine tooth placed close to the palatine-vomerine suture and
(2) a shallow step present at the maxilla-premaxillary contact,
which implies that the premaxillary alveolar margin is slightly
discontinuous with that of the maxillary tooth row. The former
character is clearly an autapomorphy of Pintosaurus, and may
represent a modification of the condition present in earlier Per-
mian taxa (e.g., owenettids) or could reflect a unique dietary
specialization. The later character was not observed in other
procolophonoid taxa, but it also could be related to the acquisi-
tion of a different, probably duraphageous diet.
Pintosaurus displays strong differences, especially in the mar-
ginal and palatal dentition, when compared to the Lower Triassic
species from Brazil. The probably coeval Procolophon pricei
(Lavina, 1983) and Procolophon brasiliensis (Cisneros and
Schultz, 2002) from Brazil possess only a single parachoanal vo-
merine row of relatively large pointed teeth, which seems to be
continued onto the pterygoids. The presence of teeth over the
palatine is hard to confirm on the holotype and only known
specimen of P. brasiliensis, due to poor preservation of that area
(pers. obs.). With respect to the cranial anatomy, there is insuf-
ficient information for comparison, but the Brazilian Procolo-
phon specimens are larger, more robust, and more heavily con-
structed, and all of them possess labiolingually expanded maxil-
lary and dentary teeth.
The hypothesis of a Gondwanan origin for the Procolopho-
noidea was recently proposed by Modesto et al. (2002) based on
their observation that the most basal members of the group were
Gondwanan in distribution. The discovery of Pintosaurus and its
identification as a basal procolophonoid provides additional evi-
dence in support of this biogeographic hypothesis, and it is re-
flected by both the Late Permian age of the sediments that
yielded the Uruguayan taxon and the morphologically transi-
tional status. Thus, Pintosaurus suggests that evolutionary events
including the acquisition of some derived cranial morphology
among primitive members of Procolophonoidea arose early in
Gondwanan regions of Pangaea.
CONCLUSIONS
Pintosaurus magnidentis, from the Buena Vista Formation of
Uruguay, is a new basal procolophonoid that lies on the procol-
ophonoid stem leading to Procolophonidae. Pintosaurus is either
the sister taxon of Coletta seca, from the Lower Triassic Lystro-
saurus Assemblage Zone of South Africa, or it may be the sister
taxon of a clade that includes Coletta and Procolophonidae. De-
spite this uncertainty, it is clear that Pintosaurus may be included
in a grade of procolophonoid taxa that are characterized by sev-
eral plesiomorphies shared with owenettids but not with procolo-
phonids.
The phylogenetic appraisal of Pintosaurus permits an analysis
of the evolution and status of the characters used previously for
identification of the major procolophonoid groups, revealing
some to be homoplasies. For the most part, the procolophonoid
phylogeny reported by Modesto et al. (2001) was supported by a
new analysis that included the new Uruguayan taxon. The un-
expected paraphyletic status of Owenetta could be biased by the
small data set used in the phylogenetic analysis and also may be
influenced by the selection of the characters, which were chosen
to determine the phylogenetic position of the new Uruguayan
procolophonoid. The preceding arguments, bearing in mind that
we were unable to examine personally any materials assigned to
either species of Owenetta, suggest that further studies are
needed for a more rigorous understanding of the phylogenetic
status of this genus.
Although the upper part of the Buena Vista Formation prob-
ably includes lowermost Triassic strata, the Permian age of the
intraformational conglomerates, established previously on the
basis of biostratigraphic and lithostratigraphic assumptions
(Goso et al., 2001), is now supported by the discovery of a con-
tinental faunal assemblage of Permian affinities (Pin˜eiro et al.,
2003). Pintosaurus, therefore, is the first record of a Permian
procolophonoid from South America.
Finally, the Permian age and paleogeographic distribution (in
what was western Gondwana) of the new Uruguayan procol-
ophonoid supports the hypothesis that this group of parareptiles
originated and first diversified in Gondwanan Pangaea, as first
proposed by Modesto et al. (2002).
ACKNOWLEDGMENTS
We thank Nora Lorenzo, Ce´sar Goso, and Mariano Verde all
of whom participated in the collection of the material. Santiago
Carreira, Guillermo D’Elı´a, and Fernando Pe´rez Miles helped us
with the cladistic analysis. We especially thank Dr. Sean
Modesto for language revision and his valuable comments about
procolophonoid relationships and Dr. Ross Damiani for provid-
ing images of Coletta and Thelegnathus specimens deposited in
the collections of the Bernard Price Institute for Paleontological
Research. Our most sincere thanks also go to the reviewers
Drs. Nicholas Fraser and Ross Damiani for their helpful sugges-
tions, which improved the final draft of the manuscript. Part of
our research was supported by the Programa de Desarrollo de
Ciencias Ba´sicas (PEDECIBA), Facultad de Ciencias, the Comi-
sio´n Sectorial de Investigacio´n Cientı´fica (CSIC) and the gov-
ernment of Cerro Largo County.
LITERATURE CITED
Barberena, M. C., D. C. Arau´jo, and E. L. Lavina. 1985. Late Permian
and Triassic tetrapods of southern Brazil. National Geographic Re-
search 1(1):5–20.
Bossi, J., and R. Navarro. 1991. Geologı´a del Uruguay. Departamento de
Publicaciones, Universidad de la Repu´blica, Montevideo, Uruguay,
970 pp.
Broom, R. 1939. A new type of cotylosaurian, Owenetta rubidgei. Annals
of the Transvaal Museum 19:319–321.
Carroll, R. L., and W. Lindsay. 1985. Cranial anatomy of the primitive
reptile Procolophon. Canadian Journal of Earth Sciences 22:
1571–1587.
Cisneros, J. C., and C. Schultz. 2002. Procolophon brasiliensis, n. sp., a
new procolophonid reptile from the Lower Triassic of Southern
Brazil. Neues Jahrbuch fu¨r Geologie und Pala¨ontologie, Monat-
shefte 11:641–648.
Colbert, E. H. 1946. Hypsognathus, a Triassic reptile from New Jersey.
Bulletin of the American Museum of Natural History 86:225–274.
Colbert, E. H., and J. W. Kitching. 1975. The Triassic reptile Procolo-
phon in Antarctica. American Museum Novitates 2566:1–23.
deBraga, M. 2003. The postcranial skeleton, phylogenetic position, and
probable lifestyle of the Early Triassic reptile Procolophon trigoni-
ceps. Canadian Journal of Earth Sciences 40:527–556.
deBraga, M., and R. R. Reisz. 1996. The Early Permian reptile Acleis-
torhinus pteroticus and its phylogenetic position. Journal of Verte-
brate Paleontology 16:384–395.
Goso, C. G. Pin˜eiro, H. de Santa Ana, A. Rojas, M. Verde, and C. Alves.
JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 24, NO. 4, 2004820
2001. Caracterizacio´n estratigra´fica de los depo´sitos continentales
cuspidales neope´rmicos (Formaciones Yaguarı´ y Buena Vista) en el
borde oriental de la Cuenca Norte Uruguaya. XI Congreso Latino-
americano de Geologı´a, III Congreso Uruguayo de Geologı´a CD-
ROM.
Gow, C. E. 1977a. Tooth function and succession in the Triassic reptile
Procolophon trigoniceps. Palaeontology 20:695–704.
Gow, C. E. 1977b. Owenetta in perspective. Palaeontologia africana 20:
115–118.
Gow, C. E. 1977c. New procolophonids from the Triassic Cynognathus
Zone of South Africa. Annals of the South African Museum 72(6):
109–124.
Gow, C. E. 2000. A new procolophonid (Parareptilia) from the Lystro-
saurus Assemblage Zone, Beaufort Group, South Africa. Palaeon-
tologia africana 36:21–23.
Huene, F. von. 1912. Die Cotylosaurier der Trias. Paleontographica 59:
69–102.
Huene, F. von. 1920. Ein Telerpeton mit gut erhaltenem Scha¨del. Cen-
tralblatt fu¨r Mineralogie, Geologie und Pala¨ontologie 1920:189–192.
Ivakhnenko, M. F. 1973. Skull structure in the Early Triassic procolo-
phonian Tichvinskia vjatkensis. Paleontological Journal 7:511–518.
Ivakhnenko, M. F. 1974. New data on the Early Triassic procolophonids
of the U.S.S.R. Paleontological Journal 8:346–351.
Ivakhnenko, M. F. 1979. Permian and Triassic procolophonids of the
Russian Platform. Akademiya Nauk SSSR 164:1–80. [Russian]
Ivakhnenko, M. F., V. K. Golubev, Y. M. Gubin, N. N. Kalandadze, I. V.
Novikov, A. G. Sennikov, and A. S. Rautian. 1997. Permian and
Triassic tetrapods of Eastern Europe. GEOS Moscow, 216 pp. [in
Russian]
Laurin, M., and R. R. Reisz. 1995. A reevaluation of early amniote
phylogeny. Zoological Journal of the Linnean Society 113:165–223.
Lavina, E. L. 1983. Procolophon pricei sp. n., um novo re´ptil procolo-
fonı´deo do Tria´ssico do Rio Grande do Sul. Iheringia, Se´rie Geo-
logica 9:51–78.
Lee, M. S. Y. 1995. Historical burden in systematics and the interrela-
tionships of “parareptiles.”Biological Review 70:459–547.
Marsicano, C., D. Perea and M. Ubilla. 2000. A new temnospondyl am-
phibian from the Lower Triassic of South America. Alcheringa 24:
119–123.
Modesto, S., and R. Damiani. 2003. Taxonomic status of Thelegnathus
browni Broom, a procolophonoid reptile from the South African
Triassic. Annals of Carnegie Museum 72:53–64.
Modesto, S. P., R. J. Damiani, and H.-D. Sues. 2002. A reappraisal of
Coletta seca, a basal procolophonoid reptile from the Lower Triassic
of South Africa. Palaeontology 45(5):883–895.
Modesto, S., H.-D. Sues, and R. Damiani. 2001. A new Triassic procol-
ophonoid reptile and its implications for procolophonoid survivor-
ship during the Permo-Triassic extinction event. Proccedings of the
Royal Society of London B 268:2047–2052.
Olson, E. C. 1947. The family Diadectidae and its bearing on the classi-
fication of reptiles. Fieldiana, Geology 11:1–53.
Pin˜eiro, G., M. Verde, M.Ubilla, and J. Ferigolo. 2003. First basal syn-
apsids (“pelycosaurs”) from the Upper Permian-?Lower Triassic of
Uruguay, South America. Journal of Paleontology 77:389–392.
Price, L. I. 1946. Um procolophonı´deo do Tria´ssico do Rio Grande do
Sul. Boletim da Divisa˜o de Geologia e Mineria 122:7–26.
Reisz, R. R., and D. Scott. 2002. Owenetta kitchingorum, sp. nov., a small
parareptile (Procolophonia: Owenettidae) from the Lower Triassic
of South Africa. Journal of Vertebrate Paleontology 22(2):244–256.
Romer, A. S. 1956. Osteology of the Reptiles. University of Chicago
Press, Chicago, 772 pp.
Seeley, H. G. 1988. Researches on the structure, organization, and clas-
sification of the fossil Reptilia. VI. On the anomodont Reptilia and
their allies. Proceedings of the Royal Society of London 44:381–383.
Small, B. J. 1997. A new procolophonid from the Upper Triassic of
Texas, with a description of tooth replacement and implantation.
Journal of Vertebrate Paleontology 17:674–678.
Spencer, P. S. 2000. The braincase structure of Leptopleuron lacertinum
Owen (Sauropsidae: Procolophonidae). Journal of Vertebrate Pale-
ontology 20:21–30.
Sues, H.-D., and D. Baird. 1998. Procolophonidae (Reptilia: Parareptilia)
from the Upper Triassic Wolfville Formation of Nova Scotia,
Canada. Journal of Vertebrate Paleontology 18:525–532.
Sues, H.-D., P. E. Olsen, D. M. Scott, and P. S. Spencer. 2000. Cranial
osteology of Hypsognathus fenneri, a latest Triassic procolophonid
reptile from the Newark Supergroup of eastern North America.
Journal of Vertebrate Paleontology 20:275–284.
Swofford, D. I. 1993. PAUP: Phylogenetic Analysis Using Parsimony,
version 3.1.1. Illinois Natural History Survey, Champaign.
Received 1 April 2003; accepted 27 January 2004.
APPENDIX 1
Description of characters and character states used in the phylogenetic
analysis. Characters were polarized with respect to Macroleter,Owenetta
kitchingorum, and Owenetta rubidgei as the outgroup. All characters are
treated as unordered. Most of the characters have been modified from
Modesto et al. (2001, 2002) whereas characters 3, 10, 12, and 20 are new.
(1) Premaxillary nasal process: extending dorsally (0); anterodorsally (1).
(2) Premaxillary tooth number: five or more (0); four (1); three (2);
two (3).
(3) Premaxillary tooth size: subequal (0); first or second teeth larger
and more robust than others (1).
(4) Premaxillary ramus of maxilla: contributing to border of external
naris (0); not contributing (1).
(5) Maxillary depression: absent (0); present (1).
(6) Maxillary tooth number: twelve or more (0); ten or less (1).
(7) Maxillary tooth basal cross-section: subcircular (0); labiolingually
expanded (1).
(8) Maxillary tooth crowns: subcircular (0); labiolingually expanded (1).
(9) Maxillary cheek teeth: not inset (0); inset from lateral margins of
skull and mandible (1).
(10) Dentary teeth bite: medial to upper (0); contact between upper and
lower teeth (1).
(11) Vomerine dentition: consisting of shagreen of small denticles (0);
consisting of one or two rows of parachoanal teeth (1); entirely
absent (2).
(12) Vomerine teeth or denticles along medial vomerine suture: present
(0); absent (1).
(13) Pterygo-palatine teeth: present and aligned antero-laterally (0);
present and aligned antero-medially (1); absent (2).
(14) Prefrontal: contributing to skull table (0); confined entirely to or-
bital margin (1).
(15) Orbit: terminating at level of anterior margin of pineal foramen (0);
extending posteriorly to level of posteriormost point of pineal fo-
ramen (1); extending posteriorly well beyond pineal foramen (2).
(16) Postfrontal: contacting postorbital (0); separated from postorbital
(1); absent and area ocuped by parietal (2).
(17) Quadratojugal lateral surface: spineless (0); with single large-based
and posteriorly directed process (1); with two or more radiating
spines whose bases are circumscribed by distinct sulci (2).
(18) Postparietal: present (0); absent (1).
(19) Suborbital foramen: formed ventrally by palatine, ectopterygoid
and pterygoid (0); by palatine and ectopterygoid alone (1); by pala-
tine, ectopterygoid and maxilla (2).
(20) Dentary incisiviform teeth: not procumbent (0); procumbent (1).
(21) Dentary incisiviform teeth number: two or more (0); one (1).
(22) Jaw articulation: in line with marginal dentition (0); positioned be-
low level of marginal dentition (1).
(23) Entepicondylar foramen of humerus: present (0); absent (1).
APPENDIX 2
Character-taxon matrix used for the phylogenetic analysis.
Missing data are coded by “?”; uncertainty for states 1 and 2 are coded
by “A”and polymorphisms for states 0 and 1 are coded by “B”.
510152023
Macroleter 00000 00000 00000 00000 000
Owenetta kitchingorum 00000 00000 00000 00000 001
Owenetta rubidgei 00000 00000 00000 00110 00?
Pintosaurus 02000 10000 112?? ???1? ???
Coletta 01001 0000? 11?00 00110 00?
Contritosaurus 02001 1100? 10000 A??20 00?
Tichvinskia 02101 11100 10102 20110 000
Procolophon trigoniceps 11111 11111 11101 11121 000
Scoloparia 131?0 1111? ???02 221?1 01?
Hypsognathus fenneri 03110 11110 21212 22121 110
Thelegnathus 11?00 11111 ???01 111?B 00?
Leptopleuron 131?0 11111 1??12 221?1 11?
PIN
˜EIRO ET AL.—A NEW PROCOLOPHONOID FROM URUGUAY 821
