(“Pelycosaurs”) from the Upper Permian

Home , Mycterosaurus

FIRST BASAL SYNAPSIDS (“PELYCOSAURS”) FROM THE UPPER PERMIAN-?LOWER TRIASSIC OF URUGUAY, SOUTH AMERICA Author(s): GRACIELA PIÑEIRO, MARIANO VERDE, MARTÍN UBILLA, and JORGE FERIGOLO Source: Journal of Paleontology, 77(2):389-392. Published By: The Paleontological Society DOI: http://dx.doi.org/10.1666/0022-3360(2003)077<0389:FBSPFT>2.0.CO;2 URL: http://www.bioone.org/doi/full/10.1666/0022-3360%282003%29077%3C0389%3AFBSPFT %3E2.0.CO%3B2

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research. J. Paleont., 77(2), 2003, pp. 389±392 Copyright ᭧ 2003, The Paleontological Society 0022-3360/03/0077-389$03.00

FIRST BASAL SYNAPSIDS (``PELYCOSAURS'') FROM THE UPPER PERMIAN-?LOWER TRIASSIC OF URUGUAY, SOUTH AMERICA

GRACIELA PINÄ EIRO,1 MARIANO VERDE,1 MARTIÂN UBILLA,1 AND JORGE FERIGOLO2 1Departamento de PaleontologõÂa, Facultad de Ciencias, IguaÂ 4225, CP 11400, Montevideo, Uruguay, Ͻ[email protected]Ͼ and 2Museu de CieÃncias Naturais, FundacËaÄo ZoobotaÃnica do Rio Grande do Sul. Av. Dr. Salvador FrancËa 1427, Jardim BotaÃnico, CEP 90690-000, Porto Alegre, RS, Brasil

INTRODUCTION is related to the ®nal withdrawal of the sea, and the upper part of INTHEIRmonograph Review of the Pelycosauria, Romer and this section is characterized by continental and strong ¯uvial in- Price (1940), proposed that the earliest synapsids (``pelycosaurs'') ¯uences. The fossils described in this paper were recovered from were cosmopolitan, despite the observation that amniotes ap- the conglomerates in the middle part of the Buena Vista Forma- peared to be restricted to the paleotropics during the Late Car- tion. boniferous and Early Permian (290±282 Ma). Romer and Price Institutional abbreviation: FC-DPV, ColeccioÂn de Vertebrados (1940) accounted for the scarcity of terrestrial tetrapods, including FoÂsiles, Departamento de PaleontologõÂa, Facultad de Ciencias, ``pelycosaurs,'' in Lower Permian beds elsewhere to the absence Universidad de la RepuÂblica, Montevideo, Uruguay. of coeval continental deposits beyond North America and Europe. DESCRIPTION Indeed, most workers recognized a geographical and temporal gap The fossil material was found during recent ®eldwork in Cerro between Permo-Carboniferous ``pelycosaurs'' and therapsid syn- Largo County, northeastern Uruguay (Fig. 1), and consists of dis- apsids. Recent research has con®rmed that varanopid and caseid articulated postcranial elements, including small dorsal and caudal ``pelycosaurs'' were components of therapsid-dominated Late vertebrae and fragmentary appendicular elements. Measurements Permian faunas preserved in Russia and South-Africa (Tatarinov of the dorsal vertebrae are listed in Table 1. and Eremina, 1975; Reisz, 1986; Reisz et al., 1998; Reisz and FC-DPV 1182, 1183, 1189, 1194, 1199, 1200 and 1333, com- Berman, 2001). prising dorsal and caudal vertebrae (Figs. 3, 4.1±4.3), are char- In this note, we report the ®rst record of basal synapsids for acterized by the following morphology: vertebrae amphicoelous South America (Fig. 1) and discuss its paleogeographic and tem- and notochordal, with a short centrum; notochordal pit above the poral implications. These remains come from outcrops assigned central point of the subcircular rim; vertebral body excavated in to the Buena Vista Formation (Goso et al., 2001; Fig. 2), which lateral view, with a bevelled ventral surface for the intercentrum, ®lled part of the ParanaÂ Basin during Late Permian and probably neural arch completely fused to the centrum, lacking a suture; Early Triassic (Bossi and Navarro, 1991). The ParanaÁ Basin is moderate development of a ventral keel in the anterior presacral one of the major sedimentary basins of Gondwana, covering an vertebrae; neural spine laterally compressed and anteroposteriorly area of nearly 1.5 million km2, in Brazil, Argentina, Paraguay and short with a slightly excavated base; postzygapophyses little ex- Uruguay (FrancËa et al., 1995; ZalaÂn et al., 1990). At the beginning panded beyond the lateral margin of the centrum and tilted antero- of basin development the Devonian marine and brackish condi- ventrally and transverse process short. Although some of these tions prevailed, but subsequently, during the Late Permian, there was a gradual change to continental sedimentation and the de- position of fossil-bearing red beds. Buena Vista Formation is characterized by reddish ®ne sandstone interbedded with lentic- ular clay layers and intraformational conglomerates (Fig. 2), (Bos- si and Navarro, 1991; Goso et al., 2001). The paleoenvironment

FIGURE 1ÐMap of the study area. The asterisk shows the geographic FIGURE 2ÐStratigraphic section of the Buena Vista Formation (Cerro position of the Buena Vista Formation outcroups in South America. Largo County, Uruguay). 389 390 JOURNAL OF PALEONTOLOGY, V. 77, NO. 2, 2003

TABLE 1ÐMeasurements (in mm) of dorsal and caudal vertebrae. Spine meet the neural arch and the neural spine, although incomplete, height and total height could not be measured because the neural spines are not complete. L: length; H: height. is clearly a blade-like structure. In the morphology of the centrum and neural arch, FC-DPV 1194 resembles more the vertebrae of Specimen # Region Centrum L Centrum H sphenacodontid synapsids than those of varanopids. FC-DPV 1182 Dorsal 10.5 6.2 FC-DPV 1183 Dorsal 11.5 7.5 DISCUSSION FC-DPV 1194 Dorsal 10.9 7.2 The only con®rmed basal synapsid from the Southern Hemi- FC-DPV 1199 Dorsal 11.4 6.4 FC-DPV 1200 Dorsal 11.0 7.5 sphere is Elliotsmithia longiceps from the Upper Permian Tapi- FC-DPV 1189 Caudal 10.2 6.2 nocephalus Assemblage Zone of South Africa (Reisz et al., 1998). FC-DPV 1333 Caudal 9.4 4.8 Elliotsmithia was identi®ed by these authors as a varanopid FC-DPV 1333 Caudal 9.5 4.8 (ϭ``varanopseid'') and basal member of a clade including Aero- saurus, Varanops and Varanodon (``Varanodontinae'' of Reisz and Berman, 2001), but study of a second, slightly more complete characters are present in basal reptiles, this combination of fea- specimen (Modesto et al., 2001), suggests that Elliotsmithia forms tures is found only in basal synapsids of the group Varanopidae a clade with Mycterosaurus and Mesenosaurus (``Mycterosauri- (ϭ``Varanopseidae''). It is unlikely that the material is attributable nae'' of Reisz and Berman, 2001). Elliotsmithia and Mesenosau- to captorhinids and parareptiles because these reptiles have trans- rus (from Eastern Europe) are the only known Late Permian var- versely expanded neural arches (Laurin and Reisz, 1995). Am- anopids; all other taxa are restricted to the Lower Permian of phicoelous and notochordal vertebrae are present also in early North America. diapsids (Carroll and Currie, 1991), but no Paleozoic diapsid pre- The morphology of the Uruguayan vertebrae identi®es a var- serves the suite of vertebral features described here (Reisz, 1981; anopid ``pelycosaur'' with close af®nities to Mycterosaurus lon- Currie, 1981). giceps (Berman and Reisz, 1982; Reisz et al., 1997) and Mesen- FC-DPV 1194 (Figs. 4.4, 5), an anterior dorsal vertebrae, osaurus romeri (Reisz, personal commun., 2000). The close re- shows a deep, longitudinally excavated centrum, stronger ventral semblance of the Uruguayan varanopid to these taxa may re¯ect keel and more deeply excavated neural arch base than the other the retention of primitive characters of the group but our material vertebrae. The transverse processes are broken just where they reveals that varanopids ranged more widely in Gondwana than

FIGURE 3ÐDrawings of varanopid presacral vertebrae from the Buena Vista Formation, Late Permian-?Early Triassic, Cerro Largo County, Uruguay. In lateral (1, 4), anterior (2, 5) and posterior (3, 6) views. (1±3 from specimen FC-DPV 1183; 4±6 from specimen FC-DPV 1200). Scale bars ϭ 1 cm. PALEONTOLOGICAL NOTES 391

FIGURE 4ÐVertebrae from the Buena Vista Formation, Late Permian-?Early Triassic, Cerro Largo County, Uruguay. 1, FC-DPV 1183 in lateral view (scale bar ϭ 5 mm); 2, 3, FC-DPV 1200 in anterior and lateral views (scale bar ϭ 10 mm); 4, FC-DPV 1194 in lateral view (scale bar ϭ 5 mm). suggested by recent studies of South A®can varanopids (Reisz et Formation. To date these are restricted to an incomplete skull of al., 1998; Modesto et al., 2001). In addition, the presence of a basal temnospondyl closely allied to the Dvinosaurus-Tupilako- varanopid ``pelycosaur'' in South America corroborates previous sauridae clade (Marsicano et al., 2000), which suggests a Late ideas of close af®nities of Permian insects and crustaceans from Permian or perhaps Early Triassic age. Other fossil remains from both Uruguay and Brazil to those from Eastern Europe and North the Buena Vista Formation include several specimens of other American (Pinto, 1972; Pinto et al., 2000). The evidence from temnospondyl amphibians and procolophonoid reptiles (Goso et insects, crustaceans, and now varanopid synapsids is indicative of al., 2001). Some of the temnospondyl remains share diagnostic extensive faunal exchange between these regions towards the characters with Capitosauridae, but retain a primitive pattern more close of the Paleozoic. typical of Permian temnospondyls. One of the procolophonoids Most of our material is referable to Varanopidae, but a single is very similar to Coletta seca (Modesto et al., in press) and Con- specimen FC-DPV 1194 (Figs. 4.4, 5), appears to represent a tritosaurus simus (Ivachnenko, 1974) from the Lower Triassic of second basal synapsid taxon. The stronger ventral keel and more South Africa and Russia, respectively, in that the Uruguayan form deeply excavated neural arch of FC-DPV 1194 (with respect to also displays a mixture of features from Permian and Triassic the other vertebrae) are suggestive of a sphenacodontid af®nitty representatives of the group. So, due to the lack of index fossils (Romer and Price, 1940; Reisz, 1986; Sumida and Berman 1993). and the ambiguous clues that the assemblage gives about the age, This is an exciting possibility, but more complete material is need- it remains highly dif®cult to assign either a Late Permian age or ed to investigate the identity of FC-DPV 1194. a Lower Triassic one to the Buena Vista Formation. A third pos- Finally, the age of the Buena Vista Formation was based on sibility is that the formation preserves a transitional fauna that correlation with Brazilian lithostratigraphic units (Rio do Rasto includes the last ``pelycosaurs.'' Alternatively, because there is no and Sanga do Cabral Formations), which have been considered other evidence of ``pelycosaurs'' in strata younger the Upper as Upper Permian and Permo-Triassic by most authors (Andreis Permian Tapinocephalus Assemblage Zone of South Africa, the et al., 1980; Acevedo et al., 1985; Barberena et al., 1985; Faccini, presence of basal synapsids in the Buena Vista Formation could 1989) and Sanga do Cabral Formation was recently assigned to indicates that this unit was deposited initially during the Late the Lower Triassic (Cisneros and Schultz, in press). However, this Permian. However, if the Lower Triassic age of the Buena Vista correlation was based in lithological similarities rather than bio- Formation suggested by Bossi and Navarro (1991) can be sub- stratigraphic correlation, owing to the scarcity of fossils of bio- stantiated, the presence of ``pelycosaurs'' could be explained by stratigraphic utility that have been collected from the Buena Vista two alternative hypotheses: 1) Taking into account the geological setting discussed above, the ``pelycosaur'' remains could represent Upper Permian materials that have been reworked by the currents that generated the intraconglomerates; 2) the ``pelycosaurs'' were contemporaneous with the putatively Lower Triassic members of the fauna. The former hypothesis is plausible, given the damage to the extremities of the known elements. The latter hypothesis is equally plausible, yet it implies that the Uruguayan ``pelycosaurs'' are the youngest known representatives of an oth- erwise Paleozoic assemblage, greatly extending their temporal range into the beginning of the Triassic.

ACKNOWLEDGMENTS We thank R. R. Reisz and S. P. Modesto for valuable comments on material identi®cation and for bibliography. A. Mones provid- ed additional help with the bibliography. We specially thank N. Scott and S. P. Modesto for language revision. A. Rojas and N. Lorenzo for assisting us in the ®eld work, and Christian Sidor and an anonymous reviewer for their comments on the manu- script. This research was supported by Pedeciba-BiologõÂa and FIGURE 5ÐDrawing of FC-DPV 1194, a dorsal vertebra from the Buena CSIC-Uruguay and the government of Cerro Largo County. Vista Formation, Late Permian-?Early Triassic, Cerro Largo County, Uruguay. In lateral (1), anterior (2), and posterior (3) views, showing REFERENCES the excavated neural arch base and strong ventral keel. Scale bar ϭ 1 ACEVEDO, S. A., E. LAVINA,M.BARBERENA,L.FERRANDO, AND R. cm. ANDREIS. 1985. Evidencias de correlacËaÄo entre a FormacËaÄo Yaguari 392 JOURNAL OF PALEONTOLOGY, V. 77, NO. 2, 2003

(Uruguai) e as FormacËoÄes Rio do Rasto e Sanga do Cabral (Rio Grande amphibian from the Lower Triassic of South America. Alcheringa, 24: do Sul- Brasil). Pesquisas, 17:112±121. 119±123. ANDREIS, R. R., G. BOSSI, AND D. MONTARDO. 1980. O Grupo RosaÂrio MODESTO, S. P., C. A. SIDOR,B.S.RUBIDGE, AND J. WELMAN. 2001. A do Sul (TriaÂsico) no Rio Grande do Sul. Annales do XXXI Congreso second varanopseid skull from the Upper Permian of South Africa: Brasileiro de GeologõÂa, 2:659±673. Implications for Late Permian ``pelycosaur'' evolution. Lethaia, 34: BARBERENA, M. C., D. ARAUÂ JO, AND E. LAVINA. 1985. Late Permian 249±259. and Triassic tetrapods of Southern Brazil. National Geographic Re- MODESTO, S. P., R. J. DAMIANI, AND H.-D. SUES. In press. A reappraisal search, 1:5±20. of Coletta seca, a basal procolophonoid reptile from the Lower Triassic BERMAN, D., AND R. REISZ. 1982. Restudy of Mycterosaurus longiceps of South Africa. Palaeontology. (Reptilia, Pelycosauria) from the Lower Permian of Texas. Annals of PINTO, I. D. 1972. Late Paleozoic insects and crustaceans from ParanaÂ Carnegie Museum, 51:423±453. Basin and their bearing on chronology and continental drift. Annales da AcadeÂmia Brasileira de CieÃncias, 44:247±254. BOSSI, J., AND R. NAVARRO. 1991. GeologõÂa del Uruguay. Departamento PINTO, I. D., G. PINÄ EIRO, AND M. VERDE. 2000. First Permian insects de Publicaciones, Universidad de la RepuÂblica, Montevideo, 996 p. from Uruguay. Pesquisas, 27:89±96. CARROLL, R. L., AND P. J. CURRIE. 1991. The early radiation of Diapsid REISZ, R. R. 1981. A diapsid reptile from the Pennsylvanian of Kansas. reptiles, p. 354±424. In H. P. Schultze and L. Trueb (eds.), Origins of Museum of Natural History, University of Kansas Special Publications, the Higher Groups of Tetrapods: Controversy and Concensus. Cornell 7:1±74. University Press, Ithaca, New York. REISZ, R. R. 1986. Pelycosauria. In P. Wellnhofer (ed.), Handbuch der CISNEROS, J. C., AND C. L. SCHULTZ. In press. Procolophon brasiliensis, PalaÈeoherpetologie, Pt. 17A. Gustav Fisher Verlag, Stuttgart, 102 p. n. sp., a new procolophonoid reptile from the Lower Triassic of South- REISZ,R.R.,AND D. BERMAN. 2001. The skull of Mesenosaurus romeri, ern Brazil. Neues Jahrbuckh fuÈr Geologie und PalaÈontologie, Monat- a small varanopseid (Synapsida, Eupeycosauria) from the Upper Perm- shefte. ian of the Mesen River Basin, northern Russia. Annals of Carnegie CURRIE, P. J. 1981. The vertebrae of Youngina (Reptilia: Eosuchia). Ca- Museum, 70:113±132. nadian Journal of Earth Science, 18:815±818. REISZ, R. R., D. DILKES, AND D. BERMAN. 1998. Anatomy and relation- FACCINI, U. 1989. O Permo-TriaÂssico do Rio Grande do Sul. Uma anaÂlise ships of Elliotsmithia longiceps Broom, a small synapsid (Eupelyco- sob o ponto de vista de Sequencias Depositacionais. Unpublished MSc. sauria: Varanopseidae) from the Late Permian of South Africa. Journal dissertation. Universidade Federal do Rio Grande do Sul. Porto Alegre, of Vertebrate Paleontology, 18:602±611. Brasil, 113 p. REISZ, R. R., H. WILSON, AND D. SCOTT. 1997. Varanopseid synapsid FRANCËA, A., E. MILANI,R.SCHNEIDER,P.LOÂ PEZ,M.LOÂ PEZ,S.SUAÂ REZ, skeletal elements from Richards Spur, a Lower Permian ®ssure ®ll near Fort Sill, Oklahoma. Oklahoma Geology Notes, 57(5):160±170. H. SANTA ANA,F.WIENS,O.FERREIRO,E.ROSSELLO,H.BIANUCCI, ROMER, A. S., AND L. I. PRICE. 1940. Review of the Pelycosauria. Special R. FLORES,M.VISTALLI,F.FERNAÂ NDEZ SEVESO,R.FUENZALIDA, AND Paper Geological Society of America, 28:1±538. N. MUNÄ OZ. 1995. Phanerozoic correlation in Southern South America, SUMIDA,S.S.,AND D. S. BERMAN. 1993. The pelycosaurian (Amniota, p. 129±161. In A. J. Tankard, R. SuaÂrez, and H. Welsink (eds.), Petro- Synapsida) assemblage from the Late Pennsylvanian Sangre de Cristo leum Basins of South America. Memoir 62. Formation of Central Colorado. Annals of Carnegie Museum, 62:293± GOSO, C., G. PINÄ EIRO,H.DE SANTA ANA,A.ROJAS,M.VERDE, AND 310. C. ALVES. 2001. CaracterizacioÂn estratigraÂ®ca de los depoÂsitos conti- TATARINOV,L.P.,AND I. V. EREMINA. 1975. New data on the morphol- nentales cuspidales neopeÂrmicos (Formaciones YaguarõÂ y Buena Vista) ogy of Ennatosaurus (Pelycosauria, Caseomorpha) from the Lower Ta- en el borde oriental de la Cuenca Norte Uruguaya. XI Congreso Lati- tarian deposits on the Pinega river. Paleontological Journal, 9(4):510± noamericano de GeologõÂa, III Congreso Uruguayo de GeologõÂa. (CD- 514. ROM). ZALAÂ N, P. V., S. WOLF,J.C.J.CONCEICËAÄ O,A.MARQUES,M.A.AS- IVACHNENKO, M. 1974. New data on the Early Triassic procolophonids TOLFI,I.S.VIEIRA,V.T.APPI, AND A. ZANOTTO. 1990. Bacia do of the USSR. Paleontological Journal, 8(3):346±351. ParanaÂ, p. 135±168. In G. P. De Raya Gabaglia and E. J. Milani (eds.), LAURIN, M., AND R. REISZ. 1995. A revaluation of early amniote phy- Origem e evolucËaÄo das Bacias Sedimentares, Rio de Janeiro (Petro- logeny. Zoological Journal of the Linnean Society, 113:165±223. braÂs). MARSICANO, C., D. PEREA, AND M. UBILLA. 2000. A new temnospondyl ACCEPTED 19 APRIL 2002