Naturwissenschaften DOI 10.

1007/s00114-011-0858-0

ORIGINAL PAPER

New stem-sauropodomorph (Dinosauria, Saurischia) from the Triassic of Brazil

Sergio F. Cabreira & Cesar L. Schultz & Jonathas S. Bittencourt & Marina B. Soares & Daniel C. Fortier & Lcio R. Silva & Max C. Langer

Received: 22 August 2011 / Revised: 7 October 2011 / Accepted: 13 October 2011 # Springer-Verlag 2011

Abstract Post-Triassic theropod, sauropodomorph, and ornithischian dinosaurs are readily recognized based on the set of traits that typically characterize each of these groups. On the contrary, most of the early members of those lineages lack such specializations, but share a range of generalized traits also seen in more basal dinosauromorphs. Here, we report on a new Late Triassic dinosaur from the Santa Maria Formation of Rio Grande do Sul, southern Brazil. The specimen comprises the disarticulated partial

skeleton of a single individual, including most of the skull bones. Based on four phylogenetic analyses, the new dinosaur fits consistently on the sauropodomorph stem, but lacks several typical features of sauropodomorphs, showing dinosaur plesiomorphies together with some neotheropod traits. This is not an exception among basal dinosaurs, the early radiation of which is characterized by a mosaic pattern of character acquisition, resulting in the uncertain phylogenetic placement of various early members of the group. Keywords Dinosauria, Late Triassic . Santa Maria Formation . Archosauria . Phylogeny

Communicated by: Robert Reisz Electronic supplementary material The online version of this article (doi:10.1007/s00114-011-0858-0) contains supplementary material, which is available to authorized users. S. F. Cabreira : L. R. Silva Museu de Cincias Naturais, Universidade Luterana do Brasil (ULBRA), Av. Farroupilha 8001, 92425-900 Canoas, Rio Grande do Sul, Brazil C. L. Schultz : M. B. Soares Departamento de Paleontologia e Estratigrafia, Instituto de Geocincias, Universidade Federal do Rio Grande do Sul (UFRGS), Av. Bento Gonalves 9500, 91540-000 Porto Alegre, Rio Grande do Sul, Brazil J. S. Bittencourt : M. C. Langer (*) Laboratrio de Paleontologia de Ribeiro Preto, FFCLRP, Universidade de So Paulo, Av. Bandeirantes 3900, 14040-901 Ribeiro Preto, So Paulo, Brazil e-mail: mclanger@ffclrp.usp.br D. C. Fortier Departamento de Geologia, UFMG, Av. Antnio Carlos 6627, 31270-901 Belo Horizonte, Minas Gerais, Brazil

Introduction Previous to the discovery of Saturnalia tupiniquim, from the Santa Maria Formation of southern Brazil (Langer et al. 1999), the oldest known members of the sauropodomomorph lineage were relatively massive, medium to large sized dinosaurs, recovered from Norian aged strata, mainly in Europe, Argentina, and South Africa (Galton 1990). Since then, an unsuspected diversity of smaller, more gracile stem-sauropodomorphs (sensu Langer 2003) was recovered from Carnian deposits of South America, especially from the Ischigualasto Formation, in Argentina. This includes Panphagia protos (Martinez and Alcober 2009), Chromogisaurus novasi (Ezcurra 2010), and possibly also Eoraptor lunensis (Martinez et al. 2011), formerly nested within Theropoda (Sereno et al. 1993). Here, we report on a disarticulated but otherwise well-preserved skeleton of a new stem-sauropodomorph from the Late Triassic of southern Brazil (Fig. 1), including most of the skull bones (Fig. 2).

Naturwissenschaften

Fig. 1 Pampadromaeus barberenai, locality map and skeletal silhouette, showing preserved elements of the holotype (ULBRA-PVT016)

Systematic palaeontology Dinosauria Owen 1842 sensu Padian and May (1993) Saurischia Seeley 1887 sensu Gauthier (1986) Eusaurischia Padian, Hutchinson & Holtz 1999 sensu Langer (2004) stem-Sauropodomorpha Huene 1932 sensu Salgado et al. (1997; see Langer 2003) Pampadromaeus barberenai new genus and species Etymology The generic name is derived from the Quechua word pampa (=plain), in reference to the grassland landscape that covers parts of Rio Grande do Sul, and dromaeus, variant of Greek (=runner), in reference to the probable cursoriality of the animal. The specific epithet honors the Brazilian palaeontologist Mrio C. Barberena. Holotype ULBRA-PVT016 (Museu de Cincias Naturais, Universidade Luterana do Brasil, Canoas), disarticulated partial skeleton preserved in a single mudstone block (Figs. 1 and 2), including most skull bones, parts of the mandible, incomplete vertebral column, and apendicular skeleton (see details of the preserved elements in the Electronic Supplementary Material). Some of the elements were extracted, but most were only superficially exposed, retaining their original position in the block. Type locality and horizon Reddish mudstones of the Alemoa Member, Santa Maria Formation, Rosrio do Sul

Group (Zerfass et al. 2003), exposed at the site known as Janner or Vrzea do Agudo (531734.20 W, 29 39 10.89 S), located about 2 km west of the town of Agudo, Rio Grande do Sul, Brazil (Fig. 1). Based on the cooccurrence of index fossils such as Exaeretodon and Hyperodapedon, the site can be assigned to the upper portions of the Hyperodapedon Assemblage Zone (Langer et al. 2007). Correlation to the better constrained Ischigualasto Formation, northwestern Argentina (Martinez et al. 2011), suggests a late Carnian age; ca. 230228 Ma. Diagnosis It is distinguished from all other well-known Triassic dinosauromorphs by the following unique combination of character states (see extended diagnosis in the Electronic Supplementary Material): head longer than two thirds of the femoral length, premaxilla with short subnarial process, concave ventral margin of the premaxillamaxilla articulation, no inset of first premaxillary or dentary tooth, premaxilla and dentary with unserrated mesial-most tooth crowns, most teeth lanceolate with coarse denticles along the carinae, sacral vertebrae with dorsoventrally expanded ribs, femur with reduced medial tuberosity (t in Novas 1996; Fig. 3) in the head and well-developed trochanteric shelf, epipodium significantly longer than femur. Description The premaxilla has a shallow narial fossa, four alveoli, and a subnarial process that does not reach the nasal. On the

Naturwissenschaften

Fig. 2 Pampadromaeus barberenai, ULBRA-PVT016 (holotype), skull elements reversed as needed to left lateral (total skull length slightly over 10 cm). a Right premaxilla and maxilla. b Right frontal in ventral view. c Right jugal. d Right prefrontal. e Right postorbital and left squamosal. f Left parietal in ventral view. g Left pterygoid in ventral view. h Left dentary. i Right surangular, angular, articular, and

prearticular in medial view. a Angular; aof antorbital fossa; ar artigular; d dentary; emf external mandibular fenestra; en external naris; f frontal; l lacrimal; m maxilla; n nasal; nf narial fossa; p parietal; pa prearticular; pf prefrontal; pm premaxilla; pmd promaxillary depression; po postorbital; q quadrate; sa surangular; sq squamosal

medial side, the maxillary process of the premaxilla forms a horizontal shelf below the palatal process of the maxilla. The dorsal and rostral rami of the maxilla meet forming a low, but well-defined angle at the rostral margin of the bone. Reconstruction of the premaxillamaxilla articulation suggests that the former is ventrally projected. This is reminiscent of the subnarial gap of theropods, in the sense that the alveolar margin is bent upwards (Rauhut 2003), but no diastema is evident. The antorbital fossa has a deep rostral depression in a position equivalent to that of the promaxillary fenestra/foramen described for Zupaysaurus rougieri (Ezcurra 2007), Syntarsus kayentakatae (Tykoski 1995), and cf. Dracovenator regenti (Yates 2005). This leads to a rostral furrow that apparently opens onto the medial surface of the bone, but no aperture is laterally visible. There are about 20 maxillary tooth positions and six large foramina around the antorbital fossa, the latter of which opens caudally. The lacrimal has a long, nearly vertical ventral ramus, the ventral portion of which is excavated by the antorbital fossa. The prefrontal lacks a

sheet of bone covering the lacrimal, which is present in various basal sauropodmorphs, but not in P. protos (Martinez and Alcober 2009). The postorbital forms a pendant eminence over the dorsocaudal border of the orbit, as seen in other basal dinosaurs (Sereno and Novas 1994; Martinez et al. 2011). The squamosal has a strap-shaped ventral process, as typical of early sauropodomorphs (Yates 2003). The caudal ramus of the jugal is forked and not ventrally arched. A medial ridge on the ventral surface of the pterygoid holds a row of about 15 positions for small rudimentary teeth, as previously recognized only in E. lunensis and Eodromaeus murphi among dinosaurs (Martinez et al. 2011; Nesbitt 2011). The dentary bears about 20 tooth positions, extending from the rostral tip of the bone. Medial to the glenoid, the articular is pierced by the foramen for the chorda tympani, as in Silesaurus opolensis (MCL, personal observation, 2007), some dinosaurs (Yates 2005; Nesbitt 2011), and various pesudosuchians (Gower 1999). Most tooth crowns in the central portion of the upper and lower rows are

Naturwissenschaften Fig. 3 Simplified phylogenetic relations of Pampadromaeus barberenai among basal dinosaurs, based on the analyses of the cited matrices (see Electronic Supplementary Material). Dotted lines and underlined lettering indicate nonEusaurischia saurischian grade; gray lines and lettering indicate sauropodomorph lineage; full black lines and lettering indicate theropod lineage. Sauropodomorpha applied sensu Langer (2003). D Dinosauria, E Eusaurischia

lanceolate, with an expanded base, a convex mesial margin, and a sigmoidal (convex at the base and concave apically) or nearly straight distal margin. More distal teeth are shorter, but retain this general shape that is typical of various basal dinosaurs (Barrett et al. 2011). There is some imbrication between adjacent crowns, with the mesial edge of the crowns medially overlapping the distal edge of the tooth mesial to it. Carinae are coarsely serrated, their denticles forming oblique angles with the crown margin. The most mesial teeth on both premaxilla and dentary are narrower and bear less evident serrations, as seen in coelophysid theropods (Tykoski and Rowe 2004). Mesial dentary teeth of other stem-sauropodomorphs such as S. tupiniquim (Langer et al. 1999) and P. protos (Martinez et al. 2011) are broader and bear serrated carinae, but the preserved elements may not belong to the very tip of the jaw. The atlantal neurapophysis has a plate-like, dorsally expanding epipophysis that does not extend caudal to the postzygapophysis. Cervical ribs have elongated and straight shafts. As typical of saurischians (Langer and Benton 2006), trunk vertebrae have well developed laminations surrounding the diapophysis, and hyposphenehypantrum auxiliary articulations. As in many basal dinosaurs (Langer 2003; Langer et al. 2011), two vertebrae form the entire sacrum, with ribs attached to the cranial half of the respective centrum. The ribs are dorsoventraly expanded and ventrally connected to one another. The circumscribed intercostal space has a small ventral aperture, medial to the caudal half of the first centrum, and a large dorsal opening bordered by unexpanded transverse processes. The scapular blade is cranially inclined, and the humerus has a broad

distal end, as typical of sauropodomorphs (Langer and Benton 2006). The ilium has a deep dorsal lamina, but the preacetabular ala is short and pointed. The long postacetabular ala bears an evident brevis fossa. The acetabular wall has a nearly straight ventral margin, as seen in S. tupiniquim, Guaibasaurus candelariensis, and P. protos. The femoral head is offset and inturned, with the long axis forming an angle of 40 to the intercondylar line. It has a flat craniolateral surface and subtle medial tubera (Nesbitt et al. 2009). The articular surface has well-developed transverse groove and facies articularis antitrochanterica (Langer 2003). The lateral surface of the proximal portion of the femur has a crescent-shaped dorsolateral trochanter, a knob-like lesser trochanter, and a well-developed trochanteric shelf. The fourth trochanter expands as an asymmetrical, sharp flange. The epipodium is nearly 20% longer than the femur. The cnemial crest of the tibia is cranially projected and the lateral condyle is set at the center of the lateral margin of the proximal articulation. Metatarsal I is significantly shorter than metatarsals II and III.

Phylogenetic analyses and discussion Phylogenetic analyses (see details in the Electronic Supplementary Material) based on four recent studies on the early radiation of dinosaurs (Martinez and Alcober 2009; Ezcurra 2010; Nesbitt et al. 2010; Martinez et al. 2011) unambiguously place P. barberenai on the sauropodomorph stem (Fig. 3), with minor variations that include its sistertaxon relationship to either P. protos, Sauropodomorpha (sensu Langer 2003), or Sauropodomorpha plus S. tupini-

Naturwissenschaften

quim. However, despite this recurring position, the affinities of P. barberenai to sauropodomorphs are not supported by strong tree statistics, neither by many or uncontroversial anatomical traits. In conjunct, these include the following: (1) highly homoplastic characters (mean Consistency Index based on the four analyses indicated under brackets) as lanceolate tooth crowns (0.53) with expanded bases (0.30) that overlap one another (0.26), four premaxillary teeth (0.47), interpostzygapophyseal notch in proximal tail vertebrae (0.20), humerus longer than or subequal to 0.6 of the femoral length (0.43), supracetabular crest extensive to the pubic peduncle of the ilium (0.28), proximal femoral articulation with straight transverse groove (0.28), and proximal portion of the tibia with a cranially displaced fibular condyle (0.33); (2) characters codified as missing data for most fossil taxa, e.g., medial foramen on the articular (Nesbitt 2011); and (3) features with a seemingly more consistent distribution, such as strap-shaped ventral process of the squamosal, proximal tail vertebrae with long neural spines, humerus with broad distal end (relative to the length of the bone), and ilium with a subtriangular preacetabular ala and long pubic peduncle. However, P. barberenai lacks some typical sauropodomorph traits such as a small head and longer crowns on the rostral part of the tooth series (Martinez and Alcober 2009). The analyses also revealed numerous autapomorphies for P. barberenai, but most of them correspond to homoplastic traits more common to other dinosauromorph groups, and none is really unique to that taxon. Theropod features (Rauhut 2003; Tykoski 1995) include concave ventral margin of the premaxillamaxilla articulation, unserrated teeth in the rostral tip of the upper and lower jaws, promaxillary depression, and deep dorsal iliac lamina. Other putative theropod traits of P. barberenai, such as a forked caudal process of premaxilla (Rauhut 2003) and a concave caudal margin of the iliac lamina in dorsal profile (Tykoski 1995), are also seen in sauropodomorphs and S. opolensis. Likewise, dinosaur plesiomorphies (Langer and Benton 2006) retained or reverted in P. barberenai include a deep acetabular medial wall with straight ventral margin and narrow sacral transverse processes that do not roof the intercostal space. This seems also to be the case of the palatal teeth, known in early members of the archosaurs lineage (Nesbitt 2011), putative basal pseudosuchians (Wu and Russell 2001), some pterosaurs (Kellner 2003), and basal dinosauromorphs such as Lewisuchus admixtus (JSB, personal observation, 2011), although hitherto unrecognized in dinosaurs other than E. lunensis and E. murphi. The outcomes of the phylogenetic analyses (Fig. 3) stress the uncertainties concerning the relationships of various basal dinosaurs. For example, G. candelariensis, E. lunensis, and herrerasaurs appear alternatively basal to the sauropodomorph-theropod dichotomy or at the base of

either lineage. This most likely results from the ambiguous distribution of traits, which became characteristic of certain dinosaur clades, among these and other basal dinosauromorphs such as S. opolensis, S. tupiniquim, P. protos, E. murphi, and Eocursor parvus. Phylogenetic reconstructions of basal dinosaurs are further problematical because those features accumulated over a relatively short period of time (Irmis 2011), in forms with recurring body structure and inferred habits (i.e., medium-sized, bipedal, and omnivorous). In fact, the early dinosaur radiation may represent a segment of evolutionary history that is particularly hard to reconstruct, and its poor constraint is not surprising. Only additional finds and, especially, more detailed phylogenetic studies will facilitate development of a more stable evolutionary framework for the placement of many basal dinosaurs.
Acknowledgments Aspects of this work were funded by Conselho Nacional de Desenvolvimento Cientfico e Tecnolgico, Universidade Luterna do Brasil (to S.F.C), and Fundao de Amparo Pesquisa do Estado de So Paulo (to M.C.L. and J.S.B.). This is contribution no. 25 of Laboratrio de Paleontologia, FFCLRP-USP. Thanks to Luiz Flvio Lopes for the photographic work, to Mara Massarani for the drawing of Fig. 2, and to three anonymous reviewers for their comments.

References
Barrett PM, Butler RJ, Nesbitt SJ (2011) The roles of herbivory and omnivory in early dinosaur evolution. Earth Environ Sci Trans R Soc Edinburgh 101:383396. doi:10.1017/S1755691011020111 Ezcurra MD (2007) The cranial anatomy of the coelophysoid theropod Zupaysaurus rougieri (Upper Triassic, Argentina). Historical Biol 19:185202. doi:10.1080/08912960600861467 Ezcurra MD (2010) A new early dinosaur (Saurischia: Sauropodomorpha) from the Late Triassic of Argentina: a reassessment of dinosaur origin and phylogeny. J Syst Palaeontol 8:371425. doi:10.1080/14772019.2010.484650 Galton PM (1990) Basal SauropodomorpphaProsauropoda. In: Weishampel DB, Dodson P, Osmlska H (eds) The Dinosauria. University of California Press, Berkeley, pp 320344 Gauthier J (1986) Saurischian monophyly and the origin of birds. In: Padian K (ed) The origin of birds and the evolution of flight. Memoirs of the California Academy of Sciences, no. 8, pp 155 Gower DJ (1999) The cranial and mandibular osteology of a new rauisuchian archosaur from the Middle Triassic of southern Germany. Stuttgarter Beitr Naturkde B 280:149 Irmis RB (2011) Evaluating hypotheses for the early diversification of dinosaurs. Earth Environ Sci Trans R Soc Edinburgh 101:397 426. doi:10.1017/S1755691011020068 Kellner AWA (2003) Pterosaur phylogeny and comments on the evolutionary history of the group. In: Buffetaut E, Mazin J-M (eds) Evolution and palaeobiology of pterosaurs. Geological Society, London, pp 105137 Langer MC (2003) The sacral and pelvic anatomy of the stemsauropodomorph Saturnalia tupiniquim (Late Triassic, Brazil). PaleoBios 23(2):140 Langer MC (2004) Basal Saurischia. In: Weishampel DB, Dodson P, Osmlska H (eds) The Dinosauria, 2nd edn. University of California Press, Berkeley, pp 2546

Naturwissenschaften Langer MC, Benton MJ (2006) Early dinosaurs: a phylogenetic study. J Syst Palaeontol 4:309358. doi:10.1017/S1477201906001970 Langer MC, Abdala F, Richter M, Benton MJ (1999) A sauropodomorph dinosaur from the Upper Triassic (Carnian) of souther Brazil. C R Acad Sci IIA 329:511517 Langer MC, Ribeiro AM, Schultz CL, Ferigolo J (2007) The continental tetrapod-bearing Triassic of south Brazil. In: Lucas SG, Spielman JA (eds) The global Triassic. New Mexico Mus Nat Hist & Sci Bull 41:201218. Available at http://econtent. unm.edu/u?/bulletins,298 Langer MC, Bittencourt JS, Schultz CL (2011) A reassessment of the basal dinosaur Guaibasaurus candelariensis, from the Late Triassic Caturrita Formation of south Brazil. Earth Environ Sci Trans R Soc Edinburgh 101:301332. doi:10.1017/ S175569101102007X Martinez RN, Alcober AO (2009) A basal sauropodomorph (Dinosauria: Saurischia) from the Ischigualasto Formation (Triassic, Carnian) and the early evolution of Sauropodomorpha. PLoS One 4:e4397. doi:10.1371/journal.pone.0004397 Martinez RN, Sereno PC, Alcober OA, Colombi CE, Renne PR, Montaez IP, Currie BS (2011) A basal dinosaur from the dawn of the dinosaur era in southwestern Pangaea. Science 331:206 210. doi:10.1126/science.1198467 Nesbitt SJ (2011) The early evolution of archosaurs: relationships and the origin of major clades. Bull Am Mus Nat Hist 352:1292. doi:10.1206/352.1 Nesbitt SJ, Irmis RB, Parker WG, Smith ND, Turner AH, Rowe T (2009) Hindlimb osteology and distribution of basal dinosauromorphs from the Late Triassic of North America. J Vert Paleontol 29:498516. doi:10.1671/039.029.0218 Nesbitt SJ, Sidor CA, Irmis RB, Angielczyk KD, Smith RMH, Tsuji LA (2010) Ecologically distinct dinosaurian sister group shows early diversification of Ornithodira. Nature 464:9598. doi:10.1038/nature08718 Novas FE (1996) Dinosaur monophyly. J Vert Paleontol 16:723741 Padian K, May CL (1993) The earliest dinosaurs. New Mexico Mus Nat Hist & Sci Bull 3:379381 Rauhut OWM (2003) The interrelationships and evolution of basal theropod dinosaurs. Spec Pap Palaeontol 69:1213 Salgado L, Coria RA, Calvo JO (1997) Evolution of titanosaurid sauropods. I. Phylogenetic analysis based on the postcranial evidence. Ameghiniana 34:332 Sereno PC, Novas FE (1994) The skull and neck of the basal theropod Herrerasaurus ischigualastensis. J Vert Paleontol 13:451476. doi:10.1080/02724634.1994.10011525 Sereno PC, Forster CA, Rogers RR, Monetta AM (1993) Primitive dinosaur skeleton from Argentina and the early evolution of the Dinosauria. Nature 361:6466. doi:10.1038/361064a0 Tykoski R (1995) Osteology, ontogeny, and relationships of the coelophysoid theropods. The University of Texas, Austin Tykoski RS, Rowe T (2004) Ceratosauria. In: Weishampel DB, Dodson P, Osmlska H (eds) The Dinosauria, 2nd edn. University of California Press, Berkeley, pp 4770 Wu X-C, Russell A (2001) Redescription of Turfanosuchus dabanensis (Archosauriformes) and new information on its phylogenetic relationships. J Vert Paleontol 21:4050 Yates AM (2003) A new species of the primitive dinosaur, Thecodontosaurus (Saurischia: Sauropodomorpha), and its implications for the systematics of early dinosaurs. J Syst Palaeontol 1:142. doi:10.1017/S1477201903001007 Yates AM (2005) A new theropod dinosaur from the Early Jurassic of South Africa and its implication for the early evolution of theropods. Palaeontol Afr 41:105122. doi:10.1038/nature07855 Zerfass H, Lavina EL, Schultz CL, Garcia AGV, Faccini UF, Chemale F Jr (2003) Sequence stratigraphy of continental Triassic strata of southernmost Brazil: a contribution to Southwestern Gondwana palaeogeography and palaeoclimate. Sedim Geol 161:85105. doi:10.1016/S0037-0738(02)00397-4

Naturwissenschaften
Electronic Supplementary Material for:

New stem-sauropodomorph (Dinosauria, Saurischia) from the Triassic of Brazil

Sergio F. Cabreira1, Cesar L. Schultz2, Jonathas S. Bittencourt3, Marina B. Soares2, Daniel C. Fortier4, Lcio R. Silva1 and Max C. Langer3*

Departamento de Biologia. Universidade Luterana do Brasil, ULBRA. Canoas, Rio Grande do Sul, Brazil.
2

Instituto de Geocincias, Universidade Federal do Rio Grande do Sul, UFRGS, Brazil.

Laboratrio de Paleontologia, FFCLRP, Universidade de So Paulo, Ribeiro Preto-SP, 14040-901, Brazil.

4

Departamento de Geologia, UFMG, Belo Horizonte-MG, 31270-901, Brazil.

* Correspondence and requests for materials about the character list and character-taxon matrix presented below should be addressed to M.C.L. (mclanger@ffclrp.usp.br)

This file includes: 1. Details on the type-material of Pampadromaeus barberenai. 1.1. Block containing the type-material of Pampadromaeus barberenai. 1.2. Anatomical details of ULBRA-PVT016. 2. Character-state scoring for Pampadromaeus barberenai. 3. Extended differential diagnosis of Pampadromaeus barberenai. 4. Details of the parsimony analyses. 5. Character analyses 6. Additional references

1. Details on the type-material of of Pampadromaeus barberenai. 1.1. Block containing the type-material of Pampadromaeus barberenai. Most of the skeletal remains attributed to P. barberenai were preserved scattered over a single mudstone block, a photograph of which, taken during early stages of preparation (2006) is shown in Fig. S1. In addition, assorted bones were collected from the perimeter of the block, including tail vertebrae, right humerus and ulna, right ilium, tibiae, metatarsals and pedal phalanges. Based on their corresponding size and on the absence of duplicated elements, all these bones are believed to belong to a single individual. Fig. S2 shows a bone map of the block.

Fig. S1 Mudstone block containing most of the skeletal remains attributed to ULBRA-PVT016, holotype of P. barberenai (during early stages of preparation)

As a whole, the holotype of P. barberenai contains the rostral portion of the right side of the skull, with semi-articulated premaxilla, maxilla (both bearing teeth), lachrymal, and palatine, as well as isolated skull bones: right nasal, frontal, jugal, prefrontal, and

postorbital, left parietal, squamosal, and quadrate, parts of the braincase (supraoccipital and paroccipital processes), and both pterygoids (incomplete right bone); parts of both mandibuar rami, including left dentary (with teeth) and surangular, and right partial dentary (with teeth) and articulated surangular, angular, prearticular, and articular; incomplete vertebral column including right atlantal neurapophysis, various trunk and tail vertebrae, an articulated two-vertebrae sacrum, various ribs (neck and trunk) and haemal arches; right scapula, partial left scapula, right humerus and ulna; right ilium lacking the ischiadic peduncle, left ilium lacking the postacetabular ala, proximal left ischium, partial left and nearly complete right femora, proximal half of left tibia, right tibial shaft, both fibulae, various metatarsals (including complete left metatarsal I) and pedal phalanges.

Fig. S2 Bone map of the mudstone block containing ULBRA-PVT016. Abbreviations: a, right angular; an, right atlantal neural arch; ar, right articular; cr, cervical rib(s); cv, caudal vertebra(e); d, dentaries; f, right frontal; f, femora; ha, haemal arch(es); ind, indeterminate bone; j, right jugal; jb?, lower jaw bone?; la, right lacrimal; mI; left metatarsal I; left ilium; m, right maxilla; n, right nasal; p, left parietal; pa, right prearticular; pal, right palatine; pf, right prefrontal; ph, pedal phalanx; pl?, osteoderm?; pm, right premaxilla; po, right postorbital; pt, pterygoids; q, left quadrate; s, sacrum; sa, surangulars; sc, right scapula; sq, left squamosal; tr, trunk rib(s); tv, trunk vertebra(e). (l) and (r) indicate left and right sides when both bones are preserved

1.2. Anatomical details of ULBRA-PVT016. Figs S3-7 depict details of the P. barberenai anatomy (based on its type material).

Fig. S3 P. barberenai (ULBRA-PVT016). Medial view of the rostral portion of the right side of the skull, including premaxilla, maxilla, lachrymal, and palatine (above). Details of the premaxilla (below right). Details of the palatal ramus of the maxilla (below left)

Fig. S4 P. barberenai (ULBRA-PVT016). Right dentary in lateral view (above). Details of the rostral tip (below right). Details of central teeth (below left)

Fig. S5 P. barberenai (ULBRA-PVT016). Right ilium in medial view

Fig. S6 P. barberenai (ULBRA-PVT016). Proximal portion of the left femur (lateral view)

Fig. S7 P. barberenai (ULBRA-PVT016). Left pterygoid (ventral view) showing medial a row of rudimentary teeth on the palatal process.

2. Character-state scoring for Pampadromaeus barberenai. Table S1 depicts the character-state scoring for P. barberenai in the four data-matrices employed in the analyses conducted here. These data-matrices represent different views of early dinosaur evolution: Martinez and Alcober (2009) is based on a previous study by Langer and Benton (2006); Ezcurra (2009) is mostly based on the analysis of Yates (2007); Martinez et al. (2011) stands for a more complete study in preparation by Paul Sereno; Nesbitt et al. (2010) is a multi-authored contribution that was expanded, with nearly equivalent results for ornithosuchians, in Nesbitt (2011).

Table S1: Character-state scoring for Pampadromaeus barberenai. (A=1/2) Martinez and Alcober 2009 Ezcurra 2010
001?111??1 ??010??0?1 ?1010??0?? 00???00211 ?0?0?0??0? ?????????0 ?????????? 0?01000?00 10??1????? ????001001 0001111100 ?????????? ??210011?? ?0?1101?01 1?3?00???? ??00??0000 00000?1??0 ?????????? ?1100?A101 ?????????? ?????????? ???00?0111 0?1100???? ?0001012?? ?????????? ?1?1???0?? 0????????? ???????? 00??1?011? 10?1??0??? ???0000000 ?????????1 0000?0??00 1?0?1????? 11010202?? 11011100?0 ?????????? 11?1???? ??01110000 ??0??????? ?100?00201 01?????10? ??00001111 1?01111101 ?????????? ?0011200?? ???????012 ??????1100 1100??????

?0???000?1 10????111? 0??0010?10 ??????0?0? 00?11?0??0 ?????????? ?????????? 0?000????? ??????????

000000?110 1????????? 0101000000 01??001?11 1????00?10 ?????????? ?????????0 ?010?????? ??0??0???0

Nesbitt et al. 2010

?0???????? ?????????? 10010????? ?????0???1 ?10??????? 01??0?A??0 ??????????

?????1?0?? ??????02?? ?????????? ?00?0????? 1????????? 00???????? ??????????

Martinez et al. 2011

??0?0010?1 ???0100??? ?????????? ?????????

01?1?0???? ?0?1???0?? ???????000

??00?0??00 ????????20 1?1?00??0?

3. Extended differential diagnosis of Pampadromaeus barberenai. Distinguished from all other well-known Triassic dinosauromorphs by the following unique combination of character states: head longer than two-thirds of the femoral length (short head is characteristic of sauropodomorphs; Langer and Benton, 2006), most teeth lanceolate with coarse denticles (unknown in herrerasaurs, Eodromaeus, and basal neotheropods; Martinez et al., 2011), premaxilla with short subnarial process and femur with well-developed trochanteric shelf (the reverse is seen in Eoraptor; Langer and Benton, 2006), concave ventral margin of the premaxilla-maxilla articulation and unserrated mesialmost tooth crowns (known only in some neotheropods; Tykoski and Rowe, 2004), epipodium significantly longer than femur (the reverse is seen in Saturnalia; Langer, 2003), no inset of first premaxillary or dentary tooth (the reverse is seen in silesaurids and sauropodomorphs; Sereno, 2007; Nesbitt et al., 2010; Martinez et al., 2011), deep sacral ribs and femoral head with reduced medial tuberosity (the opposite is seen in Marasuchus and lagerpetids; Novas 1996; Nesbitt et al., 2009). The ilium of Pampadromaeus differs from that of Chromogisaurus (Ezcurre 2010) for its deeper dorsal lamina, less caudally extensive supracetabular crest, and smoother cranial portion of the brevis shelf, whereas its scapula differs from that of Panphagia (Martinez and Alcober 2009) for its slenderer blade and shorter and deeper acromial process. In addition, the articular of Panphagia lacks a well developed foramen for the chorda tympani, medial to the glenoid.

4 - Details of the parsimony analyses. A series of four parsimony analyses were performed based on the character-taxon matrices of Martinez and Alcober (2009), Ezcurra (2009), Nesbitt et al. (2010), and Martinez et al. (2011). The only modifications, other than the scoring of Pampadromaeus barberenai (Table 1), were those proposed by Bittencourt and Kellner (2009) for Staurikosaurus pricei in the data-matrix of Langer and Benton (2006), incorporated into the matrix of Martinez and Alcober (2009). The character-taxon matrices were modified in Mesquite v.2.6 (Maddison and Maddison, 2009), and TNT v.1.1 (Goloboff et al., 2003; Goloboff et al., 2008) was employed to perform heuristic/exhaustive searches. Zero length branches were not collapsed. Other parameters (e.g., constrains, additive sequence of characters states) follow the original studies. Bootstrap proportions, Bremer support decay indices, and tree statistics were calculated using TNT v.1.1. Bootstrap analyses were conducted with one thousand replicates and the same parameters employed for the primary MPT searches.

4.1. Analysis based on Martinez and Alcober (2009) An implicit enumeration search revealed tree (03) most parsimonious trees (MPTs) with 197 steps, summarized here in a strict consensus tree (Fig. S8). Tree statistics: CI

(consistency index, excluding uninformative characters) = 0.533; HI (homoplasy index, excluding uninformative characters) = 0.467; RC (rescaled consistency index) = 0.297; RI (retention index) = 0.558.

4.2. Analysis based on Ezcurra (2010). A heuristic search (Addition sequence = random; Branch-swapping algorithm = TBR; 10.000 replicates; hold = 10) revealed four (04) most parsimonious trees with 1,171 steps, summarized here in a strict consensus tree (Fig. S9). Tree statistics: CI (consistency index, excluding uninformative characters) = 0.376601; HI (homoplasy index, excluding uninformative characters) = 0.623399; RC (rescaled consistency index) = 0.255811; RI (retention index) = 0.679262. 4.3. Analysis based on Nesbitt et al. (2010). A heuristic search (Addition sequence = random; Branch-swapping algorithm = TBR; 10.000 replicates; hold = 10) revealed three (03) most parsimonious trees with 757 steps, summarized here in a strict consensus tree (Fig. S10). Tree statistics: CI (consistency index, excluding uninformative characters) = 0.457181; HI (homoplasy index, excluding uninformative characters) = 0.542819; RC (rescaled consistency index) = 0.321476; RI (retention index) = 0.703170. 4.4. Analysis based on Martinez et al. (2011) An implicit enumeration search revealed five (05) most parsimonious trees (MPTs) with 248 steps, summarized here in a strict consensus tree (Fig. S11). Tree statistics: CI (consistency index, excluding uninformative characters) = 0.613; HI (homoplasy index, excluding uninformative characters) = 0.387; RC (rescaled consistency index) = 0.490; RI (retention index) = 0.799. An unconstrained analysis reveals a set of 10 MPTs of 247 steps, the strict consensus of which shows Silesauridae forming a clade with Heterodontosauridae and Genasaura+Lesothosaurus.

Fig. S8 Strict consensus of 3 MPTs based on the inclusion of Pampadromaeus barberenai in the character-taxon matrix of Martinez and Alcober (2009). Bootstrap proportions and Bremer support decay indices are successively indicated for each clade 9

Fig. S9 Strict consensus of 4 MPTs based on the inclusion of Pampadromaeus barberenai in the character-taxon matrix of Ezcurra (2010). Bootstrap proportions and Bremer support decay indices are successively indicated for each clade

10

Fig. S10 Strict consensus of 3 MPTs based on the inclusion of Pampadromaeus barberenai in the character-taxon matrix of Nesbitt et al. (2010). Bootstrap proportions and Bremer support decay indices are successively indicated for each clade

11

Fig. S11 Strict consensus of 5 MPTs based on the inclusion of Pampadromaeus barberenai in the character-taxon matrix of Martinez et al. (2011). Bootstrap proportions and Bremer support decay indices are successively indicated for each clade

5. Character analyses Tables S2-5 list all unambiguous synapomorphies (recovered using TNT v.1.1) of the least inclusive clade containing Pampadromaeus barberenai in the strict consensus trees of Figs S7-10 (indicated with a black circle). Each line of the tables successively includes: number and definition of the characters in the original studies, direction of the transformation in the present analyses, character state scores for P. barberenai, and number of MPTs that include the synapomorphy.
Table S2: Partial synapomorphy list based on Nesbitt et al. (2010).

0>1 5 Premaxillary teeth, number: 3(0); 4 (1); 5 (2); 6+ (3); 0 (4). Articular, foramen on the medial side: absent (0); present and medial to the 0>1 92 101 150 216
glenoid (1). Tooth, crown: not mesiodistally expanded (0); mesiodistally expanded above root in cheek teeth (1). Humerus, distal end width: narrower or equal to 30% of humerus length (0); greater than 30% of humerus length (1). Femur, proximal surface: rounded and smooth (0); transverse groove that is straight (1); transverse groove that is curved (2).

1 1 1 1 1

3 3 1 3 3

0>1 0>1 0>1

12

Table S3: Partial synapomorphy list based on Martinez et al. (2011).

0>1 2 margin (1). 0>1 19 Nasal, posterolateral process: absent (0); present (1). Squamosal, ventral process, shape: transversely compressed flange (0); slender 0>1 24 45 69 70 76 79 81 90 95 122 125
prong 3 or more times basal width (1). Dentary tooth 1, position: terminal (0); inset (1). Humerus, deltopectoral crest, length relative to humeral length: 30% or less (0); 35-44% (1); 45% or more (2). Manus, length (longest digit) relative to humerus + radius: 20-30% (0); approximately 40% (1); 50- 70% (2). Manus digit I phalanx 1, rotation of axis through distal condyles: no rotation or slight ventrolateral rotation (0); rotated 45 ventromedially (1); rotated 60 ventromedially (2). Ilium, preacetabular process, shape: tab-shaped (0); strap-shaped (1); subtriangular (2); semicircular (3). Ilium, preacetabular process, attachment scar: absent (0); present (1). Ischium, mid shaft, cross-sectional shape: oval or elliptical (0); subtriangular (1). Ischium, antitrochanter, anteroposterior length relative to adjacent length of the articular surface for the ilium: greater (0); less (1). Astragalus, fibular facet, primary orientation: dorsolateral (0); lateral (1). Astragalus, anteromedial corner, shape (dorsal view): subrectangular (0); anteriorly projecting at least 25% width of the medial side of the astragalus (1).

External naris, size: small, set within tapered snout end (0); large, expanded narial

? ? 1 0 ? ? ? 2 ? ? ? ? ?

5 4 4 2 4 4 4 5 4 5 3 5 5

0>1 1>2 2>1 0>1 0>2 0>1 0>1 0>1 0>1 0>1

Table S4: Partial synapomorphy list based on Martinez and Alcober (2009).

0>1 14 length 24 Maxillary/dentary tooth crowns unexpanded (0) or rostro-caudally expanded at the 0>1 26 28 48 50 90
base (1) Lanceolate crowns present in none (0), some (1), or most (2) maxilla/dentary teeth Tooth crowns on the rostral quarter of the tooth-bearing areas of the upper and lower jaws are about the same height (0) or significantly higher than more caudal teeth (1) Humerus longer than or subequal to (0), or shorter than 0.6 of (1) the length of the femur Humeral distal end is narrower than or equal to (0), or wider than (1) 0.3 of the total length of the bone Caudolateral flange of distal tibia short and does not project (0) or projects caudal to the fibula (1)

Ventral ramus of the squamosal wider (0) or narrower (1) than a quarter of its

1 1

1 3 3 1 1 1 2

01>2 2 0>1 0 1>0 0>1 1>0 0 1 ?

13

Table S5: Partial synapomorphy list based on Ezcurra (2010).

0>1 75 semilunate and wider than high (1). 131 Relative elongation of the anterior cervical centra (cervicals 35): lengths of the 0>1
centra less than 2.5 times the height of their anterior faces (0); lengths 2.54 times the height of their anterior faces (1); the length of at least cervical 4 or 5 exceeds 4 times the anterior centrum height (2). Postzygodiapophyseal lamina in cervical neural arches 48: present (0); absent (1). Laminae of the cervical neural arches 48: well developed tall laminae (0); weakly developed low ridges (1). Length of base of the proximal caudal neural spines: less than (0), or greater than (1), half the length of the neural arch. Position of postzygapophyses in proximal caudal vertebrae: protruding with an interpostzygapophyseal notch visible in dorsal view (0); placed on either side of the posterior end of the base of the neural spine without any interpostzygapophyseal notch (1). Transverse width of the distal humerus: is less than (0), or greater than (1), 33% of the length of the humerus. Length of the pubic peduncle of the ilium: less than (0), or greater than (1), twice the anteroposterior width of its distal end. Position of the posterior end of the fibular condyle on the proximal articular surface tibia: anterior to (0) or level with (1), the posterior margin of proximal articular surface. Supraacetabular crest of ilium: not extended along the pubic peduncle or only at the base of the peduncle (0); extended along the pubic peduncle as a faint ridge (1); extended along the entire pubic peduncle and contacts the distal end as a well developed crest (2). Iliac blade in dorsal view: straight or slightly laterally curved along the whole of its anteroposterior extension (0); strongly laterally curved, with a deeply concave lateral border (1).

Shape of the supraoccipital: diamond-shaped, at least as high as wide (0);

? ?

4 4

142 143 184 185

0>1 0>1 0>1 0>1

? ? 1 1

4 4 4 4

211 252 304 363

0>1 0>1 1>0 1>2

1 1 0 2

4 4 4 4

372

0>1

6. Additional References
Bittencourt JS. Kellner AWA (2009) The anatomy and phylogenetic position of the Triassic dinosaur Staurikosaurus pricei Colbert, 1970 Zootaxa 2079:156. Goloboff PA, Farris JS, Nixon KC (2003) TNT: Tree analysis using new technologies. http://www.zmuc.dk/public/phylogeny Goloboff PA, Farris JS, Nixon KC (2008) TNT, a free program for phylogenetic analysis. Cladistics 24:774786. Maddison WP, Maddison D R (2009) Mesquite: a modular system for evolutionary analysis. Version 2.6. http://mesquiteproject.org Nesbitt SJ (2011) The Early Evolution of Archosaurs: Relationships and the Origin of Major Clades. Bull Am Mus Nat Hist 352:1292. Sereno PC (2007) Basal Sauropodormorpha: historical and recent phylogenetic hypothesis, with comments on Ammosaurus major (Marsh, 1889). In: Barrett PM, Batten DJ (Eds) Evolution and palaeobiology of early sauropodomorph dinosaurs. Spec Pap Palaeontol 77:261289

14