Aetosauria is an early-diverging clade of pseudosuchians (crocodile-line archosaurs) that had a global distribution and high species diversity as a key component of various Late Triassic terrestrial faunas. It is one of only two Late Triassic clades of large herbivorous archosaurs, and thus served a critical ecological role. Nonetheless, aetosaur phylogenetic relationships are still poorly understood, owing to an overreliance on osteoderm characters, which are often poorly constructed and suspected to be highly homoplastic. A new phylogenetic analysis of the Aetosauria, comprising 27 taxa and 83 characters, includes more than 40 new characters that focus on better sampling the cranial and endoskeletal regions, and represents the most comprenhensive phylogeny of the clade to date. Parsimony analysis recovered three most parsimonious trees; the strict consensus of these trees finds an Aetosauria that is divided into two main clades: Desmatosuchia, which includes the Desmatosuchinae and the Stagonolepidinae, and Aetosaurinae, which includes the Typothoracinae. As defined Desmatosuchinae now contains Neoaetosauroides engaeus and several taxa that were previously referred to the genus Stagonolepis, and a new clade, Desmatosuchini, is erected for taxa more closely related to Desmatosuchus. Overall support for some clades is still weak, and Partitioned Bremer Support (PBS) is applied for the first time to a strictly morphological dataset demonstrating that this weak support is in part because of conflict in the phylogenetic signals of cranial versus postcranial characters. PBS helps identify homoplasy among characters from various body regions, presumably the result of convergent evolution within discrete anatomical modules. It is likely that at least some of this character conflict results from different body regions evolving at different rates, which may have been under different selective pressures.
Keywords: Triassic, Aetosauria, Chinle Formation, Phylogeny, Partitioned Bremer Support, Pseudosuchia
The goal of phylogenetic systematics is to determine phylogenetic relationships of organisms based on shared homologous character states, and to use this information to interpret the evolutionary histories of clades, or monophyletic lineages of organisms, as well as the histories of various evolutionary character transformations (e.g., Wiley & Lieberman, 2011). This presents special challenges for vertebrate groups with extensive carapaces of dermal armor like those of aetosaurian and ankylosaurid archosaurs, which are comprised of hundreds of individual osteoderms (e.g., Desojo et al., 2013). Whereas these osteoderms may be common in the fossil record, they are generally dissociated from the rest of the skeleton prior to burial (Heckert & Lucas, 2000). It has been asserted for aetosaurians that osteoderms provide an exhaustive source of phylogenetically informative character data above and beyond that provided by the underlying skeleton (e.g., Long & Ballew, 1985; Heckert & Lucas, 1999; Parker, 2007), but it has also been argued that, while informative, these data may be plagued with phylogenetically confounding homoplasy (Parker, 2007; Parker, 2008a). The specific goal of this paper is to confront these assertions analytically, first by undertaking an expanded phylogeny of aetosaurian archosaurs based on the largest taxonomic sample yet assembled, using a suite of characters that samples both osteoderms and endoskeletal characters; and second, by applying a new method (Partitioned Bremer Support) to assess character support and conflict within an entirely morphological dataset.
Aetosaurians are quadrupedal, pseudosuchian archosaurs characterized by antero-posteriorly shortened skulls with upturned snouts, and heavy armor carapaces, as well an armor plastron (Fig. 1; Walker, 1961; Desojo et al., 2013). They had a global distribution during the Late Triassic and are often used as index fossils for biostratigraphic correlations (Heckert & Lucas, 1999; Desojo et al., 2013). The paramedian osteoderms possess a diagnostic surface ornamentation that allows for assignment of osteoderms and associated material to specific taxa, although as previously mentioned some of these osteoderm characters may be homoplastic (Long & Ballew, 1985; Parker, 2007). Accordingly it has been argued that characters from the lateral osteoderms may be more phylogenetically informative than those from the paramedian series (Parker, 2007).
Skeletal reconstruction of an aetosaur (Stagonolepis robertsoni) showing the extensive carapace and associated armor in dorsal (A) and lateral (B) views.
When Long & Ballew (1985) first proposed a taxonomy of North American aetosaurs based exclusively on osteoderm characters, they recognized only four taxa (Desmatosuchus, Typothorax, Calyptosuchus, Paratypothorax). Much new work based upon many new specimens reveals that the particular osteoderm character combinations proposed by Long & Ballew (1985) in fact can occur in many other unique combinations, resulting in the establishment of many new taxa from North America based almost solely on osteoderms (e.g., Zeigler, Heckert & Lucas, 2003; Martz & Small, 2006; Spielmann et al., 2006; Lucas, Hunt & Spielmann, 2007; Parker, Stocker & Irmis, 2008; Heckert et al., 2015). Moreover, it has been demonstrated that aetosaurs with nearly identical osteoderm character combinations can differ significantly in the other portions of the skeleton, especially in the cranial elements, indicating even more taxonomic potential (Desojo, 2005; Desojo & Báez, 2005; Desojo & Báez, 2007; Desojo & Ezcurra, 2011). Finally, aetosaurian osteoderm characters are not intraorganisimally homogeneous (i.e. characters can differ depending on position within the same carapace) and capturing this variation in the construction of phylogenetically informative characters is challenging (Harris, Gower & Wilkinson, 2003; Parker, 2007; Parker, 2008b; Desojo et al., 2013).
Although early studies did focus on character change across broadly defined carapace regions such as the cervical and caudal regions (e.g., Long & Ballew, 1985; Heckert & Lucas, 1999), more recent studies have sought to detail variation within those subregions (Martz, 2002; Parker, 2003; Parker, 2008b; Schoch, 2007; Parker & Martz, 2010; Heckert et al., 2015). Potentially further complicating this situation is our general lack of data regarding character transformations affected by ontogenetic variation as well as differences caused by individual and sexual dimorphism (Taborda, Cerda & Desojo, 2013; Taborda, Heckert & Desojo, 2015). Overall though, the rich source of character data present in aetosaurian osteoderms provides the systematist with a broad canvas on which to construct a detailed phylogenetic hypothesis, presuming of course that the changes in osteoderm characters are indeed phylogenetically informative (Parker, 2007) and that the homology of these characters can be determined (e.g., Harris, Gower & Wilkinson, 2003; Parker & Martz, 2010; Heckert et al., 2015).
At present we do not have an appropriate sample size across all clades to capture all of intraorganisimal character variation that occurs across the aetosaurian carapace and plastron. Indeed, many taxa are currently only known from a handful of associated osteoderms (e.g., Tecovasuchus, Apachesuchus, Rioarribasuchus), with the current challenge simply lying in determining the proper position of these osteoderms within the carapace (Lucas, Heckert & Hunt, 2003; Martz & Small, 2006; Spielmann et al., 2006; Parker, 2007; Lucas, Hunt & Spielmann, 2007; Parker & Martz, 2010; Spielmann & Lucas, 2012). As more discoveries are made, particularly of associated and articulated specimens, our increased understanding of positional variation should allow for more precise placement of isolated osteoderms leading to stronger determinations of homology of individual osteoderms (Parker, 2007; Parker & Martz, 2010; Heckert et al., 2015).
For this study all previously recommended characters used for determination of aetosaurian systematics were reviewed (Parrish, 1994; Heckert, Hunt & Lucas, 1996; Heckert & Lucas, 1999; Desojo, 2005; Parker, 2007; Desojo, Ezcurra & Kischlat, 2012; Roberto-Da-Silva et al., 2014; Heckert et al., 2015). Characters were discarded if found to be generally uninformative or ambiguously scored. The retained characters, as well as new characters, have been rewritten to be more descriptive and thus hopefully easier to interpret and score. Although the retention and construction of many characters and associated character states would presumably lead to better resolution and clade support (Hillis, Huelsenbeck & Cunningham, 1994), the goal of any phylogenetic analysis is accuracy, and this should not come at the expense of artificial resolution by including ambiguously written characters (Slowinski, 1993). Thus, the overarching goal of this project was to recover phylogenetic trees that seem logical given our anatomical understanding of aetosaurians, rather than highly resolved and supported trees that appear problematic and nonsensical in these regards. The matrix of Parker (2007), which has been used as the basis for many recent phylogenetic analyses (Parker, Stocker & Irmis, 2008; Desojo, Ezcurra & Kischlat, 2012, Roberto-Da-Silva et al., 2014; Heckert et al., 2015), is dominated by osteoderm characters. This is problematic given the large amount of discovered homoplasy in this dataset (Parker, 2007; Desojo, Ezcurra & Kischlat, 2012), and in light of the underlying assumption that osteoderm characters provide the main phylogenetic signal for the clade irrespective of the rest of the skeleton (Desojo, 2005; Parker, 2007; Parker, 2008b). For these reasons, this study sought to increase the number of non-osteoderm characters, as suggested by Desojo (2005) & Desojo, Ezcurra & Kischlat (2012). This presents challenges because of the relative infrequency of aetosaurian postcranial remains, which are lacking for many taxa or sometimes obscured by articulated carapaces. One of the best sources for aetosaurian postcranial bones is the Placerias Quarry in northeastern Arizona (Long & Murry, 1995). However, owing to a lack of association with diagnostic osteoderm material, most of these postcranial elements cannot unequivocally be referred to species (Parker, 2014; Parker, 2005a; differing from Long & Murry, 1995). Fortunately, there is cranial material preserved for many aetosaurian taxa and almost every known skull, with the exception of some elements from the Placerias Quarry and the Post Quarry (Texas), are unambiguously associated with osteoderms allowing for a precise taxonomic referral. Thus, the present analysis was able to significantly expand the number of cranial characters utilized.
The original basis for aetosaurian phylogenetic characters and character transformations is a table of information published by Long & Ballew (1985:58) where comparisons are provided between various North American taxa, establishing a key early character-based taxonomic scheme for aetosaurians (also see Walker, 1961). Several of these characters are still utilized in recent phylogenetic analyses. The first computed phylogenetic analysis of aetosaurians (Parrish, 1994) examined 15 characters (six osteoderm, nine non-osteoderm) and eight taxa. However, nine of those characters are parsimony-uninformative for the ingroup, and there are several incorrect scorings and typographical errors that affect the analysis; thus the published tree is neither well-resolved, nor accurate in its character state distributions (Harris, Gower & Wilkinson, 2003). Heckert, Hunt & Lucas (1996) expanded on Parrish’s (1994) work, inflating the matrix to nine taxa and 22 (potentially 23) characters (17 armor, five non-armor). That study was also affected by some scoring errors, as well as the lack of use of a non-aetosaurian outgroup to root the resulting trees (Harris, Gower & Wilkinson, 2003), but did include many new characters that have been used in subsequent aetosaurian phylogenetic studies. Furthermore that study was the first to unambiguously recover the major clades Desmatosuchinae and Typothoracisinae (sensuParker, 2007).
Heckert & Lucas (1999) aimed to expand the matrix of Heckert, Hunt & Lucas (1996), mainly to determine the phylogenetic relationships of a new taxon, Coahomasuchus kahleorum. Their published matrix consisted of 13 in-group taxa and 60 characters. However, 26 of these characters as coded were parsimony uninformative, and as noted by Harris, Gower & Wilkinson (2003) the published matrix included several typographical errors. When corrected, that matrix produced a tree that was different from the published one. Harris, Gower & Wilkinson (2003) were critical of several other aspects of this study, including the ad hoc deletion of taxa from the matrix when safe methods to determine appropriate taxon deletion were available (e.g., Wilkinson, 1995a), and character constructions that inflated seemingly non-independent character suites and biased the resulting tree (composite versus reductive coding; Rowe, 1988; Wilkinson, 1995b). Nonetheless, the study by Heckert & Lucas (1999) built further upon the character list of Heckert, Hunt & Lucas (1996) and represents a very important progression in our understanding of aetosaurian systematics.
The most recent core phylogenetic analysis of aetosaurians (Parker, 2007) focused on the lateral osteoderms, whereas the previous studies had focused more on characters of the paramedian osteoderms (Heckert, Hunt & Lucas, 1996; Heckert & Lucas, 1999). Parker (2007) noted that aetosaurians could roughly be divided into three groups based on the overall anatomy of the lateral osteoderms. This translated into a phylogenetic analysis (16 in-group taxa, 37 characters) that recovered three distinct clades: Aetosaurinae, Desmatosuchinae (Heckert & Lucas, 2000) and Typothoracinae. Whereas support for Desmatosuchinae and Typothoracinae was strong, especially for the subclade Paratypothoracini, Aetosaurinae was unresolved and weakly supported. This became especially apparent when other taxa were subsequently added to the matrix, causing significantly different tree topologies and character support (Parker, Stocker & Irmis, 2008; Desojo, Ezcurra & Kischlat, 2012). Indeed, a recent study (Desojo, Ezcurra & Kischlat, 2012) failed to recover Aetosaurinae as a clade, with Aetosaurus ferratus as the only member by definition (Heckert & Lucas, 2000). Desmatosuchinae is always recovered and well-supported, but relationships within the clade are not always fully resolved (e.g., Parker, Stocker & Irmis, 2008); however, Typothoracinae remains well-supported and resolved. Nonetheless, criticisms of the Parker (2007) dataset include the lack of endoskeletal characters as well as some scoring errors (see Desojo & Ezcurra, 2011; Desojo, Ezcurra & Kischlat, 2012; Heckert et al., 2015).
Materials and Methods
In order to test these questions about taxon sampling, character independence, and tree topology, the matrix has been expanded to include more taxa and characters. The new matrix (Appendix A) utilizes 83 characters for 26 ingroup taxa. The characters are well-divided between anatomical regions, with endoskeletal characters constituting the majority (34 cranial, 16 axial/appendicular, 33 osteoderm).
The 26 in-group taxa include the majority of aetosaurian taxa currently considered valid (Desojo et al., 2013; Roberto-Da-Silva et al., 2014; Heckert et al., 2015). They are listed below, and this study is the first to investigate the phylogenetic positions of Adamanasuchus eisenhardtae, Apachesuchus heckerti, Stagonolepis olenkae, Redondasuchus rineharti as well as a new taxon, Scutarx deltatylus gen. et sp. nov. Other taxa are rescored (e.g., Coahomasuchus kahleorum; Typothorax coccinarum) based on new referred material.
Taxa excluded from this analysis include Acaenasuchus geoffreyi (Long & Murry, 1995; Redondasuchus reseriHunt & Lucas, 1991; Typothorax antiquumLucas, Heckert & Hunt, 2003; and Chilenosuchus forttaeCasamiquela, 1980). Acaenasuchus and Chilenosuchus were excluded because Chilenosuchus presently scores as a taxonomic equivalent (sensuWilkinson, 1995a) of Typothorax coccinarum, and newly recognized material, including vertebrae and fused osteoderms, of Acaenasuchus casts doubt on its aetosaurian identify (M. Smith, personal communication, 2014). Redondasuchus reseri is poorly known and presently scores as a taxonomic equivalent of Redondasuchus rineharti; whereas Typothorax antiquum represents an ontogenetic stage of Typothorax coccinarum rather than a distinct species (Parker, 2006; Parker & Martz, 2011; Martz et al., 2013). In any case, in this matrix Typothorax antiquum and Typothorax coccinarum are taxonomic equivalents (i.e., they are scored exactly the same, and thus can obscure relationships in the data if both are included; Wilkinson, 1995a), so the less complete, Typothorax antiquum, is excluded.
Revueltosaurus callenderi is included in the analysis as an outgroup because it is currently recovered as the sister taxon of Aetosauria (Nesbitt, 2011). Furthermore, it is known from several specimens, which preserve nearly the entire skeleton (Parker et al., 2007). Postosuchus kirkpatricki is utilized as an outgroup because it is relatively complete, well-described and illustrated (Weinbaum, 2011; Weinbaum, 2013). Furthermore, it represents a more crownward clade (Paracrocodylomorpha) within Pseudosuchia providing a deeper optimization of character states than can be provided by Revueltosaurus. Both of these taxa have been utilized as outgroups in previous phylogenetic studies of the Aetosauria (e.g., Heckert & Lucas, 1999; Parker, 2007; Desojo, Ezcurra & Kischlat, 2012; Heckert et al., 2015). Unfortunately neither Postosuchus nor Revueltosaurus can presently be scored for lateral osteoderm characters and therefore these characters have been scored as inapplicable for these taxa. Furthermore, most of the paramedian osteoderm characters were scored as inapplicable for Postosuchus because even though Postosuchus possesses trunk osteoderms, the homology of characters such as ornamentation pattern and presence of certain processes cannot be determined.
A previous work (Parker, 2007) incorporated many scorings from past studies (Parrish, 1994; Heckert, Hunt & Lucas, 1996; Heckert & Lucas, 1999) some of which were later determined to be erroneous (Schoch, 2007; Desojo & Ezcurra, 2011; Desojo, Ezcurra & Kischlat, 2012; Heckert et al., 2015). Therefore, for this study the matrix was scored from scratch and the scorings completed from carefully studying materials first hand for most taxa, and using photos and the literature for any not studied first-hand (Stagonolepis olenkae, Aetosaurus ferratus, SMNS 19003 (Desojo & Schoch, 2014), Stenomyti huangae, Redondasuchus rineharti, Gorgetosuchus pekinensis, Polesinesuchus aurelioi). Much effort was directed toward detecting and fixing typographic errors, which can have a major effect on the final tree topologies (Harris, Gower & Wilkinson, 2003). Scoring completeness is shown in Supplemental Table 1 for each taxon, with inapplicable characters counted as scored. Completeness scores range from 98% (80 of 82) for Desmatosuchus smalli, which is known from several skulls and skeletons; to 22% for Apachesuchus heckerti (18 of 82), which is known only from five paramedian osteoderms. The average completeness score was 60%. The major factor causing incompleteness is a lack of skull material, which affected all taxa that scored lower than 50%. Because aetosaurians are generally identified by armor characters, there are no taxa that consist solely of cranial material, in contrast with many other groups (e.g., synapsids, dinosaurs).
The electronic version of this article in Portable Document Format (PDF) will represent a published work according to the International Commission on Zoological Nomenclature (ICZN), and hence the new names contained in the electronic version are effectively published under that Code from the electronic edition alone. This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix http://zoobank.org/. The LSID for this publication is: urn:lsid:zoobank.org:pub:841F81C7-A4AE-4146-94FE-DFE0A6725634. The online version of this work is archived and available from the following digital repositories: PeerJ, PubMed Central and CLOCKSS.
Institutional abbreviations – AMNH, American Museum of Natural History, New York, USA; ANSP, Academy of Natural Sciences of Drexel University, Philadelphia, Pennsylvania, USA; CPE2, Coleção Municipal, São Pedro do Sul, Brazil; DMNH, Perot Museum of Natural History, Dallas, Texas, USA; DMNH, Denver Museum of Nature and Science, Denver, Colorado, USA; FMNH, Field Museum, Chicago, IL, USA; FR, Frick Collection, American Museum of Natural History, New York, USA; MCCDP, Mesalands Community College Dinosaur Museum, Tucumcari, New Mexico, USA; MCSNB, Museo Civico di Scienze Naturali Bergamo, Bergamo, Italy; MCP, Museo de Ciencias e Tecnología, Porto Alegre, Brazil; MCZ, Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts, USA; MCZD, Marischal College Zoology Department, University of Aberdeen, Aberdeen, Scotland, UK; NCSM, North Carolina State Museum, Raleigh, North Carolina, USA; NHMUK, The Natural History Museum, London, United Kingdom; NMMNH, New Mexico Museum of Natural History and Science, Albuquerque, New Mexico, USA; MNA, Museum of Northern Arizona, Flagstaff, Arizona, USA; PEFO, Petrified Forest National Park, Petrified Forest, Arizona, USA; PFV, Petrified Forest National Park Vertebrate Locality, Petrified Forest, Arizona, USA; PVL, Paleontología de Vertebrados, Instituto ‘Miguel Lillo’, San Miguel de Tucumán, Argentina; División de Paleontología de Vertebrados del Museo de Ciencias Naturales y Universidad Nacional de San Juan, San Juan, Argentina, SMNS, Staatliches Museum für Naturkunde, Stuttgart, Germany; TMM, Texas Memorial Museum, Austin, Texas, USA; TTUP, Museum of Texas Tech, Lubbock, Texas, USA; UCMP, University of California, Berkeley, California, USA; ULBRA PVT, Universidade Luterana do Brasil, Coleção de Paleovertebrados, Canoas, Rio Grande do Sul, Brazil; UMMP, University of Michigan, Ann Arbor, Michigan, USA; USNM, National Museum of Natural History, Smithsonian Institution, Washington, D.C., USA; VPL, Vertebrate Paleontology Lab, University of Texas at Austin, Austin, Texas, USA; YPM, Yale University, Peabody Museum of Natural History, New Haven, Connecticut, USA; VRPH, Sierra College, Rocklin, California, USA; ZPAL, Institute of Paleobiology of the Polish Academy of Sciences in Warsaw, Warsaw; Poland.
The phylogenetic study by Nesbitt (2011) is currently the basis for most studies of archosauriform relationships (e.g., Nesbitt & Butler, 2013; Butler et al., 2014). This study utilizes the format used in that study for the listing of terminal taxa and characters to make this work compatible.
Adamanasuchus eisenhardtae (Lucas, Hunt & Spielmann, 2007)
Holotype – PEFO 34638, partial skeleton including paramedian and lateral osteoderms, several vertebral centra, and a partial femur (Lucas, Hunt & Spielmann, 2007).
Referred Material – PEFO 35093, osteoderm fragments, nasal fragment; PEFO 36806, osteoderm fragments.
Remarks – Lucas, Hunt & Spielmann (2007) refer a lateral osteoderm (UCMP 126867) to Adamanasuchus eisenhardtae without explanation other than noting a 2007 personal communication from Andrew Heckert. They neither figure nor describe the specimen, but list its provenance as the Placerias Quarry near St. Johns, Arizona and attribute it as another Adamanian record of Adamanasuchus eisenhardtae. Examination of UCMP 126867 confirms the identification of the element as an aetosaurian lateral osteoderm; however, the specimen was collected from PFV 075 (Karen’s Point) in Petrified Forest National Park and not from the Placerias Quarry. PFV 075 is in the Martha’s Butte beds of the Sonsela Member, which are Revueltian in age (Parker & Martz, 2011), thus this would represent a range extension of this taxon up into the Sonsela Member and into the Revueltian biozone. This specimen differs from the holotype of Adamanasuchus eisenhardtae in possessing an extremely reduced dorsal flange and a dorsal eminence that forms a broadly triangular “spine” that projects dorsally. The outer surface of the lateral flange and the dorsal eminence bear an elongate ridge, which is located very close to the plate margin. Curiously the osteoderm lacks an anterior bar so it cannot be determined if this margin is the anterior or posterior edge. In Adamanasuchus eisenhardtae, the lateral osteoderms are more symmetrical with nearly equal lateral and dorsal flanges, and the eminence does not form a projected spine (PEFO 34638). Because of these anatomical differences and the discrepancy in the stratigraphic and locality data, the referral of this specimen to Adamanasuchus eisenhardtae is not supported.
PEFO 35093 includes osteoderm fragments that possess the unique surface ornamentation of a faint background, radial pattern, ‘overprinted’ by deep randomly developed pits. This ‘overprinting’ is characteristic of Adamanasuchus eisenhardtae and differs from other aetosaurians with a radial ornamentation pattern (Lucas, Hunt & Spielmann, 2007). An associated fragment of a nasal most likely belongs to the same specimen as it has an identical preservation and no other aetosaur specimens were recovered from the immediate area. Unfortunately, the nasal fragment is too incomplete to provide more information. PEFO 36806 is another referred specimen and consists solely of osteoderm fragments. Both PEFO 35093 and PEFO 36806 were recovered from the upper part of the Blue Mesa Member at about the same stratigraphic horizon as the holotype specimen of Adamanasuchus eisenhardtae.
Age – Late Triassic, early to middle Norian, Adamanian (Ramezani et al., 2011; Parker & Martz, 2011).
Occurrence – upper Blue Mesa Member, Chinle Formation, Petrified Forest National Park, Arizona, U.S.A. (Lucas, Hunt & Spielmann, 2007; Parker & Martz, 2011).
Remarks – Lucas, Hunt & Spielmann (2007) named Adamanasuchus eisenhardtae for a partial skeleton collected from the upper part of the Blue Mesa Member (Chinle Formation) in Petrified Forest National Park in 1996 (Hunt, 1998; Parker, 2006). Parker (2006) incorrectly assigned this specimen to Typothorax antiquum based on interpretation of comments made by Hunt (1998) regarding this specimen. In 2010, park staff revisited the type locality and finished the excavation; several paramedian and lateral osteoderms had been covered and left by the original workers and these materials were not included in the original description. The diagnosis provided by Lucas, Hunt & Spielmann (2007) does not adequately differentiate Adamanasuchus eisenhardtae from other known aetosaurians, in particular from Calyptosuchus wellesi; however, key characters found in Adamanasuchus eisenhardtae to the exclusion of Calyptosuchus wellesi are the strongly sigmoidal lateral edge, that results is a ventrolateral corner of the osteoderm that appears to have been sheared-off, and a triangular patch in the posteromedial corner of the paramedian osteoderm surface that is smooth and devoid of ornamentation. The first character state also occurs in paratypothoracins and the second is found in a new aetosaur species described below (e.g., PEFO 34616), except that in the latter taxon the triangular area is strongly raised.
Key References – Lucas, Hunt & Spielmann (2007).
Aetobarbakinoides brasiliensis (Desojo, Ezcurra & Kischlat, 2012)
Holotype – CPE2 168, partial postcranial skeleton (Desojo, Ezcurra & Kischlat, 2012). A cast of this specimen is in the Petrified Forest National Park (PEFO) collections.
Referred Material – none.
Age – Late Triassic, late Carnian – earliest Norian, Hyperodapedon Assemblage Zone (Langer et al., 2007; Martinez et al., 2011).
Occurrence – Sequence 2, Santa Maria Supersequence, Rio Grande Do Sul, Brazil (Desojo, Ezcurra & Kischlat, 2012).
Remarks – The holotype (CPE2 168) of Aetobarbakinoides brasiliensis is a fragmentary postcranial skeleton of a small aetosaurian that was originally referred to Stagonolepis robertsoni (=Aetosauroides in their hypothesis) by Lucas & Heckert (2001). The lack of open neurocentral sutures in the cervical and trunk vertebrae suggests that CPE2 168 represents a skeletally mature individual (Irmis, 2007). Despite the fragmentary preservation of the holotype, Desojo & Ezcurra (2011) were able to distinguish this material from that of other South American aetosaurs, based on the presence of discrete vertebral laminae in the trunk series, a character lacking in taxa such as Aetosauroides scagliai and Neoaetosauroides engaeus. Furthermore, Aetobarbakinoides is the only South American aetosaurian specimen with trunk vertebrae that bear accessory articular structures (i.e. hyposphene), a feature recognized previously in aetosaurians only in desmatosuchines (Parker, 2008b). Determining the phylogenetic position of this taxon is difficult because it is represented primarily by endoskeletal (non-osteoderm) material. A few osteoderms are present, but the surface ornamentation is poorly preserved. Lateral osteoderms, which have been key to phylogenetic placement (Parker, 2007), are not preserved. Furthermore, the preserved paramedian osteoderms lack their lateral edges, which, if preserved, would have provided information about the medial edges of the lateral osteoderms allowing for the scoring of some characters. Desojo, Ezcurra & Kischlat (2012) recovered Aetobarbakinoides brasiliensis as the sister taxon of the clade Desmatosuchinae + Typothoracinae; however, Heckert et al. (2015) considered it to be a ‘wildcard’ (unstable) taxon in their analysis and pruned it a posteriori from their published tree. It performed as a wildcard taxon in the present analysis as well, which is discussed in more detail below.
Key References – Desojo, Ezcurra & Kischlat (2012).
Aetosauroides scagliai (Casamiquela, 1960)
Holotype – PVL 2073, postcranial skeleton including the majority of the carapace, vertebral column, and sacrum in articulation (Casamiquela, 1961).
Referred Material – see Desojo & Ezcurra (2011).
Age – Late Triassic, Carnian, Hyperodapedon Assemblage Zone (Rogers et al., 1993; Furin et al., 2006; Martinez et al., 2011).
Occurrence – Cancha de Bochas Member, Ischigualasto Formation, Argentina; Sequence 2, Santa Maria Supersequence, Rio Grande do Sul State, Brazil (Casamiquela, 1961; Desojo & Ezcurra, 2011).
Remarks – Aetosauroides scagliai was originally described by Casamiquela (1960) & Casamiquela (1961) based on well-preserved cranial and postcranial material from the lower part of the Ischigualasto Formation of Argentina. Further material was assigned by Casamiquela (1967) who redescribed the specimens in light of the monograph on Stagonolepis robertsoni by Walker (1961). Strong similarities have been noted between Aetosauroides and Stagonolepis as well as Aetosaurus and based on element size Aetosauroides was considered to be somewhat morphologically transitional between the two European taxa (Casamiquela, 1967). In an unpublished masters thesis, Zacarias (1982) erected a second species of Aetosauroides (“Aetosauroides subsulcatus”) for material from the Upper Triassic of Brazil. All of this material has been briefly redescribed, the majority of it assigned to Stagonolepis robertsoni (Lucas & Heckert, 2001; Heckert & Lucas, 2002). Those authors argued that only superficial differences could be found between all of these specimens and that assignment of the South American material strengthened previously proposed biostratigraphic correlations between Brazil, Argentina, and the U.K., as well as to the southwestern United States. In contrast, Desojo & Ezcurra (2011) assigned the Brazilian material to Aetosauroides scagliai based on the presence of well-developed fossae on the lateral sides of the trunk vertebrae and the exclusion of the maxilla from the external naris in the skull of Aetosauroides scagliai, a character first noted by Casamiquela (1967). A phylogenetic analysis recovered Aetosauroides scagliai as the sister taxon to all other aetosaurs (Stagonolepididae) (Desojo, Ezcurra & Kischlat, 2012). Redescriptions of the Argentinian material were presented in two unpublished dissertations (Desojo, 2005; Parker, 2014), and a full redescription by Desojo and Ezcurra is in progress (J. Desojo, personal communication, 2014).
The cranial material of Aetosauroides scagliai is significant because it exemplifies the plesiomorphic aetosaurian skull condition, optimizing characters such as the exclusion of the maxilla from the external naris, frontals that are wider than the parietals, nasals that taper anteriorly, a large triangular depression present anterior to the frontals, the lack of a ‘slipper-shaped’ mandible, the lack of a basal swelling in the teeth, and the mediolaterally compressed teeth with recurved tips (Parker, 2014). The skull is significantly different from that of Stagonolepis robertsoni, Stagonolepis olenkae, Neoaetosauroides engaeus, and Calyptosuchus wellesi and that characters of the osteoderms used to unite these taxa (e.g., Heckert & Lucas, 2002) are homoplasious (Desojo & Ezcurra, 2011; Parker, 2008b).
Cerda & Desojo (2011) provide details of the osteoderm histology of Aetosauroides scagliai, although using referred specimens rather than the holotype. This adds to the increasing understanding of the bone histology of aetosaurians (e.g., Parker, Stocker & Irmis, 2008; Scheyer, Desojo & Cerda, 2013). It is possible that once histological features and their relationships with ontogenetic maturity at time of death and potential environmental effects are better known, that histological characters can be incorporated in phylogenetic analyses of the Aetosauria.
Key References – Casamiquela (1960); Casamiquela (1961); Casamiquela (1967); Heckert & Lucas (2002); Desojo & Ezcurra (2011); Cerda & Desojo (2011); Parker (2014).
Aetosaurus ferratus (Fraas, 1877)
Lectotype – SMNS 5770, specimen XVI (16) (Schoch, 2007).
Referred Material – SMNS 5770, at least 24 specimens recovered in the same block as the lectotype; SMNS 18554, articulated skeleton lacking the skull and pectoral girdle; SMNS 14882, articulated caudal segment; SMNS 12670, trunk and ventral osteoderms; MCZ 22/92G, partial skull, limb bones and vertebrae, osteoderms; MCSNB 4864, trunk osteoderms.
Age – Late Triassic, middle Norian to early Rhaetian, Revueltian (Deutsche Stratigraphische Kommission, 2005; Lucas, 2010).
Occurrence – Lower and Middle Stubensandstein, Löwenstein Formation, Germany; Calcare de Zorzino Formation, Italy; Ørsted Dal Member, Fleming Fjord Formation, eastern Greenland (Wild, 1989; Jenkins et al., 1994; Schoch, 2007).
Remarks – The genus Aetosaurus originally included two species, Aetosaurus ferratus and Aetosaurus crassicauda. Aetosaurus crassicauda is presently understood to represent a larger specimen of Aetosaurus ferratus (Schoch, 2007). Specimens of Stegomus arcuatus from eastern North American have been assigned to Aetosaurus (Lucas, Heckert & Huber, 1998); however, the majority of this material consists of natural molds that do not preserve the surface ornamentation. These specimens are assignable to Aetosaurus only on the basis of “aetosaurine” (sensuParker, 2007) synapomorphies such as a sigmoidal lateral margin of the paramedian osteoderms with a pronounced anterolateral projection, as well as their small size. Small osteoderms (e.g., NMMNH P-17165) from the Bull Canyon Formation of New Mexico referred to Stegomus (Aetosaurus) arcuatus by Heckert & Lucas (1998) possess an anterior bar, radial pattern, offset dorsal eminence, and an anterolateral projection, which are “aetosaurine” characters and not diagnostic of a less inclusive taxon. Several authors consider the lack of dorsal ornamentation, including a dorsal eminence (boss) in the osteoderms of Stegomus (Aetosaurus) arcuatus to be diagnostic of the taxon (e.g., Heckert & Lucas, 2000; Heckert et al., 2001; Spielmann & Lucas, 2012); however, the lack of ornamentation is because the type and key referred specimens consist solely of natural molds of the ventral surfaces of the osteoderms which are typically smooth and unornamented in aetosaurs.
Purported specimens of Aetosaurus ferratus from the Chinle Formation of Colorado (Small, 1998) are now considered to represent a distinct taxon, Stenomyti huangae (Small & Martz, 2013). Aetosaurus has also been recognized from Greenland and Italy. The Greenland material consists of a partial skull, postcranial skeleton and osteoderms (MCZ 22/92G; Jenkins et al., 1994). This skull possesses the following characteristics of Aetosaurus ferratus; an anteroposteriorly short jugal, a round supratemporal fenestra; and an antorbital fossa that covers the majority of the lacrimal (Schoch, 2007). The Italian material (MCSNB 4864) consists of a short series of articulated dorsal paramedian and lateral osteoderms that possess an identical surface ornamentation to Aetosaurus ferratus (Wild, 1989). This specimen is significant as it was recovered from marine sediments of Norian age and represents a potential tie point to the marine biostratigraphic record for the Late Triassic (Lucas, 1998a; Irmis et al., 2010).
In summary, Aetosaurus ferratus is presently known from Greenland, Germany, and Italy, and other purported North American occurrences cannot be substantiated (Schoch, 2007; Small & Martz, 2013). For this study Aetosaurus ferratus is scored only from the German lectotype and referred material.
Key References – Wild (1989); Jenkins et al. (1994); Schoch (2007).
Apachesuchus heckerti (Spielmann & Lucas, 2012)
Holotype – NNMNH P-31100, left dorsal paramedian osteoderm.
Referred material – NMMNH P-63427, left cervical paramedian osteoderm; NMMNH P-63426, right caudal paramedian osteoderm. Both of these specimens were originally included in NMMNH P-31100 (Heckert et al., 2001; Spielmann & Lucas, 2012:fig. 70e), but have been renumbered. Spielmann & Lucas (2012) also report that much more complete material of this taxon, including postcrania, is currently under study by Axel Hungerbühler at the Mesalands Dinosaur Museum in Tucumcari, New Mexico. This new material is also from the Redonda Formation of New Mexico; however, the new material referable to Apachesuchus heckerti only consists of a few more paramedian osteoderms, whereas the rest of the material is actually referable to Redondasuchus rineharti (J. Martz, personal communication, 2013).
Age – Late Triassic, late Norian-Rhaetian, Apachean (Spielmann & Lucas, 2012).
Occurrence – Quay Member, Redonda Formation, Dockum Group, New Mexico, U.S.A (Spielmann & Lucas, 2012).
Remarks – The holotype and paratype (referred) osteoderms were recovered in a microvertebrate assemblage found within a very large phytosaur skull and were originally assigned to Neoaetosauroides sp. because the lack of surface ornamentation of the paramedian osteoderms was thought to be diagnostic of Neoaetosauroides (Heckert et al., 2001). However, the lack of surface ornamentation of some of the osteoderms of the holotype of Neoaetosauroides is the result of overpreparation of the specimen and close examination shows that the material does have a surface orientation of radial grooves and ridges; therefore the NMMNH material cannot be assigned to that taxon. The lack of surface ornamentation in the type material of Apachesuchus heckerti appears to be a genuine feature and is considered an autapomorphy of the taxon (Spielmann & Lucas, 2012; J. Martz, personal communication, 2013). Apachesuchus heckerti is considered to possess a low width/length ratios (> 0.3) of the paramedian osteoderms; which was obtained by comparing the length of the lateral edge to the total plate length (Heckert et al., 2001; Spielmann & Lucas, 2012). However, the lateral edge of NMMNH P-31100 is greatly expanded anteroposteriorly than the rest of the osteoderm strongly skewing this ratio. The length at the center of the osteoderm is 32 mm, compared to an overall width of 104 mm. This provides a width/length ratio of 3.25, compared to the ratio of 2.5 provided by Spielmann & Lucas (2012). It is important to standardize areas of measurements when determining ratios of aetosaur osteoderms because simply using maximum length can skew results in osteoderms with abnormal shapes. This is also true for osteoderms with elongate anterolateral processes of the anterior bars (e.g., Calyptosuchus wellesi). In these cases osteoderm lengths should be taken from the main osteoderm body and not from the anterior bar. Furthermore, an unnumbered referred anterior dorsal paramedian osteoderm in the Mesalands Community College Dinosaur Museum (MCCDM) collection (field number 20080618RET002RRB) has a width of 110 mm and a median length of 28 mm for a W/L ratio of 3.92. This compares well with typothoracine aetosaurs such as Typothorax coccinarum (Long & Murry, 1995; Heckert et al., 2010).
Key References – Heckert et al. (2001); Spielmann & Lucas (2012).
Calyptosuchus wellesi (Long & Ballew, 1985)
Holotype – UMMP 13950, articulated carapace from the posterior dorsal and caudal regions, associated with a portion of the vertebral column and the sacrum (Case, 1932; Long & Ballew, 1985).
Referred Material – UMMP 7470, two trunk paramedian osteoderms, three trunk vertebrae, mostly complete, articulated sacrum; UCMP 27225, paramedian, lateral, and ventral osteoderms, partial right dentary. Numerous specimens from the Placerias Quarry from the UCMP and the MNA collections, as well as specimens from Petrified Forest National Park also can be referred to Calyptosuchus wellesi (Long & Murry, 1995; Parker, 2014).
Age – Late Triassic, early-middle Norian, early Adamanian (Ramezani et al., 2011; Ramezani, Fastovsky & Bowring, 2014; Parker & Martz, 2011).
Occurrence – upper Blue Mesa Member, Chinle Formation, Arizona, U.S.A.; Tecovas Formation, Dockum Group, Texas, U.S.A (Long & Murry, 1995; Parker & Martz, 2011).
Remarks – Case (1932) described a posterior portion of a carapace and associated pelvis and vertebral column of what he believed to be a phytosaur from the Upper Triassic of Texas. Although he discussed possible taxonomic affinities he was thoroughly perplexed by the material and thus did not assign the specimen to an existing taxon or coin a new taxonomic name. This is mainly because of the common association of aetosaurian osteoderms with phytosaur remains (e.g., Nicrosaurus kapffi, Case, 1929) and because the osteoderms of UMMP 13950 possessed a radial surface ornamentation more similar to osteoderm material then assigned to the phytosaur Nicrosaurus (=Phytosaurus) kapffi (now the holotype of the aetosaurian Paratypothorax andressorumLong & Ballew, 1985). This is unlike the surface ornamentation found in the other aetosaurian Case was familiar with, Desmatosuchus spurensis (Case, 1922). Indeed, Case (1932) tentatively suggested that UMMP 13950 may belong to the genus “Phytosaurus.” Gregory (1953a) recognized that the specimen was probably more closely related to Typothorax than to phytosaurs and hence most likely a pseudosuchian (aetosaur), but still considered the purported close similarity of the rectangular osteoderms with those assigned to some phytosaurs to be problematic for taxonomic resolution of the material.
This problem was finally resolved by Long & Ballew (1985) who correctly determined that all of the Triassic material with broad, rectangular osteoderms was referable to aetosaurians. Those authors also listed UMMP 13950 as the holotype of a new genus, Calyptosuchus wellesi. They did not redescribe Case’s specimen, but instead discussed the new taxon in terms of referred material from the Triassic of Arizona. A recent description of the taxon is by Long & Murry (1995) who mainly described referred material from the Placerias Quarry of Arizona. The referrals of material to Calyptosuchus wellesi by Long & Murry (1995) have been questioned mainly because of the recognition that the cervical lateral osteoderms assigned to Calyptosuchus wellesi by Long & Ballew (1985) & Long & Murry (1995) actually belong to a paratypothoracin aetosaur demonstrating the presence of a third aetosaur taxon in the Placerias Quarry (Parker, 2005a; Parker, 2007).
Parker (2014) carefully sorted and grouped the Placerias Quarry material based on field numbers and used the resulting associations as well as apomorphic comparisons to test these assignments. Referred elements of Calyptosuchus wellesi were redescribed and these referred specimens, as well as the holotype, are used to and score that taxon in this phylogenetic analysis. This anatomical work, in association with detailed biostratigraphic work of the Chinle Formation (Parker & Martz, 2011), has also determined that Calyptosuchus wellesi is presently restricted to the upper part of the Blue Mesa Member and that specimens of Calyptosuchus noted from the Sonsela Member (e.g., Parker & Martz, 2011) belong to a new taxon described below.
Key References – Case (1932); Long & Ballew (1985); Long & Murry (1995); Parker (2014).
Coahomasuchus kahleorum (Heckert & Lucas, 1999)
Holotype – NMMNH P-18496, much of an articulated, but crushed skeleton (Heckert & Lucas, 1999).
Referred Material – TMM 31100-437, partial skull, paramedian, lateral, and ventral osteoderms, vertebrae, limb, and girdle material (Murry & Long, 1996; this study); NCSM 23168, much of a carapace (Heckert et al., 2015).
Age – Late Triassic, ?Carnian, Otischalkian (Lucas, 2010).
Occurrence – Colorado City Formation, Dockum Group, west Texas, U.S.A.; Pekin Formation, Newark Supergroup, North Carolina, U.S.A (Heckert & Lucas, 1999; Heckert et al., 2015).
Remarks – The holotype of Coahomasuchus kahleorum is distinctive, but poorly preserved, consisting of a flattened carapace and plastron concealing the majority of the vertebrae, the posteroventral corner of the skull, the posterior portion of the mandible, and a poorly preserved braincase, as well as articulated limb and girdle material (Heckert & Lucas, 1999; Desojo & Heckert, 2004). Fraser et al. (2006) documented the first occurrence of Coahomasuchus in the Pekin Formation of North Carolina providing a biostratigraphic correlation with the lower part of the Dockum Group of west Texas. Past phylogenetic analyses have recovered Coahomasuchus kahleorum as the sister taxon of Typothorax coccinarum and Redondasuchus reseri (Harris, Gower & Wilkinson, 2003 correction of the Heckert & Lucas, 1999 dataset); as the sister taxon of an unresolved clade containing Aetosauroides, Calyptosuchus, Aetosauroides, and Aetosaurus (Parker, 2007); and in an unresolved position closer to the base of Stagonolepididae (Desojo, Ezcurra & Kischlat, 2012). Moreover, the latter authors pruned Coahomasuchus from their final tree to achieve better resolution, thus the phylogenetic relationships of this taxon are far from resolved. However, a more recent analysis by Heckert et al. (2015), utilizing a modified version of the dataset in Parker (2007) & Desojo, Ezcurra & Kischlat (2012), recovered Coahomasuchus as a non-stagonolepidid aetosaur at the base of Aetosauria. In this analysis Coahomasuchus kahleorum is coded from the holotype as well as a newly referred specimen from the Dockum Group of Texas (TMM 31100-437) previously referred to as the ‘carnivorous form’ (Murry & Long, 1996), which was recovered from the same geographical area and stratum as the type specimen (Lucas, Hunt & Kahle, 1993).
It was suggested that the holotype of Coahomasuchus kahleorum may represent a skeletally immature individual (Parker, 2003). However, histological sampling of the referred specimen TMM 31100-437, which is in the same size class, indicates that TMM 31100-437 is close to skeletal maturity (S. Werning, personal communication, 2014). Thus, Coahomasuchus kahleorum is most likely not a juvenile individual of Lucasuchus hunti or Longosuchus meadei, both which are found in the same stratigraphic horizon and localities (e.g., Parker & Martz, 2010).
Key References – Heckert & Lucas (1999); Desojo & Heckert (2004).
Desmatosuchus spurensis (Case, 1920)
Holotype – UMMP 7476, skull, nearly complete carapace, articulated cervical and dorsal vertebral column, ilium (Case, 1922).
Referred Material – see Parker (2008b).
Age – Late Triassic, early to middle Norian, Adamanian (Ramezani et al., 2011; Ramezani, Fastovsky & Bowring, 2014; Parker & Martz, 2011).
Occurrence – Tecovas Formation, Dockum Group, Texas, U.S.A., Los Esteros Member, Santa Rosa Formation, Dockum Group, New Mexico, U.S.A., upper Blue Mesa Member, Chinle Formation, Arizona, U.S.A (Long & Murry, 1995; Parker, 2008b).
Remarks – First described from much of a carapace, and associated vertebral column as well as a skull, Desmatosuchus spurensis is a well-known aetosaurian from the Upper Triassic of the southwestern United States. Despite this, confusion exists regarding characters of the dorsal armor for referral of specimens. For example all of the specimens listed by Long & Ballew (1985) from Petrified Forest National Park actually pertain to paratypothoracines, and the osteoderm of Desmatosuchus figured by Lucas & Connealy (2008:26) for the Dawn of the Dinosaurs exhibit at the New Mexico Museum of Natural History and Science is actually and osteoderm of Calyptosuchus wellesi.
Gregory (1953a) synonymized Desmatosuchus spurensis with Episcoposaurus haplocerus, a form described by Cope (1892), and the taxon was known as Desmatosuchus haplocerus for several decades, until it was determined that Episcoposaurus haplocerus was actually a nomen dubium (Parker, 2008b; Parker, 2013) although this has not been accepted by all workers (e.g., Heckert, Lucas & Spielmann, 2012). New material from the Chinle Formation of Arizona demonstrated that previous carapace reconstructions for Desmatosuchus spurensis were erroneous and the body was broader than previous believed (Parker, 2008b).
Limb and pectoral girdle for Desmatosuchus spurensis is not known from the two best preserved specimens (UMMP 7476, MNA V9300), but Long & Murry (1995) assigned isolated material from the Placerias Quarry to the taxon, which has been utilized for studies including bone histology (de Ricqlès, Padian & Horner, 2003). Unfortunately Long & Murry (1995) did not discuss the evidence for these referrals, which have been questioned (Parker, 2005a; Parker, 2008b); however, utilizing field numbers from the Placerias Quarry it may possible to refer some of this material to Desmatosuchus spurensis. For this analysis Desmatosuchus spurensis is coded from UMMP 7476 and MNA V9300.
Key References – Case (1920); Case (1922); Long & Ballew (1985); Long & Murry (1995); Parker (2008b).
Desmatosuchus smalli (Parker, 2005b)
Holotype – TTU P-9024, almost complete skull and right mandible, partial pelvis, femora, nearly complete cervical armor and numerous osteoderms from the rest of the carapace (Parker, 2005b).
Referred Material – see Parker (2005b) & Martz et al. (2013).
Age – Late Triassic, mid-Norian, latest Adamanian and possibly earliest Revueltian (Ramezani et al., 2011; Martz et al., 2013).
Occurrence – Middle section of the Cooper Canyon Formation, Dockum Group, Texas, U.S.A.; ?Martha’s Butte beds, Sonsela Member, Chinle Formation, Arizona, U.S.A (Parker, 2005b; Martz et al., 2013).
Remarks – Small (1985) & Small (2002) described new material of Desmatosuchus from the Cooper Canyon Formation of Texas. Although he noted differences in the cranial material of the new material from the holotype of Desmatosuchus spurensis (UMMP 7476), he did not feel they were of taxonomic significance. In a revision of the genus Desmatosuchus, significant differences in the lateral armor were noted between the Cooper Canyon specimens and the type of Desmatosuchus spurensis (Parker, 2003). Combined with the cranial differences noted by Small (2002) the Cooper Canyon Formation material was assigned to a new species (Parker, 2005b). Further comments regarding this taxon including a novel reconstruction of the lateral cervical armor were provided by Martz et al. (2013).
One of the problems in utilizing the non-osteoderm postcranial material of Desmatosuchus smalli is that some of it may actually pertain to an undescribed specimen of Paratypothorax from the quarry (Martz, 2008). A detailed apomorphy-based study of the aetosaurian material from the Post Quarry is needed along with a reinvestigation of field collection data to clarify some of the taxonomic assignments of the material (Martz, 2008).
Other than the Texas material, Desmatosuchus smalli is known from only one single referred lateral osteoderm from the Chinle Formation of Arizona (MNA V697), which had been assigned to Desmatosuchus by Long & Ballew (1985) as a cervical lateral osteoderm. MNA V697 actually represents a dorsal lateral osteoderm and is assigned to Desmatosuchus smalli based on the ventrally recurved spine tip, which is an autapomorphy of Desmatosuchus smalli and does not occur in Desmatosuchus spurensis (Parker, 2005b). Although MNA V697 is listed as originating from a locality in the upper part of the Sonsela Member near Petrified Forest National Park (Long & Ballew, 1985), the locality data for this specimen are ambiguous. However, if correct this would represent the only Revueltian occurrence of Desmatosuchus (Parker & Martz, 2011).
The holotype of Desmatosuchus (=Episcoposaurus) haplocerus (ANSP 14688; Cope, 1892) consists chiefly of lateral and paramedian osteoderms of the cervical and anterior trunk regions (Gregory, 1953a; Parker, 2013). Unfortunately the tips of the spines on all of the trunk lateral osteoderms are broken away so the material cannot be differentiated between Desmatosuchus spurensis and Desmatosuchus smalli. Interestingly, the shape of the cervical lateral osteoderms as well as the ornamentation of the trunk paramedian osteoderms are more reminiscent of Desmatosuchus smalli rather than Desmatosuchus spurensis, but the data are not conclusive and therefore Desmatosuchus haplocerus is considered a nomen dubium (Parker, 2008b; Parker, 2013).
Key References – Small (1985); Small (2002); Parker (2005b); Martz et al. (2013).
Longosuchus meadei (Sawin, 1947)
Lectotype – TMM 31185-84b, skull and much of a postcranial skeleton (Sawin, 1947). See Parker & Martz (2010) for detailed discussion of the status of the type materials.
Referred Material – TMM 31185-84a, partial skull and postcranial skeleton. See Long & Murry (1995) for a complete list.
Age – Late Triassic, ?Carnian, Otischalkian (Lucas, 2010).
Occurrence – Colorado City Formation, Dockum Group, Texas, U.S.A (Hunt & Lucas, 1990).
Remarks – The Works Progress Administration program in the 1930s made vast collections of vertebrate fossils from a series of quarries in strata of the Dockum Group in Howard County, Texas. This included several skeletons of an aetosaurian that Sawin (1947) described as a new species of Typothorax, Typothorax meadei. Several subsequent authors recognized the distinctiveness of this material (Long & Ballew, 1985; Small, 1989; Murry & Long, 1989) and the species was placed in a new genus, Longosuchus, by Hunt & Lucas (1990). Sawin’s original description is thorough, but affected by a lack of good comparative material as well as the poor historical understanding of the taxonomic make-up of the Aetosauria available at the time of his initial work. Thus he incorrectly reconstructed the incomplete lower jaw and pelvis, which confused aetosaur in-group relationships until these details were later corrected by Walker (1961).
Most of the Otis Chalk material remains unprepared and numerous specimens, including partial skeletons (unpublished TMM documents), referable to Longosuchus meadei are in the Vertebrate Paleontology Lab (VPL) collections at the University of Texas (Austin) awaiting preparation.
An isolated fragment of a paramedian osteoderm from the Salitral Shale (Chinle Formation) of New Mexico, assigned to Longosuchus meadei by Hunt & Lucas (1990), possesses a beveled posterior edge and a radial ornament pattern and is more likely referable to Paratypothoracini, in particular Tecovasuchus (Irmis, 2008). Lateral osteoderms from the Argana Group of Morocco assigned to Longosuchus meadei by Lucas (1998b) appear to also represent a paratypothoracin as they are strongly dorsoventrally compressed and slightly recurved (Parker & Martz, 2010). Unfortunately this cannot be tested as these specimens have been reported as lost (S. Nesbitt, personal communication, 2013). Character state scorings for this study for Longosuchus were made solely utilizing the TMM material.
Key References – Sawin (1947); Hunt & Lucas (1990); Long & Murry (1995)
New basal dinosauromorph records from the Dockum Group of Texas, USA
Article number: 19.2.21A
Copyright Society for Vertebrate Paleontology, July 2016
Plain-language and multi-lingual abstracts
Submission: 1 May 2015. Acceptance: 1 June 2016
The basal dinosauromorphs constitute an important component of the Late Triassic Dockum dinosauromorph diversity. This study introduces nine hitherto unpublished hind limb elements of basal dinosauromorphs from Garza and Randall counties in Texas. Most of these specimens consist of femoral and tibial fragments referred to Dromomeron, whereas one complete and one partial fibula resemble the morphology of Marasuchus lilloensis, but with much larger size, and are assigned to undetermined dinosauriforms. The discovery of three Dromomeron specimens in the Tecovas Formation of the Palo Duro Canyon are particularly noteworthy because these are the lowest occurrences of Dromomeron romeri and Dromomeron gregorii, and the first reported biostratigraphic overlap of the two taxa. The extended taxonomic range of D. romeri challenges its suggested replacement for D. gregorii in the Late Triassic (Adamanian-Revueltian) faunal turnover.
Keywords: Late Triassic; Texas; Dockum; Dinosauromorpha; basal groups; biostratigraphy
Final citation: Sarıgül, Volkan. 2016. New basal dinosauromorph records from the Dockum Group of Texas, USA. Palaeontologia Electronica 19.2.21A: 1-16. https://doi.org/10.26879/564
The Dinosauromorpha (Benton, 1985) was one of the first clades erected after the introduction of cladistic approach to fossil archosaur studies. Since dinosaurs (including birds) comprise the majority of the dinosauromorphs, the terminology of "non-dinosaurian Dinosauromorpha" or "basal Dinosauromorpha" is useful to cover the close outgroups to the Dinosauria (e.g., Nesbitt et al., 2009; Langer et al., 2013). Early members of this outgroup like Lagerpetonchanarensis and Marasuchuslilloensis were discovered in the Chañares Formation (Romer, 1971, 1972), and the basal dinosauromorph fossil record remained restricted to Argentina for the following decades. Larger taxonomic diversity and wider dispersal patterns of these animals during the Mid and Late Triassic were revealed only after new discoveries from the early twenty-first century (e.g., Dzik, 2003; Irmis et al., 2007b; Ferigolo and Langer, 2007; Nesbitt et al., 2009, 2010; Kammerer et al., 2012; Peecook et al., 2013; Barrett et al., 2015).
Non-dinosaurian dinosauromorphs are also represented in the Upper Triassic terrestrial sediments of the southwestern North America. Technosaurus smalli (Chatterjee, 1984) and Eucoelophysis baldwini (Sullivan and Lucas, 1999) are the two contemporary silesaurids from the Dockum Group and the Chinle Formation, respectively, which were originally considered as dinosaurs (Nesbitt et al., 2007; Irmis et al., 2007a). Probably the most significant contribution to the North American dinosauromorph studies was the discovery of the new basal dinosauromorph taxon Dromomeron (Irmis et al., 2007b). Dromomeron has two species described so far in North America, D. romeri and D. gregorii, of which the holotypes were, respectively, collected from the Hayden Quarry of the Petrified Forest Member (Chinle Formation) and the Otis Chalk Quarry 3 of the Colorado City Formation (Dockum Group) (Irmis et al., 2007b; Nesbitt et al., 2009). Together with Lagerpetonchanarensis, the two Dromomeron species are grouped in the clade Lagerpetidae, the basal-most dinosauromorph clade (Nesbitt et al., 2009). Various other specimens have been assigned to Dromomeron romeri and to silesaurids coming from the Late Triassic sediments of the Eagle Basin in northwestern Colorado (Small, 2009; Langer et al., 2013).
All of the Dockum non-dinosaurian dinosauromorphs have been collected from the Texas outcrops of the Dockum Group, including the holotype of Technosaurus smalli (Chatterjee, 1984), the holotype of Dromomeron gregorii together with their paratypes and a referred femur (Nesbitt et al., 2009; Martz et al., 2013), and an undetermined dinosauriform tibia that has affinities with silesaurids (Nesbitt and Chatterjee, 2008; Martz et al., 2013). Complementing the previous studies, the description of the unpublished non-dinosaurian dinosauromorph specimens provided here represents the most recent work on the subject.
Geological Setting and Fossil Localities
Continental sedimentation in the southwestern portion of North America during the Late Triassic was greatly influenced by a huge palaeoriver system named Chinle-Dockum (Riggs et al., 1996), which deposited large volumes of fluvial sediments. Comprising the upstream portion of this ancient river system, the Dockum Group sediments cover vast areas in the western part of Texas and the eastern part of New Mexico (Figure 1). The classical lithostratigraphic framework for the Dockum Group of Texas is provided by Lehman and Chatterjee (2005). In this scheme, the Dockum fluvial deposition is represented by two sequences: a thinner (less than 80 m) older sequence consisting of the Santa Rosa and Tecovas formations and a thicker (over 150 m) younger sequence consisting of the Trujillo and Cooper Canyon formations. The two sequences are separated by the "Tr-4" unconformity (Lucas and Anderson, 1993a). In the same work, both sequences are characterized by a fining-upward sequence that starts with basal fluvial channel sandstones (i.e., Santa Rosa and Trujillo formations) and grade into floodplain mudstone and claystone deposits (i.e., Tecovas and Cooper Canyon formations).
Sedimentology and lithostratigraphy of the Dockum Group were recently revisited in southern Garza County (Martz, 2008; Martz et al., 2013). In this work, problems with the lithologic boundaries of the Cooper Canyon Formation were recognized for the first time. Originally coined as the Cooper Member (Chatterjee, 1986), this is a thick mudstone unit overlying a massive sandstone layer in southeastern Garza County, but the name Cooper was preoccupied by a Tertiary marl unit (Cooke and MacNeil, 1952). As its suggested lithostratigraphic equivalent in the Dockum Group of New Mexico, the name Bull Canyon Formation was proposed as a nomenclatural replacement (Lucas and Hunt, 1989) with priority over the Cooper Canyon Formation, for which a new type section was measured in southern Garza County, as a replacement for the former Cooper Member (Lehman et al., 1992). Accordingly, the Cooper Canyon and Bull Canyon formations were taken as interchangeable, until it was realized that the sandstone layer, the Boren Ranch Sandstone of Frelier (1987), on which the Cooper Member was originally described (Chatterjee, 1986; Lehman et al., 1992) was misidentified as the Trujillo Formation. Therefore, the Cooper Canyon Formation sensu Lehman (e.g., Lehman et al., 1992; Lehman, 1994a, 1994b; Lehman and Chatterjee, 2005) covers nearly all the Upper Triassic sediments above the Santa Rosa Formation in southern Garza County and cannot be used as a substitute for the Cooper Member. Martz (2008) employed the name Cooper Canyon Formation for the whole sequence overlying the Boren Ranch Beds, and subdivided it into lower, middle and upper units (Figure 2). Furthermore, it is firmly established that there is no evidence for a "Tr-4" unconformity in the Dockum Group of Texas (Martz, 2008) or in the Chinle Formation of Arizona (Woody, 2006; Parker and Martz, 2011).
In order to provide a fully-established lithostratigraphy, it is needed to reconcile the stratigraphic framework of the Dockum Group in southern Garza County with the rest of the Dockum outcrops in Texas and in New Mexico. Subsequent studies indicated that the lower, middle and upper units of the Cooper Canyon Formation in southern Garza County are equivalent to the Tecovas, Trujillo and Bull Canyon formations, respectively, in northern Garza, Crosby and in other northern Texas counties, and also in New Mexico (e.g., Cather et al., 2013; Martz et al., 2013) (Figure 2). Lithostratigraphic equivalency of the Dockum Group exposures south to Garza County in Texas is pending and needs more detailed studies; nonetheless, the Colorado City Formation is considered here as a correlate of the lower part of the Tecovas Formation, following the original works of Lucas and Anderson (1993a, 1993b, 1995) (Figure 2).
There are five non-dinosaurian dinosauromorph bearing fossil localities in the Dockum Group of Texas (Figure 1). Boren Quarry (MOTT 3869) and Post Quarry (MOTT 3624) are the two well-known fossil localities from the Garza County that produced a prolific vertebrate assemblage (e.g., Lehman and Chatterjee, 2005; Martz et al., 2013). In the stratigraphic column, they roughly represent the bottom and top levels of the lower unit of the Cooper Canyon Formation, respectively (Figure 2). Technosaurussmalli (TTU-P9021, holotype) is one of the previously published non-dinosaurian dinosauromorph specimens from the Post Quarry, together with a complete left femur of Dromomerongregorii (TTU-P11282) and a left dinosauriform tibia, which cannot be assigned to a specific taxon (TTU-P11127) (Chatterjee, 1984; Sereno, 1991; Hunt and Lucas, 1994; Irmis et al., 2007a; Nesbitt et al., 2007; Nesbitt and Chatterjee, 2008; Martz et al., 2013). Another fruitful locality from the Garza County is Headquarters South (MOTT 3898) that is placed at the upper part of the middle unit of the Cooper Canyon Formation, a direct correlative of the Trujillo Formation (Figure 2).
The Otis Chalk Quarry 3 (MOTT 2000) and the "lower Sunday Canyon" site are situated in Howard and Randall counties, respectively (Figure 1 and Figure 3). The "lower Sunday Canyon" site (34 57.530' N 101 42.099' W) is located in the eastern end of the Little Fox Canyon within the Palo Duro Canyon State Park where a small tributary joins the Sunday Creek almost due north of the locality (Figure 3). The "lower Sunday Canyon" Site belongs to the lowermost portion of the Tecovas Formation and is placed slightly lower than the Otis Chalk Quarries in the stratigraphic column (Bill Mueller, personal commun., 2015) (Figure 2 and Figure 4). The Otis Chalk quarries include beds from the middle portion of the Colorado City Formation (Lucas and Anderson, 1993b; Lucas et al., 1993) (Figure 2). The holotype of Dromomerongregorii (TMM 31100-1306) and associated paratypes were collected from the Otis Chalk Quarry 3 (Nesbitt et al., 2009).
MATERIALS AND METHODS
Nine basal dinosauromorph specimens are described here. Most of the specimens (TTU-P12537X, TTU-P12539X, TTU-P18331, TTU-P20046, TTU-P10546 and TTU-P19803) were collected from the southeastern outcrops of the Dockum Group of Texas and are reposited in the Museum of Texas Tech University (MoTTU). The dinosauromorph specimens from the Dockum Group of Texas (TTU-P specimens) were collected by Sankar Chatterjee and his crew, especially by Bill Mueller and Doug Cunningham, since the 1980s. The matrix covering the bone surfaces was removed mechanically in the MoTTU fossil preparation lab, and photographs were taken by a professional photographer (Bill Mueller). Three additional specimens (WTAMU-V-8301, WTAMU-V-8302 and WTAMU-V-8303) were loaned from West Texas A&M University (WTAMU). These three specimens were recovered from a prepared pile of assorted bone fragments collected from the same stratigraphic horizon by Gerald E. Schultz from Palo Duro Canyon State Park (i.e., the "lower Sunday Canyon" Site). The corresponding information about the collection site and catalog numbers for these specimens were also obtained from G.E. Schultz. The preliminary data for all the TTU-P and WTAMU specimens were included in an unpublished dissertation (Sarıgül, 2014), whereas the voucher information for type specimens of Dromomeron gregorii collected from the Otis Chalk Quarry 3 (MOTT 2000) refers to the original publication of Sterling Nesbitt and his colleagues (2009).
ARCHOSAURIA Cope 1869 sensu Gauthier, 1986
DINOSAUROMORPHA Benton, 1985
LAGERPETIDAE Arcucci 1986sensu Nesbitt et al., 2009
Dromomeron gregorii Nesbitt et al., 2009
Figure 4, Figure 5, Figure 6, Figure 7, Figure 8
Material. Proximal end of left femur (TTU-P18331) and distal end of left femur (TTU-P20046) from the Post Quarry (MOTT 3624); proximal end of right femur (WTAMU-V-8302) and proximal end of right tibia (WTAMU-V-8303) from the "lower Sunday Canyon" Site (Figure 4).
Description and remarks. Although the damaged tip of the femoral head obliterates the anteromedial tuber, the posteromedial tuber of TTU-P18331 is well preserved (Figure 5). The trochanteric fossa is situated posteriorly. The anterior trochanter with a trochanteric shelf is distinct as a rugose ridge on the anterolateral side. The fourth trochanter bulges out more distally on the posteroventral side. The enlarged posteromedial tuber is the most conspicuous lagerpetid character of TTU-P18331, and this specimen can be assigned to D. gregorii based on the presences of an anterior trochanter with a trochanteric shelf and a fourth trochanter.
As in TTU-P11282 (Martz et al., 2013) and in all the specimens referred to Dromomeron (Nesbitt et al., 2009), the muscle scar for the M. femorotibialis externus is expressed as a rounded fossa that is slightly damaged in TTU-P20046 (Figure 6). The distal condyles are substantially enlarged compared to D. romeri, and the characteristic intercondylar groove between the medial and the fibular condyles is also highly reduced due to the relative enlargement of the latter.
Despite the fact that the anteromedial tuber of the femoral head is obliterated, the enlarged posteromedial tuber is quite distinct on WTAMU-V-8302, which represents an unequivocal lagerpetid character together with the ventral emargination located on the anteromedial side (Figure 7). As an expansion of the anterior trochanter, the trochanteric shelf provides a prominent and rugose muscle attachment surface. Only the proximal part of the fourth trochanter is preserved. An anterior trochanter with a trochanteric shelf and a fourth trochanter are typical traits of D. gregorii to which WTAMU-V-8302 is assigned.
The proximal surface of WTAMU-V-8303 is triangular and significantly longer in the anteroposterior direction (Figure 8). The cnemial crest is prominent with a small ridge on its tip, which is slightly projected laterally. A small lateral depression that stands posterior to the cnemial crest does not produce a strict sense tibial notch, since it does not extend down along the shaft. The lateral condyle is ventrally deflected and smaller than the medial, which is expanded posteriorly with a tapering tip, as in D. gregorii. The preserved portion of the shaft is mediolaterally compressed.
Dromomeron romeri Irmis et al. 2007b
Figure 4, Figure 9, Figure 10
Material. Proximal end of right tibia (TTU-P12537X) from the Headquarters South locality (MOTT 3898); distal end of right femur (WTAMU-V-8301) from the "lower Sunday Canyon" Site (Figure 4).
Description and remarks. The proximal surface of TTU-P12537X is triangular and the three sides are almost equal in length (Figure 9). The cnemial crest is small but distinct, and its tip is slightly projected laterally. Posterior condyles are aligned, not pronounced and separated by a very shallow cleft. The lateral condyle is ventrally deflected, as diagnostic of Dromomeron (Nesbitt et al., 2009). Even though the medial condyle is obliterated, the posterior condyles seem to be aligned and sub-equal in size, a character that is more comparable to D. romeri (see Irmis et al., 2007b, figure 2; Nesbitt et al., 2009, figure 4).
The slender shaft of WTAMU-V-8301 significantly widens distally with a distinct ridge on the anteromedial side that establishes its affinity to Dromomeronromeri (Irmis et al., 2007b) (Figure 10). The distal end is much larger mediolaterally than anteroposteriorly. Condyles are small and gracile as noted for D. romeri (Nesbitt et al., 2009). Medial and fibular condyles are well separated by the intercondylar groove.
Figure 4, Figure 11
Material. Proximal end of right femur (TTU-P12539X) from the Headquarters South locality (MOTT 3898) (Figure 4).
Description and remarks. The head (or capitulum) of the proximal femur is broken but its hooked morphology of a typical lagerpetid is still traceable (Figure 11). The posteromedial tuber and adjacent trochanteric fossa (facies articularis antitrochanterica) are distinct. Right below the broken capitulum, a very faint emargination is present on the anterolateral side. The femoral shaft bears no anterior trochanter or a trochanteric shelf; the absence of which was previously noted for D. romeri and also for the smaller specimens of D. gregorii. Unfortunately, the inadequate preservation prevents a species-level assignment for this specimen.
DINOSAUROMORPHA Benton, 1985
DINOSAURIFORMES Novas, 1992
Gen et sp. indet.
Figure 4, Figure 12, Figure 13
Material. Left fibula (TTU-P10546) and distal end of left fibula (TTU-P19803) from the Boren Quarry (MOTT 3869) (Figure 4).
Description and remarks. TTU-P10546 represents a complete and mediolaterally compressed left fibula (Figure 12). The proximal articulation is elongated in the anteroposterior direction and slightly altered around the edges. The anterior end of the proximal portion tapers smoothly with the missing tip and slightly curved anteromedially. The lateral corner of the posterior end is projected posterolaterally. The shaft is straight, slender, transversely compressed and considerably damaged where the attachment site for the M. iliofibularis was to be detected. The anterior and lateral borders of the distal end are rounded, whereas the posterior corner is expanded posteroventrally and becomes a prominent process. The reflection of this process on the distal surface is a small pyramidal tuberosity. There are two attachment surfaces detected on the distal surface. The lateral attachment surface is concave, extending from the posterior side towards the slightly raised medial side. This helical morphology of the lateral attachment surface probably conforms to the unevenly raised articular facet formed by astragalus and calcaneum. The medial attachment surface is located on the posteromedial side of the distal surface and bears a slight bevel expressed on the anteromedial portion. This is visible in both medial and distal views, probably articulating with the lateral edge of the ascending process of the astragalus.
The proximal end and shaft of TTU-P10546 have the typical plesiomorphic dinosauromorph morphology (e.g., Langer, 2004; Nesbitt, 2011), and the main indicative of its affinity is probably the configuration of the distal end, which differs substantially from the lagerpetid and dinosaurian condition. The lagerpetid distal part of fibula is documented only in Lagerpetonchanarensis (Sereno and Arcucci, 1993), where both anterior and posterior ends of the convex distal articular surface are elongated and tapering over the tibia and astragalus, respectively. In Silesaurusopolensis and in basal saurischians, on the other hand, the distal surface of the fibula is flattened, and it is widened in anteroposterior (or anteromedial-posterolateral) axis with a distinct posterior tuber (e.g., Novas, 1993; Bonaparte et al., 1999; Langer, 2003; Dzik, 2003, figure 13C; Sereno et al., 2013). The basal ornithischian condition is even more peculiar; the fibula is reduced to a rod, where the distal portion is significantly twisted and fused to the tibia and astragalocalcaneum (e.g., Butler, 2005; Butler et al., 2010; Norman et al., 2011). Pisanosaurusmertii somewhat differs from the typical ornithischian condition by lacking the distal torsion and yielding an expanded distal end (Bonaparte, 1976; Norman et al., 2004).
The distal portion of the fibula of Marasuchuslilloensis offers the best comparison for TTU-P10546. Following the original description of Sereno and Arcucci (1994), the distal portion of the M. lilloensis fibula possesses an elliptical distal surface that is anteroposteriorly concave but mediolaterally convex with a distinct bevel on the medial side. A prominent tuber is present on the posterior corner, as described for TTU-P10546. In both taxa, the enlargement of the posterior tuberosity forms an asymmetrical distal end uniquely among ornithodirans (Nesbitt, 2011, character 345), differing from the posterodistal inclination on the flattened lateral border of the distal fibula seen in some basal saurischians like Eoraptor lunensis and Herrerasaurusischigualastenesis (Novas, 1993; Sereno et al., 2013). However, the medial edge is more rounded and the subdivision of the articular facets, if present, is less conspicuous in the distal part of the M. lilloensis fibula (Sereno and Arcucci, 1994, figure 12A). There is also a notable size difference between the two fibulae; the maximum length of the M. lilloensis fibula is about 7 cm (Sereno and Arcucci, 1994, table 5), whereas TTU-P10546 is more than three times longer (24 cm).
Since only the distal end of TTU-P19803 is preserved, the overall morphology and size is similar to those of TTU-P10546; only slightly more compressed mediolaterally (Figure 13). The lateral and anterior borders are rounded; a prominent process occurs on the posterior side that is followed by a small pyramidal tuberosity on the distal surface. The lateral and medial articulation surfaces seen in TTU-P10546 are not discernable in TTU-P19803; the tarsal articulation surfaces are represented by a single concave surface that is raised towards the medial side. This pattern is more compatible to that observed for M. lilloensis (Sereno and Arcucci, 1994, figure 12A).
CONCLUSION AND DISCUSSIONS
The dinosauromorph record of the Dockum Formation is important for understanding the distribution of ornithodirans in southern North America. Basal dinosauromorph specimens are presently unknown from the Dockum Group of New Mexico, leaving the Texas specimens as the only current record. Previously published specimens include the type specimens of Dromomeron gregorii (TMM 31100-1306 and associated paratypes), a left femur referred to D. gregorii (TTU-P11282), the holotype of the silesaurid Technosaurus smalli (TTU-P9021), and an isolated tibia possibly attributable to a silesaurid (TTU-P11127). The new non-dinosaurian dinosauromorph specimens described here are referred to Dromomeron romeri (TTU-P1253X7; WTAMU-V-8301), Dromomeron gregorii (TTU-P18331; TTU-P20046; WTAMU-V-8302; WTAMU-V-8303) and Dromomeron sp. (TTU-P12539X), with two remaining specimens (TTU-P10546; TTU-P19803) assigned to Dinosauriformes indet. based on their morphologic similarities to the fibula of Marasuchus lilloensis, regardless of their different sizes (Figure 4, Figure 5, Figure 6, Figure 7, Figure 8, Figure 9, Figure 10, Figure 11, Figure 12, Figure 13). Despite the fragmentary nature of the specimens, it can be concluded that the Late Triassic non-dinosaurian dinosauromorph fauna of the Dockum Group was taxonomically diverse.
The stratigraphic distribution of the Dromomeron specimens also deserves special attention. Previously, the fossil record of D. romeri was restricted to the upper levels of the Chinle Formation (i.e., the Petrified Forest Member), whereas D. gregorii was collected from the lower levels of both Dockum and Chinle sequences. Since the Dockum Group and the Chinle Formation are coeval and lithostratigraphically equivalent (e.g., Cather et al., 2013), it has been questioned whether there is a possible temporal succession of the two species (Nesbitt et al., 2009). The Dromomeron fossil record in the Dockum Group of Texas supports this theory. In Garza County, specimens of D. romeri are collected from the middle unit of the Cooper Canyon Formation (TTU-P12537X), whereas D. gregorii fossils are recorded only in the lower unit of the Cooper Canyon Formation (TTU-P18331; TTU-P20046) (Figure 2 and Figure 4). Similarly, the Otis Chalk Quarry 3 (MOTT 2000) where the Dromomeron gregorii was discovered is located in a relatively low stratigraphic position within the Dockum Group (Figure 2 and Figure 4). However, the discovery of specimens WTAMU-V-8301, WTAMU-V-8302 and WTAMU-V-8303 from the same stratigraphic horizon of the "lower Sunday Canyon" Site demonstrates the ranges of the two Dromomeron species overlap (Figure 4). These findings not only document the first co-occurrence of D. romeri and D. gregorii, but also represent the lowest occurrences of both Dromomeron species in North America. Moreover, WTAMU-V-8301 extends the taxon range of D. romeri down to the lower levels of the Tecovas Formation, a situation implying that D. romeri existed during most of the Late Triassic. Concurrently, the suggested replacement of D. gregorii by D. romeri at the Adamanian-Revueltian faunal transition (Parker and Martz, 2011) seems to reflect a sampling bias.
Now, the extension of the D. gregorii range gains importance to test the possible extinction of this taxon at the Adamanian-Revueltian boundary. The Post Quarry (MOTT 3624) and the Placerias Quarry of the Sonsela Member (Ramezani et al., 2014) are the two uppermost quarries from the Dockum Group and the Chinle Formation from where D. gregorii fossils were obtained (Nesbitt et al., 2009; Martz et al., 2013). Any future discovery of D. gregorii from the overlying sequences will show that this taxon was not affected by this Late Triassic faunal turnover, just like D. romeri. Therefore, Dromomeron taxon ranges should be refined with additional sampling in the future to further support these faunal turnover inferences.
The author owes special thanks to S. Chatterjee for providing the material and workspace, to G.E. Schultz and B. Mueller for sharing their personal data and observations, and to the Museum of Texas Tech University staff for their friendly attitude.
Arcucci, A.B. 1986. Nuevos materiales y reinterpretacion de Lagerpeton chanarensis Romer (Thecodontia, Lagerpetonidae nov.) del Triassic Medio de La Rioja, Argentina. Ameghiniana, 23:233-242.
Ash, S.R. 1976. Occurrence of the controversial plant fossil Sanmiguelia in the Upper Triassic of Texas. Journal of Paleontology, 50:799-804.
Barrett, P.M., Nesbitt, S.J., and Peecook, B.R. 2015. A large-bodied silesaurid from the Lifua Member of the Manda Beds (Middle Triassic) of Tanzania and its implications for body-size evolution in Dinosauromorpha. Gondwana Research, 27:925-931.
Benton, M.J. 1985. Classification and phylogeny of the diapsid reptiles. Zoological Journal of the Linnean Society, 84:97-164.
Bonaparte, J.F. 1976. Pisanosaurusmertii Casamiquela and the origin of the Ornithischia. Journal of Paleontology, 50:808-820.
Bonaparte, J.F., Ferigolo, J., and Riberio, A.M. 1999. A new early Late Triassic saurischian dinosaur from Rio Grande do Sul state, Brazil, p. 89-109. In Tomida, Y., Rich, T.H., and Vickers-Rich, P. (eds.), Proceedings of the Second Gondwanan Dinosaur Symposium. National Science Museum Monographs no.15, Tokyo.
Butler, R.J. 2005. The 'fabrosaurid' ornithischian dinosaurs of the Upper Elliot Formation (Lower Jurassic) of South Africa and Lesotho. Zoological Journal of the Linnean Society, 145:175-218.
Butler, R.J., Porro, L.B., Galton, P.M., and Chiappe, L.M. 2010. Anatomy and cranial functional morphology of the small-bodied dinosaur Fruitadens haagarorum from the Upper Jurassic of the USA. PLoS ONE, 7:e31556. doi:10.1371/journal.pone.0031556.
Cather, S.M., Zeigler, K.E., Mack, G.H., and Kelley, S.A. 2013. Towards standardization of Phanerozoic stratigraphic nomenclature in New Mexico. Rocky Mountain Geology, 48:101-124.
Chatterjee, S. 1984. A new ornithischian dinosaur from the Triassic of North America. Naturwissenschaften, 71:630-631.
Chatterjee, S. 1986. The Late Triassic Dockum vertebrates: their stratigraphic and paleobiogeographic significance, p. 139-150. In Padian, K. (ed.), The Beginning of the Age of Dinosaurs: Faunal Change Across the Triassic-Jurassic Boundary. Cambridge University Press, Cambridge.
Cooke, C.W. and MacNeill, F.S. 1952. Tertiary Stratigraphy of South Carolina. U.S. Geological Survey Professional Paper, 243:19-29.
Cope, E.D. 1869. Synopsis of the extinct Batrachia and Reptilia of North America, Part I. Transactions of the American Philosophical Society, 14:1-252.
Dzik, J. 2003. A beaked herbivorous archosaur with dinosaur affinities from the early Late Triassic of Poland. Journal of Vertebrate Paleontology, 23:556-574.
Ferigolo, J. and Langer, M.C. 2007. A Late Triassic dinosauriform from south Brazil and the origin of the ornithischian predentary bone. Historical Biology, 19:1-11.
Frelier, A.P. 1987. Sedimentology, fluvial paleohydrology, and paleogeomorphology of the Dockum Formation (Triassic), west Texas. Unpublished M.Sc. thesis, Texas Tech University, Lubbock, Texas, USA.
Gauthier, J.A. 1986. Saurischian monophyly and the origin of birds. Memoirs of California Academy of Sciences, 8:1-55.
Google Earth 2016. “Palo Duro Canyon State Park, in part” Map. Google, retrieved in 12 February 2016.
Hunt, A.P. and Lucas, S.G. 1994. Ornithischian dinosaurs from the Upper Triassic of the United States, p. 227-241. In Fraser, N.C. and Sues, H.-D. (eds.), In the Shadow of the Dinosaurs: Early Mesozoic Tetrapods. Cambridge University Press, Cambridge.
Irmis, R.B., Parker, W.G., Nesbitt, S.J., and Liu, J. 2007a. Early ornithischian dinosaurs: the Triassic record. Historical Biology, 19:3-22.
Irmis, R.B., Nesbitt, S.J., Padian, K., Smith, N.D., Turner, A.H., Woody, D., and Downs, A. 2007b. A Late Triassic dinosauromorph assemblage from New Mexico and the rise of dinosaurs. Science, 317:358-361.
Kammerer, C.F., Nesbitt, S.J., and Shubin, N.H. 2012. The first silesaurid dinosauriform from the Late Triassic of Morocco. Acta Palaeontologica Polonica, 57:277-284.
Langer, M.C. 2003. The pelvic and hind limb anatomy of the stem-sauropodomorph Saturnalia tupiniquim (Late Triassic, Brazil). PaleoBios, 23:1-30.
Langer, M.C. 2004. Basal Saurischia, p. 24-46. In Weishampel, D.B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria, second edition. University of California Press, Berkeley.
Langer, M.C., Nesbitt, S.J., Bittencourt, J.S., and Irmis, R.B. 2013. Non-dinosaurian Dinosauromorpha, p. 157-187. In Nesbitt, S.J., Desojo, J.B. and Irmis, R.B. (eds.), Anatomy, Phylogeny and Palaeobiology of Early Archosaurs and Their Kin, Geological Society Special Publications, 379.
Lehman, T.M. 1994a. The saga of the Dockum Group and the case of the Texas/New Mexico boundary fault. New Mexico Bureau Mines and Mineral Resources Bulletin, 150:37-51.
Lehman, T.M. 1994b. Save the Dockum Group! West Texas Geological Society Bulletin, 34:5-10.
Lehman, T.M. and Chatterjee, S. 2005. The depositional setting and vertebrate biostratigraphy of the Triassic Dockum Group of Texas. Indian Journal of Earth System Science, 114:325-351.
Lehman, T.M., Chatterjee, S., and Schnable, J.P. 1992. The Cooper Canyon Formation (Late Triassic) of western Texas. The Texas Journal of Science, 44:349-355.
Lucas, S.G. and Anderson, O.J. 1993a. Lithostratigraphy, sedimentation, and sequence stratigraphy of Upper Triassic Dockum Formation, West Texas, p. 55-65. In Crick, R.E. (ed.), 1993 Southwest Section Geological Convention, Transactions and Abstracts. American Association of Petroleum Geologists, Arlington.
Lucas, S.G. and Anderson, O.J. 1993b. Triassic stratigraphy in southeastern New Mexico and southwestern Texas, p. 231-235. In Love, D.W., Hawley, J.W., Kues, B.S., Adams, J.W., Austin, G.S., and Barker, J.M. (eds.), New Mexico Geological Society Field Conference Guidebook 44. New Mexico Geological Society, Albuquerque.
Lucas, S.G. and Anderson, O.J. 1995. Dockum (Upper Triassic) stratigraphy and nomenclature. West Texas Geological Society Bulletin, 34:5-11.
Lucas, S.G. and Hunt, A.P. 1989. Revised Triassic stratigraphy in the Tucumcari basin, east-central New Mexico, p. 150-169. In Lucas, S.G. and Hunt, A.P. (eds.) Dawn of the Age of Dinosaurs in the American Southwest. New Mexico Museum of Natural History, Albuquerque.
Lucas, S.G., Hunt, A.P., and Kahle, R. 1993. Late Triassic vertebrates from the Dockum Formation near Otis Chalk, Howard County Texas, p. 245-260. In Love, D.W., Hawley, J.W., Kues, B.S., Adams, J.W., Austin, G.S., and Barker, J.M. (eds.), New Mexico Geological Society Field Conference Guidebook 44. New Mexico Geological Society, Albuquerque.
Martz, J.W. 2008. Lithostratigraphy, chemostratigraphy, and vertebrate biostratigraphy of the Dockum Group (Upper Triassic), of southern Garza County, West Texas. Unpublished Ph.D. thesis, Texas Tech University, Lubbock, Texas, USA.
Martz, J.W., Mueller, B.D., Nesbitt, S.J., Stocker, M.R., Parker, W.G., Atanassov, M., Fraser, N., Weinbaum, J., and Lehane, J.R. 2013. A taxonomic and biostratigraphic re-evaluation of the Post Quarry vertebrate assemblage from the Cooper Canyon Formation (Dockum Group, Upper Triassic) of southern Garza County, western Texas. Proceedings of the Royal Society of Edinburgh, 103:339-364.
Nesbitt, S.J. 2011. The early evolution of archosaurs: relationships and the origin of major clades. Bulletin of the American Museum of Natural History, 352:1-292.
Nesbitt, S.J. and Chatterjee, S. 2008. Late Triassic dinosauriforms from the Post Quarry and surrounding areas, west Texas, USA. Neues Jahrbuch für Geologie und Paläontologie Abhandlungen, 249:143-156.
Nesbitt, S.J., Irmis, R.B., and Parker, W.G. 2007. A critical re-evaluation of the Late Triassic dinosaur taxa of North America. Journal of Systematic Palaeontology, 5:209-243.
Nesbitt, S.J., Irmis, R.B., Parker, W.G., Smith, N.D., Turner, A.H., and Rowe, T. 2009. Hindlimb osteology and distribution of basal dinosauromorphs from the Late Triassic of North America. Journal of Vertebrate Paleontology, 29:498-516.
Nesbitt, S.J., Sidor, C.A., Irmis, R.B., Angielczyk, K.D., Smith, R.M.H., and Tsuji, L.A. 2010. Ecologically distinct dinosaurian sister group shows early diversification of Ornithodira. Nature, 464:95-98.
Norman, D.B., Crompton, A.W., Butler, R.J., Porro, L.B., and Charig, A.J. 2011. The Lower Jurassic ornithischian dinosaurs Heterodontosaurus tucki Crompton and Charig, 1962; cranial anatomy, functional morphology, taxonomy, and relationships. Zoological Journal of the Linnean Society, 163:182-276.
Norman, D.B., Witmer, L.M., and Weishampel, D.B. 2004. Basal Ornithischia, p.325-334. In In Weishampel, D.B., Dodson, P. and Osmólska, H. (eds.), The Dinosauria, second edition. University of California Press, Berkeley.
Novas, F.E. 1992. Phylogenetic relationships of the basal dinosaurs, the Herrerasauridae. Palaeontology, 35:51-62.
Novas, F.E. 1993. New information on the systematics and postcranial skeleton of Herrerasaurus ischigualastensis (Theropoda: Herrerasauridae) from the Ischigualasto Formation (Upper Triassic) of Argentina. Journal of Vertebrate Paleontology, 13:400-423.
Parker, W.G. and Martz, J.W. 2011. The Late Triassic (Norian) Adamanian-Revueltian tetrapod faunal transition in the Chinle Formation of Petrified Forest National Park, Arizona. Earth and Environmental Science Transactions of the Royal Society of Edinburgh, 101:231-260.
Peecook, B.R, Sidor, C.A., Nesbitt, S.J., Smith, R.M., Steyer, J.S., and Angielczyk, K.D. 2013. A new silesaurid from the Upper Ntawere Formation of Zambia (Middle Triassic) demonstrates the rapid diversification of Silesauridae (Avemetatarsalia, Dinosauriformes). Journal of Vertebrate Paleontology, 33:1127-1137.
Ramezani, J., Fastovsky, D.E., and Bowring, S.A. 2014. Revised chronostratigraphy of the lower Chinle Formation strata in Arizona and New Mexico (USA): High precision U-Pb geochronological constraints on the Late Triassic evolution of dinosaurs. American Journal of Science, 314:981-1008.
Riggs, N.R., Lehman, T.M., Gehrels, G.E., and Dickinson, W.R. 1996. Detrital zircon link between headwaters and terminus of the Upper Triassic Chinle-Dockum Paleoriver system. Science, 273:97-100.
Romer, A.S. 1971. The Chañares (Argentina) Triassic reptile fauna X. Two but new incompletely known long-limbed pseudosuchians. Breviora, 378:1-10.
Romer, A.S. 1972. The Chañares (Argentina) Triassic reptile fauna. XV. Further remains of the thecodonts Lagerpeton and Lagosuchus. Breviora, 390:1-7.
Sarıgül, V. 2014. Anatomy of the Late Triassic Dinosauromorphs from the Dockum Group of Texas: Their Biostratigraphic, Paleobiogeographic and Evolutionary Significance. Unpublished Ph.D. thesis, Texas Tech University, Lubbock, Texas, USA.
Sereno, P.C. 1991. Lesothosaurus, “fabrosaurids,” and the early evolution of Ornithischia. Journal of Vertebrate Paleontology, 11:168-197.
Sereno, P.C. and Arcucci, A.B. 1993. Dinosaurian precursors from the Middle Triassic of Argentina: Lagerpeton chanarensis. Journal of Vertebrate Paleontology, 13:385-399.
Sereno, P.C. and Arcucci, A.B. 1994. Dinosaurian precursors from the Middle Triassic of Argentina: Marasuchus lilloensis, gen. nov. Journal of Vertebrate Paleontology, 14:53-73.
Sereno, P.C., Martinez, R.N. and Alcober, O.A. 2013. Osteology of Eoraptor lunensis (Dinosauria, Sauropodomorpha). Journal of Vertebrate Paleontology, 32:83-179.
Small, B.J. 2009. A Late Triassic dinosauromorph assemblage from the Eagle Basin (Chinle Formation), Colorado, U.S.A. Journal of Vertebrate Paleontology, 29:182A.
Sullivan, R.M. and Lucas, S.G. 1999. Eucoelophysis baldwini, a new theropod dinosaur from the Upper Triassic of New Mexico, and the status of the original types of Coelophysis. Journal of Vertebrate Paleontology, 19:81-90.
Woody, D.T. 2006. Revised stratigraphy of the lower Chinle Formation (Upper Triassic) of Petrified Forest National Park, Arizona, p. 17-45. In Parker, W.G., Ash, S.R., and Irmis, R.B. (eds.), A Century of Research at Petrified Forest National Park, Museum of Northern Arizona Bulletin No. 62.