This article was downloaded by: [Stony Brook University]On: 30 November 2012, At: 05:34Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Journal of Systematic PalaeontologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tjsp20

Revision of the sauropod dinosaurs of the LowerCretaceous Trinity Group, southern USA, with thedescription of a new genusMichael D. D’Emic a ba Museum of Paleontology and Department of Geological Sciences, University of Michigan,1109 Geddes Avenue, Ann Arbor, MI, 48109–1079, USAb Department of Geology and Geography, Georgia Southern University, Box 8149,Statesboro, GA, 30460, USAVersion of record first published: 30 Nov 2012.

To cite this article: Michael D. D’Emic (2012): Revision of the sauropod dinosaurs of the Lower Cretaceous Trinity Group,southern USA, with the description of a new genus, Journal of Systematic Palaeontology, DOI:10.1080/14772019.2012.667446

To link to this article: http://dx.doi.org/10.1080/14772019.2012.667446

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form toanyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses shouldbe independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly inconnection with or arising out of the use of this material.

Journal of Systematic Palaeontology, iFirst 2012, 1–20

Revision of the sauropod dinosaurs of the Lower Cretaceous Trinity Group,southern USA, with the description of a new genus

Michael D. D’Emic∗

Museum of Paleontology and Department of Geological Sciences, University of Michigan, 1109 Geddes Avenue, Ann Arbor,MI 48109–1079, USA

(Received 5 December 2011; accepted 6 December 2011)

Early Cretaceous sauropods were among the first dinosaurs discovered in North America, but several aspects of their taxonomyand evolution remain poorly understood. Much of this ambiguity stems from lack of anatomical overlap among taxa and the125-year-long taxonomic confusion surrounding the sauropods Astrodon and Pleurocoelus. New discoveries have begun toremedy the first problem, but a lack of autapomorphies in their holotypes and skeletal associations among their hypodigmsrenders Astrodon johnstoni, Pleurocoelus altus and Pleurocoelus nanus nomina dubia. Herein I examine the affinities ofsauropods from the Trinity Group of Texas and Oklahoma previously referred to as ‘Pleurocoelus’ or ‘Astrodon’. Some ofthis material currently comprises the genera Paluxysaurus and Sauroposeidon from laterally equivalent strata in Texas andOklahoma, respectively. Although representative individuals of Paluxysaurus are only two-thirds the size of Sauroposeidon,bone histology of Paluxysaurus indicates that the individuals from the type locality were not near adult size. The similarprovenance, lack of morphological differences, and shared unique features support referral of Paluxysaurus to Sauroposeidon.Other sauropod remains from the Trinity Group are not referable to ‘Pleurocoelus’, ‘Astrodon’ or Sauroposeidon. Some ofthese remains comprise the holotype of Astrophocaudia slaughteri gen. et sp. nov., a basal titanosauriform diagnosed by ahyposphene–hypantrum system in the caudal vertebrae. A sauropod hind limb previously referred to ‘Pleurocoelus’ is insteadreferable to Cedarosaurus weiskopfae based on shared features of the pes. Cladistic analysis indicates that Astrophocaudiaand Sauroposeidon are members of Somphospondyli, whereas Cedarosaurus is a brachiosaurid. The Trinity Group of Texasand laterally equivalent Antlers Formation of Oklahoma exhibit similar dinosaur faunas at the generic and specific levels tothe Cloverly Formation of Wyoming. This homogeneity with respect to latitude stands in marked contrast to the latitudinalvariation in dinosaur communities that developed later in the Cretaceous.

http://zoobank.org/urn:lsid:zoobank.org:pub:FE82D372-7ADA-4870-9572-3A3F607D39CE

Keywords: sauropod; titanosauriform; Pleurocoelus; Astrodon; Trinity; Cretaceous

Introduction

Sauropod dinosaurs were common and diverse mega-herbivores in many Mesozoic ecosystems. Their statusas the largest land animals that ever evolved, as well astheir unique body plan with a long neck and tail set onan elephantine body, has fuelled studies of their evolutionand palaeobiology (e.g. Wilson 2002; Sander & Clauss2008; Sander et al. 2010; Mannion et al. 2011). Theirextreme size has also hindered such studies, contributingto the incompleteness of most fossil sauropod individualsand the difficulty of working with them (Mannion &Upchurch 2010). Nonetheless, better fossil material andmany systematic revisions in the last decade have greatlyincreased the amount of information available to sauropodresearchers (e.g. McIntosh 2005; Rose 2007; Taylor 2009;Carballido et al. 2011; Mannion 2011).

∗Present address: Department of Geology and Geography, Georgia Southern University, Box 8149, Statesboro, GA 30460, USA.Email: mdemic@umich.edu

The first sauropod described from North America wasfound in the Arundel Formation of Maryland (Johnston1859), which was originally thought to be Late Jurassicin age (e.g. Marsh 1897), but is now recognized asEarly Cretaceous (e.g. Ostrom 1970). The earlier-namedAstrodon Leidy, 1865 and the later-named PleurocoelusMarsh, 1888 were based on isolated, incomplete type spec-imens, to which later-discovered specimens from Marylandand elsewhere were referred. The validity, hypodigms andinferred affinities of these two genera have varied widelysince their naming over a century ago (Table 1). Sauropodsfrom several other regions of the world have been referredto Pleurocoelus and Astrodon (e.g. Langston 1974), butnew discoveries and analyses have challenged some of thesereferrals (e.g. Wedel et al. 2000a; Rose 2007). The contro-versial or ambiguous taxonomy of many fragmentary EarlyCretaceous North American sauropods and the discovery

ISSN 1477-2019 print / 1478-0941 onlineCopyright C© 2012 The Natural History Museumhttp://dx.doi.org/10.1080/14772019.2012.667446http://www.tandfonline.com

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

2 M. D. D’Emic

Table 1. Novel genus–species combinations of Astrodon and Pleurocoelus in order of publication, with taxonomic action taken at thetime of naming.

Author Taxa named Action

Johnston 1859 Astrodon New genusLeidy 1865 Astrodon johnstoni New speciesMarsh 1888 Pleurocoelus nanus, Pleurocoelus altus New genus, two new speciesMarsh 1896 Pleurocoelus montanus New speciesLydekker 1890 Pleurocoelus valdensis New speciesMarsh 1897 Pleurocoelus suffosus New combinationHatcher 1903 Astrodon suffosus New combinationGilmore 1921 Astrodon nanus, Astrodon altus New combinationsde Lapparent & Zbyszewski 1957 Astrodon pusillus, Astrodon montanus New species; new combinationKingham 1962 Astrodon altithorax, Astrodon atalaiensis,

Astrodon brancai, Astrodon fraasiNew combinations

of substantial new material prompt a comprehensive re-evaluation of Early Cretaceous North American sauropods.

Cretaceous sauropods of North America

The survival of sauropods into the Cretaceous of NorthAmerica was confirmed by Larkin (1910), who reportedthe discovery of a sauropod coracoid in the Early Creta-ceous Antlers Formation of Oklahoma (a lateral equivalentof the Trinity Group of Texas). Several decades later, morecomplete excavations and exploration in the Trinity Groupwere undertaken by the Field Museum of Natural History,Harvard University, and Southern Methodist University.These excavations yielded the remains of a diverse verte-brate fauna, including sauropods (Bilelo 1969). AdditionalTrinity Group sauropod remains were reported from Texas,including a hind limb from the Paluxy Formation at WalnutCreek in Wise County (Bilelo 1969) and a partial skele-ton from the underlying Glen Rose Formation near Blanco(Langston 1974; Tidwell & Carpenter 2003).

The Trinity Group sauropods collected by the FieldMuseum teams and reported by Bilelo (1969) were referredto Pleurocoelus sp. by Langston (1974), a taxonomic deci-sion that was followed by some subsequent authors (e.g.Gallup 1975, 1989; McIntosh 1990), but not more recentones (e.g. Gomani et al. 1999; Upchurch et al. 2004;Carpenter & Tidwell 2005; Rose 2007). Authors have alsodisagreed on the phylogenetic affinities of the Early Creta-ceous Texan materials referred to Pleurocoelus, with somereferring them to Brachiosauridae, and others regardingthem as more derived (e.g. McIntosh 1990; Salgado et al.1995). Fig. 1 depicts a cladogram of the major cladesand taxa discussed in this paper based on recent analyses(Upchurch et al. 2004; Rose 2007).

The Early Cretaceous North American sauropod fossilrecord has improved greatly in the last decade as severalmore complete skeletons have been described from Early

Cretaceous strata, especially in the Trinity Group andAntlers Formation of Texas and Oklahoma, respectively(Fig. 2). Two monospecific genera have been named fromthese strata: Sauroposeidon proteles Wedel et al., 2000aand Paluxysaurus jonesi Rose, 2007. Comparatively littlehas been said of the taxonomy or phylogeny of other Trin-ity sauropod fossils, although a partial sauropod skeletonreported by Bilelo (1969; SMU 61732) has been mentionedas a genus distinct from other Early Cretaceous Trinitysauropods (e.g. Tidwell et al. 1999; Wedel et al. 2000b).Here I provide a name, diagnosis, description and compar-isons for this skeleton. I then examine the anatomy andaffinities of the hind limb from the Paluxy Formationdescribed by Gallup (1975, 1989), and evaluate the validityand diagnoses of other Early Cretaceous sauropods fromthe Trinity Group. Finally, I evaluate the similarity of theserevised Trinity Group faunas to those of similar age, suchas the Cloverly Formation of Wyoming and Montana andthe Cedar Mountain Formation of Utah.

Note on the taxonomy of Astrodon andPleurocoelusThe taxonomies of Astrodon and Pleurocoelus have variedwidely according to different authors (Table 1). The typeseries of Astrodon johnstoni consists of two teeth (YPM

Figure 1. Simplified cladogram depicting the relationships ofrelevant sauropod clades. Based on the phylogeny of Upchurchet al. (2004).

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

Trinity Group sauropods 3

798) from the Arundel Formation in Maryland, USA(Johnston 1859; Leidy 1865). The type series of P. altusconsists of a left tibia and fibula from one individual(USNM 4971), and the type series of P. nanus consistsof a cervical (USNM 5678), a dorsal (USNM 4968), asacral (USNM 4969) and a caudal (USNM 4970) vertebralcentrum, all of which are unfused to their neural arches andribs, and were not recovered with the type series (Marsh1888; Lull 1911). The type series of Pleurocoelus nanusis the appropriate size to belong to one individual, but theexact provenance of each bone is uncertain. These bonescould represent a chimera of individuals or taxa (Hatcher1903).

Referrals of new material to Pleurocoelus and Astrodonand revisions of the Arundel Formation fauna drasticallychanged the taxonomy of its sauropods during the 20thcentury. Hatcher (1903: 12) believed that the ArundelFormation remains were found “. . . in essentially, andperhaps identically, the same locality and horizon . . .”, andgiven the lack of substantial variation among those sauro-pod remains suggested that only one genus and species,Astrodon johnstoni, was present. Lull (1911) agreed withHatcher (1903) that Astrodon and Pleurocoelus altus repre-sented the same taxon, but thought that Pleurocoelus nanuswas a different taxon, based on the relative frequency ofsmall and large sauropod bones in the formation. Gilmore(1921) reviewed the Arundel Formation fauna and agreedwith former workers that more than one species of sauro-pod was present, but differed from them in assigningtaxonomic value to the observed differences among thematerial, referring all of the specimens to a single genus,Astrodon, and creating the new species Astrodon nanusand A. altus. Gilmore’s (1921) taxonomy was preferred forseveral decades; for example, subsumation of Pleurocoelusinto Astrodon was followed by the influential work of Romer(1956). Kingham (1962) also referred all of the Marylandspecies to Astrodon; in addition, he referred several speciesof Brachiosaurus to Astrodon, creating A. atalaiensis, A.brancai and A. altithorax. Kingham (1962) also named A.fraasi; in resurrecting this species, he overturned Janensch’s(1929) synonomy of Brachiosaurus fraasi with B. brancai.Kingham’s (1962) assignments have not been followed bysubsequent authors (e.g. McIntosh 1990; Upchurch et al.2004; Carpenter & Tidwell 2005).

Carpenter & Tidwell (2005) redescribed much of theArundel Formation sauropod material and concludedthat the low degree of variability in the available skeletalelements indicated that only one species was present,which would be Astrodon johnstoni based on priority.They presented 10 autapomorphies in their diagnosis of A.johnstoni (based on all of the sauropod material from theArundel Formation): (1) supraoccipital crest low and wide;(2) tall, narrow foramen magnum; (3) short, wide cameratecervical vertebrae with very large pleurocoels; (4) deeppleurocoels in the dorsal vertebrae; (5) deep pleurocoels

in the sacral vertebrae; (6) posterior sacral vertebrae witha prominent groove on the ventral surface; (7) anteriorcaudal vertebrae with short centra; (8) coracoid thick withprominent lip; (9) radius with distinct oblique ridge; and(10) two small posterodistal condyles on the radius. Manyof these features are indistinguishable compared to othersauropods such as Camarasaurus (characters 1, 2, 7–10;Osborn & Mook 1921; Ostrom & McIntosh 1966, pls 46,51; Madsen et al. 1995, fig. 23), and/or are related to thejuvenile nature of the material (characters 1, 3–5; Wedelet al. 2000b; Curry Rogers & Forster 2004).

None of these autapomorphies deals with teeth or thecrus, which are the holotypic elements of Astrodon john-stoni and Pleurocoelus altus, respectively. Like Astrodonand P. altus, the syntype vertebrae of P. nanus bear nounique features, making all three taxa nomina dubia.

Carpenter & Tidwell (2005) presented an example of the‘laissez-faire’ approach outlined by Wilson et al. (2009), inwhich undocumented field associations are assumed, forc-ing conspecificity where it might not be present. Employingthis approach in the case of the Arundel Formation involves:(1) ascribing no taxonomic meaning to the range of varia-tion in the sample; (2) assuming the penecontemporaneousnature of the various outcrops of the Arundel Formation;and (3) inferring ontogenetic transformations amongspecimens in the sample. These assumptions are large. Forthe taxonomy of Astrodon and Pleurocoelus, I advocate theopposite viewpoint, the ‘tabula rasa’ approach (Wilson et al.2009), in which taxa whose holotypes are non-diagnosticare deemed nomina dubia. I employ this approach in theArundel Formation because of the especially fragmentaryand non-diagnostic type species involved, and the paucityof provenance or quarry data for nearly all specimens.

However, the ventral groove on the sacral vertebra (char-acter 6 of Carpenter & Tidwell 2005) may be unique (inthe referred specimen USNM 5666, pers. obs. 2010), butcannot be observed in the syntype specimen of Pleuro-coelus nanus (USNM 4969). Indeed, other sauropod mate-rial from the Arundel Formation does bear unique features,including a groove below the ectopterygoid/palatine articu-lar facets on the maxilla and laterally curved pedal unguals(pers. obs. 2010). However, the lack of association amongthese materials and/or the holotype of P. nanus and P. altuslimits their systematic utility. Furthermore, some bones inthe sample display marked variation, such as the deeplydivided versus flat distal ends of the first metatarsals (pers.obs. 2010). Future discoveries may yield associated skele-tons in the Arundel Formation bearing these or other diag-nostic features.

Anatomical abbreviationsacdl: anterior centrodiapophyseal lamina; acet: acetabu-lum; bo: boss; CD: caudal vertebra; cprl: centroprezy-gapophyseal lamina; haa: hypantrum articulation surface;hy: hyposphene; ncj: location of fused neurocentral

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

4 M. D. D’Emic

junction; ng: nail groove; pcdl: posterior centrodiapophy-seal lamina; pl: pleurocoel; pn fo: pneumatic fossa;pocdf: postzygapophyseal centrodiapophyseal fossa; podl:postzygodiapophyseal lamina; posdf: postzygapophysealspinodiapophyseal fossa; poz: postzygapophysis; prcdf:prezygapophyseal centrodiapophyseal fossa; prdl: prezy-godiapophyseal lamina; prsl: prespinal lamina; prz:prezygapophysis; sdf: spinodiapophyseal fossa; sprf:spinoprezygapophyseal fossa; sprl: spinoprezygapophy-seal lamina; spol: spinopostzygapophyseal lamina; tp:transverse process; tprl: intraprezygapophyseal lamina; tu:tubercle wf: wear facet.

Institutional abbreviationsASDM: Arizona-Sonora Desert Museum, Arizona, USA;DMNS: Denver Museum of Nature and Science, Denver,USA; FMNH: Field Museum of Natural History, Chicago,USA; FWMSH: Fort Worth Museum of Science andHistory, Fort Worth, USA; HMN MB. R: HumboldtMuseum fur Naturkunde, Berlin, Germany; OMNH: Okla-homa Museum of Natural History, Norman, USA; SMU:Southern Methodist University, Dallas, USA; TMM: TexasMemorial Museum, Austin, USA; USNM: United StatesNational Museum (Smithsonian Institution), Washington,DC, USA.

Systematic palaeontology

Dinosauria Owen, 1842Sauropoda Marsh, 1878

Neosauropoda Bonaparte, 1986Titanosauriformes Salgado et al., 1997

Astrophocaudia gen. nov.

Type species. Astrophocaudia slaughteri sp. nov.

Astrophocaudia slaughteri sp. nov.(Figs 3–11)

1974 ‘Pleurocoelus’ sp. Langston: 85, figs 5, 6a.1997 ‘Pleurocoelus’ Salgado & Calvo: 44, fig. 9.

Holotype. SMU 61732 and 203/73655; a tooth (SMU203/73655), two cervical vertebrae, fragments of dorsalvertebrae, 24 caudal vertebrae, approximately 20 fragmen-tary dorsal ribs, two chevrons, a distal scapular blade, partof a right ilium, and numerous fragments. The two associ-ated teeth mentioned by Rose (2007) are prezygapophysesof middle caudal vertebrae; these were glued onto theirvertebrae during preparation in August 2009. The appro-priate size and lack of duplication of elements suggest thatonly one sauropod individual was present in the quarry.

Diagnosis. Autapomorphies include: anterior-middlecaudal vertebrae with planar hyposphene–hypantrumarticulations set off from zygapophyses; anterior-middlecaudal vertebrae with prespinal lamina contactingintraprezygapophyseal lamina. Unlike Astrophocaudia,Sauroposeidon proteles (see below) has wide spino-prezygapophyseal fossae that are bounded by stronglydeveloped spinoprezygapophyseal laminae in anterior

Figure 2. A, type locality of Astrophocaudia slaughteri in Wise County, Texas, USA (star), with locations of other named Early CretaceousNorth American sauropods (dots); arrow indicates the direction of north during the Albian (taken from www.paleodb.org); scale bar =500 km. B, stratigraphical position of Astrophocaudia slaughteri relative to other Early Cretaceous North American sauropods; placementof the abbreviated name indicates the most likely age for the taxon, and the dotted lines represent age uncertainty. Abbreviations: Aby,Abydosaurus; Arun, Arundel Formation sauropod material, including ‘Astrodon’ and ‘Pleurocoelus’; As, Astrophocaudia gen. et. sp. nov.;Blanco, TMM 40435, partial sauropod skeleton from the Glen Rose Formation (Tidwell & Carpenter 2003); Bro, Brontomerus; Ced,Cedarosaurus; Clo, Cloverly Formation sauropod material (Ostrom 1970); FMNH, FMNH PR 977, hind limb of a sauropod from thePaluxy Formation referable to Cedarosaurus weiskopfae (Gallup 1989); Plx, Paluxysaurus; Sau, Sauroposeidon (holotype only); Son,Sonorasaurus. Tooth figures indicate horizons that preserve indeterminate sauropod teeth.

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

Trinity Group sauropods 5

Figure 3. Holotypic tooth of Astrophocaudia slaughteri (SMU203/73655) in: A, labial; B, ?mesial; C, lingual; D, ?distal; E,occlusal; and F, proximal views. Orientation is uncertain becauseof uncertainty in tooth position within the jaw. Scale bar = 1 cm.

caudal vertebrae. Giraffatitan, Abydosaurus, Venenosaurusand Cedarosaurus are differentiated from Astrophocaudiaby the presence of sporadically distributed fossae belowthe transverse processes in caudal vertebrae and forward-leaning neural spines in the latter two genera. Sonorasaurusis differentiated from Astrophocaudia by the reduction ofthe spinopostzygapophyseal laminae in the anterior-middlecaudal vertebrae that cause each postzygapophysis toproject far beyond the posterior margin of the neural spine.The ilium of Astrophocaudia is dissimilar from that ofBrontomerus mcintoshi in its gently curving preacetabularprocess in dorsal view.

Derivation of name. A-, non- (Greek); stropho-, twist-ing or turning (Greek); caud-, tail (Greek). The name is inreference to the tightly articulating hyposphene–hypantrumsystem in the anterior and middle caudal vertebrae, which

Figure 4. Holotypic middle to posterior cervical vertebra ofAstrophocaudia slaughteri (SMU 61732) in lateral view. Dashedlines indicate missing bone. Scale bar = 10 cm.

also resembles a star (astron; Greek) in posterior view. Thegeneric name is also a reference to Astrodon, the first EarlyCretaceous North American sauropod. The specific namehonours Dr Robert H. Slaughter, who excavated the speci-men in the 1960s.

Locality, horizon and age. The holotype comes fromWalnut Creek, southeast of Decatur, Wise County, Texas,33◦09′ N, 97◦34′ W (Thurmond 1974; Fig. 2). This localityhas also yielded the semionontid fish Lepidotes (somescales of which were found in contact with the bones ofAstrophocaudia), a theropod claw (SMU 62723), a thero-pod squamosal (SMU 61741), and the turtle Naomichelys.Bilelo (1969) reported the sauropod as coming fromWall, Texas. There is a city called ‘Wall’ in Texas, butit is several hundred kilometres from Wise County. DrWann Langston visited the Walnut Creek sauropod sitein 1984 (pers. obs. of field notes housed at SMU, 2009),and noted the location as occurring “5.6 miles south and0.9 miles west of Decatur” (which is in Wise County). Inaddition, Thurmond (1974, appendix) gave latitude andlongitude coordinates for a “Walnut Creek B local fauna”that includes a sauropod and is in accordance with thelocality given in Langston’s notes. SMU 61732 comesfrom the Trinity Group, uppermost part of the middle unitof the Paluxy Formation (Thurmond 1974), which is lowerAlbian (112.2–106 Ma) in age (Jacobs & Winkler 1998).

Description and comparisons

In the following description, abbreviations for vertebrallaminae and fossae follow Wilson (1999) and Wilsonet al. (2011), respectively. Comparisons are made whererelevant; other comparative information can be found inthe differential diagnosis or Discussion below.

ToothSMU 203/73655 (Fig. 3) is missing its enamel, but detailsof its morphology and wear are still evident. It is uncer-tain what part of the jaw the tooth comes from. The toothis 17 mm long, 5.6 mm wide labiolingually, and 6.4 mmwide mesiodistally. It is very slightly spatulate, with amidline longitudinal groove on its lingual face and a broadarch across the labial face (Fig. 3). The crown–root junc-tion is not clearly preserved, precluding precise calculationof the slenderness index (mesiodistal crown width/crownheight; Upchurch 1998). However, the slenderness indexwould be somewhere around 2.7, which is intermediatebetween values observed in narrow-crowned diplodocoidsand titanosaurs, and broad-crowned basal titanosauriformsand more primitive sauropods (Chure et al. 2010, fig.5). The slenderness index of Astrophocaudia is close tothe average value observed for Abydosaurus (Chure et al.2010). Its wear facet is nearly planar and is angled either

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

6 M. D. D’Emic

Figure 5. Selected vertebrae of the holotypic anterior-middle caudal vertebral series of Astrophocaudia slaughteri (SMU 61732) in: A,left lateral; B, anterior; and C, posterior views. Numbers below each vertebra indicate its likely position in the caudal sequence. The 10thcaudal vertebra is reversed in A. Dashed lines indicate missing bone. Scale bar = 10 cm.

mesially or distally, but not labiolingually. There is a veryslight axial twist towards the apex of the tooth, though notas great as in the upper teeth of Abydosaurus (Chure et al.2010).

Presacral vertebraeOne partial middle to posterior cervical vertebra and frag-ments of several posterior cervical or anterior dorsal verte-brae are accessioned as part of SMU 61732. Little can besaid about the morphology of the more fragmentary verte-brae, so description will focus on the middle to posteriorcervical vertebra.

The mid-cervical vertebra has been sheared dorsally onits right side, exposing the ventral face of the centrum inright lateral view (Fig. 4). Its centrum is 48 cm long, andthe partially fused neurocentral junction is still visible ante-riorly and posteriorly as well-defined furrows offsetting theneural arch pedicles. The posterior centrum is 24 cm wideand 15 cm tall but is sheared, so the actual measurementswould have been closer to 22 × 17 cm. The elongation index(EI = centrum length/posterior centrum height; Upchurch

1998) cannot be determined reliably due to the deformationmentioned above, but is between 2.8 and 3.2. The averageEI (aEI = centrum length divided by average of posteriorwidth and height; Chure et al. 2010) equals 2.5. The aEIvaries along a vertebral column in sauropods; it is gener-ally highest in the middle cervical vertebrae. The aEI ofthis vertebra of Astrophocaudia is intermediate betweenthe low values observed in basal macronarians (e.g. Cama-rasaurus) and the high values (above 4.0) observed in manytitanosauriforms (e.g. Giraffatitan, Malawisaurus; Chureet al. 2010).

The centrum is strongly opisthocoelous and has a flatto slightly concave bottom that lacks ridges, fossae orforamina. Pneumaticity is extensive in the centrum and itslateral face is highly subdivided into pneumatic cameraeand camellae that are about 1–8 cm in their longest dimen-sion, ramify into decreasingly smaller cavities, and areseparated by 1–3 mm thick bony walls. A sharp ridgeon the dorsolateral face of the centrum delimits the pneu-matic areas of the centrum from those of the neural arch.The mid-posterior cervical vertebra appears to possess the

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

Trinity Group sauropods 7

Figure 6. Prezygapophyses and hypantra of the 3rd and 6thpreserved caudal vertebrae of the holotype of Astrophocaudiaslaughteri (SMU 61732) in anterolateral view. Dashed lines indi-cate missing bone. Scale bar = 1 cm.

Figure 7. Selected vertebrae of the holotypic posterior caudalvertebral series of Astrophocaudia slaughteri (SMU 61732) in:A, left lateral; B, anterior; and C, posterior views. Numbers beloweach vertebra indicate likely position of the vertebra in the caudalsequence. Dashed lines indicate missing bone. Scale bar = 10 cm.

Figure 8. Holotypic dorsal rib of Astrophocaudia slaughteri(SMU 61732) in: A, lateral; and B, posterior views. Dashed linesindicate missing bone. Scale bar = 10 cm.

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

8 M. D. D’Emic

Figure 9. Holotypic chevron of Astrophocaudia slaughteri (SMU61732) in: A, anterior; B, proximal (line drawing); C, posterior;and D, lateral views. Dashed lines indicate missing bone. Scalebar = 5 cm.

camellate or somphospondylous condition sensu Wedel(2003), with several generations of sub-centimetre branch-ing chambers that permeate the centrum. Much of theneural arch is damaged, and the neural spine is largelymissing. The anterior centrodiapophyseal lamina and prezy-gapophyseal diapophyseal lamina are well developed, andthe fossa between them (prezygapophyseal centrodiapophy-seal fossa) is subdivided.

Caudal vertebraeLangston (1974) listed 21 caudal vertebrae as belongingto SMU 61732. In addition to these, three vertebraewere located in a drawer with other materials of SMU61732, along with a notecard, which reads: “Walnut CreekLoc. Paluxy Fm. Sauropod Loc. Turtle vertebrae” (pers.obs. 2009). These caudal vertebrae and the 21 presentedby Langston (1974) are from a single quarry and areappropriate in morphology, size, and preservation torepresent a single series. With these additional vertebrae,the holotypic caudal vertebral series includes a total of24 preserved vertebrae (Figs 5–7). Additional fragments,including zygapophyses, neural spines, and transverseprocesses, were recovered from boxes of fragments fromthe site and reattached in August 2009.

Figure 10. Holotypic partial scapula of Astrophocaudia slaugh-teri (SMU 61732) in lateral view with cross section indicated.Dashed lines indicate missing bone. Scale bar = 10 cm.

All regions of the tail are well represented, aside from theanteriormost caudal vertebrae. The anteriormost preservedcaudal vertebra is likely the eighth in the series, based oncomparisons with Giraffatitan brancai and Cedarosaurusweiskopfae. After this, there is a gap of 1–2 vertebrae, andthen a series of five consecutive caudal vertebrae. The restof the preserved tail is made up of series of one to sixvertebrae with gaps between. Estimated vertebral positionsare given in Figs 5 and 7. In the description below, vertebraewill be numbered according to their most likely anatomicalposition.

Over the course of the caudal vertebral series, the centramaintain a width:height ratio of roughly 1:1 (Table 2). Incaudal vertebra eight, the lateral walls of the centrum angleinwards ventrally and are delimited from the flat ventralface by a weak ridge. This ridge persists to about caudalvertebra 25, whereas distal caudal vertebrae are cylindricalin cross section. There are no fossae on the lateral or ventralfaces of any of the caudal vertebral centra, and all of thecaudal vertebral centra and neural arches have a solid (i.e.non-camerate or camellate) bone texture. With the neural

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

Trinity Group sauropods 9

Figure 11. Holotypic partial right ilium of Astrophocaudiaslaughteri (SMU 61732) in: A, dorsal; and B, lateral views.Dashed lines indicate missing bone. Scale bar = 10 cm.

canal held horizontally, the centrum and neural spines of theanterior and middle caudal vertebrae are nearly verticallyoriented (Fig. 5).

The anterior face of the first preserved (eighth) caudalcentrum is concave, and the posterior face is irregular inshape, but overall flat (Fig. 5). The eighth and tenth caudalvertebral centra have concave anterior faces, and posteriorfaces that are flat or weakly concave. In the next vertebra,both faces are concave, but the anterior concavity is greaterthan the posterior. This type of articulation, termed ‘plani-concave’ by Tidwell et al. (2001) has been proposed as anautapomorphy of Cedarosaurus. However, this conditionis present in several sauropods, including Camarasaurus(YPM 1905; pers. obs. 2007), Brachiosaurus altithorax(FMNH P 25107; pers. obs. 2008) and Sauroposeidon(FWMSH 93-B; see below; pers. obs. 2009). The rest ofthe vertebrae, excepting the last few, have equally weaklyconcave anterior and posterior faces (i.e. weakly amph-icoelous). The last few are a mix of procoelous and biconvexcentra as in some vertebrae referred to Giraffatitan brancai(HMN MB.R 5000; pers. obs. 2008) and some titanosaurs(e.g. Trotta et al. 2002; Calvo & Gonzalez Riga 2003).

The neural spines of the more anterior caudal verte-brae are composed anterolaterally of spinoprezygapophy-seal laminae. In caudal vertebrae 8–15, the prespinal laminarugosity reaches all the way to the intraprezygapophyseallamina. The spinoprezygapophyseal fossa is narrow andshallow. In the tenth caudal vertebra in the series, there isa slight transverse expansion of the neural spine distally.

Table 2. Measurements of the holotypic caudal vertebrae ofAstrophocaudia slaughteri (SMU 61732) in cm. e = estimatedmeasurement; d = measurement influenced by distortion.Centrum width and height measured at posterior face.

Estimated positionin series

Centrumlength

Centrumwidth

Centrumheight

8 — — —10 7.2 15.6e —11 — — —12 9.7 10.2e 10e13 9.1 10.3 9.214 8.7 9.4 8.915 7.4 10.2 8.619 8.8 6.8 7.120 9.1 6.6 7.422 8.6 6.4 6.723 8.7 7.4d 5.9d24 8 6e 6.225 7.3 6.5d 6.326 7.2 6.8 5.827 8.1 6.4 5.530 7.8 5.4 5.231 7.2 5.4 5.233 6.8 4.3 4.337 6.9 — —38 6.5 — —39 5.7 3.4 3.143 5.1e 2.6e 1.8e46 4.5 1.5 1.648 — 1.3 1.2

In all other caudal vertebrae, the lateral faces of the neuralspine are parallel. In the eighth caudal vertebra of the series,the neural spine has a ‘saddle’ at its midpoint (Fig. 5), asin some vertebrae of Venenosaurus (DMNH 40392; pers.obs. 2010) and Camarasaurus (Ostrom & McIntosh 1966,pl. 37). Some mid-posterior caudal vertebrae have a slightanterior projection on the neural spine (Fig. 7). On theneural arch, there is a small fossa in front of the postzy-gapophyses that represents a combined postzygapophysealspinodiapophyseal fossa plus a postzygapophyseal centro-diapophyseal fossa. The remnants of this fossa persist as asubtle depression until caudal vertebra 15.

The zygapophyses and hyposphene–hypantrum articula-tions undergo dramatic morphological changes along thecaudal vertebral series. In the first caudal vertebra thatpreserves zygapophyses (caudal vertebra 10), the pre- andpostzygapophyses are large and subhorizontally oriented,and the hyposphene is subequal in size to the postzygapoph-ysis. The postzygapophyses and hyposphene are both planarand meet at an angle of about 80◦ (Fig. 5). More posteriorlyin the caudal vertebral series, the zygapophyseal articu-lar surfaces decrease in size faster than the hyposphene-hypantrum articular surfaces. By caudal vertebra 20, theintervertebral neural arch articulation is represented by asingle, vertical plane, as in other sauropods (e.g. Cama-rasaurus, Giraffatitan, Mendozasaurus).

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

10 M. D. D’Emic

Dorsal ribsA total of about 20 fragmentary dorsal ribs are present;none are ‘plank-like’ (cross section more than three timeswider than broad). Some of the preserved ribs approach thiscondition (cross sectional ratio 2.8), so Astrophocaudia mayhave plank-like ribs as do titanosauriforms (Wilson 2002).The largest dorsal rib is pneumatic as in Titanosauriformes(Fig. 8). There is an oval, ridged tubercle on the anteriorhalf of the proximolateral part of the largest dorsal rib (Fig.8A). This feature is absent in the only other dorsal rib thatpreserves this portion of the shaft.

ChevronsLangston (1974) reported a single chevron with SMU61732. In the process of studying the material, a secondwas discovered. Both come from the anterior region of thetail. The more complete chevron is missing only its distal-most blade (Fig. 9). It is 24.2 cm long and has an open hemalcanal measuring 9.1 cm deep dorsoventrally. On the anteriorface of the blade, there is a flattened oval boss measuring1.2 × 4 cm that has a texture of ridges and grooves (Fig. 9).Each arm of the chevron bears a single articular facet witha medially pointed process.

ScapulaA distal scapular blade (Fig. 10) is preserved, which will bedescribed as if held subvertically. The blade has completeanterodorsal and posteroventral margins, and is almostcomplete distally. It could represent a left or right scapula,as the preserved part is symmetrical about its long axis.The preserved length is 70 cm. Its breadth ranges from17 to 38.5 cm, giving a minimum/maximum width ratio ofabout 2.3. The scapular blade is less than 1 cm thick distallyand about 3 cm thick at the centre of the broken base ofthe blade. The base of the blade is flat in cross section asin Euhelopus and titanosaurs, rather than D-shaped with abroad lateral ridge as in non-somphospondylans (Wilson& Sereno 1998). The bone has a texture of subtle, axiallyoriented ridges and grooves on the exterior face of the bone.

IliumThe acetabular region and part of the preacetabular processof the right ilium are present (Fig. 11). The preacetabularlobe of the ilium flares outward at about 45◦ along a gentlecurve, as in most titanosauriforms (Salgado et al. 1997).No evidence of pneumaticity exists in the preserved ilium.A subtle ridge extends anteroposteriorly above the pubicpeduncle on the lateral face of the ilium, as in some othersauropods (e.g. Camarasaurus; Ostrom & McIntosh 1966).This ridge helps to delimit a subtle subtriangular fossa justabove and in front of the public peduncle, as in some othersauropods (e.g. Cetiosaurus; Upchurch & Martin 2003).The ventral edge of the preacetabular process is crushedinwards. The total preserved length of the element is45 cm.

Genus Cedarosaurus Tidwell et al., 1999

Type species. Cedarosaurus weiskopfae Tidwell et al.1999.

Cedarosaurus weiskopfae Tidwell et al. 1999(Figs 12–15)

1974 Pleurocoelus sp. Langston: 85, figs 5, 6a.1989 Pleurocoelus sp. Gallup: 71, figs 1–5.1997 Pleurocoelus Salgado & Calvo: 44, fig. 9.1999 Cedarosaurus weiskopfae Tidwell et al.: 22, figs 2–11.

Holotype. DMNH 39045, a partial skeleton consisting ofdorsal and caudal vertebrae, dorsal ribs, chevrons, partialleft and right scapulae, coracoids and sternal plates, aright humerus, radius, ulna, and metacarpal IV, partial rightand left pubes, ischia, and femora, a right tibia and rightmetatarsals I, II and V.

Holotype locality, horizon and age. Early CretaceousYellowcat Member of the Cedar Mountain Formation, Utah,USA, Barremian–Aptian (Tidwell et al. 1999; Greenhalgh& Britt, 2007).

Referred material. FMNH PR 977, a partial hind limbincluding an incomplete tibia and fibula, an astragalus,five metatarsals and 11 phalanges. Paluxy Formation,Aptian–Albian (Jacobs & Winkler 1998), 20 km south ofDecatur, Texas, USA.

Revised diagnosis. Autapomorphies of Cedarosaurusweiskopfae include: radius with well-developed flangelateral to ulnar articulation (Tidwell et al. 1999), radiuswith tubercle on anterior face of shaft, one-third of theway from proximal end, metatarsal II with well-developedmedial and lateral tubercles at mid-shaft (pers. obs. 2010).FMNH PR 977 is referable to Cedarosaurus on the basis oftwo additional autapomorphies: metatarsal V only slightlyexpanded proximally, and metatarsal V around 1.5 timeslonger than metatarsal I. Also, FMNH PR 977 possessesfour well-developed pedal unguals and a phalanx onmetatarsal V, providing two additional autapomorphies forCedarosaurus.

Description

The following description focuses on material referredto Cedarosaurus weiskopfae, which consists of a partialhind limb described by Gallup (1989). For description andcomparisons of the holotype of Cedarosaurus, see Tidwellet al. (1999).

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

Trinity Group sauropods 11

Figure 12. Metatarsals I–V, referable to Cedarosaurus weiskopfae (FMNH PR 977). in dorsal (top row) and lateral (bottom row) views.Dashed lines indicate missing bone. Scale bar = 10 cm.

Crus and tarsusAbout three-quarters of the tibia and fibula are preservedin several pieces (Gallup 1975). The preserved lengths ofthe tibia and fibula are about 71 and 95 cm, respectively.The tibia is oval in cross section, and its midshaft measures16.1 cm anteroposteriorly by 8.4 cm transversely. Thefibula is roughly D-shaped in cross section with a slightlyconcave medial margin, and its midshaft measures 11.6 cmanteroposteriorly by 5.7 cm transversely. The astragalus isextremely crushed dorsoventrally. It resembles the astragaliof most sauropods (e.g. Camarasaurus) in its proportionsand rugose texture. No calcaneum was found with thespecimen.

MetatarsalsFive metatarsals are preserved (Figs 12, 13). MetatarsalsII–IV are incomplete distally, and the second and fifth arecrushed dorsoventrally (Fig. 12). The proximal articularsurface is largest on metatarsal I, and it is slightly smalleror subequal for metatarsals II–IV. All of the metatarsals haveslightly concave proximolateral faces for the articulation ofthe adjacent metatarsal. In dorsal view, the lateral margin

of each metatarsal is more tightly curved than the medial,as in most sauropods. The metatarsals increase in lengthlaterally such that metatarsal V is the longest and about 1.8times longer than metatarsal I. The long axis of the proximalend of metatarsals I and III is oriented dorsoventrally androughly orthogonally with respect to the long axis of theirdistal ends, which is oriented mediolaterally (Figs 12, 13).

Metatarsal I is subtriangular in proximal view, comingto a point dorsally. The articular facet for ungual I.1 isbevelled dorsomedially (Figs 12, 13). Metatarsals II–IV aresubrectangular proximally. Little can be said about theirdistal ends due to deformation and extensive plaster recon-struction (Fig. 12). Metatarsal V is much broader trans-versely than dorsoventrally, and is only slightly narrowertransversely at its midshaft than at its distal end (Fig. 12).

PhalangesEleven phalanges were preserved with the foot, includ-ing four unguals (Figs 14, 15). The first digit is the onlyone for which an exact phalangeal count (two) is known.Three are definitively the proximalmost phalanges of digits

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

12 M. D. D’Emic

Figure 13. Articulated pes referable to Cedarosaurus weiskopfae(FMNH PR 977) in: A, proximal; and B, anterodorsal views.Dashed lines indicate missing bone. Scale bar = 10 cm.

Figure 14. Phalanx I.1 referable to Cedarosaurus weiskopfae(FMNH PR 977) in: A, proximal; B, distal; C, ventral; and D,medial views. Scale bar = 5 cm.

Figure 15. Unguals referable to Cedarosaurus weiskopfae(FMNH PR 977). A, ungual of digit I in proximal, medial anddorsal views; B, ungual of digit II in proximal, medial and dorsalviews; C, ungual of digit III in proximal, medial and dorsal views;D, ungual of digit IV in proximal, medial and dorsal views. Scalebar = 10 cm. Dashed lines indicate missing bone.

2–4 based on their large size, and a small phalanx belongsto metatarsal V because it is too small to be the penul-timate phalanx on any of the other digits, which bearunguals. The best-estimate phalangeal formula is 2-3-3-3-1(Fig. 13), but there is a range of possibilities based on othersauropods (e.g. 2-(2-3)-(2-4)-(2-4)-1; Gonzalez Riga et al.2008). Phalanx I.1 is wedge-shaped. The remaining non-terminal unguals are roughly trapezoidal in dorsal view withconstricted midshafts and oval proximal faces. The inner-most three unguals are about 1.7 times longer than tall andhave dorsally acuminate oval proximal faces (Fig. 15). Eachbears variably developed nail-grooves. The presence of a

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

Trinity Group sauropods 13

Figure 16. Bone histology of Paluxysaurus jonesi. A, micropho-tograph of a thin section of the humerus (FWMSH 93B-10);B, microphotograph of a thin section of the femur (FWMSH 93B-10). Thin sections reveal wide vascular canals in primarily laminarfibrolamellar bone, which indicates that these individuals were notat adult size at death. Scale bar is the same for A and B.

large claw on metatarsal IV is regarded as an autapomorphyamong eusauropods (Upchurch 1995, 1998; Wilson 2002).

Sauroposeidon Wedel et al., 2000a

Type species. Sauroposeidon proteles Wedel et al., 2000a.

Sauroposeidon proteles Wedel et al. 2000a(Fig. 16)

2000a Sauroposeidon proteles Wedel et al.: 110, figs 1, 2,4.

2007 Paluxysaurus jonesi Rose: 5, figs 4–27.

Holotype. OMNH 53062, four middle cervical vertebrae.

Holotype locality, horizon and age. Early Cretaceous(Aptian–Albian) Antlers Formation of southern Oklahoma,USA.

Referred material. Referred material (see below) includesholotypic and referred materials of Paluxysaurus jonesi

Rose 2007: FWMSH 93B-10-1 through FWMSH 93B-10-51, TMM 42488. Early Cretaceous (Aptian–Albian)Twin Mountains Formation of north-central Texas, USA.(Winkler et al. 1990). The Twin Mountains Formationis laterally equivalent to the Antlers Formation. Some ofthe material reported by Ostrom (1970) from the CloverlyFormation is likewise referable to Sauroposeidon, thedetails of which will be dealt with in future work (see alsoWedel & Cifelli 2005).

Revised diagnosis. Extreme elongation of the middlecervical vertebrae (length/posterior centrum height > 6),middle cervical vertebrae with posterior expansion of thepneumatic fossa to the cotyle, neural spines perforated inmiddle cervical vertebrae, top of neural spine with broadmidline ridge flanked by small fossae at its anterior andposterior ends, middle and posterior dorsal neural spinesthat taper distally, anterior caudal vertebral centra roughlysquare in cross section, anterior caudal vertebrae with diver-gent spinoprezygapophyseal laminae (angle greater than50◦) forming wide spinoprezygapophyseal fossae, scapulawith two processes at the base of the ventral edge of theblade, humerus gracile, length/midshaft width > 7.5.

Remarks. Sauroposeidon proteles and Paluxysaurusjonesi are from laterally equivalent units and have beenhypothesized to be closely related to the brachiosauridsBrachiosaurus and Giraffatitan (Wedel et al. 2000b; Rose2007). Because of their close spatiotemporal similarityand proposed phylogenetic affinity, it is possible thatSauroposeidon and Paluxysaurus represent a single taxon.Seven autapomorphies were originally proposed to diag-nose Sauroposeidon proteles (Wedel et al. 2000a, b). Fourof these features characterize a wider group of titanosauri-forms, including extreme elongation of cervical vertebrae(e.g. Erketu, Ksepka & Norell 2006; Euhelopus, Wilson& Upchurch 2009); internal somphospondylous vertebralpneumaticity (e.g. Chubutisaurus, Salgado et al. 1997;Euhelopus, Wilson & Upchurch 2009), extremely longcervical ribs (e.g. Rapetosaurus, Curry Rogers 2009), anda ‘centroparapophyseal lamina’ (Wedel et al. 2000a, b),which is similarly developed in some large vertebrae ofGiraffatitan brancai (e.g. Janensch 1950, figs 40, 42; pers.obs. 2008) and Euhelopus (Wiman 1929, pl. 3; Wilson& Upchurch 2009). Wedel et al. (2000b) described thelatter feature as differing between Sauroposeidon and Giraf-fatitan, but existing differences in their development andorientation are minor when serial variation within Giraf-fatitan itself is taken into account. Three of the proposedautapomorphies of Sauroposeidon (Wedel et al. 2000b)are shared with Paluxysaurus: extremely thin-walled toperforate neural spines in mid-cervical vertebrae, poste-rior expansion of the lateral pneumatic fossa of the centrum(‘pleurocoel’), and the posterior placement of the diapophy-ses in larger cervical vertebrae (Rose 2007).

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

14 M. D. D’Emic

Besides the autapomorphies presented by Wedelet al. (2000a, b), Rose (2007) proposed three features todistinguish Sauroposeidon from Paluxysaurus: absence ofan anterior centrodiapophyseal lamina, wider neural arches,and a wider prezygapophyseal diapophyseal lamina thanSauroposeidon. However, re-examination of the vertebraeof both exemplars indicates that these features do not distin-guish them. There are multiple laminae beneath the armof the prezygodiapophyseal lamina (prdl) in both exem-plars (see Wedel et al. 2000b, fig. 6; Rose 2007, figs 7, 8;pers. obs. 2009). The anterior centrodiapophyseal laminaappears unusually large on the sixth cervical vertebra ofPaluxysaurus (Rose 2007, fig. 8) because this vertebra hasbeen sheared upwards and backwards on its right side.Despite this shearing, it is apparent that the vertebrae ofthe two taxa do not differ in breadth of the centrum or thewidth of the prezygodiapophyseal lamina (prdl).

Although there are no substantive morphologicaldifferences between their exemplars, the two taxa differsubstantially in size: the sixth to eighth cervical vertebraeof Sauroposeidon have centra (including condyles) that arearound 1.25 m long, whereas the largest known vertebraof Paluxysaurus (FWMSH 93B-10-30) was estimated tohave a centrum length of 0.83 m (Rose 2007). In order toevaluate the meaning of this size difference, morphologicaland histological features relevant to ontogeny are describedbelow.

Description

The following description focuses on demonstrating thejuvenile nature of exemplars of ‘Paluxysaurus jonesi’ inorder to substantiate its referral to Sauroposeidon prote-les. For description and comparisons of the holotype ofSauroposeidon proteles, see Wedel et al. (2000a, b), andfor description and comparisons of ‘Paluxysaurus jonesi’see Rose (2007).

All of the neurocentral sutures of the holotypic cervicalvertebrae of Sauroposeidon are fully fused, suggesting thatit was at or near adult size (Ikejiri 2003). In contrast, somecervical and dorsal vertebrae referred to Paluxysaurus (e.g.FWMSH 93B-10-8, -13, -30) have slight furrows repre-senting their neurocentral sutures near their condyles andcotyles (pers. obs. 2009). In addition, the last sacral verte-bra and ilia of one specimen of Paluxysaurus are not fullyfused into the sacrum (pers. obs. 2009), and one of thecoracoids is not fused to a scapula (Rose 2007), provid-ing evidence that these individuals were not fully skeletallymature (Schwarz et al. 2007). The femora and humeri in theJones Ranch quarry do not differ by more than 10% in size(Rose 2007), so all of the quarry individuals were likely ofsimilar ontogenetic age.

In order to more precisely determine the ontogeneticage of the Paluxysaurus specimens, midshaft samples of a

referred fragmentary humerus (FWMSH 93-B-10-7; Rose2007, fig. 23, subfigures 8–11) and left femur (FWMSH93-B-10) were taken with a diamond-tipped drill accord-ing to the methodology of Stein & Sander (2009). Due tothe impermeable but friable nature of the bones, stabiliza-tion with glue and extraction of the cores was difficult, andthe outermost few millimetres of bone were lost from onecore. A second sample from the femur preserves the entireperiosteum, but is slightly off-centre and distal relative tothe mid-shaft sample. It is uncertain whether or not thefemur and humerus belonged to a single individual.

Thin sections of the humeral and femoral cores reveala cortex composed of fibrolamellar bone, as in most othersauropods (Fig. 16; Klein & Sander 2008). Most of the bonetexture is parallel-fibred, and vascular canals are relativelyopen. Only one line of arrested growth is visible in eachsection. The presence of an external fundamental systemis equivocal in the humerus and one of the femoral coresdue to damage, but it was absent in the femoral sectionthat preserves the periosteum. A few secondary osteons arepresent within the outer centimetre of both cortices, butthese do not form a solid remodelling front.

The bone represented in the outer cortex of the humerusrepresents types D and E of Klein & Sander (2008), whichapproximates to these authors’ histological ontogeneticstage 8 or 9. This indicates that this individual was notat adult size. Femora and humeri of Giraffatitan brancai athistological ontogenetic stage 8–9 can be one-half to three-quarters the size of the largest known individuals of thatspecies (see Klein & Sander 2008, fig. 4G).

In summary, lack of substantial morphological or defini-tive size differences and shared unique features (see diag-nosis above) suggest that Paluxysaurus jonesi and Sauro-poseidon proteles represent the same species.

Discussion

Phylogenetic affinities of sauropod remainsfrom the Trinity Group and AntlersFormationThe proposed phylogenetic affinities of sauropods fromTrinity–Antlers strata have not been as complex as theirtaxonomic identities, with all authors agreeing that theyrepresent basal titanosauriforms or titanosaurs. As shownabove, most of the material making up ‘Pleurocoelus’from Texas discussed by Salgado et al. (1997) and otherauthors (Langston 1974; McIntosh 1990; Upchurch et al.2004) is instead attributable to Astrophocaudia (SMU61732), Cedarosaurus (FMNH PR 977) or Sauroposeidon(FWMSH 93B-10; TMM 42488). Salgado et al. (1995,1997) suggested that ‘Pleurocoelus nanus’ is the sistertaxon to Titanosauria, and ‘Pleurocoelus sp.’ from Texasis a basal titanosaur. The characters listed as supporting

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

Trinity Group sauropods 15

titanosaur or near-titanosaur affinities – for example, asingle prespinal lamina in dorsal vertebrae – are nowrecognized as features characterizing a wider clade thanTitanosauria (Carballido et al. 2011). Sauroposeidon wasdescribed as a brachiosaurid (Wedel et al. 2000a) andrecovered as such (as ‘Paluxysaurus’) in the phylogeneticanalysis of Rose (2007). Upchurch et al. (2004) and Naishet al. (2004) also suggested that Sauroposeidon was abrachiosaurid. Referral of material from the CloverlyFormation to Sauroposeidon (Wedel et al. 2000b; D’Emicand Foreman in press) augments the available data relevantto its phylogenetic position. Ostrom (1970) pointed outsome features shared between sauropod material fromthe Cloverly Formation (e.g. YPM 5449) and titanosaurs(such as a robust ulna); these features contrast with thebrachiosaurid–Sauroposeidon links proposed by Wedelet al. (2000a, b), Naish et al. (2004) and Rose (2007).

The taxonomic revisions presented above augment,combine and redistribute some apomorphies of the TrinityGroup sauropods. With this new information in hand, acladistic analysis of basal titanosauriform relationshipswas conducted, including 119 characters and 22 ingrouptaxa. Shunosaurus (Dong et al. 1983), Omeisaurus (Heet al. 1988; Tang et al. 2001) and Jobaria (Serenoet al. 1999) were used as outgroup taxa. Provenance and

reference information for each terminal taxon is listedin the Online Supplementary Material. Most characterswere culled from the literature (e.g. Upchurch 1998;Wilson 2002), and many of these characters are novel. Thecharacter–taxon matrix and character list are presentedin the Online Supplementary Material. Characters wereentered into MacClade (Maddison & Maddison 1992) andanalysed in PAUP∗ (Swofford 2000). A branch and boundsearch using the tree bisection and reconnection algorithmyielded 33 equally most parsimonious trees of length 197steps (Fig. 17; consistency index = 0.64; retention index =0.80). Decay indices for many nodes in the tree were equalto one but others, such as those supporting Brachiosauri-dae, Somphpospondyli and Titanosaurifromes, were higher(Fig. 17).

Both Astrophocaudia and Sauroposeidon are recoveredas members of Somphospondyli more derived than Liga-buesaurus, in a polytomy with Tastavinsaurus, Chubuti-saurus, Phuwiangosaurus and Euhelopus. Nine steps arerequired to move Sauroposeidon into the Brachiosauridae,and 10 steps are required to position it within Titanosauria.Templeton tests (see Larson 1994; Wilson 2002) rejectboth titanosaur and brachiosaurid affinities for Sauro-poseidon (n = 36 and p < 0.0001 for Titanosauria;n = 21, p = 0.03 for Brachiosauridae). Astrophocaudia

Figure 17. Phylogenetic relationships of Astrophocaudia, Sauroposeidon, and Cedarosaurus among basal titanosauriforms. Numbers tothe right of each node indicate the decay index; if no number is listed, decay index = 1. Abbreviations: CI, consistency index; mpts, mostparsimonious trees; RI, retention index; TL, treelength.

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

16 M. D. D’Emic

and Sauroposeidon are recognized as titanosauriformsbased on the presence of camellate presacral vertebrae,pneumatic dorsal ribs, a flared iliac preacetabular process,and caudal vertebrae with neural arches situated ante-riorly. Somphospondylan features of Astrophocaudiaand Sauroposeidon include somphospondylous vertebralpneumaticity (sub-centimetre cells with sub-millimetrewalls that permeate the vertebra) and a scapular blade witha flat cross section at its base. Sauroposeidon possessesnumerous other somphospondylan synapomorphies,including anterior dorsal vertebrae with a single prespinallamina and flat, ‘paddle-shaped’ neural spines, a mediallybevelled scapular glenoid, an ischial blade that is shorterthan the pubic blade, and a proximally embracing tibia andfibula. Cedarosaurus is a brachiosaurid titanosauriformon the basis of a high humerus-to-femur ratio, gracilehumerus, rounded humeral proximolateral corner, andanterior and middle caudal vertebrae with sporadicallydistributed, shallow fossae in lateral faces of centrum.

Several sauropod remains that are indeterminate tothe genus level have been reported from Trinity–Antlersstrata, including a coracoid (Larkin 1910), ischium (Gallup1975) and teeth (Gallup 1975; Maxwell & Cifelli 2000).Langston (1974) mentioned a partial skeleton from theGlen Rose Formation of Blanco County, Texas (TMM40435; see also Tidwell & Carpenter 2003). This specimenis a juvenile based on the lack of neurocentral fusionin some vertebrae, and the lack of fusion among thelaterosphenoids, prootics, parietals and frontals (pers. obs.2008). This skeleton represents a titanosauriform based onthe presence of camellate presacral vertebral pneumaticity,but is not diagnostic at the genus level.

Latitudinal homogeneity in EarlyCretaceous North American dinosaur faunasThe referral of Paluxysaurus and some Cloverly Forma-tion sauropod material (Ostrom 1970; D’Emic andForeman in press) to Sauroposeidon, as well as the refer-ral of the hind limb from the Glen Rose Formation toCedarosaurus, reinforces proposed faunal links among theTrinity Group, Antlers Formation, Ruby Ranch Member ofthe Cedar Mountain Formation, and the Cloverly Formation(Langston 1974; Winkler et al. 1990; Jacobs & Winkler1998). Nydam & Cifelli (2002) challenged the temporalcorrelation of the Cloverly and Antlers formations on thebasis of their disparate lizard faunas, but noted similar-ity between the Cloverly and Twin Mountains formations.However, at the spatiotemporal scale of sampling in theseformations, palaeoenvironmental biases can also explainthe observed differences in faunal compositions, especiallywhen dealing with small taxa of probable small geographi-cal range (Winkler et al. 1990).

Though separated by over 15◦ of palaeolatitude (1500km; Fig. 2), four dinosaur genera are shared between

the penecontemporaneous Twin Mountains/Antlersand Cloverly formations: Tenontosaurus, Deinonychus,Sauroposeidon and Acrocanthosaurus (D’Emic et al. inpress). The formations also share multiple non-dinosaurgenera, as well as some suprageneric dinosaur taxa suchas Nodosauridae and an unnamed clade of small basalornithopods (Jacobs & Winkler 1998). This degree offaunal similarity over 15◦ of palaeolatitude stands incontrast to the latitudinal variation present in dinosaurfaunas observed in the Late Cretaceous (Lehman 1987,2001). Decades of exploration and analyses of LateCretaceous North American strata have reinforced thislatitudinal variation (Gates et al. 2010; Mannion et al. inpress), whereas increased data and study of Early Creta-ceous faunas has reinforced its latitudinal homogeneity.Proposed explanations for Late Cretaceous latitudinalvariation (e.g. climate) could be tested against the morehomogenous pattern observed in the Early Cretaceous.

Conclusions

Historically, Early Cretaceous sauropods from NorthAmerica were referred to the genera Pleurocoelus orAstrodon from the Early Cretaceous Arundel Formationof Maryland. A lack of associations and non-diagnostictype specimens means that species of Astrodon andPleurocoelus (Table 1) are nomina dubia.

Because these Maryland species are invalid, materialspreviously referred to them were re-examined. A singlepartial skeleton previously referred to ‘Pleurocoelus sp.’from the Trinity Group represents a new taxon, Astropho-caudia slaughteri. Material from the Trinity Group desig-nated as Paluxysaurus jonesi are morphologically similarto and bear autapomorphies of Sauroposeidon, have a simi-lar spatiotemporal provenance, and were not at their adultbody size. Paluxysaurus is a junior synonym of Sauroposei-don, which is also represented in the Cloverly Formation ofWyoming (Ostrom 1970; Wedel et al. 2000b; D’Emic andForeman in press). This augmentation makes Sauroposei-don one of the best-known Early Cretaceous North Amer-ican sauropods, whereas previously it was one of the mostpoorly known. A hind limb from the Trinity Group previ-ously referred to ‘Pleurocoelus’ (Langston 1974) is referredto Cedarosaurus weiskopfae on the basis of pedal synapo-morphies. Cladistic analysis indicates that Cedarosaurus isa brachiosaurid, whereas Astrophocaudia and Sauroposei-don are members of the Somphospondyli. The revision ofthe taxonomic and phylogenetic affinities of Sauroposeidonhave implications for the affinities of fragmentary sauropodremains that have been linked to the genus from other land-masses and geological time periods (e.g. Naish et al. 2004;You & Li 2009).

The sauropod dinosaurs Sauroposeidon andCedarosaurus were widespread in western North America

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

Trinity Group sauropods 17

in the Early Cretaceous, reinforcing links among northernand southern faunas first drawn based on other dinosaurs(Jacobs & Winkler 1998). This relative homogeneitysuggests that provincialism among North Americandinosaur faunas with respect to latitude developed onlylater in the Cretaceous.

Acknowledgements

This paper represents a portion of my doctoral thesisat the University of Michigan. Thanks go to A. Panand L. Ballinger (FWMSH), K. Carpenter (DMNS), D.Winkler and L. Jacobs (SMU), M. Carrano and M. Brett-Surman (USNM), R. Cifelli (OMNH), L. Murray and T.Rowe (TMM), B. Simpson (FMNH), D. Schwarz-Wings(Humboldt Museum), W. Joyce and D. Brinkman (YPM),D. Colodner (ASDM), and J. Diffily and the mounting crewat the Reid workshop in Azle, Texas, for collections access.D. Winkler is thanked for permission to describe, and infor-mation pertaining to, Astrophocaudia; A. Pan, L. Ballingerand D. Colodner (ASDM) for permission to destructivelysample limb bones for histology; M. Sander and K. Stein(University of Bonn) for helpful technical informationabout drilling sauropod bones; D. Fisher (University ofMichigan) for use of thin-sectioning equipment; K. M.Smith (Georgia Southern University) for assitance; L. BeatiZiegler (Georgia Southern University) for computing time;and D. Nixon (FWMSH) for preparation of some bones ofAstrophocaudia. Discussions with T. Ikejiri, P. Mannion,P. Rose, J. A. Whitlock, D. Winkler and especially J. A.Wilson greatly improved the paper. T. Baumiller, C. Badg-ley, T. Ikejiri, J. A. Whitlock and J. A. Wilson are thankedfor helpful comments on earlier drafts of this manuscript. P.Upchurch, an anonymous reviewer, and editors are thankedfor comments and edits that improved the paper. Fundingwas provided by the University of Michigan Scott TurnerAward and a Horace Rackham Graduate Student ResearchGrant.

References

Bilelo, M. M. 1969. The fossil fish Lepidotes in the Paluxy Forma-tion. North-central Texas. American Midland Naturalist, 81,405–411.

Bonaparte, J. F. 1986. The early radiation and phylogenetic rela-tionships of sauropod dinosaurs, based on vertebral anatomy.Pp. 247–258 in K. Padian (ed.) The Beginning of the Age ofDinosaurs. Cambridge University Press, Cambridge.

Bonaparte, J. F., Gonzalez Riga, B. J. & Apesteguıa, S.2006. Ligabuesaurus leanzai gen. et sp. nov. (Dinosauria,Sauropoda), a new titanosaur from the Lohan Cura Forma-tion (Aptian, Lower Cretaceous) of Neuquen, Patagonia,Argentina. Cretaceous Research, 27, 364–376.

Borsuk-Bialynicka, M. 1977. A new camarasaurid sauropodOpisthocoelicaudia skarzynskii, gen. n., sp. n. from the UpperCretaceous of Mongolia. Palaeontologia Polonica, 37, 1–64.

Calvo, J. O. & Bonaparte, J. F. 1991. Andesaurus delgadoi n.g. n. sp. (Saurischia, Sauropoda) a titanosaurid dinosaur fromthe Rıo Limay Formation (Albian-Cenomanian), Neuquen,Argentina. Ameghiniana, 28, 303–310.

Calvo, J. O. & Gonzalez Riga, B. J. 2003. Rinconsauruscaudamirus gen. et sp nov., a new titanosaurid (Dinosauria,Sauropoda) from the Late Cretaceous of Patagonia, Argentina.Revista Geologica de Chile, 30, 333–353.

Canudo, J., Royo-Torres, R. & Cuenca-Bescos, G. 2008. Anew sauropod: Tastavinsaurus sanzi gen et. sp. nov. from theEarly Cretaceous (Aptian) of Spain. Journal of VertebratePaleontology, 28, 712–731.

Carballido, J. L., Pol, D., Cerda, I. & Salgado, L. 2011.The osteology of Chubutisaurus insignis Del Corro, 1975(Dinosauria: Neosauropoda) from the ‘middle’ Cretaceous ofcentral Patagonia, Argentina. Journal of Vertebrate Paleon-tology, 31, 93–110.

Carpenter, K. & Tidwell, V. 2005. Reassessment of theEarly Cretaceous sauropod Astrodon johnstoni Leidy 1865(Titanosauriformes). Pp. 78–114 in V. Tidwell & K. Carpen-ter (eds) Thunder Lizards: the sauropodomorph dinosaurs.Indiana University Press, Bloomington.

Chure, D., Britt, B. B., Whitlock, J. A. & Wilson, J. A. 2010.First complete sauropod dinosaur skull from the Cretaceous ofthe Americas and the evolution of sauropod dentition. Natur-wissenschaften, 97, 379–391.

Cope, E. D. 1877. On a gigantic saurian from the Dakota epochof Colorado. Paleontological Bulletin, 25, 5–10.

Curry Rogers, K. 2009. The postcranial osteology ofRapetosaurus krausei (Sauropoda: Titanosauria) from the LateCretaceous of Madagascar. Journal of Vertebrate Paleontol-ogy, 29, 1046–1086.

Curry Rogers, K. A. & Forster, C. 2004. The skull ofRapetosaurus krausei (Sauropoda, Titanosauria) from theLate Cretaceous of Madagascar. Journal of Vertebrate Pale-ontology, 24, 121–144.

del Corro, G. 1975. Un nuevo sauropodo del CretacicoChubutisaurus insignis gen. et sp. nov. (Saurischia-Chubutisauridae nov.) del Cretacico Superior (Chubutiano),Chubut, Argentina. Actas I Congreso Argentino de Paleon-tologıa y Bioestratigrafıa, Tucuman, 2, 229–240.

D’Emic, M. D. & Foreman, B. Z. in press. The beginning of thesauropod dinosaur hiatus in North America: insights from theLower Cretaceous Cloverly Formation of Wyoming. Journalof Vertebrate Paleontology.

D’Emic, M. D., Melstrom, K. M. & Eddy, D. R. in press. Paleo-biology and geographic range of the large-bodied Cretaceoustheropod dinosaur Acrocanthosaurus atokensis. Palaeogeog-raphy, Palaeoclimatology, Palaeoecology.

D’Emic, M. D. & Wilson, J. A. 2011. New remains attributableto the holotype of the sauropod dinosaur Neuquensaurusaustralis, with implications for saltasaurine systematics. ActaPalaeontologica Polonica, 56, 61–73.

D’Emic, M. D., Wilson, J. A. & Williamson, T. E. 2011.A sauropod dinosaur pes from the latest Cretaceous ofNorth America and the validity of Alamosaurus sanjuanensis(Sauropoda: Titanosauria). Journal of Vertebrate Paleontol-ogy, 31, 1072–1079.

Dong, Z., Zhou, S. & Zhang, Y. 1983. The dinosaurian remainsfrom Sichuan Basin, China. Palaeontologica Sinica, Series C,23, 139–145.

Gallup, M. R. 1975. Lower Cretaceous dinosaurs and associatedvertebrates from north-central Texas in the Field Museumof Natural History. Unpublished M.A. thesis, University ofTexas, Austin, 318 pp.

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

18 M. D. D’Emic

Gallup, M. R. 1989. Functional morphology of the hindfootof the Texas sauropod Pleurocoelus sp. indet. Pp. 71–74 inJ. O. Farlow (ed.), Paleobiology of the Dinosaurs. GeologicalSociety of America Special Paper, 238.

Gates T. A., Sampson, S. D., Zanno, L. E., Roberts, E. M.,Eaton, J. G., Nydam, R. L., Hutchinson, J. H., Smith,J. A., Loewen M. A. & Getty M. A. 2010. Biogeography ofterrestrial and freshwater vertebrates from the Late Cretaceous(Campanian) Western Interior of North America. Palaeogeog-raphy, Palaeoclimatology, Palaeoecology, 291, 371–387.

Gilmore, C. W. 1921. The fauna of the Arundel Formationof Maryland. Proceedings of the United States NationalMuseum, 59, 581–594.

Gilmore, C. W. 1922. Discovery of a sauropod dinosaur from theOjo Alamo Formation of New Mexico. Smithsonian Miscel-laneous Collections, 81, 1–9.

Gomani, E. M. 2005. Sauropod dinosaurs from the Early Creta-ceous of Malawi, Africa. Palaeontologia Electronica, 8, 1–37.

Gomani, E., Jacobs, L. L. & Winkler, D. A. 1999. Comparisonof the African titanosaurian, Malawisaurus, with an EarlyCretaceous North American sauropod. Pp. 223–233 in Y.Tomida, T. H. Rich & P. Vickers-Rich (eds) Proceedings of theSecond Gondwanan Dinosaur Symposium. National Museumof Tokyo Monograph, 15.

Gonzalez Riga, B. J., Calvo, J. O. & Porfiri, J. 2008. An articu-lated titanosaur from Patagonia (Argentina): new evidence ofneosauropod pedal evolution. Palaeoworld, 17, 33–40.

Greenhalgh, B. W. & Britt, B. B. 2007. Stratigraphy andsedimentology of the Morrison–Cedar Mountain Formationboundary, East-Central Utah. Pp. 81–100 in G. C. Willis, M.D. Hylland, D. L. Clark & T. C. Chidsey, Jr (eds) Central Utah– diverse geology of a dynamic landscape. Utah GeologicalAssociation Publication, 36.

Hatcher, J. B. 1903. Discovery of Astrodon (Pleurocoelus) inthe Atlantosaurus beds of Wyoming. Annals of the CarnegieMuseum of Natural History, 2, 9–14.

Haughton, S. H. 1928. On some reptilian remains from theDinosaur beds of Nyasaland. Transactions of the Royal Soci-ety of South America, 16, 69–83.

He, X., Li, K. & Cai, K. 1988. The Middle Jurassic Dinosaurfauna from Dashanpu, Zigong, Sichuan, Vol. IV: Sauropoddinosaurs (II), Omeisaurus tianfuensis. Sichuan Scientific andTechnical Publishing House, Chengdu. 143 pp. [In Chinesewith English summary.]

Ikejiri, T. 2003. Sequence of closure of neurocentral sutures inCamarasaurus (Sauropoda) and implications for phylogenyin Reptilia. Journal of Vertebrate Paleontology, 23, 65A.

Jacobs, L. L. & Winkler, D. A. 1998. Mammals, archosaurs, andthe Early to Late Cretaceous transition in north-central Texas.Pp. 253–280 in Y. Tomida, T. H. Rich & P. Vickers-Rich (eds)Proceedings of the Second Gondwanan Dinosaur Symposium.National Museum of Tokyo Monograph, 15.

Jacobs, L. L., Winkler, D. A., Downs, W. R. & Gomani, E.M. 1993. New material of an Early Cretaceous titanosauridsauropod dinosaur from Malawi. Palaeontology, 36, 523–534.

Janensch, W. 1914. Ubersicht uber die Wirbeltierfauna derTendaguru-Schichten, nebst einer kurzen Charakterisierungder neu aufgefuhrten Arten von Sauropoden. Archiv Fur Bion-tologie, 3, 15–58.

Janensch, W. 1929. Material un Formengehalt der Sauropoden inder Ausbeute der Tendaguru Expedition. Palaeontographica,Supplement 7, 2, 1–34.

Janensch, W. 1950. Die wirbelsaule von Brachiosaurus brancai.Palaeontographica, Supplement 7, 2, 27–93.

Johnston, C. J. 1859. Note upon odontology. American Journalof Dental Science, New Series, 9, 337–343.

Kingham, R. F. 1962. Studies of the sauropod dinosaur AstrodonLeidy. Proceedings of the Washington Junior Academy ofSciences, 1, 38–43.

Klein, N. & Sander, M. 2008. Ontogenetic stages in the long bonehistology of sauropod dinosaurs. Paleobiology, 34, 247–263.

Ksepka, D. T. & Norell, M. A. 2006. Erketu ellisoni, along-necked sauropod from Bor Guve (Dornogov Aimag,Mongolia). American Museum Novitates, 3508, 1–16.

Langston, W. 1974. Nonmammalian Comanchean tetrapods.Geoscience and Man, 8, 77–102.

Lapparent, A. F. de & Zbyszewski, G. 1957. Les dinosauriensdu Portugal. Memoires des Services Geologiques du Portugal,Nouvelle Serie, 2, 1–63.

Larkin, P. 1910. The occurrence of a sauropod dinosaur in theTrinity Cretaceous of Oklahoma. Journal of Geology, 28,93–98.

Larson, A. 1994. The comparison of morphological and molecu-lar data in phylogenetic systematics. Pp. 371–390 in B. S. B.Schierwater, G. P. Wagner & R. DeSalle (eds) Molecular biol-ogy and evolution: Approaches and applications. Birkhauser,Basel.

Lehman, T. M. 1987. Late Maastrichtian paleoenvironments anddinosaur biogeography in the Western Interior of North Amer-ica. Palaeogeography, Palaeoclimatology, Palaeoecology, 60,189–217.

Lehman, T. M. 2001. Late Cretaceous dinosaur provinciality.Pp. 310–328 in D. Tanke & K. Carpenter (eds) MesozoicVertebrate Life. Indiana University Press, Bloomington.

Lehman, T. M. & Coulson, A. B. 2002. A juvenile specimenof the sauropod dinosaur Alamosaurus sanjuanensis from theUpper Cretaceous of Big Bend National Park, Texas. Journalof Paleontology, 76, 156–172.

Leidy, J. 1865. Cretaceous reptiles of the United States. Smithso-nian Contributions to Knowledge, 14, 1–135.

Lull, R. S. 1911. The reptilian fauna of the Arundel Formation.Pp 173–211 in Maryland Geological Survey Volume 4: LowerCretaceous. Johns Hopkins Press, Baltimore.

Lydekker, R. 1890. Contributions to our knowledge of thedinosaurs of the Wealden, and the Sauropterygia of thePurbeck and Oxford Clay. Quarterly Journal of the Geologi-cal Society of London, 46, 182–184.

Lydekker, R. 1893. The dinosaurs of Patagonia. Anales del Museode La Plata, 2, 1–14.

Maddison, W. P. & Maddison, D. R. 1992. MacClade 3: Analysisof phylogeny and character evolution, Version 3.0. SinauerAssociates, Sunderland.

Madsen, J. M., McIntosh, J. S. & Berman, D. S. 1995. Theskull and atlas-axis complex of the Upper Jurassic sauropodCamarasaurus Cope (Reptilia: Saurischia). Bulletin of theCarnegie Museum, 31, 1–115.

Mannion P. D. 2011. A reassessment of Mongolosaurus haplodonGilmore, 1933, a titanosaurian sauropod dinosaur from theEarly Cretaceous of Inner Mongolia, People’s Republic ofChina. Journal of Systematic Palaeontology, 9, 355–378.

Mannion, P. D., Benson, R. B. J., Upchurch, P., Butler, P.,Carrano, M. T. & Barrett, P. M. In press. A temperatepalaeodiversity peak in Mesozoic dinosaurs and evidence forLate Cretaceous geographical partitioning. Global Ecologyand Biogeography. DOI: 10.1111/j.1466-8238.2011.00735.x

Mannion, P. D. & Upchurch, P. 2010. Completeness metricsand the quality of the sauropodomorph fossil record throughgeological and historical time. Paleobiology, 36, 283–302.

Mannion, P. D., Upchurch, P., Carrano, M. T. & Barrett, P.M. 2011. Testing the effect of the rock record on diver-sity: a multidisciplinary approach to elucidating the generic

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

Trinity Group sauropods 19

richness of sauropodomorph dinosaurs through time. Biolog-ical Reviews, 86, 157–181.

McIntosh, J. S. 1990. Sauropoda. Pp. 345–401 in D. B. Weisham-pel, P. Dodson & H. Osmolska (eds) The Dinosauria. Univer-sity of California Press, Berkeley.

McIntosh, J. S. 2005. The genus Barosaurus Marsh (Sauropoda,Diplodocidae). Pp. 38–77 in V. Tidwell & K. Carpenter (eds)Thunder-Lizards: the Sauropodomorph Dinosaurs. IndianaUniversity Press, Bloomington.

Marsh, O. C. 1878. Principal characters of American Juras-sic dinosaurs. Pt. I. American Journal of Science, 16,411–416.

Marsh, O. C. 1884. Principal characters of American Jurassicdinosaurs, Part VII. On the Diplodocidae, a new family ofthe Sauropoda. American Journal of Science, Series 3, 27,161–167.

Marsh, O. C. 1888. Notice of a new genus of Sauropoda andother new dinosaurs from the Potomac Formation. AmericanJournal of Science, Series 3, 35, 89–94.

Marsh, O. C. 1896. Dinosaurs of North America. US GeologicalSurvey Annual Report, 16, 133–244.

Marsh, O. C. 1897. Recent observations on European dinosaurs.American Journal of Science, 4, 413–416.

Martin, V., Buffetaut, E. & Suteethorn, V. 1994. A new genus ofsauropod dinosaur from the Sao Khua Formation (Late Juras-sic or Early Cretaceous) of northeastern Thailand. ComptesRendus de l’Academie des Sciences de Paris, 319, 1085–1092.

Martin, V., Suteethorn, V. & Buffetaut, E. 1999. Description ofthe type and referred material of Phuwiangosaurus sirindhor-nae Martin, Buffetaut and Suteethorn 1994, a sauropod fromthe Lower Cretaceous of Thailand. Oryctos, 2, 29–91.

Maxwell, W. D. & Cifelli, R. L. 2000. Last evidence of sauro-pod dinosaurs (Saurischia: Sauropodomorpha) in the NorthAmerican mid-Cretaceous. Brigham Young University Geol-ogy Studies, 45, 19–25.

Monbaron, M., Russell, D. A. & Taquet, P. 1999. Atlasaurusimelakei, n. g., n. sp., a brachiosaurid-like sauropod from theMiddle Jurassic of Morocco. Comptes Rendus de l’AcademieDes Sciences de Paris, 329, 519–526.

Naish, D., Martill, D. M., Cooper, D. & Stevens, K. A. 2004.Europe’s largest dinosaur? A giant brachiosaurid cervicalvertebra from the Wessex Formation (Early Cretaceous) ofsouthern England. Cretaceous Research, 25, 787–795.

Nydam, R. L. & Cifelli, R. L. 2002. Lizards from the LowerCretaceous (Aptian–Albian) Antlers and Cloverly Forma-tions. Journal of Vertebrate Paleontology, 22, 286–298.

Osborn H. F. & Mook C. C. 1921. Camarasaurus, Amphicoelias,and other sauropods of Cope. Memoirs of the AmericanMuseum of Natural History, 3, 247–387.

Ostrom, J. H. 1970. Stratigraphy and paleontology of theCloverly Formation (Lower Cretaceous) of the Bighorn basinarea, Wyoming and Montana. Peabody Museum Bulletin, 35,1–234.

Ostrom, J. H. & McIntosh, J. S. 1966. Marsh’s Dinosaurs. YaleUniversity Press, New Haven, 388 pp.

Otero, A. 2010. The appendicular skeleton of Neuquensaurus,a Late Cretaceous saltasaurine sauropod from Patagonia,Argentina. Acta Palaeontologica Polonica, 55, 399–426.

Owen, R. 1842. A description of a portion of the skeleton ofCetiosaurus, a gigantic extinct saurian occurring in the OoliticFormation of different parts of England. Proceedings of theGeological Society of London, 3, 457–462.

Paul, G. S. 1988. The brachiosaur giants of the Morrison andTendaguru with a description of a new subgenus, Giraffatitan,and a comparison of the world’s largest dinosaurs. Hunteria,2, 1–14.

Powell, J. E. 1992. Osteologıa de Saltasaurus loricatus(Sauropoda–Titanosauridae) del Cretacico Superior delnoroeste Argentino. Pp. 165–230 in Sanz, J. & Buscalioni,A. (eds) Los Dinosaurios y Su Entorno Biotico: Actas delSegundo Curso de Paleontologıa en Cuenca. Instituto Juan deValdes, Cuenca.

Rich, T. H., Vickers-Rich, P., Gimenez, O., Cuneo, R., Puerta,P. & Vacca, R. 1999. A new sauropod dinosaur from ChubutProvince, Argentina. National Science Museum Monographs,15, 61–84.

Riggs, E. S. 1903. Structure and relationships of opisthocoeliandinosaurs. Part I: Apatosaurus Marsh. Field ColumbianMuseum Geological Series, 2, 165–196.

Romer, A. S. 1956. Osteology of the reptiles. University ofChicago Press, Chicago, 772 pp.

Rose, P. J. 2007. A titanosauriform (Dinosauria: Saurischia) fromthe Early Cretaceous of Central Texas and its phylogeneticrelationships. Palaeontologica Electronica, 10, 1–65.

Royo-Torres, R. 2009. El sauropodo de Penarroya de Tastavins.Monograpfıas Turonlenses, 6, 1–548.

Salgado, L. & Azpilicueta, C. 2000. Un nuevo saltasaurino(Sauropoda, Titanosauridae) de la provincia de Rıo Negro(Formacion Allen, Cretacico Superior), Patagonia, Argentina.Ameghiniana, 37, 259–264.

Salgado, L. & Calvo, J. O. 1997. Evolution of titanosauridsauropods. II: The cranial evidence. Ameghiniana, 34, 33–48.

Salgado, L., Calvo, J. & Coria, R. A. 1995. Phylogenetic rela-tionships of Pleurocoelus Marsh (Sauropoda). XI JornadasArgentinas de Paleontologıa de Vertebrados, Resumenes,34, 12.

Salgado, L., Coria, R. & Calvo, J. O. 1997. Evolution oftitanosaurid sauropods. I: phylogenetic analysis based on thepostcranial evidence. Ameghiniana, 34, 3–32.

Sander, P. M. & Clauss, M. 2008. Sauropod gigantism. Science,322, 200–201.

Sander, P. M., Mateus, O., Laven, T. & Knotschke, N. 2006.Bone histology indicates insular dwarfism in a new Late Juras-sic sauropod dinosaur. Nature, 441, 739–741.

Sander, P. M., Christian, A., Clauss, M., Fechner, R., Gee, C.T., Griebeler, E. M., Gunga, H. C., Hummel, J., Mallison,H., Perry, S. F., Preuschoft, H., Rauhut, O. W. M., Remes,K., Tutken, T., Wings, O. & Witzel, U. 2010. Biology of thesauropod dinosaurs: the evolution of gigantism. BiologicalReviews, 86, 117–155.

Schwarz, D., Ikeijiri, T., Breithaupt, B., Sander, P. M. & Klein,N. 2007. A nearly complete skeleton of an early juvenilediplodocid (Dinosauria: Sauropoda) from the Lower Morri-son Formation (Late Jurassic) of north central Wyomingand its implications for early ontogeny and pneumaticity insauropods. Historical Biology, 19, 225–253.

Sereno, P. C., Beck, A. L., Dutheil, D. B., Larssen, H. C.E., Lyon, G. H., Moussa, B., Sadleir, R. W., Sidor, C.A., Varricchio, D. J., Wilson, G. P. & Wilson, J. A.1999. Cretaceous sauropods from the Sahara and the unevenrate of skeletal evolution among dinosaurs. Science, 286,1342–1347.

Stein, K. & Sander, M. 2009. Histological core drilling: a lessdestructive method for studying bone histology. Pp. 69–80in M. A. Brown, J. F. Kane & W. G. Parker (eds) Litho-dendron: The Occasional Papers of Petrified Forest NationalPark 1. Methods in Fossil Preparation. Proceedings of theFirst Annual Fossil Preparation and Collections Symposium,Petrified Forest, Arizona.

Suteethorn, S., Le Loeuff, J., Buffetaut, E., Suteethorn, V.,Taubmook, C. & Chonglakmani, C. 2009. A new skeletonof Phuwiangosaurus sirindhornae (Dinosauria, Sauropoda)

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12

20 M. D. D’Emic

from NE Thailand. Pp. 189–215 in E. Buffetaut, G. Cuny, J. LeLoeuff & V. Suteethorn (eds) Late Palaeozoic and MesozoicEcosystems in SE Asia. Geological Society of London SpecialPublication, 315.

Swofford, D. L. 2002. PAUP∗: phylogenetic analysis using parsi-mony (∗and other methods), Version 4.0b10. Sinauer Asso-ciates, Sunderland.

Tang, F., Jing, X., Kang, X. & Zhang, G. 2001. Omeisaurusmaoianus: a complete sauropod from Jingyuan, Sichuan.China Ocean Press, Beijing. [In Chinese with Englishsummary.]

Taylor, M. 2009. A reevaluation of Brachiosaurus altithoraxRiggs, 1903 (Dinosauria, Sauropoda) and its generic sepa-ration from Brachiosaurus brancai (Janensch, 1914). Journalof Vertebrate Paleontology, 29, 787–806.

Thurmond, J. T. 1974. Lower vertebrate faunas of the Trin-ity division in North-Central Texas. Geoscience and Man, 8,103–129.

Tidwell, V. & Carpenter, K. 2003. Braincase of an Early Creta-ceous titanosauriform sauropod from Texas. Journal of Verte-brate Paleontology, 23, 176–180.

Tidwell, V., Carpenter, K. & Brooks, W. 1999. New sauropodfrom the Lower Cretaceous of Utah, USA. Oryctos, 2, 21–37.

Tidwell, V., Carpenter, K. & Meyer, S. 2001. New titanosauri-form (Sauropoda) from the Poison Strip Member of the CedarMountain Formation (Lower Cretaceous), Utah. Pp. 139–165in D. Tanke & K. Carpenter (eds) Mesozoic Vertebrate Life:New Research Inspired by the Paleontology of Phillip J.Currie. Indiana University Press, Bloomington.

Trotta, M. N. F., Campos, D. A. & Kellner, W. A. 2002.Unusual caudal vertebral centra of a titanosaurid (Dinosauria,Sauropoda) from the Continental Upper Cretaceous of Brazil.Boletim do Museu Nacional, 64, 1–11.

Upchurch, P. 1995. The evolutionary history of sauropoddinosaurs. Philosophical Transactions of the Royal Societyof London, 349, 365–390.

Upchurch, P. 1998. The phylogenetic relationships of sauropoddinosaurs. Zoological Journal of the Linnean Society, 124,43–103.

Upchurch, P. & Martin, J. 2003. The anatomy and taxonomy ofCetiosaurus (Saurischia, Sauropoda) from the Middle Jurassicof England. Journal of Vertebrate Paleontology, 23, 208–231.

Upchurch, P., Barrett, P. M. & Dodson, P. 2004. Sauropoda.Pp. 259–324 in D. B. Weishampel, P. Dodson & H. Osmolska(eds) The Dinosauria. 2nd edition. University of CaliforniaPress, Berkeley.

Wedel, M. J. 2003. The evolution of vertebral pneumaticity insauropod dinosaurs. Journal of Vertebrate Paleontology, 23,344–357.

Wedel, M. J. & Cifelli, R. L. 2005. Sauroposeidon: Okla-homa’s Native Giant. Oklahoma Geology Notes, 65, 40–57.

Wedel, M. J., Cifelli, R. L. & Sanders, R. K. 2000a. Sauro-poseidon proteles, a new sauropod from the Early Creta-ceous of Oklahoma. Journal of Vertebrate Paleontology, 20,109–114.

Wedel, M. J., Cifelli, R. L. & Sanders, R. K. 2000b.Osteology, paleobiology, and relationships of the sauropoddinosaur Sauroposeidon. Acta Palaeontologica Polonica, 45,343–388.

Whitlock, J. A. 2011. A phylogenetic analysis of Diplodocoidea(Saurischia: Sauropoda). Zoological Journal of the LinneanSociety, 161, 872–915.

Wilson, J. A. 1999. A nomenclature for vertebral laminae insauropods and other saurischian dinosaurs. Journal of Verte-brate Paleontology, 19, 639–653.

Wilson, J. A. 2002. Sauropod dinosaur phylogeny: critique andcladistic analysis. Zoological Journal of the Linnean Society,136, 217–276.

Wilson, J. A. & Sereno, P. C. 1998. Early evolution and higher-level phylogeny of sauropod dinosaurs. Journal of VertebratePaleontology, 18, 1–68.

Wilson, J. A. & Upchurch, P. 2009. Redescription andreassessment of the phylogenetic affinities of Euhelopuszdanskyi (Dinosauria: Sauropoda) from the Early Creta-ceous of China. Journal of Systematic Palaeontology, 7,199–239.

Wilson, J. A., D’Emic, M. D., Curry Rogers, K. A., Mohabey,D. M. & Sen, S. 2009. Reassessment of the sauropoddinosaur Jainosaurus (= “Antarctosaurus”) septentrionalisfrom the Upper Cretaceous of India. Contributions fromthe Museum of Paleontology, University of Michigan, 32,17–40.

Wilson, J. A., D’Emic, M. D., Ikejiri, T., Moacdieh, E. &Whitlock, J. A. 2011. A nomenclature for vertebral fossae insauropods and other saurischian dinosaurs. PLOS One, 6(2),1–19.

Wiman, C. 1929. Die Kreide-Dinosaurier aus Shantung. Palaeon-tologia Sinica, Series C, 6, 1–67.

Winkler, D. A., Murray, P. A. & Jacobs, L. L. 1990. Early Creta-ceous (Comanchean) vertebrates of Central Texas. Journal ofVertebrate Paleontology, 10, 95–116.

You, H.-L. & Li, D. 2009. The first well-preserved Early Creta-ceous brachiosaurid dinosaur in Asia. Proceedings of theRoyal Society B: Biological Sciences, 276, 4077–4082.

Young, C. C. 1939. On a new Sauropoda, with notes on otherfragmentary reptiles from Szechuan. Bulletin of the Geologi-cal Society of China, 19, 279–315.

Dow

nloa

ded

by [

Ston

y B

rook

Uni

vers

ity]

at 0

5:34

30

Nov

embe

r 20

12