Saturday, February 22, 2014

What is Eoraptor revisited 1 - Outside Eusaurischia?

You all know the big Eoraptor monograph was finally published as Sereno et al. (2013) containing the first detailed description of the material.  It may have taken 22(!) years from discovery to publication, but it's an excellent paper.  I previously examined the evidence Martinez et al. (2011) provided for Eoraptor being a sauropodomorph when they announced the idea in their Eodromaeus paper.  I found numerous coding issues and errors, such as composite codings and miscodings, which when partially corrected recovered a theropod Eoraptor instead, and also noted few of the proposed theropod characters had been included in their matrix.

Since then, we have the new basal theropod Daemonosaurus, redescriptions of basal sauropodomorphs Chromogisaurus and Pantydraco, redescriptions of all heterodontosaurids, redescriptions of the near/basal-dinosaurs Saltopus and Nyasasaurus, a version of Yates' analysis finding Eoraptor to be a sauropodomorph and more characters from Nesbitt's and Ezcurra's analyses finding it to be a theropod closer to Avepoda than herrerasaurids.  So let's explore the suggested evidence for each position for Eoraptor.  I'll list all novel suggested characters from each analysis, bolding the ones that seem valid.

Outside Eusaurischia (Sauropodomorpha+Theropoda)

 Padian et al. (1999)
1. centra of posterior cervical vertebrae (6-8) subequal in length to those of anterior dorsal vertebrae (refined by Langer, 2004).  This is untrue in Eoraptor, as cervicals 6-8 are 18-23 mm and the anterior dorsals are 16-17 mm.
2. third finger of the manus longer than second finger.  As this is true in Eodromaeus and Tawa, the opposite condition in known sauropodomorphs and avepods is probably convergent.  Herrerasaurus  also has III longer than II, while Guaibasaurus is like sauropodomorphs and avepods.  Though Tianyulong has an avepod-like condition, Heterodontosaurus and derived ornithischians have III longest, so this was more likely the basal condition in Ornithischia.
3. metatarsal I contacts tarsus.  Since this is also true in Sauropodomorpha, the authors had no reason to list it.  It only excludes Eoraptor from Avepoda.

Langer (2004)
4. subnarial premaxillary process extends posteriorly to the external naris.  This is also true in Daemonosaurus, making the basal condition for Theropoda ambiguous.  As Herrerasaurus has this state as well, it's even worse if herrerasaurids are theropods.
5. radius more than 80% of humerus length.  This is untrue in Eoraptor, which has a ratio of 74%.
6. manual ungual I shorter than metacarpal I.  This is untrue in Eoraptor, which has a ratio of 100%.  Note even if the ratio is actually barely in agreement, Eodromaeus and Tawa have short unguals I too, so it would be another character convergent in avepods and sauropodomorphs.
7. metacarpal III longer than metacarpal II.  Another character also present in Eodromaeus and Tawa, again making sauropodomorphs and avepods convergent. As with the digit length comparison, Guaibasaurus has longer II while Herrerasaurus has longer III.  Also while heterodontosaurids have longer II, other ornithischians and Saltopus have longer III, suggesting the former is convergence.
8. distal end of ischium unexpanded. This is untrue in Eoraptor.
9. medial margin of distal tibia not broader than lateral margin. This is untrue in Eoraptor.

Smith et al. (2007)
10. maxillary tooth count 12-18. Eoraptor has 17, while basal theropods have (9/10)-11 and basal sauropodomorphs don't preserve the area or have unreported counts.  As Saturnalia and Panphagia have 17 and 23 dentary teeth respectively, their maxillary counts were probably within the 12-18 range or higher, not lower as in basal theropods. Thus there is no shared derived state to differ from.
EDIT: Ugh, how could I forget Pampadromaeus?!  It has at least twenty maxillary teeth, proving my point.
11. lateral surface of anterior end of nasal along the posterior margin of the external naris flat.  Pantydraco and Daemonosaurus also lack this narial fossa, though Panphagia has it.  This means Theropoda is basally ambiguous while Sauropodomorpha is barely basally derived in having the fossa, so the character is not an unambiguous eusaurischian synapomorphy.  For what it's worth, Herrerasaurus also lacks the fossa.
12. posteroventral dentary process far posterior to posterodorsal process.  This is true in basal sauropodomorphs  (Panphagia, Pantydraco) and basal theropods (Eodromaeus, Tawa).  It's not true in ornithischians (Tianyulong, Heterodontosaurus, Eocursor) though, making the condition in Saurischia's outgroup ambiguous (given Silesaurus having the opposite condition).  Again for what it's worth, Herrerasaurus has the same condition as ornithischians.
13. foramen in the ventral part of the splenial absent.  This is difficult to code as the anterior splenial is thin and often broken.  In Sauropodomorpha, Panphagia has a foramen, Lamplughsaura is illustrated without one (though by Chatterjee, whose drawings are often idealized), Plateosaurus is polymorphic, and Lufengosaurus and Adeopapposaurus have it.  In Theropoda, Liliensternus lacks one, Dilophosaurus is illustrated as lacking one but seems to be anteriorly incomplete, and Ceratosaurus has one.  Thus the basal condition in either saurischian clade is unclear though more probably present in sauropodomorphs.  Ornithischians lack the foramen, as does Staurikosaurus though the latter has a poorly preserved mandible.
14. iliac supraacetabular crest shelf-like and short, extending primarily laterally.  This is also true in basal sauropodmorphs (Panphagia, Pampadromaeus) and basal theropods (Eodromaeus, Tawa).
15. ridge on lateral side of tibia for connection with fibula absent. This is untrue in Eoraptor

Yates (2007)
16. relationship between posterolateral process of the premaxilla and the anteroventral process of the nasal a broad sutured contact.  This is untrue in Eoraptor.
17. size and position of subnarial foramen small (no larger than adjacent maxillary neurovascular foramina) and positioned outside of narial fossa.  Basal theropods (Tawa, coelophysids, Dilophosaurus) lack a subnarial foramen, as do outgroups (ornithischians, Silesaurus).  Thus there is no obvious ancestral condition for the subnarial foramen, nor evidence theropods ancestrally had one.  Herrerasaurus does have this condition.
18. pointed posterolateral process of the nasal overlapping the lacrimal absent. This is untrue in Eoraptor.
19. length of middle to posterior cervical centra (6-8) no more than the length of the axial centrum. This is unknown in Eoraptor, as the axis is fragmentary.
20. laminae bounding triangular infradiapophyseal fossae on dorsal neural arches absent. This is untrue in Eoraptor.
21. transverse width of the first distal carpal less than 120% of the transverse width of the second distal carpal. This is unknown or untrue in Eoraptor, as distal carpal I is either unpreserved or diagenetically fused to the radiale in the left carpus and ~192% the width of distal carpal II. Notably, the basal theropods Eodromaeus and Tawa and basal sauropodomorph Efraasia have a small distal carpal I though, so this is not a eusaurischian character, though Heterodontosaurus does have a large distal carpal I so that ornithischians have an ambiguous basal state. Herrerasaurus has a small distal carpal I.

Martinez and Alcober (2009)
22. no caudosacral.  This is also true in basal theropods (Eodromaeus and Tawa) and ambiguous in sauropodomorphs (true in Pampadromaeus but not Efraasia and more derived taxa).  Guaibasaurus, Sanjuansaurus and Herrerasaurus also lack a caudosacral, though Staurikosaurus may have one.  Ornithischians have a caudosacral.
23. width of metacarpal I shaft less than 35% of length. This is untrue in Eoraptor.

Bittencourt Rodrigues (2010) also placed Eoraptor basal to Eusaurischia, but this paper has yet to be translated.

What's that?  No characters were bolded?  These turned out particularly bad, with almost half (10-11) not even being present in Eoraptor.  The others are basically all also found in taxa agreed to be basal theropods (11) and/or sauropodomorphs (7).  The best character is the absent anterior splenial foramen, which depends on illustration inaccuracy of a poorly preserved and seldomly exposed element.

Next up- is it a sauropodomorph?

References- Padian, Hutchinson and Holtz, 1999. Phylogenetic definitions and nomenclature of the major taxonomic categories of the carnivorous dinosaurs Dinosauria (Theropoda). Journal of Vertebrate Paleontology. 19(1), 69-80.

Langer, 2004. Basal Saurischia. In Weishampel, Dodson and Osmolska. The Dinosauria Second Edition. University of California Press. 861 pp.

Smith, Makovicky, Hammer and Currie, 2007. Osteology of Cryolophosaurus ellioti (Dinosauria: Theropoda) from the Early Jurassic of Antarctica and implications for early theropod evolution. Zoological Journal of the Linnean Society. 151, 377-421.

Yates, 2007. Solving a dinosaurian puzzle: The identity of Aliwalia rex Galton. Historical Biology. 19(1), 93-123.

Martinez and Alcober, 2009. A basal sauropodomorph (Dinosauria: Saurischia) from the Ischigualasto Formation (Triassic, Carnian) and the early evolution of Sauropodomorpha. PLoS ONE. 4(2), e4397.

Bittencourt Rodrigues, 2010. Revisao filogenetica dos dinossauriformes basais: Implicacoes para a origem dod dinossauros. Unpublished Doctoral Thesis. Universidade de Sao Paulo. 288 pp.

Martinez, Sereno, Alcober, Columbi, Renne, Montanez and Currie, 2011. A basal dinosaur from the dawn of the dinosaur era in Southwestern Pangaea. Science. 331, 206-210.

Sereno, Martinez and Alcober, 2013. Osteology of Eoraptor lunensis (Dinosauria, Sauropodomorpha). Journal of Vertebrate Paleontology. 32(Supplement to 6), 83-179.


  1. I know it's bitch putting Eusaurischia in it's proper clade! Trust me I've tried for years.I would like Prosauropoda to come back. Sauropodomorpha is to vast, like nobody wants to go there!
    As i'm sure you know most subfamilies and families in Guaibasauridae' Saturnalinae and Sauropodomorpha are almost Theropoda in design..And the (archosaur) Smok, which is as close to a
    Theropod as anything Iv'e seen.
    I hate to ask for help again, but can give your cladeogram on Sauropodomorpha. I respect your judgement
    E-mail: rb3wi

    Thank you,
    Bobby Williams

    1. Hi. Prosauropoda still exists, it's just very small right now (~5 genera). I agree Guaibasaurus itself may end up being a theropod, as in Bittencourt Rodrigues' (2010) thesis. Saturnalia seems pretty firmly sauropodomorphan though. I'm betting Smok will prove to be a pseudosuchian once described in detail and analyzed. Finally, my sauropodomorph cladogram is online here , but it's generally taken from others' analyses so doesn't represent any particular result of my work.

  2. Do you think any of these characters are potentially relevant in light of taxa like Tawa falling outside Eusaurischia?

    1. Of course, and that's one reason I stopped doing these kinds of posts and entries in the Database comments- in most situations every taxon affects every other so you can't just say character x supports taxonomic assignment 1 without a lot of caveats. Hence the reason for phylogenetic analyses in the first place, and I find it easier to grasp and less repetitive to instead list the number of steps needed for alternative placements in good analyses.