Nesbitt and his Characters – 9 – Dinosauria

Nesbitt (2011) and nearly all prior studies considered dinosaurs monophyletic and so does the large reptile tree. The only difference are the included taxa. The large reptile tree includes Trialestes, various silesaurids and poposaurids with Pisanosaurus, all within the Dinosauria. The large tree also nested sauropods and poposaurids with ornithischians in an herbivorous clade, Phytodinosauria. On the other hand, Nesbitt (2011) followed traditional nestings with Pisanosaurus as a basal ornithischian and sauropods closer to theropods (along with pterosaurs not far off). Poposaurids were nested with rauisuchians.

Again, apologies for many of the traits here for which I have not data. Often these traits are subtle and tucked away, not commonly illustrated.

Unambiguous dinosaur characters found by Nesbitt (2011) and others include:

1) supratemporal fossa [stf] present anterior to the supratemporal fenestra;
This trait is tentatively present in Silesaurus, which preserves the frontals. Effigia rotates the stf posteriorly and drops it into a fossa, but I haven’t seen the skull in occipital view. The stf of Lotosaurus appears to be depressed in a fossa.

2) epipophyses present in postaxial anterior cervical vertebrae; Seems to be present in Effigia (could use a second opinion here).

3) apex of deltopectoral crest situated at a point corresponding to more than 30% down the length of the humerus; This trait is indeed present in theropods, sauropods and ornithischians, but not present in Pisanosaurus or other poposaurids so far as data indicates (missing in Effigia and Poposaurus).

4) radius shorter than 80% of humerus length; Not present in Pampadromaeus, Herrerasaurus, Archaeopteryx, Scelidosaurus, Heterodontosaurus and others, but present on several dinos.

5) fourth trochanter a sharp flange; A good character for dinos, but also found in Turfanosuchus and Hesperosuchus. Also found in poposaurids.

6) fourth trochanter asymmetrical, with distal margin forming a steeper angle to the shaft; Sorry, I don’t have good data for this.

7) proximal articular facet for fibula of the astragalus and calcaneum occupies less than 0.3 of the transverse width of the element; Sorry, I don’t have good data for this.

Unambiguous synapomorphies Nesbitt (2011) first found include:

1) exocciptials do not meet along the midline on the floor of the endocranial cavity; Sorry, I don’t have good data for this.

2) proximal articular surfaces of the ischium with the ilium and the pubis separated by a large concave surface; Sorry, I don’t have good data for this.

3) cnemial crest arcs anterolaterally; Sorry, I don’t have good data for this.

4) distinct proximodistally oriented ridge present on the posterior face of the distal end of the tibia; Sorry, I don’t have good data for this.

Potential synapomorphies found by Nesbitt (2011) and others include:
1) concave ventral margin of the acetabulum of the ilium;  Sorry, I don’t have good data for this.

2) Concave articular surface for the fibula of the calcaneum; Sorry, I don’t have good data for this.

3) postfrontal absent; A good character for dinos, but also found in crocs beginning with Terrestrisuchus and higher pararchosauriforms plus Doswellia. Also found in poposaurids. In crocs and pararchosauriforms it seems apparent that the postfrontal and postrobital have fused.

4) posttemporal opening absent or less than half the diameter of the foramen magnum; Sorry, I don’t have good data for this.

5) three or two phalanges on manual digit IV; A good character for dinos, four on Pisanosaurus, Scelidosaurus, Lotosaurus, Effigia, Poposaurus.

6) posterior process of the jugal is split by the anterior process of the quadratojugal; In some theropods the jugal process is so gracile that it may be overlapped rather than split. This area is often damaged. In Scelidosaurus and Heterodontosaurus the jugal is not split.  This trait is also present in Lotosaurus and damaged or missing in other poposaurids.

Potential synapomorphies found originally by Nesbitt (2011) include:

1) three premaxillary teeth; Hmmm, I found four basally in dinos, reduced in Scelidosaurus and Heterodontosaurus and missing in several poposaurids. More than four teeth in the clade of Agilisaurus, Lesothosaurus and Scuttellosaurus.

2) premaxillary narial fossa expanded in the anteroventral corner of the naris; This appears to be a good character for dinos, but extends into Turfanosuchus and Decuriasuchus, Qianaosuchus.

3) concave emargination ventral to the proximal head of the femur; Sorry, I don’t have good data for this.

4) calcaneal tuber absent; This appears to be a good character for dinos, but does not include Pisanosaurus and poposaurids, which redeveloped the tuber, then lost it again in Silesaurus + Pseudolagosuchus. A tiny tuber is present in basal crocs. By convergence the tuber is also absent in all basal reptiles upt to Prolacerta, then absent again in the Proterochampsia.

5) interclavicle absent; A good character for dinos, including Lotosaurus and poposaurids. Absent by convergence in three other unrelated clades. Fused to sternum and clavicles in pteros.

6) proximal articular surface of the humerus separated by a gap from the deltopectoral crest. This trait may be correlated to deltopectoral crest apex located 30% down the humerus. As above, this trait is indeed present in theropods, sauropods and ornithischians, but not present in Pisanosaurus or other poposaurids so far as data indicates (missing in Effigia and Poposaurus).

I think that’s it for the Nesbitt characters. Much more work to be done. Convergence is rampant. 

Tomorrow we’ll go back to a variety of subjects.

As always, I encourage readers to see specimens, make observations and come to your own conclusions. Test. Test. And test again.

Evidence and support in the form of nexus, pdf and jpeg files will be sent to all who request additional data.

Nesbitt SJ 2011.
 The early evolution of archosaurs: relationships and the origin of major clades. Bulletin of the American Museum of Natural History 352: 292 pp.

9 thoughts on “Nesbitt and his Characters – 9 – Dinosauria

  1. Effigia okeeffeae, does it have:
    Tibia with distal end bearing caudolateral flange which abuts fibula?

    Your illustration implies you can disarticulate the fibua to do this; this illustration requires you to move the fibula away from its calcaneal joint to do so. So, no.

    Supratemporal fossa anterior to supratemporal fenestra?
    None. It doesn’t take a relativel sideways view or anything to project absence of this feature: the parietal or frontal must bear a depressed region bordering the rim of the supratemporal fenestra. As is visible in Nesbitt (2007, figs.2 & 11) no such depressed region is rostral to the supratemporal fenestra, and in fact the postorbital produces a dorsal, medially elongated process that forms a higher, thicker region instead of a shallow, depressed one. There is no rostral supratemporal fossa.

    Epipophyses in post-axial cervical vertebrae?

    Deltopectoral crest apex below 30% total length of humerus?
    Unclear, as the preserved humerus is broken into fragments that represent, using a mould of the humerus, less than half the total length of the element. Despite this, only a small proximal portion of the deltopectoral crest is preserved, and Nesbitt’s limited remarks on the topic imply the mould does not clarify.

    MDIV has 2-3 phalanges?
    The manus is incompletely preserved. Md-IV-1 is preserved, but has a small articular region for mdIV-2, but no indications for further.

    Does Heterodontosaurus tucki have a:
    Caudally bifurcate jugal, receiving rostral process of quadratojugal?

    This depends on your definition of “bifurcate.” In both skulls the preserve the jugal, the caudal, post-dorsal process region of the jugal is, indeed, bifurcate, so that a dorsal process extends along the dorsal margin of the quadratojugal, while a ventral process projects ventrally, rather than caudally. This my be homologous with the caudal orientation of the process and encapsulation of the rostral process of the quadratojugal. Heterodontosaurus tucki is unique among ornithischians in having a rostral process of the quadratojugal positioned extremely high on the bone relative to the bone’s position in the skull, with a ventrally projecting “caudal” process (quadrate ramus), forming a T-shape. This morphology is partly influencing the appearance of a gap between jugal and quadratojugal ventral margins, and the presence of a “flange”.

    I have little else to remark on, but some of these have to do with explicit definitions or changes that alter expression of states from the expected value in putative sister taxa. Effigia okeeffeae has so many character traits in common with poposaurid crurotarsans because, with the bulk of the data out there pointing to it, it is a poposaurid crurotarsan. Other features in convergence, such as parasagitallization of the gain, should lead to various traits including the reduction of outlever features of the knee, ankle, elbow and wrist, including the reduction of carpals and tarsals in size, reduced processes/tuberae, etc/ This is enforced by the more ventral orientation of the scapulocoracoid joint with the humerus, and the acetabulum with the femur and an inturned, medial inclination of the femoral caput and distinct caput for the humerus. This is why these features appear so commonly among these taxa, but not that they are evidence of explicit phylogenetic sister-group-ness-hood.

  2. re: Effigia skull – Wish we had an occipital view to see the extent of the fossa vs fenestra. The ridge you rightly refer to is the rim of the fossa that otherwise surrounds the fenestra. The back of the skull rotates posteriorly, distinct from all kin, so what we have here has evolved in a whole different direction perhaps leaving some legacy traits.

    re: Effigia ankle – I added a wee bit of cartilage between the elements, just a slip. No disarticulation. There is something growing toward the fibula from the tibia.

    re: Poposaurid curotarsans – indeed, Effigia is a poposaurid, but these belong with dinos that redeveloped the ankle tuber (for whatever reason). The “Extreme convergence” noted by Nesbitt is homology. The parasagittal-ization you refer to did not work with Postosuchus, which is 99% bipedal and a rauisuchian.

    I appreciate your comments. Currently moving all of the poposaurids to the Rauisuchia adds 41 steps to the large reptile tree, two more if moved close to Ticinosuchus.

    • “Postosuchus, which is 99% bipedal and a rauisuchian.”
      I would be terribly surprised if Postosuchus was 99% bipedal. It probably was occasionally, but considering the length of the trunk, the size of the skull, and the length of the forelimbs it probably wasn’t close to an obligate biped. Poposaurus is another story, of course.

      • Moreover, the elongated nature of the forelimb material described so far, and the structure of the elbow, suggests that it was in fact often on it’s forelimbs. It was not an Effigia-level faculative biped.

    • So far, I think, the weight of higher-character count, more rigorously tested analyses (yes, I’m dissing a bit on yours, you’ll forgive me) suggest that the crurotarsan aspects of poposaurs outweigh those of any apparent dinosaurian ones. These include, despite being convergently parasagittal and cursorial, crurotarsans such as poposauroids and “sphenosuchian” grade taxa retain, among other things: interclavicles; divided and well-developed clavicles; large and blocky proximal carpal bones with little to no prehensility in the manus; a nearly imperforate acetabulum with virtually no ventral contribution to the acetabulum’s ventral margin by the pubis; a calcaneal heel and no evidence for an ascending process of the astragalus or concordant fossa on the tibia incised into the cranial surface (rather than a crest that divides fibula from tibia); a fifth metatarsal with articulated digit; not to mention various cranial features such as divided vomers, a huge prefrontal that can contact the postorbital, a postfrontal, contact between the quadrate postorbital, and so forth. These features would all need to be redeveloped, despite no intermediates, to recover poposauroids into dinosaurs (anywhere in them), and especially more grievous when trying to dump any of them into ornithischians.

      There is no extraordinary evidence (no intermediate fossils showing the transition) which permit a typical basal dinosaur form that develops into the croc-like form that poposauroids represent. While post hoc reasnoning is being used correctly when applying phylogenetic analysis and trying to find the right characters to explain the output, other analyses are doing this despite continued embracing larger and larger reams of characters, splitting characters and recognizing correlated sets of features and evaluating them accordingly. It is a good rebuttal of your methods, especially in light of Naish’s criticism (as voiced by others) that phylogenetic analysis is a tool, but not the primary form of scientific analysis when discussing phylogeny: that actually occurs during evaluation of characters, bulk data, and including/excluding and forcing phylogenetic arrangements and analyzing things afterwards. It is simply a cluster analysis, and knowing how the various algorithms work and the data behaves in connection to this is the bigger part of such analysis. But … that’s all I have to say.

      • A major problem with the Nesbitt analysis is the exclusion of all Youngina/Youngoides specimens, which in the large reptile tree lie at the base of the Archosauriformes. With these specimens, plus the Choristodera the Archosauriformes split in two with Parasuchians and Proterochampsids into a separate and distinct clade also evolving bipeds in Tropidosuchus and Lagerpeton. Then, of course, there is the inclusion of the oddball pterosaurs, which has been beaten to death. As it stands there are several untenable sisters in that clade as I demonstrated in the very first over a year ago. I’m surprised that you didn’t criticize the Nesbitt results for those very obvious problems. By contrast, also illustrated in the fourth, my analysis produces a gradual evolution and acquisition of characters. And that, my friend, really is a model matching the reality. Please remember that the most primitive known dinosaurs, like Herrerasaurus, are well on their way away from the basal form. Trialestes, to your point, had large block proximal carpals. That’s key and it unites basal crocs, like Gracilisuchus, to basal dinos. If you see any mismatches among the current list of sisters on the large reptile tree, please let me know what they are and how you would shift them. Then we’ll compare the two hypotheses side-by-side in a future blog. Thank you for your comments, Jaime. I appreciate your expertise.

    • Considering that Postosuchus and Poposaurus material has often been mixed together (which makes one wonder about niche partitioning) it may be that manus (which looks somewhat like the one described from Poposaurus from GSENM) isn’t necessarily Postosuchus. I don’t have that 2008 paper in front of me right now, though, so I can’t see what they specifically describe. It does seem to be in contrast with, as Jamie says, what has been published before for it.

      • The hand is from another P. species from North Carolina. The rest is from Arizona. So, who knows? That’s what makes it fun. Something is almost always being overturned with new evidence.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.