The Protorosaurus Wastebasket

Back in  2009
Gottmann-Quesada and Sanders produced the first comprehensive study of Protorosaurus (Meyer 1832, Tatarian, Late Permian) in over a hundred years. Protorosaurus was one of the first fossil reptiles ever described (Spener 1710). According to Gottmann-Quesada and Sanders, “large numbers” of Protorosaurus specimens have been added to collections, Only one (Fig. 6), they say, preserves a complete skull.

Unfortunately 
Gottmann-Quesada and Sanders lumped several disparate genera under the genus Protorosaurus. Evidently the genus Protorosaurus has become a phylogenetic ‘wastebasket’ for a variety of protorosaurs and other reptiles in the Late Permian.

Figure 1. The lectotype of Protorosaurus identified by Gottmann and Sanders. See below for a reconstruction and comparisons.

Unfortunately
Gottmann-Quesada and Sanders consider Diapsida the ancestral clade for Archosauromorpha and Lepidosauromorpha. The large reptile tree (now 614 taxa) does not support that old paradigm. Their analysis is based on the data set of Dilkes (1998) “because he was the first to propose a paraphyletic Prolacertiformes.” Unfortunately for Gottmann-Quesada and Sanders the Dilkes study focuses on the basal rhynchosaur, Mesosuchus, a taxon completely unrelated to Protorosaurus in the large reptile tree. The Gottmann and Sanders tree is similar to that of Nesbitt et al. (2015) we just looked at with regard to Azendohsaurus.

Relying on someone else’s tree
has become more and more of a headache for paleontologists who keep chasing their tails with untenable and falsified cladograms.

Figure 2. Results of the most inclusive phylogenetic analysis of early archosauromorphs by Gottman-Queseda and Sanders. Note the separation of Protorosaurus and Prolacerta, the nesting of Protorosaurus with Megalancosaurus and the use of suprageneric taxa. This tree suffers greatly from too few specific taxa. Pamelaria is misspelled Palmeria, the least of the many problems with this tree.

In contrast,
the large reptile tree finds that Archosauromorpha and Lepidosauromorpha are basal reptile clades (with Gephyrostegus bohemicus of the Westphalian) nesting as a closest known sister to that as yet unknown, but close to Eldeceeon, a Viséan ancestor. The Diapsida, therefore, turns out to be diphyletic with lepidosaurs on one branch and archosaurs on the other, related to each other only through G. bohemicus.

Figure 3. The Protorosauria. nests two Prolacerta specimens and three Protorosaurus specimens, along with a scattering of others. Click to enlarge.

Getting back to Protorosaurs (taxa nesting with Protorosaurus)
they nest basal to the archosauriformes and both are derived from terrestrial younginiformes. Former  protorosaurs, like Macrocnemus and Tanystropheus now nest within the Lepidosauria between Rhynchocephalia and Squamata. This new paradigm has to start sinking in and permeating the paleo world.

Gottmann-Quesada and Sanders used
144 characters, 15 hand-picked terminal ungroup taxa, two hand-picked outgroup taxa. Bootstrap and Bremer values were considered “low.”

That compares to
228 characters and 610 taxa in the completely resolved large reptile tree with generally high to very high Bootstrap values throughout. All subsets remain fully resolved. That means deletion of taxa do not affect the remaining tree topology in the large reptile tree. And all derived taxa are preceded by series of taxa with gradually accumulating character traits — unlike other traditional trees, like the Dilkes/Gottman-Quesada and Sanders tree

Figure 4. Two taxa assigned to Protorosaurus by Gottmann-Quesada and Sanders. The lower one is the new electrotype (Fig. 1). The upper one nests closer to Pamelaria and is clearly not congeneric. See how reconstructions help? Some of this is not immediately apparent in the fossils themselves.

The Gottmann-Quesada and Sanders analysis (Fig. 2) 
nested Protorosaurus with the drepanosaurid Megalancosauru and away from Prolacerta. That should have been noticed as a red flag. One can only wonder how poorly these taxa were scored for such nestings to happen.

The large reptile tree nested Protorosaurus with Prolacerta and other protorosaurs.
Which analysis would you have more confidence in?

Figure 5. The putative Protorosaurus juvenile (in situ) is actually a large Permian Homoeosaurus.

A juvenile Protorosaurus?
Gottmann-Quesada and Sanders considered the Late Permian reptile IPB R 535 (Institut für Paläontologie, Unversität, Bonn) the first and only juvenile Protorosaurus.  I added it to the large reptile tree and recovered it rather securely as a large Homoeosaurus, a long-lived taxon otherwise known from Jurassic strata. This specimen adds to the small but growing number of known Permian lepidosaurs,

Figure 6. The WMsN-P47 specimen assigned to Protosaurus, but is closer to Pamelaria. The scapulocoracoid is not fused, as proven by one scapula flipped so that the dorsal rim is in contact with its corticoid. I’ve always wondered about that inconsistency. A hi-rez image and DGS solved that problem.

The WMsN-P47 specimen that Gottmann-Quesada and Sanders assigned to Protorosaurus (Fig. 4) is actually closer to Pamelaria (see figure 7) in the large reptile tree. This specimen is too distinct to be lumped with Protorosaurus.

Gottmann-Quesada and Sanders reported
that Protorosaurus has seven cervicals. I found evidence for eight without seeing the fossil first hand. DGS techniques enable the identification and reconstruction of skull elements in the pre-Pamelaria specimen (Fig. 6) previously considered too difficult to attempt.

Figure 7. Several protorosaurs to scale including Pamelaria, Protorosaurus, Prolacerta, Malerisaurus, Boreopricea and Jaxtasuchus. Click to enlarge.

It is unfortunate
that Gottmann-Quesada and Sanders lumped all of their Protorosaurus specimens together when there is clearly a diversity of morphologies and sizes here. They did not feel the need to perform a phylogenetic analysis on the individual specimens or to create more than a single skull reconstruction (Fig. 8).

And I apologize
for earlier reconstructions created out of more than one specimen. I should never have created chimaeras. They really mess up phylogenetic analyses.

Figure 8. GIF animation of the NMK S 180 specimen assigned to Protorosaurus by Gottmann and Sanders. I was able to tease out certain palatal bones ignored by them. Reconstruction by Gottman and Sanders.

Gottmann-Quesada and Sanders mention Peters (2000)
due to that paper adding pterosaurs to the list of then considered prolacertiformes (later corrected in Peters 2007). They report, “this analysis suffers from over interpretation of poorly preserved fossils.” This is more professional BS. Either one look or rigorous examination of the fossils studied in Peters (2000) reveals that all include soft tissue and preserve every bone in articulation, which is the definition of “exquisitely preserved.”

I can only imagine
that, like Hone and Benton (2007, 2009) Gottmann-Quesada and Sanders felt the need to cite relevant literature, but shuddered at the prospect of actually dealing with non-traditional results. To their point on interpretation, mistakes were made in Peters (2000), some from under-interpretation and some from naiveté. That is why I submitted corrections (which were rejected), including Peters 2007 (which was published as an abstract). ReptileEvolution.com/cosesaurus.htm and links therein publicly repair the errors found in Peters (2000).

Gottmann-Quesada and Sanders report
the only trait uniting the Prolacertiformes [protorosaurs] are the elongated mid-cervical vertebrae. Unfortunately this trait also appears in several other clades within the Reptilia. The large reptile tree likewise did not find a single common character in the protosaurs. As in so many other clades it is the suite of traits that lump and separate them.

References
Gottmann-Quesada A and Sander PM 2009. A redescription of the early archosauromorph Protorosaurus speneri Meyer, 1832, and its phylogenetic relationships. Palaeontographica Abt. a 287: 123-220.
Meyer H von 1832. Palaeologica zur Geschichte der Erde und ihrer Geschöpfe. Verlag Siegmund Schmerber, Frankfurt a.M. 560 pp.
Peters D 2000. A redescription of four prolacertiform genera and implications for pterosaur phylogenesis. Rivista Italiana di Paleontologia e Stratigrafia 106: 293-336
Peters D 2007. The origin and radiation of the Pterosauria. Flugsaurier. The Wellnhofer Pterosaur Meeting, Munich 27.
Seeley K 1888. Research on the structure, organisation and classification of the fossil Reptilia 1. On the Protorosaurus speneri (von Meyer). Philosophical Transactions of the Royal Society, London B 178, 187–213.
Spener CM 1710. Disquisitio de crocodilo in lapide scissilli expresso, aliisque Lithozois. Misc. Berol. ad increment. sci., ex scr. Soc. Regiae Sci. exhibits ed. IL92-110.

Azendohsaurus postcrania

Figure 1. Azendohsaurus skull reconstructed with two premaxillary teeth, not four.

The new paper
on Azendohsaurus (Dutuit J-M 1972, Middle to Upper Triassic, Figs. 1-5) post-crania (Nesbitt et al. 2015, Fig. 2) has been eagerly awaited. The skull was published five years ago (Flynn et al. 2010). The description of the post-cranial bones is excellent, as was the earlier description of the cranial bones.

The is definitely a different sort of reptile…at first glance.

The big phylogenetic question is…
is Azendohsaurus closer to Pamelaria (the protorosaur archosauromorph)? Or closer to Trilophosaurus (the rhynchocephalian lepidosaur)?  Or closer to Sapheosaurus and Noteosuchus (two rhynchocephalian lepidosaurs)? Or closer to Eohyosaurus or Mesosauchus (yet other rhynchocephalians)?

Let’s test one analysis against another.

Figure 2. Reconstruction of Azendohsaurus from Nesbitt et al. 2015. Length: a bit over one meter. Manus on left. Pes on right. Click to enlarge. No matter which clade this nests in, it is an oddity. But at present it is less odd nesting with Trilophosaurus and Sapheosaurus according to the input traits.

Unfortunately too few taxa were tested in the Nesbitt et al. study. 
Nesbitt et al. tested 29 taxa and found that Azendohsaurus nested between Teraterpeton + Trilophosaurus and Pamelaria at the base of their Archosauromorpha (Fig. 3). Several mismatches occur here with great phylogenetic distances between purported sister taxa. The tree also mixes lepidosauromorphs with archosauromorphs, once again demonstrating convergence within the two major clades of reptiles.

Figure 3. The Nesbitt et al. 2015 cladogram nesting Azendohsaurus between Trilophosaurus + Teraterpeton and Pamelaria. This is not supported by the large reptile tree. Here green taxa are lepidosauromorphs. Black taxa are archosauromorphs. The proposed Allokotosauria is diphyletic. So is the proposed Azendohsauridae according to the large reptile tree. The black taxa are improperly included in this too small study on Azendohsaurus relationships.

By contrast,
the large reptile tree tested 610 taxa and maintained a long-standing and fully tested sisterhood between Trilophosaurus and Azendohsaurus. Furthermore, these two nested within the Rhynchocephalia at the transition to Mesosuchus, Eohyosaurus and the rhynchosaurs on the lepidosauromorph branch of the reptile (amniote) tree. Ancestral sisters include Sapheosaurus, Noteosuchus and tiny Leptosaurus (Fig. 6), taxa not listed in the Nesbitt et al. tree. Pamelaria continues to nest as a derived protorosaur on the archosauromorph branch of the reptiles.

Shifting
Azendohsaurus + Trilophosaurus + the rhynchosaurs and kin over to Pamelaria raises the MPTs from 9109 to 9127, a pretty small bump considering the great phylogenetic distance. The number drops to 9124 nesting this clade with Protorosaurus or the BPI 375 specimen of Youngina. Such similarity is due to convergence.

Notably,
Pameleria does not have two parallel rows of large teeth as in rhynchocephalians including Azendohsaurus and rhynchosaurs. Trilophosaurus does something odd with not two, but three laterally aligned cusps.

Figure 4. Rhynchocephalian subset of the large reptile tree with Azendohsaurus highlighted.

The nesting of rhynchosaurs and Trilophosaurus as basal archosauromorphs close to or within the protorosaurs has been a long standing problem in paleontology. They do indeed evolve away from the basic rhynchocephalian bauplan seen in Sphenodon.

Reducing the large reptile tree taxon list to that of Nesbitt et al. 2015
recovers a tree (Fig. 5) that again splits up archosauromorphs (inverted type) and lepidosauromorphs (black type).

Figure 5 A subset of the large reptile tree matched to the Nesbitt et al. 2015 taxon list. Here the tree mixes taxa from the two major branches demonstrating the convergence of the derived taxa and the importance of including a large gamut (610) of tested and verified taxa rather than fewer than 30 favorites or traditional guesses. Inverted taxa are archosauromorphs. Others are lepidosauromorphs in the large reptile tree. Note that both Pamelaria and Terterpeton are only one node away from the Azendohsaurus clade here.

 

With an odd reptile like Azendohsaurus
it is necessary to use a large gamut cladogram, like the large reptile tree, to test all the possibilities and relationships, leaving out virtually no possibilities.

Figure 6. Azendohsaurus compared to sister taxa and putative sister taxa including Trllophosaurus, Pamelaria, Teraterpeton, Sapheosaurus and Leptosaurus. Diandongosaurus is ghosted and not related. Azendohsaurus nests with Trilophosaurus in both studies. Even so it is quite distinct.

Along the way
I learned more about Trilophosaurus (Fig. 7) by going to photographs of the material after trusting published reconstructions that combined the anterior skull specimen with a mismatched posterior skull specimen.What we’ve gotten used to  seeing is a chimaera.

Figure 7. Trilophosaurus skulls, right side flipped. Note the deep jugal on two and the shallow jugal on the third. Also note the postjugal bone (deep blue), filling in the posterior jugal notch. This is a novel ossification. The asymmetry between the left and right quadrate/jugal suture appears to be natural. The palatine teeth align with the maxillary teeth, a unique trait.

I also learned
that the published Azendohsaurus premaxilla has a bit of maxilla on it (Fig. 1), reducing the premaxillary tooth count from four to two. The original researchers considered the crack the suture. No sister taxa have four teeth in the premaxilla. All have two teeth including the shallow jugal specimen of Trilophosaurus with two vestigial teeth in the premaxilla.

Figure 8. Eohyosaurus nests as a sister to the Trilophosaurus-Azendohsaurus clade.

Basal rhynchocephalia
have teeth ankylosed (fused) to the bone. In some cases the teeth are bone. Apparently when rhynchocephalians became phylogenetically miniaturized in tiny Leptosaurus, neotony re-produced regular socketed teeth of the sort one also sees in Eohyosaurus, Mesosuchus, Sapheosaurus and Azendohsaurus.

Nesbitt et al. report
“Teasing apart homology from homoplasy of anatomical characters in this broad suite of body types remains an enormous challenge with the current sample of taxa.”

Indeed that sample of taxa has to be greatly increased.
610 taxa demonstrate this amply. 29 is just too few. Too many actual sister taxa were overlooked and excluded in the Nesbitt et al. analysis. They relied on tradition rather than testing when oddly matched sister taxa nested with one another on their cladogram.

Even so,
as can be seen by the reconstructions (Fig. 5), there is still a great deal of phylogenetic distance between tested sisters. The large reptile tree minimizes this, but the distances still remain great. New discoveries will help fill these gaps, but the correct inclusion group must be used. The tree subset that includes protorosaurs and basal archosauriforms (Fig. 9) does not include Azendohsaurus, Trilophosaurus, rhynchosaurs or tanystropheids (which are all lepidosauromorphs).

Figure 8. Sapheosaurus GIF animation. This smaller ancestral sister to Azendohsaurus was overlooked and excluded by the Nesbitt et al. study.

Nesbitt et al. created two suprageneric clades.
Unfortunately the proposed clade Allokotosauria is diphyletic. So is the proposed clade Azendohsauridae according to the large reptile tree.

Figure 9.  Subset of the large reptile tree focusing on the Protodiapsida, the Diapsida, Marine Younginiformes and Terrestrial Younginiformes, including Protorosaurs and basal Archosauriformes.
Click to enlarge. Azendohsaurus, Trilophosaurus, tanystropheids and rhynchosaurs do not nest here as they do in the smaller gamut Nesbitt et al. cladogram, which improperly included them.

References
Dutuit J-M 1972. Découverte d’un Dinosaure ornithischien dans le Trias supérieur de l’zhtlas occidental marocain. Comptes Rendus de l’Académie des Sciences à Paris, Série D 275:2841-2844.
Flynn JJ, Nesbitt, SJ, Parrish JM, Ranivoharimanana L and Wyss AR 2010. A new species of Azendohsaurus (Diapsida: Archosauromorpha) from the Triassic Isalo Group of southwestern Madagascar: cranium and mandible. Palaeontology 53 (3): 669–688. doi:10.1111/j.1475-4983.2010.00954.x .
Nesbitt, S, Flynn J, Ranivohrimanina L, Pritchard A and Wyss A 2013. Relationships among the bizarre: the anatomy of Azendohsaurus madagaskarensis and its implications for resolving early archosauromroph phylogeny. Journal of Vertebrate Paleontology abstracts 2013.
Nesbitt SJ, Flynn JJ, Pritchard AC, Parrish JM, Ranivoharimanana L and Wyss AR 2015. Postcranial osteology of Azendohsaurus madagaskarensis (?Middle to Upper Triassic, Isalo Group, Madagascar) and its systematic position among stem archosaur reptiles. Bulletin of the American Museum of Natural History 398: 1-126.

wiki/Azendohsaurus

 

 

Triassic gastric pellet semi-reconstructed, better this time…

A while back
Dalla Vecchia, Wild and Muscio (1989) described a small pellet (MFSN 1891, Fig. 1) of Late Triassic bones from the Dolomia di Forni Formation of Firuli (NE Italy) as a small jumble of pterosaur bones. They tentatively referred it to Preondactylus, the only pterosaur known at the time from that formation. This was an early work for all three paleontologists.

Following the original paper
Earlier I attempted a reconstruction of the elements based on the pterosaur model. I recognized then that it didn’t turn out too well. I was working from the original drawings. Now new data has been published and a new hypothesis has been put forth that makes much more sense.

Figure 1. Several views of the Triassic gastric pellet formerly considered pterosaurian, but now considered langobaridsaurian. Elements from a surface photo, a microCT scan of the opposite side still buried in matrix, and DGS colors. Not all bones have been colored here, but employed colors are assembled in figure 2. The long cervical at upper left is 1 cm long. So is the scale bar. The pellet is about 5 cm wide.

Recently 
Holdago et al. (2015) redescribed the pellet in much greater detail using microCT acquisition. They concluded “The best candidate for the pellet is not a pterosaur, but a protorosaurian like Langobardisaurus.  Therefore, the skeletal remains could belong to a still unknown small reptile with procoelous dorsal vertebrae, rather elongate and probably procoelous cervical vertebrae with low neural arch and spine, filiform cervical ribs, at least some dicephalous dorsal ribs, elongated and hollow limb bones, and no osteoderms.”

They did not attempt a reconstruction,
so I do so here (Fig. 2) following the hypothesis that the elements belong to a langobardisaur (contra Holdgago, et al., not a protorosaur but a tritosaur lepidosaur).

Figure 2. The elements of MFSN 1891 assembled to form a langobardisaur in a bipedal pose. Some langobardisaurs have a very long neck, slender limbs and a short tail. Lots of guesswork here.

Lots of guesswork here. 
Everything is tentative. The toes could be ribs. Lots of slivers and scraps left over. More complete langobardisaurs (Fig. 3) have 8 cervicals, but they are related to tanystropheids, with 13 cervicals. Renesto et al. (2002) considered langobardisaurs as likely facultative bipeds in the manner of the many extant facultative bipedal lizards, all with sprawling hind limbs.

Figure 3. Langobardisaurus tonneloi reconstructed. Note the cosesaur-like pectoral girdle.

MFSN 1891 needs to be dissembled
in high resolution, then reassembled like a puzzle. I’d like to help if possible. Here (Fig. 2) is a first draft lo rez example leading to others of greater detail in the future. Worthwhile taking another look at the pes (Fig. 3) which greatly resembles a basal pterosaur pes with that elongate p5.1. It resembles a pterosaur pes because these two taxa are related (Peters 2000).

Figure 4. Click to view full scale on a 72 dpi screen. Tanystrachelos compared to the gastric pellet lepidosaur. The large hemal arches on the gastric pellet are the genesis of the paddle-like hemal arches on Tanytrachelos and Tanystropheus.

Compared to the tritosaur Tanytrachelos (Fig. 4)
the gastric pellet reptile has a similar number of cervicals, but longer limbs and longer cervicals. Are we seeing the origin of Tanystropheus (Fig. 5) here? Or a hatchling? The large hemal arches appear to have homologs in Tanytrachelos and Tanystropheus.

Figure 5. Tanystropheus and kin going back to Huehuecuetzpalli. Two scales here, one yellow, one white.

Then we have Fuyuanssaurus, 
a tiny tanystropheid (Fig. 6) about twice the size of the gastric pellet reptile. Unfortunately we don’t know if it was long-legged or not. Notably the skull elements of Fuyuansaurus, which we looked at earlier here were all quite slender. This is the model we should use for the gastric pellet lizard until data suggests another model.

Figure 6. Click to enlarge. Reconstruction of Fuyuanasaurus. Fraser et al. identified a strange circular object as the pubis, but no sister taxa have a circular pubis. Here it is tentatively ID’d as an egg because a standard pubis is found nearby.

References
Dalla Vecchia FM, Wild R and Muscio G 1989. Pterosaur remains in a gastric pellet from Upper Triassic (Norian) of Rio Seazza valley (Udine, Italy). Gortania 10: 121–132.
Holgado B, Dalla Vecchia FM, Fortuny J, Bernardini F and Tuniz C 2015. A Reappraisal of the Purported Gastric Pellet with Pterosaurian Bones from the Upper Triassic of Italy. PLoS ONE 10(11): e0141275. doi:10.1371/journal.pone.0141275
Peters D 2000. A Redescription of Four Prolacertiform Genera and Implications for Pterosaur Phylogenesis. Rivista Italiana di Paleontologia e Stratigrafia 106 (3): 293–336.
Renesto S, Dalla Vecchia FM, Peters D. 2002. Morphological evidence for bipedalism in the Late Triassic prolacertiform reptile Langobardisaurus. In: Gudo M, Gutmann M, Scholz J, editors. Concepts of functionalengineering and constructional morphology: biomechanical approaches on fossil and recent organisms. Senckenb Lethaea 82(1): 95–106.

 

What is Tasmaniosaurus?

Updated September 20, 2014 with a new reconstruction and nesting for this taxon.

Revised skull reconstruction of Tasmaniosaurus nesting at the base of the Erythrosuchia.

Ezcurra 2014 considered the tiny Early Triassic archosauriform Tasmaniosaurus traissicus (Camp and Banks 1978) a tiny proterosuchid, following the original assessment.

Previous authors thought the maxilla was exposed in lateral view. If so, it has a maxillary fossa. Ezcurra thought the maxilla was exposed in medial view. Proterosuchids do not have a maxillary fossa. If the premaxilla was not downturned, Tasmaniosaurus is anything but a proterosuchid. If the maxilla is exposed in lateral view, Tasmaniosaurus is anything but a proterosuchid.

The caudal vertebrae are very long, longer than 3x their width, very un-proterosuchid like. Seven in a row have no neutrals spines.

The interclavicle of Tasmaniosaurus is T-shaped with a very long and slender posterior process. Among archosauriforms, only Euparkeria has a T-shaped interclavicle. Many are I-shaped.

The femur and tibia/fibula are short and robust, so no possible biped here and the Early Triassic is a little too early for bipeds.

Tasmaniosaurus is tiny. About the size of Youngina and Euparkeria, much smaller than any known proterosuchid or erythrosuchid.

In phylogenetic analysis (not updated online yet), Tasmaniosaurus nests at the base of the Erythrosuchidae, as a sister taxon to Fugusuchus + Revueltosaurus. So, another miniaturized taxon nests basal to a large clade.

References
Camp CL, Banks MR 1978. A proterosuchian reptile from the Early Triassic of Tasmania. Alcheringa 2: 143–158.
Ezcurra MD. 2014. The Osteology of the Basal Archosauromorph Tasmaniosaurus triassicus from the Lower Triassic of Tasmania, Australia. PLoS ONE 9(1):e86864. doi:10.1371/journal.pone.0086864

The base of the Diapsida: bipeds, quadrupeds and swimmers

Earlier we looked at how often bipeds give rise to aquatic taxa. Today we’ll look at the base of the Diapsida, the clade that includes Petrolacosaurus, enaliosaurs, protorosaurs and archosauriforms and does not include lepidosaurs (which are not related and arrive at the diapsid configuration by convergence).

Figure 1. The base of the archosauromorph Diapsida, compressed to delete derived Enaliosauria. Here Eudibamus, Spinoaequalis, Tangasaurus and Thadeosaurus are basal to several major clades.

The great diapsid radiation
begins with these few taxa derived at the base of their several clades: Eudibamus, Spinoaequalis, Tangasaurus and Thadeosaurus (Fig. 2). Milleropsis is the proximal outgroup.

Eudibamus and Spinoaequalis nest at the base of the araeoscelids, which soon became extinct. The two Tangasaurus nest at the base of the aquatic enaliosaurs, which became extinct at the close of the Cretaceous. Thadeosaurus nests at the base of the protorosaurs and archosauriforms which are alive today in the form of birds and crocs.

Figure 2. Basal archosaurmorph diapsids, including Eudibamus, Spinoaequalis, two Tangasaurus and Thadeosaurus to scale. These are all sister taxa. The tangasaurs are basal to the aquatic enaliosaurs. Thadeosaurus is basal to protorosaurs and archosauriforms. Eudibamus and Spinoaequalis are basal to araeoscelids.

What do they all have in common?
These four Permian taxa are all about the same size and have a lizardy appearance, but they are not related to lizards.

The hind limb and foot are larger than the forelimb and hand. Pedal digit 1 is quite short and pedal digit 5 is quite long. The ilium is elongated, chiefly posteriorly. The limbs are gracile in most cases, the long-neck Tangasaurus the exception. The humerus is much broader distally in all. Hemal arches are deep in all except Eudibamus. The pubis has a dorsal process in all taxa. The manus is sub equal in length to the ulna. The pes is subequal in length to the tibia. None of these taxa had large teeth.

Figure 2. Tangasaurus specimens in dorsal view. They are both wide, like pancakes, with very wide anterior caudal verts. Those are ribs on the holotype (above), gastralia on the long neck specimen (below). Those large caudal ribs (transverse processes) anchor strong hind limb muscles. Those large coracoids and that large sternum anchor strong forelimb muscles.

Differences?
The skull is poorly known in several of these taxa. The cervicals are robust in most cases, Spinoaequalis and the Tangasaurus holotype are the exceptions. The scapulocoracoid is fused only in Thadeosaurus.

Speaking of the skull…
Currie (1982) reports more than 300 specimens of Tangasaurus and the related but more aquatic Hovasaurus are known, but none preserve the entire skull.

These taxa are rarely studied but they are key basal taxa in each of their clades and united by their diapsid ancestry. Probably all were active and speedy insect-eaters, whether terrestrial or aquatic.

Pedal digit 5
The lateral toe is much longer in these taxa, inherited from Milleropsis. It remains long in enaliosaurs, like mesosaurs and thalattosaurs. Pedal digit 5 becomes much shorter in Eudibamus (Fig. 1) and other araeoscelids and, by convergence, following Thadeosaurus and its terrestrial descendants.

Figure 1. Basilisk walking on water.

A modern analogy?
The extant basilisk runs bipedally through water, keeping its torso above the surface.  From Wikipedia, “Once a basilisk submerges, it continues swimming. Although this lizard stays close to water to escape terrestrial predators, it swims only when necessary because some other aquatic animals would eat the basilisk given the chance.” The soft tissue crests are sexually dimorphic.

On the the other hand…
Eudibamus was found in an upland paleograben containing no aquatic vertebrates.

On the other-other hand…
Thadeosaurus was found in split nodule in a rapidly filling deep rift valleys in Madagascar, some open to the sea. The presence of oolites replaced with collophane suggests a rich phosphate source, such as deep marine upwellings, similar to the situation of Galapagos marine iguanas. Milleropsis was also found in a split nodule that contained several specimens all living together, something that likewise occurred in Heleosaurus, its phylogenetic ancestor.

References
Berman, DS, Reisz RR, Scott D, Henrici AC, Sumida SS and Martens T 2000. Early Permian bipedal reptile. Science 290: 969-972.
deBraga M and Reisz RR 1995. A new diapsid reptile from the uppermost Carboniferous (Stephanian) of Kansas. Palaeontology 38 (1): 199–212. palass-pub.pdf
Carroll RL 1976. Galesphyrus capensis, a younginid eosuchian from the Cistephalus zone of South Africa. Annals of the South African Museum 72: 59-68.
Carroll RL 1981. Plesiosaur ancestors from the Upper Permian of Madagascar. Philosophical Transactions of the Royal Society London B 293: 315-383
Currie PJ 1984. Ontogenetic changes in the eosuchian reptile Thadeosaurus. Journal of Vertebrate Paleontology 4(1 ): 68-84.
Currie P 1982. The osteology and relationships of Tangasaurus mennelli Haughton. Annals of The South African Museum 86:247-265. http://biostor.org/reference/111508
Haughton SH 1924. On Reptilian Remains from the Karroo Beds of East Africa. Quarterly Journal of the Geological Society 80 (317): 1–11.
Reisz RR, Modesto SP and Scott DM 2011. A new Early Permian reptile and its significance in early diapsid evolution. Proceedings of the Royal Society B 278 (1725): 3731–3737.

wiki/Tangasaurus
wiki/Thadeosaurus
wiki/Spinoaequalis
wiki/Eudibamus

Oldest archosauromorph/archosauriform – svp abstracts 2013

From the abstract
Ezcurra et al. 2013 wrote, “Archosauromorphs include all diapsids closer to crocodiles and birds than to lepidosaurs. The group has a very rich Mesozoic and Cenozoic fossil record, but the Paleozoic record is restricted to a handful of Late Permian specimens. The most informative Permian archosauromorph so far discovered is Protorosaurus speneri from the middle Late Permian of Western Europe. In addition, there are several less well known putative archosauromorphs from Russia and Africa. We review these records here and include several of them in a quantitative phylogenetic analysis for the first time. This phylogenetic analysis included a broad taxonomic sampling of basal synapsids, basal diapsids and saurians. We could not find archosauromorph apomorphies in a supposed Late Permian proterosuchid cervical vertebra from South Africa (Bernard Price Institute for Palaeontological Research specimen BP/1/4220), and consider this specimen to belong to an indeterminate amniote. BP/1/4220 possesses striking features that are not present in other amniotes of which we are aware, such as posteriorly extended, wide and almost horizontally oriented accessory processes between the postzygapophyses. A problematic reptile (University Museum of Zoology, Cambridge specimen UMZC T836) from the Late Permian of Tanzania, first described in the 1950s, was recovered in the phylogenetic analysis as a protorosaur at the base of Archosauromorpha and is probably diagnosable as a new species. The position of UMZC T836 within Archosauromorpha is supported by the presence of three well-developed laminae in the cervico-dorsal neural arches and the absence of a humeral entepicondylar foramen. The supposed protorosaur Eorasaurus olsoni from the middle Late Permian of Russia was recovered within Archosauriformes, being more closely related to crown archosaurs than to proterosuchids, implying that this species may be the oldest known archosauriform. However, the fragmentary nature of the known material of this taxon and the low character support for this position means that this identification is currently tentative. Archosaurus rossicus from the latest Permian of Russia was found to be more closely related to Proterosuchus fergusi than to other archosauromorphs and represents a valid species. The revision conducted here suggests a minimum fossil calibration date for the crocodile-lizard split of 254.7 Ma. The occurrences of Protorosaurus speneri close to the paleo-Equator and UMZC T836 in high paleolatitudes of southern Pangea imply a wider paleobiogeographic distribution for archosauromorphs during the Late Permian than previously appreciated.”

Notes
If the Archosauromorpha is indeed all diapsids closer to birds and crocs than to lizards, then the large reptile tree, as discussed earlier, divides all reptiles into the new Archosauromorpha and the new Lepidosauromorpha, both of which contain diapsid-grade reptiles as subclasses. So, there’s a flip-flop here between overarching clades and subclades. That means the lepido-archo split occurred before the advent of Westlothiana, the oldest known archosauromorph at 338 mya in the Visean, Mississippian (early Carboniferous), itself a taxon derived from a sister to the most primitive archosauromorphs, Brouffia and Casineria.

Archosauriforms, Younginids and Protorosaurs
What Ezcurra et al. are really looking at are basal Archosauriformes, as mentioned in their abstract, but in the large reptile tree this clade is derived from younginids, which in turn are derived from protorosaurs and thadeosaurs. Youngina BPI 3859 and Thadeosaurus from the Late Permian 260 mya were coeval with Protorosaurus, so the last common ancestor of these three predates them, probably back to 280-300 mya. Like Garjainia and Proterosuchus, Archosaurus is also derived from Youngina. At about this time the major split between aquatic and terrestrial archosauromorphs took place as derived mesosaurs appear at 290 mya.

Lepidosaurs also appear in the Late Permian 255 mya in the form of Lacertulus coeval with Paliguana, the most primitive Lepidosauriform… no relation to the oldest archosauriforms.

References
Ezcurra MD, Butler R and Scheyer T 2013. The Permian archosaurmorph record revisited: a new species from Tanzania and the potentially oldest archosauriform. Journal of Vertebrate Paleontology abstracts 2013.

Jaxtasuchus – a new protorosaur, not a doswelliid

A new paper by Schoch and Sues (2013)
introduces a new armored archosauriform with long teeth, Jaxtasuchus salomoni (Fig. 1). It was considered semi-aquatic because it was found in Middle Triassic mudstones along with amphibians, crustaceans, and mollusks. Several incomplete skeletons are known. Schoch and Sues (2013) ran a phylogenetic analysis of 17 taxa that nested Jaxtasuchus with doswelliids, which were similarly armored. Unfortunately, that tree also nested several strange-bedfellows together, including Mesosuchus with Prolacerta, Vancleavea with Chanaresuchus, Parasuchus with Stagonolepis and Scleromochlus with Marasuchus none of whom resemble their putative sisters.

Maybe not a doswelliid
Adding what little is know of Jaxtasuchus to the large reptile tree nests it firmly with Pamelaria, a protorosaur. Protorosaurs are known for their elongated necks, but not for their armor.

In any case, it takes 10 more steps to move Jaxtasuchus to Prolacerta (which was included in the Schoch and Sues (2013) phylogenetic analysis) and 14 steps to move Jaxtasuchus to Doswellia. If valid, the long teeth and armor of Jaxtasuchus would be protorosaur autapomorphies that add new variety to this clade.

Figure 1. Reconstruction and restoration of the skull and neck of Jaxtasuchus (based on Schock and Sues 2013), along with scattered armor. The long neck and slender cervical ribs are protorosaur traits. The antorbital fenestra is shared with Pamelaria (fig. 2). No other known protorosaur has such long teeth and armor.

A new fifth instance of an antorbital fenestra!
Perhaps even more exciting than armor, Jaxtasuchus was described with an antorbital fenestra lacking a fossa. Doswellia (Heckert et al. 2012) likewise has a tiny antorbital fenestra, but not similar in design. However, a reexamination of Pamelaria reveals a very similar maxilla to Jaxtasuchus, which means it also had a previously overlooked antorbital fenestra (Fig. 2). Together these two up the total number of novel inventions of the antorbital fenestra from four to five. That’s a big deal.

Figure 2. The skull of Pamelaria from Sen 20003, with the maxilla highlighted in green. The maxilla appears similar to that in Jaxtasuchus in having an antorbital fenestra. Teeth are only present along the posterior portion of the maxilla.

Restoring the manus and pes
I reconstructed the pes and manus of Jaxtasuchus using PILs (parallel interphalangeal lines). On both the manus and pes proximal phalanges were all longer and distal phalanges were all subequal.

Figure 3. The manus and pes of Jaxtasuchus restored. Along with the cervicals (Fig. 1), these are among the most complete segments known of this reptile.

The osteoderms have a longer history
The osteoderms of Jaxtasuchus were previously interpreted as coming from temnospondyl amphibians or aetosaurs. Remains of Jaxtasuchus have been found in five localities and have been considered common. Now, courtesy of Schoch and Sues (2013) we know enough about it to consider it an armored predator with an antorbital fenestra. Thanks to the large reptile tree, which includes virtually every basal reptile clade, we can consider Jaxtasuchus a new protorosaur, rather than a doswelliid.

Figure 4. Click to enlarge. Jaxtasuchus reconstruction with armor. The small limbs might suggest a sinuous mode of locomotion, but the armor would argue against that. Perhaps this was a sit-and-wait predator, convergent with tanystropheids. The skull and neck specimen appear to come from a larger one than the post-crania, hence the estimate to match. 

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.

References
Heckert AB, Lucas SG and Spielmann JA 2012. A new species of the enigmatic archosauromorph Doswellia from the Upper Triassic Bluewater Creek Formation, New Mexico, USA”. Palaeontology (Blackwell Publishing Ltd) 55 (6): 1333-1348.
Sen K 2003. Pamelaria dolichotrachela, a new prolacertid reptile from the Middle Triassic of India. Journal of Asian Earth Sciences 21: 663–681.
Schoch RR and Sues H-D (2013). A new archosauriform reptile from the Middle Triassic (Ladinian) of Germany. Journal of Systematic Palaeontology (advance online publication) DOI:10.1080/14772019.2013.781066 online

wiki/Jaxtasuchus

New Langobardisaurus Confirms Earlier Findings

There’s a new Langobardisaurus (Figs. 1-3) with hollow limb bones courtesy of Saller, Renesto & Dalla Vecchia (2013). Langobardisaurus, with all of its oddities and wonders deservers a bit more PR. So, here ’tis.

Figure 1. The new Langobardisaurus. P10121. A little hard to see, but the neck curves up, left, down and behind the body, with the head emerging on the right. The hollow bones are crushed revealing their interiors. No soft tissue is preserved along with the fossil leaves shown here.

Here’s the abstract:
“A new specimen of the small protorosaurian reptile Langobardisaurus pandolfii is described. It was collected from the Seefeld Formation, of Late Triassic (Norian) age, in the Innsbruck area (Austria) and represents the first occurrence of Langobardisaurus outside Italy. Although preserved mostly as an impression, the find is significant

because it extends the palaeogeographic range of the genus and it is the second specimen known to date with the skull fully exposed. The preserved portions of the limb elements show that the bones are hollow, with a layer of compacta and without any trace of spongiosa. Reappraisal of all the specimens assigned to the genus Langobardisaurus reveals no significant differences between L. pandolfii and L. tonelloi, allowing to consider the latter as a junior synonym of the former.”

Not a protorosaur. Not an archosauromorph.
Saller, Renesto and Dalla Vecchia (2013) labeled Langobardisaurus a “small archosauromorph” basing this on conventional thinking linking Langobardisaurus to protorosaurs. We talked about this mistake earlier. The large reptile tree nests Langobardisaurus and its sisters with tritosaur lizards, descended from a sister to Huehuecuetzpalli and Lacertulus. Sister taxa in the large reptile tree include Cosesaurus, Tanystropheus, Tanytrachelos and Macrocnemus. This clade includes several other long-necked bipeds with sprawling hind limbs (Renesto, Dalla Vecchia and Peters 2002). So Langobardisaurus was an occasional biped, lizard-style.

Saller, Renesto and Dalla Vecchia (2013) report, “All specimens of Langobardisaurus were found in a dark limestone and dolostone that formed in relatively small and deep marine basins surrounded by shallow-water carbonate platforms.” Langobardisaurs appear to be terrestrial reptiles, so their bodies appear to have been swept into these basins from river floods along with the plant debris seen in the fossil.

Figure 2. The skull of the new Langobardisaurus in situ, above, and reconstructed below, using the DGS technique. If there was no antorbital fenestra the rostrum was at least very weak and this taxon immediately preceded a taxon known to have an antorbital fenestra, Cosesaurus. The left maxilla itself was broken into several pieces. The skull looks like the other Langobardisaurus skulls, so is likely conspecific. The dentary is tipped with a tooth-like structure. Note the very tall coronoid process.

Antorbital fenestra
Langobardisaurus (Fig. 2) appears to have had an antorbital fenestra as it now appears in two specimens (Fig. 4) and in Pteromimus (Atanassov 2001), another langobardisaur with an antorbital fenestra.

Skull bones
The premaxilla is reported as edentulous with toothlike-projections erupting from it. Certainly this morphology was distinct and provided a mechanism for prey (insect) acquisition. Perhaps these were teeth fused to the jaws in the manner of sphenodontid teeth.

Most maxillary teeth had two or three cusps, but the posterior-most maxillary and dentary teeth were much longer than the others and bore many tiny cusps. These would have acted like linear molars.

The coronoid process was tall and robust, unlike other tritosaur lizards. No stomach contents tell us what Langobardisaurus ate. But the teeth and coronoid process tell us it was probably crunchy, requiring a certain amount of oral processing.

The pectoral girdle
Earlier we talked about the coracoid and strap-like scapula of Langobardisaurus, relabeled from earlier interpretations. Here (Fig. 3) those identifications are confirmed with similar morphologies and placements.

Figure 3. The pectoral girdle of the new Langobardisaurus highlighted in colors. These elements correspond to those of an earlier Langobardisaurus with an angled coronoid and a strap-like scapula.

Hollow limbs
The hollow limb bones of Langobardisaurus are shared with members of the Fenestrasauria, including pterosaurs. So are the elongated nares, the large orbits, the elongated pedal 5.1 and the advancement of the sternum toward the clavicles.

Figure 4. Langobardisaurus tonelloi. The incomplete tail of this specimen was proabably longer based on other specimens. The the cosesaurid-type pectoral girdle. 

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.

Reference
Atanassov M 2001. Two new archosauromorphs from the Late Triassic of Texas. – Journal of Vertebrate Paleontology Abstracts 21(3): 30A.
Atanassov M 2002. Two new archosauromorphs from the Late Triassic of Texas. Dissertation.online abstract
Muscio G 1997. Preliminary note on a specimen of Prolacertiformes (Reptilia) from the Norian (Late Triassic) of Preone (Udine, north-eastern Italy). Gortania – Atti del Museo Friulano di Storia Naturale 18:33-40
Renesto S 1994. A new prolacertiform reptile from the Late Triassic of Northern Italy. Rivista di Paleontologia e Stratigrafia 100(2): 285-306.
Renesto S and Dalla Vecchia FM 2000. The unusual dentition and feeding habits of the Prolacertiform reptile Langobardisaurus (Late Triassic, Northern Italy). Journal of Vertebrate Paleontology 20: 3. 622-627.
Renesto S and Dalla Vecchia FM 2007. A revision of Langobardisaurus rossii Bizzarini and Muscio, 1995 from the Late Triassic of Friuli (Italy)*. Rivista di Paleontologia e Stratigrafia 113(2): 191-201. online pdf
Renesto S, Dalla Vecchia FM and Peters D 2002. Morphological evidence for bipedalism in the Late Triassic Prolacertiform reptile Langobardisaurus. Senckembergiana Lethaea 82(1): 95-106.
Saller F, Renesto S and Dalla Vecchia FM 2013. First record of Langobardisaurus (Diapsida, Protorosauria) from the Norian (Late Triassic) of Austria, and a revision of the genus. Neues Jahrbuch für Geologie und Paläontologie – Abhandlungen 268(1): 83-95
DOI: http://dx.doi.org/10.1127/0077-7749/2013/0319
Wild R 1980. Tanystropheus (Reptilia: Squamata) and its importance for stratigraphy. Mémoires de la Société Géologique de France, N.S. 139:201–206.

uninisubria/Langobardisaurus
wiki/Langobardisaurus

Rethinking the Diapsida (Modesto and Sues 2004)

Wikipedia documents a family tree of “the Diapsida” recovered by Modesto and Sues (2004). This is fairly conventional, but based on too few taxa, as you’ll see.

Figure 1. Tree from Wikipedia documenting the Diapsida according to Modesto and Sues 2004. Colored boxes divide taxa into the new Lepidosauromorpha and the new Archosauromorpha recovered by the large reptile tree that includes a magnitude or more taxa.

What’s interesting about this tree is if you take the pinks only you have a very good tree of the new Archosauriformes within the new Archosauromorpha that more or less matches the large reptile tree. And the same goes for the light green areas representing the new Lepidosauromorpha. Unfortunately, in the very center of things here you find the suprageneric taxon, “Squamata” and no sphenodontids higher than Gephyrosaurus. If the basal squamate Huehuecuetzpalli was added that might have linked that clade to Drepanosauridae + Tanystropheidae, as in the large reptile tree. If higher sphenodontids were added, that would link them to Trilophosaurus and rhychosaurs, as in the large reptile tree, as we saw yesterday with Lazarussuchus.

There is convergence here. That’s why there’s confusion.
Certainly protorosaurs did converge with tanystropheids, but by adding many more reptile taxa this convergence is revealed. The same goes for nesting lepidosaurs with choristoderes.

More taxa equals greater resolution and fewer strange bedfellows. Try it and see. It’s a simple experiment that any amateur or professional can do. If I can do it, you can do it.

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.

References
Modesto SP and Sues HD 2004. The skull of the Early Triassic archosauromorph reptile Prolacerta broomi and its phylogenetic significance. Zoological Journal of the Linnean Society 140 (3): 335.

Pamelaria, a large, long-necked protorosaur

Pamelaria dolichotrachela (Sen 2003) Middle Triassic, is the largest known prolacertiform (= protorosaur) and it has the longest neck of them all. (Please, remember Tanystropheus is a tritosaur lizard, not a prolacertiform/protorosaur).

Figure 1. Pamelaria, an long-necked protorosaur related to Protorosaurus.

Distinct from Protorosaurus, the skull of Pamelaria was relatively smaller with a shorter rostrum and smaller teeth. The nares were reported as confluent, and indeed they may be so, but that area of the skull was poorly preserved. The premaxilla was more robust. The postorbital was waisted at the postfrontal process. The quadrate was nearly vertical. The lacrimal was larger. The orbit was taller than long. The palate included smaller openings for the choanae due to a wider set of vomers. The parasphenoid was larger. The mandible elements were all shorter, including the teeth. The ventral rim of the mandible was straighter.

The cervicals were each longer. The tail was relatively shorter. The dorsal ribs were longer and more robust, enclosing a larger gut.

The scapula was taller and not fused to the coracoid. The forelimbs were more robust. Digits 3 and 4 were nearly equal in length. Ungual 1 was deeper proximally.

The pelvis was relatively shorter with an excavated ventral rim. The fibula was bowed away from the tibia. The foot was more robust with shorter digits. Metatarsals 2 and 3 were aligned with the base of ungual 1.

Sen (2003) considered Pamelaria a carnivore. With a bulkier body, smaller head and longer neck, Pamelaria must have looked like a sauropod, except with sprawling, lizard-like limbs (reminiscent of the earliest reconstructions of sauropods!). The small teeth and large gut suggest an herbivorous diet.

Pamelaria was a derived taxon leaving no known descendants. I see it as convergent to dinocephalians in respect to the torso, tails and limbs.

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.

References
Sen K 2003. Pamelaria dolichotrachela, a new prolacertid reptile from the Middle Triassic of India. Journal of Asian Earth Sciences 21: 663–681.