New pterosaur precursor study excludes all pterosaur precursors

Summary, for those in a hurry:
Ezcurra et al. 2020 create a chimaera from lagerpetid and protorosaur bits and pieces, then call it a pterosaur precursor close to dinosaurs. Unfortunately, this paper is another taxon omission disaster from Nature.

It is good to see more material data
appearing for lagerpetids (Figs. 4, 13), an enigmatic clade formerly known from pelvic and hind limb material, and more recently from skull bits.

Unfortunately
Ezcurra et al. follow an established history of workers omitting competing taxa in pterosaur origin papers while cherry-picking comparative taxa and employing a chimaera of disassociated and unrelated bits and pieces from a protorosaur and a lagerpetid. By contrast, the omitted pterosaur precursors are complete, articulated, preserve soft tissue and nest closer to pterosaurs in several prior cladograms.

Figure 1. From Ezcurra et al. 2020 showing the various parts used to produce the chimaera in the middle and call it a pterosaur precursor.

Figure 1. From Ezcurra et al. 2020 showing the various parts used to produce the chimaera in the middle and call it a pterosaur precursor. See Figure 2. for a valid pterosaur precursor.

Ezcurra et al. posit a pterosaur relationship
after omitting four competing candidate taxa put forth twenty years ago in Peters 2000.

In 2000 the four competing taxa
(Langobardisaurus, Cosesaurus (Fig. 2), Sharovipteryx, Longisquama) where added to four prior phylogenetic analyses and all nested closer to pterosaurs than any prior candidates. The most basal pterosaur, Bergamodactylus (MPUM 6009, Fig. 6), is also omitted from the Ezcurra et al. study.

Figure 1. CLICK TO ENLARGE. Cosesaurus reconstructed with enlarged parts of interest including a pes (foot) matching a Rotodactylus track. Here the pelvis is reconstructed according to figure 3. Shown here about life-size.

Figure 2.  CLICK TO ENLARGE. Cosesaurus reconstructed with enlarged parts of interest including a pes (foot) matching a Rotodactylus track. Here the pelvis is reconstructed according to figure 3. Shown here about life-size.

Ezcurra et al. take a moment to cite Peters 2000,
but in a misleading fashion: “However, some studies have alternatively recovered pterosaurs as the sister group … among tanystropheid archosauromorphs.”

Not archosauromorphs, “fenestrasaurs.” Peters 2007 moved them all over to Lepidosauria.

All four fenestrasaurs have an elongated pedal digit
with two phalanges longer than the metatarsal, matching basal pterosaurs. A long pedal digit 5 is not found in the new lagerpetid/protorosaur chimaera (Fig. 13).

pterosaur wings

Figure 3. Click to enlarge. The origin of the pterosaur wing and whatever became of manual digit 5?

Manus
Cosesaurus
, Sharovipteryx and Longsquama have a robust, elongated manual digit 4 on a robust metacarpal 4 rotated axially. The finger frames the wing membrane in pterosaurs. By contrast, metacarpal 4 is short and not robust in the new lagerpetid/protorosaur (Fig. 4).

Figure 3. The lagerpetid manus compared to the basal pterosaur manus. In the lagerpetid metacarpal 4 is not larger or more robust than the others. Compare to figure 3.

Figure 4. The lagerpetid manus compared to the basal pterosaur manus. In the lagerpetid metacarpal 4 is not larger or more robust than the others. Cosesaurus likewise has metacarpals of similar diameter, but metacarpal 1 is not shorter than the others. Sharovipteryx and Longsquama have a transitional manus, including the evolutionary shortening of metacarpal 1, as in Bergamodactylus. So this trait is convergent in a valid phylogenetic context. Compare to figure 3.

Coracoid
Cosesaurus
, Sharovipteryx and Longisquama have a quadrant-shaped, locked down coracoid (Figs. 2, 5). Such a shape is also found in birds and pterosaurs. This enables flapping. The authors reconstruct the new lagerpetid/protorosaur chimaera with a mobile disc-like coracoid.

Colorized sternal complex elements in Cosesaurus.

Figure 5. Click for rollover image. Colorized sternal complex elements in Cosesaurus. Coracoids in blue. Scapulae in green. Clavicles in pink. Interclavicle in red. Sternum in yellow. Reconstructed in figure 2.

Sternal complex
In Cosesaurus and Longsiquama the ventral stem of the immobile coracoid articulates with a broad sternal complex (Fig. 5) created by the migration and overlapping of robust clavicles, a single lepidosaurian sternum, and a cruciform interclavicle that are fused together in Longisquama and pterosaurs. The new lagerpetid/protorosaur chimaera has no such sternal complex.

Prepubis
Cosesaurus has a prepubis (Fig. 2), a pelvic bone found otherwise only in pterosaurs (Fig. 6). A prepubis is not found in the new lagerpetid/protorosaur chimaera.

Elongate ilium and expanded sacrum
Cosesaurus, Sharovipteryx and Longisquama have an elongate ilium, matching pterosaurs, with a much longer preacetabular process than the new chimaera. The Cosesaurus sacrum, includes four vertebrae articulating with the medial ilium. Sharovipteryx, Longisquama and the basal pterosaur, Bergamodactylus, have more sacrals between longer ilia. The new lagerpetid/protorosaur chimaera has two sacrals. More than two sacrals is common in bipedal taxa.

Figure 1. Bergamodactylus compared to Cosesaurus. Hypothetical hatchling also shown.

Figure 6. Bergamodactylus compared to Cosesaurus. Hypothetical hatchling also shown.

Attenuated tail
The tail of Cosesaurus, Sharovipteryx and pterosaurs is attenuated with tiny chevrons because the tail no longer anchors large caudofemoralis muscles. Such an attenuated tail is not present on the new lagerpetid/protorosaur chimaera, which as deep chevrons.

Pteroid
Peters 2009 documented the migration of two centrale on the wrist of Cosesaurus to the medial rim where they match the pteroid and prearticular carpal in pterosaurs.

Antorbital fenestra
The new lagerpetid/protorosaur chimaera has an antorbital fenestra without a fossa extending to the naris. All archosauriformes, some protorosaurs, all pterosaurs, and the competing candidates also have this trait. The antorbital fenestra is not close to the naris in Tropidosuchus, and is unknown in the holotype Lagerpeton.

Figure x. New mandible compared to the Triassic pterosaur Seazzadactylus where the tip is actually a tooth as in Langobardisaurus.

Figure 7. New mandible compared to the Triassic pterosaur Seazzadactylus where the tip is actually a tooth as in Langobardisaurus.

Figure x. Pterosaur precursor, Langobardisaurus, has anteriorly-oriented dentary tip teeth.

Figure 8. Pterosaur precursor, Langobardisaurus, has anteriorly-oriented dentary tip teeth as  in basal pterosaurs.

Mandible
The new lagerpetid/protorosaur chimaera mandible has an ‘edentulous and tapering anterior end’ (Fig. 7. The basalmost pterosaur, Bergamodactylus, has teeth at the tip of its mandible. So does the pterosaur employed by Ezcurra et al. (Seazzadactylus, Fig. 8). So does Langobardisaurus (Fig. 8), nesting outside the Pterosauria.

Figure from Ezcurra et al. 2020 comparing skull top of Ixtalerpeton to Prolacerta.

Figure 9. From Ezcurra et al. 2020 comparing skull top of Ixtalerpeton to Prolacerta, not close to lagerpetids. Here the pink Kongonaphon rostrum fragment is matched to the Ixalerpeton cranium.

Femur
Lagerpetid and protorosaur femora have a large trochanter for the insertion of large caudofemoralis muscles. Pterosaurs and the four fenestrasaurs lack a large trochanter because the caudofemoralis has become a vestige and is lost in pterosaurs. The great majority of the femoral muscles in pterosaurs and fenestrasaurs anchor on the pelvis and prepubis instead of the tail.

Figure x. Ixalerpeton pelvis compared to Prolacerta.

Figure 10. Ixalerpeton pelvis compared to Prolacerta.

Deltopectoral crest on humerus
The new lagerpetid/protorosaur chimaera has a long, low deltopectoral crest. Sharovipteryx and Longisquama each have a large, robust deltopectoral crest, as in pterosaurs.

Figure 3. The closest kin of Tropidosuchus are the much larger Chanaresuchus (matching Nesbitt 2011) and the smaller Lagerpeton.

Figure 11 The closest kin of Tropidosuchus are the much larger Chanaresuchus (matching Nesbitt 2011) and the smaller Lagerpeton.

In their study on pterosaur origins,
Ezcurra et al. also omitted a pair of citations of a single supertree study by Hone and Benton (2007, 2008). In the first paper, Hone and Benton announced they would test the taxa in Peters 2000 against those in Bennett 1996. In the second paper Hone and Benton omitted all taxa from Peters 2000, omitted any citation of Peters 2000, then credited both competing hypotheses to Bennett 1996. This is a glaring example of the established history of omitting pertinent and competing taxa in pterosaur origin papers. Dozens of other papers simply omit competing candidates from Peters 2000. Most of them do so without making the mistake of promising in print to include them as part of a study.

Figure from Ezcurra et al. 2020 comparing skull top of Ixtalerpeton to Prolacerta.

Figure 12. The lagerpetid maxilla associated with the lagerpetid foot in figure 13. Kongonaphon restoration based on Tropidosuchus.

Figure z. Tracing of lagerpetid pes with colors showing all four toes. Digits 1 and 2 are fused to metatarsal 3. Only digits 3 and 4 bore weight. Digit 5 was a vestige unlike basal pterosaurs.

Figure 13. Tracing of lagerpetid pes with colors showing all four toes. Digits 1 and 2 are fused to metatarsal 3. Only digits 3 and 4 bore weight. Digit 5 was a vestige unlike basal pterosaurs.

To combat that long history of omitting taxa,
www.ReptileEvolution.com tests 1775 taxa in a single phylogenetic analysis that includes lagerpetids, protorosaurs, dinosaurs and pterosaurs. In that wide gamut online study, lagerpetids nest with Tropidosuchus and the Chanaresuchidae. Pterosaurs nest with the four fenestrasaurs within a third clade of lepidosaurs, the Tritosauria, far from dinosaurs, protorosaurs and lagerpetids. Chimaera taxa (created from bits of this and parts of that) are not tested..for good reason. There is no cherry-picking here. Taxa nest wherever the software indicates they should.


References
Ezcurra MD et al. (17 co-authors) 2020. Enigmatic dinosaur precursors bridge the gap to the origin of Pterosauria. Nature (2020). https://doi.org/10.1038/s41586-020-3011-4
Peters D 2000. A Redescription of Four Prolacertiform Genera and Implications for Pterosaur Phylogenesis. Rivista Italiana di Paleontologia e Stratigrafia 106 (3): 293–336.
Peters D 2007. The origin and radiation of the Pterosauria. In D. Hone ed. Flugsaurier. The Wellnhofer pterosaur meeting, 2007, Munich, Germany. p. 27.
Peters D 2009. A reinterpretation of pteroid articulation in pterosaurs. Journal of Vertebrate Paleontology 29: 1327-1330

Cosesaurus paper on ResearchGate.net

http://reptileevolution.com/cosesaurus.htm
http://reptileevolution.com/pterosaur-wings.htm
http://reptileevolution.com/langobardisaurus.htm
http://reptileevolution.com/cosesaurus.htm
http://reptileevolution.com/sharovipteryx.htm
http://reptileevolution.com/longisquama.htm
http://reptileevolution.com/MPUM6009.htm
Review in Nature by Kevin Padian https://doi.org/10.1038/d41586-020-03420-z

Here’s the quote from Gizmodo.com that says it all:
“These creatures seem an unlikely sister group from which pterosaurs emerged, which is probably why they’ve been ignored for so long.”

Tiny Vellbergia: a juvenile Prolacerta, not a stem-lepidosauromorph

Sobral, Simoes and Schoch 2020
report on a new, tiny, Middle Triassic reptile, Vellbergia bartholomei (Figs. 1, 3) known from disarticulated skull material. Without creating a reconstruction, they considered Vellbergia a stem-lepidosauromorph different from other stem-lepidosauromorphs.

Figure 1. Vellbergia in situ, original line drawing, DGS colors apple and reconstruction. Note the quadrate break occurs during taphonomic crushing of the curved bone. Scale bar = 5mm. So the skull is about 1.5cm in length, quite tiny.

Figure 1. Vellbergia in situ, original line drawing, DGS colors apple and reconstruction. Note the quadrate break occurs during taphonomic crushing of the curved bone. The nasal, frontal and parietals are incomplete due to their juvenile state. Scale bar = 5mm. So the skull is about 1.5cm in length, quite tiny.

The authors report:
“Important morphological attributes of Vellbergia, most notably the elongate and slender jaw bones, the deeper post-dentary region of the jaw relative to the anterior region, and the far posteriorly reaching maxillary tooth row can be found on some other early diverging diapsid species, such as Prolacerta and Youngina, thus showing these features were retained into the early part of the lepidosauromorph evolutionary history as well.”

Prolacerta.

Figure 2. Prolacerta. Note the relative lengths of the manual and pedal lateral digits.

After phylogenetic analysis
in the large reptile tree (LRT, 1653+ taxa) Vellbergia nests with Prolacerta (Figs. 2, 3). A reconstruction (Figs. 1, 3) demonstrates a close affinity. Based on size and the smooth, incomplete, open sutures of the specimen, this is a juvenile. So the genus ‘Vellbergia’ is a junior synonym. The authors did not include Prolacerta in their published cladogram, but did list it in their suppdata.

Figure 3. Prolacerta and 'Vellbergia' to scale.

Figure 3. Prolacerta and ‘Vellbergia’ to scale.

Taxon exclusion
The Sobral, Simoes and Schoch taxon list did not include enough taxa to produce the basal dichotomy splitting Archosauromorpha from Lepidosauromorpha in the Viséan following their last common ancestor, Silvanerpeton. Prolacertiformes (= Protorosauria) nest in the Archosauromorpha and converge with Lepidosauriformes in many ways, hence the traditional confusion.

The LRT is available online.
Problems like this can be avoided by using ReptileEvolution.com and the LRT to double-check work before submission.


References
Sobral G, Simoes TR and Schoch RR 2020.
A tiny new Middle Triassic stem-lepidosauromorph from Germany: implications fro the early evolution of lepidosauromorphs and the Vellberg fauna. Nature.com Scientific Reports 10, Article number: 2273.

https://doi.org/10.1038/s41598-020-58883-x
https://www.nature.com/articles/s41598-020-58883-x

 

 

Litorosuchus phylogeny revised

Figure 1. Litorosuchus somnii was wrongly considered a sister to Vancleavea and wrongly considered an archosauriform. In the LRT it nests with Macrocnemus, a tritosaur lepidosaur that also has members with an antorbital fenestra. Click to see an enlarged rollover image.

Figure 1. Litorosuchus somnii was wrongly considered a sister to Vancleavea, which was wrongly considered an archosauriform. In the LRT Litorosuchus nests with Jaxtasuchus, a protorosaur, which was also wrongly considered an archosauriform, that also has an antorbital fenestra. Click to see an enlarged rollover image. (See figure 2).

Earlier we looked at
Litorosuchus (Fig. 1,2; Li et al. 2016; Middle Triassic; 2m in length), an armored aquatic protorosaur (previously mis-nested in the large reptile tree (LRT, 1327 taxa) as an armored, aquatic macrocnemid). The recent work on the various specimens attributed to Macrocnemus raised red flags on Litorosuchus, which did not quite fit the elusive, but required ‘gradual accumulation of traits’ that tells you your phylogenetic analysis reflects actual events in evolution.

Protorosaurs and macrocnemids
have been traditionally confused with one another and wrongly linked together due to their many convergent traits. The former is a member of the new Archosauromorpha. The latter belongs to the new Lepidosauromorpha.

Figure 2. Litorosuchus in situ with a new tracing of the inverted and displaced posterior premaxilla with several teeth.

Figure 2. Litorosuchus in situ with a new tracing of the inverted and displaced posterior premaxilla with several teeth.

Some re-scoring led to a re-nesting
of Litorosuchus with an incompletely known aquatic armored protorosaur also with an antorbital fenestra, Jaxtasuchus (Fig. 3; Schoch and Sues 2012; Middle Triassic; originally wrongly considered a Doswellia relative due to taxon exclusion; estimated length: 70-80cm).

Figure 2. The armored aquatic protorosaur, Jaxtasuchus, is smaller and less complete than Litorosuchus.

Figure 3. The armored aquatic protorosaur, Jaxtasuchus, is smaller and less complete than Litorosuchus.

Only a few scored traits differ
in Litorosuchus and Jaxtasuchus. Only a few related protorosaurs share an antorbital fenestra, developed by convergence with various chroniosuchid macrocnemid, fenestrasaurYoungina, Youngoides and archosauriform taxa.

Apologies
for the earlier mistake. Reexamination (= testing) is just one of the many processes of science. That often happens here, but a bit rare out there in the world of paleontology.

References
Li C, Wu X-C, Zhao L-J, Nesbitt SJ, Stocker MR, Wang L-T 2016. A new armored archosauriform (Diapsida: Archosauromorpha) from the marine Middle Triassic of China, with implications for the diverse life styles of archosauriforms prior to the diversification of Archosauria. The Science of Nature 103: 95. doi:10.1007/s00114-016-1418-4
Schoch and Sues 2012. A new archosauriform reptile from the Middle Triassic (Ladinian) of Germany. Journal of Systematic Palaeontology, 2013. http://dx.doi.org/10.1080/14772019.2013.781066

wiki/Prolacertiformes
wiki/Protorosauria
wiki/Jaxtasuchus
wiki/Litorosuchus

Müller et al. discuss Lagerpeton ‘sisters’ without Tropidosuchus

Müller, Langer and Dias da Silva (2018) report:
“Despite representing a key-taxon in dinosauromorph phylogeny, Lagerpertidae is one of the most obscure and enigmatic branches from the stem that leads to the dinosaurs.”
It’s taxon exclusion, yet again.
Lagerpeton (Fig. 1) is obscure and enigmatic because it is NOT in the stem that leads to dinosaurs. We discussed that earlier here in 2011. Langer is aware of the better, more inclusive, option because he sent me the reference for Novas and Agnolin 2016 discussed here.
“Recent new findings have greatly increased our knowledge about lagerpetids, but no phylogenetic analysis has so far included all known members of this group. Here, we present the most inclusive phylogenetic study so far conducted for Lagerpetidae. …Finally, quantification of the codified characters of our analysis reveals that Lagerpetidae is one of the poorest known among the Triassic dinosauromorph groups in terms of their anatomy, so that new discoveries of more complete specimens are awaited to establish a more robust phylogeny.”
Tropidosuchus is known from complete skeletons (Fig. 1) and is the sister to Lagerpeton in the more inclusive large reptile tree (LRT, 1173 taxa).
Problem solved!

Figure 3. The closest kin of Tropidosuchus are the much larger Chanaresuchus (matching Nesbitt 2011) and the smaller Lagerpeton.

Figure 1. The closest kin of Tropidosuchus are the much larger Chanaresuchus (matching Nesbitt 2011) and the smaller Lagerpeton.

Ixalerpeton
Müller et al. nest Ixalerpeton (Fig. 24) with Lagerpeton without testing other candidates. In the LRT Ixalerpeton nests among basal protorosaurs, far from Lagerpeton. And yes, it might have been the only bipedal protorosaur, exciting news that was completely overlooked due to taxon exclusion. Or not. The foot would be helpful, but is not known.

Figure 2. Ixalerpeton bits and pieces reconstructed. This taxon nests with protorosaurs.

Figure 2. Ixalerpeton bits and pieces reconstructed. This taxon nests with protorosaurs.

Which protorosaur was closest to Ixalerpeton?
Ixalerpeton nests between the AMNH 9502 specimen of Prolacerta and Czatkowiella, (Fig. 3), taxa omitted by the Müller et al. 2018 study.

Figure 1. Czatkowiella harae bits and pieces here reconstructed as best as possible. Note the size difference here between the large maxilla and the small one.

Figure 3. Czatkowiella harae bits and pieces here reconstructed as best as possible. Note the size difference here between the large maxilla and the small one. This taxon is close to Ixalerpeton. The skull roof and occiput are comparably similar.  Perhaps Ixalerpeton had a longer neck based on this sister.

References
Müller RT,  Langer  MC & Dias-Da-Silva  S 2018. Ingroup relationships of Lagerpetidae (Avemetatarsalia: Dinosauromorpha): a further phylogenetic investigation on the understanding of dinosaur relatives. Zootaxa 4392(1): 149â158
Novas FE and Agnolin FL 2016 Lagerpeton chanarensis Romer (Archosauriformes): A derived proterochampsian from the middle Triassic of NW Argentina. Simposio. From Eventos del Mesozoico temprano en la evolución de los dinosaurios”. Programa VCLAPV. Conferencia plenaria: Hidrodinámica y modo de vida de los primeros vertebrados. Héctor Botella (Universitat de València, España) 2016

SVP 2017 abstracts: Does Malerisaurus nest with Azendohsaurus?

The short answer is
no.

You might recall
we looked at Malerisaurus earlier.

A different take comes from the Nesbitt et al. 2017 SVP abstract:
“Mediation of some of these challenges is now possible with the recently recognized early
archosauromorph clade Allokotosauria. This clade contains disparate, ecologically
diverse (faunivores and herbivores), and typically larger bodied (1-3 meters in length)
archosauromorphs (Azendohsaurus, Trilophosaurus), but to this point, plesiomorphic,
early-diverging allokotosaurians have not been identified.

“Here, we recognize specimens assigned to the enigmatic taxon Malerisaurus from both present-day India and western Texas as members of Allokotosauria, and more specifically, the Azendohsauridae. The recognition of Malerisaurus as both an allokotosaur and an azendohsaurid has also helped identify other fragmentary remains of close relatives from Triassic deposits across Pangea including India, elsewhere in North America, and Africa. As such, Allokotosauria had a near Pangean distribution for much of the Middle to Late Triassic. Allokotosauria represents one of the oldest successful clades of archosauromorphs that achieved a wide geographic distribution and both taxonomic and ecomorphological diversity.”

The Protorosaurs, Malerisaurus, Prolacerta, Protorosaurus, Pamerlaria and Boreopricea.

Figure 1. Click to enlarge. The Protorosaurs, Malerisaurus, Prolacerta, Protorosaurus, Pamerlaria and Boreopricea. It is easy to see why these taxa become confused with Trilophosaurus and Azendohsaurus. More taxa solves the problem. 

In the
large reptile tree (LRT 1050 taxa) Malerisaurus nests with other protorosaurs within the new Archosauromorpha, sharing many traits by convergence with the Allokotosauria. This clade of currently three taxa (Trilophosaurus, Azendohsaurus and horned Shringasaurus) nests within the Rhynchocephalia between the derived taxa, Noteosuchus and Mesosuchus all within the new Lepidosauromorpha.

I’m guessing,
based on past performance (I was not in Calgary), that Nesbitt et al. did not add any or many basal rhynchocephalians to their cladogram, so this new odd nestings appears to join other odd nestings, likely victims of taxon exclusion.

References
Nesbitt SJ et al. (nine co-authors) 2017. The ‘strange reptiles’ of the Triassic. The morphology, ecology, and taxonomic diversity of the clade Allokotosauria illuminated by the discovery of an early diverging member. SVP abstracts 2017.

Ozimek in water: Two new hypotheses

The hyper-slender limbs
of the protorosaur, Ozimek (Fig 1), are unique within the clade Tetrapoda. They are so slender that one wonders how Ozimek was able to move about, with or without the additional burden of that large skull on the end of that long slender neck. We looked at this taxon earlier here and here.

FIgure 1. Ozimek skeleton in vivo. Water and grainy lake bedding are indicated. Neutral buoyancy is one answer to the riddle of those hyper-slender limbs.

FIgure 1. Ozimek skeleton in vivo. Water and grainy lake bedding are indicated. Neutral buoyancy is one answer to the riddle of those hyper-slender limbs.

Originally
(Dzik and Sulej 2016) Ozimek was described as a glider related to Sharoviptyerx, but it is much larger, not related, no lateral membranes are present either in Sharovipteryx or Ozimek, the ribs of Ozimek enclose a cylindrical torso, unlike the flat torso of Sharovipteryx,  and most importantly the limbs in Ozimek are much more slender, relative to those in Sharovipteryx despite being many times larger.

Phylogenetically
the large reptile tree (LRT) nests Ozimek with Prolacerta. There are no gliders in that clade. Jaxtasuchus is another long-necked protorosaur, but it was armored with bony scutes, and was not directly related to Ozimek, yet similar in size. While the metapodials are not compact in any protorosaurs, neither do they spread widely, so the manus and pes of Ozimek were likely narrower (Fig. 1) than originally restored by Dzik and Sulej.

Let’s remember
that gliders may be slender, but their ‘wing struts’ must berobust enough to support their entire weight on extended unsupported limbs, multiple ribs, as in Draco, or extended dermal rods, as in Coelurosauravus. Gliders may also have strong pectoral and pelvic girdles to anchor those gliding limbs. Ozimek will never be described as anything but weak and slender. If it was the size of a fly it might have glided, but at the size of a praying mantis or Sharoviipteryx, that possibility is less likely, let alone at its current size, 3x longer and 9x heavier than Sharovipteryx.

Tanystropheus underwater among tall crinoids and small squids.

Figure 2. Tanystropheus in a vertical strike elevating the neck and raising its blood pressure in order to keep circulation around its brain and another system to keep blood from pooling in its hind limb and tail.

Converging on the long-necked tritosaurs,
Tanystropheus (Fig. 2) and Langobardisaurus (Fig. 3), Ozimek had similar overall proportions,  but with more slender limbs. Tanystropheus is best considered an underwater biped (Fig. 2) based on many geologic clues. Langobardisaurus may have been amphibious, but facultatively bipedal either wet or dry.

Figure 2. Langobardisaurus compared to Ozimek and its sister, Prolacerta.

Figure 3. Langobardisaurus compared to Ozimek and its sister, Prolacerta.

Geological setting
Dzik and Sulej report, “the [Ozimek] fossils under study occur in the one-meter thick lacustrine horizon where the dominantspecies are aquatic or semi-aquatic animals. Shallow freshwater conditions existed at deposition. This taken together with conchostracans occurring within the fauna suggest an abundance of periodic ponding at a lake shore.” Ponds are generally placid bodies of water. Dzik and Sulej report, periodic flooding events transported terrestrial taxa to the ponds during redeposition events. The strata also contain large parasuchians (phytosaurs), large metoposaurs, lungfish and other fish. “Most of the articulated Ozimek gen. nov. specimens were found at the boundary between the red upper and grey lower units of the lacustrine horizon, but not in the grainstone lenses.”

Neutral buoyancy
Ozimek does not have the robust solid ribs that would indicate it was a bottom dweller. The slender limbs were hollow, but not pneumatic, so they were likely filled with bone marrow, according to the authors. In any case, their very slenderness minimizes the amount of air or fat they can contain.

After wrestling with various niche scenarios for Ozimek
the morphological and analog evidence indicates that IF Ozimek was a slow-moving reptile supported by a placid aquatic medium, slender limbs could evolve. The long neck raised and lowered the skull, both for breathing and prey capture. Lateral motions were limited. The long cervical ribs would have kept the neck fairly straight, like a flexible fishing rod. Ozimek was likely a sit and wait predator in shallow ponds. It is difficult to envision it as a giant glider or a terrestrial predator. The limbs were too slender.

Figure 4. Ozimek hitching a ride on top of Metoposaurus.

Figure 4. Ozimek hitching a ride on top of Metoposaurus as a possible parasite, cleaner, or egg predator.

A second hypothesis:
Sometimes when taxa have unusual traits, those develop due to a relationship with other individuals, or even other genera. If Ozimek entertained a remora-like lifestyle, hitched to the top of a giant, slow-moving Metoposaurus (Fig. 4), it would have been protected by the  bulk of its patron and able to feed on leftovers from the giant predator’s meals. Or maybe it fed on parasites attached to Metoposaurus. Or on worms stirred up when Metoposaurus was settling in. Or on Metoposaurus eggs when they were laid. In that role, the limbs of Ozimek would not have needed to remain ambulatory because Metoposaurus would have done the walking. The long fingers and sharp claws (unguals) of Ozimek might have helped anchor it to the back or the flat skull of the giant amphibian.

References
Dzik J and Sulej T 2016. An early Late Triassic long-necked reptile with a bony pectoral shield and gracile appendages. Acta Palaeontologica Polonica 61 (4): 805–823.

 

Czatkowiella: the terror of mixed up puzzle pieces

Fissure fossils
are wonderful in that they are typically uncrushed and readily extracted. They are also horrible in that many bones are fragments, few are attached to each other and worst of all…several sizes, species and genera can get mixed up together. That’s the case with Czatkowiella harae (Fig. 1).

Here’s how Borsuk−Biaynicka and Evans 2009 described the scenario:
“The description that follows is based on isolated and fragmentary bones extracted from a microvertebrate−bearing deposit containing the remains of at least four other small reptiles. Of these, Czatkowiella is overlapped by the archosauriform Osmolskina (Borsuk−Białynicka and Evans 2003) in the upper end of its size range, by Pamelina (Evans 2009) in the middle part of its range, and by Sophineta (Evans and Borsuk− Bialynicka 2009b) at the extreme lower end. This presents something of a challenge in terms of attributing elements, particularly with Osmolskina which is the more closely related taxon. For the smaller diapsids, we have used a combination of fit for individual elements and the general rule that if a bone of the same morphology occurs through a wide size range it is more likely to be Czatkowiella, since there is little variation in jaw size for the two small lepidosauromorphs. The structure of the dentition (using scanning electron microscopy) permits association of tooth−bearing elements, the maxilla then forming a template around which to fit other skull bones.”

Figure 1. Czatkowiella harae bits and pieces here reconstructed as best as possible. Note the size difference here between the large maxilla and the small one.

Figure 1. Czatkowiella harae bits and pieces here reconstructed as best as possible. Note the size difference here between the large maxilla and the small one.

Czatkowiella harae 
(Borsuk−Biaynicka & Evans 2009) Early Triassic ~250 mya, was recently described as a long-necked sister to Protorosaurus Here it nests between Ixalerpeton and Malerisaurus, two taxa that were not tested originally. Several sizes are known, all from disarticulated bone fragments. The number of cervicals is unknown but sister taxa have 8. Hopefully all of the bones assigned to this genus actually belong to this genus.

Figure 2. Borsuk−Biaynicka and Evans tested only Prolacerta and Protorosaurus among the Protorosauria.

Figure 2. Borsuk−Biaynicka and Evans tested only Prolacerta and Protorosaurus among the Protorosauria.

Distinct from Protorosaurus,
the skull of Czatkowiella was longer, lower, flatter and wider. The maxilla ascended in a process just aft of the long naris. The jugal had a stub quadratojugal process. The descending process of the squamosal was gracile and acute. The upper temporal fenestra was smaller. The lateral processes of the parietal were more posteriorly oriented. The cervicals and dorsals were shallower with lower neural spines.

If the two maxillae are indeed conspecific, then the shape of the maxilla changes with ontogeny.

The Borsuk−Biaynicka and Evans cladogram 
suffers from massive taxon exclusion as it nests

  1. Lepidosauromorpha  is wrongly nested as a sister clade to Protorosaurus + Czatkowiella and the rest of the taxa listed below.
  2. The gliderCoelurosaurus, is wrongly nested as a sister clade to Protorosaurus and the rest of the taxa listed below.
  3. Coelurosauravus, is wrongly nested as a sister clade to Ichthyosauria + Thalalattosauria and the rest of the taxa listed below.
  4. Ichthyosauria + Thalalattosauria is wrongly nested as a sister clade to Choristoderes + Turtles  but it is a sister to the clade Sauropterygia.
  5. Choristodera is wrongly nested as a sister clade to Tanystropheus, Macrocnemus, Prolacerta and the rest of the taxa listed below.
  6. Prolacerta is wrongly nested as a sister clade to Trilophosaurus + Rhynchosauria but it is correctly nested as a sister to the Archosauriformes.

Still not sure
how Prolacerta and Protorosaurus get separated in cladograms produced by PhDs.

Borsuk−Biaynicka & Evans 2009 were using outdated taxon sets,
but then, these preceded the publication of the large reptile tree by a few years.

Quick reminder,
Tritosaurs, including Tanystropheus and Macrocnemus are not related to protorosaurs in the large reptile tree despite the many convergent traits.

References
Borsuk−Biaynicka M and Evans S E 2009. A long−necked archosauromorph from the Early Triassic of Poland. Paleontologica Polonica 65: 203–234.

wiki/Czatkowiella

Ozimek volans: homology and analogy

Earlier we looked at the new protorosaur
Ozimek volans (Fig. 1) here and determined by phylogenetic analysis that it was a sister to Prolacerta, not Sharovipteryx.

Today, just a short note
about its homology with Prolacerta and its purported and invalid analogy with the unrelated membrane gliders Sharovipteryx and Cynocephalus.

Figure 1. Ozimek volans compared to its homolog sister, Prolacerta, and to two putative analogs, Sharovipteryx and Cynocephalus, all to scale. Note the lack of climbing claws and the weakness of the limbs and girdles in Ozimek.

Figure 1. Ozimek volans compared to its homolog sister, Prolacerta, and to two putative analogs, Sharovipteryx and Cynocephalus, all to scale. Note the lack of climbing claws and the weakness of the limbs and girdles in Ozimek, adorned here with hypothetical membranes.

Floating is just one niche possibility
based on the weakness of the muscle anchors in Ozimek. I have never seen such skinny arms and legs, so I am at a loss for a suitable niche for it.
I don’t see large climbing claws,
long manual digits, large muscles and their anchors on Ozimek that one finds on Cynocephalus. If it were it otherwise, I might support the gliding hypothesis.
Gliding animals need strong limbs
and muscle anchors not only for supporting their total weight in the air, but also for climbing trees and the momentum shock of both take-off and landing. In this regard, Ozimek appears to be quite a bit weaker than either Cynocephalus or Sharovipteryx. If it was like Sharovipteryx the diameter of the limb bones should have been scaled up to deal with the magnitude greater mass.
Sharovipteryx has elongate ilia and pectoral elements with short arms, plus seven sacrals, all lacking in Ozimek, its putative sister.
Sharovipteryx does not have a lateral membrane
Old and bad reconstructions of Sharovipteryx used to add a membrane between imagined long forelimbs with short fingers and the longer hind limbs. No one has ever seen such a membrane in the fossil. No sisters have such a membrane. Rather a uropatagium trails each hind limb, as in pterosaurs and Cosesaurus. Phylogenetic bracketing adds a pterosaur-like brachiopatagium behind each tiny Sharovipteryx forelimb, but it is likewise not visible in the fossil. The Dzik and Sulej team counts on the validity of the fantasy lateral membrane between the limbs to make their Ozimek a glider. But it was never there in any case.

Figure 1. Dzik and Sulej are so sure that their Ozimek was a spectacular big sister to Sharovipteryx that they gave a model gliding membranes and used the largest disassociated humerus for scale. More likely it was an aquatic animal that did not move around much underwater.

Figure 2. Dzik and Sulej are so sure that their Ozimek was a spectacular big sister to Sharovipteryx that they gave a model gliding membranes and used the largest disassociated humerus for scale. No membranes are present lateral to the pancaked ribs in Sharovipteryx and so this patagium on Ozimek, lacking such ribs, is also based on fantasy.

Prolacerta is also hollow-boned,
and is the sister of Ozimek in the LRT. No tested taxon, including Sharovipteryx, is phylogenetically closer.
Langobardisaurus analogy
Overall, Ozimek looks like a big, skinny Langobardisaurus (Fig. 3).

Figure 2. Langobardisaurus compared to Ozimek and its sister, Prolacerta.

Figure 3. Langobardisaurus compared to Ozimek and its sister, Prolacerta to scale. Structurally, Ozimek was similar to Langobardisaurus, but had much longer, weaker limbs and girdles and despite a long list of similarities, still nested with Prolacerta.

Langobardisaurus had the same long neck
and big skull as seen in Ozimek, but is not related, The girdles are larger and the limbs are more robust in the smaller Langobardisaurus than in the larger Ozimek. So, whatever Langobardisaurus was doing, Ozimek might have been doing, but more slowly, cautiously and secretly, perhaps like a spider.

Protorosaurs and Tritosaurs
appear on opposite sides of the LRT, but closely resemble one another such that macrocnemids and langobardisaurs were both considered protorosaurs (even by me) before the LRT showed macorcnemids and langobardisaurs actually nested with tritosaur lepidosaurs. The convergence is amazing and potentially confusing unless a rigorous analysis is performed. The LRT has been successful in separating such convergent taxa and continues to do so.

References
Dzik J and Sulej T 2016. An early Late Triassic long-necked reptile with a bony pectoral shield and gracile appendages. Acta Palaeontologica Polonica 61 (4): 805–823.

Ozimek volans: long and skinny, but not a glider

Updated a few hours later
with a phylogenetic analysis nesting Ozimek with Prolacerta.

A new and very slender
Late Triassic (230 mya) reptile from lake sediment, Ozimek volans (Dzik and Sulej 2016; ZPAL AbIII / 2512; Figs. 1-3) appears to look like a variety of taxa on both sides of the great divide within the Reptilia: macrocnemids and protorosaurs. Based on the long, thin-walled neck bones, Ozimek was originally considered a possible pterosaur or tanystropheid, but Dzik and Sulej nested it with Sharovipteryx (Fig. 1), the Middle Triassic gliding fenestrasaur, and considered it a big glider (Fig. 3).

Figure 1. Three in situ specimens attributed to Ozimek. The largest humerus (purple) is scaled up from the smaller specimen. These are 80% of full scale when viewed at  72 dpi. To me, that 2012 ulna looks like a tibia + fibula and the 2012 humerus looks like a femur, distinct from the 2512 humerus.

Figure 1. Three in situ specimens attributed to Ozimek. The largest humerus (purple) is scaled up from the smaller specimen. These are 80% of full scale when viewed at  72 dpi. To me, that 2012 ulna looks like a tibia + fibula and the 2012 humerus looks like a femur, distinct from the 2512 humerus.

The large reptile tree
(LRT) does not nest the much larger Ozimek with tiny Sharovipteryx, but with Prolacerta (Fig. 2). While lacking an antorbital fenestra, Dzik and Sulej consider Ozimek an archosauromorph. They also consider Sharovipteryx an archosauromorph.  Like all fenestrasaurs, Sharovipteryx has an antorbital fenestra by convergence with archosauromorpha.

Figure 2. Reconstruction of Ozimek with hands and feet flipped to a standard medial digit 1 configuration and compared to Sharovipteryx and Prolacerta to scale. Note the short robust forelimbs and elongate pectoral elements of Sharovipteryx, in contrast to those in Ozimek.

Figure 2. Reconstruction of Ozimek with hands and feet flipped to a standard medial digit 1 configuration and compared to Sharovipteryx and Prolacerta to scale. Note the short robust forelimbs and elongate pectoral elements of Sharovipteryx, in contrast to those in Ozimek. Compared to Prolacerta the girdles are much smaller, indicating a much smaller muscle mass on the limbs, probably making it a poor walker. Perhaps it floated to support its weight.

Sediment
The authors report on the limestone concretion, “the fossils under study occur in the one-meter thick lacustrine horizon in the upper part where the dominant species are aquatic or semi-aquatic animals. These also include the armored aetosaur Stagonolepis, possible dinosauriform Silesaurus, crocodile-like labyrinthodont Cyclotosaurus, and the predatory rauisuchian Polonosuchus.”

Figure 1. Dzik and Sulej are so sure that their Ozimek was a spectacular big sister to Sharovipteryx that they gave a model gliding membranes and used the largest disassociated humerus for scale. More likely it was an aquatic animal that did not move around much underwater.

Figure 3. Dzik and Sulej are so sure that their Ozimek was a spectacular big sister to Sharovipteryx that they gave a model gliding membranes and used the largest disassociated humerus for scale. More likely it was an aquatic animal that did not move around much underwater due to its weak musculature. The model was built based on crappy reconstructions of Sharovipteryx.

Forelimbs
Dzik and Sulej take the word of Unwin 2000, who did not see forelimbs in Sharovipteryx (and illustrated it with Sharov’s drawing), rather than the reports of Sharov 1971, Gans et al. 1987 and Peters 2000 who did see forelimbs. The latter three authors found the  forelimbs were short with long fingers, distinct from the gracile forelimbs and short fingers found in Ozimek. So, that’s one way to twist the data to fit a preconception. New specimens often get a free pass when it comes to odd interpretations, as we’ve seen before in Yi qi and others.

Manus and pes
In the reconstruction it appears that the medial and lateral digits are flipped from standards. This is both shown and repaired in figure 2.

According to the scale bars
the ZPAL AbIII/2511 specimen is exactly half the size of the ZPAL AbIII/2012 specimen. That issue was not resolved by the SuppData  The humerus shown in the 2012 specimen is not listed in the SuppData. Even so, the authors also ally another large humerus (2028) to Ozimek, and this provides the large scale seen in the fleshed-out model built for the museum and the camera (Fig. 3).

Built on several disassociated specimens
the reconstruction of Ozimek (Fig. 2) is a chimaera, something to watch out for.

Initial attempts at a phylogenetic analysis
based on the reconstruction pointed in three different directions, including one as a sauropterygian based on the illustrated dorsal configuration of the clavicles relative to the coronoids. If the clavicles are rotated so the vernal rim is aligned with the anterior coracoids the dorsal processes line up correctly with the indentations on the scapula (Fig. 2), alleviating the phylogenetic problem.

Lifestyle and niche
Sharovipteryx has an elongate scapula and coracoid, traits lacking in Ozimek. Sharovipteryx also has an elongate ilium and deep ventral pelvis, traits lacking in Ozimek. The limbs are so slender in Ozimek, much more so than in the much smaller Sharovipteryx, that it does not seem possible that they could support the large skull, long neck and long torso in the air – or on the ground. This is a weak reptile, likely incapable of rapid or robust locomotion. So instead of gliding, or even walking, perhaps Ozimek was buoyed by still water. Perhaps it moved its spidery limbs very little based on the small size of the available pectoral and pelvic anchors for muscles, despite those long anterior caudal transverse processes. Those might have been more useful at snaking a long thin tail for propulsion.

If we use our imagination,
perhaps with a large oval membrane that extended from the base of the neck to fore imbs to hind limbs Ozimek might have been like a Triassic water lily pad, able to dip its skull beneath the surface seeking prey, propelled by a flagellum-like tail. Not sure how else to interpret this set of specimens.

References
Dzik J and Sulej T 2016. An early Late Triassic long-necked reptile with a bony pectoral shield and gracile appendages. Acta Palaeontologica Polonica 61 (4): 805–823.

wiki/Ozimek (in Polish)

A unique Late Triassic dinosaur assemblage

Cabreria et al. 2016
bring us two new taxa, Ixalerpeton and Buriolestes from the Late Triassic. Originally considered a lagerpetid and a carnivorous sauropodomorph. “This is the first time nearly complete dinosaur and non-dinosaur dinosauromorph remains are found together in the same excavation, clearly showing that these animals were contemporaries since the first stages of dinosaur evolution.”

Figure 1. Ixalerpeton bones (above) as originally reconstructed (below) and a new reconstruction (middle).

Figure 1. Ixalerpeton bones (above) as originally reconstructed (below) and a new reconstruction (middle) as a protorosaur with a small scapula, as in Malerisaurus, and short cervicals, as in Boreopricea.

Ixalerpeton polesinensis
(ULBRA-PVT059) was originally considered a laterpetid (close to Lagerpeton, which they considered a dinosauromorph), but distinct from Lagerpeton in several ways. The large reptile tree (LRT) nests Ixalerpeton among the basal protorosaurs. Protorosaurs were not part of the original inclusion set presented by Cabreira et al., so taxon exclusion may be an issue here. As in the protorosaur, Malerisaurus, the scapula was not large in Ixalerpeton. As in another protorosaur, Boreopricea, the cervicals were not elongated.

Cabreira et al. report
“The parietal and frontal bones of Ixalerpeton polesinensis form a skull roof broader than that of most early dinosaurs. A large postfrontal fits laterally to the frontal, as more common to non-archosaur archosauromorphs.” In other words, these are clues that the inclusion set needs to be expanded. Taking a look at the pelves of candidate sisters provides some clues to the affinities of Ixalerpeton, but, of course, all the traits count.

Figure 2. Ixalerpeton pelvis compared to Lagerpeton, Tropidosuchus, Chanaresuchus, Prolacerta and the SAM K 7710 specimen of Youngina.

Figure 2. Ixalerpeton pelvis compared to Lagerpeton, Tropidosuchus, Chanaresuchus, Prolacerta and the SAM K 7710 specimen of Youngina. The latter two are taxa that frame Ixalerpeton in the LRT. No perfect matches here, but the vertical pubis on two of them match them.

Buriolestes schultzi
(ULBRA-PVT280) was originally considered a carnivorous basal sauropodomorph. Several cladograms were presented based on the addition of these taxa to prior analyses. The LRT nests Buriolestes with the basal theropod, Tawa, which was part of their analyses, but none of their topologies match the topology of the LRT, in many regards based on taxon exclusion.

Figure 2. Buriolestes reconstructed along with skeletal elements, some of which have been colorized for segregation.

Figure 2. Buriolestes reconstructed along with skeletal elements, some of which have been colorized for segregation.

Cabreira et al. report
Buriolestes schultzi corresponds to a sauropodomorph dinosaur, as indicated by

  1. a mandible tip with a ventrally inclined dorsal surface
  2. a deltopectoral crest that extends for more than 40% of the humeral length.”

Drawings of Tawa (Fig. 4) indicate a straight or elevated mandible tip, but the fossil has a ventrally inclined dorsal surface.

Indeed Tawa does not have a 40% deltopectoral crest, but it also does not have a reduced antebrachium. Similarly in T. rex, Segisaurus and other reduced forelimb theropods the deltopectoral crest extends relatively further down the short humerus.

Figure 4. Skull of Tawa. Note the descending mandible tip.

Figure 4. Skull of Tawa. Note the descending mandible tip not reflected in the drawing.

The Cabreira et al. study finds:

  1. Dinosauromorpha is composed of Lagerpetidae and Dinosauriformes. (but see this recent post in which Novas and Agnolin 2016 nest Lagerpeton with Tropidosuchus in the Chanaresuchidae)
  2. Lagerpetidae is composed of Lagerpeton, Ixalerpeton, and Dromomeron. (but see above)
  3.  Ixalerpeton and Dromomeron are sister taxa. (not when tested with protorosaurs)
  4. Dinosauriformes includes Marasuchus and a clade with all other members of the group. (Marasuchus nests with a clade of theropods that are not often included in analyses in the LRT.
  5. Saltopus, Lewisuchus, and Pseudolagosuchus form a polytomy with Dinosauria. (additional taxa in the LRT disrupt this polygamy and other taxa nest as outgroups to Dinosauria).
  6. Dinosauria of composed of the Saurischia and Orithischia lineages. (in the LRT the division is between Theropoda and Phytodinosauria).
  7. Asilisaurus is the sister group of all the other ornithischians, including Silesauridae. (in the LRT Asilisaurus and Silesaurus are poposaurs, not related to ornithischians.)
  8. Silesauridae is composed of Eucoelophysis, Silesaurus, Sacisaurus, and Diodorus. (only Silesaurus and Sacisaurus are tested in the LRT)
  9. Silesauridae is the sister-clade of the group composed of broadly accepted ornithischians. (not in the LRT)
  10.  Herrerasauria is the sister group to all other saurischian dinosaurs. (in the LRT, all other dinosaurs)
  11.  Herrerasauridae is composed of Staurikosaurus, Herrerasaurus, and Sanjuansaurus. (not in the LRT where Staurikosaurus is close, but at the base of the Marasuchus clade at the base of the Theropoda)
  12. Herrerasaurus and Sanjuansaurus are sister taxa (not tested in the LRT)
  13. Tawa and Chindesaurus are sister taxa. (not tested in the LRT)
  14.  Guaibasaurus, Eodromaeus, Tawa + Chindesaurus, and Daemonosaurus are saurischians belonging neither to Theropoda nor to Sauropodomorpha (i.e. non-Eusaurischia). (in the LRT Daemonosaurus nests at the base of the Ornithischia.)
  15. Eusaurischia is composed of the theropod and sauromopodorph branches. (paraphyletic in the LRT)
  16. Buriolestes is the sister group of all other sauropodomorphs. (nests with the basal theropod Tawa in the LRT)
  17. Eoraptor is the sister group of all other sauropodomorphs with the exception of Buriolestes. (Eoraptor is close to the base of the Phytodinosauria in the LRT).
  18. Pampadromaeus is the sister group of Panphagia, Saturnalia + Chromogisaurus, and all other sauropodomorphs. (confirmed in the LRT, but Pampadromaeus is also the sister group of the Ornithischia in the LRT)

References
Cabreira SF et al. (13 authors) 2016. A unique Late Triassic dinosauromorph assemblage reveals dinosaur ancestral anatomy and diet. Current Biology (2016), http://dx.doi.org/10.1016/j.cub.2016.09.040